Language in transnational education trajectories between the Soviet Union, Israel and Germany: Participatory research with children

The paper focuses on language in transnational education and puts children’s perspectives in the spot-light. In light of increasing transnational mobility, their voices are of particular significance: How do transnational children – children with migration experience – perceive the role of languages in educational trajectories? In order to answer these questions, a qualitative study was conducted with children of Soviet immigrants who were socialized in a Hebrew-speaking education system and who are today pupils in Germany. The findings from group conversation and language portraits allow deep insights into children’s perspectives on multilingual practices and highlight the importance of the environment – in this case, a German school that became part of transnational education by offering opportunities for students with migration experience.


Introduction
Multilingualism is a social reality, impacted by a long history of transnational mobility including children in the midst of their education.Thus, while children's transnational education is likely affected by language considerations that arise from being transnationally mobile, we do not know how children view this situation.This is because, first, language has not been discussed extensively in the research on transnational education, and, second, research is often conducted on or for children but not with them.This paper focuses on language in transnational education and puts children's perspectives in the spotlight, particularly when considering the following questions: How do transnational childrenchildren with migration experience -perceive the role of languages in educational trajectories?How does their perception as a multilingual speaker develop in this process?To answer these questions, I will first present a literature review that links research on multilingualism in education to transnational studies (section 2).I will then turn to transnational children and argue why their experiences with language use must gain more attention (section 3).The main part of the paper will present a qualitative study (section 4) framed by a specific institutional context -a German school that became part of transnational education by adjusting to students with migration experience (4.1).To access children's perspectives, data were collected through language portraits and group conversation with children of Soviet immigrants who were socialized in a Hebrew-speaking education system and who are today pupils in Germany.In the methodological section, I will discuss the opportunities and limitations of this method (4.2).The qualitative data allow for deep insights into how transnational children perceive language use and highlight the importance of the environment, as will be shown in the findings (section 5).The paper will conclude with implications for further research and practices in transnational education (section 6).

