Minoritising a regional language in multilingual preschools in Dutch Limburg: Teachers’ and toddlers’ choices between Dutch and Limburgish

Aims and objectives: We discuss how children with various language backgrounds interact in preschool playgrounds in Limburg, the Netherlands. This paper addresses the question how power dynamics between Dutch and/or Limburgish, and other languages are enacted in, by, and through language choice in preschool settings, and to what extent this leads to social (in)equality. Approach: This paper incorporates data collected during ethnographic fieldwork and discusses literature about language socialisation, multilingualism, and language policy. Data and analysis: All data were collected at three preschools in Southern Limburg with over 30 children and more than five teachers. The data were analysed using various theoretical perspectives. Findings and conclusions: We conclude that the ways in which, when, how, and in which activities children and teachers select languages show a social order. This order renders Limburgish and other languages than Dutch unequal in the earliest educational setting that children encounter. No concrete language policy has been developed for children who speak other home languages than Dutch and/or Limburgish. Children discover quickly that using Dutch is more important than other languages in the preschool. Consequently, children, as individual agents, will start acting accordingly. Originality: Through its ethnographical approach, this paper offers a unique insight in the multilingual landscape including regional language use at preschools in Southern Limburg. This approach is based on actual, observed behaviour instead of reported behaviour or behaviour stipulated by language policies. Significance/implications: The initial stimulus for this research is societal: 60.8% of the Limburgish participants in a study reported to speak Limburgish, yet this amount is decreasing, and concerned parents and municipalities requested insights why their children prefer Dutch over Limburgish at home soon after attending preschool. This research aims to provide answers why this happens and how to ensure a more linguistically equal preschool.

The 'Methodology' section discusses the methodology used in this study. 'The use of Limburgish in preschools in the municipality of Eijsden-Margraten' section reveals actual daily language practices in three preschools in Limburg focussing on teacher-child, and peer interactions and language choice, and discusses them. Finally, the section 'Discussion and conclusion of language choice at preschool' contains a discussion.

Language policy and language choice as a contextualised process
The aim of this paper is to shed light on the power dynamics between Dutch and/or Limburgish, and other languages in preschool settings in Dutch Limburg. Every language user associates particular languages with specific kinds of speakers and practices within a social, political, and economic hierarchy, and language choice is thus never a socially neutral activity (Cenoz & Gorter, 2019;Garrett, 2007, p. 252). In this respect, the use of Dutch and/or Limburgish and/or other languages, and language choice both by adults and children in preschool is always an emotional and contextualised process (Pavlenko, 2007).
Limburg can be characterised as diglossic: Dutch and Limburgish are used by many speakers in varying situations. Dutch is most often used in formal situations and most prestigious in education whereas many inhabitants attach great importance to speaking Limburgish in informal, daily, and in cultural activities that are important in the construction of regional identities (Cornips, 2020). Of the inhabitants of the Dutch province Limburg, 48 percent report to speak Limburgish. Limburgish is a spoken variety primarily but might be written by many on various social media platforms. If the dominant language at home is Limburgish, 37 percent of those active on social media use Limburgish as well. (Schmeets & Cornips, 2021). Furthermore, speakers distinguish Limburgish from Dutch since they speak of translating Limburgish into Dutch or vice versa. They also easily code-switch between Limburgish and Dutch in multiparty interactions depending on whom one is orientating to in the conversation (Cornips, 2013). Children raised in Limburgish will also be raised in Dutch, and almost all children raised monolingually in Dutch will understand Limburgish (Cornips, 2020) since Limburgish is a vital language in the public domain.
Language policy is a powerful instrument for achieving political goals and justifying ideological choices (Nic Craith, 2010). The ideal place for propagating a standard language is within schools and its aim is to teach children the standard language (Haugen, 1966, p. 927) which is needed for their educational success (De Backer et al., 2017). Teaching in the standard language is 'frequently based on the idea that each language is best learned through immersion and thus in isolation from the others' (Günther-van der Meij et al., 2020, p. 3). Importantly, '[. . .] educators dynamically negotiate if and when languages [. . .] are to be used in education' (Menken & García, 2010, p. 258). This means that a language policy can be handed down, but in addition to that '[ . . . ] educators respond to the policy according to their own perspective -ideological, linguistic, cultural, educational, and pedagogical' (Menken & García, 2010, p. 259). There is a differentiation between what is declared by a language policy by, for instance, the Dutch law; a perceived language policy, that is, what teachers' thoughts and interpretations are on how a certain language should be used; and a practised language policy, that is, how a language is actually used by teachers and children (Bonacina-Pugh, 2012). Therefore, what is written in a language policy does not fully equal teachers' beliefs, which in turn does not equal what is actually happening. Both children and teachers can be individual agents in the language hierarchy at preschools (Fogle, 2012).
The umbrella preschool management organisation Spelenderwijs meaning 'in a playful manner' in Dutch Limburg coordinates programmes at many preschools in municipalities in the south of Limburg. This private organisation declares in its pedagogical policy plan (Spelenderwijs, 2018, p. 5) that both Dutch and the regional language may be used in order to address children in 'their' preschools, with the regional language being Limburgish. Consequently, minority languages such as Yiddish and Roma cannot be used and are lower in the hierarchy. Dutch is recommended to be used in group activities whereas the use of Limburgish should be restricted to individual interaction between teacher and child. As a result, Spelenderwijs promotes language separation, that is, Limburgish is linked to the individual Limburgish-speaking child and Dutch to the group of children regardless of their language background (Dutch, Limburgish, or other).

