Beyond Vulnerability: Syrian Refugees in Urban Spaces in Turkey

Since 2011, 5.6 million people have fled Syria due to ongoing conflict. In Turkey alone, 3.6 million Syrians are confronted with a series of constraints once in the host country. This paper analyses, within the context of urban exile in Turkey, the different experiences and survival strategies of Syrians who are modulated by particular relations of race, class and gender. It aims to explain how refugees manage to create their own visibility in this new space full of limitations, and further explores how their newfound participation in these urban areas can deconstruct dominant representations of refugees, who are otherwise seen as threats or as voiceless victims. In all, this paper aims to go beyond the vulnerability of refugees, without neglecting the violence they endure. To do so, the study was conducted using a series of semi-structured interviews, complemented by an ethnological approach.


Introduction
Since 2011, 5.6 million people have fled Syria seeking refuge in several countries, notably neighbouring Turkey, Lebanon and Jordan. So far, the Turkish Government has registered 3.6 million Syrian refugees (United Nations High Commissioner for Refugees [UNHCR] 2018), with most (90%) living outside camps in urban areas. At the beginning of the crisis in the territory of el-Cham, 1 Syrians could cross the Turkish border quite easily, with the risk of summary deportations being low. However, from 2015 onwards, policy changes and border closures began, and Turkey repealed a six-year agreement on 8 January 2016, which allowed visa-free entry for Syrian citizens. With the hardening of policies to deter the influx of people, different forms of abuse began to arise, particularly at the land border. Numbers may vary, but according to the Syrian Observatory for Human Rights (2019), until 2019 at least 417 civilians have been killed by Turkish border guards while trying to cross in search for safety. In addition to summary deportations, Turkey has also built a wall along most of its 911 km border with Syria.
Two major events will affect the lives of Syrians in Turkey: the 'Statement' signed on March 2016 between the European Union (EU) and Turkey to deal with the so-called 'migration crisis', 2 and the Turkish coup d'état attempt in July and its subsequent implications. Within this context of tension, the war of words intensifies and refugees are caught in the middle of the battlefield, both nationally and internationally. For example, in November 2016 during a speech in Istanbul, Turkish President Recep Tayyip Erdoğan threatened Europe with allowing refugees to cross into the continent (New York Times 2016). At that time, the media concluded that if the speech were not well received by Europe, nobody would dare to react 'because the migratory threat is very real' (Reuters 2016). Internally, while the Turkish Government calls Syrians 'brothers' it also emphasises them as a 'burden' (Middle East Institute 2017). In this sense, the victim narrative remains as such as long as they are not in 'our territory'-they are the victim of 'others', but not the victim of the nation's migratory policies, which reinforce their vulnerable condition (Mainwaring 2016: 290). This double portrait-victim, as victim of others, or potential danger once in 'our land'-guides the implementation of current racialised, 3 classist 4 and gendered 5 migration policies.
Upon leaving their country in crisis and arriving in a chaotic Turkey, further facing conditions that heighten their vulnerability amid attempts to leave home for Europe (or further abroad), Syrians live from crisis to crisis, stigmatised and exploited. They are perceived as a threat and are subject to political games, treated as a second-class population. Despite this, life continues in exile. They find new destinations, or stay in Turkey, while fighting the stigma and violence of everyday life.
