Keywords

1 Introduction

Translanguaging has been defined as ‘the ability of multilingual speakers to shuttle between languages, treating the diverse languages that form their repertoire as an integrated system’ (Canagarajah, 2011, p. 401). While theoretically exciting, language teachers have reported struggling with how to make use of learners’ full linguistic repertoire within the popular task based language teaching (TBLT) approach in the language classroom (e.g. Carless, 2004, 2008). This chapter examines the relationship between translanguaging and TBLT in depth, answering the question of how both approaches can co-exist in the language classroom. The translanguaging paradigm shift is beginning, and this is a time for the well-established TBLT to work with and alongside it.

We take an applied sociolinguistics approach,Footnote 1 asking what roles learners’ full linguistic repertoires, including first languages and affiliated first acquired linguistic resources (L1s), play in oral task performance in the language classroom. This is a key question when language teachers use interactive classroom tasks, especially in the many EFL contexts in which learners often share a common L1 (Storch & Wigglesworth, 2003; Storch & Aldosari, 2010). We have decided to use the term “L1” to encompass all of an individual’s naturally acquired linguistic resources during youth, and we contrast this with “L2” to mean any socially defined ‘language’ that is purposefully learned in addition to L1. By utilising this terminology, we maintain a bridge between TBLT and translanguaging, using TBLT recognised terminology while making room for translanguaging conceptualisations of the linguistic repertoire, a bridge which encompasses the essence of this chapter. We then illustrate our discussion with data from Vietnamese learners of English participating in classroom speaking tasks, as well as interviews with the students.

Finally, this chapter provides a mandate for TBLT to embrace ‘the intuitive communicative strategies multilinguals display in everyday life’ (Canagarajah, 2011, p. 401). We maintain that TBLT has space for translanguaging. Furthermore, actively making and maintaining space for translanguaging in TBLT will both enrich TBLT’s research agenda and extend the impact of TBLT beyond its typical second/foreign language education sphere of influence.

2 Situating L1 in Translanguaging and TBLT Research

In education, traditional codeswitching analysis (where the respective authors have used the term “codeswitching” and consider languages to be discreet, definable, and separable linguistic systems) and the monolingual bias result in “L1” use (or use of the full repertoire, as described in the section above) often being overlooked or discouraged (May, 2014). However, since translanguaging considers use of the full repertoire highly beneficial and argues that when learners use their existing linguistic resources they can better acquire new linguistic resources (García & Wei, 2014; Weber, 2014), there is an inherent challenge posed to dominant views of a monolingual norm in education. Where learners are cut off from portions of their linguistic repertoire, either through social or other factors, growing the rest of their repertoire is harder. Due to its grounding in a sociocultural framework, translanguaging further emphasises that use of the full linguistic repertoire is natural: speakers naturally use whatever features best suit the social and linguistic situation (Canagarajah & Wurr, 2011; García & Wei, 2014).

However, this exciting approach to languages education hits some barriers when it comes to applying it in one of the most popular forms of language teaching today – Task Based Language Teaching (TBLT), which traditionally sits within a cognitivist/psycholinguistic paradigm (Skehan, 1998). TBLT is, simply put, teaching through tasks.Footnote 2 As explained further by Ellis (2003, p. 16),

A task is a workplan that requires learners to process language pragmatically in order to achieve an outcome that can be evaluated in terms of whether the correct or appropriate propositional content has been conveyed… A task is intended to result in language use that bears a resemblance, direct or indirect, to the way language is used in the real world. Like other language activities, a task can engage productive or receptive, and oral or written skills and also various cognitive processes.

Questions concerning the role of L1 in the teaching and learning of additional languages have occupied language teachers and applied linguistics for as long as either has existed (Hall & Cook, 2012; Howatt, 1984; Cook, 2001). However, the field of TBLT has been relatively slow to engage with this question directly, especially in the formative years of the development of the field from the 1980s to early 2000s (although more recent research shows promise, cf. Moore, 2017).

One explanation for the reluctance in TBLT research to engage with the L1 question can be found in TBLT’s roots in communicative language teaching (CLT), an approach that privileged the native speaker teacher and tended to adopt a subtractive view of L1 use in the classroom. Traditional CLT discouraged L1 use because it reduced opportunities for L2 input processing and communicative practice (Cook, 2001). The theoretical roots of TBLT in Long’s Interaction Hypothesis (1996) and Krashen’s Input Hypothesis (1985) gave credence to this subtractive view.

