Bidialectal pre-school: enacting participation frames through linguistic and other semiotic means

ABSTRACT

This paper analyses how teachers and toddlers enact participation frames in bidialectal early education in Limburg, the Netherlands. Teachers’ language choice is often context-bound as they use the national language, Dutch, for instruction and the regional language, Limburgish, for playful or social-emotional situations with individual children. Drawing on ethnographic data generated during 4.5 months of fieldwork in a bidialectal pre-school, I address how teachers and toddlers use the two language varieties, respectively, as well as other semiotic means to shape situational participation in multiparty interaction. My multi-modal analysis of selected video- and audio-recordings of interactions of two teachers and the target child Felix as well as varying other participants shows that teachers may use Limburgish to move into a personal conversation amongst colleagues in front of the children. In contrast, they use Dutch to stage conversations which they intend to be overheard by the children. Closely investigating children’s orientation towards participatory statuses and their interactional consequences, it becomes evident that children co-create participation frames initiated by the teachers at times and subvert them at other times.

KEYWORDS:

• Participation
• bidialectalism
• early education
• multiparty interaction
• overhearing
• language socialisation

1. Introduction

The southern province of Limburg in the Netherlands is known for a wide-spread use of dialect,¹ which is a cornerstone of the construction of local and regional identities (Thissen 2018). Local dialects got recognition as a regional language under the umbrella term Limburgs (Limburgish) by the Dutch government in 1997. As a regional language, Limburgish may be used in pre-schools in Limburg in addition to the national language Dutch (‘Wet kinderopvang/Law on childcare, art. 1,55).

For children, Early Childcare and Education (ECE), including pre-schools, constitutes the first step beyond the more intimate home sphere into active involvement in societal institutions. In ECE, children between 2 and 4 come into contact with the language use of teachers and peers which plays an important role in their language socialisation (Schwartz 2018). Language socialisation is essentially the process of learning to use language in ways that are deemed socially meaningful through the use of language (Ochs and Schieffelin 1986).

As prior research shows, pre-schoolers in Limburg frequently do not speak Limburgish in pre-school but only Dutch, even when both parents and teachers, at least partially, use Limburgish with the child (Cornips 2020b). There seems to be a link between this phenomenon and the teachers’ context-dependent code choice, where Limburgish is commonly used for emotional and one-on-one situations and Dutch for instruction and organisation, especially when the whole group is addressed (Cornips 2020b; Morillo Morales and Cornips 2022).

Such a context-bound code choice suggests different participatory affordances, especially since teachers are commonly aware if Limburgish is a home language of an individual child. Participation, understood as ‘actions demonstrating forms of involvement performed by parties within evolving structures of talk’ (Goodwin and Goodwin 2004, 222) is, however, not pre-determined. Instead, participation frameworks are a common effort and achievement by speakers, hearers, and other participants (Goffman 1979). Participation frameworks describe the organisation of constellations of participants orienting toward one another, and toward an action at hand across different participatory roles (Goffman 1979; Goodwin 2000). Consequently, these frameworks develop situationally and dynamically, not only through linguistic means but also through the use of other semiotic resources such as gaze, touch, gestures, body positioning as well as spatial and temporal means (Goodwin 2007b).

This paper addresses the question how diverse forms and constellations of interactional participation evolve between teachers and toddlers against the particular background of the bidialectal linguistic landscape of pre-schools in Limburg. Since participation frameworks are pivotal to the organisation of language socialisation, I set out to investigate the role of Dutch and Limburgish, respectively, as well as other semiotic resources in toddlers’ and teachers’ shaping of situational interactional participation.

2. Local background

2.1. Limburgish dialects and Dutch

In the following section, I will discuss the use of Limburgish and Dutch in the Dutch province of Limburg and briefly introduce linguistic differences between the two varieties in order to facilitate a holistic understanding of the data in the local context.

According to recent numbers, 48 percent of the inhabitants of the province of Limburg speak Limburgish (Schmeets and Cornips 2021). Speakers may acquire Limburgish as part of their bi- or multilingual upbringing in combination with Dutch and/or other languages (Cornips 2013; Extra 2004) or, more seldomly, in the case of migrants from outside Limburg as an L2 (Vousten 1995). New speakers of Limburgish might, however, experience linguistic othering as they are not perceived as ‘authentic’ dialect speakers by the local population (Cornips 2020a). Limburgish is foremost an oral language, but it is visible in a number of street signs (Thissen 2018) and used widely on Social Media (Jongbloed-Faber, van Loo, and Cornips 2017). Children who grow up monolingually in Dutch are generally likely to passively understand Limburgish due to its high vitality in the public domain (Morillo Morales and Cornips 2022)

On a syntactic level, phenomena which distinguishes Limburgish from Dutch (but not necessarily from varieties spoken across the national borders with Germany) include the Ripuarian reflexive adjunct middle, like reported by Cornips (2013, 379) in the following example:

Table

(1)	Limburgish:	Der	sal	singt	sich	legt
	Dutch:	De	zaal	zingt	(-)	goed
	English:	the	hall	sings	(refl)	easily
	‘this hall has good acoustics’ (lit. … sings well).’

