Translanguaging in the context of mainstream multilingual education

ABSTRACT

In the context of multilingual education, translanguaging has been put forward as a means of including several languages in education. However, teachers often assess translanguaging-based approaches as being too vague and idealist. This study discusses data from two settings (Luxembourg and Netherlands) in which teachers working in design-based projects operationalised the concept of translanguaging in order to include both migrant and minority languages in mainstream education. Examples from each dataset will be discussed in order to show the different functions of translanguaging in the two settings. Analyses of classroom transcripts provide insights into how official translanguaging can be used as pedagogical strategy to acknowledge migrant languages, achieve less language separation in traditional immersion models and to increase content understanding. Based on teachers’ own reflection on their use of translanguaging and on iterative interpretation of excerpts of the data, the study provides an overview of the functional use of different languages within moments of official translanguaging.

KEYWORDS:

• Translanguaging
• multilingual education
• minority languages
• immigrant languages

1. Introduction

Across European schools, the number of multilingual pupils is currently rising; migrant pupils and newcomers now co-exist much more with minority languages (Aronin & Hufeisen, 2009; Vertovec, 2007). It is thus an imperative that schools cater for equal school success of their multilingual pupils. While the European discourse on multilingualism is a highly favourable one, reality shows that those who are socialised in more than one language are often underperforming in European schools (OECD, 2016).

This increase of multilingual pupils has led to the investigation of models of multilingual education (MLE; Cenoz, 2009; Hobbs, 2012) as means to improve school outcomes of multilingual pupils. According to Cenoz and Gorter, ‘multilingual education refers to the use of two or more languages in education, provided that schools aim at multilingualism and multiliteracy’ (2015, p. 2). It is thus an umbrella-term for various school approaches including several languages of instruction, also for those aiming at fostering elite bilingualism. Yet, in many recent programmes framed within a MLE perspective, one common feature is the active inclusion of pupils’ family languages as a resource in instruction. In terms of including pupils’ languages in instruction, the term translanguaging has been put forward not only to describe multilingual practices that include ‘the full range of linguistic performances of multilingual language users’ (Wei, 2011, p. 1224), but also to propose a pedagogical approach in which such practices are systematically used in education (Cenoz & Gorter, 2015; Duarte, 2016; García, Johnson, & Seltzer, 2017). Still, and although an increasing body of literature points towards a growth in implementation, translanguaging approaches are still not widely spread in European mainstream education and face several implementation challenges (Ticheloven, 2016).

The main aims of the present paper are (a) to explore translanguaging practices across two mainstream education settings, (b) identify their pedagogical functions, and (c) to reflect on the ways in which teachers implement the concept of translanguaging for their specific purposes.

2. From immersion to MLE

Recent research on strong bi- and trilingual school models has offered evidence for the potentials of using multilingualism for raising academic achievement (Beetsma, 2002; Duarte, 2011; Duarte & Pereira, 2011; Francis & Lesaux, 2006; Rolstad & Mahoney, 2005). In addition, mainstream schools which used multilingualism as a resource for learning have yielded positive academic results for all pupils (Bourne, 2013; Bührig & Duarte, 2014; Creese & Blackledge, 2010; Dirim, 1998; Gogolin & Neumann, 1997; Moodley, 2007; Rolff, 2006). However, the common approach towards teaching multilingual pupils is immersion in the official languages of instruction of national curricula, which is closely related to the so-called monolingual bias in language teaching and is operationalised in the strict separation of the languages of instruction (see the notions of ‘two solitudes’, Cummins, 2008; and ‘separate bilingualism’, Creese & Blackledge, 2010).

Based on the idea that pupils and teachers bring diverse linguistic knowledge that can actively be used as a resource for learning, a recent trend regarding language teaching and learning has been termed the multilingual turn in language education (Conteh & Meier, 2014). Similarly, Flores and Baetens-Beardsmore (2015) describe the benefits of heteroglossic approaches in which minority and immigrant languages are incorporated in instruction. Cenoz and Gorter (2011, 2015) refer to the ‘Focus on Multilingualism’ approach in which the natural multilingual practices of pupils are closely related to the ways in which languages are taught. Currently, however, the theoretical development of teaching approaches that use pupils’ multilingual competences is more advanced than empirical research on the implementation and effectiveness of such models (Herzog-Punzenberger, Le Pichon-Vorstman, & Siarova, 2017).