A second contributing factor involves the ‘second language’ settings (c.f. foreign language or multilingual settings) in which TBLT research in its formative years was carried out and which typically involved research participants drawn from classes of adolescent or young adult pre-sessional ‘English as a Second Language’ (ESL) students at universities or colleges in North American or the United Kingdom (e.g. Bygate, 1996; Doughty & Pica, 1986; Gass & Varonis, 1985). Students in such classes do not always share common L1s with each other and/or with the teacher and so their L1s are often denied a role in classroom discourse. In research conducted in such contexts, researchers often choose to conveniently avoid L1 use by placing learners in groups with learners from other L1 backgrounds to perform communication tasks (e.g. Doughty & Pica, 1986; Newton, 2013). Given this situation, it is hardly surprising that native speakerism has dominated much of the research and theorising around TBLT, with the monolingual native speaker treated, by default, as the ideal model or target, and the full linguistic repertoire of learners made all but invisible.

However, there are notable exceptions. For example, Antón and DiCamilla (1998) examined the socio-cognitive functions L1 played in collaborative dialogue between learners engaged in a writing task. L1 was used to maintain intersubjectivity (i.e. to establish a shared perspective on the task), to scaffold work on the task, and to regulate mental activity (i.e. through private speech). L1-use has also been a consistent theme in task-related research conducted in Canadian bilingual and immersion contexts, notably by Merrill Swain and colleagues. Swain & Lapkin (2000), for instance, investigated the ways that grade 8 French immersion students in Canada used L1 to perform a jigsaw and Dictogloss task. While L1 was shown to be used most frequently for the purpose of task management, it was also used to clarify aspects of grammar and vocabulary, and to a lesser extent for interpersonal talk.

A small collection of TBLT studies over the last 10–15 years has continued this tradition of investigating L1 use in task-based interaction. Storch and Aldosari (2010) found a ‘modest’ amount of L1 use in pair work by EFL students at a Saudi Arabian college, with the type of task having a stronger influence of L1 use than proficiency pairing. L1 was used primarily for task management and deliberations over vocabulary. Lasito and Storch (2013) compared L1 use in pair and small group work by adolescent Indonesian EFL learners, finding greater L1 use in pairs where it was primarily used for task management and for dealing with unfamiliar vocabulary. Moore (2013, 2017) investigated L1 use by university-level Japanese EFL learners as they worked in pairs to prepare for an oral presentation. Moore found a relatively high level of L1 use (28% in the 2013 data), although with individual learners displaying consistently high or low L1 use patterns. L1 talk occurring across all categories of talk, though, particularly focused on procedural matters and off-task talk. Additionally, in a study which drew on interviews with English teachers and teacher educators in Hong Kong, Carless (2008) highlighted the complexity of the issues surrounding L1 use in task-based classrooms in this context, concluding that “a balanced and flexible view of MT [mother tongue] use in the task-based classroom” (p. 336) is needed.

While the studies discussed above typically focus on adolescents or young adults performing tasks, there is also an emerging strand of TBLT research that is concerned with the viability of TBLT for young learners in pre-school or primary school contexts. In this research, L1-use in peer-peer interaction is a noticeably more consistent theme. For example, Tognini and Oliver (2012) reported on the use of L1 in primary and secondary foreign language classrooms in Australia, with a subset of their data focused on tasks. They found that, predictably, L1 use increased as tasks became more demanding, and was used more in pair or group work in which students had to work collaboratively to construct a text, such as a role play or arguments for a debate. Azkarai and García Mayo (2017) investigated the use of L1 by Spanish primary school EFL learners, with a particular focus on the effect of task repetition on L1 use. They found relatively modest L1 use (although more than has been reported in studies involving adult participants), with its main functions being to appeal for help (i.e. to find a word), for metacognitive talk, or as borrowings from Spanish to keep the task flow. L1 use fell significantly in task repetitions. Additionally, Shintani (2014) showed how beginners in pre-school used L1 during input-based tasks to complete the tasks successfully. The learners used L1 for meta-talk and to communicate with the teacher. Over the five weeks of data collection, the learners began to rely less on their L1 and to use English more.

