Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

A classroom is often conceptualised as a site where only one teacher is interacting with a class of students and the interactional roles are predefined (Creese 2006). However, in a team-teaching approach, two teachers jointly instruct the students (Dafouz and Hibler 2013). In bilingual classrooms, team teaching can be challenging, especially if the teachers are assigned different language roles (e.g. Dillon et al. 2015). We know, however, only a little about how languages are actually used by collaborating team teachers: “[T]here is a lack of discourse-based classroom research examining how teacher collaboration occurs” (Park 2014, p. 35). Although the research field on e.g. translanguaging practices in bilingual classrooms is rapidly growing as part of a paradigm shift in bilingual education in the twenty-first century (García 2009), the focus is often on a single teacher interacting bilingually with student(s) and on elementary (or later) school contexts (see e.g. García and Li Wei 2014). Moreover, in most studies on team teaching with a language focus, the language of instruction is English only (see further below). Importantly, only a few studies are available on preschool teachers’ bilingual language use (see Gort and Pontier 2013; Pontier and Gort 2016; Schwartz and Asli 2014).

Table 1 The structure of the circle times: activities, leadership, language allocations and participant roles
Full size table

There are a growing number of guides for teachers that describe how to implement team teaching in general (e.g. Sileo 2011), but the guides mostly lack any description of how to use language(s), especially in bilingual classrooms. Dillon et al. (2015) have reported on teachers’ uncertainty on how to implement team teaching in a bilingual classroom context. In our study, we seek to answer this question by focusing on the dynamic patterns of interaction found between teachers in a bilingual preschool classroom in Finland. As the context, we have selected a team-teaching routine that is typical in Finnish preschools, circle time. In this classroom, one of the two kindergarten teachers had a predefined role as a Finnish speaker while the other had a bilingual, Swedish/Finnish-speaking role. We will analyse the participant roles in terms of leadership and possible changes in the roles, as well as how language use was allocated during two team-taught circle times. We will also make a more detailed analysis of the interaction between the teachers and relate this to their predefined language roles. Before introducing the study, we will address the concept of team teaching, focusing especially on bilingually team-taught classrooms, and explain the sociocultural and educational context of the preschool.

2 Bilingual Team Teaching and its Implementations

Team teaching is often used as an umbrella term for a range of different approaches to teacher collaboration. These approaches are in the literature labelled co-teaching (e.g. Cook and Friend 1995; Friend 2008; Friend et al. 2010; Pontier and Gort 2016), team teaching (as a more specific term, see e.g. Friend et al. 2010; Dafouz and Hibler 2013), partner teaching (e.g. Bronson and Dentith 2014) and collaborative and cooperative teaching (see e.g. Liu 2008, p. 105). The terms are sometimes used interchangeably by the same author(s), and they are not always accompanied with strict definitions of the approach for which they are being used. Common to all the approaches is, however, that two or more teachers share responsibilities, while the extent to which and how the teachers are expected to collaborate in the stages of planning, organisation, instruction delivery and evaluation varies (cf. e.g. Dafouz and Hibler 2013; Friend et al. 2010; Liu 2008; Perry and Stenwart 2005; Sandholz 2000). For the purpose of this study, we use the umbrella term team teaching. In the analysis section, we apply the definition given by Dafouz and Hibler (2013, p. 97), who define team teaching as “a pedagogical approach where two teachers collaborate simultaneously and share in the instructional process for the same group of students within a given subject matter in the same classroom”. We thus focus on what is going on inside the classroom as a part of the actual delivery of instruction, rather than the phases and processes of planning, organisation or evaluation. This is due to the fact that the data collection design (classroom observations) only allows us to discuss the delivery stage of team teaching.

With regard to the status and educational role of the teachers, some researchers find it essential for team teaching that the teachers enjoy the same status, as is the case when two professional teachers work together and share the teaching responsibilities (e.g. Friend et al. 2010). Other researchers also include cases where one main teacher and one teaching assistant collaborate in the classroom (see e.g. Carless 2006). The teachers in the preschool classroom that we examined shared the same professional background (qualified preschool teachers) and had the same status and educational role. In the following, we first present previous research on team-taught bilingual classrooms in general, and after that we discuss the bilingual language practices that have been applied in preschool classrooms.

2.1 Previous Research on Bilingual Team Teaching

Team teaching has been argued to have great potential for bilingual education (Bahamonde and Friend 1999; Dillon et al. 2015). However, the vast majority of research on team teaching with a language focus has been conducted in language classrooms where English serves as both the goal and the means of instruction. They are thus not bilingual classrooms perse ==> per se, although at least one of the teachers is bilingual. One strand of research concerns EFL (English as a Foreign Language) classrooms, for example in Asia, where a native English teacher and a non-native English teacher collaborate (e.g. Carless 2006; Liu 2008; Park 2014), or English-medium CLIL (Content and Language Integrated Learning) classrooms (Dafouz and Hibler 2013). Another body of research examines mainstream English-medium classrooms where a bilingual ESL (English as a Second Language) teacher assists the class teacher (e.g. Arkoudis and Creese 2006; Gardner 2006). Although other languages than English are occasionally reported to be used during the team teaching (e.g. in Carless 2006), the language used for interaction between the teachers is almost always only English.

Studies from team-taught EFL classrooms provide evidence of the benefits as well as the challenges of team teaching. Carless (2006) and Liu (2008), who observed and interviewed teachers and students from different team-taught EFL classrooms, found that students in team-taught classrooms had multiple opportunities to listen to and speak in English and to observe the two teachers demonstrating dialogues or question and answer routines. However, Careless and Liu also found that one teacher’s relative lack of ability in the other’s native language could be a barrier to collaboration. Other challenges that have been reported include the anxiety and confusion that can arise as a result of uncertainty as to which role each teacher should have in the classroom (ibid.), or as a result of the collaboration having been imposed on the teachers (Davison 2006).

Mainstream or CLIL classrooms also add to the challenge the question of how to integrate and distribute teaching resources to address content as well as language (Arkoudis and Creese 2006; Dafouz and Hibler 2013; Dillon et al. 2015). With reference to children under school age, Dillon et al. (2015) present a study carried out on team teaching in a dual language (English and Arabic) early years education programme in Abu Dhabi. In this programme, two teachers were placed in each kindergarten classroom: an English-medium teacher with primary responsibility for certain subjects (e.g. English and Maths) and an Arabic-medium teacher who was responsible for other subjects (e.g. Arabic and Islamic Studies). The programme had been in a state of “constant change and fluidity” (ibid., p. 22) since a recent reform of national education and the survey carried out as part of the study therefore had as one of its aims to explore the teachers’ experiences of team teaching so far. The findings showed a generally positive attitude among the teachers as far as collaboration and partnership were concerned (although some teachers were more positive than others) and the belief that a team-teaching environment “can better facilitate the learning process” (ibid. p. 28). However, the teachers asked for more training on models of team teaching as they felt uncertain about how to work with them. Dillon et al. (2015) further found that the challenge in the “area of teaching Arabic and English Literacy simultaneously in an interwoven co-constructed dialogue continues to be an area for concern” (p. 30). The need for further research on actual language practices in team-taught bilingual preschool classroom is thus urgent.

