UvA-DARE (Digital Academic Repository) Turkish-Dutch mosque students negotiating identities and belonging in the Netherlands

: For Muslim communities in the West, teaching Islam to children in mosques is an important part of the intergenerational maintenance of young people’s heritage, identity, and communal ties. However, children with a migrant background beneﬁt from fostering identiﬁcation and feelings of belonging both to their heritage community and to the majority society. This study seeks to understand the negotiation of sense of belonging and self-identiﬁcation as Turkish, Dutch, and Muslim, in a sample of mosque students in The Netherlands (N = 29). It is based on interviews conducted with Turkish-Dutch students (ages 6–16) during ﬁeldwork in the mosques between March and December 2017. The study reveals that most respondents identify as Dutch-Turkish and have a stronger sense of belonging to The Netherlands than to Turkey. Nevertheless, there is also a group of mosque students who struggle with growing up between two cultures. Furthermore, the analyses of the mosque student’s perspectives show that they are negotiating a space for the possibility of identifying as Muslim and feeling a sense of belonging to The Netherlands without feeling Dutch. The study concludes by discussing a number of key issues arising from the ﬁndings, namely the role of the mosque context, the development of reactive identity, and the meaning attribution by the students.


Introduction
Mosque education followed by Muslim children (i.e., extracurricular supplementary Islamic education provided in mosques) is one of the most salient ways of socialization into the identity, value system, and practice of Islam. Instilling a sense of belonging to the greater supranational Islamic community (umma) and to the ethnic Islamic community of the mosque is seen as a religious duty Islamic associations (djama'a) have towards their youngsters. Previous studies on the pedagogical practices, curriculum content, and learning goals of Turkish mosque education in The Netherlands have shown that according to imams and mosque teachers, one of the primary objectives of providing mosque education is to strengthen children's Muslim and Turkish identities . The findings of Altinyelken and Sözeri (2019, p. 60) also illustrate that solidifying the Muslim and Turkish identities of the children was seen as equally important by their parents, as according to them: "only then they could build on some other identities". The reinforcement of Islamic identity and sense of belonging to the Muslim community is achieved in the mosque classes through participation in performative rituals in which the children join together with the rest of the congregation: like the communal prayer of namaz, or reciting the Qur'an and learning Arabic together (Berglund and Gent 2017). Such participatory learning interactions play a defining role in the development of Muslim children's sense of self and help to stimulate their bonding with co-ethnic peers and adults.
Strengthening Muslim children's feelings of religiosity, identification, and belonging to their heritage communities, however, is a politically contested matter in Western societies. The reasons for the contestation in public debates fall into two main discourses of fear. First, mosque education (which is not inspected by formal educational authorities) is often perceived as linked to religious radicalization (Gholami 2017;Cherti et al. 2011;. Second, there are pronounced concerns among European policymakers particularly about Diyanet mosques (the Turkish Presidency of Religious Affairs) teaching civic allegiance to the Turkish state which sponsors the mosque classes as a part of their diaspora governance policies (Baser and Féron 2021), and, hence, alienating young Muslims from the mainstream Western society where they live (Sözeri et al. 2022).
Against the backdrop of those political anxieties and discourses of fear, the current paper aims to shed light on the discourses on self-identification employed by the mosque students when negotiating their belonging both to The Netherlands and the Turkish-Islamic communities they come from. To that end, we address the following questions which are often discussed in relation to the integration of youngsters with a Muslim background. First, we explore the national self-identification of the mosque students: how do they identify themselves and which identity do they prioritize? Second, we investigate mosque students' sense of belonging to The Netherlands, their country of birth, Turkey, the country of origin of their parents or grandparents, and to Muslim communities in general.

