Welcoming immigrants in Istanbul: Gendering faith-based and professionalised hospitality

This article examines the hospitality practices of pro-migrant civil society organisations in Istanbul. Drawing from qualitative interviews, we focus on intersecting gendered, professionalised and faith-based aspects of pro-migrant activities and explore the ways that politically and morally charged ambivalences of hospitality practices are articulated and negotiated. Moreover, by contextualising Turkey’s religious and geopolitical particularity as a gatekeeper of Europe, we work with Derrida’s concept of plural laws to investigate hospitality practices towards refugees in Istanbul. Civil actors’ intentions and attempts to be good citizens, Muslims, and care providers expose the intimate aspects of hospitality – a segue into discourses of displaced subjects’ (gendered) deservingness. By portraying how macro–micro, global–local and public–private relations condition hospitality practices, we observe how globalisation is lived intimately, influencing perceptions of deservingness and the prioritisation of displaced subjects’ needs.

The reception that many immigrants receive in their flight to safer communities can be examined in relation to more general practices of hospitality. Hospitality is characterised by recognising others through the offering of space and time (Dikeç et al., 2009). It is at once an intimacy and a set of conditioned and conditional relationships (Derrida, 2000) The colonial genealogy and Eurocentrism implicated in discussions of the 2015 'refugee crisis' (Fiorenza, 2021: 23), when more than a million refugees arrived at the borders of the European Union (EU), led to the neglect of crises in neighbouring countries in the Global South. In Turkey, the so-called 'refugee crisis' began in 2011 with the first mass arrival of Syrians at its borders. At the time, European states did not see this as a 'crisis' for the Turkish government but as a situation of the willing hosting of 'guests' who needed government protection (Özden, 2013: 10). Later, due to the continuation of the war as well as increasing restrictions in Europe and the steadily growing numbers of 'guests' whose stay extended longer than anticipated, Turkey then recognised its own type of 'refugee crisis' (Bélanger and Saracoglu, 2019: 284). The (European) 'refugee crisis', which began in 2015, also influenced the course of events in Turkey. Aiming to control the crossing of displaced people from Turkey to the Greek islands, in March 2016, European states and Turkey agreed to what has become known as the EU-Turkey deal. Providing a €6 billion fund for hosting refugees in Turkey, EU member states agreed to take one Syrian refugee from Turkey for each person arriving irregularly on the Greek islands. This gatekeeping system has been the source of several disputes between the EU and Turkey over the years. Claiming to have not received all the promised funding, Turkey has repeatedly threatened to withdraw from the deal. As a consequence of Turkey's strategic geo-social position, the country now has more international non-governmental organisations (INGOs) and has received more funding for civil society to provide humanitarian aid to migrants. In this political and economic context, South-South migration has been shaped by the interventions of Northern states in migratory channels and processes.
In what follows, we first describe our theoretical framework. We then discuss the socio-economic and political context of Turkey within Southern geography, outlining the methodology of the study. The latter sections of the article examine pro-migrant civil society hospitality practices from the perspective of three related themes: faith-based motivations, gender relations, and professionalisation. Derrida's (2000) deconstruction of the term 'hospitality' has inspired rich theoretical debates on the unequal power relations between the host and the guest, especially through the tension between hostility and hospitality incited by conditionality. The (im)possibility of unconditional hospitality has been a fundamental concern for scholars (see, for example, Dikeç et al., 2009;Kelly, 2011;Lynch et al., 2011;Norton, 2015). Derrida traces an 'insoluble antinomy, a non-dialectizable antinomy' between 'the law of absolute, unconditional, hyperbolical hospitality' and: expectations, rules, norms, and prejudices. Strictly speaking, plural laws are the very conditions of hospitality. They provide the concrete spatial and temporal context of an ideational unconditional hospitality. We work with this understanding of plural laws to investigate civil society hospitality practices towards refugees. More specifically, we explore the relationships that pro-migrant civil society actors have with migrants in Istanbul. In our analysis, we approach the context of hospitality (i.e. the plural laws) by delineating three themes: faith, gender and professionalised relations. In our understanding, these three themes set the parameters of the plural laws through which the hospitality of pro-migrant civil society actors is instantiated. Dikeç et al. (2009: 