Language as an issue of transnational education research
This section will show why language constitutes a crucial issue in educational research and why this perspective is important for the field of transnational education.
Transnational mobility has an impact on language practices: multilingualism, while already present, continues to increase and diversify within and across national borders.Yet, the exact number of children who grow up multilingual is often not documented.For Germany, indirect data can be found in statistics on the so-called migration background.The microcensus, for example, collects data on children "with an immigrant background" who attend day-care facilities and "speak predominantly not German at home" (Report 2016, p. 40).According to these surveys, in western Germany, they constitute up to 50 percent, with an increasing tendency (ibid.).These figures neither reflect the language constellations (languages of parents, siblings, common family languages) nor the actual language practices of the children.A more differentiated picture exists for the federal state of Hamburg: In 2015, 42.8 percent of the four-and-a-half-year-old children in the families spoke languages other than or next to German (Lengyel 2017, p. 159).
At the same time, in most countries, including Germany, the education system itself is nationally oriented, meaning it is also monolingually oriented: (1) Language -mostly the nation state's language -is required for accessing school, as an institution of education.Initial school examinations, for example, which assess the children's level of development and thus their preparedness for school, are based on a (monolingual) norm.For children of preschool age who do not comply with this norm in the context of their family socialization, this often leads to setbacks (Voet Cornelli et al. 2013).(2) Language is also the criterion of selection and source of segregation.For example, teachers' recommendations for secondary schools are often based on an idea of a linguistic norm and not on the actual performance of the child, which results in selection/segregation of children, including the transfer of minority children to schools for children with special needs (Gomolla/Radtke 2009) p. 282).For pupils who are new to the education system, the transition from special language classes (German as second language) to regular education is inconsistent and can take up to two years (Dewitz et al. 2018).(3) Language is also the medium for how knowledge is transmitted and evaluated.Certain linguistic norms, such as the academic register, are implied but not explicitly communicated or taught (Morek/Heller 2012).This monolingual orientation has been confirmed for educational setting across countries (Fürstenau 2016 for Germany, Thoma 2018 for Austria; Schnitzer 2015 for Swiss; Pulinx/van Avermaet 2015 for Belgium, Young 2017 for France).For children who are not prepared to comply with the required norm, this may influence their participation in learning processes and affect their access to education (Gogolin 2010).
Consequently, this approach causes migration-related multilingual students to underperform at school; but, this is not the only effect.Empirical studies show that these monolingual practices at school can also shape one's self-perception as a multilingual person (Putjata 2017).As reconstructed in biographical research, individuals remember that being banned from using their family language at school represented a negative experience.As a reaction to this experience, they consciously start to orient themselves towards the legitimate state language in all domains of life, starting from what they read for leisure up to who they choose as friends and partners.Finally, they develop a negative perception of multilingualism and report that, today, they are reluctant to speak the immigrant language (Putjata 2017).Furthermore, a study on linguistic identity has shown that language management at school, as the place of socialization, shapes children's community imagining and their self-perception as legitimate member of it (Putjata 2018).An important part of this process is played by the indexical function of language -namely, the significance that is assigned to different languages by educational policy measures: In a context in which only the state language is accepted, interviewees develop the image of a monolingual community and perceive their own multilingualism as a deviation from that norm.The state language has been assigned an integrative function; it is as important for participation in society as it is in education.Therefore, for multilingual individuals, integration in this community means to discard all obstacles, including immigrant languages (Putjata 2018).
This brief research review shows that language is more than just means of communication and knowledge transmission: In the perspective of educational research, language is crucial for participating in learning and educational process and, thus, in society at large.At the same time, language management in educational settings shapes children's language attitudes, language development and their self-perception as an equal member of the community.This is of extreme importance, particularly considering that educational institutions should not only transmit knowledge but also support the children's individual development (Fürstenau 2017).Following the presented empirical evidence, researchers across the world and across disciplines have started to call for a paradigm shift -a multilingual turn.This turn involves accepting transnational mobility and questioning the monolingual norm in education (Meier 2017).