On a morphological level, Cornips (2013, 380) mentions formation of the plural through an umlaut, as exemplified through the example of the English “bud/buds”:

Table

(2)	Limburgish (from Tongeren):	“knoep/ knüp”
	Dutch:	“knop/knoppen”

Limburgish is further characterised by phonological differences to Dutch which include a voiced velar fricative /ɣ/ in onset position (e.g.: /ɣ/ember, ginger) in contrast to Standard Dutch as spoken in the North of the Netherlands where a voiceless uvular fricative /χ/ is produced (e.g.: /χ/ember, ginger). The voiced velar fricative /ɣ/is, however, also part of a regionally flavoured variety of standard Dutch spoken in Limburg and marks Limburgians when speaking Dutch (Cornips 2020a). Limburgish is an umbrella category which includes six main variants and high variation occurs between these variants, especially on a lexical level (Camps 2018).

Speakers of Limburgish and Dutch commonly perceive the two varieties as ‘expressions of two distinct linguistic identities’ (Cornips 2020a, 7). In that sense, Limburgish is considered a ‘natural’ way of expression, mainly for the everyday informal, emotional, and/or familiar domain, whereas Dutch is understood as a ‘neutral’ language that is more suitable for formal affairs like business and education (Cornips 2020a). Speakers may code-switch and translate between Limburgish and Dutch in socially meaningful ways (Morillo Morales and Cornips 2022).

2.2. Early education in the Netherlands

After having introduced the two language varieties and the local context in question, I will now turn to the domain of Early Childcare and Education. Official ECE in the Netherlands includes day-care-centres, pre-schools (so-called ‘peuterspeelzalen’ = toddler play salons), and childminders. Since 2018, all of these different forms have legally been harmonised under the umbrella category of childcare in the legislation of the Ministry of Social Affairs and Employment. Parents, their employers, and the government share childcare costs by means of childcare benefits. The number of children in official childcare has increased from 447 720 in 2012 to 522 920 in 2019 with an average of about 58.6 h of attendance per month in 2019 (Centraal Bureau voor de Statistiek 2020; Rijksoverheid 2019). 89% of toddlers between 2.5 and 4 years old attended some form of childcare in the Netherlands in 2019 which implies that official childcare has a broad reach (Vroom 2019).

Whereas day-care-centres are intended for children between 0 and 4, pre-schools target children between 2/2.5 and 4. Pre-schools have a stronger educative character than day-care-centres and aim explicitly at preparation for the school. Children enter school upon reaching the child’s fourth birthday. Parents can choose one form of childcare for their children or combine different forms, so that some children go to both a pre-school and a childminder or day-care-centre. Additionally, there are certain facilities that carry out specific early and pre-school education programmes. These programmes entail early intervention such as additional time for engagement in pedagogical activities in the pre-school. They are attended by children who are considered to have a so-called ‘developmental delay’, one of which concerns language development (Rijksoverheid, n.y.).

2.3. Language policies and ideologies in early education in Limburg

The national law on childcare was put into place in 2005 with the intention to better organise finances and introduce certain quality requirements to childcare (Vermeer and Groeneveld 2017). The law defines the national language, Dutch, as the working language of childcare facilities and permits the use of the regional languages (Frisian, Low Saxon, and Limburgish) as additional working languages wherever these are ‘in lively use’ (art. 1.55 ‘Wet kinderopvang/Law on childcare,’ own translation). As discussed in section 2.1, Limburgish is clearly ‘in lively use’ in the province of Limburg. However, as common for regional minority languages, it is subject to strong language ideologies which attribute it to the family context and the cultural as well as the emotional domain (i.e. the ‘language of the heart’) rather than to the one of educational and economic achievements (Cornips 2020a).²

These ideologies are also reflected in language policies in pre-schools: teachers commonly use Limburgish to provide emotional support to individual children in one-on-one situations while they use Dutch to address the whole group, in instruction contexts as well as to structure the day at pre-school (Morillo Morales and Cornips 2022). This context-dependent language use has implications for the organisation of attention as Dutch signals to all children to pay attention whereas Limburgish does not require the same. Cornips (2020b) and Morillo Morales and Cornips (2022) have shown that children themselves co-create the dominance of Dutch in childcare facilities. Limburgish-speaking children switch to Dutch as soon as a Dutch-speaking child starts to interact with them. On the other hand, Dutch speaking children never switch to Limburgish in order to learn it. Socialisation into such language hierarchies at that age has been identified to be amongst the reasons that many children actively only speak Dutch themselves even if their parents raise them in Limburgish at home (Cornips 2020b).