The consistent orientation in all of these studies is towards the productive functions played by L1 in task performance, especially its role in metacognitive talk, task management and appeals for help. It is interesting to note that the topic of L1 use features so prominently in research involving young learners (and by default, often beginners). To put it another way, L1 use appears to be infrequently reported in task-based research except when it involves young learners, which is an interesting emerging trend. This is also possibly why TBLT and translanguaging studies have not spoken to each other much, as the majority of translanguaging research to date has focused on teenage or adult learners (see Seals and Olsen-Reeder, 2019 for notable exceptions to this).

The TBLT studies reviewed above have all deliberately focused on L1 use. It is possible that other studies may report L1 data without it being a main research focus (e.g., Ellis and Yuan, 2004). However, Plonsky and Kim’s (2016) meta-analysis of 85 TBLT studies involving learner production found that only six explicitly accounted for L1 use in the data. We conclude that the majority of TBLT studies involving learner production either design L1 use out of the data (e.g. by requiring tasks to be performed in L2 only), do not report it, or exclude it from analysis when it does occur. The third of these options is illustrated in Park’s (2010) study on pre-task planning and task-based interaction in which examples of language related episodes containing both L1 and L2 are supplied but with no mention of this L1 data, or indeed any reference to L1 use.

Given this situation, it is not surprising that our search for TBLT studies which adopt an explicit translanguaging view of language has come up almost empty handed. There are positive signs in this direction however. Moore’s research (2013, 2017), for example, draws on the concepts of languaging (Swain, 2009) and translanguaging (García, 2009) to frame L1 use as ‘situated discursive practice’ (p. 240). Moore’s (2013) conclusion that L1 use is a ‘naturally occurring phenomenon in the L2 classroom as in bilingual communication’ (2013: 251) points promisingly to the synergistic relationship between TBLT and translanguaging that we are advocating for in this paper. In addition, recent years have seen researchers engaging with TBLT across a much broader range of contexts, including primary school language classrooms (e.g., García Mayo & Imaz Agirre, 2016; Newton & Bui, 2017), bilingual and multilingual classrooms (e.g., Álvarez & Pérez-Cavana, 2015) and English Medium instruction settings (e.g. Moore, 2017). We are hopeful that, as a consequence, the field will continue to engage more critically with a broader range of orientations and perspectives on the nature of language than has traditionally been the case.

3 Similarities and Differences in Principles

We have established that there is room for translanguaging in TBLT empirical research and that it likely is indeed present but often not focused on. But what about its fit theoretically? In investigating the characteristics of a translanguaging-friendly pedagogy, García & Sylvan (2011) established that there are eight guiding principles:

  1. 1.

    Celebrating heterogeneity and singularities in plurality

  2. 2.

    Collaboration among students

  3. 3.

    Collaboration among faculty

  4. 4.

    Learner-centred classrooms

  5. 5.

    Language and content integration

  6. 6.

    Plurilingualism from the students up

  7. 7.

    Experiential learning

  8. 8.

    Localised autonomy and responsibility

These can then be compared to the guiding principles of TBLT. Predictably, TBLT treats ‘a task’ as the core element in classroom language teaching and learning. Definitions of tasks have proliferated over the years but centre around the key elements in Ellis’s (2003: 4) definition of a task as “an activity which requires learners to use language, with an emphasis on meaning in order to attain an objective”. More recently however, Ellis (Ellis, 2018; Ellis & Shintani, 2013) has argued that this and other similar definitions fail to make a crucial distinction between ‘task’ as an educational unit of planning (task-as-workplan) and ‘task’ as the activity that learners engage in when they perform a task (task-in-process). Ellis argues that the starting point for TBLT should be the former - task-as-workplan, which is defined with respect to four key features:

  1. 1.

    The primary focus on meaning; learners are primarily focused on comprehending and/or producing messages (Ellis, 2018, p. 12).

  1. 2.

    There is some kind of gap. The task-as-workplan presents the learners with a gap which needs to be filled with information, reasoning or opinions.

  2. 3.

    Learners need to use their own linguistic and non-linguistic resources. Rather than the task-as-workplan providing learners with the language resources they need to complete the task learners need to “draw on their own existing linguistic resources (potentially both L1 and L2) [emphasis added] and their non-linguistic resources (e.g. gesture; facial expressions) for comprehension and/or production” (Ellis, 2018, p. 12).

  3. 4.

    There is an outcome other than the display of language. The task-as-workplan specifies what communicative outcome the task is designed to accomplish, and learner performance is measured not in whether language is used correctly but on whether this outcome is established (Ellis, 2008, p. 12).

