Research into practice: The task-based approach to instructed second language acquisition

Abstract

This article discusses the phenomenon of task-based language teaching (TBLT) in instructed additional language settings. It begins from the premise that, despite considerable theoretical and empirical support, TBLT remains a contested endeavour. Critics of TBLT argue that, particularly with regard to time-limited foreign language instructional contexts, TBLT's learner-centred and experiential approach to second language acquisition fails to provide an adequately structured environment that allows for sufficient exposure to frequent language, and processing and practising of grammatical form. At the same time, differences emerge between how TBLT is conceptualised in theory and how TBLT is operationalised practically in many additional language classrooms. These realities signal the need to look at the interface between theory, research and practice. The article considers what current research into TBLT has not succeeded in getting through to classrooms, what has succeeded in getting through reasonably well, and what has been over-applied. It is concluded that the under- and over-application of theory and research in practice highlight the difficulty in identifying exactly what TBLT is or should be in instructed contexts. The article proposes a way forward to strengthen the effectiveness of the TBLT endeavour.

1. Introduction

1.1 TBLT – a contested endeavour

Task-based language teaching (hereafter TBLT) has been developing over several decades as a learner-centred and experiential pedagogical approach arising from the tradition of communicative language teaching (CLT). TBLT focuses on the use of ‘tasks’ in additional language (L2) classrooms. More particularly, students’ interactive engagement with tasks aims to integrate several components of the language learning process (such as focusing on pragmatic meaning, giving due attention to grammatical form, cognitive processing and scaffolding opportunities). On this basis TBLT arguably drives second language acquisition (SLA) forward more adequately than other CLT approaches that may place greater emphasis on an individual component of the process (Nunan 2004; De Ridder, Vangehuchten & Seseña Gómez 2007; Willis & Willis 2007). TBLT thereby challenges some of CLT's more established approaches, in particular the Presentation-Practice-Production (PPP) sequences that, following on from the grammar-translation tradition, focus primarily on teacher-led grammatical instruction and grammatical practice exercises.

As Van den Branden, Bygate & Norris (2009) assert, there are both theoretical grounds and empirical evidence to support a belief that TBLT is able to meet all the requirements for successful SLA in a variety of contexts and among a range of learners. In instructed contexts where exposure to the target language is constrained and largely reserved for the classroom, TBLT has been shown to be effective (see, e.g., Shintani 2016).

There are, however, theorists who have railed against the TBLT endeavour, particularly in time- and input-limited contexts, claiming its inadequacy and inefficacy (e.g., Klapper 2003; Bruton 2005; Swan 2005). Alleged limitations focus on the failure of TBLT to provide adequate exposure to frequent language, failure to explain grammatical concepts and rules sufficiently, and failure to offer sufficient practice with the language. Swan, in particular, has argued for a more teacher-led, structured approach.

Also, despite considerable theoretical and empirical support, TBLT in practice can end up looking very different to TBLT in theory or empirical study. This can lead to a perception of discordance between its advocates (e.g., theorists and researchers) and its recipients (e.g., teachers). Such dissonance can result in the parodying of TBLT as ‘the long-awaited elixir of language teaching’ (Richards & Rodgers 2014: 177) promulgated by the leaders of ‘a major religion’ whose devotees misuse it, with consequences ‘similar to the mismatch often arising between the orthodoxy of the priesthood and the pragmatic syncretism of adherents’ (Hadley 2013: 194).

Such hyperbolic caricaturing of TBLT, alongside critique of TBLT, signals the need to take a close look at the interface between theory, research and practice. With time- and input-limited instructed contexts principally in mind, this article looks at three different interfaces between research and practice – what current research in the TBLT space has not succeeded in getting through to classroom teaching, and why; what has succeeded in getting through reasonably well; and what has been over-applied.

