Transforming language brokering policies at school: Learning from students with transnational biographies

Within discourses on education and transnationalism, much importance has been given to language, both for the educational success of migrants and their transition into the labor market. Among the many perspectives on multilingualism, there is a growing body of research on language brokering. In contrast to professional interpreting, this is often associated with interpreting without formal qualifications or payment in mostly informal settings. However, language brokering is embedded in complex social structures, and students who interpret often act within complex power relations. In our article, we address experiences with education-related language brokering. While focusing on the perspectives of the students, we will address the roles they take during language brokering activities between different actors and the power relations they experience. Our data were collected as part of an ethnographic research project which aimed to explore the experiences of refugee students following their arrival in Austria and during the adjustment period in their new sociolinguistic and educational context. Empirically, the article is based on a data set of small stories and reflections about language brokering. The analysis was carried out through contrastive comparisons between different cases. The paper will conclude with suggestions on the introduction of critical language brokering practices and policies in educational institutions.


Introduction
Within discourses on education and transnationalism, much importance has been given to language, both for the educational success of migrants and for their transition into the labor market. Language-related questions take on additional relevance in the context of forced migration (cf. Le Pichon-Vorstman, 2020; Morrice et al., 2021;Niedrig, 2015). Among the many different perspectives on multilingualism there is a growing research body on language brokering (Abreu and O' Dell, 2017;Cline et al., 2014;Orellana and Phoenix, 2017). The practice of language brokering differs from professional interpreting primarily in that children and young people interpret without formal qualifications and payment, often in informal arrangements within educational institutions. This means that language brokering is embedded in complex social structures and power relations, and students who interpret often play a critical role in dealing with unequal power relations between institutions, on the one hand, and marginalized individuals and families on the other. Frequently, these language brokering interactions even involve consequences for the interpreting student. In our article, we address experiences with language brokering which students are called on to do at school. While focusing on the perspectives of individuals who regularly perform the role of interpreters, we will address the following questions: • • What roles do students take during language brokering activities between different actors (peers, teachers, parents, social workers etc.) at school? • • What power relations are linguistically minoritized students confronted with during their brokering activities? • • How are language brokering activities embedded in the individual transnational biographies of students?
By drawing on the students' narratives we aim at understanding language brokering through their experience and perspectives. Thereupon we will reflect on what can be learnt from these insights for the implementation of critical language brokering policies at schools. The data presented in the article were collected in the context of an ethnographic research project which aimed to explore the experiences of refugee students following their arrival in Austria and during the adjustment period in their new sociolinguistic and educational context. Empirically, the article is based on a data set of small stories (Bamberg and Georgakopoulou, 2008;Schnitzer, 2020) and reflections about language brokering by students of different linguistic backgrounds. The analysis was carried out using contrastive comparison between different cases. The paper concludes with suggestions on language brokering policies in educational institutions within a broader societal and political context.

Research context and methodology
The project "ZwischenWeltenÜberSetzen" ("Translating Wor(l)ds"), 1 which was carried out between 2017 and 2019, aimed to reconstruct the biographical experiences and competencies of young people around the age of 17 who had come to Austria in the course of refugee migration and were schooled together in a so-called "transition class" for 1 year. "Transition classes" are based within mainstream schools and aim to prepare refugee students to attend regular classes within 1 year. They offer some general education in mathematics, English and other school subjects, but they primarily focus on teaching and learning the German language. As has been noted, the somewhat politically naïve institutionalization of "transition classes" for refugee students failed to recognize the heterogeneity of educational pathways and experiences as well as language repertoires of the students, which led to a number of social and pedagogical challenges (Dausien et al., 2020). The 16 students we were working with throughout our project differed not only in their backgrounds but also in their legal status in Austria (some of them were still seeking asylum, while others had been granted asylum), a difference which shaped both their position in society and their plans for educational transitions (see Thoma and Langer, 2022) and the future in general. Our focus was on young people's experiences in the process of arriving in the Austrian school system and with multilingualism and translation. School was conceived as a space of belonging and education, one that could open up or close off possibilities of articulation. Students were addressed as competent subjects with multiple experiences of translating between different languages and social worlds, which they explored together with researchers.
Theoretically, our view of multilingualism in schools is inspired by a speaker-oriented perspective that understands language against the background of biographical trajectories (Thoma, 2018(Thoma, , 2020 and as an interactive field of language practices. In doing so, we consider multilingualism in its interplay of linguistic resources, practices, and ideologies (Androutsopoulos, 2018: 197). Methodologically, we worked with ethnographic field notes and conducted interviews with students and teachers. We did not conduct in-depth biographical narrative interviews (Schütze, 2008) during the project, as the need to tell a consistent and "true" life story is inextricably linked to interviews in the context of asylum procedures and also involves the risk of retraumatization (Thielen, 2009). Therefore, we worked with different formats of "small stories" (Bamberg and Georgakopoulou, 2008) around arriving in Austria, school, languages and translation. In addition, we were sensitive to listen to spontaneous biographical narratives beyond methodological guidance (Völzke, 2005) and to give them space. The project was designed to be participatory and involved the young people as actors in the research process.