This paper takes a participation and co-operative action perspective to understand children’s pre-school socialisation into the local diglossic situation better. The following section briefly introduces this perspective.

3. Conceptual background: participation and co-operative action in language socialisation

Children’s socialisation takes place throughout their participation in a multitude of situations across different participatory roles (de León 2011). This is in line with Goffman’s notion of footing (1979), which constitutes a diversification of the traditional model of hearer and speaker, acknowledging that participation in multiparty interaction can entail different statuses. These different statuses can, as they stand in relation to each other, lead to a variety of participation frames (Goffman 1979). Goffman’s classification of different types of speakers mainly relates to intertextual complexities, calling into question who produces the talk (i.e. the ‘animator’) and who is being quoted (i.e. the ‘author’). With regards to hearers, Goffman distinguishes between ratified and unratified hearers. The ratified category includes both ‘addressed recipients’, i.e. those expected to orient toward the talk and take the next turn, and ‘official hearers’, those who are expected to listen but who are not addressed. Bystanders whose participation is not ratified are divided into ‘inadvertend hearers’, who overhear the talk, and ‘advertent hearers’, who intentionally listen and thereby eavesdrop on the ongoing conversation.

Language socialisation studies have pointed out that children are socialised into participation across different statuses of hearers even when they are not in the position of the addressed recipient (Blum-Kulka and Snow 2002; Chaparro 2020; de León 2011). Despite a long-lasting strong focus on dyadic interaction in language socialisation studies, children commonly spend more time as participants in multiparty interaction than in dyadic interaction which requires them to navigate different participatory statuses (Blum-Kulka and Snow 2002). Children are also socialised when they are bystanders as they can pick up on social roles and different ways of talking, including the situational use of different language varieties, through observing changes in talk that produce and are produced by changes in participation framework (Blum-Kulka and Snow 2002). In addition, overhearing has been found to be a robust means for vocabulary acquisition, and even children as young as two years old are able to closely focus on third-party interaction and draw from it for their own (language) development (Akhtar 2005).

Multimodal studies have highlighted that participation relies on more than talk alone, but is rather achieved through a combination of linguistic and other semiotic means like body positioning, gaze, touch, and gestures (Goodwin 2007b). Furthermore, occasions for participation are dynamically emerging between speakers, hearers, and other participants and not limited to the pre-determined roles defined by Goffman. To grasp the situatedness of participation as a common achievement by all participants who engage together in ‘constitute[ing] their life worlds’ (Goodwin and Goodwin 2004, 240), Goodwin has reconceptualised participation as co-operative action, a perspective this paper embraces. Accordingly, as people interact, they ‘inhabit each other's actions’ (Goodwin 2013). This reconceptualization draws the attention to the collaborative nature of participation and highlights the constant reflexive orientation processes speakers and hearers engage in (Goodwin and Goodwin 2004, 235). For example, Goodwin has shown how a teenage girl and her father collaboratively organise a homework activity interactively, using bodily, material and other resources to organise their participation and the activity (Goodwin 2007b). Building on data of archaeologists making a map of dirt, Goodwin shows how gestures function in an environmentally-coupled manner when they operate on speech and other semiotic resources, and vice versa (Goodwin 2000, 2018).

To that end, socialisation sites that were traditionally seen as classical one-way-input settings have recently been reconceptualised from a co-operative perspective. For example, researchers have described how young children agentively participate in early literacy practices like parental or caregivers’ storytelling through gaze, verbal completions, and gestures (Burdelski 2019; Cekaite and Björk-Willén 2018; Evaldsson and Abreu Fernandes 2019). As such, participation can take many forms, and language socialisation is co-created on a moment-to-moment basis across a variety of participation frameworks across time (de León and García-Sánchez 2021).

4. Methodology

4.1. Research objectives

Based on the discussed research problem and literature, this study has the following research objectives:

• Understand the potential consequences of the use of Dutch, Limburgish, and other semiotic resources for the organisation of collaborative action and participation frameworks, and investigate how children orient to different participatory statuses in bidialectal multi-party interaction in ECE.

• Understand how participation in everyday multi-party interaction in ECE contributes to children’s language socialisation into the diglossic situation of Dutch Limburg.

4.2. Methods

Data for this study stems from 4.5 months of ethnographic fieldwork in a pre-school in the Limburg, the Netherlands which was spread out between October 2020 and May 2021. I generated linguistic ethnographic data (fieldnotes, audio- and video-recordings) in the pre-school on two mornings a week.³ This study is part of a bigger project on language socialisation in the Southern German-Dutch border region which has received ethical clearance by the Ethical Review Committee of Maastricht University.