9) argued that the perversion of unconditional hospitality is not an accident or aberration but a consequence of the experience (of hospitality) itself. The antinomy between 'the law' and 'plural laws' further highlights the dynamic relationship between hospitality and hostility. Hence, the existing norms, rules and power relationships under which hospitality is instantiated also hold the potential for hostility, that is, occasions when there is no hospitality, no welcome, but instead a mostly exclusionary or ignorant attitude. Hospitality relationships between civil society actors and migrants in Istanbul encapsulate the unsteady borders between hospitality and hostility with potential for change due to faith-based concerns and gender dynamics, as well as the role of the professional/bureaucratised workplace in shaping support activities. This is not to say that faith, gender or the professionalisation of support determine the conditions of hospitality and hostility. Rather, they provide the social setting within which vulnerable migrants seek hospitality. Faith and gender are recognised as conditions of hospitality (see Darke and Gurney, 2000;Hamington, 2010;Siddiqui, 2015), and migration in turn influences the extent of hospitality (see ; Dağtaş, 2017;Gibson, 2003;Kelly, 2006). In addition to contributing to these larger discussions, the study of these conditions in Turkey has the potential to identify a broader range of hospitality practices (Fiddian-Qasmiyeh, 2020).

Hospitality and conditionality
Various forms of secularism have dominated humanitarian agencies. However, the rise of the US Christian right, political Islam, and equality concerns, along with the simultaneous fall of socialism and communism over the past decades, have led to a rapid increase in faith-based initiatives (Hasu, 2013) in civic educational practices and hospitality initiatives. All three monotheistic religions consider hospitality a virtue (Siddiqui, 2015). In Turkey, Islam informs the social and political acts of the majority of faith-based organisations (Mackreath and Sağnıç, 2017;Sunata and Tosun, 2018). In addition to faith-based influences, gender shapes the power relations of hospitality, with feminist scholars highlighting how women are largely responsible for hosting and caring practices at home (Hamington, 2010). However, as Darke and Gurney (2000:83) warn, it is an oversimplification to see women solely as hosts and men as guests, with empirical analyses showing that in the context of migrant hospitality in Istanbul, it is much more common to see women as guests -in other words, as 'deserving' refugees. As a result, women employees are more in demand among civil society organisations. Recognising gendered power relationships and stereotypes makes women's situatedness in relationships of hospitality (and hostility) visible.

Contextualisation of the national setting and methodological reflections
As the country hosting the largest number of Syrian refugees, Turkey plays a prominent role for the contemporary Syrian diaspora. Turkey's diplomatic, economic and humanitarian expansion over the last two decades manifests its new political positioning and complicates a dichotomised conception of the Global North and South. Turkey has a long history of accession negotiations with the EU. However, especially during the last decade, there has been a reluctance to implement EU reforms and an increase in undemocratic measures (Soyaltin-Colella, 2020: 6). Additionally, the increasing tensions emerging from the EU's reluctance to accept Turkey's membership have led Turkey to adopt a more diplomatic approach to non-Western foreign policy (Donelli and Gonzalez-Levaggi, 2016). Turkey's growing role in the Global South, particularly in regard to diplomatic networks, interregional politics and cooperation, refugee hosting and humanitarian assistance, has contributed to its growing influence in the political landscape. While considering Turkey part of the Global South, we underline the importance of not reducing the Global North and Global South to homogeneous, reified entities (Fiddian-Qasmiyeh, 2020). Our aim is to recognise the distinct socio-economic and political context of Turkey with regard to refugee hosting.
After long being recognised as a country of considerable emigration, Turkey has slowly transformed into a transit country for immigrants. However, more recently, it has become a destination country for asylum seekers from neighbouring countries in the Middle East and beyond (Baklacioglu, 2009). Turkey has undergone legislative as well as administrative changes over the last ten years with the so-called Europeanisation of asylum policy, along with other institutional and legislative transformations (Baklacioglu, 2009;İcduygu and Kirişci, 2013). Not surprisingly, Turkey's Europeanisation of asylum policy has involved two contradictory trends: humanitarianism and securitisation (Baklacioglu, 2009). The Europeanisation process is not limited to the governmental level but includes engagement in different transnational networks as well as financial support for pro-migrant European organisations.