Language brokering and language education policies in schools
We contextualize this article at the nexus of theories on language brokering and on language education policies in order to show how the combination of these two perspectives can enhance our understanding of language brokering as part of language education policies in education institutions.
Over the last two decades, there has been a growing interest in language brokering across a variety of disciplines, such as childhood and development studies (Cline et al., 2011;Orellana and Phoenix, 2017), education (Pena-Díaz, 2019), linguistics, (Tse, 1996), and psychology (Crafter and Iqbal, 2020). Despite theoretical and conceptual differences, which are reflected in the use of different terms, such as "procedural brokering" (Lazarevic, 2017), "family interpreting" (Valdés, 2003), "family brokering" (Hall and Sham, 2007: 18), "community interpreting" (Ahamer, 2014), and "child brokering" (Katz, 2014), language brokering can be described as a process in which communication between speakers of different languages is mediated. Language brokering differs from professional interpreting in that primarily children and young people interpret without formal qualifications or payment, often in informal arrangements within institutions and families. It is thus embedded in a variety of power asymmetries, such as disparities in the symbolic capital of languages (Orellana, 2009: 27), racial and cultural power differences (Ishimaru et al., 2016), intergenerational power or power related to chronological age (Hall and Sham, 2007;Orellana, 2009), and hierarchical power dynamics between parents and educators (Ishimaru et al., 2016). Language brokering has been defined as "a particular form of the translingual practices that are part and parcel of growing up in places where more than one language is spoken" (Orellana et al., 2014: 311). Some researchers point to non-normative relationships in families and problematize language brokering against the background of parentification (e.g. Titzmann, 2012). Other researchers, however, have shown how child language brokers in migrant families occupy "shifting positions of authority and power during brokering events" (Eksner and Orellana, 2012: 215). By revealing that child-adult and learner-teacher roles are not static, they have challenged normative Western models of learning and development as well as notions of developmental roles (Eksner and Orellana, 2012: 215). Furthermore, it has been argued that language brokers take active roles to achieve desired outcomes and challenge issues they disagree with. Some studies also indicate that young language brokers learn the skills of citizenship by engaging in decision-making processes that affect social outcomes and thus learn to become citizens through the work of interpreting (Bauer, 2010). Furthermore, researchers have shown that the non-normative roles young people take when language brokering can conflict with normative institutional expectations, such as regular school attendance . While some authors conceptualize language brokers as "institutional agents" (Martinez-Cosio and Iannacone, 2007) emphasizing their critical role in institutions, we focus on individual experiences in language brokering processes and their interactional, educational and biographical significance.
To understand language brokering policies in school, our study draws from a conceptual framework on policy as practice (Bartlett and Vavrus, 2014;Sutton and Levinson, 2001) which sees education policy as a practice of power and aims not only at scholarly interpretation but also at the democratization of policy processes (Levinson et al., 2009: 768). Such a critical approach is not necessarily centered around official and codified forms of policy but includes unauthorized and unofficial policy in unwritten forms, which are negotiated and (re)organized in ongoing institutional memory and practice (Levinson and Sutton, 2001;Levinson et al., 2009). Thus, it allows us to analyze the (re)production of social order through policy in a given domain. It also contributes to a better understanding of language policy, since "language policy exists even where it has not been made explicit or established by authority" (Spolsky, 2004: 8). Since language policies of institutions can be best understood from a study of their language practice or beliefs (Spolsky, 2004: 8), we understand institutional ways of dealing with language brokering as part of language education policies that are more or less intentionally put into practice by schools. Hence, teachers, school heads and professionals in school boards who are "at the epicenter of this dynamic process" (Menken and García, 2010: 1), act as ideology brokers (Blommaert, 1999) and can thus perpetuate and/or question and transform social inequalities related to multilingualism. Although the practices related to language brokering are multidimensional and can include such different actors as pedagogical staff, students, parents, and community members with their varied language repertoires and perspectives, most of the written educational and policy documents in schools do not contain any suggestions or norms related to language brokering. Finally, understanding language brokering practices as part of language education policies allows us to reconsider how actions and interpretations of subjects are interwoven with broader institutional and social structures.
Prior work has revealed that language brokering in different social, cultural and institutional contexts (such as schools, hospitals, stores, restaurants, and home) is embedded in complex power relations, involving different demands and "differing degrees of support and/or pressure" (Orellana, 2017: 69) for manifold texts, tasks, and activity settings. Using observations and audio-taping of live translation episodes, researchers have examined how young language brokers in parent-teacher-conferences communicated teachers' evaluations to their parents and how students manage their paradoxical and simultaneous positions as translators, as interlocutors being talked about, and as co-narrators-in-translation (García-Sánchez et al., 2011;Orellana, 2009;Sánchez and Orellana, 2006). This research has also revealed how language brokers in these conferences are subject to multiple forms of surveillance (Reynolds et al., 2015). Using vignette methodology, researchers have inquired individual stances of youth on language brokering and competing demands of autonomy and connectedness in the relationship between children and parents . With narrative methods, researchers have inquired how adults who have grown up brokering for their parents construct their childhood language brokering experiences and how adults' views of their own language brokering practices as children change over time (Bauer, 2013;Orellana and Phoenix, 2017). Our work builds on these various lines of research and focuses on experiences of students who are conducting language brokering. Since school is one of the main sites in which language brokering practices take place (e.g. Orellana, 2017: 71-72), and since it is embedded in specific cultural and institutional settings and power relations, our research focuses specifically on language brokering at school by eliciting students' reflections on their experiences in this context. Our data and analysis are therefore centered around situations of language brokering in educational institutions as narratively reconstructed by students, and we contrast them with experiences in other settings, such as home or social work settings. More specifically, as there has been identified a need for expansion of research into new populations and contexts (Orellana, 2017: 74), we focus on newly arrived youth whose residence status is not yet secured and whose languages have a particularly low prestige in the institutional context (see Alpagu et al., 2019). Moreover, we do not limit ourselves to a reconstruction of young people's perspectives, but make suggestions on how schools can do more justice to young people's practices and achievements as language brokers and transform their educational policies along ethical considerations and in line with young people's experiences and perspectives.