Within the vast literature devoted to the Middle East and the study of migrations in the region, the Syrian question is gaining more space but remains limited. In the available literature, authors have described the humanitarian responses (see Canefe 2018;Çelik and İçduygu 2018), while others focus on security issues, management of borders or integration (see Balci 2012;Balci and Tolay 2016;Şimşek 2018). Some fit into the study of social movements and the theory of collective action (see Al-Abdullah 2018;Mahmud 2016), but many will denounce the politicisation of the Syrian refugee situation, inserting this instrumentalisation in a broader context of hardening migration policies on a global scale (see Freedman, Kivilcim and Baklacıoğlu 2017;Marcou 2017). In a study conducted with displaced Syrians in Lebanon, Akesson et al. (2018) examined the ethical implications of qualitative research while exploring the everyday mobilities of Syrian families in the country. That said, thorough analysis of survival strategies developed by Syriansthat is modulated by particular relations of race, class and gender-in a context of urban exile in Turkey remains nevertheless in the shadow. It is important to note that relations of race (or ethnicity), class and sex (or gender) constitute the forms of social categorisation and hierarchy also in exile, which will frame, empower and constrain the social actors. These cumulative experiences can produce heterogeneity within each category, which is co-produced through multiple power relationships (Pfefferkorn 2011;Poiret 2005). With these elements, this paper proposes to analyse how refugees transform their experiences into a source of contestation and fight against this 'crisis', of which they are considered at the centre. It also aims to explain how refugees overcome the constraints and categorisations imposed by an environment of exception, and further focuses on the labour market dynamics and the change of roles. By questioning public policies, this paper will show how the reductive and passive status granted to refugees is limited www.crimejusticejournal.com IJCJ&SD 36 2020 9(3) and makes problematic a real understanding of social relations developed in exile. Overall, the idea is to, without neglecting the violence they suffer, go beyond the vulnerability of refugees. Although the notion of vulnerability is central to the definition of refugee status (Cambrézy 2007: 3) in many contexts, it is imperative to understand that vulnerability is not a condition in itself, but rather created not only by the conditions in the place of origin that forced their departure, but also by policies of (non-)reception and the harsh conditions of exile (Freedman 2017: 38). To do so, the study is based on a series of semi-structured interviews conducted with different profiles of refugees during three months in Turkey in 2017. A total of 33 interviews were conducted with Syrians, including three group interviews. Of the respondents, 53% were women compared to 47% men, with 63% aged between 19 and 39. Further, only 29% of interviewees declared possessing any post-secondary degree. Questions about ethnicity, identity or confession were not automatically included for discussion. This approach seemed the best choice because (1) the sectarian character of the war in Syria deepened some existing divisions that were transferred to exile, causing tensions, suspicion and fear (Balanche 2018). Further, (2) analysing Syrian society through the theory of religious confession and ethnic minorities-which notably opposes Sunnis to Alawites, including Christians and Kurds-seems incomplete and problematic, not only because these categorisations are often arbitrary but also because Syrians themselves frequently do not identify with these divisions. The idea was not to produce a one-dimensional scheme of Syrian society, even because some communities may be considered a minority twice-for example, from the ethnic point of view and from the religious point of view. 6 In addition, the intersectionality of race, class and gender can be cumulative, having different effects on an individual's experience (Belkhir and Barnett 2001: 158). (3) In Turkey, the host country, minorities have also been historically conceived as a 'problem' (Karaosmanoğlu 2010;Minority Rights Group International n.d.), thus, stigmatising identification with any group. Consequently, refugees could be targeted twofold. (4) Finally, given the tense context, this approach seemed to respect the ethical guidelines to conduct research with displaced people, that must go beyond the preoccupation with a rigorous methodology and the mere compliance with notions of only doing 'no harm' (Krause 2017;Pittaway, Bartolomei and Hugman 2010). In that sense, when interviewees would spontaneously mention the subject, it would be discussed. Following this approach, some interviewees declared themselves, for example, Kurds, Turkmens and Sunnis.
One key concern in the interviews was representativeness and preventing research from contributing, even involuntarily, to silencing the voices of people who are often already overlooked (Krause 2017: 9). As such, we used snowball interviews in different contexts and cities. Since the population is difficult to identify in urban spaces, this approach also helped to ensure greater participation. In addition, the method proved its relevance for assessing the importance of networks (solidarity, friendship, professional, family, etc.) and, more specifically, what Pierre Bourdieu (1980) calls social capital in the construction of strategies in exile. Many informal discussions also took place with refugees, local communities, members of associations and international organisations. Although a beginner, the researcher's previous study of Arabic eased the introduction process during interviews, and, in many cases, a person who could translate worked also as a facilitator. Indeed, the use of interpreters during field research is not new, as they help not only with the translation aspect of discussions, but also with the identification of potential respondents. In complex places such as Turkey, where access to refugees is comparatively difficult, we relied on locals for support. This allowed us, in particular, to benefit from the social capital local facilitators enjoyed, and, in turn, helped build ties between the researcher and the respondents. Despite positive contributions in relation to the survey, the presence of someone other than the interviewer and the respondent did not come without compromise. In this context, particular attention was given to the relationship between interviewee and translator, with the different power relationships that this interaction imposes. For example, one female interpreter preferred to consult strictly with women in interviews whenever possible. Some interviews were also conducted without an intervening facilitator or translator, which functioned as a way of measuring the effect that a third person has during interviews, ensuring the objectivity of analysis.