Methodology
This paper is one of the few observational studies on toddlers that may reveal how influential teacher-child (De Houwer, 2020), and peer interactions promote type of discourse strategies (Blum-Kulka & Gorbatt, 2014;Rickert, subm) and language choice between a European regional language, that is, Limburgish, standard language, and so-called migrant languages. The municipal authorities of the municipality under study, that is, Eijsden-Margraten in the south of the province of Limburg invited us to observe language practices in preschools with a focus on the use of Limburgish, and report the results to them. The municipal authorities requested the participation of several preschools and subsequently informed the management of Spelenderwijs, who in turn sent a letter to each of the parents of the children at the preschools and gave informed consent for conducting the field work: that is, they received a form and gave permission for the researcher's presence and the recording of material. Preschool teachers and parents of the children were informed by Spelenderwijs about the aim of our research and the presence of the researcher (and first author) during preschool activities.
The observational approach taken in this study yields more detailed insights on the interactional context and a more complete understanding of bilingual behaviour in which verbal and non-verbal behaviour occurs (Blum-Kulka & Gorbatt, 2014;Mifsud & Vella, 2018, p. 58), thus, what teachers, peers, and children can and cannot do and say (perceived and practised language policy; Bonacina-Pugh, 2012). On a critical note, the mere presence of an 'outsider' (here: the researcher) influences behaviour as well since every 'participant' is aware of an extra person who usually is not present. However, being present is as close to reality as one can get. All observations were made at three preschools in the municipality of Eijsden-Margraten on 13 days between February and June 2016.
Two preschools were open during the morning hours (8:45-11:45) whereas the third one operated all day (between 8:45-11:45 and 13:00-15:30). Most ethnographic fieldwork was conducted by joining in the daily programme, for example, by answering teachers' questions together with the children, doing the same things everyone else was doing, and observing as much as possible from a distance how the children interacted. One of the main guidelines in the participant observation was to lend a helping hand wherever possible, without interfering in the activities and proceedings at the preschool, for instance, by placing chairs in a circle or cleaning up. In addition to not interfering, it was just as important not to breach existing language routines. Therefore, an utterance in Dutch by a child or teacher always led to a response in Dutch by the first author, and similarly for Limburgish.
Field work notes were made by the first author and at the moment when trust was established with teachers and children, and after observing which discourse strategies, that is, language choice moments were interesting, for instance, during the so-called circle conversations and free play outside, recordings were made. In some cases, solely taking notes using a notebook would have provided insufficient details about the use of paralinguistic elements such as pitch, volume, or intonation. Recordings were made using the recording application of a Samsung Galaxy S5 (in silent mode). Audibility of the recording and visibility of the device were key considerations in how to position the mobile phone. The device may have distracted the children for a few seconds, but usually they quickly continued with their activities. The teachers were aware of the recording side of this study.
The languages used and the numbers of attending children and teachers varied enormously between the three preschools and between morning versus afternoon hours. The number of children in all preschools varied between a minimum of 3 and maximum of 15. The first author observed in the Margraten preschool on 23 February that 14 children were present during the morning hours who all spoke Limburgish whereas 7 children were present during the midday hours of whom 4 spoke Limburgish, 2 Dutch, and 1 another language. The exclusive use of Limburgish was never found among teachers and children in two preschools. The presence of teachers varied between three and six for all preschools under study including internship teachers. With the exception of one teacher who has been residing in Limburg for 25 years but chooses to interact in Dutch, all teachers used Limburgish. 5 For this paper, all observations, field work notes, and recordings were analysed using VLC Media Player, and recordings were transcribed in Microsoft Word.