Understanding language brokering through students' experiences
On the school's website, the languages Spanish, French, Italian and Czech are listed under the heading "Linguistic Diversity." In the logic of a commercial school, these languages are introduced as "business languages," "world languages" and as international languages. Overall, this "celebration of diversity" (Piller, 2016: 19) reveals a commodification-oriented perspective on languages (see Heller, 2010). In the linguistic hierarchy which both the website and the curriculum revealed, German is immediately followed by the above-mentioned languages, which have a high prestige and a fixed place in the curriculum, while the less prestigious migration languages, which are part of many of the students' linguistic repertoires, don't have any relevance in the institution's selfimage or in teaching. In line with this invisibilization of marginalized languages is the lack of an explicit institutional policy on language brokering. Our ethnographic analysis of the institutional practices and the experiences of students, however, revealed that the institution sees interpreting mainly as an instrument for giving and receiving information and that students who speak marginalized languages are expected to provide language brokering work in and for the institution (see also Dausien et al., 2020). Such a monolingual habitus (Gogolin, 2008) is not surprising considering that only the German teacher in the transition class has training in German as a second and foreign language and that both the majority of the teachers and the principal have not had the opportunity to engage with language and multilingualism during their university studies, as these topics are still not solidly anchored in Austrian teacher education. Our data, however, show that the students themselves experience interpreting in complex ways. Presenting these different layers and facets of interpreting, we wish to provide insight into roles and (power) relations that ought to be considered in the context of language policies in schools.
Over the course of one school year, during which 12 workshops were held, we worked with various methods and media that were intended to enable, but not enforce, the articulation of autobiographical narratives. We used approaches from adult education and pedagogical biography work (Behrens-Cobet, 1999;Dausien, 2011) and adapted them to the work with youth. In addition, we invited the students to our university, to the cinema, the museum, and to a hiking tour in order to create a basis of trust outside of the institution and to enhance possibilities for engaging in discussions which were not only related to school. In a few workshops in school, we worked with moderated storytelling sessions, in which the young people had the opportunity to tell stories about a specific topic in a small circle. In one of these workshops, the topic was "translating" and the experiences the young people had with different forms of translation and the different social processes around them. After talking about their experiences, the 13 students participating that day discussed similarities between their stories and aspects that they considered relevant to the topic. We will now present some of these points together with extracts of the students' accounts and their explanations to illustrate the multilayered perspectives on translating the students shared with us. The work of interpreting (or being interpreted) reported by the students took place in various social contexts. Some experiences were simply about buying clothes or explaining the rules of a game to others. Many others dealt with bureaucratic settings involving challenging communication processes in asylum procedures, hospitals, schools or interacting with the police. In their narrations and concluding reflections on the experiences the students expressed that interpreting for them had to do with the ability to speak for oneself and to help others. They also pointed to complex questions of power and trust, (in)dependence as well as "standing in between." From the many stories told by the young people which were situated in very different settings, we will present and analyze narratives from two students. Their accounts were selected for the purpose of this article because they deal with language brokering related to school and education decisions in the Austrian educational context.

Becoming independent from interpretation
In one of the workshops with the transition class, we had a moderated session in which the students had the chance to talk about their experiences with translation. Haadiah 2 tells us about her first experiences arriving in a new school when she was still beginning to learn German:

Es gab auch eine Junge aus Afghanistan bei uns in alte Klasse. Aber er war schlimm, er hat mich_ er hat mir nicht geholfen, dann hab ich immer versucht -Deutsch lernen und mich verbessern -deutsche Sprache und auf EIGENE Füße stehen.
There was also a boy from Afghanistan with us in old class. But he was bad, he did me_ he did not help me, then I always tried -learn German and improve -German language and stand on my OWN feet. 3 Haadiah mentions a classmate who, like herself, came from Afghanistan. Characterizing him as "bad," she presents her classmate as a person who did not help her. The expectation of solidarity between two students who share national origins and presumably also a common language is thus unmet here. However, Haadiah says that she continuously "tried" to learn German and was able to improve her knowledge of the school language with the aim of standing on her own feet, emphasizing the word "own." She continues with her account of the concrete situation, referring to the same boy: In the reconstruction of a situation in which she asked her classmate to interpret something for their teacher, she recognized the product of the interpretation as deviating from her statement. The "then," which announces a temporally subsequent consequence, is repeated here ("then I thought"), and learning German is presented as a necessity without alternative in order to be able to solve one's own problems. Thus, this narration can also be read as an argument for having to become independent of the help of others. The moderator asks how she noticed that the interpretation was not correct, and she says: Haadiah begins by contradicting the assessment of her non-existent knowledge of German and pointing out that she already knew "a little" German. She substantiates and explains this by saying that her grammar was still a little error-ridden and that she therefore did not want to speak for herself thus pointing to the aspect of shame in language learning processes (e.g. Liyanage and Canagarajah, 2019). However, her request for a correct translation into the language of the teacher was not fulfilled: In the interpretation of the boy there was a mistake in content, namely a "wrongly" used word, which Haadiah says she had not used. The phrase "something like that" could indicate that Haadiah considers "that" word used by the boy to be inappropriate or misleading.
After a short break, the moderator asks the group if anyone else has had experiences with "wrong" translations. Since nobody comes forward, the moderator asks what translating means for the young people, and what from their point of view is important about it. Haadiah answers: Die richtigen Worte auf seine oder ihre Muttersprache verstehen -wissen -erklären -deutlich (2) und genau sagen, was (der) Person will. Was übersetzt wird.
Understanding the right words in his or her 5 native language -knowing -explaining -clearly (2) and saying exactly what (the) person wants. What is translated.
She begins by referring to the "mother tongue" and emphasizes that it is important to understand the "right" words. Understanding is followed by "knowledge," although it is unclear whether this refers to knowledge related to the aforementioned words or to metalinguistic knowledge. In any case, Haadiah additionally names "explaining" and concretizes this by saying "clearly" and "exactly" what a person "wants." The emphasis on linguistic precision in language brokering is juxtaposed with a translational performance that captures not only what a person "says" but what he or she "wants." In Haadiah's view, a language broker must convey not only the content of what is said but also the intention of the person speaking.

Getting responsibility with a widening linguistic repertoire
While in the last section, it was central to Haadiah to free herself from dependence on incorrect interpreters and to gain greater agency in German, the following passages show that a larger linguistic repertoire can lead to new difficulties: Früher habe ich auch sehr geholfen mit meiner Mutter. Sie hat immer gesagt äh sie lernt B1 gerade [. . .]. Dann wenn sie etwas_ die Aufgabe nicht verstehen kann, dann kommt sie zu mir und sagt mir: "Hilfe mir. Wie geht das? Wie soll ich das äh schreiben? Oder wie kann ich das lösen?" Dann sag ich: so, so und so. Dann ((lacht)) auf einmal sag ich: "Ja, immer muss ich das machen. Immer soll ich äh euch helfen, ihr könnt das nicht und ich will das nicht. Immer schreib ich eure Hausaufgaben." "Ja, sag mir, wie schreib ich das? Ist das richtig, wie ich geschrieben habe?" [. . .] Ja und hab ich auch einmal gesagt, äh wenn wir nicht äh positiv Asylbescheid bekommen, wenn negativ oder muss man wieder zurück zu unsere äh Heimatland gehen oder ein andere_ ein anderes Land, dann würd ich nicht die andere Sprache lernen. ((lacht)) Alle kommen zu mir, übersetzen übersetzen. ((lacht))" In the past I also helped a lot with my mother. She always said uh she is learning B1 6 right now [. . .] In this account, Haadiah refers to a phase in which she supported her mother in the context of a German language course. The re-enacted dialogue refers to regular requests for support and to concrete questions concerning linguistic or content-related problems with homework. Haadiah's reported speech "Then I say: so, so and so" refers to a well-rehearsed approach to solving such tasks. However, the evidence of dealing with her mother's homework for the German course is then broken by Haadiah pointing out the burden of the ongoing obligation. Her phrasing in the plural 7 shows that helping others, not just her mother, is not something she is eager to do. "I always write your homework" reflects a reversal of the intergenerational hierarchy; and the mother and perhaps other adults are put in the position of needing help. Via a restatement of speech, Haadiah refers exemplarily to questions she is continually confronted with. To show how much she sees the assistance as a burden, she recounts that she once jokingly said that if she had to move again to her home country or another linguistic context, she would no longer learn the other language. This "joke" isn't presented as a mere thought experiment but as a realistic possibility as she refers to the ongoing asylum process. By pointing to this uncertainty about her future residence, she emphasizes the fragility in transnational educational pathways, especially in the context of migration and flight, and the associated complexity of motivation in language acquisition processes.