Theoretical Underpinnings
Two interrelated concepts are central to the analyses in the study: a sense of belonging and identity. It has been shown that a lack of a sense of belonging to the group they are members of puts people at a higher risk of depression, loneliness, and emotional distress (Sargent et al. 2002). Young people with a migration background are a particularly vulnerable group as they are constantly faced with questions such as: where do they belong, where are they from, and who they are (Safi 2013). Those who fail to develop identification and a sense of belonging with regard to both their heritage culture and to mainstream society are the ones who are most marginalized and are at increased risk of radicalization (Lyons-Padilla et al. 2015). Published studies have pointed out that children with a migrant background benefit from fostering identification and feelings of belonging both to their heritage community and to the majority society (Morrow 2004). Some scholars have argued that Muslim faith schools play a vital role in helping Muslim children develop a secure sense of identity and belonging which are, in turn, associated with higher self-esteem (Parker-Jenkins 2002, cited in McCreery et al. 2007. Furthermore, in social identity research, the concept of belonging is closely linked with an individual's identification with a religious, ethnic, or national community. Some studies, for example, use minority children's national self-identification as a measure for their national belonging (Gharaei et al. 2018), and vice versa (Kabir 2008). Faas (2009, p. 304), who studies the identity negotiation processes of Turkish youngsters in Germany and England, conceptualizes social identities as "the communities young people feel they belong to". Others use the concepts of national belonging and national identity interchangeably (Verkuyten and Martinovic 2012). Verkuyten and Martinovic (2012) base their conceptualization on Tajfel and Turner's social identity perspective (Tajfel andTurner 1979, cited in Verkuyten andMartinovic 2012). According to this, "cultural identity involves a sense of belonging to one or more cultural groups and the feelings associated with group memberships" (Verkuyten and Martinovic 2012, p. 84). This implies that individuals may identify simultaneously with more than one national community. Verkuyten and Martinovic (2012, p. 87) describe this phenomenon as a dual identity: "a sense of national commitment and belonging without distancing oneself from one's ethnic group". With that line of reasoning, a religious identity might add additional layers to the dual identity turning it into a hybrid identity constituted from multiple belongings (i.e., encompassing also a sense of belonging to a community of Muslim believers). In other words, the students might be identifying with multiple "I-positions" of the self that are in a dialogical and hierarchical relationship with each other (i.e., some identities/I-positions being more dominant in the self than the others, depending on the context of interactions, as described by Hermans 2012, pp. 6-7).
Likewise, Östberg (2000), who studied the role of Islamic nurture and identity management in the lives of Pakistani children in Norway, found out that the Muslim children in her study expressed a "complexity of social belonging" to their family and relatives, to Norwegian classmates and teachers, to the Pakistani community, to other Muslims, to their neighborhood, and to Norwegian society. According to Östberg, the children did not have difficulty combining these diverse aspects of social belonging. On the contrary, she claims that the complexity of social belonging contributed to the development of multiple cultural competencies, strong reflexivity, and "integrated plural identity" by the children (2000, pp. 94-100). Östberg's concept of "integrated plural identity" is similar to Verkuyten and Martinovic's concept of dual identity: it implies that the children were able to integrate different feelings of belonging and their corresponding identities without losing a sense of self. Östberg argues that while formal schooling and contacts with non-Muslim peers socialized children into being Norwegian, mosque education played a role in bringing children's Pakistani and Islamic identities together.
Similar to Östberg (2000) and Verkuyten and Martinovic (2012), we base our understanding of identity on post-structuralist theories of identity according to which national identities are fluid and context-dependent (Erikson 1968). Correspondingly, we take the identities and sense of belonging of the Dutch-Turkish children attending a mosque education to be plural and negotiable rather than essentially ascribed. This means that children's self-identification might be characterized by a negotiation between being Muslim, Turkish, and/or Dutch, depending on the actors with whom they interact (e.g., imams, mosque teachers, school teachers, parents, peers), and the context and the messages from their social environment. Furthermore, informed by Rumbaut (2008) and Çelik (2015), we take into account that processes of perceived exclusion and discrimination by peers or significant adults in children's pedagogical contexts might trigger the formation of reactive ethnic and religious identities. In other words, that if confronted with cases in which their Turkish or Muslim identities are being devalued, they might defensively attempt to protect them by showing enhanced identification with the threatened parts of their selves. Lastly, we also consider the role of friendships formed at school or at the mosque in the identity negotiation of the children, as previous research has established that friendships might affect national identification and group belonging (Agirdag et al. 2011;Tolsma et al. 2012).