Transnational children and their perspective
So far, the issue of a 'multilingual turn' in education, as presented in section 1, has been studied with focus on institutions.In this section, I will argue why individual children's perspectives, as adopted in this paper, are important, and I will present what research from this perspective has been conducted so far.
A multilingual turn in education has been, for many years, based on normativetheoretical discourse with focus on sociopolitical and psycholinguistic arguments.These arguments underline how important it is to include immigrant languages in regular education: This would, on the one hand, allow for a cognitive transfer in learning processes and, on the other hand, help to overcome the deficit-oriented perspective on immigrant languages (Cummins 2010).In recent years, these arguments have been supported by empirical research that confirms the positive outcomes from the use of multilingual methods in classroom (Wei 2018).On the institutional level, several research projects have focused on the implementation of immigrant languages as part of general school development, for example in Germany (Fürstenau 2016) and the Netherlands (Duarte 2016).These studies show that schools can become 'transnational educational spaces' when they offer opportunities for children and families with migration experience.
As presented here, the issue on multilingualism in education is often researched from the perspective of institutions and/or their agents.Only rarely do individuals themselvestransnational children -get their chance to speak.
The field of linguistic studies, on the other hand, often focuses on multilingual children.However, in second language research, this focus is often on children's competences (reading, writing etc.) as compared to 'native speakers'; thus, this arguably results in the enumeration of deviations from the monolingual norm, as, stated recently: "The majority of children with a migration background suffer from language deficits in one or both languages" (Eisenwort et al. 2018, p. 99).These studies reveal the underlying assumption that one person equals one language, and they promote, in research and public discourse, a deficit-oriented perspective on migration-related multilingualism.This deficit perspective is highlighted, for example, in the still-prevailing myth on 'semilingualism' -'half-knowledge' of the second language coupled with a 'half-knowledge' of the first language (Tracy 2008).
The first shift in the research discourse has emerged over the last 20 years with the advent of 'perceptual linguistics', which looks at language attitudes and ideologies (Roth et al. 2018;Goltsev 2019;Olfert 2019).Its aim is to systematically describe multilingual people's perceptions of different languages and how they position themselves in relation to them.Research on language attitudes often works with quantitative surveys that ask individuals to name or to choose the language they use in particular situations -with family or friends, while reading or watching TV, etc.Thus, these questionnaires imply that the individual must choose between the languages.Furthermore, in migration studies these questionnaires are used to measure one's integration in a particular society (Prashizky/ Remennick 2015), arguably reproducing the still-prevailing assumption that one nation or one community equals one language.Open questions in a biographical approach can help to loosen this link, as has been shown in studies on experience with multilingualism in lifespan (König 2018;Putjata 2018).Yet, these studies focus on adults and rarely include children's perspective.Following the pedagogical perspective, not only adult speakers but also children and adolescents should be included in research to allow them to speak for themselves based on their experiences and interests (Plewnia/Rothe 2011).With regard to childhood research, Heinzel (2012, p. 23) underlines: "The question of how children can have their say in research situations and whether their views are sufficiently understood by adult researchers are central problems of qualitative childhood research, which endeavours to conduct research 'from the perspective of children' ".Yet, childhood research rarely focuses on language, and when so, mostly on second language acquisition (Gogolin/Eckhardt 2013) or indirectly, when researchers describe children's behavior, e.g.family language rejection, in ethnographic research (Panagiotopoulou 2017, p. 262).
Thus, drawing on the research lacuna presented in sections 1 and 2, I conducted an exploratory study that focuses on language in transnational education from the perspectives of children.This study draws on data from an emerging research project "LiMmud: Listening to the voices of children on multilingualism and transnational education" and will be presented in the following sections.

Study: Children's perspective on language use in transnational education
Regarding the research frame, the relationship between language and education described so far has as its reference framework the nation state.In the next section, I will turn to a transnational perspective and provide a theoretical framework by building on Bourdieu's model of school as a linguistic market (Bourdieu 1990), and Blommaert's order of indexicality and 'spatial turn' (Blommaert 2010).

Theoretical and contextual background of the study
According to Bourdieu (1990), language constitutes a particular form of capital since languages have different values within the social space, and these values correspond to the prevailing hierarchy of societal power relations.A certain way of speaking, i.e. state language or minority language, reflects a person's origin and their social background.In everyday conversation, speakers of a minority language can subconsciously be perceived as less skilled, while the majority language functions as cultural capital.These hierarchical differences in language status are socio-politically constructed, and school plays a predominant role in this process.Here, knowledge of the legitimate language is rewarded, while deviations -i.e. the use of minority languages -are ignored or even sanctioned.Further, the marks given in the school certificate reflect how skilled the student is with the 'right' cultural capital.These credentials constitute a decisive criterion for students' educational careers, with linguistic capital becoming a determinant for success.Thus, school reproduces the prevailing ideas about the legitimacy of certain language practices within the existing hierarchy of society (Bourdieu 1990, p. 20).On the individual level, this is where speakers learn the value of their language.Through teachers' feedback, the individual realizes which language practices are considered legitimate and positions him-or herself on the linguistic market of the school (Putjata 2017).
Regarding transnationality, the mechanisms of linguistic exchange at the 'market' of school, as described so far, put the nation state at the center.If, however, the perspective is extended and transcends the national frame, the way in which linguistic practices are viewed can arise in a new light, since the value of a language resource is never absolute but can only be determined in relation to its environment.The 'spatial turn' of the 20th century in sociolinguistics has worked to expand the concept of multilingualism not as something that individuals possess or do not possess but as something that develops its value in relation to space and only through interaction.Consequently, multilingualism can appear on a continuum between 'valued asset' or 'dismissed as having no language' (Blommaert 2005, p. 197) depending on the context.The present study draws on a specific context: The school Aleph 1 -a German school that has become part of transnational education.
Aleph is situated in a big city and was founded in the 18th century with the aim of giving children from socially deprived families access to education.From the very beginning, it was open for children of all confessions, but particularly for Jews.This fact may be the reason why it became a place of transnational education: Historically motivated, Jews have been subject to transnational movements over centuries and increasingly in the last decades.Today, about 45 percent of the school's pupils are from families with an immigrant, mostly Soviet, background.The school has reacted to this shift in the student body by adapting to the linguistic needs of transnational children: Besides Hebrew and English as foreign languages, children can also choose Russian in addition to or instead of further foreign languages like French or Spanish.Furthermore, the language curriculum was extended to include a special language course -"German as a second language".Thus, the school became part of the transnational trajectories of families that cannot be classified as the international elite; hence, the school does not offer transnational education at a place 'from above' but it offers this education to those seeking social mobility.With respect to transnational children, this theoretical and contextual framework opens up new questions: How do transnational children -children with multiple migration experiences -perceive the role of languages in this place of transnational education?How do their perceptions as a multilingual speakers develop in this context?

