Language Comparison as an Inclusive Translanguaging Strategy: Analysis of a Multilingual Teaching Situation in a German Primary School Classroom

This chapter highlights the practical implementation of language comparisons in a multilingual classroom. Following the key-incident-approach, we analysed a specific teaching situation in a German primary school in order to point out how the teacher shapes a translanguaging strategy that includes all children and their knowledge, and how she deals with the associated challenges.


Introduction
In educational projects all over the world, it has been stated that educators should encourage their students to use their entire linguistic repertoire to think, reflect and extend their inner speech (for an overview see Hélot 2013). The present chapter focuses on language comparison as a didactical core element of multilingualism-sensitive classrooms. The use of language comparisons has been promoted in the context of Translanguaging Pedagogy, and we will trace it back to Hawkins' (1984) ideas of "Awareness of Language". Although language comparisons play an important role in the conception of didactic approaches to multilingualism, their practical implementation in the classroom can qualify as a research desideratum. In this contribution, we address this desideratum by analysing a classroom situation. The aim of the analysis is to work out how a teacher shapes the language comparison in a multilingual learning group and how she deals with the associated challenges. One challenge frequently mentioned with regard to multilingual classrooms in migration societies is the fact that the children use many different family languages about which the teacher herself knows very little. The analysis in this article is led by the question as to how the teacher can include the linguistic knowledge of the children in class and use it for joint language learning in the group despite this challenge. The paper is based on data from a research project ("Multilingualism as a field of action in intercultural school development", short MIKS-project) in which the teaching staff of primary schools was assisted and supported in implementing multilingual didactic approaches in the classroom. An important precondition for language comparisons in multilingual settings was met: teachers in the MIKS-project schools were open and willing to engage with the pupils' home languages.

Translanguaging Pedagogy
To analyse language education in multilingual classrooms, it is useful to adopt a pedagogical concept of translanguaging that involves all languages in a classroom, as the one offered by Ofelia García and others: Translanguaging can be defined as a process by which students and teachers engage in complex discursive practices that include all the language practices of students in Language Comparison as an Inclusive Translanguaging Strategy … order to develop new language practices and sustain old ones, communicate appropriate knowledge, and give voice to new socio-political realities by interrogating linguistic inequality. (García and Kano 2014, p. 261) As we know, the inclusion of children's multilingual language practices into class is by no means the norm but rather the exception in most schools worldwide (see Hélot 2013). Against this background, teachers who want to develop translanguaging pedagogy in multilingual classes are facing major challenges. It is not merely a matter of including all the language experiences of all students; in addition, new language practices should be jointly developed in the multilingual learning group-involving all children in the communication-with the aims of learning something and of questioning language hierarchies in social contexts. It was not without reason that one of the MIKS project's focal points was the support of school development processes. For a single teacher, it is hardly possible to implement translanguaging pedagogy in this comprehensive sense. Rather, it needs the cooperative development of a "multilingualism-friendly school culture", as Katrin Huxel observes in her case study of a MIKS project school (Huxel 2018). This corresponds to the categorization of translanguaging pedagogy as "transformative" pedagogy (García and Wei 2014).
With regard to the lessons in every single classroom, many questions remain open as well. Some researchers point out that translanguaging among children is most likely to occur naturally once a school decides to be open to all children's languages (Creese and Blackledge 2015;Vogel and García 2017). This expectation is based on the finding that multilingual children are accustomed to access their full linguistic repertoire for communicative purposes, even in educational contexts. Accordingly, teachers can support the learning processes of individual children already by permitting all languages in class and by welcoming multilingual activities. This is not enough, however, if we adopt the comprehensive understanding of translanguaging pedagogy outlined above and the premise that teaching should support co-constructive learning in the group. Following Suresh Canagarajah (2011), we therefore base the analysis of classroom interaction in this paper on an understanding of translanguaging as a "social accomplishment" (p. 4). In this understanding, translanguaging "is an interactive achievement that depends on aligning one's language resources to the features of the ecology to construct meaning" (p. 5). The teacher is therefore faced with the task of shaping "translanguaging as a form of social practice" (p. 5) in the classroom, drawing on all children's language experience and knowledge, and rendering it useful for joint learning in the group. With this in mind, we view translanguaging as an inclusive teaching strategy. The professional actions of a teacher who implements translanguaging as an inclusive teaching strategy can be characterized by using three core components described by García et al. (2017, p. xii): "stance", "design" and "shifts". According to the authors, a "translanguaging stance sees the bilingual child's complex language repertoire as a resource, never as a deficit" (p. xiii). The second component, design, relates to planning and organizing a multilingualism-sensitive classroom. The third component, shifts, means that the teacher is essentially prepared for dynamic multilingual acting and thinking in the classroom and is willing to take up the children's multilingual knowledge spontaneously. These core components are-as the analysis in this paper will showsuitable to capture the characteristics of teacher action in shaping a multilingual teaching situation. When shaping such a situation according to our understanding, all students should get the opportunity to develop cooperative metalinguistic awareness and mutual understanding, to recognize the value of all languages, and to jointly challenge language hierarchies. These goals of translanguaging pedagogy can be traced back to the concept of language awareness, as will be illustrated below.