Predefined teacher roles and tasks may carry with them certain expectations on teacher discourse practices in the classroom, albeit with allowance made for deviations from these expectations. Park (2014) was particularly interested in the dynamics of team teaching, floor alternation and how teachers jointly coped with unforeseen instructional and interactional issues. The setting was a Korean EFL elementary school classroom which was team taught by a native English-speaking teacher and a non-native English-speaking teacher. The main objective of the study was to examine the participation patterns in which the teachers interacted with students and with each other. In contrast to e.g. Dafouz and Hibler (2013), where the teachers had fairly predefined and stable roles in the classroom, in this classroom the roles in terms of leadership changed as they alternated in taking the floor and leading an activity: when one of the teachers played the lead role, the other stepped aside and acted as “a silent but vigilant co-participant by remaining attentive to the ongoing talk”, but they intervened whenever appropriate (ibid. p. 36). The teachers made use of each other’s language competencies as a resource in the unfolding discourse. Hence, roles can be predefined and certain constraints set for action, but as Park convincingly showed, the leadership may also shift during the course of a class (see also Gardner 2006).

2.2 Language Practices in Team-Taught Bilingual Preschool Classrooms

A common distinction used in the literature for language practices in bilingual teaching is, on the one hand, that of language separation and, on the other, the flexible use of two languages (e.g. García 2009). Language separation is common in bilingual education programmes with an additive framework, frequently used examples of which are immersion bilingual education and dual language education in the USA, which showcase separation practices on many levels: teacher-determined (one teacher–one language), time-determined (half a day in one language and half a day in the other), place-determined (different classrooms for different languages) and subject-determined (different subjects taught in different languages) (García 2009, p. 292ff). A recent development is the increasing acceptance in educational settings of the use of flexible language practices. An increasing number of studies have examined flexibility in language practices in bilingual school settings. In many studies, these practices are labelled translanguaging, referring either to a bilingual teaching methodology where input and output are systematically varied, or to a scaffolding approach designed to engage (emergent) bilingual students and use their bilingual resources as strengths (García 2009; Lewis et al. 2012). In our study, we use the term flexible language practices to refer to how two languages are used in combination and concurrently by the teachers in the classroom.

Studies on language practices in bilingual team-taught preschool classrooms are rare. One exception is the study of Gort and Pontier (2013), who looked at a Spanish-English dual language preschool in the US. Although the research focus was on how the teachers mediated bilingual interactions with the children, and thus not on team teaching as such, the examples that were examined also included cases where two teachers created some joint dialogue. In the first of the two classes they looked at, a language separation policy was followed. There they found examples of the application by the teachers of a separation strategy that was labelled tandem talk, i.e., “a type of collaborative bilingual practice where a pair of speakers coordinates the use of two languages so that each maintains the use of monolingual speech in a bilingual conversation” (Gort and Pontier 2013, p. 234). In the other classroom, which followed a language-by-time-of-day separation strategy, Gort and Pontier showed how one of the teachers temporarily departed from the target instructional language (Spanish) to help a child to connect to previous experiences and support the child’s engagement (in English). The other teacher maintained the target language (Spanish) and continued the meaning negotiation, acknowledging the contributions just made by the teacher and child in English. In a more recent study, Pontier and Gort (2016) found flexible bilingual as well as monolingual performances by both teachers, despite predefined monolingual roles during shared readings of English and Spanish storybooks. The teachers’ performances “reflected their collective bilingual repertoire, drawing on their distributed bilingual expertise” (Pontier and Gort 2016, p. 96). Language use was found to depend partly on each teacher’s official language designation (whether as a monolingual English or a monolingual Spanish model for the children), but also on the children’s contributions as well as on the two teachers’ collective content and pedagogical knowledge as to how best to support children’s oral language, vocabulary and narrative genre development. In a dual-language Arabic/Hebrew kindergarten in Israel (Schwartz and Asli 2014), the Hebrew-speaking teacher and the Arabic-speaking teacher, both bilingual to a certain extent, used their languages flexibly, making use of what Schwartz and Asli label translanguaging strategies, i.e. the frequent use of inter- as well as intra-sentence code-switching, within as well as across teacher turns. In contrast to the tandem talk found in Gort and Pontier (2013), where the use of monolingual speech was maintained, the teachers in the Schwartz and Asli study thus “shared the instruction by following each other and not separating the languages” (2014, p. 27).

In our study, we will examine in detail the dynamicity between two teachers with different predefined pedagogical and language roles, during one particular type of preschool routine, circle time. The insights we gather will add to the hitherto small body of research that has been carried out on bilingual team teaching in preschool settings.

3 The Finnish Setting

In order to situate the bilingual team teaching addressed in this article in a broader socio-linguistic and sociocultural context, we will briefly describe the status of the two languages used in the preschool, Finnish and Swedish, as well as the Finnish educational system. We will also give an overview of Finnish preschools as an educational context in general and of the target preschool in particular.

3.1 The Sociocultural and Educational Context of Finland

Finland is a bilingual country where Swedish is, by status, an official language equal to Finnish, but a minority language in terms of numbers of speakers: at the end of 2015, 88.7% of the population were registered as Finnish speakers, 5.3% as Swedish speakers and 6.0% as speakers of other languages (Official statistics of Finland 2015). A majority of the 316 municipalities are officially unilingual Finnish-speaking (266 municipalities); the remainder are either bilingual (32 municipalities) or unilingual Swedish-speaking (17 municipalities). Although many citizens are bilingual, the Finnish social system is to a large extent built on separate Swedish- and Finnish-speaking institutions, which produces what can be described as a system of parallel monolingualism (Heller 1999). Schools are administratively either Finnish or Swedish medium and the other national language (Swedish in Finnish-medium schools, Finnish in Swedish-medium schools) is a compulsory subject from the age of 12 years. Although attitudes towards Swedish among speakers of the majority language are today generally positive, there are certain political movements arguing for reducing the status and space of Swedish in Finland (e.g. Lindgren et al. 2011; Hult and Pietikäinen 2014).