3. Translanguaging in the context of MLE

García’s work on ‘translanguaging’ (2009) is probably the most well-known approach pointing towards the relevance of using family languages in instruction (García & Wei, 2014). The concept refers to the use of the learner’s full language repertoire in teaching and learning (García et al., 2017). García and Kano (2014, p. 261) refer to translanguaging in education as

a process by which students and teachers engage in complex discursive practices that include ALL the language practices of ALL students in a class in order to develop new language practices and sustain old ones, communicate and appropriate knowledge, and give voice to new sociopolitical realities by interrogating linguistic inequality.

In its original formulation as coined in Welsh by Williams (2002), it referred to the deliberate practice of alternating the language of input and the language of output, the basic idea being that one language reinforces the other in order to raise understanding as well as pupil’s activity in both languages (Lewis, Jones, & Baker, 2012). According to Allard,

[t]ranslanguaging includes flexible language practices such as code-switching, co-languaging, and others, though the term extends the understanding of these practices as “dynamic and functionally integrated” in ways not previously captured by a focus on the alternation between two separate codes (2017, p. 117).

Empirical research has focused on analysing classroom interaction by zooming in on the ways translanguaging is used for constructing meaning, acquiring knowledge and negotiating power in diverse classrooms. An array of studies has underlined the advantages of a translanguaging pedagogy at different levels of school performance and for both migrant and minority languages: for example, as a means of balancing the power-relations among languages in the classroom (2017, p. 117), in protecting and promoting minority languages (Cenoz, 2017), for raising participant confidence and motivation (Creese & Blackledge, 2010), as a maximiser of learning literacy skills (Hornberger & Link, 2012), for empowerment and language learning (Latisha & Young, 2017) and for higher cognitive engagement in content-matter learning (Duarte, 2016).

Criticism to translanguaging-based approaches stresses its lack of empirical verification in terms of tangible effects on educational outcomes. In addition, teachers often complain that its goal is too philosophical and lacks a clear definition with regard to pedagogical tools (Ticheloven, 2016). In sum, although enjoying positive echoing in research, the implementation of translanguaging approaches in mainstream education does not yet belong to the pedagogical status quo in most European schools. On the one side, a translanguaging pedagogy clashes against prevailing monolingual ideologies often translated into immersion models for language teaching which lead to strict language separation. On the other side, ideas of teachers in relation to the value and functions of pupils’ additional languages lead translanguaging practices to be perceived as ‘illegitimate’ in mainstream education (Kamwangamalu, 2010). As a consequence, minority and migrant languages are often left out of mainstream education. In addition, many projects looking at translanguaging are not conducted within mainstream education but work with pull-out designs. Translanguaging research has studied the use of either migrant or minority languages but not focused on both types of multilingual speakers comparatively. The present paper presents two projects including a translanguaging-based approach conducted (a) within mainstream education without pull-out designs, and (b) including both migrant and minority languages alongside national and foreign languages. Through this inclusive approach, the paper aims to answer the following research questions:

1. RQ1: What are the main functions achieved through a translanguaging-based pedagogy in the two settings?

2. RQ2: What supports teachers in giving form to the concept of translanguaging to serve their specific purposes?

4. Methodology

The paper is based on two empirical studies conducted in two contexts and for different educational levels (see Table 1).

Table 1. Overview of empirical studies.

	Educational level/age	National context	Languages involved	Sample	Methodology
Study 1	Pre-school (5-year olds)	Luxembourg	Luxembourgish and Portuguese	5 schools, 9 teachers, 55 pupils	Intervention with mixed-methods design
Study 2	Primary education (6– 10-year olds)	Friesland, Netherlands	Dutch, Frisian, English and migrant languages	5 schools, 10 teachers, 85 pupils	Design-based study with videography

The paper will first present each individual study in terms of its design and implementation in relation to a translanguaging-based pedagogy, then present exemplary classroom transcripts of the projects and finally discuss the main lessons that can be learned from the two settings on the functional use of translanguaging as based on empirical research.

4.1. Study 1 – translanguaging at pre-school level in Luxembourg

4.1.1. Setting

Luxembourg is a trilingual country in which Luxembourgish (a Germanic language), French and German co-exist in different societal areas. This is also the case in education. Pre-school education (until children are aged 6) is done in Luxembourgish, then children are alphabetised in German, and French is gradually included as school subject. In addition, about 17% (STATEC, 2017) of the population in Luxembourg is of Portuguese origin, which makes it the largest group of migrants. This number is even higher among primary school children.