The nature of translation also requires reflection. We cannot forget that 'all translations contain a certain level of interpretation of what has been said, and content … can be lost' (Krause 2017: 17). To cope with www.crimejusticejournal.com IJCJ&SD 37 2020 9(3) that, we developed verification mechanisms. In addition to reminders during interviews, if a translation was not yet satisfactory, the audio recorded during discussions would be retranslated while guaranteeing the anonymity of interviewees. Moreover, it is interesting to consider that every speech is a discursive translation of a thought (Maitilasso 2014). Thus, the narrative of their exile is a form of translation.
Sociologists are also responsible for translation of the field he or she is studying. Thus, although it is well known that the interpretation of certain words can be problematic, we argue that translation has an essential role, notably to share knowledge. As access to the field was particularly difficult during this period, mostly because Turkey restricted research with Syrian refugees, 7 the investigation was limited to three cities: Istanbul, Ankara and Gaziantep, near the border with Syria. These cities were chosen based on their relevance. First, note that Gaziantep is a border city, sharing historically many ties with Syria. 8 Nowadays, Gaziantep is among the cities that host most Syrians in Turkey, not only in absolute number but also in percentage of the population. It is also important to note that it was in Gaziantep that the Syrian Interim Government established its offices. Istanbul, in turn, is the biggest host city for Syrians in absolute number, with more than 547,000 registered refugees living there (Ministry of Interior Directorate General of Migration Management 2019). Finally, Ankara, as a capital city, had strategic importance for the research, given most organisations reside there. Use of this methodology was complemented by taking an anthropological and ethnological approach to these urban spaces in Turkey, particularly to support the work of local NGOs, observe spaces of socialisation and to visit Syrians in their homes.

From Crisis to Crisis
Although Turkey receives the largest number of Syrians globally, they have never had refugee status despite being registered in the country. It is important to point out that Turkey signed the 1951 Geneva Convention with a geographical limitation, which means that only European citizens are eligible for 'refugee' 9 in the country. The Syrians, who at the beginning of the crisis were thought to return quickly to their homes, were, therefore, considered to be 'guests'. To fill the legal gap and regulate the sudden increase in Syrians crossing the border, Turkey set-up in 2013 a series of mechanisms, starting with the Directorate General of Migration Management (Göç İdaresi Genel Müd) within the Ministry of Interior, which is now responsible for implementing migration-related policies. In addition, the Turkish Government relies on other ancient institutions such as, for example, the Directorate of Disaster and Emergency Situations, the Turkish Red Crescent and a number of new humanitarian associations and organisations often with religious links, more or less related to the government (Pérouse 2013: 86). In practice, this guarantees multi-level control over registered individuals. It is within this context that Turkey established the system of 'Temporary Protection' (TP) for Syrians, which provides access to a few social services including health and education. However, these rights remain partly formal due to the lack of resources, issues related to the registration process, and language barriers that prevent people from accessing those rights. In addition, the temporary nature of this protection hampers a long-term policy towards Syrian refugees in the country-although, after more than eight years the scenario indicates to a protracted situation. Thus, as guests they can never truly feel at home. This sentiment was echoed by several people interviewed for study, who demonstrated discomfort with their temporary situation, considered too unstable, which prevents them from thinking about the future.
Alongside this, Europe has continued attempts at closing and securing its borders. On 18 March 2016, this culminated to the signing of a controversial 'Statement' between the EU and Turkey, which has since increased the risks for refugees. Qualified as a statement, and not exactly as an agreement, to avoid the legal consequences that such a document might imply, it stipulates that Turkey agrees to readmit all those who arrived on the Greek Islands in an 'irregular way'-all categories included. In exchange, the EU promised to liberalise the Schengen visa policy 10 for Turkish nationals, to pay EUR€6 billion in humanitarian aid and to reopen Turkey's EU accession process. attempting to enter the Greek Islands following the Statement. It must be emphasised that 'people readmitted to Turkey … could not have access to a lawyer and most of them were jailed' (Preiss 2016: 37), challenging the status of 'safe country' granted to Turkey (European Commission 2020).
Several factors have affected the already difficult life of Syrians in Turkey, including the coup attempt on 15 July 2016, followed by declaring a state of emergency, subsequent changes in various State institutions and organisations, and the proposed constitutional amendment introducing a controversial new presidential system. In some areas, tensions with the host community have increased, resulting in acts of violence. 11 Several Syrians interviewed said that the coup attempt and the actions that followed-such as the closing of universities, the deportations of Syrians arbitrarily accused of being linked or sympathising with the coup, and various conflicts with the local community-reminded them of the crisis back home. At the same time, this chaos heralded itself as a new crisis to be fought, except this time with increased precaution and distance, as the fear of being involved in Turkish internal affairs and becoming subject to new forms of persecution is huge.