Methodology: Research with children and by children
To answer these questions, I conducted an exploratory study with transnational children who attend the school today.Keeping in mind the importance of children's own perspectives on their experience, the study was designed in a manner that would allow for actively including them in the research process.Thus, the data were collected using language portraits and group conversation. 2These different instruments were used complementarily to allow for a complex picture of the situation.From the very beginning, children were informed about the objective of the research and were asked to investigate languages, how and where they are used.A researcher diploma 3 confirmed this joint research: children were given researcher diplomas to show that they were considered equal partners in the research effort (see Figure ).
The first instrument was language portraits, a technique used quite frequently nowadays (Busch 2018).Children were asked to visualize their linguistic repertoire using the outline of a body silhouette (see Figure ).This tool opened a conversation and invited children to investigate their own multilingual practices and the practices they perceive in their environment, at school, with friends and at home, such as, for example, in statements like: "I had two languages at home, Russian and Hebrew, that's why I got them on the legs, because of the house, on the one side like this and on the other like that.I've got English in the ears, because I understand English better when I can hear it.I learned English through hearing, and German as well.I learned German mostly at school and it is also the language I feel most secure.I also write the most in the language, that's why I've got it at the hand.I would like to learn Italian, that's why I've got it in the heart."The portraits served as a starting point and were followed by a group conversation with siblings and friends from the same school.This method of data collection was chosen as it allows accessing the subjective perceptions of the children: On the one hand, group conversations can help minimizing the involvement of the researcher and, on the other hand, the conversation within a familiar peer group can have a motivating effect on participation and the well-being of the children (Heinzel 2012).The questions aimed to elicit narratives on language biographies and experience with language use.After the story was told and only if no statements were made on language, further questions were asked going back to particular moments and focusing on what the interviewee remembered on the role of the language.This method was chosen because by reconstructing their experience with the role of the language, interviewees not only reveal their subjective theories on language acquisition or language use but also reflect on their perceived role in society, with certain Lucas languages becoming an asset or a disadvantage.In doing so, they make use of the prevailing linguistic hierarchy, where social positions are allocated to specific groups of speakers.Hence, they always position themselves and others (König 2018).It is exactly this positioning, the construction of roles for speakers of different languages in a society and the expression of language perception as negotiated in the examined transnational education setting, that is in focus of the study.
Data were collected from one group conversation 5 with three children: two friends from the same school aged 12 and one older sister aged 19.The interview took place at home and lasted 62 minutes, allowing the participants to tell their story without being interrupted.The languages used were Russian, German, Hebrew as well as greetings in different languages that the children brought up in their language portraits.Switching between the languages was consciously exemplified by the interviewer, so that the children would feel free to present their multilingual practices.
For the analysis, all data were transcribed and coded by four researchers 6 using the annotation tool MAXQDA.The analysis followed a deductive-inductive approach (Mayring 2010).The first step was theoretical coding.Based on Bourdieu's theory, considerations about language and education are expected to appear in topics such as language usefulness for life in general, for one's school career and in the discussion of language legitimacy in interaction.Hence, the code 'language as capital' was used to embrace all the statements a participant made on certain languages, their usefulness and their value.The code 'language use' contained mentions of language use in teacher and peer interactions or statements about linguistic norms in educational contexts.The third code, 'self-perception', focused on statements about the participants' status as multilingual speakers at school and in the society.The second step involved reconstructing the argumentation as well as considering whether language practices were mentioned when they were not asked about at first; this step reveals the perceived importance of language in a particular situation.After coding all the interviews, comparison strategies were applied to find patterns between the interviews (Nohl 2010).This step allowed us to find similarities and differences such as in: "[Sandra:] I also translated what she said in Russian for him.[...]: I think that helped both of us a lot." as compared to "[Jasmina:] I had a friend who spoke Russian [...].Because I had a lot of contact with him, I spoke less German, which was not so good."Comparing these two statements allowed us to see how some children differently perceived the use of different languages; for example, in a situation of translation, using a different language may be seen as positive (helpful) by some children, whereas others criticized it.After finding initial similarities and differences, the group conversations were analysed again, this time focusing on data that would confirm or disprove the patterns found.If the coding differed between the four researchers, this was resolved in a joint discussion.