Despite a history of hosting undocumented migrants from former Soviet republics and transit migrants from Middle Eastern and African countries (Toksöz et al., 2012), migration was not a public concern in Turkey until the arrival of Syrians in 2011. There were approximately 3.6 million officially registered Syrians in Turkey as of November 2020 (Göç İdaresi Genel Müdürlüğü [General Directorate of Migration Management], 2020), and a growing body of literature has emerged on migration to Turkey (Dağtaş, 2017;Koca, 2015;Özden, 2013). Like other cosmopolitan cities, Istanbul attracts both internal and international migration. Nearly 500,000 of the 3.6 million Syrians in Turkey are officially registered in Istanbul. Since the beginning of 2018, the General Directorate of Migration Management has not accepted new applications from Syrians for registration in Istanbul except in extraordinary situations. (Karakaş, 2018). As a result, the population of unregistered Syrians without any access to social services has grown.
Pro-migrant civil society organisations have developed during the last two to three decades in Turkey with diverse characteristics in regard to their expertise, activities and level of organisational complexity. Professionalisation, institutionalisation and bureaucratisation of civil society organisations are contested concepts. A common thread is how professionalisation (or the lack thereof) is defined through the funding (or lack thereof) that an organisation receives. As we discuss in the next section, our findings suggest that funding generates and expands professionalisation in the field while simultaneously increasing bureaucratisation.
Some organisations engage in rights-based advocacy, whereas others provide specific social services and/or humanitarian aid (Mackreath and Sağnıç, 2017). Most of the organisations have institutionalised structures, whereas network-type activist organisations are relatively scarce (Erdoğdu and Şenses, 2015). National-based pro-migration civil society organisations multiplied with the arrival of Syrian refugees (Mackreath and Sağnıç, 2017). 1 The arrival of international funding to support mainly Syrian refugees has influenced pro-migrant civil society activism (Mackreath and Sağnıç, 2017). We have found that some humanitarian aid projects have highly structured guidelines and procedures due to the demands of funding agencies. Consequently, we identify an increasing trend towards the professionalisation of pro-migrant civil society organisations. Compartmentalisation in organisational structures and task divisions have often led to project coordinators becoming administrators while young social workers carry out face-to-face work with refugees. Social workers are university graduates who are relatively well paid in comparison to those holding similar jobs in the private sector. It is also important to note that Turkey imposed a state of emergency from July 2016 to July 2018, which constrained civil society activities. More than a thousand civil society organisations were closed, and the government instituted strict controls, such as mandating bureaucratic procedures for organisations, including having to seek permission to conduct fieldwork. During this period, some organisations moved away from rightsbased activism in favour of aid-based projects (Altunkaynak-Vodina, 2019).
According to statistics from the Interior Ministry, there are currently approximately 900 humanitarian aid associations and 100 human rights advocacy associations in Istanbul (Karakaş, 2018). There are no data showing how many of these organisations specialise in migration. The first author contacted the main organisations that engage in either rights advocacy or humanitarian aid, focusing on those in neighbourhoods with high numbers of refugees and aiming for a diversity of expertise. The neighbourhoods were Esenler, Fatih, Beyoğlu, Şişli and Sultanbeyli. Four of the sample organisations were established specifically to support migrants/refugees. The other six have a broader remit but also focus on migrants and the protection of their rights and/or provision of services. One is a women's organisation and another is faith based. Given the heterogeneity within pro-migrant organisations, we sought out organisations that provided examples of diverse hospitality practices. All the organisations have institutionalised structures and, except for the faith-based organisations, have taken part in internationally funded projects. By institutionalised structures, we mean the formal structures and rules of the association, including membership, in contrast with non-institutional ones, referring to the networks of activists and/or volunteers not directed by bureaucratic structures and procedures. Our findings thus mostly represent the views of civil society actors who work in institutionalised and more or less professionalised organisations as the most common form of civil society activism in Istanbul.