It is worth stating at the outset the standpoint from which I make the observations that follow. It seems I am ranked by Hadley (2013: 194, 195) as among the TBLT priesthood, portrayed as ‘a true believer,’ guilty of ‘rosy prognostication’ and ‘clear bias,’ and unable to discern (or unwilling to own) the reality of messy classrooms. On the contrary, I am very mindful of the difficulties teachers may encounter when implementing TBLT as innovation in real-world classrooms and time-limited instructional contexts. This ‘Thinking Allowed’ piece enables me to present a personal viewpoint shaped largely by my work as a language teacher educator with a keen research interest in the interface between TBLT in theory and TBLT in practice. My aim is not to present a comprehensive overview of all research into or theoretical underpinnings of TBLT. Rather, I wish to highlight what I see as several key pedagogically important issues and provide a perspective on where I see theory, research and practice interacting and not interacting successfully.

1.2 The theory-research-praxis nexus

The interface between theory, research and practice is an important one for me as a language teacher educator. In particular, my work with teachers at the chalk face who continually bring their joys and struggles with TBLT to my attention leads me to reflect on questions such as these: Does TBLT need to reframe itself to accommodate more traditional elements? Or do teachers need to be challenged to be more open and receptive to innovation?

On the one hand, Markee (1997) argues that often SLA theory and research are ineffective in promoting changes in practice because they do not address the real-world concerns of teachers in classrooms. Teachers’ experiences with TBLT must be allowed to inform theoretical arguments about TBLT going forward. On the other hand, as Van den Branden (2009a: 666) puts it, whatever the nature of the innovation, there is a tendency for teachers to ‘teach in the way they themselves were taught, and show strong resistance toward radically modifying the teaching behavior that they are so familiar with.’ A mismatch between theory and practice does not have to suggest problems with TBLT itself. Rather, it suggests problems as the enactment of TBLT interacts with teachers’ beliefs about effective teaching.

1.3 A fundamental issue – teacher cognition

When it comes to implementing any pedagogical approach, including TBLT, teachers inevitably bring their own beliefs and understandings about effective pedagogy with them into their own classrooms. Nunan (2004: 6), for example, argues that, although often not immediately visible, everything teachers do in classrooms is ‘underpinned by beliefs about the nature of language, the nature of the learning process and the nature of the teaching act.’ There are several sources of influence on teachers’ beliefs, with consequent impact on classroom practices (Pajares 1993; Phipps & Borg 2007). It is recognised that teachers’ beliefs are strongly shaped by their own early learning experiences, and may be resistant to change. Indeed, all beliefs held by teachers may become filters of new information and experiences, influencing teachers’ interpretations in the present. However, when teachers’ beliefs are acknowledged and addressed within teacher education programmes, there is evidence to suggest that new understandings can be established successfully (Cabaroglu & Roberts 2000; Borg 2003).

Seen from the above standpoint, my work as a language teacher educator is firmly grounded in the principle of reflective practice, and an understanding that facilitating critical thinking about past and present beliefs alongside actual experiences in the classroom will enhance the likelihood of changes both to beliefs and to future practice (Bullock & Muschamp 2004; Chien 2013). A consequence of the intersection between teachers’ experiences and teachers’ beliefs is the need to provide adequate research-informed teacher education initiatives that can support teachers more effectively and help them to mediate a middle path between reframing of TBLT (i.e., acknowledging that more established pedagogical processes may have a role to play) and embracing innovation (i.e., trying out something new and unfamiliar). As with determining the strengths and limitations of any approach, stakeholders need opportunities to try out, explore and test TBLT against theoretical claims and research findings. In this vein I provide the following reflections on the theory-research-praxis nexus.