The pre-school is attended by toddlers between the age of 2;5 and 4;0, most of whom go there twice a week, while some (those with an indication for early intervention) come four times a week. The usual group size is 16 children with two teachers.

While I conducted participant observation and generated data, my own role in the pre-school used to shift situationally. Whereas I was mainly an observer in formalised situations like morning circles, I also acted as an assistant to the teacher with easy tasks like handing out food etc. This facilitated my access in the pre-school. On yet other occasions, I blended in with the children during free play when they welcomed me, which they commonly did – with or without my video camera. I took an ethnographic ethics approach of ‘practices of withness’ (Dennis and Huf 2020), foregrounding my involvement with the community of the classroom. Such an approach leaves room for different positionalities at different times as a chance for building relationships with the children and teachers, and for learning through my own relational entanglements with them (for an extensive discussion of my approach see: Rickert under submission). I myself am not a speaker of Limburgish and usually actively used Dutch (my L2) in the pre-school. In doing so, I certainly contributed to the (re-)production of linguistic hierarchies and specific participatory frames which I will take into account whenever relevant in the following analysis sections.

In the analysis section, I present three extracts of multiparty interaction between the two teachers, Lieke and Helena, and a child, Felix, as well as other participants who vary across the extracts including myself. Child Felix is raised with Limburgish at home, and the teachers are aware of that. In one-on-one situations, the teachers commonly address Felix in Limburgish. Felix himself pre-dominantly uses a regionally flavoured form of Dutch with a few words of dialect sometimes. Felix generally took a dominant position amongst the children at pre-school as he was very pro-active and talkative, and the teachers devoted a lot of attention to him. Focusing on his case provides for consistency in investigating participation frames as they emerge in multiparty interactions that include a child who understands both Limburgish and Dutch.

The examples presented are the result of reviewing multiparty-interactions between the target child, the teachers and others in the audio and video data. Upon making a collection of occurrences (see e.g. Burdelski 2021), I was on the lookout for changes in participation frames co-created between teachers and children that occur linked to code switches. I had previously observed the relevance of code-switching for the organisation of participation throughout my fieldwork. The cases discussed in the analysis section were chosen since they were found to be particularly rich for an analysis of the processes of intimization, staging, and subverting frames which were a common theme throughout the data and will be discussed in more detail in the upcoming sections.

5. Analysis

5.1. Switching languages, shifting frames

In the following section, I discuss a situation in which teachers, toddlers, and I co-create and orient toward different action that is unfolding simultaneously. Extract 1.1 shows the transcript of a video-recording. In the translation, Limburgish is underlined with partly dotted lines while Dutch is not. This mode was chosen as Dutch occurs more often than Limburgish throughout the transcripts.

Extract 1.1 takes place just before the daily fruit break and thereby, at a relevant socialisation site (mealtime, e.g. Cekaite and Evaldsson 2019). Teacher Lieke and child Felix (3;6) stand at the door of the classroom and take a look outside to see if they can spot any easter eggs. The other children are seated around a table. Teacher Helena sits on a stool in the second row and I stand on the opposite side of the table. All children, teacher Helena and me (the researcher) direct their attention to teacher Lieke and child Felix, who are initially looking for the eggs.

Extract 1.1

As teacher Lieke walks back to her chair, she gazes at her colleague to identify the intended recipient of her subsequent speech (Goodwin 1981). Lieke says in Dutch that she always has the same experience (l. 3), i.e. as she explains subsequently, that she could never find eggs at home (l. 5). As can be seen in the picture, teacher Helena and teacher Lieke look at one another while teacher Lieke walks to her chair. Thus, they ‘simultaneously display (…) and propose (…) a state of coparticipation in collaborative action’ (Robinson 2006, 88). Teacher Helena expands her engagement in the ongoing action between the teachers through recycling child Felix’s explanation of why eggs could not be found. Felix explains in line 1: ‘They are hidden for real’, and Helena subsequently draws on this explanation when she tells Lieke: ‘(…) therefore you also don’t see them, they are hidden for real’ (l.4). The repetition is used to enforce Felix’s reasoning in a joking way, as is common in recycling in classroom interactions (Cekaite and Aronsson 2004). Whereas the two teachers already talk to one another as Lieke and Felix walk back toward the table, this conversation remains linked to the prior action of looking for eggs as well as Felix’s (the ‘author’ in a Goffmanian sense, (Goffman 1979)) involvement. As teacher Helena connects to Felix’ prior talk and uses his sentence for her own interactional goals, she engages in a form of ‘format tying’ (Goodwin 1990). Format tying is a form of imitating a prior speaker’s talk in a slightly adapted way in order to fit the current project. As the conversation thereby links to Felix’ involvement, children may, and also do, listen to the conversation. This participation frame changes when teacher Lieke sits down again, as can be seen in the next part of the extract:

Extract 1.2

When teacher Lieke sits down, she keeps talking to her colleague Helena, and also looks at her, i.e. away from the children, as can be seen on the picture of extract 1.2. Now, she starts to tell a story about personal Easter memories, directed at her colleague Helena. Teacher Lieke’s switch from Dutch to Limburgish (l.6) marks the opening of a frame of personal storytelling amongst colleagues, in front of the children. Conversational storytelling can contribute to identity and group membership building (Dressel and Satti 2021). Here, it contributes to the situational construction of adult and child identities and thereby leads to a shift in participation frame. This is co-constructed through topical choice as the theme of nostalgia that the two address by means of the story is not very accessible for the children.

While the two teachers speak in Limburgish, child Mia (2;5) initiates a second, simultaneous collaborative action. In the embodied participation framework (de León 2011), I stand in front of the children while child Mia faces me and the two teachers are located behind her. I have wet wipes in my hands as I am assisting the teachers with some tasks. Cleaning the hands with wet wipes is part of the routine before eating fruit at the pre-school.

As can be seen in the picture in Extract 1.2, child Mia quickly turns around to look at teacher Lieke only to orient toward me afterward again. While gazing at me, she also reaches her arms out to me. Mia’s reaching out can be considered an ‘environmentally-coupled gesture’ (Goodwin 2007a) as it emerges and only becomes meaningful in interrelation with the material environment, i.e. it aims at the wet wipes in my hands. Following child Mia’s embodied turn, I orient toward the collaborative action suggested by her and start to distribute the wet wipes (l.8).

While teacher Lieke tells the story in Limburgish to her colleague, me and the children direct our attention to the activity of cleaning the hands now. In the sense of ‘guided participation’, i.e. interactional routines which structure children’s participation in multi-party interaction (Rogoff 1990), the teachers usually sing a song while cleaning the hands with the children. Felix now agentively breaks with the hierarchy in the pre-school. He supposedly remarked that the teachers themselves were not orienting toward the activity of cleaning the hands and would not initiate the song as usual, and consequently, he reinforces the activity himself through singing in Dutch (l.10).

The situation continues as portrayed in the following:

Extract 1.3

Expressing our orientation toward the collective action of cleaning the hands, the child Leon and me start to join in Felix’ singing in Dutch. Simultaneously, teacher Lieke keeps on addressing her colleague Helena in Limburgish, keeping up the story-telling amongst colleagues that the two have moved into. I, the researcher, actively take part in the co-creation of the simultaneous participation frames. I turn my focus away from the storytelling frame between the colleagues which is co-created in Limburgish and actively shape the simultaneous participatory frame of cleaning the hands. Adding to this, I reinforce the activity by joining in child Felix’ singing in Dutch. In this way, I also orient toward the action taking place in Dutch rather than to the one-on-one interaction co-constructed in Limburgish between the teachers.

At some point, the child Finja tries to claim teacher Helena’s attention by showing her something on her arm (see picture, extract 1.3). It is only when teacher Lieke walks to the door for demonstration purposes related to her story-telling, that teacher Helena takes a quick look at child Finja’s arm. As soon as her colleague starts talking, she directs her attention to her again, and keeps the connection to child Finja merely by touch. Teacher Helena thus engages in ‘bodily emotion socialisation’, mediated by the touch with child Finja (Cekaite and Holm Kvist 2017), and simultaneously orients to her colleague’s storytelling by gaze and body positioning. Looking away can communicate a low engagement in the collaborative action (Goodwin 1981) and Helena shows a higher level of engagement in the storytelling of her colleague.

Shortly after, teacher Lieke’s story comes to an end and she bodily orients toward the group of children again:

Extract 1.4

When Lieke sits down again in extract 1.4, she also gazes toward the children again (see picture). She closes her story, and thereby the one-on-one-frame, in Limburgish: ‘Hoa, I will never forget this’ (l.12). Then, when she redirects her attention to the collaborative action of cleaning the hands, she switches to Dutch by first joining in our singing: ‘♫RITS RATS♫’, followed by the instruction ‘oh clean very well, very well’ (l.12).

The example shows how the teachers might use Limburgish, in combination with other semiotic resources such as gaze to move from a conversation in which the children are ratified into a personal dyadic conversation amongst colleagues. In contrast, they may use Dutch to signal their involvement in co-operative action with the whole group. The example also illustrates that even children who are very young and do not speak yet, like Mia (2;5) can successfully initiate new collaborative action through environmentally-coupled gestures. My position as an adult who orients toward this newly initiated action as well as the subsequent singing in Dutch, legitimises child Mia’s embodied turn and responds to it.