Language Awareness
In the scientific literature on Language Awareness, the following definition attributed to B. Gillian Donmall (1985) is popular: "Language Awareness is a person's sensitivity to a conscious awareness of the nature of language and its role in human life" (as cited in Garret and James 1993, p. 109). This concept has been used for decades as a basis to develop teaching that combines linguistic education in the school and majority language, in foreign languages, in the student's home languages and in minority languages; today, it is of outstanding importance for approaches to multilingualism didactics. Eric Hawkins, who developed the concept in the 1970s, is considered the founder of Language Awareness. In the scientific discussion on foreign language didactics since the 1960s, the focus has been increasingly on teaching communicative competences, contrasting it to didactics that focus on teaching grammar. In this context, Hawkins' notion of "Awareness of Language", which he describes in his book with the same title (1984), was special and innovative. From Hawkins' point of view, one important goal of language teaching is to reflect language, thus rendering students' implicit knowledge about language accessible and making it explicitly the subject of the lesson. Unlike didactic approaches that focus on communication skills, "Awareness of Language" is about developing metalinguistic knowledge: "our new curriculum topic will seek to give pupils confidence in grasping the patterns in language" (Hawkins 1984, p. 4). In addition, Hawkins developed the concept not only from a linguistic but also from a pedagogical perspective: We are seeking to light fires of curiosity about the central human characteristic of language which will blaze throughout our pupils' lives.
[…] Above all we want to make our pupils' contacts with language, both their own and that of their neighbours, richer, more interesting, simply more fun. (Hawkins 1984, p. 6) Our request to use linguistic diversity in the classroom for joint learning in the group and to design inclusive situations can also be traced back to Hawkins' thoughts on "Awareness of Language" (p. 3). According to Hawkins (1984), when children "come from different language backgrounds" and "tell one another about their language experiences", all children "can feel that they have something to contribute. Experiences that they share […] can, if properly handled, unite children". Hawkins' ideas live on programmatically in current concepts of multilingualism didactics. However, there are still hardly any answers to the question as to how multilingualism in the classroom should be "properly handled" in order to make joint learning possible. This question arises in particular with regard to learning groups in which the children have experiences with a great variety of languages the teacher is barely familiar with. Our analysis of a teaching situation in this paper aims at this desideratum.

Language Comparison
Comparing linguistic forms of expression is, according to Hawkins (1984), the core element of a didactic that follows his concept of Awareness of Language: A contrastive study, at an appropriate level, of […] patterns [in language] with those met in other languages (foreign languages studied in class as well as the ethnic minority languages of classmates) will be part of […] growing insight into the way language works to convey meaning. (p. 5) In research, foreign language didactics were the first to come up with language comparisons as a teaching strategy-a didactic approach that soon was discussed controversially (for an overview see Ticheloven et al. in press, for an example see Mehlhorn 2011). Due to the complexity of language comparisons, it is not easy to clearly outline the subject of a language comparison in the classroom and to use the comparison for linguistic learning. Nevertheless, language comparisons play an important role in recent concepts of multilingualism didactics. But the scientific study of concrete experiences in the classroom has just begun. The scientific discussion is often about challenges, obstacles, and the reasons why language comparisons are rarely carried out in multilingual learning groups (see Bredthauer 2019;Ticheloven et al. in press). The unease of teachers when dealing with languages they do not understand and about which they know nothing is an important issue. One of the few empirical studies on language comparisons in "hyperdiverse-multilingual classes" (p. 5) in German schools (Bredthauer 2019) is based on six expert interviews with teachers who teach language subjects at secondary level and declare that they regularly conduct language comparisons to incorporate the knowledge of their multilingual students. These teachers take the plunge of accepting that the classic role allocation in school changes when the students act as experts in their languages in class and when there are things the teacher does not know. Bredthauer elaborates, however, that the surveyed teachers take on an important "accompanying and moderating function" (p. 15) when guiding language comparisons, a task the respondents perceive as highly demanding (2019, translated from German). Hawkins (1984) attempts "to challenge pupils to ask questions about language" (p. 4). In light of the role that teachers play in including and comparing the students' languages, we would like to add that it can be equally important to challenge teachers to ask questions about language. The fact that language comparisons in multilingual classes can stimulate joint learning not only among the children, but may also include the teacher as a learner will be depicted in the classroom analysis in this paper.