2. Research findings that have not been well applied

2.1 Challenges of turning theory into practice

The essential pedagogical drivers for TBLT were as means to address the inadequacies that began to become evident with more traditional realisations of CLT. Two polarisations of CLT serve to illustrate the perceived inadequacies (although it must be conceded that, in practice, CLT has been operationalised in a range of ways that sit at different points between the two extremes). So-called strong CLT, with its zero-grammar or focus on meaning approach (Long 2000) influenced by immersion contexts and innatist thinking (i.e., there is no need to teach the students anything about grammar; they will acquire the rules as they go along), meant that students developed inadequate grammatical competence. This was presumably because the rules underpinning the language had been only partially processed and assimilated. In contrast, so-called weak CLT, drawing on what Long calls a focus on forms approach influenced by more traditional teaching practices (i.e., paying overt attention to grammatical forms, taught deductively), meant that students were hindered in their development of adequate fluency. This was presumably because the rules were too much in the foreground and there were limited opportunities for authentic communicative interaction. TBLT has been proposed as ‘an attempt to harness the benefits of a focus on meaning . . . while simultaneously, through use of focus on form (not forms), to deal with its known shortcomings, particularly rate of development and incompleteness where grammatical accuracy is concerned’ (Long & Norris 2000: 599). ¹

Empirical studies into a range of variables (e.g., task sequencing, task repetition, task complexity, uptake of feedback) came subsequently to influence thinking about how TBLT can promote SLA. Robinson's (2011) valuable summary makes it apparent that theorising and empirical findings have been generally in favour of TBLT. However, a limitation of empirical studies into the effectiveness of tasks, which Robinson acknowledges, is that they are commonly carried out in controlled experimental settings. It can be difficult to draw conclusions from them that are generalisable to actual real-world classroom environments.

To underscore the challenges for TBLT at the chalk face, the dimension of research that has investigated TBLT from the teachers’ perspective has brought to our attention the reality that teachers hold a range of interpretations and understandings of TBLT (Carless 2003, 2007, 2009; Van den Branden, Van Gorp & Verhelst 2007; Andon & Eckerth 2009; Van den Branden 2009b; Xiongyong & Samuel 2011; East 2012; Zheng & Borg 2014). It is clear that several dimensions of theory and research are not getting through to practice.

2.2 Understanding ‘task’

A crucial challenge for TBLT in practice is teacher uncertainty about exactly what a task is for purposes of TBLT. From a theoretical perspective a range of operational definitions of task have been developed (Samuda & Bygate 2008). Ellis (2009b) brings together the common strands apparent across these definitions, including: a primary focus on meaning (rather than on grammatical form); students’ reliance on their own resources (in contrast to being told what language to use); and an outcome to the task beyond the use of language (so that language is not being practised for the sake of practice). Much empirical research into task efficacy has been built on these fundamental understandings of task (see, e.g., Foster & Ohta 2005).

However, conceptual understandings of task often fail to get through to teachers. Teachers often struggle with differentiating a task from the kinds of more structured communicative activities that would have dominated the weak CLT classroom and a PPP model. For example, teachers may construct a simple transactional role-play scenario (waiter and customer in a café comes to mind), believing it to be a task because ostensibly it focuses on meaning (negotiating requests) and has an outcome in mind (receiving the requested item or items). Such a role-play, depending on how it is set up, may be highly structured, requiring the use of pre-learnt phrases and specific grammatical forms. The exercise may be useful to practise context-specific language, but it is not a task, essentially because its focus is not predominantly on meaning and non-linguistic outcome, but on practising pre-determined forms.

Even when teachers do have a level of understanding of what a task is, they may bring their own prior experiences and beliefs to bear on their understanding, and make a range of context-specific adaptations to task scenarios (Van den Branden 2009b), reshaping a given task so that it loses key features that originally made it a task. Of course, flexibility with the notion of task gives teachers considerable freedom to make context-specific choices about what they consider to be suitable tasks to foster student engagement, and how they will implement those tasks, and this is a potential strength. Problems can occur, however, when there is misunderstanding about even the basic notion of task and its place in teaching and learning sequences. If, for example, students should ideally be encouraged to rely on their own resources when completing the task, too much teacher intervention may detract from the realisation of this goal. Thus, tasks in practice can often end up being quite different to tasks in theory.