5.2. Staging conversations

In terms of the topic and form of storytelling, the conversation between the teachers in the example of section 5.1 showed many similarities to the personal conversations which the teachers (and I as fieldworker) had during lunches after the children had left. The two teachers are also good friends, and warmly included me in personal conversations throughout my fieldwork. Usually, the teachers used Limburgish in such situations but sometimes switched to Dutch for me. I conclude that the personal conversation the teachers Lieke and Helena move into in section 5.1 could also have taken place in a similar way, and importantly also in Limburgish, between the two of them independent from their communication with the children. However, in their work with the children, the two teachers also engage in another form of dyadic conversation. Namely, they commonly stage conversations between each other which would never happen in the same way if the children were not present. In such conversations, the message the teachers aim to bring about is actually intended for the children to be overheard (Goffman 1979) rather than solely directed to the other teacher. The teachers say to one another, in Dutch, e.g. statements like ‘I don’t know what we can still teach them. They really know everything about the animals on the farm’ (from fieldnotes 5/5/2021).

The following situation is an example of such a staged conversation. At first, Felix explains to the teachers why you should run away when the Easter bunny comes, and subsequently, the teachers Helena and Lieke engage in a dyadic conversation about the same topic as well as about the children’s performance on the topic of Easter in front of the children.

Extract 2

All speech takes place in Dutch in this extract. In the first part (l. 1–5), teacher Lieke aligns with Felix who explains that you would have to run away when the Easter Bunny comes so that he does not see you. Teacher Helena takes a questioning position (l.2: ‘run away?’), leading to Felix’ and Lieke’s collaboration on the reasoning (l.3, Felix: ‘(…) because otherwise he sees us and that’s not allowed.’; l.4, Lieke: ‘we may not see what he is hiding’; l.5, Felix: ‘no, that’s a surprise’). Lieke and Felix enforce the argument mutually here, resulting in a triadic constellation in which Helena pretends to be in the learning position. Some of the other children follow the conversation and look toward the speakers, respectively, while others focus on finishing their fruit or taking looks into each other’s fruit boxes.

In the second part, Lieke introduces a change in participation frame through starting to talk about the children to Helena rather than with them. This manifests in her use of third person plural when she says ‘(…) they are right’ (l. 6). With the personal pronoun ‘ze/they’, Lieke refers to all children here, even though it was only Felix who explained how to behave when seeing the Easter Bunny. Now, a participation frame in which the two teachers play an active, conversing role, while the whole group of children becomes intended overhearers, unfolds. From a language socialisation perspective, overhearing is an exercise in observation, attention, and inference as well as participation for young children (de León 2011).

Helena animates Felix’s prior explanation of hidden eggs being a surprise (l. 7) again. By then, most children direct their attention toward the teachers’ conversation, as can be seen on the second picture. One child even turns around to look to the teachers and participate in the ongoing action in an embodied way. The two children who do not seem to follow the teachers’ conversation are aged 2;5 and 2;8, respectively, and are not yet socialised into the organisation of attention to the same extent as the others. In the last two utterances of the excerpt, the teachers indirectly praise the children for their knowledge, jokingly expressed in form of a pretend-complaint by Helena: ‘(…) then again, they know all about the next topic’ (l.9), and a pretend-justification by Lieke: ‘yes well, they asked (…) I have to give them an answer’ (l.10). I suggest the terms ‘pretend-complaint’ and ‘pretend-justification’ here as the acts are clearly staged and performed on issues which do not require a serious complaint or justification.

In contrast to the situation discussed in section 5.1, where the teachers moved into a personal conversation in front of the children, the conversation between the teachers in the example at hand is not intimised as they seem to intend the children to hear their praises. This is enacted, in one way, by the use of Dutch instead of Limburgish. Dutch is the variety the teachers usually use to communicate to the whole group and Limburgish the teachers’ default variety for personal conversations. The teachers’ bodily orientation also suggests ratified participation on the side of the children. This can be seen on the second picture of Extract 2, in which Lieke and Helena do not change their body positioning but just gaze at one another as they move from the triadic conversation with Felix into the part where only the two of them speak. Their way of talking about the children using third person plural, however, makes clear that the situation turned into a dyadic conversation between the teachers now. The children may and should participate as ratified listeners but are not expected to participate as speakers, also since the teachers do not gaze at the children but instead at one another. Most children take up their role as overhearers by silent participation as well as their bodily alignment and gaze toward the teachers.