In institutionalised organisations that received international funding, a certain degree of professionalisation has emerged. This bureaucratising effect has influenced professional civil society workers rather than volunteers. In the end, 10 organisations were contacted, and 12 semi-structured interviews were conducted, 6 of them with women. The youngest of the interviewees was in his late 20s, and the oldest was in his 50s; most were in the age range of 35 to 40. All were university graduates in social sciences. More than half (8) of the interviewees were with project coordinators/ assistants who did not usually work face-to-face with refugees. Thus, our interviews were with people who played a significant part in pro-migrant civil society organisations, and their ideas helped us to develop a fuller understanding of the relevant issues.
The names of the organisations and interviewees were anonymised. The interviews lasted 40 to 60 minutes. In general, the interview questions addressed the character of the civil society organisations' activities and the main problems that they faced in their encounters with refugees and shifting national policies on immigration. The interviews provide the textual field for exploring the divergent and contingent intersections of the discourses that constitute the interviewees' ideas of hospitality practices. The common themes that surfaced repeatedly in the course of the interviews shape the analytical themes of this article. We specifically focus on the reflections referring to the themes of gender, faith and professionalisation. Most of the research participants spoke in the terminologies of the projects, using such terms as 'beneficiaries' to denote refugees.

Faith-based hospitality
The development of faith-based organisations in Turkey, their role in changing migration politics, the social policies of the state, and their proximity to the governing Justice and Development party have been extensively examined (Atalay, 2019;Göçmen, 2014). However, research on faith-based organisations' involvement in refugee protection has emerged only more recently (Danış and Nazlı, 2018;Tezel-McCarthy, 2017), and more research is needed to develop a robust contextual framework.
The director of a faith-based organisation, 50-year-old Ahmet, elaborates on his lengthy faith-motivated involvement in charity. On the organisation's political goals, he asserts: We create public awareness, put forward an argument, protest or make a press release about abuses of rights, fundamentally about the oppression, injustice and contradictions facing Muslim people in Turkey, in the region or in other parts of the world. . . . Apart from that, we try to provide aid, as much as possible, to destitute places that we can see and reach.
Underscoring a shared faith with Syrian refugees, Ahmet frames their problems in a larger political context, 'where a group of Muslims are struggling for independence at home from a leader whom they do not support anymore'. As our conversation centred on Syrian refugees, Ahmet expressed his disappointment in the rise of nationalist, xenophobic and racist attitudes. While (idealistically) detaching religious discourses from racism, Ahmet believes that public opinion in Turkey is divided between those who are nationalist and racist and those who are motivated solely by religion to help Syrians. He clarified: Turkish society has a nationalist, fascist cultural side. Let's be honest. But there is also this: part of the Turkish population looks after Syrians solely with Islamic motives. They share their property, their homes. . . . They take care of Syrian brothers and sisters (kardeşlerimiz) right from the beginning. And they do this to all others. For example, Afghans are coming to Erzurum. And in fact, they are not escaping a war and are here for economic reasons. Islamic circles also take care of them.
Conversely, hostility, according to Ahmet, is the result of nationalism and/or fascism. By referring to Syrians generally as kardeş ('brothers' and 'sisters'), Ahmet recalls the Islamic story of Muhajir and Ansar, which circulates widely within faith-based circles in different migratory contexts. Siddiqui explains that the Muhajir (the first group of people who converted to Islam under the threat of persecution by the Quraysh clan of Mecca, and who migrated to Medina with the Prophet Mohammed) and the Ansar (the local habitants of Medina who helped them) 'convey the strong bond which emerged through faith, migration and community' (Siddiqui, 2015: 53). This migration is closely associated with the origin of Islam.