2.3 Form or forms

A second arena in which theory and research do not necessarily get through to practice concerns attention to grammar. From a theoretical perspective, grammatical accuracy within TBLT is acquired principally through the notion of focus on form (Long 2000). A central aspect of focus on form, which is largely concerned with implicit processing, is that it entails ‘briefly drawing students’ attention to linguistic elements . . . in context, as they arise incidentally in lessons whose overriding focus is on meaning, or communication’ (Long 2000: 185). On this basis, learners first need to notice or pay attention to the grammatical rule or form during the interactional process (Schmidt 1990, 1993, 2001). Once noticed, the grammatical principle will become acquired and accessible for subsequent use.

How teachers conceptualise and enact attention to grammar may be quite different to the principle of focus on form outlined above. Once more, teachers’ prior beliefs and understandings may play a role in this. The scenario of the teacher maintaining an up-front and top-down role (the teacher at the front of the class carefully explaining grammatical rules) is widespread (East 2012). Indeed, Larsen-Freeman makes this point in a recent ‘Thinking Allowed’ piece, arguing (2015: 263) that grammar instruction ‘has been relatively unaltered by research findings. It remains traditional for the most part, with grammar teaching centered on accuracy of form and rule learning, and with mechanical exercises seen as the way to bring about the learning of grammar.’

In summary, teachers’ misunderstandings about the concept of task and a tendency for many teachers to hold onto more traditional or teacher-fronted approaches to grammar represent but two of several phenomena where theory and research have often had inadequate influence on classroom practice. As Robinson's (2011) overview makes clear, a wide range of research has investigated a number of variables. Findings have supported the notion that TBLT has positive effects on SLA. However, much remains that has not found its way into teachers’ practices.

3. Research findings that have been well applied

There are, however, aspects of theory and research in the TBLT space that I believe are being embraced by teachers in their practices. Arguably the most valuable application in classrooms of theory and research has been the increased emphasis on helping students to engage in meaningful, authentic interactive tasks that promote fluency. Indeed, Brown (2007: 218) speaks of a ‘new wave of interest’ in the L2 classroom whereby teachers are ‘treating the language classroom as a locus of meaningful, authentic exchanges among users of language,’ with L2 learning seen as ‘the creation of meaning through interpersonal negotiation among learners.’

When it comes to seeing the practical application of TBLT theory and research, it is useful to consider those studies that have investigated stakeholder perspectives. Whilst the findings of such studies may be critiqued on the basis that they often rely on self-reports rather than experimental studies into task efficacy, and may therefore be subject to biases such as the halo effect, a cogent counter-argument is that they serve to illustrate stakeholders’ perceptions about the effectiveness of what is happening in classrooms. They can also illustrate teachers’ changing beliefs by virtue of reflective practice, and provide an important window into teachers’ actual thinking and decision-making. Various presentations of teachers’ and students’ engagement with tasks serve to underscore the positive potential of TBLT in practice.

Leaver & Willis (2004), for example, present a range of accounts of task-based initiatives for a variety of foreign languages, with a particular focus on tertiary programmes. These accounts report improvements in students’ language proficiency and enhanced student satisfaction and enthusiasm. Tasks were: seen to promote student engagement and risk-taking, with evaluations signalling ‘that students like tasks,’ and course results indicating ‘that students learn from tasks’ (p. 65) (courses for Slavic languages); viewed as ‘a tool that helps [students] carry out specific real-life activities’ (p. 77), contributing to a course that was perceived by students as ‘engaging and informative,’ ‘really fun,’ and ‘very interesting’ (p. 78) (courses for Spanish); considered to be ‘an excellent choice to motivate students and promote higher levels of proficiency,’ creating ‘a low-anxiety learning environment where students can try out their ideas and practice their language to develop confidence’ (p. 140) (courses for Japanese).