As to early childhood education and care (ECEC), the authorities in Finland are obliged to provide ECEC for all children under school age (7 years). ECEC comprises day care for children under the age of six and a Reception year for six-year-olds. In this article, however, we use ‘preschool’ to refer to pre-primary education for children aged 1–6 years. Governed by legislation, preschool services must be offered in the official languages of Finland, Finnish and Swedish, as well as the heritage language Sámi (mainly offered in the Sámi region in northern Finland in government-funded language nests). As stated in the National Core Curriculum for Pre-Primary Education (NCCPE), preschools should also support the language and culture of children with a Romany background or an immigrant background, as well as of those using sign language (NCCPE 2010). In some Finnish-Swedish bilingual areas in Finland there are language immersion preschools that operate in Swedish for Finnish-speaking children and in Finnish for Swedish-speaking children (see e.g. Björklund and Mård-Miettinen 2011). Some preschools operate entirely or partly in a foreign language, mostly in English, but also in French or Russian, using a CLIL approach. The framework of CLIL ranges from immersion-like high-intensity long-term programmes to low-intensity short-term forms, the latter referred to in Finland as ‘language showers’. According to Nikula and Marsh (1997, pp. 24–26) and Mehistö et al. (2008, p. 13), language showers are weekly or daily foreign language activities intended to familiarise children with a foreign language and to develop positive attitudes to language learning.

Debates about language education policy in Finland are only rarely and implicitly concerned with preschool education. The ideology of separating languages in both educational and family contexts is, however, strong in Finland, especially as regards the two national languages, Finnish and Swedish. Both language groups have their own schools and there is a consensus among teachers (see e.g. Palviainen and Mård-Miettinen 2015) as well as among parents (Palviainen and Boyd 2013) that the best model to use to stimulate early bilingual language development is a one-person one-language strategy (e.g. Barron-Hauwaert 2004).

3.2 Finnish Preschool As an Educational Context

In Finnish preschools teachers usually work in teams, in contrast to primary school classes, which are typically taught by one teacher. The official staff-child ratio in a preschool group with children under 3 years old is one staff member to four children, and with over 3-year-olds the ratio is one to eight (Act on children’s day care 239/1973). In preschools, children are either divided according to age into under 3-year-olds and 3–6-year-olds or placed in so-called sibling groups, where siblings of the same family are placed in the same group regardless of their age. A typical Finnish preschool group has three adults and 21–25 children (or 12 children if they are all under the age of 3). (Kirves and Sajaniemi 2013, p. 94).

Finnish early childhood education practices and pedagogy combine care, education, and teaching (the so-called Educare approach, see e.g. Kalliala and Tahkokallio 2003). The staff team in a Finnish preschool group is multi-professional, typically including kindergarten teachers with a tertiary level degree (Bachelor of Education or Master of Education or Social Science) and nurses or social educators holding a vocational qualification in the field of social welfare and healthcare. Some preschool groups have temporary assistants and trainees; these are, however, always supervised.

The teacher-initiated activities in a preschool group are usually planned as well as carried out on a team basis (Venninen et al. 2014). The way the team distributes the work is at least to some extent connected with the qualifications and level of responsibility of the team members. Since teachers are pedagogically responsible for the activities that are organised in preschool, one of them usually does the detailed planning on their own and the other staff members share in implementing the plan. Many teams apply a rotating responsibility for pedagogical matters on a monthly or weekly basis, or a rotating system based on routines (e.g., one is responsible for circle time for an agreed length of time, and the other for working in small groups).

It is emphasised in the national preschool curriculum (NCCPE 2010) that there should be flexibility in the organisation of preschool activities, with the children working at different times in large groups, small groups and individually. Large-group activities have been found to be rare (Ojala and Talts 2007). The routines in a Finnish preschool group are basically the same as in preschools in any country: circle time, small group activities, and free and adult-led play indoors and outdoors (cf. Ojala and Talts 2007; Zaghalawan and Ostrosky 2011). As part of the daily routine, children in Finland have three meals (breakfast, lunch, and a snack) and rest for approximately one hour in the afternoon.

In this article, we focus on the circle time routine. Circle time is a common daily whole-group routine that often, but not always, takes place in the morning. It often lasts from 10 to 30 min (Emilson 2007; Zaghalawan and Ostrosky 2011) and is typically led by one teacher, although others may also contribute. Circle time can thus be described as a formal routine during which the discourse is controlled by the leading teacher, often clearly sitting apart from the others, for example on a separate chair. The children sit in a semicircle around the leading teacher and the other staff members sit either among the children or outside the circle, participating in the activities and disciplining the children. The aims of circle time have been found to be social, cognitive, and informative: the children are training various social and conversational skills as well as learning academic content (e.g. Emilson 2007; Zaghalawan and Ostrosky 2011). Various case studies have reported that the activities during circle time most often include songs, reading aloud, discussions, drama, arts, exercises, play, the weather, and the roll call (e.g. Rubinstein Reich 1993; Zaghalawan and Ostrosky 2011), and many of them include ritualised elements such as a calendar ritual (cf. Maloney 2000).

3.3 The Preschool Context in this Study

The bilingual preschool classroom examined here was situated in a preschool unit in a Finnish-speaking municipality with around 130,000 inhabitants. Out of these, around 300 were registered as Swedish speakers. The preschool unit housed two sections: a Finnish-medium section for children from Finnish-speaking families and a Swedish-medium section for children from Swedish-speaking (or Swedish-Finnish-speaking bilingual) families. The two sections operated separately but were led by the same director.

The classroom under study here belonged to the Finnish-medium section, where bilingual pedagogy was introduced in 2012. The aim with this pedagogy was to familiarise monolingual Finnish-speaking children with the other national language, Swedish, and to develop positive attitudes towards language learning. A bilingual Finnish-Swedish kindergarten teacher was appointed to develop and implement the bilingual pedagogy. The staff therefore had predefined language roles: the bilingual teacher was to communicate bilingually in Finnish and Swedish with the children, while the Finnish-speaking teachers would use Finnish.

The activities and routines in this preschool were typical of a preschool in Finland (see above). The three kindergarten teachers had rotating responsibility for planning the activities. The routine that is focused on in this study, circle time, was always team taught by at least two of the teachers, one of them acting as the leader while the other was a co-teacher. In the current study, one of the Finnish-speaking teachers was in charge of both circle times, with the bilingual teacher as a co-teacher. Thus, the staff not only had predefined language roles but also predefined pedagogical roles.

Taking these predefined language and pedagogical roles as the point of departure, the general aim of the present study was to examine how the two teachers cooperated bilingually in two team-taught circle times. More specifically, our research questions were: How can the team teaching dynamics be described in terms of (changes in) leadership, participant roles and turn-taking? How are the predefined language roles reflected in the teachers’ actual language use and how does the language use relate to previous studies on language strategies related to separation and flexibility?

4 Methods and Material

4.1 Research Design and Data Collection

The data analysed in the current study was collected as part of a larger ethnographic study in which we followed the implementation of bilingual pedagogy in the Finnish preschool for two academic years (August 2012–May 2014). In the first year of the data collection there were 22 monolingual Finnish-speaking children in the group, aged between 1 and 6 years, and three teachers: the bilingual kindergarten teacher and two Finnish-speaking kindergarten teachers. In the following year the group expanded to include 28 children and another bilingual teacher, a nurse with training in children’s care and upbringing, joined the class.