Beyond Vulnerability: Developing Survival Strategies
The omnipresence of violence (symbolic or physical) perpetrated by different actors, and the stigmatisation that Syrians face, viewed as an internal and external threat or as passive voiceless victims 12 -only benefiting from humanitarian aid and spending the resources of host countries-together reinforce the insecurity and vulnerability of people in exile. Further, dominant representations in the media and in public discourse, despite the variety of profiles, only reinforce these stereotypes and negatively affect one's experience as a refugee. At the same time, refugees are framed as passive and lacking agency. With the change in their legal, social and economic status, combined with the prolongation of exile (especially in the context of ongoing social and political tension), most refugees are still trying to find their way in Turkey. The emergence of new possibilities, which are characteristic of new beginnings, is extremely important to overcome the situation, even to challenge in practice the categorisations to which they are subject.
Faced with this scenario of distress, Syrian refugees have developed strategies, which vary according to the actors and the context in which they are inserted. Between the duty to fight the 'crisis', of which they are the protagonists and must face consequent stigmatisation, and the need to survive in an adverse and hostile environment, the possibilities are limited. The development of strategies is balanced with the option of taking distance when confronted with different forms of danger-the latter of which also inevitably becomes a form of survival strategy. In this context, one key priority for refugees in exile is to find work, as indicated in a variety of different fields of literature (Cambrézy 2007: 14;Safi 2011;Zetter and Ruaudel 2018). Since aid is insufficient or even non-existent, work remains a path to certain freedom-the beginning of the reconstruction of human bonds, while simultaneously disrupting the stigma attached to the perception of humanitarian aid. Regarding the labour market, it is important to point out that Turkey adopted in January 2016 the Regulation on Work Permits of Foreigners under Temporary Protection. According to this regulation, persons under 'temporary protection' may apply for a work permit six months after completing TP registration. There is a quota of 10%, which means that for every Syrian, 10 Turkish citizens must be employed. People under TP can only benefit from a work permit exemption in two sectors: agriculture and livestock. In this scenario of restrictions, which prevails in most receiving countries (Zetter and Ruaudel 2018: 4), the vast majority of people work without a permit and on a temporary basis, vulnerable to exploitation. This is the case for 22-year-old Mustapha, 13 who lives in the district of Solfasol Malhesi, 7 km from the city centre of Ankara. Rent is cheaper there and many Syrians live in the area. Originally from Northern Syria, Mustapha tells about his resignation on his way home from work at a small factory in the region: I just quit … he paid all the other employees except me. It's been weeks … I spoke to him but he did not want to pay me, so I resigned.
www.crimejusticejournal.com IJCJ&SD 39 2020 9(3) Work without a permit, which is evident in many host countries (Zetter and Ruaudel 2018: 5), is often the target of Turkish authorities and the local community due to a supposed disloyal competition between Turks and Syrians in the labour market, and particularly in less specialised sectors where Syrians would likely work for less and without the required regulations. However, resorting to engage in undocumented work is rather the result of quotas and the marginalisation of Syrians in Turkey, as well as (more broadly) the restrictions imposed by public policies on refugees. This puts them in a vulnerable position, rife with exploitation. Therefore, the practice of working without a permit is for many refugees a survival strategy within an unequal system. For those interviewed, undocumented work experience was often ambiguous: although access to the labour market ensures one's ability to secure the basic needs of their family, it is also a source of frustration due to its instability. As such, many Syrian refugees feel humiliated by employers who take advantage of their condition by underpaying and abusing them. Moreover, without access to qualified positions, such as engineering, refugees often suffer from downgrading. The challenges imposed by the language only increase the disadvantages experienced by most Syrians in the labour market. For Zeid (50, originally from Aleppo), who currently lives with his wife and three sons in Gaziantep, the first key difficulty regards their status as 'alien' under TP: I am a teacher in Syria … I taught for about 23 years there … Here, I found a job only temporarily, for three months, six months, nothing else. I looked for another one … Two months ago, I went to a Syrian bakery; I only worked there for one day. I could not continue. I am old. I wanted to bring money for my family, but I could not … There were very difficult conditions … Nowadays, I have no salary, so I had to send my children to factories to work just to bring money home … They will work until I can find a job.