Results
Lucas, Sandra and Jasmina are all transnational children -children with migration experience.Lucas and Jasmina are siblings, born in Israel.Their parents moved from the Soviet Union to Israel in the early 1990s, after the collapse of Soviet Union.In 2009, they moved to Germany.At this point, Lucas was 3 and his sister 10.Today, their father lives in the Netherlands, their mother in Germany, one of the grandmothers in Israel and the other grandmother in Russia; they have further relatives in Canada and the US.Sandra is a friend of Lucas.She attends the same class at Aleph (their school; see context background in 3.1).She was also born in Israel, as a child of Soviet migrants.In 2014, when Sandra was 8, the family moved to Germany.Before this, Sandra's family considered Canada as a possible place to move because they had extended family networks there, but they rejected it because Canada was too far from close family members.When they decided to move to Germany, Sandra's parents consciously chose a school that would offer Hebrew as foreign language "So that she [Sandra] could be good in something right from the beginning." 7At the point of data collection, Sandra was 12 years old and had lived in Germany for 4 years; Lucas was also 12 and had lived in Germany for 9 years, as had his sister Jasmina, aged 19.
The findings reveal that in this educational context, children (1) perceive the significance of multilingual resources in (German) language learning process; (2) differentiate between languages as linked to different contexts rather than judging their importance or value; (3) see multilingual practices as a challenge for non-experienced teachers and develop a perception of multilingualism as indispensable in everyday communication with friends and family.In the following, I will elaborate on these results and directly quote the most concise examples from the language portraits and the group conversation.The qualitative data allow me to further extend the analysis beyond the quoted examples.

Significance of multilingual resources while acquiring German
The first finding is that children perceived their multilingual resources as helpful as they access the German education system.Sandra started learning German already in Israel, and, at the same time, her family consciously chose a school that would offer Hebrew.However, what was not foreseen was the usefulness of Russian, as a language spoken by students and by the German-as-Second-language teacher from Georgia.When Sandra and Jasmina speak about their first time at school, they remember: [Sandra:] She was really nice.I had a German course with her.Whenever everyone in the class had German, me and another boy, we went to the course and learned German.And that helped me a bit too.So grammar and so.[Interviewer:] So this was in German?[S:] No, that was in Russ, no, well, that was difficult, because the other boy, he came from Israel and he didn't know Russian.And she spoke Russian to me and simple German to him.And I also translated what she said in Russian for him.