The focus of the study was on the bilingual kindergarten teacher, who was responsible for developing the pedagogy. The data included three 60-min interviews with her as well as two half-day video-recorded observations of her interaction with the children. The interviews, which have been reported in more detail in Palviainen and Mård-Miettinen (2015) and Palviainen et al. (2016), were of a stimulated recall type: the bilingual teacher watched the video-recordings of her interaction with the children in small group activities that she had led and commented on and explained her own bilingual language practices. The principles she described herself as following and the practices she applied are explained further in the following (Sect. 4.2).

As for the recordings, the first one was made in December 2012, and a follow-up recording was made one year later, in December 2013. The video-recorded observations included different preschool routines such as meals, small group interaction and circle time. Although most of the recorded data included only the bilingual teacher and the children, there were also activities in which other teachers were present. For the purpose of the current study, we selected the two circle time sessions which involved the bilingual kindergarten teacher, Johanna, and the Finnish-speaking kindergarten teacher, Heikki. As the data does not include any interviews with Heikki and the interviews with Johanna did not include questions on teacher collaboration, the focus of the current study is an examination of the instruction delivery – what is observed in the classroom – rather than the stages of planning, organisation or evaluation.

The circle time routine was selected as it is typically team taught in Finnish preschools. In this case, it involved one teacher with a predefined bilingual role and one teacher with a monolingual role. The routine is, further, fairly structured, and includes several activities and rituals and set educational roles, which makes it suitable for the analysis of teacher collaboration. As there is limited previous research in this particular area, we developed an analytical framework which included the following categories: activity, participant roles, language allocation and interaction (see Sect. 4.4). First of all these categories are explored from a quantitative point of view (Sect. 5.1), and then the categories are elaborated on and examined in more detail, showing how the teacher collaboration played out in practice in terms of turn taking and language use (Sect. 5.2).

The recording of Circle time 1 was 30 min in length and of Circle time 2 was 29 min. The video-recorded data was transcribed by one of the authors and checked by the other two. The transcription key can be found in the Appendix.

4.2 Participants

The two teachers who were team teaching the circle times in our data, Johanna and Heikki, were qualified preschool teachers with long experience. The teacher with the predefined bilingual role, Johanna, was a bilingual kindergarten teacher with 20 years’ experience of Swedish-medium and Finnish-medium preschool education in Finland and in Sweden, as well as Swedish immersion in Finland. She grew up in Sweden in a Finnish-speaking family and went to a mainstream Swedish school. She completed a lower degree in childcare in Sweden, moved to Finland, and completed a Bachelor’s degree in the field of childhood education in Swedish in Finland. She also did a module in pre-primary and primary education and enrolled in in-service courses on immersion education. By the time of the data collection she had lived and worked in a Finnish-dominated area of Finland for over 15 years.

Johanna had the main responsibility for planning and implementing bilingual pedagogy in the class in all types of activities. Previous analyses of her bilingual practices when leading small group activities have shown that she followed certain principles in her use of the two languages: (a) she flexibly and continuously alternated between the two languages, (b) she used the two languages to equal amounts across activities, (c) she adhered to responsible code-switching, avoiding direct translation between the two languages and using the two languages for different purposes, (d) she used rich scaffolding structures (body language, contextualisation, verbalisation of actions, repetition of words and routines, etc.), and (e) she adjusted her bilingual speech to the needs of individual children (for more details see Palviainen et al. 2016). She also emphasised the importance of acting as a role model of a bilingual speaker.

The teacher with the predefined monolingual Finnish-speaking role, Heikki, was a Finnish-speaking kindergarten teacher with 29 years’ experience in Finnish-medium preschool education in Finland. He held a Bachelor’s degree in the field of childhood education from a Finnish-medium university, which he had complemented with a module in pre-primary and primary education and several in-service courses on preschool education. He grew up in a monolingual Finnish-speaking context but studied Swedish as a subject in secondary school (6 years, two classes a week). He had a positive attitude towards Swedish and had a relatively high proficiency in Swedish. Hence, although his predefined language role was monolingual Finnish, he had bilingual skills that he could make use of. Fieldwork observations in the preschool classroom made it clear that he used only Finnish when interacting with the children in small group activities that he led.

4.3 Categories of Analysis

All the recorded and transcribed material was categorised and analysed for activity, participant roles and language allocation. After this first step, four interactional exchanges were selected for more detailed analysis of the interaction, i.e. turn taking and language use. They included two examples of occasions when the teachers kept their predefined language roles and two examples of when they deviated from them.

Activities

‘Activity’ has been defined by Linell (2009, p. 190) as “an overall structured action sequence that somehow form[s] a global whole. An activity is protracted in time and is carried out in a social situation, during a whole social encounter or a phase thereof.” We have used this as a way of demarcating the activities in our data: opening sequence(s), a main message and closing sequence(s) (Linell 2009, pp. 203–204). The activities have been categorised according to their content, following Zaghalawan and Ostrosky (2011), who use activity types such as academic (topics like numbers, letters, etc.), calendar (date, etc.) or reading books.

Participant Roles

With regard to participant roles, we see the circle time as a social routine “that is accomplished by all of the parties involved (teachers and children alike) in and through temporally unfolding interaction” (Park 2014). From the data, it is clear that the circle time routine is by its nature a three-party interaction involving the two teachers and the children. However, as our focus is on team teaching, we examine primarily the roles of and collaboration between the two teachers. We take into account that the type of engagement and the expectations of the participants differ from one activity to another and that the roles of the teachers in terms of leadership may change even during the course of one activity (Gardner 2006; Park 2014). We operationalise the leading teacher of an activity as the one who initiates the activity, i.e., has a turn that frames the activity (e.g., “Now we are going to play a game”) or who is given the leading role by the other teacher (e.g., “Now children, listen to Teacher”). We further specify the role of the non-leading teacher as being either a co-leader or a participant (active or silent). Using Park (2014) as the point of departure, we distinguish three types of leadership:

  • Single leadership: The leading teacher instructs and disciplines the children and the other teacher takes a child-like participant role, either as a silent or an active (child-like) participant.

  • Co-leadership: The teachers jointly run the activity but the leading teacher is responsible for the instructional content and leads the nomination of speakers. The other teacher accompanies the leader with disciplinary and/or instructional turns, having either the same types of turns (e.g. both teachers give instructions) or different types (e.g. the leading teacher instructs and the other teacher disciplines).

  • Alternated leadership: The roles and tasks change during the course of the activity, so that the leading teacher becomes the non-leading teacher and vice versa.

Language Allocation

We calculated the percentages for language use – of Swedish and Finnish – on the basis of the number of words used in each language. This was done for each activity and for each teacher. The quantitative results are presented in Sect. 5.1. The outcomes are analysed in relation to the predefined language roles of Heikki (monolingual Finnish use) and of Johanna (bilingual use of Swedish as well as Finnish).