Furthermore, the following ten methodological principles (MPs) for TBLT were proposed by Michael H. Long (2009, 2015). He claims that these are “universally desirable instructional design features motivated by theory and research findings” (Long, 2009, p. 376).

MP1::

Use task, not text, as unit of analysis

MP2::

Promote learning by doing

MP3::

Elaborate input

MP4::

Provide rich input

MP5::

Encourage inductive “chunk” learning

MP6::

Focus on form

MP7::

Provide negative feedback

MP8::

Respect learner syllabi and developmental processes

MP9::

Promote cooperative collaborative learning

MP10::

Individualise instruction

As the guiding principles above show, there are divergent areas in the focus of translanguaging and TBLT that pose challenges when finding how they might work together. This is reflective also of the hesitancy that often accompanies discussions of translanguaging and TBLT in the same space. When examining TBLT and translanguaging’s divergent areas with respect to theory, there are two main areas of difference. Firstly, translanguaging has a broadly descriptive focus when looking at qualities of dynamic plurilingual pedagogies, while TBLT has a more granular focus on pedagogy – what to do and how to do it. Second, translanguaging has a much more explicit agenda of learner empowerment and metacognitive awareness of language and power, while, traditionally at least, TBLT has situated itself with a broadly cognitive orientation to SLA and focused on the impact of generalizable features of task design on language acquisition.

Where then is the common ground? In fact, when conducting a comparative analysis between the guiding principles of both TBLT and translanguaging, it may be surprising to some to find that two approaches that have developed in two such seemingly different directions (one socioculturally and one cognitively) actually have a lot in common, as elaborated in the bulleted list below:

  • Both translanguaging and TBLT align in their focus on student collaboration. Students working together to negotiate meaning is key to both approaches.

  • TBLT and translanguaging advocate for the importance of content and language integration. Stemming from a historical background in communicative language teaching for both has led to this recognised need of language learning’s usefulness to understand meaningful material.

  • TBLT and translanguaging advocate for experiential learning. It is not enough to merely receive language instruction passively; learners must actively take part in their acquisition of sociolinguistic knowledge.

  • Both approaches recognise the criticalness of learner-centred classrooms. Long gone are the days where a teacher-centred approach was deemed the most appropriate to language learning. Now it is understood that successful language learning and teaching must take place in an environment in which learners own their learning.

  • Pedagogy must be more fluid and needs-responsive. This recognition of learner differences emphasises both inter-learner and intra-learner difference across time. In order for learners to keep growing their sociolinguistic repertoire, language teaching must be adaptable to their needs.

  • There is a need to focus on functional, communicative language use. Similar to the second and third commonalities, language use must have a communicative purpose‚ and the material that is being taught in language classrooms must be functional for the learners (i.e. teaching to learners’ communicative needs).

In all, based on the literature and on these six principles, it seems evident that translanguaging and TBLT are not as far apart as they first may seem, and in fact have quite a lot in common. To illustrate how the two approaches can work together, we present an analytical illustration in the next section of this chapter.

4 Applying Theory to Practice

In this section of the chapter we explore the potential for complementarity in the analysis of classroom task-interaction data from a TBLT perspective and from a translanguaging perspective. The data we use here are drawn from a study carried out in EFL classrooms at a secondary school in Hanoi, Vietnam (Nguyen, 2013; Newton & Nguyen, 2019). The original purpose of the study was to understand how the EFL teachers in this school went about teaching with tasks in the speaking lessons, and how learners engaged with and performed the tasks. The researchers identified the well-established practice shared by all the EFL teachers in the school of giving learners the opportunity to first rehearse a speaking task in pairs or groups, and then for as many pairs or groups as time allowed to perform the task again publicly in front of the class. Data included video and audio recordings and field notes from observations of 45 lessons from nine Grades 10, 11 & 12 classes (15–18 year old learners) taught by nine different teachers. Interviews were conducted with the teachers and with selected students. One of the research questions addressed in the study concerned the quantity and purpose of L1 use in task rehearsal, which brought our attention to the presence of this data for a qualitative translanguaging analysis as well. We are aware that terminology is a challenge here, but that is in the nature of trying to start a conversation across approaches. As explained near the beginning of this chapter, we use the terms L1 and L2 here in a way that attempts to bridge understandings across TBLT and translanguaging research: “L1” refers to all of an individual’s naturally acquired linguistic repertoire in their youth, while “L2” refers to the socially defined ‘language’ that is the current target of the language learning taking place.