4.1.2. Aims

Due to the large achievement gap between Luxemburgish and Portuguese-speaking pupils, the Ministry of Education decided to initiate a pilot-project in order to include Portuguese in pre-school education and describe the effects of this intervention on the language skills of children and on their ability to transfer between their languages (Duarte & Quintus, 2016). The main aim was to use the competences in the family language as a resource to learn Luxembourgish by exploring connections between languages.

4.1.3 Design

The study had an explorative pre-post-design with an experimental (n = 40 pupils) and a control group (n = 15 pupils). During the intervention, a Portuguese-speaking assistant worked closely with the mainstream teacher for 3 hours per week, and 5 hours per school were filmed and analysed using conversational analysis (Heritage & Clayman, 2010). Table 2 provides an overview of the sample.

Table 2. Sample of study 1.

	Gender and languages	Experimental group	Control group	Total
Teachers	F	8	2	10
	M	2	1	3
Pupils	F	20	8	28
	M	20	7	27
Pupils’ family language	Port.	21	7	28
	Port. + Lux.	16	5	21
	Other	3	3	6

4.1.4 Relation to translanguaging

The concept of translanguaging was perceived in two different ways in order to fit the educational context of Luxembourg. This was done in a design-based approach in which teachers co-construct their own classroom intervention via cycles of development and experimentation (Anderson & Shattuck, 2012). On the one hand, translanguaging was used as a planned-in systematic activity for three hours a week when the Portuguese teacher would teamteach with the Luxembourgish. This has been termed pedagogical translanguaging (Cenoz, 2017). On the other hand, participating teachers agreed to explore the fluid discursive practices taking place inside the classroom and encourage pupils to use peer–peer interaction in a translanguaging modus to cognitively engage with new content. This approach has been referred to as spontaneous translanguaging (Cenoz, 2017). Very often spontaneous translanguaging was niched in moments in which pedagogical translanguaging was already taking place.

4.2 Study 2 – translanguaging at primary education in the Netherlands

4.2.1 Setting

Friesland is an officially bilingual province in the North of the Netherlands with Dutch co-existing in a diglossic situation with West-Frisian (a Germanic language), the second official language of the Netherlands. As a response to regional multilingualism, trilingual schools with Dutch, English, and Frisian were set up. The three languages of instruction are kept apart from each other, by attributing different days of the week to instruction in the different languages. Thus, these trilingual schools opted for a triple immersion approach. Evaluation of the model attested higher levels of proficiency in the Frisian and English languages and comparable performance in Dutch as well as more positive language attitudes for pupils attending trilingual schools (Riemersma & De Vries, 2011). In addition, the region has welcomed a growing number of migrants in the last years, mostly asylum seekers and refugees (CBS, 2017). An average of 10% of the inhabitants of Friesland has a migrant background.

4.2.2. Aims

After two rounds of exploratory workshops to determine the needs of teachers in schools in Friesland, concrete aims for the research project were jointly formulated. Schools, researchers and teacher training programmes in the Friesland region agreed on two main aims for the project:

1. achieve less language separation between the three official instruction languages in the curriculum (Dutch, Frisian, and English);

2. implement strategies to include migrant languages in the current trilingual models (Duarte & Jellema, 2017).

4.2.3. Design

In order to achieve these aims, a design-based approach was applied (Cobb, Confrey, diSessa, Lehrer, & Schauble, 2003), in which experiments are developed and implemented in a cyclic design. As a result, the 10 participating teachers developed their own tailored research questions and didactical experiments, which they implemented, reflected upon, and improved. In each school, implementation occurred in two classrooms of pupils with ages ranging between 7 and 12 years of age. Video-observations were conducted during the implementation phases with the aim of investigating how teachers translated the concept of translanguaging (Duarte & Jellema, 2017). In each class, we observed 5 hours of implementation in each of the classrooms (a total of 37.5 hours of footage). For coding, we looked for critical incidents in the data which are ‘particular events or occurrences that might typify or illuminate very starkly a particular feature of a teacher’s behaviour or teaching style for example’ (Cohen, Manion, & Morrison, 2007, p. 404). Table 3 provides an overview of the sample.

Table 3. Sample of study 2.