[I:] In Hebrew? [S:] Yes. [I:] And that helped him? [S:] I think that helped both of us a lot."
[Jasmina:] I had a friend who spoke Russian, so I was lucky, because he translated to me at the beginning […].And I felt more secure, because, I mean, I didn't understand a word.And yes, maybe, because I had a lot of contact with him, I spoke less German, which was not so good.But I was happy and it was nice to be able to talk to him, because there was someone who understood me.Sandra describes the situation as "difficult" but at the same time as "helpful".The interviewer herself appears surprised as she asks "And that helped?"Sandra reflects on this situation as something that "helped both of us a lot", not only herself but also her Israeli peer-student.Jasmina also perceived Russian as helpful.However, Jasmina's and Sandra's perception reveal a difference: Jasmina judges this situation as "not so good" and considers whether speaking Russian was at the expense of German.This statement reveals the unquestioned importance of the German language on the one hand and the importance of being able to communicate and to be understood on the other.Thus, the two languages are opposed and result in an ambivalent perception of language resources as being, at the same time, both useful and hindering.In contrast, no judgment or ambivalence appears in Sandra's statement.The translation from Russian into Hebrew while learning German is stated first as a neutral and then, after the interviewer's question, as helpful.These differences can be understood in terms of the theoretical framework: Findings from children's perspectives reveal the importance of all linguistic resources in learning processes, as presented in section 1.1, on cognitive transfer.However, while Sandra experiences the use of multilingual practices as legitimate in a formal learning contextas an accepted capital, and herself as an expert and bearer of this capital -Jasmina relies on it in an informal setting, namely, in communication with a Russian-speaking peer, and doubts the legitimacy of this practice.

Differentiation between languages as linked to different contexts
Compared to the literature review, both Jasmina and Sandra perceive that the use of all languages is important and useful.Children in the presented literature report having experienced the need to decide between the languages they speak; and as state language is perceived as more important, they reject family languages by consequence.Today, they report that they are reluctant to speak the immigrant language (see section 1).In contrast, in Jasmina's and Sandra's statement, the languages themselves are not valued as more or less important but rather as practices linked to particular contexts.When asked what languages they speak to whom at school, the children stated: [Lucas:] All lessons are in German, Russian in Russian and Hebrew in Hebrew.In the Hebrew lessons we all speak Hebrew.But that depends on the group.For example, in Sandra's, in the best group, everyone only speaks Hebrew.In our group, most of us can read and write, but we don't understand everything.So, we also speak German.In the English lessons we speak English and in the Russian lessons Russian.And in the other subjects we speak German.[I:] And with each other:?[L:] At school, I speak mostly German.But sometimes also Russian, not very often.And Hebrew also sometimes with some friends from Israel.
[Jasmina:] Sometimes in Russian, sometimes in Hebrew and in spare time...I had a teacher with whom I spoke Hebrew.And he teaches in German.But I liked it, and it gives you the feeling of a closeness and it's a different atmosphere when you talk in Hebrew.So yes, and that's why it was more pleasant to speak in Hebrew, because we both just, yes, it' s different.But in class it is easier for us if the teacher speaks German, because we read German texts and explain them in German [...].
[Sandra:] Many of our teachers speak Russian or Hebrew.And then outside of school, when you are in your spare time, you automatically speak Russian or Hebrew.And then you have a completely different feeling.You are not a pupil, it's just out of the school.[Lucas:] Exactly, yes, if you ask in another language, yes, how are you doing, so if you talk about the weekend, if you just talk to the teachers about free time, then it's just different.
In these statements, no valorization of languages appears as a language that would be more important, in terms of indexicality (Blommaert 2005).However, children do assign different purposes to different languages: While German is marked as the language they "read and write", as a medium of learning, Hebrew and Russian are linked to informal conversation.Purposeful switching to Russian makes them perceive it as a switch to a different mode -"not a pupil" and teacher talking, but "we (are) both just (talking)" as people discussing their week-end.This can be compared to the discourse in research literature on academic and everyday register (Cummins 2000).However, children perceive and differentiate between the situational purpose not only regarding different registers but also for different languages.In terms of Bourdieu, the languages appear as capital that changes its value depending on the situation.Furthermore, these statements reveal a perception of multilingualism as normality for a person and in school as a context.They do not link 'one person' or 'one community' to 'one language', as compared to research findings with children who have experienced monolingual school settings.By including migrationrelated languages in the curriculum, as Russian was made an official language in this school, made it an institutionalized capital (Bourdieu 1990).But multilingual teachers also seem to play an important part as role models: Using different languages in contexts and giving feedback on multilingual practices as legitimized, they function as 'ideology brokers' (Blommaert 1999, p. 9; Plöger/Putjata 2019) who produce and reproduce multilingual ideologies at school.Through this positive feedback, children perceive multilingual practices as normal and themselves as worthy members of a multilingual community.