Interaction

In order to analyse the dynamics of teacher collaboration and the distribution of language use between the teachers on a more detailed level than a quantitative overview allows, short extracts from the interactional exchanges were selected from four of the activities, two in which the two teachers held on to their predefined roles and two in which they deviated from them. These examples were analysed from two points of view: how the turn-taking between the teachers happened (cf. Park 2014), and how the teachers played out their language roles and how they used their resources. The dynamics of teacher turn-taking were analysed by identifying the transition-relevance places, recognised by the participants as a place where speaker change was possible (Sacks et al. 1974), and examining who gave and who took the turn and for what purposes. The teachers’ language strategies were analysed on the basis of categories from Gort and Pontier (2013) and Palviainen et al. (2016): flexible language practices (continuous language alternation, code-switching, translation, and scaffolding structures) or language separation strategies (separation by person, place, time or subject; cf. García 2009).

5 Results

In Sect. 5.1 we analyse the activities included in the two circle times to ascertain the collaborating teachers’ participant roles and language allocation during the activities. In Sect. 5.2 we analyse in more detail the patterns of turn-taking and language strategies in the four team-taught activities.

5.1 Overview of the Circle Time Activities

The educator with the main pedagogical responsibility for both of the circle times examined here was the Finnish-speaking kindergarten teacher, Heikki. His leading role was signalled by the fact that he was sitting on a separate “leader’s chair”. The children were seated on three benches which together formed a circle. The bilingual kindergarten teacher, Johanna, mostly sat on one of the three benches, with the children.

As far as structure is concerned, the two circle times consisted of similar activities: academic play, reading a book, discussion, information, and the roll call ritual. Still, the circle times were not identical: Circle time 1 comprised several short activities and one longer book reading activity (Activity 1.6: 9 min and 26 s), whereas Circle time 2 was dominated by two longer activities, one with play (Activity 2.3: 15 min and 24 s) and another with book reading (Activity 2.5: 8 min). Moreover, Circle time 1 started with a good morning ritual while Circle time 2 started with information about the programme for the day. However, the two circle times ended with the same type of activity, where the teachers told the children about the transition to the next routine.

We identified 22 activities within the two circle times. A closer look at how leadership and participant roles were distributed among them reveals certain patterns. Heikki was the principal leader of both circle times and was the leading teacher in 15 of the 22 activities. Johanna was the leading teacher in four activities, and the leadership shifted from Heikki to Johanna in two activities. The most common type of collaboration was co-leadership (16 out of 22 activities), where Heikki typically acted as the leader and Johanna as the co-leader, to be followed by single leadership by either of the teachers (four cases) and alternated leadership (two cases). The most frequent participant role by the non-leading teacher was to act as a co-leader, indicating a norm that both teachers should have an active role. One exception seems to be the roll call rituals, which were led by only one of the teachers (Johanna in 1.3 and Heikki in 2.2). For either of the teachers to act only as a participant (active or silent) was very rare. One example of where Johanna had an active participant role was in a song activity (1.5), when she disciplined the children while Heikki was playing the guitar and leading the song. The dynamics of the changes in leadership roles in this preschool classroom show how predefined roles may be put flexibly into practice (cf. Park 2014).

As for language allocation in general, we can see that both languages were clearly present in the majority of the activities: in 19 out of 22 activities, both languages were used. The teachers had predefined language roles, but the overview shows that these roles were used flexibly, as in the study carried out by Pontier and Gort (2016). In our data, the predefined roles were kept in half of the activities (11 out of 22): Heikki then used Finnish only and Johanna used Finnish as well as Swedish (see e.g. 1.4 and 1.13). In Johanna’s case, bilingual use often meant that Swedish dominated (see e.g. Activity 1.3, with 84% Swedish and 16% Finnish). When the teachers deviated from their predefined language roles, we found two major patterns: Johanna spoke only Swedish (e.g., Activities 1.7–1.9) or Heikki used Swedish in addition to Finnish (e.g., Activities 1.1 and 2.7). In the following (Sect. 5.2) we will provide and discuss examples of collaboration between the teachers both when they held on to their predefined language roles and when they deviated from them.

To conclude the quantitative analysis, we can see that although both teachers used Finnish as well as Swedish during the two circle times, the relative amounts differed. In both the circle times, Heikki used predominantly Finnish while Johanna typically communicated bilingually: two thirds of the words she uttered were in Swedish and one third in Finnish. Previous analyses of Johanna’s language use when leading small group activities on her own showed that she then followed the principle of using equal amounts of both languages when communicating with the children (Palviainen et al. 2016). In this team-taught circle routine, she seemed to aim to counterbalance the dominance of Finnish (as the leading teacher, Heikki, was communicating predominantly in Finnish) by increasing her own use of Swedish. Heikki also helped to make this possible, either by giving her space (see Excerpt 1 below in Sect. 5.2), or by responding to her initiatives in Swedish (Excerpts 3 and 4 below). Overall, however, taking both teachers into account, the dominant language used with the children in most of these activities was Finnish. This was due to the fact that Heikki, who primarily used Finnish, was the leading teacher in most of the activities, whereas Swedish was the dominant language in only two of the activities, which Johanna was leading on her own (1.1 and 1.3).

5.2 Qualitative Analysis of Turn-Taking and Language Strategies

The quantitative overview of language use in the circle time activities revealed the pattern of the teachers holding on to their predefined roles in half of the activities: Heikki stuck to his monolingual Finnish-speaking role and Johanna to her bilingual role. In one sense, these cases can be seen as resulting from a teacher-determined language separation strategy (c.f. Sect. 2.2 above), in that a certain type of language use is connected with a certain teacher. However, in contrast to many studies that have examined language separation strategies where one person speaks one language (e.g. Gort and Pontier 2013), in our study Johanna is one person who speaks two languages, while Heikki’s role is to be one person who speaks one language. However, as the analysis of language allocation in the activities above has shown, one or both teachers in our study deviated from this pattern in the remaining half of the activities. This is evidence of the teachers’ flexible application of their predefined language roles, something that has also been found in some previous studies in bilingual classrooms (Gort and Pontier 2013; Pontier and Gort 2016; Swartz and Asli 2014). In the circle times as a whole, we can even argue that both teachers – Johanna as well as Heikki – were applying a one person speaks two languages practice and thus served as models of adults who can communicate in both official languages (Pontier and Gort 2016). This was a principle emphasised by Johanna when she herself was leading small group activities (Palviainen et al. 2016).

The first example to be discussed is one in which Heikki maintains his predefined role of speaking only Finnish while Johanna uses Finnish as well as Swedish. Excerpt 1 is from an activity where the children and the teachers are playing a memory game (Activity 2.3). Utterances in Finnish are given in bold text and in Swedish in normal text, with English translations written in italics (the full transcription key can be found in the Appendix).