Accounts of primary school classroom practices presented by Van den Branden et al. (2007) also note advantages for learners similar to those identified in Leaver & Willis (2004). In three primary school contexts (Dutch-medium schools in Brussels, and English as an additional language in Hungary and Hong Kong) a task-based approach has stimulated ‘higher levels of involvement and motivation’ (p. 119); task repetition has helped students to ‘lower their anxiety and boost their confidence and motivation,’ with language seen as being used ‘for real purposes’ (p. 150); and task innovation has led to ‘positive impact’ for both teachers and learners in terms of ‘increasing their use of English’ (p. 252).

The conclusions of Van den Branden's report on work with teachers in real classrooms (2009b) suggest that, although in practice TBLT is complex and subject to the influence of teachers’ beliefs and perceptions, teachers (who were very well supported by teacher education initiatives) were in principle in favour of TBLT. They welcomed a task-based approach because they saw it as efficacious, and contributing to motivation, both extrinsic (the need to develop L2 communicative proficiency relevant to the context) and intrinsic (tasks were perceived as inherently motivating).

One project I recently undertook explored the possibility of challenging and changing school-based foreign languages teachers’ prior beliefs as a result of critical reflection on TBLT in theory and practice. I investigated two different courses, one for pre-service (beginning) teachers and the other for in-service (practising) teachers. Each course had a dedicated focus on, and clear exploration of, the theory of TBLT. Each provided opportunities for participants to try out TBLT in classrooms. Participants were subsequently required to reflect on their practical experiences in light of the theory. It was found that approaching TBLT in this way showed that it could be positively embraced, by both beginning and more experienced teachers, especially because of its promotion of meaningful and authentic language use through tasks in ways that are motivating for the learner.

For example, one beginning teacher (East 2014a: 269) argued, ‘in view of what I've seen and what I've experienced in my pre-service teaching experience, a well-considered task-based communicative approach results in a far more engaging, learner-centred, and authentic language learning experience than other methods observed.’ This teacher went on to conclude that, with TBLT, students were ‘more likely to become confident communicators of the language and have fun.’

One practising teacher (East 2014b: 694) noted shifts from the beginning to the end of the course. She initially stated heavy reliance on a more traditional teacher-led communicative approach. She explained, ‘[m]y planning before implementing TBLT involved much more structured, teacher-directed activities as well as a big focus on grammar. Then I normally let the students go and practise what I had taught.’ By the end of the course, and having implemented tasks in her classroom, she noted, ‘when recently speaking to my students about the difference in their learning, they said they felt like their learning was more hands-on and it allowed them to be more creative.’ Reflecting on feedback elicited from her students, she observed that, as a consequence of experimenting with tasks, ‘the purpose of their learning was made much more explicit and [it was] much more fun.’ This teacher concluded that there was now ‘more time spent using the language by the students which is the main goal instead of me doing it all, modelling, correcting, a big focus on grammar, etc.’

In summary, many teachers’ engagement with TBLT ideas has led in practice to increased emphasis on meaningful, authentic interactive tasks that students find motivating and that develop students’ fluency in the target language.

4. Research findings that have been over-applied

The positive outcomes noted above also signal the risk of a potentially imbalanced approach in classrooms. There are two intersecting dimensions which I believe demonstrate an over-application of research and theorising. The first is the misunderstanding that TBLT is all about spoken interaction. The second is that TBLT is all about pair and group work.

4.1 TBLT is all about speaking

Much research work in the TBLT space over several decades has focused on ‘interaction . . . to provoke negotiation for meaning’ (Bygate, Skehan & Swain 2001: 3), and, as a consequence, oral tasks (Skehan & Foster 1998; Ellis 2009a; Ahmadian 2012). Wigglesworth's (2001) useful overview of studies into the effects of different variables on students’ spoken proficiency serves to underscore this apparent emphasis. Furthermore, theoretical rationales for task demands are more clearly articulated for spoken production than they are for writing, reading and listening (Robinson 2011). This can lead to the impression that a task in TBLT is synonymous with a spoken task. This impression is strong and pervasive, simply by virtue of the fact that many tasks do indeed foster spoken interaction (Hughes 2011).