Materials and Methods
The data used in this study comes from a larger research project on the teaching practices of the mosques and their role in the integration of Turkish-Dutch children (Sözeri 2021). The focus on the Turkish community is not accidental: the Turkish-Dutch are the most numerous Islamic community with a migration background in The Netherlands (CBS 2022). In the current study, the voices of the children are central. The fieldwork was conducted by the first author between March and December 2017 at two mosques affiliated with the largest Turkish-Islamic organizations providing mosque education to the Turkish diaspora abroad: Diyanet and Milli Görüş. Whereas Diyanet is governed by the Turkish state, Milli Görüş is established and run by migrants . At the time of the fieldwork, Diyanet reported to the researchers that a total student population of 15,500 registered in its 147 mosques in The Netherlands, and Milli Görüş enrolled 5000 students in its 45 mosques. The Diyanet mosque was attended by approximately 320 students, and the Milli Görüş by about 300 students. The mosques were not randomly chosen by the researchers. Access to the mosques was granted by the respective chairs of the mosque associations, and they selected mosques with a large student population. It is worth remarking that negotiations for access took place at a time of heightened polarization and distrust within the Turkish diaspora due to the failed coup attempt by Gülenists (followers of the Islamic preacher in exile, Fethullah Gülen) in Turkey in July 2016, and social tensions due to a Dutch-Turkish diplomatic crisis in March 2017. It is likely that the positionality of the first author as a young female, native Turkish speaker, who was herself raised as a member of an indigenous Muslim minority in an Eastern European country (and, hence, not linked to political polarizations in Turkey and in The Netherlands), might have played a role in building trust with the chairs of the mosques, the children, and their parents.
The interviews with the students focused on their views about their national and religious identities. Additionally, questions about their sense of belonging asked whether the children see their future in The Netherlands, Turkey, or somewhere else. Also, to get a glimpse of the role of inter-ethnic peer friendships in children's social belongings, the children were asked whether they have more friends at the mosque or at school, and the names of their five best friends. At the start of the interviews, the children were offered paper and coloring markers and asked whether they would like to draw the mosque, their house, and their school during the interview. All participants, except the oldest two, were enthusiastic about drawing. Drawing served as a child-friendly ice-breaker which allowed the researcher to connect more easily with the children. At the end of each interview, the children were rewarded with small gifts (e.g., coloring pencils, a pen, or a notebook) as a token of appreciation for their participation.
In total, 29 semi-structured interviews were conducted. We spoke with 14 students at the Diyanet mosque, and 15 mosque students at the Milli Görüş mosque. The students participating in the interviews were a self-selected group, with ages between six and 16, chosen based on their expression of interest in talking to the first author. They were informed about the aims of the study, the confidentiality of their participation, and their right to withdraw at any point. Some children needed repeated confirmation that their answers will not be shared with their friends, parents, and mosque teachers. Parental consent for the students was obtained via active consent letters signed by their parents. All students in the sample were born in The Netherlands. Below, Tables 1 and 2 present the demographic characteristics of the students, and the occupation and country of birth of their parents.  The respondents were offered the choice to talk in Turkish or Dutch, and depending on their level of proficiency, the interviews were often conducted in a mixture of both languages. The interviews lasted for about 25 min on average. At the Diyanet mosque, one student did not want their voice to be recorded but allowed us to take notes instead. The audio recordings were verbatim transcribed and all transcriptions, including the notes from the interview, were coded for relevant concepts using the software program Atlas.ti version 8.4.23.0. Some examples of codes (derived from the theoretical underpinnings of the study) are "Islamic self-identification", "bi-cultural", "mono-cultural", "belonging", "intra-ethnic friendships", "inter-ethnic friendships", "othering", and "exclusion". The second author also coded half of the data and codes were compared with the first author as a measure of inter-coder reliability. Subsequently, the coded texts were thematically analyzed for emerging patterns of meaning (Braun and Clarke 2012). Quotations used in the article were translated into English by the first author.
Lastly, it needs to be noted that the positionality of the first author who conducted the interviews may have influenced the findings in a variety of ways. In the current study, it is possible to define the first author as an "in-betweener" of the field. That is, someone who, due to certain characteristics, could be perceived simultaneously as an insider and as an outsider of the studied population. For example, the children might have identified the researcher as an outsider to their tight-knit community as they have observed that she is not a strictly practicing Muslim herself (e.g., telling her that they knew she would not wear a headscarf outside of the mosque). Besides respectfully abstaining from communal prayers, the first author did not have any prior affiliation to any of the mosque associations either. These might have hindered establishing rapport and trust with some of the children. On the other hand, the children might have perceived the researcher as "one of them" and, hence, might have been more open about their experiences because of recognizing the researcher as a Turkish native speaker who has an understanding of what it means to grow up in a non-Muslim country due to her membership to a Turkish minority in another European country.