The use of Limburgish in preschools in the municipality of Eijsden-Margraten
In order to address the question as to how power dynamics between Dutch and/or Limburgish, and other languages are enacted in, by, and through language choice in preschool settings, and to what extent this leads to social (in)equality, we discuss our observations regarding (a) what language knowledge is transferred in the so-called kringgesprek 'circle conversation', (b) how the teachers use and switch between Dutch and Limburgish, and (c) how the children use Dutch, Limburgish, and other languages in peer-group interactions (see Blum-Kulka & Gorbatt, 2014;Von Suchodoletz et al., 2014, p. 510).

Preschool education: transferring knowledge in kringgesprek (circle conversation)
Teachers and children sit on small chairs aligned in a circle for the so-called kringgesprek (circle conversation) which takes place several times per day. In a circle conversation, teachers and children perform several activities together such as listening to a story read by the teachers, putting an object in the centre of the circle and talking about it, answering teachers' questions such as 'How was your weekend?', or singing for someone's birthday. In this way, knowledge is being transferred and children are learning. Although the overwhelming majority of the children speak Limburgish, teachers clearly show an inclination to speak Dutch. For example, one of the teachers announces in a circle conversation to all children what the next part of the programme will be: cutting out a small flower shape on paper: (1) A teacher addressing the group in Dutch: 'We gaan zo een bloemetje prikken!' 'We will punch out a small flower in a minute!' Contrastingly, teachers show a tendency to use Limburgish in dialogues between themselves and Limburgish-speaking children in one-to-one interactions, as shown in example (2). The examples in (1) and (2) took place shortly after each other, both produced in the circle setting: (2) A teacher addressing Iris in Limburgish: 'Wils se 'n hertje prikke?' 'Do you want to punch out a small heart?' In general, teachers avoid the use of Limburgish when addressing children who only speak Dutch. Consequently, Dutch-speaking children and children speaking other languages learn fairly quickly that a teacher's message in Limburgish is not directed at them but at children who use Limburgish. Frequently occurring strategies such as speaking Limburgish with an individual child, and Dutch while addressing all children will be entrenched more deeply in an individual's memory, consequently routinising and automatising them (Schmid, 2014). Aside of audible evidence of frequency effects, frequency also has an impact on the stored information within an individual's mind, that is, on the information which gets entrenched (Ibbotson, 2013;Tomasello, 2000). Knowledge about which language to select in which context is argued to be the result of stored knowledge about its usage, obtained in past experiences. Entrenchment occurs passively as well: when a teacher consistently uses Limburgish while speaking to an individual child and Dutch to the group of children in a specific context with a specific aim, the established language selection will be entrenched in the child's mind. Specifically, a child will learn that this is the way in which she or he or the group will be addressed in each context and for which purpose. Contrasting with entrenchment, which takes place within an individual's mind, conventionalisation occurs in the group of children (Schmid, 2015, p. 11).
Children with either Dutch or Limburgish as their home languages will reply in their home language, regardless of whether their language choice differs from the teacher's or not. This is illustrated in example (3). A teacher asks the entire group in the circle in Dutch which colour a specific object has in its centre. All Dutch-speaking children reply with 'Green!' in Dutch, while all Limburgish-speaking children do the same, but in Limburgish: (3) A teacher addressing the group in the circle in Dutch: 'Welke kleur is dit?'