Navigating between different expectations while interpreting
In the following, Haadiah recounts an experience with language brokering that was situated in a decision-making process concerning the next step of her educational pathway. In this situation, she had to choose between attending a "regular" school or an education program called "Youth College" which is mainly targeting young refugees who are no longer of compulsory school age 8 : This passage gives us clues to communication procedures and decision-making processes in the setting in which Haadiah finds herself. She presents herself as an active agent who wants to have a say in deciding her own educational path. However, she also expresses uncertainties and doubts: she doesn't seem to be able to decide or know which option to choose and becomes "nervous." The tutor takes an essential role in this process, providing knowledge about education and schools, making suggestions in this regard, and initiating action. She also calls in the parents who are portrayed as significant figures in the decision-making process, but nevertheless do not appear to act on their own initiative. In this constellation, Haadiah acts as a mediating link, and she explains her role in this form of indirect communication: "Then I have to translate, I was very nervous." Even in the personal conversation with all participants present, Haadiah has a mediating function between the actors. On the one hand, the reference to her own nervousness follows from the nervousness mentioned earlier, which can possibly be explained by an uncertainty about the decision. On the other hand, the function as language broker can be understood as a new and potential further reason for her nervousness, although this causality is not made explicit in the narrative. However, the narrator already makes clear that the activity of brokering is not easy for her in a situation where she is nervous (for whatever reason).

Yes -then -my father and my mother then said uh: "Choose something, either Youth College or -school." Then I chose the school.
The reported speech reveals that Haadiah is subjected to institutional surveillance (see also Reynolds et al., 2015): The tutor puts her under pressure to give reasons ("Why don't you go?") and counters her reservations. Besides the two variants offered to Haadiah, which are both unpleasant for her, the tutor does not suggest any other alternative. While Haadiah already has a lot of local institutional and cultural knowledge and is aware of the different values of the educational pathways available for choice, her parents most probably don't possess this knowledge, which is why they obviously cannot offer any further alternatives either. Thus, they ask the daughter to choose between the two suboptimal variants. This is how Haadiah's story at our partner school starts.
Interestingly, at this point Haadiah only reproduces questions and advice directed to her by the tutor, although the conversation was initiated by the tutor to talk to the parents. As in the narration about helping her mother with her homework, the parents appear as secondary characters she cannot count on in terms of support and advice. Haadiah's point of "standing on one's own feet" thereby gains even more relevance. She positions herself as a person who-on different levels-is responsible for her own communication (by conducting and interpreting the conversation), her own decisions, her own life. The parents are not directly addressed (at least not in the narrative). Instead of answering the tutor's questions, Haadiah first interprets what was said for her parents. Again, she reiterates her nervousness about the situation. At this point, the parents appear for the first time as speakers and call on the student to act, to make a decision by taking up the tutor's suggestions. The narrator concludes the narrative with the outcome of the situation, "Then I chose school." She thus reappears as an agent. Although it can be stated that she merely chooses from two options-two options she herself isn't very happy about-and her scope of action is rather limited, she presents herself as deciding agent who has the last word in the discussion.
After the main exchange above, there is an opportunity for follow-up questions to the narrator. Haadiah is asked if the conversation with the parents took place on the phone or if all the people were together in one room. She responds by explaining that everyone involved "sat together." She then goes back to the conversation already described and illustrates it with an excerpt from the conversation: wir haben gemeinsam gesessen, ja "Ich will das nicht" ((lacht)) Und mein Betreuer hat gesagt: "Wieso willst du nicht?" Dann meine Mutter hat gefragt: "Was sagt sie?" ((lacht)) Sie sagt mir: "Wieso gehst du nicht?" Dann meine Mutter hat wieder gefragt: "Ja, wieso gehst du nicht?"

we were sitting together, yes "I don't want this" ((laughs)). And my tutor said, "Why don't you want to?" Then my mom was like, "What is she saying?" ((laughs)) She's telling me, "Why don't you go?" Then my mom asked again, "Yeah, why don't you go?"
Here it becomes clear that Haadiah takes on several roles in the conversation. On the one hand, she is the content and addressee of the conversation. The tutor and the parents (in this case the mother, as the father doesn't appear as a talking character in the narration) are interested in finding out the reasons for her hesitation and refusal to go to school ("Why don't you want to?," "Yes, why don't you go?"). The student is thus directly addressed from all sides, and prompted to answer and to make a decision. At the same time, however, her mediating function as a broker becomes obvious ("Then my mother asked: 'What is she saying?' -She tells me: 'Why don't you go?'"). The narration reveals the demanding situation in which Haadiah finds herself. She has to negotiate so much simultaneously. On the one hand, she is supposed to come to a decision and realize what she wants and needs. On the other hand, at the same time, she is also supposed to explain the situation to her parents and respond to the tutor's order, so that she has not only to translate in all directions but is also exposed to a double demand for legitimacy. This account is followed by another question during the workshop aimed at finding out how Haadiah felt in the situation described. Haadiah answers as follows: Ja, ich wollte das nicht übersetzen ((lacht)), damit meine Mutter es nicht versteht, was ich will oder ich will nicht in die Schule_ zur Schule gehen. Dann muss ich übersetzen. Es gab niemand (übersetz).
Yeah, I didn't want to translate that ((laughs)) so my mother doesn 't understand it, what I want or I don't want to go to school_ to school. Then I have to translate. There was no one (translate).
Her thoughts about the activity of language brokering at the end reveal a struggle: Haadiah does not want to interpret and would prefer to withhold information (about herself and her wishes, possibly also her insecurity) from her mother. On the one hand, her essentially powerful position as a broker becomes apparent, as she is the only one who has access to the languages of all those involved in the conversation and who, through her mediation, could withhold, add or change information. Yet she does not seem to be interpreting with any form of strategy or self-interest. She does not present the act of brokering as a happy opportunity for influence, but rather as a necessary evil, something she is obliged to do because no one else is present to perform this function. It can therefore rather be assumed that she interpreted as "neutrally" as possible. The background to this could also have been that she had not yet made her educational decision and, rather disoriented, would have preferred not to decide at all. She finally makes the decision only after both the tutor and the parents have asked her to do so.
Haadiah's narrative presents an example of how students can experience a situation in which their linguistic competencies are called upon to enable communication in interactions. The student, who has to act as an interpreter between tutor and parent, fulfills this role "nervously." The situation is particularly interesting in that she not only participates in the conversation by translating but is also the addressee and subject of the conversation, which is about a decision regarding her further educational path and thus about her future life. We do not learn from Haadiah's perspective whether the tutor's language brokering policies were deliberately employed or whether they had pragmatic or economic motivations. In any case, the passages suggest that the tutor does not address the complex position in which Haadiah is put and that language brokering by students when their own future is at stake appears to be a common practice. Interesting in this case are Haadiah's remarks on feelings and attitudes toward her "duty" to interpret in the reported situations as well as the way she positions herself as an acting and deciding person in an apparently overwhelming situation, which can be addressed through this narrative research angle.