The entire data set from which the following excerpts come contains many examples of what we see as translanguaging behaviours, although none of the data were originally collected for translanguaging analysis. This reinforces that translanguaging is a natural response and practice that can be located in much TBLT research. Further, these data were originally analysed through a codeswitching lens. The behaviour termed ‘translanguaging’ below and ‘codeswitching’ originally, on the surface, look similar in form. However, there is an important distinction between codeswitching and translanguaging found in the applied theoretical lens. As explained by Seals (forthcoming, p. 3):

Chiefly, translanguaging is a macro lens through which language use can be viewed that acknowledges all parts of the linguistic repertoire as connected and equally valid. It is a position actively aligned with critical pedagogy… Within a translanguaging lens, it is entirely possible to have micro units of analysis such as codeswitching/codemeshing, etc. Therefore, a translanguaging lens does not preclude the existance or use of codeswitching and codemeshing. However, naming translanguaging is also naming an activist position.”

Through a re-analysis of the data below (originally analysed via a TBLT lens), we found that translanguaging (incorporating L1) was used by speakers for three key purposes: task management, giving or seeking assistance, and negotiation of meaning. Translanguaging therefore fulfils a crucial function in TBLT practice – it allows students to manage the task, to collaborate, and to extend their control over the language and forms used.

A summary of L1 use is presented in its original quantitative form in Tables 13.1 and 2. Data for both turns and words are presented because the turns varied considerably in length. Table 13.1 shows that L1 turns constituted around 43% (964 turns) of the total; L2 made up 38% (860 turns) and the use of both together made up 19% (422). On average, in each task rehearsal, the students produced 20 L1 turns (M = 20.08), 18 L2 turns (M = 17.91), and 9 translanguaged turns (M = 8.79). A Friedman test showed a significant difference in the size of the different types of turns (χ2 = 21.027, p = .000). A follow-up Wilcoxon signed ranks test indicated that the mean of L1 turns and L2 turns was significantly different from translanguaged turns (Z = −5.117, p = .000 and Z = −3.958, p = .000), but L1 and L2 turns did not differ statistically from each other (Z = −1.085, p = .278). This indicates that in task rehearsal, students produced L1 and L2 turns in similar amounts.

Table 13.1 Amounts of L1 and L2 use by turn in task rehearsal
Full size table
Table 13.2 Amounts of L1 and L2 use by word in task rehearsal
Full size table

With regards to the amounts of L1/L2 use by word, Table 13.2 shows that in total students produced more L1 words (55.8%) than L2 words (44.2%). On average, in each rehearsal, students used more L1 words (M = 164.37; SD = 122.49) than L2 words (M = 130.22; SD = 93.13). A paired-samples t-test showed no significant difference between these two means, t(47) = 1.643, p = .107. This again shows that the students used L1 in roughly equal amounts to the L2 target.

It is clear from these data that during dialogic rehearsals, the students used L1 substantially. This finding contrasts with findings of other studies on the amount of L1 use during pair and group work.Footnote 3 For example, Storch and Aldosari (2010) found a limited amount of L1 use with 15 pairs of EFL Arabic learners (7% for L1 words, and 16% for L1 turns). Other studies also found similar low proportions of L1 use. For example, Swain and Lapkin (2000) found that in their talk in preparation for a written task, Grade 8 French immersion students used L1 for 29% of the turns in the jigsaw task and 21% in the dictogloss task. However, the considerable amount of L1 use in Vietnamese classroom data reported above echoes Guk and Kellogg’s (2007) finding that the Korean EFL learners in their study used L1 in 47% of their utterances. Alegría de la Colina and García Mayo (2009) also found high L1 use – 55–78% depending on the tasks (jigsaw, dictogloss, text reconstruction). Alley (2005) also found students used English L1 predominantly in group work (71%), though for different mediating functions.

The considerable amount of L1 in the rehearsal stage in the present study can be explained in several ways. First, the students used L1 extensively because they treated public performance as final, and rehearsal as preparatory. Second, under time pressure (there were time limits for task rehearsal, on average 5 min), the students used L1 to sort out their ideas and marshal language resources to express the messages they wanted to convey. The following excerpt (translated into English) between the researcher (R) and students (focus group interview) illustrates these points.