Excerpt 1

1

Children:

[((laughing))]

2

Johanna:

[((gazes at Heikki and firmly raises her hand))]

3

Heikki:

okei] (.) [((nods to Johanna))]

[okay] (.) [((nods to Johanna))]

4

Children:

[((cont. laughing))]

5

Johanna:

[((makes a silencing gesture towards the children with her hand))]

6

Heikki:

[sen (.) hei sit kuunnelkaas] nyt on Johanna raukka katotaan riittääkö Johannan muisti ((points to Johanna))

[and its’ (.) hey let’s listen ] now it’s poor Johanna let’s see if Johanna has enough memory ((points to Johanna))

7

Johanna:

Johanna lähti retkelle [och jag tog me mej en (.) e de en bäb is?] (.) va de en bäbi-?

Johanna went on a trip [and I took along a (.) is it a ba by?] (.) was it a bab-?

In this scene, all the children are sitting in the circle, Heikki is sitting on the leading chair, and Johanna is sitting on one of the benches with the children, with one of them on her lap. The game is led by Heikki, who has given the instructions in Finnish, and the game, involving the children, has been carried out entirely in Finnish.Footnote 1 Up to the point which is illustrated in Excerpt 1 , Johanna has been silent, but she has been listening carefully. At this moment, in the middle of the game activity, the children have burst out laughing, as one of the children has said something funny (line [1]). Johanna recognises this as a transition-relevance place where she has a chance to get the speaker turn, and she raises her hand very determinedly, at the same time gazing at Heikki, the leading teacher (line [2]). She thus makes use of the institutionally pre-allocated turn-taking system and the way of asking for her turn, as pupils in a classroom would do (Sahlström 1999). Heikki sees Johanna and nods towards her to acknowledge her request for a turn (line [3]). He then explicitly gives Johanna the turn by instructing the children (in Finnish) to listen carefully to what Johanna will say (line [6]).

When Johanna gets the turn from Heikki, she plays the game according to the rituals and says in Finnish (in line [7]) the framing clause, “Johanna went on a Christmas trip”, thus signalling to the leading teacher and the children that her turn is intended as a contribution from a participant in the game. After the framing clause in Finnish, Johanna code-switches (line [7]) and repeats in Swedish each item that the children have previously mentioned in Finnish. In this way, she gives the children access to the Swedish word equivalents. The speech turn exchange which occurs in lines [2–6] thus means that Heikki makes it possible for Johanna to carry out her language teaching task. Excerpt 1 is an example of the type of separation strategy (or a specific type of tandem talk, cf. Gort and Pontier 2013) that it has been determined beforehand will be used in this preschool classroom: Heikki will communicate in Finnish and Johanna will communicate bilingually in Finnish and Swedish. In her bilingual language use, Johanna applies intra-sentence code-switching in line [7], producing the framing clause in Finnish and then continuing in Swedish in order to give the children Swedish equivalents for the items previously mentioned in Finnish. She helps the children to make connections between the item labels in the two languages by various scaffolds, clearly pointing out whose item she is addressing, e.g., with questions like the one in line [7]. However, she sticks to her principle of not providing direct translations (cf. Palviainen et al. 2016).

In the following excerpt, from the co-led activity 1.14, the two teachers are instructing the children about the activity that is going to take place immediately after the circle time routine (baking gingerbread biscuits).

Excerpt 2

1

Heikki:

Ja nyt me tehään sellanen et jos toi Johanna ottaa ne leipojat ensi matkaas ja tästä sinne ehkä kolme tai kaks mahtuu kerrallaa mukavasti.

and now we’ll do it like this that if Johanna takes those bakers along first and for this maybe there’s room for three or two nicely at a time

2

Johanna:

hej ja tror ja tar faktiskt ((eye contact with Heikki)) tre ja hej väntas jag tar ett två tre fyra ((counts with her fingers)) å det är (.) ((points at Miia)) Miia vill du komma å baka pepparkaka? jo: ((stands up and looks at Isabella)) och Isabella får också komma. ((makes a gesture to Miia to join her)) Miia kom.

hey I think I’ll take actually ((eye contact with Heikki)) three yes and hey wait I’ll take one two three four ((counts on her fingers)) and that is (.)((points at Miia)) Miia do you want to come along to bake gingerbreads? yeah. ((stands up and looks at Isabella)) and Isabella too may come along. ((makes a gesture to Miia to join her)) Miia come.

3

Miia:

((stands up))

4

Johanna:

å Isabella de e två. ((holds up two fingers))

and Isabella that’s two. ((holds up two fingers))

5

Johanna:

kukas nyt ei oo ollu sitten vielä?

who else hasn’t been along yet?

6

Heikki:

eh: ei oo ollu

eh:m hasn’t been

7

Johanna:

[((looks around)) ].

8

Heikki:

[toi toi toi toi Timo ei oo ollu].

e:r e:r e:r Timo hasn’t been.

In line [1], Heikki, the leading teacher, tells the children in Finnish that a certain number of children will join Johanna in the baking activity. Johanna then recognises the transition-relevance place, takes her turn, and continues in Swedish (lines [2–4]). Her task is to select the children for the activity, and when she turns to them, she accompanies her use of Swedish with non-verbal gestures (counting on her fingers, pointing to the children, gazing) and stresses certain words (numbers when counting), hence making use of scaffolding structures (e.g. García 2009 p. 329–336) and sheltered instruction (Gort and Pontier 2013). In line [5] she switches to Finnish and Heikki continues in Finnish [6 and 8].

In this teacher interchange – as in Excerpt 1 above – a predetermined separation strategy is applied: Heikki uses Finnish and Johanna uses both Finnish and Swedish. When Johanna takes the turn in line [2], she continues smoothly in Swedish with the topic that Heikki introduced in Finnish in line [1] (the number of children to be selected for the baking activity). An equally smooth shift of language occurs in line [5], when Johanna flexibly code-switches into Finnish to think aloud which of the children have not yet done any baking. Heikki then joins Johanna to use Finnish (in line [6]). In this extract, Johanna applies a policy of flexibly and continually switching between the two languages, as she did when leading small group activities on her own (reported in Palviainen et al. 2016). Translation as a strategy is not employed in the interchange in Excerpt 2, since both the teachers understand both of the languages used in the classroom.

There were some instances when either or both of the teachers took another language role than their predefined one. In Excerpt 3 (from Activity 1.1), Johanna sticks to her predefined bilingual language role while Heikki abandons his predefined role and speaks only Swedish.