Getting involved in discussions on peers' educational trajectories
Our next empirical example of language brokering by students in school is an extract from an interview conducted with Sami, another student of the "transition class" in the school we were cooperating with. Sami was not present the day of the above mentioned moderated storytelling session, but he was one of the students we were able to stay in touch with even after the year of research at the school. He is also one of the students with whom we decided to do a biographical interview. As mentioned above, we generally did not conduct in-depth biographical interviews with the students from the "transition class" for ethical reasons as we considered this to be a potential risk for them in the context of ongoing asylum procedures. Unlike most of his classmates, Sami was already granted asylum, and after 1 year of getting to know him, we felt certain enough to ask him for his life story.
During the interview it became clear that Sami is very familiar with brokering activities and supported others with translations whenever he had a free day. When asked about language brokering experiences at school, he remembered one incident. At the beginning of his narration, however, he hesitates and asks the interviewer whether he should really report from the conversation which took place at the principal's office, thereby showing his awareness of his role as an intermediary dealing with sensitive and personal information. The interviewer starts to reassure him and tells him that he doesn't have to tell anything he doesn't want to. Sami states that he will only talk about the essential information of the conversation and not go into detail.
He then introduces the incident that occurred at school a month before the interview. A student didn't pass the school year and wanted to start working. The school principal called for her father, and since there was no common language to communicate, he called Sami in to interpret a discussion between himself and the student's father. The principal's request for translation did not seem to be unusual for Sami or in need of any explanation as he did not comment or elaborate on it any further in his narration. However, his introduction into the scene reveals some hesitation about how much detail to give:

S:
Und der Direktor hat ihren Vater eingeladen (4)  Samit's hesitation does not refer to the fact that he got involved, without being asked, in a conversation between the principal and a colleague's father on the subject of the other student's future educational path. Rather, his hesitation relates to the legitimacy of his narrative: Sami is apparently aware that he had access to sensitive information and data, and that his involvement in a parentprincipal conversation could be seen as problematic by others. He then starts: S: Ja und als der Direktor der den Vater gesagt hat das = das Mädchen will nicht mehr lernen sondern arbeiten gehn der Vater war schockiert und er war zuckerkrank und ihm gings schlecht. I: mhm S: Ja. Ihn = ihn war also nachher ich bin zu dem Mann gegangen und ich hab ihn gefragt warum es ihm so schlecht gegangen und so und so. Sie kann auch (unv.) neben den Arbeit lernen wenn sie will und so. Und aber er er machte sich Sorge dass sein Mädchen nicht mehr lernt.