Certainly, a cursory reading of the theoretical literature around tasks would appear to encourage this limiting understanding. When, for example, Nunan (2004) argues that TBLT places emphasis on students learning to communicate ‘through interaction in the target language’ (p. 1) or ‘by communicating’ (p. 8), we can be left with a distinct impression that TBLT really is only about speaking. Furthermore, Philp, Adams & Iwashita (2014) underscore the valuable learning potential of peer-to-peer interactions from two theoretical perspectives – cognitive (e.g., Long 1983, 1996) and sociocultural (e.g., Vygotsky 1978) – thereby also reinforcing a potential bias in TBLT towards speaking.

An emphasis on speaking tasks in practice is borne out by several studies. For example, for the most part the tasks discussed by Van den Branden et al. (2007), drawing on a range of contexts, are speaking tasks. Additional data arising from my recent study among pre-service teachers (see earlier), some of which is presented in East (in press), reveal the same phenomenon. When, after having been introduced to TBLT in theory, the beginning teachers I worked with were left to their own devices to conceptualise and enact a task in a real-world classroom, they tended to resort principally to some form of spoken interactive task. Spoken interaction was elicited exclusively or predominantly in all but one of 16 tasks I considered for inclusion in that chapter. (The one exception was a writing task, with students required to respond, in Spanish, to a letter from an Argentinian boy, Julio, about their hobbies and free time activities.)

In fact, research into tasks has been broader than a focus on speaking. Byrnes & Manchón (2014) bring together a recent collection of work that focuses exclusively on writing, framing a range of activities from a task-based perspective. Task-based approaches to reading have been discussed, for example, by Green (2005), and Mounts & Smirnova (2011). Ellis (2001) argues for the efficacy of what he refers to as ‘non-reciprocal tasks’ and (2003) addresses the value of listening tasks. Moreover, an understanding that communicative approaches should foster an integrated skills approach acknowledges the reality that, when speaking, for example, listening is a necessary, important component (see, e.g., Hinkel who suggests [2010: 115] that TBLT is arguably ‘the most widely adopted model of integrated language teaching today’).

4.2 TBLT is all about pair and group work

Allied to a view that tasks are all about speaking is the perception that pair and group work must be central to the task-based classroom. Certainly, group work offers potential advantages (Long & Porter 1985) and supports Brown's (2007) assertions around the promotion of meaningful, authentic exchanges and interpersonal negotiation. However, over-application of this principle can lead to a learner-centredness that downplays other instructional and interactional patterns.

In practice, as Ellis (2009b) makes clear, pair and group work are not essential to TBLT. Individual and whole-class work are also important. Furthermore, the learner-centred implication of pair and group work is that, at least during task performance, teachers stand on the side-lines. Van den Branden (2009b: 284) asserts, however, that the teacher is ‘a crucial interactional partner in task-based language classrooms.’ As Shintani (2016) demonstrates, teacher intervention and feedback during the task phase of a lesson are valuable and productive elements. Teachers have a clear role to play in providing crucial feedback during task execution (Lyster 2004).

As Ellis (2009b) also makes clear, teachers do not necessarily understand where tasks fit into other aspects of programmes. Skehan (1996), for example, speaks of strong and weak realisations of TBLT. With strong TBLT the task is everything, and task execution is seen as ‘the necessary and sufficient condition of successful second language acquisition’ (Nunan 2004: 21). In a weak TBLT model (what some might label ‘task-supported’) tasks are still central, but students’ engagement in these tasks fits within a more structured approach which may be both preceded and followed by teacher-led moments and focused instruction. The weak TBLT framework provides scope for whole-class teacher input in both pre- and post-task phases of a task-focused lesson.