The following questions, derived from theory and state of research, guide the analysis of a multilingual classroom situation: How does the teacher include the children's knowledge? To which extent is the teaching situation inclusive? In which ways does joint learning take place in the group? What are the goals of language comparison and what is the subject of comparison?

Research Project and Methods
The project "Multilingualism as a field of action in intercultural school development" (short MIKS-project) was funded by the BMBF (Federal Ministry of Education and Research) for six years (2013-2019). 1 MIKS includes a training programme for professionalization and school development, which aims at developing constructive approaches to incorporating students' home languages into the classroom (Fürstenau 2016). The training programme was developed in three primary schools during the first phase of the project (2013-2016) and then implemented and scientifically investigated at 17 project schools in North Rhine-Westphalia from 2016 to 2019. It comprises training days with a focus on the development of psycholinguistic and sociolinguistic knowledge and of possible multilingual teaching strategies, trial phases in which teachers experiment with multilingual strategies in their own lessons, and guided reflection exercises (for details see Lange 2019). During the trial phases, teachers could choose or design ways of multilingual teaching that fit their own classroom and teaching style. In this regard, according to Mechthild Dehn the MIKS project is an "impulse project" (as cited in Fürstenau 2019): Teachers receive knowledge input and are supported in designing multilingual lesson units, but at the same time, they develop their own multilingual teaching practices. The scientific investigation of the MIKS project included both qualitative (i.e. participant classroom observation, interviews with teachers and head teachers, observation of training days) and quantitative methods (survey questionnaires before and after the training programme). All project schools had a linguistically diverse school population with monolingual and multilingual students born in Germany as well as newcomer students. From the 17 project schools, four focus schools were selected for participant observation based on their explicit interest to be a focus school. Results from the scientific study of the MIKS project show that multilingual classroom situations differ greatly, e.g. depending on whose languages are incorporated, by whom, for which reasons and in what manner (see Dlugaj and Fürstenau 2019;Gilham and Fürstenau 2019). Project school teachers reported to use language comparisons more frequently than other multilingual teaching practices (Ticheloven et al. in press). Based on an analysis of observation protocols, we distinguished three levels of teacher initiative (Ticheloven et al. in press): responding to student-initiated comparisons (1), improvising spontaneous comparisons (2), and planning lessons that include comparisons (3). In this paper, we conduct an in-depth analysis of a classroom situation on the micro level. The data basis is one of 92 protocols from classroom observations in which a teacher conducts a language comparison in a planned manner. The situation illustrates a successful handling of the challenges outlined above. The teacher creates an inclusive classroom situation in which individual multilingual children can contribute their knowledge and, at the same time, the group can learn jointly. In addition, the teacher ensures that all children can understand the goal of their joint conversation about language and the subject of comparison. When choosing the teaching situation for analysis, we thus followed a key incident approach (Erickson 1977;Wilcox 1980). The objective was to "write it up so that others can see the generic in the particular, the universal in the concrete, the relation between part and whole […]" (Erickson 1977, p. 61). The results shed light on micro-level teaching patterns that represent and corroborate the bigger and more abstract theory of comparing languages in a multilingual setting.