	Gender and languages	Participating schools (n = 5)	Total
Teachers	F	8	10
	M	2
Pupils	F	173	330
	M	157
Pupils’ family language	Dutch	103	330
	Frisian (and Dutch)	205
	Other (Arabic, Polish, Swedish)	23

4.2.4. Relation to translanguaging

Translanguaging was used to create moments of less language separation, as the alternation of languages in different phases of one single language class. Baker describes this process in the following way: ‘To read and discuss a topic in one language, and then to write about it in another language, means that the subject matter has to be processed and “digested”’ (2011, p. 289). The principle chosen by the teachers for this was to provide new input in the language in which pupils were less proficient and then have pupils discuss content and language with peers in another language (Duarte & Jellema, 2017, p. 23).

5. Classroom scenes and results

For the present paper, all sequences in which two or more languages played a role in classroom communication were coded as translanguaging and detached from the larger dataset. After extensive viewing of these video excerpts and analysis of the transcripts, scenes for the present article were selected based on two criteria:

1. fragments were found by teachers and researchers to be representative of the dataset they were taken from, in terms of the pedagogical use of the several languages and the interactive practices observed;

2. fragments show different pedagogical approaches in relation to the use of translanguaging.

For each study, a transcript will first be provided, followed by a paraphrase of events and an analysis of language use of translanguaging in each sequence. In the end, a discussion of translanguaging in both studies will be carried out in order to sort out the pedagogical functions implemented in the two settings presented.

5.1. Study 1 – example of translanguaging in pre-school in Luxembourg

The fragment was selected from the dataset of videographic observations of this pre-school class with about 60% of Portuguese-speaking pupils (see Table 4). In average, pupils were 5.3 years old. For this class, the teacher displayed different plates containing ingredients used for baking. Some of the ingredients have a similar appearance in terms of colour and form (e.g. sugar and flour). She then starts a question and answer phase in order to elicit different verbs associated with eating, tasting and cooking in the Luxembourgish language.

Table 4. Translanguaging sequence from a pre-school class in Luxembourg.

Teacher 1 (Luxembourgish):	Kucke emol. (Look). Kuck emol, ob s du d’Miel fënns. (Look here if you can discover the flour.) Also op Portugisesch war et … ? (So how do you say flour in Portuguese?)
Teacher 2 (Portuguese):	Diz outra vez. (Say it again.)
Teacher 1 (Luxembourgish):	So mer emol nach eng Kéier, wéi seet ee schonn erëm op Portugisesch? (So one more time, how do you say this in Portuguese again?)
Several children (Portuguese):	Farinha. (Flour.)
Teacher 1 (Luxembourgish):	Voilà ! OK Mengs du et wier dat? Pointing to plate: (Right, ok. Do you think it is that?)
Pupil 1(Luxembourgish):	Jo (Yes.)
Teacher 1 to pupil 1 (Luxembourgish):	A wéi kanns du dat dann elo erausfannen? (And how can you find out?) Kuck emol, hei ass awer och wäiss (Look here, this one is also white.)
Pupil 1 (Luxembourgish):	Jo (Yes.)
Teacher 1 to pupil 1 (Luxembourgish):	Ha, ha an do ass och wäiss. (Haha, and this one is also white)
Pupil 1 (Luxembourgish):	Jo (Yes.)
Teacher 1 to pupil 1 (Luxembourgish):	Ho, ho. Mengs du et wier dat doten? (Do you think it is that one there?)
Pupil 1 (Luxembourgish):	Jo (Yes.)
Teacher 1 to all pupils (Luxembourgish):	Majo, a wat kann een dann elo maachen fir eraus ze fannen, wat fir eng (…)? (But what can we do to find out in which (…)?)
Teacher 2 (Portuguese):	O que (…) (What (..))
Pupil 2 (Luxembourgish):	(…) Ech weess et ass. Ech weess wou deen ass. (… I know it. I know where it is.)
Teacher 1 to all pupils (Luxembourgish):	Lauschtert emol eng Kéier no. Wéi – kënne mer erausfannen – wat fir eng – also op wat fir engem Teller, dass d’Miel ass? (Listen one more time now. What can we do in order to find out in which plate the flour is?)
Pupil 2	Euhm
Teacher 1 to all pupils (Luxembourgish):	Wéi ma mer dat elo? (How can we do that?)
Teacher 2 (Portuguese):	Vocês perceberam bem? (Have you understood it correctly?)
Pupil 3 (Portuguese):	Sim. (Yes.)
Teacher 2 (Portuguese):	O que é que a C. perguntou? (What has miss C. asked?)
Pupil 3 (Portuguese):	Eu acho que temos que provar. (I think we have to taste it.)
Pupil 4 (Portuguese):	Onde está. (Where it is.)
Teacher 2 (Portuguese):	A hei. Muito bem. (A hei, very well.)
Pupil 4 (Portuguese):	Temos que procurar onde está. (We have to look for where it is.)
Teacher 2 (Portuguese):	Temos que procurar e tu disseste …  (We have to look for and you said …)
Pupil 3 (Portuguese):	… provar (… taste.)
Teacher 2 (Portuguese):	Provar. Temos que provar (Taste. We have to taste it⁠.)
Pupil 2 (Luxembourgisch):	Op dee rou/op deen orangen (It’s in the orange plate.)
Teacher 1 to all pupils (Luxembourgish):	Ok, an d’Lilly hat och eppes gesot. (Ok, and Lilly also had an idea).
Lilly (Luxembourgish):	Mir musse schmaachen. (We have to taste it.)
Teacher 2 (Portuguese):	Provar. (Taste.)
Teacher 1 to all pupils (Luxembourgish):	Ah, voilà! (Ah, we got it.)
Pupil 4 (Luxembourgish):	Ech hunn dat gesot – am Portugiseschen. (I had already said that in Portuguese).
Teacher 1 to pupil 4 (Luxembourgish):	Ma tiptop, dat ass ganz gutt. (Super, that is great.)
Teacher 1 to pupil 1 (Luxembourgish):	Also probéiere mer elo fir et ze schmaachen. (So now let us try and see if we can taste it.)