Zeid's testimony opens up a course for reflection with regards the refugee experience in Turkey and the strategies developed in response to a so-called 'crisis', in which Syrians are supposed to be the protagonists. Zeid, who was a teacher in Syria with financial stability, now suffers from a social, cultural and professional downgrade; his sons must work and he continues to seek for a job, while fighting the stigma to which he and his family are subject. Indeed, the strategies developed in this context may vary among Syrians. Factors such as age, facilities to learn a new language, gender, pre-exile economic, social and cultural capitals, and consequent changes following exile all shape the nature and implementation of individuals' coping strategies. Naturally, there is no single approach to manage the constraints of displacement, simply because of the heterogeneity within the Syrian community, and the multiplicity of structures and contexts that respond to these strategies.
Note here the use of minors' work as an example. This negative strategy (Ballet et al. 2010) affects several families, like Zeid's, including those whose social, economic and cultural capital were important in Syria but has since dwindled in Turkey. In practice, implementation of this negative strategy would be different among different groups. For example, families who have experienced a downgrade can determine, implicitly, an age limit for sending their children to work, which is around 14 years old. Meanwhile, the poorest of families, who have not necessarily experienced a significant downgrade in circumstance since their situation in Syria was already difficult, can resort to sending minors aged less than 10 years old to work. In fact, it was not uncommon to find in the streets of Istanbul children under 10 selling napkins or candies.
In relation to Zeid, who was forced to send his two eldest sons (one of whom is aged only 16) to work in a factory, it is possible to observe a clear limit in the use of child labour as a survival strategy. He, for example, was shocked when asked if his younger child, who was nine and a half at the time of interview, was also working: 'Of course not, he is at summer school.' Zeid's decision against sending his youngest to work and his surprise when questioned about that possibility are indicative of a desire to fight the dehumanisation and stigmatisation that he and his family experience in forced exile, while simultaneously maintaining his values prior to seeking refuge. Evidently, Zeid does not want to move away from what he was before this forced downgrading, notably at the level of belonging to a specific social class that possess an expressive www.crimejusticejournal.com IJCJ&SD 40 2020 9(3) cultural capital. Indeed, several interviewees spoke of the 'shame' and 'humiliation' to define their feelings about the social downgrading that exile represents.

Intersectionality Beyond Borders
While exile often dismantles most of one's capitals-economic, social, symbolic and cultural-this does not mean that power structures have been completely disrupted in exile. Indeed, class, gender and race are reproduced at different scales within the group, consequently modulating the different strategies for survival.
For example, Syrians with economic capital have created their own businesses and organisations, which are not forbidden in Turkey, and tend to hire mostly, if not only, Syrians. This can be an example of a collective strategy. Notably, Oubai (28, originally from Aleppo) opened an IT company in Gaziantep: Oubai's testimony reveals a particular inventiveness-to hire only Syrians, without a permit, and then to flee in the case of verifications by the Ministry of Labor. To deceive the authorities in this case is to fight the limitations of their condition. 14 The creation of restaurants and companies represent the restructuring of Syrian society in exile. The preexisting networks or the new ones created in Turkey are the basis for this collective strategy. Indeed, it must be emphasised that this is not a uniform dynamic. Instead, it varies by city, and especially by group, due to pre-existing inequalities or even inequalities that arise with the condition of exile. In that sense, both Gaziantep, a border city with a geographical concentration of Syrians, and Istanbul, which also contains a high number of Syrians, were most representative in the Turkish region, as they presented many new Syrian business and organisations. However, in Ankara, for example, Syrian isolationism is the norm, giving space to individual survival strategies. It should also be noted that while this may correspond to a collective strategy capable of strengthening the group's powers, it does not mean that different forms of exploitation over more deprived Syrians by ones belonging to a more privileged social strata does not exist. These relations of power and domination are exemplified by Syrian factory owners in the Gaziantep region who will hire other Syrians (including minors) under a regime of extreme precariousness, often with very low wages and long working hours (Financial Times 2017; Hurriyet Daily News 2018). They justify a burden of instability on their Syrian co-nationals due to the constraints of exile, while recognising that they have become more enriched in Turkey than in Syria. Here, it is not a question of underestimating the forces that represent the structures of exile, 15 especially in relation to the limitations imposed by the labour market (underlined through the example of Oubai), but to recognise the reproduction of power structures in exile.
Many Syrians from wealthier strata (despite the difficulties experienced in exile) reject the possibility of registering with the government for TP, 'because this would represent to ask for a refugee ID'. 16 It is possible to observe a strategy of negotiation within the double categorisation that could be imposed on those in exile-that is, as refugees, with all the stigmas this might imply, but also in terms of social class. For example, in Istanbul several attempts to distinguish Syrians who resort to begging, among other groups, have been documented, with 'campaigns … even organised and relayed by young Syrian activists on social media denouncing that "the beggars of Istanbul are not Syrian" ' (Al-Abdullah 2018: 36).
