Translanguaging Strategies in Superdiverse Mainstream Norwegian ECEC : Opportunities for Home Language Support RESEARCH

In Norway, 92% of all children between 1 and 5 attend early childhood education and care (ECEC), and 18% of these children are minority language speakers. The Framework Plan for Content and Tasks of Kindergartens (Ministry of Education, 2017, p. 24) states that ECEC staff shall ‘help ensure that linguistic diversity becomes an enrichment for the entire group of children and encourage multilingual children to use their mother tongue while also actively promoting and developing the children’s Norwegian/Sami language skills’. In this paper, we present a study of how home language (HL) support takes place within the context of Norwegian ECEC, focusing on the strategies used by the staff to promote HL use. After analysing 26 narratives from practice, we found that the most common strategies employed were initiating activities that encourage HL use, facilitating metalinguistic conversations and consulting/involving language experts. The strategies available depend on contextual factors, such as the number of children present and the languages spoken by both children and staff. The HL support strategies are discussed in light of the interplay between teachers’ language ideologies, planned actions and spontaneous responses in situations where children’s HLs are involved inspired by the theories in García, Johnson and Seltzer’s study (2017). Translanguaging Strategies in Superdiverse Mainstream Norwegian ECEC: Opportunities for Home Language Support RESEARCH


INTRODUCTION
In this study, we discuss how home languages (HLs) are supported in monolingual mainstream early childhood education and care (ECEC) centres in the Oslo region. In Norway, mainstream ECEC centres provide preschool education for children aged 0-6 years. Although ECEC attendance is voluntary, the majority (92%) of children in Norway attend ECEC. Many ECEC centres in urban areas are characterised by superdiversity (Budach & de Saint-Georges, 2017), with a large number of languages represented in each classroom/group.
In Norway, ECEC centres typically group children by age, so that toddlers (0-3-year-olds) are separated from older children (3-6-year-olds); however, some centres include mixed age groups. There are usually between 9 and 24 children in each group and 3-4 members of staff. Some of the staff hold a bachelor's degree in early childhood education (ECEC teachers), while staff with no relevant or only lower level education are referred to as teacher assistants. All ECEC centres in Norway, both municipal and private, are required to follow the Framework Plan for the Content and Tasks of Kindergartens (Ministry of Education and Research, 2017). The Framework Plan promotes a child-centred approach to teaching and learning as a basis for all-round development. ECEC activities are characterised by a combination of free play and play-based learning. Language learning is usually considered in terms of learning through participation in play and everyday activities. However, more formal language teaching activities led by staff also take place, especially for children who require extra language support.
While the main language in Norwegian ECEC centres is Norwegian, the Framework Plan (Ministry of Education, 2017) states that ECEC staff shall support children's multilingualism by promoting their development in both their HL and Norwegian or Sami, as well as ensuring that 'linguistic diversity becomes an enrichment for the entire group of children ' (p. 24). However, relatively little is known about how this requirement is met in practice (see also Alstad, 2013;Giaever, 2018;Lindquist, 2018Lindquist, , 2019, particularly in ECEC centres characterised by superdiversity. In our study, we explore how ECEC staff encourage and support the children's HLs across various formal and informal situations. By home languages (HLs), we refer to all languages other than Norwegian that are used in the children's homes (see Connaughton-Crean & Ó Duibhir, 2017). In the following, we firstly present theoretical perspectives and previous research on HL support in mainstream ECEC centres, before describing the methods used in this study, and finally, going on to analyse and discuss our data.

TRANSLANGUAGING AND HL SUPPORT IN MAINSTREAM MONOLINGUAL EDUCATIONAL CONTEXTS
Maintenance of the HL(s) is a constant concern for many bi-/multilingual families raising their children in communities where the HLs have minority status (see also Hashimoto & Lee, 2011). Research has shown that starting formal education in a majority language can lead to reduced HL use in bi-/multilingual children due to the dominance of the society language, monolingual instructional practices in education where HL(s) are not seen as relevant and valued, and HLs having lower status compared to the language of the society/instruction (Cummins, 2005;De Houwer, 2017Jin et al., 2017;Wong Fillmore, 2000). Although Cummins (2005) more than a decade ago called for action to re-consider how HLs can be supported within mainstream education, not much research has been done on strategies for HL support in mainstream settings. In this paper, Cummins points out that teachers should acknowledge and appreciate the children's proficiency in their HL(s) in the classroom so that the children can benefit from the use of bilingual strategies, e.g. using multimedia resources or producing dual-language books.