5.1.1. Paraphrase of excerpt

In this sequence, the teacher starts by asking one pupil the initial question that will be the topic throughout the fragment: how to find out where the flour is in the displayed plates. Immediately after this introduction, she asks the pupils how to say flour in Portuguese. Most of the pupils answer at once and stand up in order to be able to see the plates better. She then turns to the girl standing by the plates and asks again in which of the plates she thinks the flour is, pointing towards the fact that most ingredients are white. The teacher then specifies her question: ‘What can we do in order to find out in which plate the flour is?’ This is the turning point in the sequence, as the Portuguese teacher takes over the interaction by asking the Portuguese pupils if they understood the question. Several pupils suggest answers, including looking for the flour and tasting the plates. The teacher summarises all provided information in Portuguese. The Luxembourgish teacher, who does not speak or understand Portuguese, then asks a Luxembourgish-speaking girl to tell her answer (‘we have to taste it’). The Portuguese teacher repeats the verb in Portuguese (‘provar’) and one of the boys explains to the Luxembourgish teacher that they had already said it in Portuguese. At the end of the sequence, the first girl is told that she can start tasting what is displayed in the plates.

5.1.2. Analysis of excerpt

This sequence is an example of the use of two languages in joint-construction of both content- and language-related knowledge (Duarte, 2016) through the alternation of languages by two distinct native speakers. Following an iterative analysis of the data and a discussion of the video footage with the involved teachers, it was determined that translanguaging plays a role here at two different levels. First, the Luxembourgish-speaking teacher asks for the word ‘flour’ in Portuguese, thus valorising pupils’ knowledge of their first language in official classroom communication. According to García and Wei, such use of translanguaging ‘develops the weaker language in relationship to the dominant one’ (2014, p. 224). Furthermore, it promotes the integration of those who are emergent bilinguals with those who have a fuller use of bilingualism in a classroom (Lewis, Jones, & Baker, 2012). Such instances of translanguaging in which a direct translation of a word or concept was asked for and integrated into overt classroom interaction happened somewhat often in the data. They fulfill a symbolic function, in that they are not aimed at increasing language or content learning but rather at acknowledging pupils’ proficiency in the home languages as a valuable resource.

Further on in the sequence, the Luxemburgish teacher stimulates the Portuguese-speaking teacher to make a summary of information discussed in Luxemburgish so far and to gather answers from the pupils. This Portuguese sequence is thus embedded in official interaction and illustrates a use of translanguaging, such as it was originally described by Williams (2002), referring to the alternation of two languages to reinforce each other and raise both understanding of content and language. Lewis, Jones, and Baker (2012) point out that this type of cognitive processing is relevant for retaining and developing bilingualism, rather than just for emergent bilinguals at the initial stages of their bilingual continuum. In such sequences in the data, both teachers and researchers agreed that translanguaging fulfills an epistemological function, as the main aim is to secure and enhance knowledge of both content and language by using pupils’ family languages.