Excerpt 1 Interview with Students About L1 Use

R::

Do you use Vietnamese when you work with each other?

S1::

Yes, a lot.

R::

Why?

S1::

Time for preparation [rehearsal] is often limited, if we use English right

away, it is very time-consuming. So we use Vietnamese first to be quick to prepare ideas and find English words later.

S2::

Also my friends might not understand all that I say if I say all in English.

R::

Umh huh. Do you think when you use Vietnamese, you’ll lose opportunities to speak English?

S1::

Not really, because the final thing is to speak up there in front of everyone.

R::

When you are up there, you can use only English?

S3::

Yes, because we have already prepared for it!

(Student focus group interview- 11D)

The students’ use of L1 also reflects their familiarity and comfort working with each other. Research into pair talk in EFL contexts (e.g., Storch & Aldorsari, 2010) has shown that when students become comfortable working with each other, they tend to use more L1 in their interaction. Crucially though, whether it was considered acceptable to use L1 was largely set by teachers. In these classes, by implementing a rehearsal phase in the first half of each of the speaking lessons in which L1 use could occur freely, the teachers had adopted an ideology which permitted translanguaging.

5 Functions of L1 and Translanguaging

In line with studies on L1 use in pair/groupwork, the current data show the students using L1 for a variety of functions. One such function was to discuss and resolve language problems. For example, students use L1 to explicitly ask for assistance concerning English words/phrases to express their intended meanings, weighing language solutions and giving explanations. Excerpt 2 displays this. Furthermore, this excerpt allows us to qualitatively analyse the discourse to investigate the use of translanguaging in this space.

Excerpt 2 Translanguaging to Request and Provide Metalinguistic Assistance

S3::

Air chi hèo?(Air what?)

S1::

Air pollute … air pollute phải không?(is it air pollute?)

S3::

Air polluted

S4::

Air pollution. Pollution là sự ô nhiễm

(Pollution is the state of being polluted)

S2::

Polluted là bị ô nhiễm (Polluted is passive)

Here S3 uses L1 to ask for the word that collocates with ‘air’. S1 provides the answer in English L2 but is uncertain and shifts back to L1 for a confirmation check. S4 and S2 then offer solutions and metalinguistic explanations, translanguaging as they do so, drawing upon L1 to provide deeper meaning while highlighting the focus word through L2.

Students also translanguaged to generate ideas, scaffold, and self-regulate as illustrated in Excerpt 3. In this task rehearsal, translanguaging is used by students as a form of bridging and negotiation of meaning. Bridging, like scaffolding, is where a gap exists in a student’s linguistic repertoire and they then acquire a corresponding lexical item from another language. However, unlike scaffolding, the use of translanguaging to fill the gap with a new lexical item is intended to retain lexical items in both languages, not simply in a target language and then to discard the other feature.

Excerpt 3 Translanguaging to Generate Ideas During Rehearsals

01:

S2: Bởi vì khi đau ốm … khi ill

(Because when they are sick … when ill)

02:

S1: They old (.) when they old (.) they old (.) they are old chứ! (should be they are old!) … sick or old!

03:

S2: They are sick (.) or old (.) their children will nuôi dưỡng (take care of) will erm

04:

S1: Take care of

05:

S2: Take care of

06:

S1: Them. Nếu có nhiều con thì erm nguồn lao động sẽ nhiều (If they have many children, they will have a good labour force) … if they have many children

07:

S2: Công nhân là workers (‘Workers’ is workers) … have workers

08:

S1: Then their family sẽ có nhiều người làm việc

(will have many workers)

Gia đình họ (.) (Their family) their family will have nhiều (a lot of) a lot of

S2 starts by providing metalinguistic commentary in line 01, almost entirely in L1. S1 then connects this meaning to English L2, but with false starts. S1 later reflects on his language use by self-correcting, saying ‘they are old’ instead of ‘they old’, with the emphatic Vietnamese ‘chứ’ directed to himself. This indicates his obvious noticing of the difference between the target-like form and his language production, showing the triggering of mental processes that lead to modified output (Swain & Lapkin, 1995, pp. 372–373). In other words, S1 has utilised translanguaging to extend his analysis beyond semantic processing to syntactic processing (Swain, 1995). Through this process, S1 is using translanguaging between L1 and L2 as a useful cognitive tool for accessing performance output in English L2.