Strategies for supporting HLs have been studied in some other contexts: in the family, community schools, bilingual educational programmes and in the work of bilingual support teachers. Kwon's (2017) study on HL support in the family, for instance, identifies three strategies used by parents with their children to support HL: use of transnational media, frequent visits to home countries, including children's attendance in public schooling there, and using literacy resources brought from the home countries. Research on community school settings and on bilingual education shows that even in bilingual educational contexts, monolingual ideologies are so prominent that they may impede effective language learning (Creese & Blackledge, 2010; A number of studies on bilingual preschool education have documented different strategies used by teachers to mediate communication where several languages are used (Gort & Pontier, 2013;Gort & Sembiante, 2015;Schwartz, 2018;Schwartz & Asli, 2014;Schwartz & Palviainen, 2016). These studies reveal flexible and strategic use of two languages to negotiate and construct meaning and support the children's bilingual development. In a study of bilingual teachers' language practices in one bilingual English-Spanish preschool, Gort and Sembiante (2015) show how the teachers perform bilingually in their collaborative engagement in the classroom, while at the same time adhering to the preschool's parallel monolingual language policy. Although each teacher performed monolingually in the designated languages, they managed together to create a dynamic multilingual environment by using bilingual recasting of each other's turns, accepting the children's responses in both languages and meeting the language preferences of individual children, as well as positioning themselves as models for flexible bilingual language use. Palviainen et al. (2016) interviewed bilingual teachers about their use of multiple languages in bilingual preschools in three different contexts (Finnish-Swedish, Russian-Finnish and Arabic-Hebrew). The five teachers that participated in the study mentioned the following language practices in their reflections: the flexible use of two languages in spite of a strict language separation policy, responsible code-switching in certain situations rather than direct translation, contextual and linguistic support to enhance understanding, adjustments for individual children, and role-modelling for bilingual language use.
Although traditional models of bilingual education and HL support in community schools and in the family provide us with knowledge on some HL support strategies, these educational settings typically require an overlap between the children's and to teachers language repertoires, which is often unattainable in superdiverse groups (Vandenbroeck, 2018). As the overlap between the language repertoires of staff and children within Norwegian ECEC groups tends to be rather limited (Romøren et al., 2021), teachers need to look for alternative approaches to support children's HLs in such settings. Some of the recent approaches to multilingual education in superdiverse settings take Cummins' proposal for action (2005) a step further towards a multilingual turn in mainstream education, and consider languages brought to the classroom by both children and teachers to be a resource for learning and for supporting children's functional bi-/multilingualism (Cenoz, 2017;Cenoz & Gorter, 2015;Cummins, 2019;García et al., 2017, García & Li Wei, 2014Sierens & Van Avermaet, 2014). These pedagogical approaches promote the use of multiple languages in the classroom to help children develop multilingual and multicultural identities. Although a number of different terms have been used for such approaches (see an overview in Cummins, 2019), the most popular concept is translanguaging. Translanguaging has been used to describe multilingual individuals' flexible and dynamic language practices, as well as pedagogical approaches that deliberately employ multilingual repertoires in educational contexts (García et al., 2017). Dynamic language practices thus entail that teachers and children draw on their entire language repertoires and manage to successfully navigate the multilingual terrain (García et al., 2017), which also includes 'shuttling' between languages (Canagarajah, 2011) even when participants do not necessarily share linguistic repertoires.
In translanguaging pedagogies, teachers take on a translanguaging stance by considering the children's multilingual repertoires as central to their learning. They intentionally design learning activities that draw on the children's full linguistic repertoires and flexibly shift to the children's linguistic needs during learning activities (García et al., 2017). These three pedagogical strandsstance, design and shifts -are key aspects of the translanguaging pedagogy framework. García et al. (2017, p. 28) point out that these three strands 'are interrelated and form a sturdy but flexible rope'.