Excerpt 3

1

Johanna:

å gomorron Isabellas= ((points to Isabella’s nose))

and good morning Isabella’s= ((points to Isabella’s nose))

2

Children:

=NÄ SA

=NO SE

3

Johanna:

nyt mäpäs otan jotain ihan uutta vänta gomorron Lauras ((stands behind Laura and points to Laura’s eye)) (.) hej gomorron Lauras=

now I’ll take something completely new wait good morning Laura’s ((stands behind Laura and points to Laura’s eye)) hi good morning Laura’s=

4

Children:

=ÖGA

=EYE

5

Johanna:

ö:ga. ((approaches Heikki)) o gomorron Heikkis= ((taps Heikki’s back))

e:ye. ((approaches Heikki)) and good morning Heikki’s= ((taps Heikki’s back))

6

Children:

=selkä.

=back.

7

Heikki:

rr[ry:gg].

ba:ck.

8

Johanna:

[rygg] ((continues to tap Heikki’s back)) m: go morr on Heikki.

[back] ((continues to tap Heikki’s back)) yeah good morn ing Heikki.

9

Heikki:

((nods)) öh tack. gomorron.

((nods)) ehm thanks. good morning.

The excerpt is from a good morning ritual, led by Johanna. It is a ritual well known to the children as they have practised it many times before. As illustrated in line [1], the routine is that Johanna says, in Swedish, “Good morning, Isabella’s” and points, for example, to Isabella’s nose, and the children know that they are expected to respond in chorus with the correct body part in Swedish (lines [2] and [4]). The same ritual formula is repeated for each child.

Up to the turn change in line [5], Heikki has been quiet and has not participated in the children’s chorus responses. When all the children sitting in the circle have been addressed and had their turn in this ritual, it comes round to Heikki, who is the last one in order. Johanna addresses him in the same way as she has addressed the children, saying in Swedish the ritual formula “Good morning, Heikki’s” while tapping him on the back (line [5]). Both Johanna and Heikki then wait for the children to respond “back” in Swedish, but apparently this is a new word for them and they respond in Finnish (selkä, line [6]). Then Heikki recognises and utilises the transition-relevant place to provide the Swedish word for “back” (rygg, line [7]), partly overlapping with Johanna’s uttering the word (line [8]). In the two last turns, which also conclude this activity, Johanna changes the ritual formula somewhat by saying, more conventionally, “Good morning, Heikki” in Swedish. This polite greeting leads to the expectation of a polite response. Heikki responds appropriately, with “Thanks” and “Good morning” in Swedish.

A similar type of separation strategy to the one used in the previous excerpts (1 and 2, above) is used in Excerpt 3 : Heikki acts as a monolingual speaker and Johanna as a bilingual speaker. However, in Excerpt 3 Heikki, despite his predefined role as a Finnish speaker, uses only Swedish in the collaboration with Johanna. Johanna, in turn, makes flexible use of both languages, behaving in accordance with her predefined language role. After using only Swedish in line [1], Johanna code-switches to Finnish in line [3] to make the children attentive to what she will say next (“Now I’ll take something completely new”). This is an example of her strategy of using Finnish and Swedish for different purposes, which was also found when she was leading small group activities on her own (Palviainen et al. 2016).

By using Swedish in this situation Heikki assists Johanna in the role of language teacher (to provide a new term in Swedish) and at the same time his use of Swedish means that he acts as a model of an adult who can communicate in both official languages (Palviainen et al. 2016; Pontier and Gort 2016). Moreover, as Carless (2006) found in the context of a Japanese team-taught classroom where both teachers used English, the “presence of two teachers was useful in allowing them to model dialogues, demonstrate question and answer routines naturally” (p. 246). Johanna and Heikki provided the children with a model of how to greet people in Swedish (lines [8–9]) and also sent the message that Swedish “is a tool for communication” (Carless 2006, p. 246).

The final sequence to be analysed is from a discussion activity (2.7) which, on the whole, was characterised by bilingual language use by both teachers, but where a micro-level analysis shows how both teachers smoothly transfer from the monolingual use of Finnish to the almost exclusive use of Swedish.

Excerpt 4

1

Heikki:

kuunte[leppa mitä tehään sitten]

lis[ten to what we are going to do next]

2

Johanna:

[((eye contact with Heikki, raises her right hand and rotates her fingers)) ]

3

Children:

((noise))

4

Heikki:

((hushes)) Johannalla oli käsi pystyssä ((points to Johanna))

((hushes)) Johanna has raised her hand ((points to Johanna))

5

Johanna

((looks at children)) hei mua jäi vähän kiinnos[taan]=

((looks at children)) hey I was just wonder[ing]=

6

Heikki:

[jo:]

[yea:h]

7

Johanna:

=kun mä kuuntelin tota tarinaa nii ni ymmärsinks mä oikein et se Viiru hyppi niinkun pöydällä?

=when I heard that story yeah if I understood it right that Findus jumped like on the table?

8

Children:

jo:.

yea:h.

9

Johanna:

oliks se niin?

was it so?

10

Child:

ja:

ye:s

11

Johanna:

((looks at Heikki))

12

Heikki:

no ei se tä ti ainakaan mun mielest [ollu]

well the aunt ie didn’t at least I [think]

13

Johanna:

[men Heikki] kan du visa ((spreads her hands)) hur Viiru kan hoppa på bordet? vi vill se.

[but Heikki] can you show ((spreads her hands)) how Findus could jump on the table? we want to see it

14

Child:

Nej

No

15

Heikki:

((laughs)) måste jag?

((laughs)) do I have to?

16

Johanna:

jo du måste (.) ((smiles)) hur ((spreads her hands)) hoppar Viiru på bordet?

yeah you have to (.) ((smiles)) how ((spreads her hands)) does Findus jump on the table?

17

Heikki:

((looks around)) xxx finns inte bord

((looks around)) xxx there is no table.

18

Child:

(xxx)

19

Johanna:

vänta. Heikki visar.

wait. Heikki will show us.

20

Heikki:

((stands up, smiles, looks for a table))

21

 

((noise))

22

Johanna:

((points at a rug)) de här är bordet (.) mattan är bordet.

((points at a rug)) this is the table (.) the rug is the table.

23

Heikki:

ai mat tan e bordet

oh the rug is the table

24

Johanna:

så där hoppar Viiru (.) titta. ((points to Heikki))

that’s how Findus jumps (.) look. ((points to Heikki))

25

Heikki:

((laughs))

This exchange was preceded by a typical story-reading activity, in which Heikki sat in his chair reading aloud from a picture book in Finnish while the others listened (Activity 2.5). The story was about some familiar characters: an obstinate little cat with the name of Findus (Viiru in Finnish), who lived in a cottage with his master, Pettson (Pesonen). In the beginning of Excerpt 4 , Heikki intended to initiate a new activity and asked the children to pay attention so as to be ready to receive some instructions (line [1]). What happens at this point is that Johanna recognises the transition-relevance place, gazes at Heikki as she raises her hand and eagerly rotates her finger to show that she would like to get the turn (line [2]). Heikki acknowledges this and gives her the right to speak (line [4]).