'Green!'
Limburgish-speaking children replying in Limburgish: 'Greun!' 'Green! During lunch -another informal group activity while sitting in a circle -Limburgish-speaking children were assembled at one table -which occurred very rarely -while the other table was mixed between Dutch-and Limburgish-speaking children. The teachers asked the respective tables whether anyone would like something to drink. At the 'Limburgish' table, the teacher asked the question in Limburgish while another teacher asked the same question in Dutch at the 'mixed' table. Similarly, during another group activity, a teacher was reading in Dutch 6 about bears, and two children were listening. One child spoke Limburgish while the other child spoke Dutch. The teacher asked the Limburgish-speaking child a question in Limburgish and repeated, that is, translated the question individually for the Dutch-speaking child, as shown in (4) and (5) respectively: (4) A teacher to John in Limburgish: 'Bis doe bang veur bere?' 'Are you afraid of bears?' The same teacher to Sarah in Dutch: 'Ben jij bang voor beren?' 'Are you afraid of bears?' The example in (6) shows an interaction between a teacher and Giulia (pseudonym), who both speak Limburgish as their home language. In interaction (6), the teacher repeats Giulia's Dutch utterance in Dutch. In doing so, Giulia will learn that all she has to speak is Dutch (Limburgishspeaking teachers will adapt and respond to her in Dutch), and the teacher will be able to provide covert feedback to her by repeating; in this case that her utterance is understandable and correct, of which the importance is described in Spelenderwijs' language policy (Wouterse-Schmitz, 2005, p. 8): (6) Giulia addressing a teacher in Dutch although both have Limburgish as their home language: 'Een groot worstje heb ik gemaakt!'

'I have made a large sausage!'
The teacher reacting to Giulia in Dutch: 'Ja, een groot worstje!'

'Yes, a large sausage!'
Repetition is one of the tools at a teacher's disposal for stimulating a specific language choice (De Houwer, 2020) in the circle and for socialising children with a view to which language to use for knowledge transfer. In example (7), Limburgish-speaking Frits (pseudonym) states in Limburgish that the caterpillar in Eric Carle's The Very Hungry Caterpillar is about to nibble on a leaf, which the Limburgish-speaking teacher confirms and repeats in Dutch.
Frits addressing a teacher in Limburgish: 'Kiek, dao geit d'r knabbele!' 'Look, he is going to nibble there!' The teacher replies to Frits in Dutch: 'Ja, daar gaat hij knabbelen!' 'Yes, he is going to nibble there!' All children were paying attention to the conversations between Frits and the teacher in (7) as in similar conversations although not all children were directly addressed, they overheard the conversations.
Concretely, as confirmed by talking to teachers, Dutch is stipulated as the instruction language for the entire group in the circle conversation in the language policy, teachers find the acquisition of Dutch important, and this is reflected in their perceived and practised language policy: they socialise the children into the usage of Dutch when transferring knowledge and in Limburgish addressing Limburgish-speaking children only in informal dyadic or multiparty interactions.

The 'private' interaction: teacher's use of and switch to Limburgish
In example (8), Lotte tells a teacher in Limburgish that she has watered the flowers, but in this case, the other children are not paying attention. As expected from the above, the teacher repeats her question in Limburgish. (8) Lotte addressing a teacher in Limburgish: 'Ich höb de bloemetjes water gegeve!'

'Oh! Have you watered the flowers?'
Repetition like in (8) occurred frequently in every type of interaction but whether the teacher will accommodate (Giles, 2008) to the language choice of the child -Limburgish or Dutch -is dependent on whether transfer of knowledge takes place in the circle and/or whether the interaction is considered 'private' between teacher and child, without the other children as overhearers (Bell, 1984). In the former case, the teacher's repetition of the child's utterance will always take place in Dutch, no matter the language chosen by the child. In the latter case, the teacher may repeat the children's Dutch utterances in Limburgish. To illustrate this alternation, in example (9), Vera is upset that her mother is not coming to pick her up at the end of a long day. The teacher rephrases her utterance as a question, and in Limburgish. (9) Vera to a teacher in Dutch: 'Mammie komt niet'.

'Mommy is not coming'.
The teacher replying to Vera in Limburgish: 'Kömp ze neet?' 'Is she not coming?' Second, in example (10), Iris says in Limburgish that she has paint at home and the teacher again reformulates her statement as a question in Limburgish. Vera and Iris have Limburgish as their home language and the other children paid no attention to the conversation: (10) Iris addressing a teacher in Dutch: 'Ik heb ook verf thuis!'

'I also have paint at home!'
The teacher replying to Iris in Limburgish: 'Höbs se ooch verf thoes?' 'Do you also have paint at home?' For children, time for playing outside means time for spreading out all over the playground. During this part of the programme, children have the chance to decide for themselves what they would like to do, and with whom. Children do activities like hide-and-seek, a bicycle race, or making sand cakes in the sandpit at the playground outside where they often leave some of their things behind. How a teacher deals with this in terms of language choice is shown in example (11). The teacher asks all the children in a loud voice -in Dutch -whose coat she has just found. Antonio, a young boy, tries to claim it -in Limburgish -after which the teacher answers him, only to tell him in Limburgish that the coat is definitely not his. The teacher replying to Antonio in Limburgish: 'Nae, dae is neet van dich!'