A number of studies from ECEC contexts provide evidence that translanguaging can be used systematically as a pedagogical tool, irrespective of the teacher's language skills. For example, de Olivera et al. (2016) describe how a monolingual English-speaking teacher gradually learns Spanish from her Latino students and uses it as a pedagogical tool to support their literacy development. Their study shows how the teacher develops and uses her emerging knowledge of Spanish to scaffold students' learning, specifically when emphasising instruction, reinforcing key concepts, checking comprehension, managing the classroom, relating to students and providing encouragement. Mary and Young (2017) observed in their study how a teacher in a mainstream preschool in France, picked up a few words from the children's home languages in interaction with them and with that, managed to use translanguaging as a pedagogical tool in her daily classroom practices. Her translanguaging had specific purposes: to meet the children's basic needs especially when they were new in the classroom, to make connections between home and school contexts, to scaffold the children's learning of the majority language by building on their prior knowledge, and to strengthen the children's engagement with literacy. Her use of the children's home languages thus created a safe learning environment for the children and helped the teacher to connect and build relationships with them. This encouraged the children's participation in whatever language they felt comfortable with and promoted positive attitudes to their home languages and cultures. Another study of teachers' translanguaging practices in preschools and primary schools in Luxemburg and the Netherlands (Duarte, 2020), suggests that pedagogical translanguaging may have three main functions: symbolic (aimed at acknowledging and valorising HLs, e.g. when a teacher asks the children about a word in the children's HLs), scaffolding (aimed at linking knowledge of children's different languages, e.g. when a teacher initiates counting in different languages), and epistemological (where different languages are actively used to enhance both content and language knowledge, e.g. when the teacher involves the children in a conversation in their HLs about something discussed in the classroom). Whereas the symbolic and scaffolding functions can be used when the teacher has little or no proficiency in the children's HLs, the epistemological function requires that the teacher is a proficient user of the children's HL.

HL SUPPORT IN NORWEGIAN MAINSTREAM ECEC CENTRES
HL support does not appear to be a common practice in Norwegian ECEC centres (Andersen et al., 2011;Lindquist, 2019;Sadownik, 2018). Bilingual teaching assistants are sometimes hired to support HLs (Tkachenko et al., 2015), but this strategy has lost popularity in Norway as ECEC centre owners often prioritise children's majority language skills (Rambøll Management, 2014). Despite this, there are a few case studies documenting how HLs are supported in mainstream Norwegian ECEC centres. Alstad (2013) describes two main HL support strategies: the use of bilingual assistants in separate HL training sessions and fostering metalinguistic conversations. Fredriksson and Lindgren-Eneflo (2019) specify four HL support strategies used in Swedish ECEC centres: promotion of different languages, promotion of different cultures, use of materials and activities, and use of children, teachers or parents with command of the relevant languages. Puskás and Bjørk-Willén (2017) describe a Swedish ECEC group with particular focus on three languages. In this group, multilingual teachers use three languages interchangeably throughout the day, emphasising one language at a time in certain time-limited activities. In addition to these studies, there are some researcher-initiated projects that try out new HL support strategies, i.e. the use of interpreters (Kanstad, 2018) and listening to fairy tales on an iPad (Jaeger, 2019).
Although HL support and translanguaging pedagogies may have positive outcomes for children, the teachers interviewed in Puskás and Björk-Willén (2017) characterise the Swedish framework plan's expectations of both supporting the children's HLs and teaching the children Swedish as a dilemma. They find it concerning that many of the children may not develop their bilingual competence since they are not given sufficient opportunities to speak Swedish, because they are not supported in their HL, or a combination of both. Moreover, the teachers articulate other dilemmas related to a lack of mutual understanding and exclusion. Similar dilemmas are also discussed in Fredriksson and Lindgren-Eneflo (2019), suggesting that ECEC teachers often feel uncertain about their pedagogical choices when it comes to translanguaging pedagogy and HL support.