In summary, the understanding that the efficacy of TBLT is largely a consequence of interaction (whether understood from a cognitive or a sociocultural perspective) appears to influence a good deal of teachers’ thinking about tasks. This understanding can lead to the perception that TBLT is all about speaking, and downplays the importance and utilisation of a variety of tasks that exploit a range of skills and different ways of working.

5. Conclusion

Overall, it seems that some of the key theorising and research around task efficacy is beginning to get through to those teachers who have encountered, and are engaging with, TBLT ideas and concepts, in particular around more meaning- and fluency-focused work and promoting spoken communicative proficiency. Furthermore, teachers drawing on TBLT ideas report increased confidence, enjoyment and motivation among learners. Nevertheless, there can be misunderstanding about what a task is, leading either to lack of application of theory (calling something a task that is not a task, or, as Samuda 2005 puts it, detaskifying a task) or over-application of theory (assuming that tasks must always be about spoken interaction, or about pair and group work). There can also be misunderstanding about the purposes of tasks, once more leading to lack of application of theory (maintaining strong teacher-led grammar instruction elements) or over-application of theory (assuming that the teacher has no role to play, at least at the task execution stage). That theory and research into TBLT are both under- and over-applied in practice highlights the challenge of identifying exactly what TBLT is or should be, because significant differences emerge in the ways in which its proponents conceptualise the approach (Nunan 2004).

In turn, the fact that TBLT theory and research are at times under-applied and at other times over-applied reflects the reality that teachers are working in a wide range of diverse contexts. Some may wish to embrace TBLT for themselves, either as an overarching framework or with a view to dipping their toes into the water; others may find themselves in contexts where TBLT is being mandated from above or driven by assessment requirements; still others, by contrast, may be open to TBLT ideas, but are working in contexts where they are constrained by curricular or assessment mandates that undermine the TBLT endeavour.

What do the above assertions mean for the future of TBLT? At the start of this article I posed the questions whether TBLT needed to reframe itself, or whether teachers needed to be more receptive to innovation. I suggested that a middle path would be useful. That is, even though TBLT's underlying theoretical precepts do not, in my view, require substantial reconceptualisation, TBLT must be open to mediating itself in light of practice. In reality, teachers who, for whatever reason, are aiming to implement TBLT, require support with implementation in ways that pay equal respect to theory, research and practice (Markee 1997). I believe that three intersecting components will enable us to provide this: (1) challenging stakeholders (whether teachers, policy makers or assessment implementers) to consider the proposals of TBLT, and the findings of research into task efficacy, seriously; (2) acknowledging the genuineness of teacher perspectives on TBLT in practice; (3) articulating more clearly to stakeholders the conceptual broadness of TBLT. I would argue that TBLT is (or is becoming) sufficiently broad in scope that it is able to accommodate elements such as clearly planned and teacher-fronted components, including top-down grammar teaching, without compromising its essentially meaning-focused tenets (see, e.g., Ellis 2009b). This does not mean to say that traditional practices that clearly hinder the development of communicative proficiency are not to be challenged. The three components I suggest above provide a middle path that embraces challenge whilst recognising constraints.

As a final word, I was curious to find myself apparently ordained, in Hadley's (2013) thinking, to the office of priest in the new world order. I have to confess here that I do not think I am worthy of that office, at least not in an orthodox way. I can see all too clearly the problems that can beset TBLT as teachers seek to work out its tenets in real-world classrooms. In practice, this means that I will continue to evaluate the theoretically and empirically-led claims to the efficacy of TBLT against the real experiences of teachers, just as I will continue to challenge teachers’ limiting beliefs through my work as a teacher educator. I hope these commitments will keep me sufficiently grounded, but in ways that enable the TBLT endeavour to be enacted more successfully in time-limited instructional contexts.