Syrians identified as Kurds and Turkmens represent particular cases within the refugee community in Turkey. 17 Turkmens (who speak Turkish) were once highly stigmatised in Syria; however, in exile they could benefit from historic ties, notably in border cities such as Gaziantep, and forge entirely new positions www.crimejusticejournal.com IJCJ&SD 41 2020 9(3) for themselves. Due largely to their language skills they could more easily enter the job market in Turkey, and some of them even found positions in the formal labour market. It was not uncommon to find Turkmens working in NGOs in positions such as translators. Their social and cultural capital was, in this case, fundamental to developing their survival strategies in exile, to subvert the social hierarchies to which they were subject back in Syria, and to fight the stigmatisation they may experience as refugees.
Kurds in turn encounter a more complex scenario. Stigmatised in both societies, but at the same time enjoying similar cultural and social capitals at least when it comes to the Kurdish community in Turkey, their experience is conflicting. At one hand, as they also speak Kurdish and many had previous family or friendship ties with other Kurds in Turkey, they could use these forms of capital to enhance their power in exile, notably in the first years of seeking refugee. For example, many chose to live in neighbourhoods (in Turkey) with already established Kurdish communities, such as in the peripheries of Gaziantep or yet in the low-income region of Tarlabaşı in the Beyoğlu district in Istanbul. On the other hand, within the Turkish context, Kurds are already highly marginalised, which, in turn, affects new arrivals.
Take 47-year-old Ghada's experience in registering for a TP card: We went to the police station and they told us that we should bring the owner of the house, but he refused it, because he is Kurdish and he was afraid of going to the police.
Evidently, Kurdish communities from Syria living in Turkey are targeted twofold, both as refugees and as Kurds, whereby even the most elementary activities such as registering can become an impossible task.
In turn, new power relations have emerged based on this double portrait of Kurds within the restrictive structures of Turkish exile, creating two different categories: Kurds from Syria and Kurds form Turkey, who are hierarchised. The symbolic power of Kurds from Turkey relies on their belonging to Turkey, while those from Syria are frequently described as Syrians only. The label 'refugee' is harder to be disrupted in that case.
Refugees who cannot find jobs and those who cannot work must try to find their place in the new society that surrounds them. The simultaneous fight against victimisation and against the precariousness of exile forces many to construct other strategies, highlighted by the need to change roles and identities while in exile for survival. For example, Aida (28, originally from Aleppo) lives with her husband and four children in the periphery of Ankara: I go to shops to buy fruits and vegetables, everything for the house, because all the time my husband is at work … In Syria, he used to do that and I stayed at home, it was the responsibility of the man to do that … The first time, I felt that it was very difficult for me, but now I am used to it.
During exile, Aida has had to acquire other responsibilities within her household. To account for her husband's absence at home to carry out many of the tasks that traditionally were his responsibility, she had to reposition herself in both domestic and public spheres. This is an example of change of roles in exile, exhibited in this context as a parameter experienced in many refugee households. Like Aida, women have repositioned themselves in the social sphere, joining (in many cases) the labour market for the first time.
In Gaziantep and its surrounds, women work with sorting spices or in shoe production on demand from their homes. Indeed, those who had significant cultural capital and were already working in Syria were able to better position themselves in the labour market in Turkey. For example, in NGOs it is possible to find many highly and formally educated women, notably responsible for conducting face-to-face encounters with beneficiaries of the organisation. Although they remain a minority, some could even be employed in the formal sector. Evidently, despite facing difficulty, refugee women try to raise their voices and find a place in exile, sometimes even appropriating themselves of gender stereotypes for their own www.crimejusticejournal.com IJCJ&SD 42 2020 9(3) benefit, as observed within the economic activities that they develop. Increasingly, these are often linked to social constructions that separate ideas of 'femininity' and 'masculinity'.
It is important to highlight that although this change in roles (whether as an individual or even a family strategy) has permitted some forms of female empowerment in exile, these women are often less paid than men in already precarious positions. This ultimately demonstrates the transfer of vulnerabilities, related to individuals' previous categorisations, to other spheres in new contexts. Unfortunately, this only further reinforces the stereotypes to which they are subject.