The literature on HL support strategies in ECEC is scarce and predominantly consists of case studies concerning one or a small number of groups or teachers. The present study supplements this body of research with data from a larger sample of ECEC centres. Further, we focus on the teachers' strategies used in superdiverse groups with many HLs represented and where the teachers do not necessarily have direct access to the languages in question. Our research 5 Tkachenko et al. Journal of Home Language Research DOI: 10.16993/jhlr.41 questions are: 1) What strategies for HL support emerge from the narratives? and 2) How can these strategies be related to García et al.'s (2017) framework for translanguaging pedagogy at the levels of stance, design and shifts?

METHOD
Our data comprises 26 narratives from practice written by ECEC student teachers. All student teachers following the early childhood education teacher training programme spend designated periods in ECEC centres as part of their studies, and many student teachers also work part time in ECEC. In our study, the student teachers collected data from the centres they worked in, as part of the coursework on multilingualism. They were asked to submit narratives from practice 'where other languages than Norwegian were used, or where there was a potential for the use of other languages'. The original dataset included 32 narratives, but some narratives were excluded because they a) did not involve children; b) did not involve the use of HL; or c) did not include any explicit or implicit information concerning teachers' actions or intentions to promote HL.
The data collection was reviewed by the Norwegian national organ for ethics in research. The student teachers provided written consent for use of their narratives in the project and were instructed on anonymisation of narratives. Children, parents and ECEC centres involved in the project were informed about the study, but since the student teachers were the formal participants in the study, consent from children, parents and teachers was not required.
Narratives from practice are short texts written by practitioners describing a situation that took place as part of their work. This is a common way of documenting ECEC practices as a basis for critical discussion in early childhood education, and it is a genre that student teachers are familiar with (see also Ødegaard & Økland, 2015;Rothuizen et al., 2019). Such narratives do not necessarily describe the events exactly as they happened (event-as-lived). The aim is rather to reconstruct the events with a focus on how the author interprets the event (event-as-told) (Connelly & Clandinin, 1990;Cortazzi, 2002). Such narratives could have also been constructed based on several events, thereby illustrating typical situations.
Since Norwegian ECEC centres traditionally have non-hierarchical structures, along with a high proportion of unskilled workers in the sector, we use the terms 'staff' and 'teachers' interchangeably here. The narratives analysed in the present study are written by student teachers who are presumably narrating their own actions and experiences, substantial proportion of the teachers referred to in the narratives are therefore likely to be teachers in training. As most of the ECEC teachers (and ECEC student teachers) in Norway are female, we refer to the narrators and the teachers as 'she'.
Our analysis takes the form of a conventional qualitative content analysis (Hsieh & Shannon, 2005;Lundestad, 2019). We worked through the narratives in several rounds, developing coding categories and identifying patterns concerning the teachers' strategies. Not all of our narratives had a distinct narrative structure involving, for example, a clear timeline, a specific major incident or a conclusion. However, some of them did take this form, thereby allowing for the use of approaches from narrative analysis (e.g. Herman & Vervaeck, 2019; Rothuizen et al., 2019), in particular when trying to understand the author's positioning towards the events they described. This positioning can be helpful in analysing the pedagogical strategies used in the various narratives. However, the strategies presented in our analysis are inevitably open to several interpretations, and we acknowledge that our limited information about the situations makes our interpretations vulnerable. We nevertheless believe that the pedagogical choices that are (implicitly or explicitly) represented in our narratives can provide an interesting window into challenges and ambitions concerning teachers' strategies to support HLs in Norwegian ECEC.