5.3. Subverting frames

In the previous examples, I showed how participation frames initiated by the teachers were, to a large extent, co-created by the children who (re-)organised their attention in correspondence to the emerging frames. However, children are agentive beings who can also challenge frames and claim different forms of participation for themselves. In section 5.1, I showed how children actively initiated and co-constructed a simultaneous frame to an intimised one between the teachers. In what follows, I discuss how a child challenges a frame that is being established between the adults in the pre-school and claims participation in their interaction.

The audio-recorded situation takes place during pick-up time at the end of the day at pre-school. The children and teacher Helena stand in front of the window to spot arriving parents. Teacher Lieke accompanies children whose parents have arrived to the gate. In the following, Felix and teacher Helena discuss by which means of transport Felix’ mum would come, when teacher Lieke enters the classroom after having brought the child Ilya to his dad:

Extract 3 (NEL: Dutch; DIA: Limburgish dialect)

As annotated on the left side of the transcript, the situation unfolds in three major frames. In part A, child Felix discusses with teacher Helena if his mother will pick him up by car or rather by bike. The conversation between the two of them takes place in Dutch. When teacher Lieke enters the classroom (part B), she starts to engage her colleague Helena in a conversation amongst colleagues and adults in Limburgish. Here, the two teachers comment on the child Ilya’s positive attitude toward staying at pre-school. Helena reports that Ilya would even have already said that he would go home with teacher Lieke, an information that child Felix alludes to subsequently. Part 3 of the situation is marked by Felix turning around to orient to the teachers and loudly, in Dutch, asking ‘Really?’ (l.9). After he joined the conversation, it proceeds in Dutch.

When teacher Lieke first enters the classroom, she and her colleague Helena enact an inter-adult/colleague frame in a comparable way as discussed in section 5.2. They talk in Limburgish and additionally embark on a topic pertinent to their teacher role, i.e. an observation of the children’s behaviour. I, as an adult in the pre-school, feel eligible to participate in the conversation and contribute with a confirmative question ‘No?’ in line 6. While my participation seems legitimated as it does not lead to any emotional interactional consequences by the teachers, Felix’ sudden active participation seems out of place in the conversation. As he turns around and suddenly asks ‘Really?’ in Dutch (l.9), he subverts the participation frame that was previously mainly achieved by the teachers in Limburgish. The teachers’ co-creation of a participation frame that is set up amongst the colleagues, then makes him an eavesdropper in part B of the transcript, the part the teachers may make sense of as ‘their’ conversation. In comparison to overhearers, eavesdroppers take this role in the participation framework without the speakers being aware that someone listens to them (Goffman 1979).

Beside the code switch that Felix introduces when he enters the conversation, his position as a child also plays a role in his making of participation statuses. It is usually not in his capacities to evaluate his peers’ behaviour. Goodwin and Kyratzis (2007) note that children in peer conversations frequently make use of linguistic resources from the adult culture like control act forms to contest social hierarchies. During fieldwork, I have observed several instances where Felix commented on a peers’ behaviour and was indirectly or directly told not to do so by the teachers. Now, he claims ratified participation in a conversation in which the teachers discuss the case of a particular child, which contributes to the subversion of the frame.

The unexpectedness of Felix’s claim of (focal) participation manifests as teacher Helena flinches and makes a surprised ‘ps::::’-sound (l. 10) at the same time as I, who observe the situation, start laughing (l. 11). Felix’ initial question seems to require explanation after teacher Helena’s and my reactions, so that Felix subsequently elaborates on the reasons why you cannot take children home in a low voice (l. 12). Teacher Helena aligns with his explanation and confirms, in Dutch, that she has a very small car. Child Felix has successfully subverted the participation frame and his contributions as a – now ratified – participant gets taken up by the adults who continue with him in a triadic participation frame in Dutch.

6. Discussion

Official language policies in Limburgish Early Childcare and Eduation centres commonly only make a difference between Dutch as main language and Limburgish as a language that might be used with individual children, without taking the complexities of multi-party frameworks into account. As common in diglossic situations, this language policy reflects an ideologically-grounded local status imbalance (Schiffman 1993) which children are socialised into. Scholars of language socialisation and language shift have brought to the fore that children are socialised into ways to handle this imbalance, e.g. ‘restricting use of a particular language to particular domains (…) and cultivating proficiency in a particular language as a means of coping with entrenched social hierarchies’ (Garrett 2011, 516). As the data shows, both cases apply in Dutch Limburg; The intimization of a participation frame performed by the teachers in Limburgish indexes the use of Limburgish for the private domain, whereas the use of Dutch for staged conversations enforces the status of Dutch as a relevant language in the educational domain. This confirms Cornips’s (2020b) results, which show that Dutch is the dominant language in bidialectal pre-school in Netherlandic Limburg, and the language used for group situations. These, as I have shown, also include situations which might on the surface look and sound like conversations between teachers, but in which the children take the status of intended overhearers (see section 5.2). In such situations, silence constitutes a relevant form of participation on the children’s side (see Schultz 2009).