Ahmet disassociates the policies of his faith-based organisation from competition for international funding. Explaining his organisation's disengagement from all funding processes, Ahmet thinks that applying for funding is most common among leftist circles. By stressing the importance of individual financial, material and emotional contributions, Ahmet evaluates his organisation's work as more 'meaningful and valuable'. This resonates with Siddiqui's description of Islamic hospitality as 'a divine imperative' that becomes 'an act of worship' with the potential for spiritual transformation (Siddiqui, 2015: 16-17), reflecting some faith-based organisations' scepticism towards European interventions. Furthermore, Ahmet talks about an 'amateur-spirited, non-institutional informal organisation of faith-based circles across Istanbul' who communicate through a WhatsApp group, supporting vulnerable displaced people. This informal organisation, according to Ahmet, is distinct from the institutionalised organisation he currently works in. It is in some ways preferable, as he believes 'institutionalisation and bureaucratisation kill the spirit of actions. . . . The essential is to give from your own pocket, your own budget. This is meaningful and valuable.' According to Ahmet, not all institutionalised organisations are funding dependent and aid based, but most of the time, they are relatively politicised. While connecting to the general understanding of hospitality as the giving of time and space to others (Dikeç et al., 2009), Ahmet identifies his faith and Muslim identity as the sole motivation for his hospitality work. In conversation with Ayşe, a journalist in her late 40s, we talked about the nature of activism among faithbased groups. Due to her engagements in organising pro-migrant faith-based activities, Ayşe differentiates between the Syrians who arrived early to Turkey in 2011 and those arriving now: Religious organisations provided for the needs of immigrants solely with a faith-based motivation and with the resources they already have on hand, without any expectations in return. However, today, all these support activities for immigrants are not being conducted solely with a faith-based motivation or only as a 'free hospitality'. Instead, it is more like a business or a professional activity since all their costs in supporting immigrants are paid by state or non-state funds.
By recognising the presence and effects of funding among faith-based organisations, Ayşe offers a different, less idealised narrative than Ahmet. This speaks to both the heterogeneity of existing faith-based organisations in Turkey (Turhan and Bahçecik, 2020) and the ways that they are represented and seen differently by local activists. 2 It is important to note that a faith-based dimension of hospitality does not emerge only in faith-based circles and discourses. References to faith in determining the power relationships of hospitality are also visible in secular organisations. For example, in describing their relationships with refugees, civil society workers often refer to the faith of the supported person and share instances of how religious beliefs affect the relationship. In non-faith-based organisations, the emergence of faith between refugees and civil society actors disrupts the much-used faith-based/non-faith-based dichotomy.
The problematic dichotomous approach to faith-and non-faith-based civil society organisations is most vividly manifested in the dilemma of whom to support (other faith groups, LGBTQ refugees, undocumented migrants, etc.) and in how each organisation identifies the nature (such as Islamic or rights-based) of its hospitality towards refugees. That is, the conditionalities of hospitality or hostility change across faith-based groups and between and within faith-based and non-faith-based groups (Mackreath and Sağnıç, 2017), reproducing the division between 'deserving' and 'undeserving' refugees (Dhaliwal and Forkert, 2015). The concept of deservingness, as Homes and Castañeda (2016: 13) argue, 'displaces responsibility from historical political and economic policies supported by powerful actors in Europe and the United States and instead locates it in displaced people'. In this process, the (un)deservingness of refugees, as shaped by gender, race, class and generation, is held in tension in the need to assert both vulnerability and potential future economic productivity.

Gendering hospitality practices
Hospitality is a gendered phenomenon. The gendered aspects of the hospitality practices of pro-migrant organisations -like those of any other association -can be examined empirically, methodologically and conceptually. Gendered concerns influence understandings of and/or assumptions about the guest and host and reinforce existing power relations (Darke and Gurney, 2000;Hamington, 2010;Stivens, 2018). For example, we found that gender plays a significant role in who is considered an (un)deserving refugee. Furthermore, research participants' accounts suggest that gendered stereotypes and expectations influence the recruitment process and to whom specific tasks are assigned and hospitality practices targeted. At times, such practices intersect with faith-based concerns about gender roles and relations. In doing so, gender and faith shape both the professionalisation of an organisation and the way that civil society practices are (un) professionally achieved.
Mahmut, a 30-year-old manager of one of the community centres of a large humanitarian aid organisation, gave examples of gendered practices of hospitality. The community centre is in a conservative neighbourhood where Syrian refugees live in large numbers. Mahmut explains: As much as possible, we prefer to have women employees. Of course, we have male employees as well, but it is very difficult to reach the male Syrian population. Also, it is mainly women [migrants] who visit the community centre and women [migrants] want to work with and talk to women [employees].