In the following, we firstly outline our main generalisations from the whole dataset comprising 26 narratives, focusing on the teachers' strategies that emerge from our material. For each strategy, we provide some short examples from our data. We then go on to present one of the narratives, which we have called 'Gatito lives here', which illustrates several of these strategies. The original narratives were written in Norwegian. We translated 'Gatito lives here' and excerpts from several other narratives into English, attempting to stay as close to the original wording

ANALYSIS AND RESULTS
Seventeen different languages were mentioned in our narratives, excluding Norwegian. Among the languages mentioned, English was the most frequent (12 narratives), often alongside other languages. Most of the narratives describe situations involving few children, and only four narratives involved larger groups of children. Although a large percentage of children and staff in ECEC speak languages other than Norwegian (Haugset et al., 2015;Ministry of Education, 2020), the narratives reflect that the shared linguistic repertoire between teachers and children is often limited: In five narratives, children and adults spoke the same HL, in nine narratives, some adults had some proficiency in some of the children's HL, while in ten narratives, the central participants did not speak the same HL. In two narratives, it was difficult to tell which languages were known among the children and staff. The number of different languages that surfaced in this relatively small dataset, along with the limited overlap in languages spoken by staff and children, supports our claim that at least some Norwegian ECEC groups can be described as superdiverse (Budach & de Saint-Georges, 2017).
In general, the ECEC staff supported HL use across various situations by directly or indirectly initiating the use of HLs or deliberately supporting someone else's initiative to use HLs. Our analysis suggests that HL support in ECEC takes place both as part of pedagogically pre-planned activities and as a response to spontaneously occurring situations: 12 narratives in our data described pre-planned use of HLs, 13 described spontaneous use of HLs, and one narrative was difficult to categorise. When relating this to García et al.'s (2017) concepts for translanguaging pedagogies, the pre-planned actions can be interpreted as examples of translanguaging design whereas the spontaneous use of HLs can be seen as translanguaging shifts.

STRATEGIES FOR HL SUPPORT
After several rounds of analysis, we developed three over-arching categories that seemed to capture the teachers' pedagogical strategies as they emerged from the narratives: a) initiating activities that encourage HL use; b) facilitating metalinguistic conversations; and c) consulting or involving language experts. The strategies sometimes resulted from pre-planned actions for translanguaging design, while others originated from the teachers' spontaneous decisions to build on learning opportunities that arose in the specific situations, i.e. translanguaging shifts. Below, we elaborate on what these three categories involved, and illustrate them with examples from the data.
Initiating activities that encourage HL use Several narratives described the use of music, digital tools, book-reading and counting as stepping stone for including HLs. Singing in multiple languages was portrayed as something that the children enjoyed and requested: 'After singing "Father Jacob" in Norwegian and English, Mari asks if they can sing it in Somali.' Other narratives describe how digitally available material can be used when the staff lack proficiency in an HL: 'The mother had written down the names of the songs in Norwegian. We couldn't find them on Spotify, but we found them on YouTube'. In a narrative describing book-reading, story books in Chinese were made available to the children: 'A member of staff was on holiday in China and bought some China-books [sic] that we have at the ECEC centre.' Finally, counting was described in many narratives, often leading to the use of several different HLs: 'I started counting in German. In the beginning, he [the child] looked at me with a questioning look.'

Facilitating metalinguistic conversations
Some narratives involved metalinguistic conversations about languages. These conversations were often initiated by staff: 'Knut [teacher] says that he can say "good morning" in English, and Gabi says, "I know Somali, do you know Somali?"' Our data also included situations where the conversational initiative came from the child, but where the staff developed the topic in a metalinguistic direction: 'One day at the playground, I overheard two girls who have Chinese as their mother tongue who talked about Ingvild, an assistant: "She can speak Chinese." "How do you know that?", I asked'. The strategy of initiating or facilitating metalinguistic conversations was quite well represented across our narratives and appeared to allow the use of multiple languages in larger groups where not all participants in the conversation had equal command of the languages in question.

Consulting or involving language experts
Several narratives involved the use of language experts, whereby members of staff, parents or children were used to translate, explain or in other ways share language knowledge. When children either took on or were given the role of language expert, the staff and children were portrayed as learning bits of an HL, as seen here when a teacher and a child read together: 'After a couple of pages, he [the child] first said the word in Norwegian and then in Vietnamese. I [the teacher] repeated the word in Vietnamese and he laughed, obviously because my pronunciation was not completely "correct"'. Not all such invitations were immediately accepted by the children, such as in this example when the staff pointed out that a child knows several languages: '"You are really lucky! Maybe you can teach us something?" The girl gets a bit embarrassed, and initially won't say anything, and the adults leave the topic as the conversation moves on to other things.' When staff members were used as language experts, they were often brought into situations they initially were not a part of. In one narrative, a conflict between two Urduspeaking children is resolved by involving a teacher from another group: ' […] can you wait a moment and then I will get Samir who works in another group and can speak your mother tongue'. The use of expert staff also appeared as part of pre-planned activities: 'We made an appointment to exchange staff some days during outdoor hours […] we also brought this adult along for excursions.' In some narratives, the narrators themselves feature as the expert, entering situations where they observe colleagues failing to understand something a child has said.