Analysis: Comparing Languages in a Multilingual Primary School Classroom
The analyzed teaching situation was observed at Hollyhock School, one of the MIKS focus schools. The school is located in the center of a German medium-sized city. In the school year 2017/2018, about 230 students attended the school, 57% of which, according to school management, use another family language besides German and 70% have a migrant background. At the school there are heritage language classes for the languages Albanian, Arabic, Greek, Portuguese, and Turkish. According to the headmistress, the school participated in the MIKS project because they feel that multilingualism and German as a second language classes are challenging and that an existing school development group should be supported in its work. We analyse a teaching situation in an inter-year class 1/2 where the children are six to eight years old. Topic of the lesson is the lexical category verbs. At the beginning of the lesson, the teacher Ms Steffens asks the children if they know what verbs are. Some children know that verbs are "doing words" ("Tuwörter"). Ms Steffens lets the children tell her verbs in their basic form and in first person singular and writes them on the board: spielen-ich spiele [play-I play], laufen-ich laufe [run-I run], etc. Then she draws the children's attention to the common feature of the German verb form in first person, i.e. the ending -e 2 : 'What do you notice now? What do these [she stresses the word and points to the first person forms] words all have in common?' She calls on a student. The student stands up, points to the ending of each word on the board and says, 'e, e, e, e.' Ms Steffens nods and says: 'There are many children in our class who speak a language other than German. Can any of you say spielen [play] in another language? How do you say spielen in your language? I can say it in English. In English you say play. Which other languages do you know?' As an introduction to the subject, Ms Steffens initiates a ten-minute conversation in which the teacher and the children jointly compare languages. Subject of comparison are the verb forms in first person singular. With the question "Which other languages do you know?", the teacher addresses the whole learning group. During the following conversation, she never addresses individual children with specific expectations. Rather, it is up to the children to decide which linguistic knowledge they want and are able to contribute. During the conversation, the children introduce seven languages of their own accord: Arabic, Turkish, Persian, Albanian, Bulgarian, Polish, and Kurdish. The teacher develops the blackboard presentation by letting the children dictate verb forms and writing them down phonetically (see Fig. 1: 2nd line Arabic, 3rd line Persian, 4th line English, 5th and 6th lines Bulgarian, and 7th line Kurdish). The teacher herself adds the English form "I play" on the blackboard once none of the children raises the hand any more. Thereupon a boy contributes Bulgarian and writes "I play" in Bulgarian using Cyrillic script on the board after the teacher wrote the phonetic version. In the course of the conversation, other children point out that while they may speak Turkish, Albanian and Polish, they do not know the verb forms; therefore, these languages do not appear on the blackboard. In the protocol is noted, for instance, "Another student says somewhat desperately: 'I can speak Polish, but I don't know what it means'". The teacher does not comment on this. Below, we will analyse in detail chosen passages from the observed sequence with regard to the teacher's pedagogical approach and her creation of an inclusive discussion in class.
Following the introductory question "Which other languages do you know?", the teacher picks a student who raises his hand: He says an Arabic word. She asks: 'Tell me your language.' He replies: 'Arabic.' She repeats: 'Arabic? And how do you say this in Arabic? Say it again.' He repeats the word to play again in Arabic. She imitates him and asks grinning: 'Do I pronounce that correctly?' The boy and the other Arabic-speaking students shout 'no' and laugh. She laughs, tries it again and says: 'Oh man, that's difficult. Can you pronounce that?' The students shout 'yes', laugh, and repeat the word loudly a few times. She repeats it with them and the students try to help her with the pronunciation. She tries it a few more times and then says: 'Now, of course I can't speak Arabic. Do you also know how to say I play in Arabic?' She looks briefly at an orange note on her desk. 3 An Arabic-speaking girl puts her hand up and says two Arabic words. Another student corrects her pronunciation. Then Ms Steffens says: 'I've had it written down for me; that's how it is written.' She writes innaa aleabu 4 on the board below ich spiele, then looks at all students and says: 'This is not Arabic writing. These are the German letters.' She turns to the Arabic-speaking boy who answered first and asks him to say the word again. He repeats it; she repeats it as well, is briefly corrected in her pronunciation by the students and then corrects herself again.
The Arabic-speaking children contribute their knowledge, and the teacher positions herself as a learner ("Do I pronounce that correctly?", "Now, of course I can't speak Arabic"). In advance, the teacher had asked someone to write down "I play" in Arabic using Latin script ("orange note"). But that was her only preparation for this language comparison, and the spelling is corrected by the children in the further course of the language comparison (see below). Not only Arabicspeaking children, but all children participate and correct each other and the teacher when it comes to the pronunciation of Arabic ("'Can you pronounce that?' The students shout 'yes', laugh and repeat the word loudly a few times"). Then, the teacher assumes again the teaching role by directing the children's attention to the writing ("This is not Arabic writing. These are the German letters").