Results
To make better sense of the student's accounts, it might be useful to start with the observation that all children reported being previously asked questions about their ethnic and religious identity by peers, teachers, or strangers. This implies that Muslim children are sensitized to being different at a very young age, as early as six years old.

Identity or Identities?
Many children commented that they are not bothered by questions about who they are or where do they come from, but there were also those who found it annoying to be repeatedly questioned about their identity. For example, Selim said that he was so irritated that he bought a t-shirt with the Turkish flag on it and started wearing it at school in order to stop the questioning. Selim is an example of a child who identified himself only as Turkish and Muslim. There were nine other children who just like him identified themselves mono-culturally, only with their heritage identity (i.e., as Muslim and Turkish). Some of them equated being Dutch with being non-Muslim: Aylin said she cannot be Dutch because "the Dutch eat pork and believe in the Bible". Some of them remarked that they could see the possibility of being simultaneously Turkish and Dutch only if one of their parents would also be native Dutch. Others, like Mina, pointed out the importance of being accepted as a member of the mainstream group: "Officially, I am Dutch but I don't define myself as Dutch. Dutch people also don't see me as Dutch." She implied that if she would be perceived as Dutch by the majority society, that would influence how they see themselves as well.
On the other hand, nineteen children, or almost two-thirds of all student participants, identified themselves bi-culturally. These are the children with dual or integrated plural identities in Verkuyten and Martinovic (2012) and Östberg's (2000) terms. When asked about their reasoning in combining their heritage and Dutch identities, many said "because my parents are Turkish but I am born and raised here", "because I can speak both languages", or "because my parents were born in Turkey but I have a Dutch passport". For many, feeling both seemed to come naturally. Our conversation with Deniz exemplified the flexibility with which he switched between identities when talking with his friends at school: Deniz: Well, yes, sometimes I say "why are you against me, I'm just Dutch". They tease me by calling me "Turk", and then I say "I'm Dutch". And then they say "What are you now, last time you said you were Turkish, and now all of a sudden you say that you are Dutch". Then I say "Of course I am Dutch!" [emphasis added].
Researcher: Do you find these questions annoying?

Researcher: Do you understand them?
Deniz: Because simply this is who I am, indeed, a Turk and also a Dutch.
It was interesting that some children were very specific in detailing how Dutch or Turkish they feel without being prompted about that. For instance, while two students said that they are "half Dutch, half Turkish", three others felt that they were "more Turkish", and another one: "more Dutch". Ahmed even specified that he feels "80% Dutch and 20% Turkish". Being that specific about a perceived division between their identities might be indicative of the frequency with which the children have been in situations where they had to think and talk about their self-identification. From very early on, identity negotiation visà-vis other ethnic and religious minorities and the majority group becomes an internalized process that students with a Muslim background might enact without questioning much of the reason behind the identity negotiation itself.
Also, it was evident that some children struggled with the fact that they also identify themselves as Dutch. After defining himself bi-culturally, Yasin apologetically added: "it is not a bad thing to be both". Another child, Azra, commented that she feels like she would offend her family if she says that she is Dutch: "as if I don't belong together with them". Yet another child, Sinan, remarked that he gets angry when he is faced with racist remarks and, in such cases, he feels more Turkish than Dutch. He remembered the time when Dutch extreme-right politician Geert Wilders came to their town. He seemed to become defensive of his Turkishness when perceiving a threat to it. Besides illustrating the context-dependent nature of children's choice of identification, this also shows how reactive nationalism can be enacted when the students are faced with perceived exclusion and racism.
Regardless of whether they identified mono-or bi-culturally, many children agreed that what matters most to them is being Muslim, because "God is not going to ask you whether you were Dutch or Turkish, but what you believed in" (Murat), and "then you can go to paradise" (Rasim). Some children found it easier to talk about personal issues that are bothering them with their mosque teachers rather than with their school teachers, and when asked why they would explain it with their Turkishness. In other words, for the mosque students, a shared Turkish and Muslim identity with the teacher implied being understood by them better than by their school teachers. This is an example of the intertwined nature of identity, trust, and cultural competencies when it comes to bonding between mosque teachers and students.