Parents were also used as experts in our narratives, for example when the staff asked them to provide a list of keywords in their HL with a purpose to integrate them into daily activities or write them on posters on the wall. Parents also provided information in more spontaneously occurring situations, for example when discussing events that took place earlier in the day: 'When the girl was picked up, I told the mother this [what had happened]. The girl repeated the word and the mother translated it as "rice"'.
Below, we present the narrative 'Gatito lives here' as an illustration of all the strategies presented above. We will consider the narrative in light of the interplay between design, stance and shifts, based on García et al. (2017), discussing how teachers' strategies at different levels contribute to bridging the worlds of home and ECEC, encouraging children's active participation and expressing positive attitudes toward languages.

STANCE, DESIGN AND SHIFTS IN 'GATITO LIVES HERE'
Participants: Aron (4 years), Maja (3 years), Eva (4 years), three teachers. Direct citations are provided in Norwegian, with an explanation/translation in English in parenthesis.
1 We are sitting at the table finishing our crafts activity. All the children have drawn themselves and their families. The children have also made their own houses and now they are about to cut and paste their families into their houses.
2 The children are talking about who lives together with them in their houses. One of the children asks me who I live with, and I answer that I live with my husband and my daughter. Eva asks if I have any animals at home, and I answer "no".
3 Aron enthusiastically declares that 'dahiho dor dej' (= Norwegian 'gatito bor der', 'gatito lives there', pronounced with some deviations from adult language). I ask him what he said, and he repeats it. Maja explains: 'Aron sier at Gatito bor hos han' ('Aron says that Gatito lives with him'). I ask who Gatito is, and the children look at each other smiling. Aron responds 'dahiho er dudedat' (= 'Gatito er en pussekatt', 'Gatito is a pussycat'). I understand that we're talking about a cat and confirm this by saying that Gatito is probably a cute cat. Aron nods and smiles.
8 I am very glad that we have multilingual children and staff with us, because these resources are useful when situations occur where you need to explain or inform in languages other than Norwegian in the ECEC. In this conversation, I needed help in both Spanish and Polish. And I saw that the conversations got better because of the adults' contributions based on the children's interests. 9 The children controlled the whole conversation based on the focus of the conversation, and they got very involved and curious about each other. And we adults were also curious about what different words meant and how the language comprehension differs from child to child. We have been thinking a lot about how we can build on this topic concerning language and family, and many good ideas have emerged from both children and teachers.
In 'Gatito lives here', the teacher indirectly encourages HL use by means of an activity, and the focus on the activity gradually develops into a metalinguistic conversation where the children demonstrate their proficiency in the HLs. Although the teacher who starts the activity does not have command of all the HLs brought into the conversation, she makes use of language experts: other children and teachers who speak the children's HLs.

Bridging the worlds of home and ECEC
During this crafts activity, the children represent their homes and families (1), which leads to the children becoming interested in talking about their families and pets (2). The use of HL may be encouraged by including activities linked to homes/families since it can generate conversations about the children's home environment. In this example, it leads to the introduction of Gatito, a cat with a Spanish-sounding name (3). In line with García et al. (2017), we interpret the deliberate use of activities that encourage HL use as an example of translanguaging design, whereby the teachers create opportunities to bring the children's linguistic and cultural identities into conversations. At the same time, this design also places responsibility on the teacher to make numerous moment-to-moment decisions during the conversation (translanguaging shifts) that are in line with the translanguaging pedagogy principles. The last sentence of the narrative (9) also illustrates how design, shift and stance are interconnected, as the narrator comments on how this experience inspires the staff to develop their strategies for later pedagogical initiatives and activities.