He further noted that, with courses for migrant women, the women ask if the instructor will be male or female before deciding to participate. Mahmut understands that women prefer other women as instructors because otherwise they would have to attend in their burka and veil, which he thinks they find uncomfortable. Thus, in this narrative of hospitality, both the host and guest as women are contextualised within the gendered mores of faith. Olivius (2014: 35) demonstrates how, in humanitarian aid, perceptions of vulnerability are calibrated based on gender, and provision of equal access to services for both genders is required. Thus, attention to gender is seen as legitimate because it constitutes one dimension of vulnerability as a criterion for provision of resources and assistance (Olivius, 2014: 35).
It was also apparent that such responsiveness to gender has resource implications. Mahmut recounts a case in which community centre workers drove migrant women to a social event, but the women did not want to be left 'alone' with the male driver. To fulfil the centre's goals for equitable service provision, a female social worker had to accompany them. Another example that Mahmut shares relates to the presence of a male translator when a migrant woman had to meet a lawyer regarding ongoing domestic abuse. Because of the woman's discomfort in talking to a male translator, the community centre arranged for a female translator.
Gendered concerns also influence and reflect existing stereotypes, imaginaries and expectations of the guest (and/or the host). Mahmut believes that women refugees are keener to participate in self-development programmes: Day by day, women participate in more and more activities. They want to learn new things for the sake of their children. They are aware that they are in a different country. For example, even conservative women seek a good education for their own children. They also recognise the importance of learning the new language to live in this country.
Mahmut's description of migrant women as agents bolsters his perception that refugee women, not refugee men, should be the primary focus of support activities. The interviews suggest that for these civil society groups, the usual 'target group' of civil society activities is women (and children). Such findings support Olivius's (2014: 71-2) conclusions regarding the instrumental use of gender in humanitarian aid. Olivius considers women 'strategic partners' (2014: 70) whose involvement brings greater success in achieving the goals of humanitarian programmes. Similarly, as articulated by Mahmut, women are seen as being more intent on succeeding in integration-related programmes than men. For Olivius, traditional gender roles and existing power relations are not adequately problematised by basic needs or instrumentalist approaches to gender issues. The former emphasise women as victims and vulnerable subjects, whereas the latter naturalise certain gender norms. Such criticisms are shared by scholars who emphasise women as more deserving refugees, as they are imagined as victims, whereas men are imagined as 'irregular' or 'economic' migrants (Freedman, 2015;Schrover and Moloney, 2013). Here, the promotion of gendered concerns, gender equality and women's empowerment are narrated as addressing the patriarchal hierarchies affecting the lives of many immigrant women. However, this emphasis on patriarchal structures in immigrant women's culture fails to consider the impact of patriarchal and gendered xenophobia in the host country and its effect on immigrant women (Farahani, 2018). The power relations that researchers and humanitarian and pro-migrant organisations themselves perpetuate through their activities and assumptions made about immigrant men and women are often also neglected (Lokot, 2019).
The funding agencies of the Global North demand professionalisation of pro-migrant organisations in demonstrating gender equality, particularly regarding human rights, economic inequalities, education and gender-based violence (Chaney, 2016). However, in confirming gendered norms, civil society hospitality practices assume that women are the primary caregivers of children (and men), whereas men are imagined as engaged in public employment. Such assumptions are bolstered by a heteronormative framing. Only two of the research participants, Zeynep, a project coordinator in a refugee organisation in her early 30s, and Ali, a mid-20s field worker, mentioned LGBTQ refugees. Ali said: When they are in our office, we do not see them either as men or women. We ask how to address them because we don't know about their sexual orientation. We need to establish trust. When we miss a detail, it is impossible to build trust.
Recognition of LGBTQ concerns is uncommon unless the support programme specifically targets LGBTQ refugees, which is itself rare.