'No, that one is not yours!'
In sum, in their perceived and practised language policy, teachers socialise children into the usage of Limburgish in informal and/or 'private' interactions but only those children who are known to speak Limburgish at home.

Among peers
Peer interaction provides a more complicated picture regarding the use of Limburgish versus Dutch. In interaction (12), Sem engages in Dutch with Gino (the researcher) and Gino responds in Dutch, but Max also replies to Dutch-speaking Sem, but in Limburgish: Dutch-speaking children, however, never switch to Limburgish, neither to their teachers, nor to their peers. They, thus, can simply keep on speaking Dutch. Moreover, when a child interacts in Limburgish with a Dutch-speaking child, the latter can interrupt the Limburgish child in Dutch, as illustrated in (13) Here, Maria actively tries to steer Tim into the usage of Dutch and shows unwilling agentic behaviour to engage in a Limburgish conversation. A similar observation has been made in the Finnish and Swedish context by Bergroth and Palviainen (2017, p. 22), where in their study one girl (Alisa) tried to steer another girl (Sara) into using Swedish with her. Peer interaction may thus show an asymmetrical process: the Limburgish-speaking Tim switches to Dutch when addressing Dutch-speaking Maria, but vice versa Maria or any other Dutch-speaking child will never switch to Limburgish. Furthermore, example (14) shows that a Dutch-speaking child -Maria -may interrupt a peer -Tim -using Limburgish. In contrast, Maria engages and actively participates in the following conversation in (14) with Tim when Tim speaks Dutch:   (13) and (14) (Bergroth & Palviainen, 2017, p. 14). He shows enthusiastic agentic behaviour for willingness and speaking (and learning) the dominant Dutch language whereas Maria shows unwilling agentic behaviour to learn Limburgish as an L2 regardless of the type of activity and the interlocutor (the teacher or peer) (Schwartz et al., 2021, p. 9).

Children's interactions with other home languages than Dutch or Limburgish
Some children have a different home language, such as Arabic, Turkish, or Italian. These children were either born in the Netherlands or in a different country and migrated to the Netherlands with their parents and they might have relatively little experience with Dutch and/or Limburgish when they arrive at preschool at the age of 2. It has been observed that the teachers only speak Dutch with these children, so, neither Limburgish nor their home language. However, Spelenderwijs' language policy does not prohibit the usage of Limburgish with these children. Therefore, the exclusive use of Dutch is an example of a differentiation between the actual language policy and teachers' behaviour based on their perspectives (Menken & García, 2010), since some teachers in this study considered learning Limburgish, in addition to Dutch and their home language, as a burden. The interactions in (15) and (16) both occurred within a short interval. In (15), a teacher asks Remco -who speaks Limburgish -in Limburgish, whether he would like to use another colour while painting. Shortly later, the teacher asks Adil -who speaks Arabic at home -in Dutch which colour he would like to have: The interactions in (17) and (18) below occurred almost simultaneously while both children were sitting at an almost equal distance from the teacher. Max with Limburgish and Aziz with Arabic as his home language were both busy colouring and were not making eye contact with the teacher. Both children could discern without making any eye contact which of the teacher's utterances was intended for whom. Aziz does not look up when the teacher speaks in Limburgish but responds when he hears the teacher posing a question in Dutch: (17)  Ayaan, a Somali girl who has only been in the Netherlands for a brief period of time, has acquired enough Dutch to express what she wants or does not want, and she notices when an utterance is directed at her. When a teacher asks Ayaan in Dutch not to cut in a flower shape on a paper, Ayaan replied, not verbally (see Blum-Kulka & Gorbatt, 2014), but physically by nodding. Even though she ended up cutting in the flower shape anyway, she clearly felt she was being talked to. However, when Joshua, a Limburgish-speaking child, addresses a group of children, including Ayaan, in Limburgish, Ayaan, similarly to Aziz did not feel addressed and did not reply to Joshua's utterance.
Similarly, during one of the observation days, a teacher was reading a story together with a group of children. One of them was a Dari-speaking girl. She was listening attentively to the teacher's story, but once the teacher switched to Limburgish in order to engage in an individual conversation with a child, she lost her concentration and left. Fogle (2012) specifies that not acting, like Ayaan does in reaction to being addressed in Limburgish is also a form of individual agency. In doing so, peers will learn to address her in Dutch, just like teachers address her in Dutch because of their perceived language policy, and these children will display unwilling agentic behaviour (Schwartz et al., 2021) towards the usage of Limburgish.
In comparison to their Limburgish-and Dutch-speaking classmates, children speaking a different home language appear to be more silent, confirming Blum-Kulka and Gorbatt's (2014) analysis, for they tend to play alone more and are left alone more by the other children. Usually, these children seek contact with a teacher. At one of the preschools, three children speaking a different language, two of them of Moroccan and one of Afghan descent, often sought contact with each other. Whenever there was interaction between children speaking another home language and a Dutch-or Limburgish-speaking child, the former child would often reply very shortly or would not reply at all, as shown in examples (19)  During observation time, a boy tried opening a conversation in Limburgish with a boy speaking another home language. However, this boy did not reply. Subsequently, the Limburgish-speaking boy continued playing without him instead of rephrasing his utterance in Dutch. Quite often, even if these children understand Dutch, they do not reply verbally but do fulfil the desired task, as shown for Mahmud, who speaks Turkish, in example (21). Children speaking a different home language are aware of their situation. They know that it is unlikely that anyone will speak their home language in preschool. Ayaan, who would routinely choose to speak Somalian with her brother, still prefers to speak the little Dutch she knows, and not Limburgish at all, over her more advanced knowledge of Somalian, as shown in examples (22)  Concretely, what is declared by the language policy is also perceived practised, namely that for children with other home languages, the preschool is the ideal place for teaching the standard language needed for their future educational success (with the erasure of respective home languages and Limburgish).