S:
Yes, and when the director told the father that=the girl didn't want to study anymore but wanted to work, the father was shocked and he was diabetic and not well. I: mhm S: Yes. I went to the man and I asked him why he was in such a bad condition and so on. She can also (incompr.) study besides work if she wants to and so. And he was worried that his girl would not continue studying anymore. I: mhm S: He wants_ so he said "As long as I live I can afford her_ and she=she so I can afford what she wants then I want her to go learn because=because so that will help her a lot in the future." Yes. And yes so now I knew why=why the parents want the=the sh_ the=the children to always learn because_ /(laughing) why I don't know/ but I mean yes they always want only the best for their children. And sure if you educate and=and study then he becomes the best. I mean believe.
Sami mainly states what the one said ("the girl didn't want to study anymore") and how the other reacted ("the father was shocked"). He does not provide any more information about the conversation of the two but continues his narration by pointing to his own conversation with the father of his classmate later on. By approaching the father and asking him why he has been unwell, he takes action and leaves his role as intermediary. Even more, he starts explaining the classmate's options to him ("study besides work") somehow speaking for her and defending her point of view. It seems that by leaving his role as an interpreter between principal and father, he is taking up the role of a mediator between father and daughter and as an expert for the Austrian educational system. In his narration Sami then moves on to the father's statement concerning his wishes for his daughter's future and concludes with his own takeaway from this conversation; namely, that he now understands better why parents always want their children to study because they always wish the best for them. He confirms the father's view with a final comment of his own ("And sure if you educate and=and study [. . .]"). This elaboration of his own conversation with the father might have been caused by his effort not to tell too much about the situation in the principal's office when he was still the interpreter and thereby bound to confidentiality. Another reason might be that his focus lies on his own experience and on the aspects he has learnt from this incident. Earlier on in the interview he explained that he likes interpreting because he often gets to hear people's experiences and life stories and thereby learns a lot. This could be a strong motive for his brokering activities for other people, which is contrary to Haadiah's experience cited above, who seems to see interpreting more as a form of being responsible, which can be overwhelming and burdensome. However, Sami's example reveals the language brokering policy put into practice by the principal. The principal, perhaps unintentionally, not only gives the student the responsibility for linguistic translation but also confronts him with an interaction in which his classmate is subjected to institutional surveillance and in which he is given access to information that is actually subject to professional confidentiality. In addition, Sami becomes engaged in mediation and conflict resolution.

Comprehensive discussion
The narrations show a wide range of elements which are crucial for a broader understanding of language brokering activities and related roles and power relations which are experienced by students in educational institutions. The comparison of the analyzed narratives reveals several aspects of language brokering: Haadiah's first narrative shows that reliance on brokering is experienced as constraining and can be seen as a motivation to learn or improve a language. However, as her second account shows, an expanded linguistic repertoire that makes someone independent of the language brokering work of others while enabling the support of others linguistically can also lead to limitations. We don't learn from the narrative data if the language brokering event has been experienced as reversing power relationships or as a way to co-construct knowledge and learn together (e.g. Eksner and Orellana, 2012). However, the analysis revealed that the roles and responsibilities taken by the actors are not static. Comparing the reported dialogues, we can see a major difference between language brokering in the private sphere and in educational institutions: Haadiah's narrative on helping with her mother's homework shows that the reported situation allowed her to articulate her arduous and time-consuming involvement in language brokering and to voice her displeasure to her mother. However, in the reported situations at educational institutions, the commitment to language brokering is not questioned in the reported dialogues, neither by Haadiah nor by Sami. This points, on the one hand, to the power of the institution to employ students as language brokers whenever they are needed and, on the other hand, to institutionalized practices behind which there is no explicit language brokering policy. The retrospective narrative within a research setting that is not connected to the authority of school enables insights into the students' reflections on the challenges they experienced: Haadiah refers to her nervousness in the narrated situation several times and she addresses the unquestioned and laborious obligation to translate. Sami does not address his feelings in the narrated situation, however, his hesitation about whether or how much of the conversation he should tell hints at his entanglement as a language broker between a classmate, her father, and the principal, and refers to his role conflict, which is brought up again in the interview situation and expanded by the question of whether he should pass on confidential data to the researcher.
Both stories pose important questions concerning the practice of language brokering in educational settings relevant to be addressed by pedagogical and administrative staff: What are the consequences of these different forms of translation for young language brokers (interpreting conversations concerning one's own issues and interpreting conversations about peers)? Which power relations and social roles-between parents and children, institutions and families, among peers etc.-are connected to these interactions and how can these often challenging and even conflicting situations be handled in a responsible way? We will conclude by discussing implications and suggestions for critical language brokering policies in schools.