Professionalised hospitality
An increase in civil society actors' involvement in providing support to displaced people has occurred in tandem with a growing professionalisation. In general, the professionalisation process, as Maloney and Deth (2010) discuss, has significant impacts on organisational structures and recruitment, the ability to mobilise and engage an active citizenry, funding opportunities, lobbying activities, and the capacity to influence policy-making and outcomes. In the context of the Global South (with all its differences), as Al-Ali (2003) rightly points out, funding (im)possibilities generated by international agendas create a competitive milieu among different organisations. Such rivalry can incentivise temporary projects that respond to available funding rather than the pressing needs of local communities.
Troubled by the increasing professionalisation of civil society organisations, Zeynep, a project coordinator, talked about the difficulties and contradictions that she faced daily in her work to meet the requirements of funding agencies. Nevertheless, Zeynep's critical perspective on professionalisation is not common among pro-migrant civil society organisations in Istanbul. Fatma, a communication specialist for a humanitarian aid organisation, sees professionalisation positively: This organisation was better known in the international arena than in Turkey [in the early years of Syrian migration]. It has credibility. As a result of this credibility, it received offers [from international partners] to cooperate in the implementation of certain policies.
Fatma interprets receiving international funding and working with international partners as increasing the credibility of her organisation through professionalisation. However, research also shows that access to international funds for refugee protection creates competition within civil society in Turkey (Al-Ali, 2003;Mackreath and Sağnıç, 2017). Zeynep, however, was more critical than Fatma of the consequences of international projects for the lives of refugees, especially in relation to the role of professional workers: 'I feel conscientious responsibility. I see my work in my dreams. However, there are people who do the same work as a nine-to-five job.' Civil society workers' differing perspectives and experiences of professionalisation influence how they describe and evaluate the relationships that they form with refugees. In most of the interviews, refugees are not referred to as a group with whom civil society actors can or should have close relationships, such as friendships. Most of them define refugees in the terminology of the projects, often as 'beneficiaries', which is common among professionalised humanitarian organisations (Rozakou, 2017). For example, Ali believes that a professional distance should be kept from refugees and that workers should not develop friendships with them: I cannot give my own phone number or my social media accounts, for example. This relationship must not turn into something personal. There is this risk that they may see us as heroes/heroines. This should not happen. It should be clear that I am there to help. I am not doing something impossible to do. And it is not personal help.
In contrast, Mehmet, who is in his late 20s and works as a social-cohesion expert in a large humanitarian aid organisation, is open to the idea of forming friendships with refugees. He refers to a 'Syrian friend' at work who used to be a 'beneficiary' but is now working as a translator in the organisation. Mehmet explains how refugees who have crossed the Mediterranean or have escaped Syria have affected him: These people are seen as victims, but actually they are like heroes/heroines of their own lives . . . . They survived. They hang on to a difficult life. When you are together with those people, when you work with them, provide them services as beneficiaries, when you do cultural activities together, you take part in their stories. You take a part in their destiny. Even if [the job is] difficult, your faith deepens because those people who look like victims are actually heroes/ heroines'. Such a self-reflexive understanding of the trust developing between a social worker and the 'beneficiary' is narrated as disrupting traditional working relationships between refugees and civil society workers. However, the inherent imbalance between global and local actors and displaced people exposes complex webs of power that frequently leave refugees in a vulnerable and dependent situation, that is, in a gendered positioning. Here, we are assisted in unpacking how unequal relationships can complicate (if not hinder) the achievement and maintenance of friendship bonds in gendered local and global relationships (e.g. Bajramović, 2018). Further, we would like to underline the importance of studies that have revealed how existing power imbalances involving humanitarian organisations and the communities that they serve create circumstances in which survival sex to fulfil essential needs becomes a daily reality in local economies (Westendorf, 2020).
In considering the professionalisation of hosting relationships and the trade-offs between psycho-social distance and closeness in such relationships, research participants also drew attention to the emotional demands of their work. Zeynep highlighted how field workers continuously bear witness to stories of violence, resulting in significant emotional labour in the field but less awareness among workers of the personal effects of secondary trauma. She states that only in the last few years have aid projects included budgets for initiatives to protect the well-being of field workers, including group therapy.