Discussion and conclusion of language choice at preschool
'The use of Limburgish in preschools in the municipality of Eijsden-Margraten' section detailed how (a) children are socialised in the usage of Dutch in vital group activities such as the circle by teachers, (b) how teachers accommodate to a child's language preference in individual conversations, (c) how there is a linguistic hierarchy between Dutch-and Limburgish-speaking children regarding the choice for a specific language, and (d) how languages other than Dutch and Limburgish are at the bottom of the language hierarchy.
Analysing our observations, we can point out the following. Teachers socialise children in an educational context with Dutch, Limburgish, and other languages in different ways. The language policy of the umbrella preschool management organisation under study advises teachers in its pedagogical plan to use bidialectal discourse strategies, that is, both Limburgish and Dutch but no other languages are allowed to be used in teacher and child interactions. However, this declared policy, what is stipulated, differs from the perceived language policy, what teachers' think about how the languages should be used. Teachers mention that it is vital to learn Dutch well and should be a priority. Limburgish, on the contrary, is mentioned to have a sentimental value, but it can also, in the words of the teachers, be a burden for children who speak another language like Turkish. These opinions lead to a practised language policy which effectively contains a separation of children who are socialised in both Limburgish and Dutch, and those who are socialised in Dutch only. Moreover, the language pedagogical plan promotes language separation in teachers' practices with different communicative goals, that is, Limburgish is linked to the individual Limburgish-speaking child, and Dutch to the group of children regardless of their language background resulting in subtractive bidialectism (García, 1998). Dutch-and other languages-speaking children, however, may show 'actively not acting' behaviour when being addressed in Limburgish by a peer. Thus, despite the various accommodations made to give Limburgish a place within the preschool context, there is still a very clear reproduction of standard language ideology. Altogether, there is clearly an issue of language status and hierarchy in preschools in Limburg. (Teachers') repetitions and interruptions are important linguistic practices (Moore, 2011, p. 209) to create and maintain language status and hierarchy. Dutch appears to be the only language of instruction, which places that language and the peers who speak it in a more powerful position than Limburgish and its speakers, and Dutch and Limburgish in a more powerful position than the migrant languages and their speakers (Wright & Tropp, 2005, p. 311). This social order effectively renders Dutch, Limburgish, and other languages, in that hierarchical order, unequal in the earliest educational setting that children will encounter. This linguistic inequality affects the children at the playground, because an unequal hierarchy of languages leads to children having unequal opportunities to use the language they are comfortable with.
One way to counteract this social order, in addition to teachers' training in multilingualism, would be to implement the so-called holistic model from a functional multilearning perspective that combines knowledge and teaching approaches. In this perspective, all languages brought into school are considered as 'didactic capital' (Günther-van der Meij et al., 2020, p. 6).