Self-perception as a worthy member of a multilingual community
Despite the positive perception of the multilingual practices presented above, the children also remembered negative situations when asked about their experiences at school: [Lucas:] I once asked in math, in a test, what is a 'measure' and the teacher answered "We are in Germany here."And hasn't answered the question […].She said, she can't say it to me, because we are in Germany.
[Sandra:] We had a close text with many difficult words.And she [the teacher] probably thought it was okay.So me and another friend who came from Israel and who had even more difficulty than me, we asked what the word means?And she said: "We are in Germany".I believe, she did not think anyone could have any problems.She thought it was ok.No idea.Somehow, it was for me... Very bad of her, that I realized that I, how can I say, that she doesn't want to help me.
"It was for me… very bad of her", remembers Sandra and even goes further to explain what exactly she felt as a negative experience: the experience of being helpless, and not being helped.However, it is worth noting that Sandra also explains this situation as a lack of experience on the teacher's side.In contrast to the findings on self-perception of migration-induced multilinguals in the national framework described in section 2.2, here the negative response by teachers does not result in the student perceiving him-/herself as a deviation from the norm, and, thus, lead the student to reject their family language in the process of integration; rather, it is the behavior of teacher that Sandra perceives as deviating from the norm and that needs to be explained: "I believe, she did not think […] She thought, it was ok." As consequence of these experiences, the children developed a self-perception of multilingualism as normal and were keen to maintain and enhance their multilingual practices: [Lucas:] I try to speak a lot of Russian with parents, and we also have Russian lessons since 7th grade and I try to read more.Yes, I've learned how to write that and so and so, and read.And in Russian we learn new declinations, verbs, that sort of thing.In Hebrew, we have to do different tasks, now we had different past forms.
[Jasmina:] But now I'm afraid to forget Hebrew.English less because I know that most people speak English.I don't want to forget any of the languages.
In the language portraits and also in the narratives around their language lives, children speak about multilingual communication as self-evident and do not differentiate in terms of language value, nor do we find in their narratives the terms 'home language', 'school language' or 'mother tongue', which is also in contrast to the results of the literature review as presented in 2.2.
These findings underline how important it is to include children's perspectives in the research and reveals further questions about children's perspectives on topics like identity construction or educational justice.At the same time, the qualitative data reveal the importance of the educational environment in transnational trajectories.These findings will be discussed in the following section with implications for educational practices.