5.2. Study 2 – example of translanguaging in primary education in Friesland

The fragment for this study was selected from a classroom of second graders (aged 7–8) in a trilingual school in Friesland (see Table 5). The average age of the 15 pupils recorded is 7.2 years. The majority of them are bilingual in Frisian and Dutch. In this class, there are two Polish-speaking pupils and one pupil with Arabic as family language. The teacher, however, is not fluent in these two languages. The sequence takes place during the opening of the day when the pupils are in a circle and greet each other. According to the triple immersion programme followed by the school, the language of instruction of the day is Frisian.

Table 5. Translanguaging sequence from a primary school in Friesland.

Teacher (Frisian) with 5 pupils in front of her:	En dizze bern dy sille even wat tsjin jim sizze. (And these children will now tell you something). Moatte jim mar hiel goed lústerje. (You have to listen very carefully.) J. mei begjinne. (J. can start.)
Pupil 1 (Dutch):	Goeiemorgen, welkom allemaal. (Good morning, welcome everyone.)
Pupil 2 (Frisian):	Goeiemoarn, wolkom allegearre. (Good morning, welcome everyone.)
Pupil 3 (English):	Good morning everybody.
Pupil 4 (Polish):	Dzień dobry, wszyscy. (Good morning, welcome everyone.)
Pupil 5 (Arabic):	صباح الخير sabah alkhyr (Good morning.)
Teacher (Frisian) while 5 pupils sit down:	Dankjewol. Keurich (Thank you. Well done.) Wolkom allegearre. Wat fijn dat jim der binne. (Welcome everyone. It’s great that you are all here.) It is wer in nije dei en kinne wy allegearre leuke dingen dwaan hjoed. (It is a new day and we can⁠ all do nice things today.) En wy hawwe (…) (And we have.)
Pupil 6 (Dutch):	(…) We zijn er vandaag. (We are here today.)
Teacher (Frisian):	… Ja, wy binne der hjoed wer. (Yes, we are here today). (Faces pupil sitting beside her) En ik wol dy freegje wolsto Ponpon helje? (And I want to ask you: could you get Ponpon?) (Turns to pupil sitting on her other side) En wolsto ek even mei. (Could you go with him?)
	These two pupils stand up and get the doll (Ponpon).
Teacher (Frisian):	Sa. Sille wy even wat tsjin ‘ e Ponpon sizze, ja? Dêr geane wy. (So. Shall we say something to Ponpon? There we go.)
Teacher and pupils (English):	Good morning Ponpon. How are you?
Pupil 1 (English):	Fine, thank you.
Teacher to all pupils (Frisian and Dutch):	No, dêr geane wy. (Frisian) Maandag, dinsdag, woensdag, donderdag, vrijdag, zaterdag en zondag. (Dutch) (Now, there we go. Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, Sunday) No even yn it Ingelsk. (Now in English).
Teacher and pupils (English):	Monday, Tuesday, Wednesday, Thursday, Friday, Saturday and Sunday.
	De woansdei. No, juster wie it … ik haw it yn myn hân, (The Wednesday. Well, yesterday it was … , I have it in my hands)
All pupils (Frisian):	13 desimber (The 13^th of December!)
Teacher (Frisian):	En hjoed 14 desimber! (And today it’s the 14^th of December!) In ien en in fjouwer. (It’s a one and a four.) No, dêr geane wy. (Now, there we go).
All pupils (Frisian):	Ien, twa, trije, fjouwer, fiif, seis, sân, acht, njoggen, tsien, alve, tolve, tretjin, fjirtjin. (One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen.)
Teacher (Frisian):	No yn it Nederlânsk. (Now, in Dutch).
All pupils (Dutch):	Één, twee, drie, vier, vijf, zes, zeven, acht, negen, tien, elf, twaalf, dertien, veertien. (One, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen.)
Teacher (Frisian):	En wy kinne it ek yn it Ingelsk, hin? (And we also can do it in English, right?)
All pupils (English):	One, two, three, four, five, six, seven, eight, nine, eleven, twelve, thirteen, fourteen.
Teacher (Frisian):	En K., dy is hiel knap, want Kuba syn opa en oma wenje noch yn Poalen, hin, K.? (En K. is really smart because his grandma and grandpa still live in Poland, right, K.?)
Polish-speaking pupil:	Ja (Yes.)
Teacher (Frisian):	Kinsto ek yn it Poalsk telle? Wolst it foar ús dwaan? (Can you also count in Polish? Can you do it for us?)
Polish-speaking pupil (Polish):	(Stands up and starts counting) Jeden, dwa, trzy, cztery, pięć, sześć, siedem, osiem, dziewięć, dziesięć, jedenaście, dwanaście, trzynaście, czternaście. (One, two, three, four, five, six, seven, eight, nine, eleven, twelve, thirteen, fourteen.)
Teacher (Frisian):	K. krijt fan ús in hiele dike …  (K. gets from us a really big …)
Teacher and pupils:	Duim! (all pupils put their thumbs up). (Thumbs up!)
Teacher (Frisian):	Mar wy hawwe N. ek yn ‘e klasse, en N. komt fan Syrië. En N. kin yn Arabysk telle, hin? Mar N. begjint noait mei de tomme, N. begjint mei de pink te tellen. Moatte jim mar ris sjen. (But we also have N. in our class and N. comes from Syria. En N. can count in Arabic, right? But N. never begins with her thumb, she begins counting with her little finger. Look at it when she counts).
Arabic-speaking pupil (Arabic):	واحد, ثنان , ثلاثة, أربعة, خمسة, ستة, سبعة, ثمانية, تسعة, عشرة, أحد عشر, اثنا عشر, ثلاثة عشر, أربعة عشرة. Wahid, athnan, thlathe, arbahe, khmse, stte, sbahe, thmaniya, tsahe, eshrahe, ahd eshr, ‘iithnaan eshr, thlatht eshr, arbahet eshr. (One, two, three, four, five, six, seven, eight, nine, eleven, twelve, thirteen, fourteen)
Teacher (Frisian):	Wat knap, hin. Dankjewol. (That’s really good, right? Thank you). Hé, en Ponpon, de big Ponpon en de small Ponpon, dy meie wer nei hûs ta gean, en dan sizze wy bye, bye. (And Ponpon, the big Ponpon and the small Ponpon, they can go home again. And we say bye, bye).
All pupils:	Doei, lieve Ponpon. (Bye, dear Ponpon).