S2 continues his contribution by starting from what S1 has said, and in line 03, mid-turn, he uses Vietnamese ‘nuôi dưỡng’ to regulate his L2 search. Sensing his interlocutor is having difficulty finding the needed word, S1 translanguages across speaker turns, offering the correct phase ‘take care of’, which S2 uses in his talk. At line 06, S1 completes S2’s utterance by adding the pronoun ‘them’ after the verb and keeps generating content in L1, followed by mapping that L1 meaning to L2 words via translanguaging. In lines 3–6, S2 is seen bridging, with input from S1 to acquire the feature “take care of” in L2.

Similarly, in lines 7–8, S2 and S1 translanguage to build upon meaning from L1 to retrieve L2 resources to express the message they want to convey. In brief, Excerpt 2 shows translanguaging being used as a sociocognitive mediating tool (Lantolf &Thorne, 2007), lending support to other studies that show how learners translanguage to mediate their use of L2 (Algería de la Colina & García Mayo, 2009; Alley, 2005; Antón & DiCamilla, 1998, Brooks & Donato, 1994; Guk & Kellogg, 2007; Storch & Aldorsari, 2010; Storch & Wigglesworth, 2003; Swain & Lapkin, 2000).

Excerpt 3 also demonstrates the collaboration and negotiation of meaning. The discussion in lines 02–03, 03–04, and 06–07 involves students refining, clarifying and negotiating meaning through all the linguistic resources they have, therein showing the usefulness of a positive stance towards translanguaging in the language learning space. The students are actively searching to find what they want to say, and how they want to say it, using every skill at their disposal, without the reservations that the monolingual bias brings. The task is better completed through students’ accessing their full linguistic repertoires.

Finally, Excerpt 4 below shows an example of what occurred when the students devised the opening and closing for the performance before acting it out.

Excerpt 4 L1 for Metadiscursive Commentary

S1::

Xí lên bắt tay đồ rứa nghe chưa. How are you đồ rứa nghe.

(When we are up there, remember to shake hands and the like. Also remember to ask ‘how are you’, and the like.)

S2::

Hi teacher, hi kids. You look very beautiful today! [Laugh] [Joking]

S1::

Oh, thank you! [Laugh] [Joking]

Excerpt 4 shows the use of students’ full linguistic repertoires to give voice to the metadiscussion. S1 uses L1 features to discuss how to proceed. The return to the performance is then signalled by S2 through the use of L2, the target language. Through using their full linguistic repertoire and translanguaging, the students are able to shift their conversation between task preparation and task performance.

While previously a codeswitching analysis within TBLT research might have viewed the excerpts above as “resorting” to the L1, a translanguaging analysis encourages a more nuanced view, while simultaneously drawing upon critical pedagogy framing. The use of L1 during L2 task preparation allows for a much more sophisticated negotiation of meaning to take place. Additionally, as shown in Excerpt 4, students are able to take on multiple “voices”, therein constructing two distinct but connected discussions simultaneously: the task performance in L2, with task preparation and discussion incorporating L1. Though some may argue this is due to a gap in the students’ English proficiency, translanguaging posits that this is actually a complex negotiation of the linguistic and social setting, without which students would lose an important tool to discursively negotiate and complete the task.

The learners’ verbal reports further confirm the roles of translanguaging as demonstrated above. For example, 29 of the 54 students said that they use L1 to prepare ideas or meanings first and then connect these established meanings to L2 forms:

  1. 1.

    Thinking in Vietnamese is powerful. I can think of millions of ideas that my limited English cannot express them all. My friends can help translate what I think into English. (HVT-10A)

  2. 2.

    I speak in Vietnamese first to search for and present ideas, and after that I turn those ideas into English. It’s like matching meanings to the English words. (NKHH-11F)

This indicates that the students view L2 as a bridge that connects meaning to L2 forms, which in turn convey that meaning across the bridge. Students further said they use L1 to sustain communication during L2 negotiation of meaning:

  1. 3.

    For example, in the middle of communication, I don’t know to express certain ideas, I don’t know what else to do but use Vietnamese to move on; I cannot let ideas flow out in English, only Vietnamese can help. (HDH-12I)

  2. 4.