In the public sphere, there are different motives for supporting migrants and different relationships between migrants and civil society actors (Genç, 2017;Rozakou, 2017). Left-leaning network-activist organisations, such as the Migrant Solidarity Network (Göçmen Dayanışma Ağı) or the We Want to Live Together Initiative (Birlikte Yaşamak İstiyoruz İnisiyatifi), are critical of the consequences of global capitalism and nationalism. They generally refer to migrants as 'friends', not necessarily to denote an intimate relationship but to highlight attempts to establish a non-hierarchical relationship based on solidarity (Genç, 2017). However, such solidarities between migrants and civil society actors are not common in Istanbul. 3 The reasons for this may range from Turkey's socio-political context to its migration-related patterns, and a satisfying explanation of the scarcity of such approaches is a topic for further research.

Conclusion
In this article, we have examined how and in what ways the politically and morally charged ambivalences of hospitality practices are articulated and negotiated, focusing on intersecting gendered, professionalised and faith-based aspects of pro-migrant activities in different civil society organisations in Istanbul. In so doing, we have paid attention to the connections between a global and localised politics of hosting, moving beyond early, emergency welcoming practices (Ticktin, 2016). By contextualising the religious and geopolitical particularity of Turkey in the Global South and its shifting political position (as a gatekeeper of Europe), we have shown how the hospitality practices of civil societies -with varying organisational structures and goals -are conditioned by global as well as local discourses and demands.
In the context of unconditional hospitality, according to Derrida, there are conditional laws (political, juridical and moral, among others) that regulate the continuously negotiable relations between host and guest. In the case of faith-based pro-migrant groups, the refugee is welcomed through discourses of gendered religious obligation by those who share the same religion. Sensing the gaps left unattended by political actors, the public engagements of Muslim faith communities challenge the traditional secular and Christian approaches underlying civil society organisations and funding agencies from the Global North. This, of course, raises the question of whether such a religiously inflected hospitality extends to non-religious people or those from other faith backgrounds. In our sample, hospitality practices were mostly extended to those who share the faith of civil society actors, with the idealised image of a refugee mostly understood to be a vulnerable woman. Thus, gender and faith intersect to create a caricature of deservingness: a practising Muslim woman, socially conservative and open to policies of integration. Here, it is worth noting that the gendered image of deservingness extends to secular civil society organisations. Men, however, are not generally thought of as a deserving group and are rarely prioritised as subjects of care. Such gendered understandings of refugees also influence the priorities of international projects. Thus, in professionalised organisations, the 'beneficiaries' are mainly women and children. The professionalisation of promigrant civil society organisations, in addition to being characterised by employment of paid staff, creates not only a particular structure but also a specific language, i.e. terms such as 'beneficiary', and practices such as meeting deadlines and filling quotas for internationally funded projects. In the process, the conditions of hospitality are transformed, and the content and form of available support changes.
Civil society actors' intentions and attempts to be good citizens, good Muslims, and good care providers expose the intimate aspects of hospitality, segueing into discourses of the (gendered) deservingness of displaced subjects as new, would-be citizens. The personal, moral, politicised and professionalised confrontation of civil society actors with the globalised challenges of displacement demonstrates how the intimate and global blend and flow into each other (e.g. Wilson, 2012). By portraying how macro and micro, global and local, and public and private relations condition hospitality practices, we observe how globalisation is lived intimately, influencing perceptions of deservingness and the prioritisation of the needs of displaced people.
The presence of Muslim refugees and immigrants in the Global North has further heightened orientalist anxieties about Islam. Cultural and political contestations about migrancy and European identity are embedded in Eurocentric secularist narratives of modernity and Islam as an Other vis-à-vis European civilising narratives (Yeğenoğlu, 2012). Focusing on the particularities of South-South migration, an emerging question is whether Muslim solidarities minimise hostility towards (Muslim) refugees in Muslim countries. Studies on refugees' life conditions in the Global South (e.g. Toğral, 2016;Yarbakhsh, 2015) demonstrate that the cultural, religious and linguistic similarities between displaced peoples and host countries fail to guarantee a decline in institutional and everyday xenophobic practices. We suggest that research on the shifting energies between hospitality and hostility needs to pay attention to how the image of the hospitable host (both as an individual and as a nation) is constituted in the situated meeting points between public policies and attitudes and civil society funding and practices, as well as through empirical frames.