Although respondents expressed feelings of accomplishment to act, despite facing severe limitations, their changing roles do not come without discomfort, particularly for women. Generally, they often demonstrated frustration or fatigue with the accumulation of responsibilities caused by a very unequal distribution of domestic work between men and women. 18 Also, in some cases, these changes can be a cause of argument for couples, disrupting identities and previous social constructions. In this case, it should not be forgotten that 'structural gender inequalities will have a significant impact on men's and women's migration routes and experiences' (Freedman 2017: 32).

Conclusion
Refugees are often viewed as threats or as voiceless victims (Agier 2006: 159). Evident from this study, the participation of Syrians attempting to build new lives in urban areas in Turkey deconstructs many dominant (and delimiting) representations of those in exile. Many have been forced to develop different survival strategies to cope with new conditions, and these are (whether individual or collective) often based on the experiences and the repertoires of life of refugees. Indeed, much is equally driven by social, economic and cultural factors, crossed with relations of race, gender and class.
As demonstrated, exile has repositioned Syrians in the social sphere and forced many, living under increasingly precarious circumstances, to develop diverse strategies for survival, notably regarding the labour market and the changing roles of refugees. Nonetheless, the ongoing reproduction of power relations-whether of class, race or gender-persists in exile, and is frequently exacerbated by the constraining conditions of migration. Although some women could, for example, benefit from a change in status due to new hierarchical relations of 'family', or even in the case of some Turkmens securing greater work opportunities due to their comparatively greater language skills (which provides evidence of the fight against the reproduction of power in exile and a real change in social hierarchy), it must be highlighted that these transformations remain limited.
It is interesting to note that the constraining nature of exile also imposes the transfer of previous vulnerabilities imposed in some categories of individuals' 'new' lives. Notably, the transfer of financial instability, particularly for women who are historically paid less than men, ends up reinforcing gender, class and race stereotypes, increasing one's vulnerability, in turn.
Indeed, the newfound participation of Syrians in urban spaces certainly reveals the agency of refugees and the emergence of a new social sphere in exile. However, their survival strategies also evidence the failure of public policies of reception, which only reinforce a host of inequalities. Thus, faced with increasingly uncertain circumstances, Syrians in exile must continue the fight in Turkey. 1 Until the nineteenth century, Syria was named Bilad el-Cham. However, today some Syrians still refer to themselves as being 'from el-Cham', notably to describe the region of Damascus and its surroundings. 2 As Jane Freedman (2017: 38) would argue, use of the word 'crisis' to describe the influx of Syrian refugees in Europe from 2014 onwards can be criticised if we consider that 'political leaders use the word "crisis" to justify the failure of a European political response to this "crisis" and to legitimise exceptional measures such as the closing of borders, the detention of refugees, or the return of refugees to Turkey'. 3 Here, we refer to this as 'a social relationship with concrete manifestations (stigma, discrimination, pressure, exploitation) for racialized individuals or groups' (Devriendt, Monte and Sandré 2018: 17). It is important to underline that we understand the limits related to use of the word, but it is a question of supporting a reflection on discriminations in exile, based on a sociohistorical process. We also share the analysis proposed by many researchers that highlight the use of other words (e.g., ethnic groups) to designate groups can also create these unequal relations describing the 'body of the other' (Gayon 2007: 273-297;Rozenberg 2007: 19-33). 4 Pierre Bourdieu's definition of 'class' in terms of field will be used as a starting point to reflect on class relations in exile. This is because he proposes a formulation based on the relationship between 'dominants' and 'dominated' from many different positions, thus, allowing us to conceptualise it in terms of 'class fractions'. This is important because in exile, with the brutal change in individuals' various capital, we can understand the relations between the 'dominant individuals' within the group of 'dominated', such as the one who suffered a downgrading but retains some cultural capital, or the relations between the poorest host community members and Syrians living in the same neighbourhood / building (Bourdieu, 2013: 31-32.) 5 It should be emphasised, as Jane Freedman (2007: 16) points out, that 'gender is a concept which has been employed by feminist research since the 1970s in order to make a distinction between the fixed characteristics of sex and the socially constructed notions of masculinity and femininity'. Thus, 'masculinity' and 'femininity', like the results of social constructions, vary according to the context in which they are inserted. 