Other children raise their hands and introduce various languages. The observation protocol indicates that in this process, the children and their teacher jointly discover the linguistic knowledge present in the group. It becomes clear that all parties are very interested in this knowledge, as is shown e.g. in the following situation: Ms Steffens writes manbasi miconam 5 on the board. The pupil Zahir Zia nods contentedly.
[…] She turns to him: 'That means I play? In which language?' He looks startled, points with both hands to his upper body and says in a high voice: 'In my language.' Ms Steffens laughs and says: 'What is your language?' He says again in a high voice: 'In Persian.' An exclamation of admiration 'oaah' can be heard from some children.
Multilingualism and engaging with different languages do not yet seem to be a normal part of class in this group, since the children meet the knowledge of Persian with "admiration", and the teacher did not know that Zia Zahir speaks Persian. She knew that some children in the group speak Arabic, but the family languages of other children seem to be new to her. This also applies to Julian. Only when no more children raise their hands and the teacher contributes her own English language skills (she writes "I play" on the board), Julian pipes up: Julian says he knows it in Bulgarian. Another student loudly calls his name and points to him: 'Julian can do that.' Ms Steffens looks at him in surprise: 'Julian, Bulgarian! I play?' He quickly says the Bulgarian words for it. It is followed by a 'Woaah' of admiration that sweeps through the classroom.
Julian dictates, and the teacher writes "as igraje" on the blackboard. Obviously, the children understood that in the current situation-unlike in most classroom situations-knowledge in languages other than German is required. The situation is inclusive since all children focus on the multilingual knowledge available in the group ("Julian can do that."). "Admiration" may also be evoked by the appreciation in the classroom context. Julian is further able to write "I play" in Bulgarian on the blackboard using Cyrillic script, and the teacher as a moderator can contribute her knowledge of various scriptures: "Look, Julian writes in Cyrillic letters. Have you ever seen that?" Some students exclaim: "Like cursive handwriting." She nods and says: "Yes, it looks a little bit like cursive handwriting." Another student says: "Our writing is different, too." This creates a new occasion for comparison, this time on the level of writing. Julian is not the only one in the group who knows about different writing systems ("Our writing is different, too"). This is probably a knowledge that has not played a major role in the classroom so far but becomes meaningful now by way of the inclusive setting.
In a next step, Ms Steffens wants to write the first person forms for another verb-to run-on the board: She says that now they will also conjugate the verb to run multilingually and starts again with Arabic. She asks: 'Can anyone say I run in Arabic?' A student puts her hand up and says two Arabic words. Another student heckles: 'No, that means walk.' Ms Steffens: 'It means I walk? And run is yet another word?' She frowns. The Arabic-speaking students think and look at each other, but do not answer. Ms Steffens says: 'But ana definitely. Does it mean I? Does ana mean I?' She writes the word ana on the board and asks: 'and run?' A student says: 'ana irkot.' The other Arabic-speaking students repeat it and Ms Steffens writes 'ana irkot' on the board. I am surprised how attentively and quietly the other students listen. Ms Steffens says: 'Look, I'm wondering right now. That means [she points to the word innaa alaebu on the board] I play and that means I run [she points to ana irkot]?' The students nod 'yes'. Ms Steffens asks: 'Why is here [she points to the word innaa] another word for 'I' than here [she points to the word ana]?' An Arabic-speaking student says: 'That's wrong.' Ms Steffens: 'This is misspelled? Is it also ana?' The student nods and Ms Steffens says: 'Who knows how that is written. We can't write Arabic. But then this should also be an ana.' She changes the word innaa to ana. The students nod.