Sense of Belonging
We tried to get an insight into children's sense of belonging by asking them where they would like to live in the future and why (for a discussion of the role played by people's ideas about their future and their sense of belonging in the present, see (May 2019)). Almost half of the children, or 14 out of 29, replied that they would like to live in The Netherlands. Ten children said they would rather relocate to live in Turkey, while three children said that they cannot decide or have no preference. Interestingly, two students expressed their wish to move to Los Angeles, USA, "because rich people live there and it has sea" (Ahmed). More than half of the children who expressed their desire to move to Turkey justified this by saying that they miss their relatives and family members who still live in Turkey. Two children said they would relocate because of the sea and sun, and one thought he could have a better football career in Turkey rather than in The Netherlands (Sinan). Among the ten students who saw their future in Turkey, only three had identified themselves as Turkish only. In other words, seven of the children who did not identify themselves as Dutch would rather see their future in The Netherlands. Although preference for a country to live in is only a partial indication of a sense of belonging, this finding still hints that identity and belonging are not conceptually interchangeable as often seen in the literature (Verkuyten and Martinovic 2012). That is to say, it points to the possibility of feeling a sense of belonging to The Netherlands without feeling Dutch and the other way around.
The children who expressed more attachment to The Netherlands provided various reasons for their choice. Along with the statement that The Netherlands is a nice country, an argument that was often repeated was that The Netherlands has better schools and a better education system allowing for more social mobility, better infrastructure, and more job opportunities than Turkey. Many children said they would like to live in The Netherlands because they speak better Dutch than Turkish. Some expressed that it is a safer country with lower criminality (Mina, Selim), and that mattered for them. According to this group of children, "Turkey is for old age and retirement" (Kenan, Murat), or "only for holidays" (Salih, Umut).
When it comes to inter-ethnic friendships, ten students out of 29 said that they have more friends at the mosque, twelve had more friends at school, and seven students claimed that their friends are equally distributed between the school and the mosque. We also asked the students to name their five best friends and discussed their ethnic backgrounds. Regardless of where they met their friends, all children had more co-ethnic friendships than inter-ethnic ones. To illustrate, the pool of best friends resulted in 91 Turkish and 39 non-Turkish friends in total. When interpreting these results, it is important to take into account the observation by the parents (based on fieldwork data collected for the larger research project) that most children already have limited contact with native Dutch peers due to the high concentration of children with migrant background in their neighborhoods and schools. Table 3 sums up the distribution of student responses per conceptual category. Since all of the students identify as Muslim, Muslim identification is not included in the table. Table 3. Overview of student responses per conceptual category.

Number of Students Identity
Bi-cultural (e.g., Turkish and Dutch) 19 Mono-cultural (e.g., Turkish only) 10 Belonging Would rather live in The Netherlands 14 Would rather live in Turkey 10 Does not know yet/cannot decide 3 Would rather live elsewhere 2 Friendships More friends at school 12 More friends at mosque 10 Equal number of friends at mosque and school 7 Note: There were two students with Afghan backgrounds in the sample. One of them identified herself as bi-cultural, and the other one as mono-cultural (Afghan). Both said they would rather live in The Netherlands. Those were also the only two interviews conducted completely in Dutch (with no Turkish words used).
Children's perspectives about the importance of background in choosing friends differed from each other. Some argued that it does not matter whether their friends are Muslim because "after all, we all speak the same language (Dutch)" (Salih) and "you have to be friends with whomever you want to be friends with regardless of whether they are Muslim, Turkish or Dutch" (Murat), but "what matters is whether they lie to you or not" (Elif). Contrastingly, Emine said she rather plays with Turkish children because Dutch children do not want to play with her and do not understand her well. Mina expressed a different problem that she encountered when she wanted to make friends with Dutch children at school: "I know many children whose parents do not allow them to play with children who are not Dutch". Again, this remark highlighted the important role of perceived acceptance by members of the mainstream society when it comes to the development of feelings of belonging.