    Sometimes we have to stop talking in English to use Vietnamese to give explanations. Sometimes we have very brilliant ideas but can’t express them in English, so using Vietnamese enables us to speak out ideas, which we later translate into English, and move on with our communication. (TTHL-10B)

Seen from a traditional TBLT perspective, the students’ comments above would indicate that students use L1 because they lack L2 resources, or are unable to access their L2 resources quickly enough. However, via translanguaging, the script is flipped, and instead, L1 is seen to be beneficial in bridging students’ cognition. Taken together, the students appear to view L1 as a mediating tool to support producing meaning in L2. This result corroborates the findings of previous research (e.g., Alley, 2005; Brooks-Lewis, 2009; Kim & Petraki, 2009) where students perceived the usefulness of using L1 in learning L2. As V. Cook (2001) argues:

Bringing the L1 back from exile may lead not only to the improvement of existing teaching methods but also to innovations in methodology. In particular, it may liberate the task-based learning approach so that it can foster the students’ natural collaborative efforts in the classroom through their L1 as well as their L2. (p.419, italics added)

The examples above, as well as corroborating prior research, show that TBLT pedagogy must accept the “fact of life” (Stern, 1992, cited in V. Cook, 2001, p.408) that “two [or more] languages are permanently present” (V. Cook, 2001, p.418). It then follows that if students are prohibited from using L1 in classrooms, they will be denied access to this useful tool (see Alegría de la Colina & García Mayo, 2009; Brooks-Lewis, 2009; Storch & Aldosari, 2010; Swain & Lapkin, 2000). Given that L1 use has been reported as one of the deterrents to the implementation of TBLT in EFL Asian contexts (e.g., Carless, 2008; Pham, 2007) and as one of the teachers’ ‘fears’ (Alley, 2005), the findings here concerning translanguaging as a cognitive tool further highlight the need to rehabilitate L1 in the TBLT space.

Overall the students’ positive perceptions of task performance following rehearsal were in line with the teachers’ dedication to bringing the task through to the public performance, which they described as the ‘happy ending’ of the task-based speaking lessons. In the current study, students made use of translanguaging to a considerable extent, and this played an important mediating function. By translanguaging between their available linguistic resources, students were able to successfully negotiate the task and perform it in the still required target form (i.e. English L2), therein taking nothing away from the TBLT agenda and instead assisting students through access to their full linguistic repertoire. This additionally shows that there is space for both TBLT and translanguaging in the language classroom, both empirically and theoretically. This illustrative analysis shows a beginning to how we might integrate TBLT and translanguaging in research and practice.

6 Concluding Discussion: What Is Gained from Integrating Translanguaging and TBLT?

As the above discussion of principles and illustrative analysis have shown, much is to be gained from integrating translanguaging and TBLT instead of pursuing differing research and pedagogical tracks. The above illustration showed how a focus on a TBLT pedagogy drove a shift to a learner-centric pedagogy (e.g. group work). This in turn opened up space for translanguaging. Consequently, translanguaging became a critical tool by which learners were able to successfully negotiate and perform the speaking tasks, therein further developing their sociolinguistic gains. The learners themselves also expressed the necessity and usefulness of translanguaging, particularly during the task rehearsal phase. The learners were empowered in their language learning and built up their confidence for the task performance, having already metadiscursively discussed how they would perform the task and why.

This empowerment of learners is further highlighted through a focus on translanguaging, making sure that this crucial element of learning is not lost in the classroom. For teachers, a translanguaging perspective offers a socially accountable and theoretically justifiable basis for harnessing learners’ full linguistic repertoires in the classroom, even if the ultimate aim is proficiency in target language (as is often the case in TBLT classrooms). That being said, through the adoption of a translanguaging perspective, teachers and learners are also more fully able to realise the meaning making goals of TBLT, freed from the constraint of only making meaning in the target language.

Additionally, translanguaging in the classroom more closely mirrors actual real-world language use, particularly in multilingual settings (which is indeed the majority of the world). This helps TBLT classrooms to maintain a more functional purpose, better suiting learners for the real communicative practices they will encounter outside of the classroom.

In sum, translanguaging issues a challenge to TBLT. It asks what TBLT is doing to challenge the status quo and to offer more opportunities to more learners, particularly those from minority communities and contexts who are trying to navigate the dominant systems of learning. As stated by Long (2015, p. 4) in discussing TBLT: ‘a responsible course of action… is to make sure that language teaching (LT) and learning are as socially progressive as possible.’ Engaging with translanguaging is one way that TBLT might more fully realise this socially progressive goal.