The protocol contains various indications that in this situation everybody involved is stimulated to think about language while the teacher and several Arabic-speaking children work out together how to write "I run" in Arabic using Latin script on the blackboard. The joint reflection begins on the semantic level as the children orally translate "I run" into Arabic (running vs. walking). The teacher thinks ahead ("'It means I walk? And run is yet another word?' She frowns."). The teacher lets the children go on pondering by not insisting on a quick answer but first writing down the part of the translation the children seem to agree on: ana. "Does ana mean I? […] and run?" Ms Steffens asks. Now a student dictates "ana irkot" and other children agree by repeating the wording. The pondering has obviously led to a joint result. Even the children who cannot translate 'I run' into Arabic seem to be pondering (the observer is "surprised how attentively and quietly the other students listen"). In the next step, the teacher once again pursues the matter of comparing languages and discovering regular patterns. She draws attention to the translation of the word I into Arabic, or rather to its phonetic representation in Latin script, because there are two different spellings, "innaa" and "ana" on the blackboard. Although it was the teacher who introduced the spelling "innaa" ("orange note", see above), a student now dares to say, "That's wrong." The teacher, in turn, accepts the student's correction without questioning it ("then this should also be an ana") and corrects the spelling on the blackboard. The teacher arranges the situation in such a way that the children have the last word, thus honouring the children's knowledge. At the same time, she observes something the group has in common, which is that apparently none of those present has learned to write Arabic script ("Who knows how that is written. We can't write Arabic."). In the role of teacher, Ms Steffens draws the attention of the class to the significance of different writing systems. However, with the spelling "ana" established by the children, everyone seems to be satisfied ("The students nod."). With the aim of making different languages visible for comparing them, the children incidentally accomplish the feat of assigning sounds of the Arabic language to Latin writing.
After the group has agreed on Arabic, a student speaks up and names a language that has not yet been considered, but which is spoken by many children at Hollyhock School: Kurdish. "Ms Steffens: 'Oh, we forgot Kurdish. Do you know it?' Another Kurdish-speaking girl laughs and says, 'Anyone can do that'." The teacher's wording in the first person plural-"We forgot Kurdish"-can be seen as a further inclusive approach to multilingualism. By using we, the teacher makes it clear that Kurdish as a language is significant for all in the group, not only for those who speak Kurdish at home. Quickly the girls dictate the teacher I play and I run in Kurdish. Subsequently, Zahir Zia dictates I run in Persian-man midoam-and explains that man means I. 6 The teacher takes up the example to explicitly compare the first person forms: joint learning. She draws the students' attention to both similarities and differences between the verb forms: As a last example, she takes the Bulgarian word and asks Julian for the translation of I run. He says it and she writes as titscham on the blackboard. She asks again: 'What is now the same in Bulgarian?' Nikolas raises his hand and says that the first word as recurs. Ms Steffens nods and underlines the as; then she asks about the ending: 'Is the ending the same, too?' The students shout: 'No.' A student says: 'No, but as means I'. Ms Steffens: 'That's possible. But as you can see, endings are not necessarily the same in every language. Here we have different endings [she points to the Arabic translations], in German the ending is always the same and in Persian it is the same, too. But in Bulgarian it is different. Thus, it can differ from language to language.' Although Julian is the only student in the group who speaks Bulgarian, the teacher clearly addresses all children with the question "What is the same in Bulgarian?" Nikolas recognized that the word "as" recurs. Then Ms Steffens asks the children to look at the endings of the Bulgarian first person forms written in Latin script on the blackboard ("Is the ending the same, too?"). The protocol states that "The students [i.e. all and not just the boy who speaks Bulgarian] shout: 'No.'" The student who then objects "but as means me" refers to the meaning of a word in a language that is new to her and shows what she has learned. The teacher answers "that's possible", making it clear that she herself is not the one who speaks Bulgarian, and continues the joint reflection on language. Finally, the teacher uses all the verb forms gathered on the board for a short talk on the subject of comparison-verbs in first person singular-and points out with regard to the observed regularities and discrepancies that "it can differ from language to language." The gist of what all children should learn in this German lesson is also explicitly stated by Ms Steffens: "In German, the ending is always the same."