Conclusion
In this article, we have argued that language brokering can create a site of tension for young people, especially when they translate content that relates to their own or their peers' educational paths and futures. Thus, they become brokers not only between different sides and their arguments but also of the content of the translated text. In addition, they unintentionally gain access to confidential data and problems that institutions do not prepare them to handle. We have drawn on narrations from students who regularly engage in language brokering activities for their families and for pedagogical professionals. Our methodological approach has allowed us to reconstruct how young language brokers experience and handle the complex and contradictory roles they have to assume when implicit language brokering policies are enacted at school without reflecting on possible consequences and entanglements. Drawing on narrations as in the case of interviews and small stories, the students' thoughts, feelings and attitudes toward the reported interaction and the way they position themselves through their narratives (e.g. as actors who speak and decide for themselves or who support others) can be addressed. The results show that students, even if they interpret regularly, sometimes feel insecure in their roles and do not always enjoy performing them. While the narratives indicate that the articulation of dissent is possible in the private sphere, their work as language brokers is so taken for granted in the institutional hierarchy that it cannot even be questioned in the retrospective narrative.
Our article concludes by reflecting on the relevance of these insights on language brokering revealed by the students' perspectives for school administrators and pedagogical staff: What can be learnt from these experiences for language brokering policies and practices at school? We would like to highlight three points that are relevant for the implementation of critical language brokering policies in schools: First, our findings testify to the relevance of language brokering in transnational educational pathways, especially if the right for asylum is at stake. As a common practice for students in migration societies, language brokering becomes relevant especially when places of residence are temporary and when students don't know where they will live in the future. Language brokering then takes place in critical and sensitive decision-making processes in bureaucratic settings and in interactions with institutions. "Transition classes" are biographical time-spaces of structural uncertainty (Draxl and Thoma, 2022), and since educational transitions are highly uncertain if the residence status is not secured (Thoma and Langer, 2022), situations in which students have to interpret add further facets of uncertainty and precarity. As Haadiah's example shows, she has already thought about further migration and what it would mean for learning other languages. However, her example challenges the normative expectation that learning the dominant language after a migration is for one's own benefit. Rather, a linguistic repertoire that includes the dominant language and enables young people to engage in language brokering activities can also be limiting because individual needs have to be put aside in favor of the needs of others.
Second, our analyses revealed that language brokering practices are used in schools without being linked to or governed by explicit policies. This comes as no surprise, considering that Austrian language integration policies are characterized by an "assimilationist monolingualism" (see Flubacher, 2021) and that educational policies are still oriented toward the dominant language of instruction (Dirim et al., 2018). While our ethnography has revealed that some teachers were concerned about multilingualism and language ideologies in the institution (Dausien et al., 2020), most of them and the principal saw the language brokering services of students as natural and took them for granted. Rather than attributing such a pedagogical perspective to the individual responsibility of professionals, we see the "will to policy" (Levinson et al., 2009) as subject to political circumstances and practical contingencies and, more concretely, to the language integration policies and the marginalization of multilingualism in teacher education curricula. Therefore, it is necessary that multilingualism becomes an integral part of teacher education and that schools create reflective spaces that enable teachers to recognize languagebrokering practices at school as socially constructed rather than natural, thus opening opportunities for policy-making.
Third, building on the previous points, we advocate for the introduction of critical language brokering policies in educational institutions that take the complex ethical issues and power constellations that were addressed through the students' narratives into account. Considering the often lacking infrastructure and support for interpreting processes in schools, it could be a starting point to implement training programs inside the schools for raising awareness about and reflecting on the potentially conflicting roles and hierarchies that students (as well as multilingual pedagogic staff members or parents) are probably involved in when interpreting at school. Also the aspects of confidentiality, responsibility and conflicting interests have to be addressed and reflected. Rather than establishing explicit and rigid language brokering policies which could be constraining as well, we advocate for the creation of spaces in which language brokers can express their experiences, thoughts and feelings, including the right to say "no" and the possibility to ask for appropriate and flexible supports to meet the requirements of differently challenging situations. During our workshops we had good experiences in addressing language brokering and collectively reflecting upon related implications and challenges. We thereby also noticed the students' high awareness of entanglements and conflicts regarding language brokering. Creating such spaces and possibilities for exchange can empower students in their roles and rights. It can also generally draw attention to the dilemmas and difficulties of language brokering at school.
In addition, support structures such as supervision and training for young people and their parents in dealing with students' roles as language brokers should not be set aside in favor of pragmatic solutions and symbolic acts. Establishing such support structures requires professionalization of pedagogical and administrative staff at schools and school boards to ensure that brokering processes can be adequately framed with an eye to interactional challenges and complex power relations. Qualified support teams could supervise and support language brokering teams consisting of teachers, parents and students who volunteer for certain occasions and are not taken for granted. Ideally, institutionalized interpreting services would be established within and across schools that allow students to circumvent the conflicting roles, especially when brokering about themselves.

Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The research project "ZwischenWeltenÜberSetzen" ("Translating Wor(l)ds") was funded by the Austrian Federal Ministry for Science, Research and Economy in the funding scheme "Sparkling Science", Grant number SPA06-229.

ORCID iD
Nadja Thoma https://orcid.org/0000-0001-9084-9638 Notes 1. The project "ZwischenWeltenÜberSetzen" ("Translating Wor(l)ds") was funded by the Austrian ministry of Science and Education within the funding scheme "Sparkling Science," Nr. SPA06-229. 2. All personal names have been anonymized. 3. In the presentation of the data, we have refrained from linguistic corrections. We understand norm deviations not as deficits but as snapshots in the language acquisition process. These are sometimes also creative new creations, which mark searching, learning and discovering movements in and between languages and reveal a learning process in which learners acquire a language with its structures and rules. 4. "Auf einmal" means "suddenly" in German, while "einmal" means "once." For reasons of content, we have opted for the translation "once." 5. In this formulation, as in other places in the material, we noticed that Haadiah, as well as some of her classmates, are concerned with the representation of women and men. 6. Haadiah refers to the Common European Framework of Reference for Languages. 7. "Euch" and "eure" are plural forms of "you/r" in German. 8. "Jugendcollege" ("Youth College") is a modular educational program in Vienna for youths and young adults between the ages of 15 and 21 who are no longer of compulsory school age. The target groups were originally persons granted asylum and beneficiaries of subsidiary protection as well as asylum seekers, and later also other groups that meet barriers on their educational pathway. The "Youth College" offers basic education through various modules on different levels (German, English, mathematics, communication technologies and elective modules like health and social issues) as well as social and educational counseling (City of Vienna, 2018). 9. The gender of the person who is called "meine Betreuer" ("meine" is a female pronoun, while "Betreuer" is a male noun) is not clear. There are indications for both variants.