Discussion
This study aimed at delving into the perspectives of Turkish-Dutch mosque students regarding their self-identification and their negotiation of a sense of belonging. The findings show that two-thirds of the children in this study identified bi-culturally as Turkish-Dutch, while one-third identified monoculturally as only Turkish. Also, the majority of the children envisaged their future in The Netherlands rather than in Turkey and had more friendships in the culturally mixed and secular educational space of the school than in the culturally homogenous and religious space of the mosque. Nevertheless, their strongest identification and primary sense of belonging seemed to be with the Muslim community.
Indeed, the children in this study prioritized their Muslim identity over their national belonging regardless of whether they self-identified as bi-cultural or as monoculturally Turkish. The possibility to identify monoculturally as Dutch was also present, it was left open to the children to self-identify, but there were no children who identified as Dutch only, while all children identified as Muslim. This resonates with the findings of previous ethnographic studies on British Muslim youth that observe a trend in the identification of second and third generation as primarily Muslim, leaving ethnic and national identities as secondary (Modood et al. 1994;Lewis 2002;Field 2011, cited in Scourfield et al. 2013. Nevertheless, the majority of the children in this study expressed stronger belonging to The Netherlands than to Turkey and seemed to be successful in combining their different cultural and religious identities without experiencing distress. This implies that the students distinguish between religious belonging and belonging to a country/homeland, and the one does not necessarily imply the other. In the light of previous research showing that national identification with the majority is lowest among young Muslims , this is an important finding, especially considering that the sample of children in this study is likely to have much higher levels of Islamic religiosity than their Turkish-Dutch peers who do not attend mosque education. Future research investigating the role of Islam in the lives of Turkish-Dutch children and Muslim children with other ethnic backgrounds who do not attend mosque classes might shed more light on this. The findings, however, also indicate that there is a group of mosque students who do struggle with growing up between two cultures. This is also reflected in the results about inter-ethnic friendships: although most children report that background does not matter for them when making friends, the great majority of them are friends with co-ethnics regardless of whether they meet them in the mosque or at school. The preference for coethnic friendships should be interpreted cautiously as residential and school segregation are significant challenges for contact with native Dutch peers. Moreover, in large Dutch cities like Rotterdam, Amsterdam, and the Hague, the Turkish communities are characterized by levels of segregation that are higher than those of other ethnic minorities (Koopmans 2010). That being said, the findings imply that student experiences of exclusion in attempts to make Dutch friends, Islam-related discrimination, and bullying at school, are seen as significant obstacles to children's sense of belonging and identification as Dutch.
There are a number of key issues that come to the fore from this study that need further attention. These are, namely, (i) the role played by the space where the interviews were conducted: the mosque classroom, (ii) the development of reactive religious and ethnic identification, and (iii) the meaning given by the children to their identities and belonging.