Conclusions
We chose the analysed teaching situation as a key incident because it is productive for answering the research questions formulated above. The analysis provides guidance on how a teacher may create an inclusive classroom situation in which individual children can contribute their knowledge and, at the same time, the group can learn jointly. Furthermore, we can see in the analysed situation how a teacher may ensure that all children understand the goal and subject of language comparison and jointly develop language awareness. To conclude, we will map out how the characteristics of the teacher's actions can be captured with the three core components of translanguaging pedagogy devised by García et al. (2017, p. xii): "stance", "design" and "shifts". How does the teacher in the analysed teaching situation succeed in activating the multilingual knowledge of first-and second-graders for language comparison and in making it the subject of conversation? It is noteworthy that the teacher addresses her questions consistently to the whole group, not to individual children. It is up to the children themselves to decide whether they want to contribute their knowledge or not. We attribute the lively participation of the children not least to the stance of the teacher: The way in which the teacher treats the children's statements displays a permissive, appreciative, curious and error-friendly stance. By way of genuine questions and great interest, the teacher seems to encourage the children to talk about their family languages, even though they have not been used to it in the classroom context. Besides, the teacher seems to literally infect all children in the group with her interest in languages, which helps to develop a conversation in which they actually think together about language. This becomes apparent, for example, in the fact that children express and discuss observations about their classmates' languages which they do not understand themselves. The joint reflection is supported by the teacher's moderation, who takes the children's contributions seriously and picks up on them, repeatedly making it clear that the group ("we") has a common interest in all represented languages. Thus, the teacher ensures an inclusive classroom situation in which the group can learn jointly not only by her stance, but also by explicit announcements. Hawkins (1984) assumes in his conception of "Awareness of Language" that "[a]ll can feel that they have something to contribute" (p. 3). This is obviously true in the analysed situation, regardless of whether the children grow up monolingual or multilingual.
The analysis has shown that, during their conversation about languages, teacher and children jointly discover the linguistic knowledge existing in the group. In the process, the knowledge of individual children receives great recognition-even "admiration", according to the observation protocol. We therefore assume that addressing the existing multilingualism is a new experience for the learning group. The analysis thus provides an insight into a classroom situation, in which the transformative potential of translanguaging pedagogy becomes apparent. According to Vogel and García (2017), translanguaging pedagogy even has "the potential to transform relationships between students, teachers and the curriculum" (p. 10). Our analysis of the teaching situation contains references to such a transformative process, since it is indeed true that "teachers and students learn from each other, and all language practices are equally valued." (Vogel and García 2017, p. 10). The described transformations take place as part of a school development process. The teaching staff of Hollyhock School, where the teaching situation was observed, decided to address multilingualism as a field of action for school development by participating in the MIKS project. The teachers try out multilingual didactic approaches in the classroom and thus gain new experiences.
A special experience for the teachers consists in planning and designing lessons in which contents-more specifically: languages-are addressed, about which they know very little themselves. This makes a careful design no less important, as the analysis has shown. The observed teacher guides the language comparison in a well-planned manner. In the analysis, we mapped out how the teacher in her role as a moderator ensures that all children can understand the goal and the subject of language comparison. The teacher does not lose sight of the subject of the lesson (verb forms in first person singular) and repeatedly tells the children explicitly what they need to understand (regularities and irregularities). Ultimately, this multilingual language comparison is a planned part of German language class, and the goal is to understand verb forms of the German language. Nonetheless, the art of teaching consists in maintaining openness to the children's knowledge and thoughts despite the plan, and this is where the shifts come into play. García et al. (2017) describe it as a challenge in the classroom "to follow the dynamic movement of the translanguaging corriente" (p. xiii). Although the observed teacher has a plan, she has to deal with the fact that it is completely unclear which linguistic knowledge the children will contribute and what the linguistic basis for the comparison of verb forms will be. The teacher repeatedly talks with the children about questions without knowing the answers. The "shifts" consist in the "moment-by-moment decisions" (García et al. 2017, p. xiii), e.g. when a child dictates something in a language the teacher does not understand, and she writes it in phonetic spelling on the blackboard. Such shifts are only possible if the teacher takes the children's knowledge seriously and is prepared to leave safe ground. Findings from the MIKS project indicate that this willingness among teachers varies considerably depending i.a. on their own experiences with language learning (Gilham and Fürstenau 2019).
It is a task for further research to investigate how teachers in multilingual teaching situations deal with linguistic uncertainties, and which strategies they pursue when applying multilingualism didactics in order to use languages they know very little about. To prepare for a lesson, the teacher may of course read up on languages spoken by children in her class as part of the design. This approach played only a minor role in the teaching situation analysed above (the teacher had Arabic words written in Latin script on a slip of paper in advance). Other lessons we observed at the MIKS focus schools were based on, e.g., cooperating with multilingual parents who contribute their linguistic knowledge to lessons. In any case, multilingual classroom situations have in common that insecurities on the teacher's side can hardly be avoided, that is to say, moments of not-knowing will occur. This conflicts with the self-image of some teachers. Therefore, studies on how teachers in multilingual teaching situations can deal constructively with insecurity and not-knowing will be useful to advance research on multilingualism didactics.