5.2.1. Paraphrase of excerpt

The sequence starts with the morning greeting in the three official languages of the school and in two other languages of pupils present in the class (Polish and Arabic). One pupil is representing each of these languages standing up in front of the group. Greetings are quickly done in the different languages after each other, indicating that it is a routine. The teacher sticks to the Frisian instruction language to manage this activity. Afterwards, she asks two pupils to get a doll (Ponpon) used to communicate in different languages. She then asks questions in Frisian so that the pupils identify the day of the week. Next, the days of the week are recited in the three languages of the school by both pupils and teacher. The teacher then wishes to know the exact date and after they mention the 14th of December she indicates that pupils should count up to fourteen using the three languages again. Once this is done she asks both the Polish- and the Arabic-speaking pupils to also count to 14 in their languages. She gives them compliments for this. In addition, she mentions that counting in Arabic with the fingers means starting with the little finger, which is the opposite of how the other pupils are used to. She interacts with these pupils also in Frisian.

5.2.2. Analysis of excerpt

Discussion with the teachers on the video footage from this sequence led to the identification of two functions of translanguaging. First, it is employed as a bridge between the instruction language of the day (Frisian) and the other two languages of instruction within trilingual instruction (Dutch and English). As such, it allows for less language compartmentalisation than in the traditional triple immersion programme the school used to follow, in which instruction languages were kept strictly apart. It is thus a tool to break with what Cummins calls the ‘language solitude’ premise within mainstream education (2008), describing the fact that languages are often compartmentalised and separated in education, ‘as if they belonged to different nation-states or different speech communities’ (García & Wei, 2014, p. 227).

Second, the involved teachers specifically mentioned using translanguaging as a scaffold to link knowledge in the three languages of instruction to knowledge in the home languages Polish and Arabic, by including them systematically in the daily routines of the group. This is carried out without further explanations by the teacher, suggesting that it is a routine to greet, recite the days of the week and count in several languages. However, it is noticeable in the sequence that the teacher mostly sticks to the Frisian language herself and uses it to elicit the other languages. Both researchers and teachers see translanguaging here thus as fulfilling a scaffolding function offering temporary bridges between languages which allow pupils to build links between official instruction languages and between home and school languages. These scaffolding moments acknowledge all different languages by giving them the same role and relevance in daily classroom routines. In addition, and from the perspective the Polish- and Arabic-speaking pupils, translanguaging as a scaffold renders their family languages as an exceptional resource, as seen in the collective thumbs up given to the pupils for counting in their home languages.