Active participation and expert roles
In 'Gatito lives here', the activity and conversation inspire children to claim expert roles and actively contribute with their multilingual resources. Although four-year-old Aron is not always understood by the narrating teacher, three-year-old Maja helps by explaining what Aron meant (3). Another child, Eva, introduces the topic of English words, firstly by asking whether the teacher knows certain animal terms, and later by elaborating on language use in her family, a trip to Australia and about the animals she saw there (4). This seems to inspire Aron to share 9 Tkachenko et al. Journal of Home Language Research DOI: 10.16993/jhlr.41 his knowledge of Spanish, and Maja again provides a translation of his Spanish words (5). The activity attracts a second boy, who joins the conversation by presenting his proficiency in Polish (6). Although the narrating teacher does not understand Polish, she invites a colleague to help. A third Spanish-speaking colleague also joins the conversation after a while. Altogether, the three adults and four children use four different languages, and they all try to communicate and show approval of the others' language proficiency by imitating and translating words between languages (7).
By engaging in the children's contributions, despite not understanding everything, the narrating teacher demonstrates an openness for translanguaging -a translanguaging stance -which might in turn encourage the children to bring more of their multilingual resources into the conversation. The fact that the teacher is aware of the languages spoken by her colleagues allows her to make use of staff experts, in line with the strategy of involving experts discussed above. We consider awareness of the language resources available to be an element of translanguaging design, which makes it possible for the teacher to make shifts, adapting flexibly to the conversation as it develops.

Positive attitudes toward the use of different languages
The teacher is portrayed as actively promoting positive attitudes towards use of the children's multilingual repertoires throughout the conversation by responding positively and acknowledging the children's initiatives, even when she does not understand. This may serve to illustrate the teacher's translanguaging stance and her willingness to take translanguaging shifts in the course of the conversation.
The teacher is depicted as having an open and explorative attitude, as well as 'shuttling' between languages used by the others to co-construct meaning (Canagarajah, 2011, pp. 4-5). This may in itself be considered a strategy that promotes HLs in ECEC. In order to successfully navigate the diverse language repertoires brought into the conversation by the children, the teacher is described as using strategies to positively affirm the children's initiatives (3), negotiating meaning when she does not understand (3), and resorting to the help of 'language experts' (children (3, 5) and colleagues (6)). She is also shown to build on her own multilingual competence (e.g. by following up on Eva's questions about the meaning of some English words (4)), and to position herself as a language learner by repeating the words in the children's HLs (7). These strategies resemble the strategies used by one of the teachers in Alstad's study (2013). Through these strategies, the teacher is role-modelling for the children how to navigate in flexible language practices. The narrative signals that an open attitude towards different HLs is also shared by the children, as they were willing to bring elements of the conversation into their activities later that day (7).
The teacher's own reflections indicate that she sees languages other than Norwegian as a useful resource, and she appreciates the help she receives from the teachers who share the linguistic repertoires with the children (8). Furthermore, the teacher stresses the importance of building on the children's HLs in her pedagogical work. We interpret the positioning of the narrator as evidence that she has taken up the translanguaging stance and manages to be flexible in language practices, which also leads staff to transform and further develop their translanguaging design (García et al., 2017).
Our analysis of the narrative shows how translanguaging stance, design and shift are interrelated and how a combination of the three can contribute to creating spaces where HLs are actively used in ECEC. We have also seen examples of how initiating activities that encourage HL use, facilitating metalinguistic conversations and consulting or involving language experts can be important components in translanguaging design. In addition, we have seen how realisation of these strategies is dependent on the teacher's translanguaging stance and shifts, i.e. positive attitudes towards the children's HLs and multilingualism, and her willingness to build on the available multilingual resources for meaning-making and mutual learning.