When Johanna has got the turn, she looks around at the children and reconnects to the story they have just heard by referring to an odd event, when the cat Findus jumped on a table (lines [5–7]). She asks the children for confirmation of whether that was actually the case and they confirm that it was (lines [8–10]). This is done in Finnish, as was the case when she wanted to engage the children in discussions when she was leading the children bilingually on her own (Palviainen et al. 2016). She then turns to Heikki, who also confirms it in Finnish (lines [11–12]). Then an imperceptible alternation in the leadership comes about when Johanna starts to take the lead and at the same time the language they both use changes to Swedish: she addresses Heikki in Swedish (line [13]) and he responds in Swedish (line [15]).

Johanna has a somewhat odd request that seems to take Heikki by surprise: Johanna asks him to act out what it looked like when the cat in the story, Findus, jumped on the table (line [13]). He replies with a laugh and asks, in Swedish, “Do I have to?”. She repeats her request and Heikki starts to play along, stands up and looks around to find a suitable table (lines [17–20]). Johanna eventually comes up with the idea that the rug may serve as a table and Heikki accepts that (line [22–23]). In this interaction, a separation strategy based on content was used (cf. García 2009): the initial teacher turns [lines 1–12] were most probably performed entirely in Finnish in order to engage the children in the topic. When moving over to a more concrete content in line [13], Johanna as well as Heikki switched to Swedish. The nature of the activity made possible the use of concrete Swedish verbs that the children probably already knew (show, jump, look).

The latter part of the exchange (lines [13–25]) is carried out almost entirely in Swedish. There are two exceptions to this: Johanna uses the Finnish name Viiru instead of the Swedish Findus [lines 13, 16 and 24], and Heikki utters a Finnish interjection (oh) [line 23]. The first of these code-switches into Finnish can probably be explained as a decision by Johanna to scaffold understanding: the name Viiru is familiar to the children from hearing the story, whereas the name Findus is not. The use of the Finnish interjection oh by Heikki is likely to be simply the spontaneous expression of surprise that it seems to be. The sequence shows that teacher collaboration and turn taking can create space for spontaneity, improvisation and playfulness (cf. Park 2014) even in a bilingually team-taught classroom, when the teachers understand both languages. It also clearly shows the flexible attitude the teachers can have to their predefined language roles.

6 Concluding Discussion

This study aimed to examine how bilingual team teaching is implemented in a preschool classroom in Finland, in a typical preschool routine, circle time. The focus was on how the team-teaching dynamics played out in terms of participant roles, language use and allocation, and teacher collaboration. One of the teachers, Heikki, assigned to the class as a Finnish speaker, was the predefined leader of both circle times and co-taught them with Johanna, who had been assigned responsibility for bilingual pedagogy in the class. Previous studies reported by Carless (2006) have indicated that the fact that teachers have different, predetermined language roles might cause confusion and anxiety as the teachers are unsure “about their respective roles in the classroom” (p. 344). We found no evidence of this in our data. On the contrary, our data showed that both teachers flexibly communicated monolingually as well as bilingually in the various circle time activities, changing language roles in a similar way to that shown by Park (2014), Pontier and Gort (2016), and Schwartz and Asli (2014).

The analysis of turn-taking patterns showed that the appointed bilingual teacher, Johanna, made use of the possibilities opened up by the transition-relevance places and actively asked the teacher in charge (the Finnish-speaking teacher Heikki) to give her the turn, and he, for his part, gave the floor to the bilingual teacher to create opportunities for the children to receive Swedish input whenever she asked for it. On some occasions (see Excerpts 3 and 4 above), Heikki even responded in Swedish to Johanna’s initiatives and, in doing so, assisted her in the teaching of Swedish to the Finnish-speaking children and served as a model of a person using two languages. This showed that the Finnish-speaking teacher had a positive attitude to bilingual pedagogy in the classroom. It also showed that the two teachers trusted each other and gave each other space to play their own roles (cf. Liu 2008).

The extensive dynamicity in our data was facilitated by the fact that the teachers had similar cultural and educational backgrounds and were both proficient in both the languages that were used in the classroom. Hence, they made use of “their collective bilingual repertoire” as a resource (Pontier and Gort 2016, p. 96). They were able to move fluently through the activities even if the language of communication changed in the course of an activity. Heikki, the teacher with the predefined monolingual Finnish role, used Swedish to support and assist the bilingual teacher in her role of language teacher, for example by serving as a bilingual model for the children (cf. Carless 2006). Successful bilingual team teaching is, however, even possible in contexts where the teachers have quite different language backgrounds, as shown by e.g. Gort and Pontier (2013) and Pontier and Gort (2016). In these contexts, it is important to agree upon the principles for language use and participant roles in order to ensure that the activities happen fluently and without confusion (Carless 2006).

Despite Heikki’s active involvement in the bilingual activities, the overall responsibility for bilingual pedagogy in the class under study lay clearly with the bilingual teacher, Johanna. We identified similar practices in her way of working bilingually during these team-taught circle times to those found when she was leading small group activities on her own (c.f. Palviainen et al. 2016): she alternated between the languages, made use of responsible code-switching and avoided direct translation, and made use of scaffolding structures to enhance understanding. There was, however, one major difference as regards the use of the two languages: while she used Swedish and Finnish more or less equally in the small group activities, she used relatively more Swedish than Finnish during the circle times. This was probably in order to compensate for the dominance of Finnish in the circle times, as the leading teacher, Heikki, used mostly Finnish. Johanna thus seemed to adapt her own language use to the communication situation.

In focusing on a bilingual preschool class in which the teachers used Finnish and Swedish, the study has provided insights into a field which has hitherto been dominated by research on English-medium primary or secondary school classes. The aim of this study has not been to examine the impact of bilingual team teaching on children’s bilingual language learning, but rather to give examples of how smooth bilingual teacher collaboration can be carried out. The bilingual team-teaching practices thus serve as examples of “good classroom practice” (cf. Carless 2006) for the collaborative creation of bilingual learning environments to support children’s content and language learning and cognitive development, which may be of value to teachers in the field and teacher educators. One limitation of the study was that the interviews with the bilingual teacher were focused on her language practices in small group activities that she led herself, rather than on team teaching. Nor did the study include interviews with the monolingual teacher. Interviews that included teachers’ comments on their own team-teaching practices and ideologies would have added further important insights, and we therefore recommend future research initiatives on team teaching in bilingual preschool classrooms to combine observations with teacher interviews. In addition to studies of the delivery of bilingual instruction, there is a need for further research on the planning, organisation and evaluation stages (cf. e.g. Liu 2008), as well as on how team teaching affects bilingual language development in children.

Finally, we believe that the analytical model that we have developed, combining as it does macro- and microanalyses of participant roles and languages practices, will make it possible for other researchers to carry out comparative studies of team teaching in other bilingual preschool settings around the world. This can lead to our learning from each other and developing good bilingual team-teaching practices, and also to learning how specific languages with different linguistic characteristics and statuses can be combined and team taught in preschool classrooms.