Previous literature on bilingual pre-school has demonstrated how children constantly move in and out of one another’s conversations (Chaparro 2020). My analysis shows that teachers engage in similar practices when they move into more private/intimate or professional conversations amongst colleagues and then move back to another participation frame, engaging with the children. When doing so, they use linguistic resources in distinct ways to construct their professional identities as educators (Ochs 1993) and their identities as befriended colleagues. ‘Teacher talk’ (Huth 2011) is performed in Dutch, while ‘(semi-)private talk’ may happen in Limburgish.

The child Felix’ participation in all cases presented in this paper highlights the complexity of the co-creation of participation frames: As Felix knows Limburgish, it becomes clear that opportunities for participation are much more complex than being simply a matter of language variety. Felix is expected to react in situations in which he is addressed by the teachers in Limburgish but not when Limburgish is used as a means that the teachers deploy to enact a personal or professional conversation amongst colleagues or adults.

Previous studies have shown how overhearer statuses can be socialised (de León 2011; de León and García-Sánchez 2021). Chaparro (2020, 14) found such overhearer statuses and their inherent silent participation to be ‘a critical part of both second language socialisation and socialisation into a bilingual classroom community’. While my findings resonate with this, it is important to say that in pre-schools in bidialectal Limburg, Limburgish may also be used to arrange for non-participation, as demonstrated in section 5.1.

7. Conclusion

In this article, I discussed how teachers and toddlers use their linguistic and other semiotic resources to shape situational interactional participation in pre-school interaction. I investigated the role of Dutch and Limburgish, respectively, as well as that of other semiotic resources, for the co-creation of participation frames.

It resulted that teachers may use Limburgish to move into a one-on-one conversation in front of the children whereas they use Dutch for staged conversations which they intend to be overheard by the children. Hence, code switches, in combination with other interactional resources, have consequences for the organisation of attention. Participation involves constant attentiveness and attuning of the participating actors (Cekaite and Björk-Willén 2018). A child suddenly claiming focal participation in a way that is not attuned to the current participation framework might therefore situationally seem out of place for other participants, like in the situation in extract 5.3. The way the teachers switch from Limburgish to Dutch when child Felix enters the conversation there, confirms the importance of code choice for establishing participation frames.

Children actively co-shape participation frames and thereby inhabit the action of peers and teachers (Goodwin 2013). While such a co-shaping can, in practice, mean an orientation toward a certain frame, and a reproduction thereof, it is also characterised by children’s agency (Schwartz 2018). Children can, e.g. subvert participation frames which are set up amongst adults and claim ratified participation. Furthermore, children succeed, even at a very young age, in initiating new collaborative action through embodied means and the performance of environmentally-coupled gestures. The analytic approach of participation frames and co-operative action provides a way not to prioritise speech by starting from a specific linguistic code (Chaparro 2020) but rather to embrace the multi-modal dynamic organisation of interaction. As such, it can account for young children’s ways of communicating through embodied means and their ongoing language socialisation. For example, in situations which entailed staged conversations, the children’s silence constituted a relevant form of participation. While this paper focused on multiparty interaction which include teachers, the question as to how peers enact similar shifts in participation frames remains open for future research.

Socialising interactional participation is an important part of children’s becoming of members of the bidialectal community they live in. A socialisation into a specific code choice across participation frames and contexts thus impacts children’s own bidialectal competences and language attitudes, ultimately in favour of the national language Dutch.

Acknowledgements

I would like to thank Leonie Cornips and Gunther De Vogelaer as well as two anonymous reviewers for helpful comments on earlier versions of this paper. Thanks to Sara Atwater for proofreading the manuscript. All remaining shortcomings are my own.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 847596.

Notes on contributors

Marie Rickert

Marie Rickert is a PhD candidate at the Faculty of Arts and Social Sciences, Maastricht University and the Institute of Dutch Philology, University of Münster. She researches language socialisation in early education on both sides of the German-Dutch border. Her wider research interests include language ideologies in education, multilingualism and multi-modality, and L2 acquisition in the context of linguistic diversity and migration.

Notes

1 While Limburgish officially has the status of a regional language, the speakers mainly refer to it as dialect (‘dialect’ or ‘plat’, see Cornips, 2020b).

2 It must be noted, however, that the boundaries between dialect and Dutch are not always clearly defined, as Dutch can resemble dialect phonetically sometimes, and speakers also use mixed forms.

3 The data includes 102:19 h of audio-recordings of both formal and non-formal activities throughout the day as well as 15:01 h of video data complemented by extensive fieldnotes.