DISCUSSION AND CONCLUDING REMARKS
In this paper, we have shown that there is room for HL support across a range of different situations in Norwegian mainstream ECEC. HL support is performed through various strategies that we have grouped within the three categories: 'consulting language experts', 'initiating 10 Tkachenko et al. Journal of Home Language Research DOI: 10.16993/jhlr.41 activities that encourage HL use' and 'facilitating metalinguistic conversations'. These categories resemble the strategies described in previous research (e.g. Alstad, 2013;de Oliveira et al., 2016;Duarte, 2020;Fredriksson & Lindgren-Eneflo, 2019;Mary & Young, 2017;Puskás & Björk-Willén, 2017), while in the present study, they are based on a larger selection of ECEC groups. As illustrated in 'Gatito lives here', the combination of these strategies can allow the children to participate in conversations about language, to share their experiences and knowledge, and to meet positive attitudes towards their HLs.
Our second research question was how the strategies observed in the narratives could be related to García et al. s (2017) framework for translanguaging pedagogy at the levels of stance, design and shifts. Our analysis of 'Gatito lives here' reveals the interplay between these three pedagogical dimensions. In 'Gatito', we interpret the choice of activity along with the teacher's knowledge about the linguistic resources available among the children and staff as examples of design, whereas the teacher's ability to build on the children's contributions during this activity can be considered shifts. Regardless of whether the pedagogical choices made were pre-planned or spontaneous, they were centred around meaning-making through the use of different languages, and by the staff attempting to accommodate the children's linguistic needs and respectfully respond to their language practices. This shows how the teacher's firm translanguaging stance may have influenced her actions. Such interrelations between multilingual practices and language ideologies are according to Ganuza and Hedman (2017) absolutely necessary for translanguaging pedagogy. The teacher's reflections also show how she and her colleagues plan to further develop their practice.
The translanguaging practices revealed from our data are not necessarily typical of Norwegian ECEC groups. Lindquist (2018Lindquist ( , 2019 states that the teachers in his study are not particularly interested in children's (possible) HL expressions. In Alstad (2013), only one of the three teachers used translanguaging practices to build on the children's multilingual repertoires. Since the student teachers in our study collected narratives from practice as an assignment in their training in multilingualism, our data might show the student teachers' first attempts at a translanguaging approach in ECEC groups where this is not in fact part of everyday procedures. This may be the reason why some narratives express insecurity among both staff and children when HLs come into play. Examples include narrators being uncertain about whether switching between languages was the appropriate choice, children withdrawing after prompts to use HLs, or teacher frustration when a song in what is assumed to be the HL of a child is not understood by that child. The expressed insecurities might also be manifestations of the mismatch between stance, design and shifts, which can lead to dilemmas about HL support, as expressed in other studies (Fredriksson & Lindgren-Eneflo, 2019;Puskás & Björk-Willén, 2017). One could perhaps expect that such insecurities could be seen as a starting point for joint reflections among staff about their practices and dilemmas, which in turn might help practitioners to realise the Framework plan's intentions of HL support and make multilingual resources an enrichment for the entire group of children.
To support teachers in their attempts to include HL support in mainstream ECEC settings, it is important to address all three strands of translanguaging pedagogy in professional development (Kirsch & Duarte, 2020). In superdiverse ECEC groups, where children and teachers do not necessarily share the same language repertoires, the choice of appropriate translanguaging strategy might depend on the degree of overlap in the language repertoires. When the teachers and/or children share an HL, they can use that HL in communication. In our data, such epistemological use of translanguaging (Duarte, 2020) mostly involved few participants. However, the strategy of involving language experts in these conversations has the potential to expand the HL use to include more children in these dynamic and flexible language practices. If the teachers have limited repertoires in HLs, they can make use of language experts (parents, children, staff) or ensure that they include activities or materials (books, music, art) that allow the inclusion of HLs without relying on the teacher's proficiency. Although symbolic and scaffolding-based (Duarte, 2020), such interactions are described in many of our narratives as sparking the children's use of HLs, not only with the staff but also among peers. Teachers' multilingual repertoires can be made visible as a means to signal the value of speaking multiple languages, and to model flexible language practices, which in turn may valorise HLs and break up the domination of the majority language. We therefore suggest that symbolic and scaffolding-based strategies may indirectly lead to more epistemological uses of HLs, even if these conversations do not necessarily involve staff or they take place in other arenas.