The Context of the Mosque Classroom
The theoretical premise that the sense of belonging and the sense of identity of the students are context-dependent and fluid presupposes that they might identify differently in different spaces (i.e., Östberg 2000;Verkuyten and Martinovic 2012;Hermans 2012). It also implies that their identity negotiation might change according to the actors with whom they interact. The interviews in this study have been conducted one-to-one in mosque classrooms. In other words, the students have been made to think about their sense of belonging and identification while they were situated in a space they associate with learning to be a good Muslim and with the Turkish-Islamic community. It is likely that they might have given (slightly) different answers if the study would have taken place within their school classroom and if the questions would have been asked by a native Dutch researcher, instead of a Turkish one. The scope of this consideration is beyond noticing a limitation of the study. Rather, it pertains to the essence of the dynamics of identity negotiation and belonging. To illustrate, many of the children have not requested themselves to be registered to follow mosque classes, but the initiative has been taken by their parents or grandparents as a way to ensure immersion into religious knowledge and intergenerational identity maintenance (Sözeri and Altinyelken 2019). Hence, for many children, mosque classes are seen as an extension of their family and home culture, to the extent that being a good Muslim also becomes equated with being a good son or daughter. This is what we learn from the comments of those students who perceive identifying as Dutch as a betrayal to their family. Alongside this, these responses also tell us that, although this is not the case for all students, some children seem to have internalized the mainstream societal perception that identifying as Dutch is mutually exclusive with identifying as Turkish and Muslim. We do know from a recent study by de Jong and Duyvendak (2022) that some young Turkish-Dutch Muslims find ways to develop coping strategies against the mainstream nativist discourse that help them to reclaim their belonging to The Netherlands.

The Development of Reactive Identification
That being said, the findings of this study provide us with a glimpse into the early development of reactive ethnic and religious identification. The great majority of the children in this study are primary school students, and they already have developed an awareness of being different, due to experiences with othering in a variety of situations. Their exposure to the discourses of fear circulated in the public space by far-right politicians and the media seems to contribute to the development of a reactive identification with their Turkish and Islamic identities. In other words, in line with the findings of extant studies (Çelik 2015;Lüders et al. 2016) minoritized children, just like adults, are prone to identify more strongly with the identities they feel are being threatened. As a result, they might engage in demonstrative enactment of reactive nationalism (e.g., wearing a t-shirt with the Turkish flag at school) as a defense mechanism, in an attempt to appropriate and protect parts of their identities and belonging that are important to them. Moreover, to protect themselves from identity threats they might also engage in coping strategies such as academic disengagement at school that might be detrimental to their educational success (Verkuyten et al. 2019). We already know that school teachers have an important role in decreasing students' feelings of exclusion and identity threat (Bergamaschi et al. 2022). Future research might explore further whether and how pedagogic contexts such as mosque classrooms can also play a protective role in dealing with students' perceptions of identity threat.
It is important to note that this phenomenon does not occur without the active involvement of the significant adults in the children's pedagogic environment. During the fieldwork, parents and mosque teachers alike have repeatedly expressed that one of the goals of providing mosque education to the Muslim children is to equip them with the necessary knowledge about Islam so that the children would be able to defend themselves against Islam-related bullying and intolerant remarks addressing Muslims. This is a characteristic of the provision of mosque education that has been born out of the peculiar circumstances of raising Muslims in a socio-political climate that is perceived as hostile to Islamic education and Muslim communities .

The Meanings Given by the Children to Their Identities and Belonging
Despite the limited time-space of this fieldwork, it is possible to see that the vocabulary of the mainstream discourses around belonging and identities is reflected in the meaning-making processes of the children. That is to say, markers such as linguistic proficiency, being born in the country, bureaucratic ties such as a passport, having a social network in the country (e.g., friends and family), their Islamic heritage, and religiously or culturally accepted norms of behavior (such as eating halal or not) all enter into the identity negotiation made by the mosque students. Furthermore, the findings tentatively suggest that the absence of identification as Dutch is not necessarily associated with the lack of a sense of belonging to The Netherlands or a lack of future-oriented perspective towards the country, for example. In those cases, the mosque students engage in the rationalization of their choices by referring to better educational opportunities, better chances for career advancement, and overall better financial outlooks. Yet, in other cases, children's sense of belonging and identification is shaped by less rational and more affective reasoning: such as feelings of being better understood by teachers who share a Muslim and Turkish identity with them.
All in all, the comprehensive and well-thought reasoning of the children suggests that the processes of meaning attribution precede the fieldwork and are likely to have been shaped in reiteration to the different pedagogic contexts they navigate. The mosque classrooms are only one of those formative contexts, but one in which the children learn to attribute more meaning to being a Muslim rather than Turkish and Dutch. It yet remains to be explored how their identification and belonging changes through time as they pass through different stages in their lifepaths, and whether and how those processes differ from young people with a migrant background who have not been exposed to the same Islamic nurture and religious education.