6 Classifications that are frequently used to describe the Syrian society vary. They can be confessional (such as Christians, Alawites, Druzes and Sunnis), but also ethnic or even 'national' (such as Kurds, Turkmens and Armenians), yet linked to different ways of living, such as bedouins (Seurat 2012). Currently, no recent official data on the different confessions in Syria exist. According to different sources, the vast majority of people are Muslim (between 71.6% and 83%), notably Sunni (68.4%), with Christians comprising the second biggest group (11.2%) (Izady 2020). Concerning ethnicity, the statistics are also vague. Data from the 1980s show that Druzes comprise around 3% of the population. Meanwhile, Kurds, considered more an ethnic minority than confessional (as the vast majority are Muslim), represent 7% of the population (Seurat 2012). 7 Nothing was formally established in the laws, but in April 2015 the Turkish Ministry of the Interior informed academicians that from that date they would need prior authorisation before conducting research on Syrian refugees living in the country. In practice, it was a ban, with many academics reporting that their research was blocked. The government justified the measure by announcing that it was to protect Syrians; however, academicians denounced an intellectual control and the silence of any opposition. 8 As inheritance from the Ottoman period, some territories were claimed by the two modern States, including the Sandjak (administrative subdivision) of Alexandretta, also called Hatay and Aintab Sandjak, known today as Gaziantep. Given to Turkey in exchange for its support for the Franco-British alliance in World War II, these territories were part of the Aleppo province in Syria prior to this division. The dispute lasted decades and it was only in 2004, during a period of friendship between the two, that the government of Damascus acknowledged the annexation by Turkey. In the region, Turkish and Arabic languages are often spoken. 9 The definition of 'refugee' is always contradictory and arbitrarily categorised. Officially and in accordance with Article A(2) of the 1951 Convention Relating to the Status of Refugees, a refugee is a person who has fled their country 'because of a wellfounded fear of persecution for reasons of race, religion, nationality, membership of a social group or political opinion'. Until today, the 1951 Convention remains the key legal document for determining who is entitled to refugee status, their rights and the legal obligations of governments towards refugees. It must be outlined that the Convention institutionalises an individual refugee definition-the so-called statutory refugee. According to Valluy (2009: 151), this is because 'it is not a question of hosting mass displacements'. These displacements, which are most common today, remain predominantly under the supervision of the United Nations High Commissioner for Refugees. The organisation, created in 1951, has a different status, notably without geographical limits, but only allows access to limited rights by refugees, precisely humanitarian rights. 10 Turkish citizens currently need a visa to visit one or several of the countries of the Schengen Area, which includes 26 countries. Within the agreement, Turkey demands visa-free entry to Turkish passport holders. 11 Daily reports of violence are noted in some parts of Turkey. During the night of the coup attempt in 2016, in some areas, such as the outskirts of the capital Ankara, Syrians were removed from their homes by Turkish residents who disagreed with their presence in the country. Various episodes of tension and violence are reported on occasion in the media, leaving Syrians to live in fear and, in some cases, stimulating stereotypes that spark tension between communities (Hurriyet Daily News 2017a, 2017b). 12 As Sayad (1999: 8) recalls, this criminalisation is often intrinsic to the act of migration. Regarding the victimisation experienced, Agier (2008: 291) emphasises the role of what he calls the 'humanitarian government', composed by a number of actors, notably United Nations agencies, international organisations and NGOs. According to him, these humanitarian institutions perceive refugees-the individuals they assist-as the 'absolute victim', contributing to their stigmatisation. 13 All names have been changed to guarantee protection and anonymity of the interviewees. 14 It must be noted that use of the word 'deceive' does not imply any form of judgement, nor does it question their experience in exile. Rather, it emphasises that labour market policies, as they currently exist, impose strong constraints on Syrians, thus, opening up space for the creation of new strategies to overcome these limitations.
www.crimejusticejournal.com IJCJ&SD 44 2020 9(3) 15 For the debate between structure, exploitation and the modes of production in capitalism, see Burawoy (1976). 16 A Syrian woman who works at a Syrian NGO justifies why, despite facing a multitude of difficulties, she rejects to register with the Turkish Government to access some social services: 'We did not register because they explained to us that this would mean to register as a refugee and we do not want that'. 17 Within these two groups, there are Turkmens Maktoumim, who are not registered but speak Turkish, and Kurds Maktoumin, who are also not registered but speak Kurdish. These two communities were viewed as semi-nomad in Syria and some of them had a late sedentarisation in the 2000s. They are not the particular subjects of this paper, as other studies focus on them specifically (see Al-Abdullah 2018). 18 In other contexts, several researchers have already observed more nuanced positions on 'the emancipation of migrant women', emphasising the strengthening of traditional roles or even greater control of men over women in the land of migration (Catarino and Morokvasic 2005: 12;Freedman 2007;Morokvasic 1984). Islamic feminism has criticised Western feminism, highlighting its effect on endogenous production, particularly with regard to the accumulation of work for women in the Middle East (Latte Abdallah 2010).