6. Discussion

The present paper looked at the use of translanguaging-based approaches in two settings of mainstream education. A distinction can be made between natural and official translanguaging (Williams, 2012). While natural translanguaging occurs spontaneously in classroom interaction in order to enhance subject or language-related understanding, official translanguaging refers to explicit strategies employed by teachers in order to use several languages in class. The excerpts presented are examples of official translanguaging as they reflect planned-in and systematic activities on behalf of the involved teachers. They have also been jointly selected by and analysed with the participating teachers. As seen in the examples, translanguaging offered a safe framing for the use of different languages by both teachers and pupils.

The first research question of the paper aimed at identifying the main functions achieved through a translanguaging-based approach in the two settings. Within the described translanguaging spaces, various practices serve different functions depending on (a) whether the aim of the teachers is to acknowledge or actively use the different languages; (b) whether the teachers are proficient in the languages involved in the translanguaging moment, and (c) the types of languages involved.

Instances of official translanguaging with a symbolic function aim at acknowledging and valorising migrant languages within mainstream education and require from the teacher no proficiency in those languages. A scaffolding function is achieved when temporary but systematic bridges towards other languages are incorporated in everyday teaching, thus attributing equal value to all languages. Teachers require no knowledge of migrant languages to do this, as long as pupils are perceived as the experts for their own family languages. Similar aims can be reached by scaffolding the acknowledgement of various instruction languages present within the teaching model (in this case Dutch, Frisian, and English). Finally, official translanguaging can also fulfill an epistemological function when the different languages are actively used to enhance both content- and language knowledge. This is suitable for exploring migrant languages in their full potential as learning instruments. To this end, a teacher proficient in those languages is needed to interact with the pupils. Table 6 provides a summary of these functions.

Table 6. Overview of functions of official translanguaging.

Functions of official translanguaging	Aims (acknowledgement or use)	Proficiency of teacher in the language	Types of languages
Symbolic function	Acknowledgement	No proficiency is needed	Migrant
Scaffolding function	Acknowledgement & use in daily routines	No proficiency is needed (except in the instruction languages)	Migrant and minority
Epistemological function	Use for content- and language learning	Teacher (or assistant teacher) is proficient	Migrant

The second research question referred to the ways teachers shaped the concept of translanguaging and relate to the methodological approach of the studies. While at the beginning of the projects teachers reacted sceptically to the concept of translanguaging, the design-based approach (Cobb et al., 2003) allowed them to develop their own didactical experiments and first implement those in their teaching at a small scale. It was a thus step-wise process until a translanguaging pedagogy could be established in these classes. In order for this to succeed, teachers needed to (a) create safe spaces in which to experiment with multiple languages in classroom and (b) operationalise the concept of translanguaging for their own context and particular aims. This design-based approach was successful in fostering ownership of the developed translanguaging-based approaches which was then translated into the translanguaging practices presented in the excerpts.

7. Conclusion

The present paper aimed at discussing examples of implementation of translanguaging-based approaches in two contexts of mainstream education, in order to identify different functions of translanguaging and to reflect on the ways in which teachers use the concept to serve their specific purposes. According to García and Wei ‘adopting a translanguaging lens means that there can be no way of educating children inclusively without recognizing their diverse language and meaning-making practices as a resource to learn’ (2014, p. 227). This could be seen in the two contexts. The presence of a Portuguese-speaking teacher in Luxembourg allowed pupils to explore their multilingual repertoires to acquire new knowledge with high cognitive involvement, whereas the Frisian example showed how official translanguaging can acknowledge different languages and incorporate them into classroom routines. The translanguaging spaces displayed here enabled pupils to actively use their dynamic plurilingual practices for learning. The typology of pedagogical functions within official translanguaging developed in this paper can guide future teachers in the development of their own translanguaging experiments.

Regarding the implementation of translanguaging-based pedagogies, design-based research offered the necessary support for teachers to progressively operationalise the concept of translanguaging for their own contexts. Bottom-up and tailored approaches owned by the teachers themselves can thus be a way to promote translanguaging-based approaches within MLE.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Ministry of Education, Children and Youth (Luxembourg) and by the Regional Government of the Province of Fryslân (Netherlands).