Discussion and implications
The aim of the study was to reconstruct children's perspectives of language and their own multilingual practices in transnational education, and the findings reveal the importance of the environment.While we know that language attitudes (e.g., pride or shame) factor into the process of multilingual development and affect how languages are used, maintained or rejected, these attitudes are not individual traits but are rather the product of discursive power relations as they are experienced by children at school.The qualitative data from this study, with language portraits and group conversation, allow for deep insights into this process: On their transnational education trajectories, the participating children very consciously perceive and reflect the value and the importance of their linguistic practices.In the presented context -a school that adjusted to the transnational experiences of its students -this environment leads children to perceive languages as important individual assets and multilingualism as the norm.In German support classes, the use of Russian and Hebrew while acquiring German lets children experience the importance of all linguistic resources and their use as legitimate.The constructive implementation of immigrant languages for learning purposes results in children having the self-perception that they are an expert and their resources are indispensable.At the same time, the participating children experience the legitimacy of these immigrant languages, as Russian and Hebrew are integrated in the official school curriculum.As such, they can further develop all of their language skills; but, this is not the only benefit: By being graded on them, children receive the important feedback that these languages are also significant for academic success.This feedback is extraordinary, as home language classes in Germany are mostly complimentary and on a voluntary basis.Besides German support and immigrant language classes, there is a third major factor that appears important in children's narratives: multilingual teachers.Their presence and their linguistic behavior sets an example for how children can develop a pattern for their own multilingual practices.As compared to the existing research body, this pattern contradicts the linkage between one person/one community/one setting and one language.Thus, the participating children learn to rely on and consciously apply all their linguistic resources for different purposes.
These results are limited to a specific context and the very small sampling of participants, including siblings.Further limitations are set by methodological issues: whereas the participatory approach reveals its importance for child-centered research, it should be included not only for data collection, but also for data evaluation.Finally, the instrument of language portraits bears some dangers: First, the eliciting question (to locate the languages in the body) predetermines for the participants the division into languages (and thus the understanding of languages as isolated, closed constructs (against dynamic, fluid, transcendental); Second, it implies a justification for the separation of languages (why do I have language in my hand, in my heart, etc.), and thus, assigns different functions/positions/roles to the languages that the participants would not have thought of in this way.Being an exploratory study, these findings are of extreme importance for further research design: more group conversations with children from the same school, using situations from everyday school life rather than language portraits as impulse for joint data collection (as presented in Dlugaj/Fürstenau 2019) and active integration of children not only in data collection but also data evaluation.
Yet, despite these limitations, the findings present an example of how a school can transform itself in times of transnational mobility.The language diversity management at Aleph constitutes a strong step for individual migration trajectories.Being an official place of education, it allows for a formal investment in linguistic capital.In addition, positioning transnational children not as immigrants in need of help but as experts and equals to their peers can help to shift the discourse about the value of immigrant languages, which is still predominantly deficit oriented.This, however, would require further political measures that would ensure the relevance of diverse languages for further education and work careers.Yet, as revealed in the group conversation with children, these findings seem limited and children may still experience monolingual-oriented teachers' behavior.Thus, further research should include not only a larger sampling with more children, but also classroom observation and interviews with teachers, in order to better understand the underlying logic.
This study contributes to a thoughtful understanding of children's perspectives on their migration experience and reveals further questions for studies into this topic.For communities and societies that are interested in transnational education, the findings presented here promote further discussion on the possibilities and limitations of a 'multilingual turn' and on the role of immigrant languages and multilingual teachers in this process.

Notes
1 The name of the school has been pseudonymized. 2 Additional data were gathered through interviews with the core family.However, as the present study involves children and their perspectives, these data are not used for this evaluation.3 The researcher diploma was designed by the "House of the little researchers", for further information see www.haus-der-kleinen-forscher.de. 4 The names of all participants have been pseudonymized.5 Further group conversations are in preparation.The first group conversation serves as exploratory approach to optimize the research design.6 Data was analyzed by graduate students at the University of Münster as part of teacher training assuring research-based learning.The importance of research-based learning for the development of teaching professionalism is, however, not the focus of the paper and cannot be discussed here.7 Interview data with the mother are not subject to analysis and will be discussed elsewhere.

Figure :
Figure: Picture of a language portrait and a researcher diploma for Lucas 4