Issues in Teachers Reinterpretation of a Task-Based Innovation in Primary Schools

DAVID CARLESS
Hong Kong Institute of Education Hong Kong SAR, China

Although task-based teaching is frequently practiced in contemporary English language teaching, it is underresearched in state school settings. This article contributes to lling this gap in the literature by using qualitative case study data to explore how a task-based innovation was implemented in three primary school classrooms in Hong Kong. Analysis of classroom observation and interview data shows how the case study teachers reinterpreted a new curriculum in line with their own beliefs and the practical challenges occurring in their school contexts. Drawing on classroom episodes, the article highlights three issues that proved problematic when the tasks were implemented: use of the mother tongue, classroom management or discipline problems, and the quantity of target language produced. Implications for the design and implementation of task-based pedagogies in primary school contexts are discussed.

uch of the research into task-based teaching has focused on adult classes in ESL contexts. As Candlin (2001) observes, however, little empirical research has been conducted on task-based teaching in school EFL1 contexts. This article seeks to ll that gap by using qualitative case study data to analyse how a task-based innovation was implemented in the Hong Kong primary school context. This cross-curricular innovation, the so-called target-oriented curriculum (TOC),2 was a form of outcomes-based education in which students progressed toward specied learning targets by carrying out tasks (Carless, 1997; Clark, Scarino, &

1 Postcolonial Hong Kong has characteristics of both a second and a foreign language context, but the distinction is not central to this article. 2 TOC has now been superseded by a reform called Learning to Learn (Curriculum Development Council, 2001). For studying English, task-based approaches still form a prominent component of the new syllabus, although the term TOC is no longer used.

TESOL QUARTERLY Vol. 38, No. 4, Winter 2004

639

Brownell, 1994; Morris, 2000). TOC was based on constructivist learning principles and argued that pupils needed to be involved in developing their own learning. Task-based teaching and learning were central to this philosophy. For studying English, task-based approaches, referred to as task-based learning, are also a key component of recent syllabi (Curriculum Development Council, 1997, 2002) that replaced the communicative syllabi, which were adopted in name but not widely implemented in the classroom (Evans, 1996). The task-based syllabus was seen as enhancing the communicative one, which was largely unsuccessful because traditional textbooks did not complement the communicative approaches, a large number of teachers were either untrained or not trained to teach the subject, and coursework focused on passing examinations that did not support communicative approaches (Evans, 1996; Lai, 1993). Both task-based and communicative orientations toward teaching contrasted with earlier practice that was traditional and teacher centred, and often included a decontextualised focus on grammatical form, or what the Education Commission (1994) describes as a concentration on the formal features of the language at the expense of encouraging students to use the language (p. 25). To begin to analyse an innovations success, it is necessary to learn how teachers are carrying it out in classrooms at schools where the innovative curriculum is supposedly being implemented. Curriculum mandates frequently do no match classroom practices (e.g., Fullan, 1999; Markee, 1997). Moreover, although task-based teaching is very much in vogue, it is underresearched in young learner classrooms. I therefore discuss the notion of task and indicate how the Hong Kong primary school context has interpreted it. Then I describe the research context and outline the qualitative research methods used to discover how teachers reinterpreted the task-based innovation. I present three critical classroom episodes and focus on three problematic issues arising from teachers reinterpretations: using the mother tongue (MT), dealing with discipline challenges, and producing the target language (TL). I also analyse the extent to which task-based reasonably describes what was going on in the classrooms under discussion and sketch some issues for further research.

NOTION OF TASK
The denition of task has received much attention in the literature (e.g., Breen, 1987; Bygate, 2000; Bygate, Skehan, & Swain, 2001; Candlin, 1987; Ellis, 2000, 2003; Skehan, 1998, 2003). Skehan (1998) denes tasks using four criteria: Meaning is primary, it works toward a goal, it is
640 TESOL QUARTERLY

outcome-evaluated, and it is related to the world outside the classroom. Skehans denition is well accepted for the ESL adult context. Turning toward task-based teaching with young learners, Cameron (2001) summarises tasks key features as follows: They have coherence and unity for learners (from topic, activity and/or outcome), meaning and purpose for learners, clear language-learning goals, a beginning and an end, and involve the learners actively. Because young learner contexts have limited English language resources, they are more likely than speakers who have had greater TL exposure to be involved in what Skehan (1996) refers to as weak forms of task-based teaching (p. 39). In a weak approach, tasks are roughly comparable to the production stage of a presentation-practice-production method. TOC denes task using ve main elements (Clark, Scarino, & Brownell, 1994; Curriculum Development Council, 1997): a purpose or underlying real-life justication for doing the task that involves more than simply displaying knowledge or practicing skills a context in which the task takes place that may be real, simulated, or imaginary a process of thinking and doing a product or the result of thinking and doing a framework of knowledge and skills

TOC distinguished tasks from exercises, which were dened as learning activities that help acquisition of specic information and skills (Education Department, 1994, p. 19). Exercises were intended as language input during the pretask stages of task-based teaching (Curriculum Development Council, 1997). Ellis (2003) points out, however, that distinguishing exercises from tasks is not easy because some activities may contain features of both. The three denitions of tasks just outlined have a great deal in common and some differences. For example, both the Skehan and TOC denitions highlight real-world aspects (context in TOC), while Cameron does not mention this aspect explicitly. Some of the features, for example, language goals, process and product (TOC), or beginning and end (Cameron, 2001), are useful for dening tasks, but they are not exclusive to task-based teaching (Widdowson, 1998). I return to the issue of denitions in the concluding section.

IMPLEMENTATION ISSUES IN SCHOOL SETTINGS

As a development within communicative approaches (Littlewood, 2004; Nunan, 1999), implementing task-based syllabi entails some familiar
ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION 641

issues: using MT during pair- or group-work activities, dealing with classroom management challenges, and issues in producing TL. Although I deal with the themes separately, they obviously overlap when, for example, pupils overuse MT during a task, thereby challenging classroom management and reducing TL production.

Mother Tongue Use

If one of the main purposes of task-based teaching is to stimulate communication in the TL, one might think that using the MT was a barrier, what Prabhu (1987) referred to as a last resort (p. 59). More recently, however, a more complex picture has emerged of the tension between using MT and using TL. Carless and Gordon (1997) used data reported by a group of inservice Hong Kong primary school teachers experimenting with taskbased teaching. The teachers identied pupils use of Cantonese as the most prominent difculty that occurred during tasks because this practice conicted with the teachers espoused goal of learners using English. Nikolov (1999), researching young learners in Hungary in an opportunistic sample of 111 classes, found that during pair or group work, students used the MT more frequently than English, although she claims that most of the childrens discourse was on-task. She concludes that she was shocked by how infrequently children in her corpus heard or used English during pair or group work. Pinter (1999), also researching young Hungarian learners, in this case 10 pairs of high achieving 10year-olds, found that students frequently used MT, in particular when working out what to do prior to commencing the task and when they did not know words or expressions in TL. In Turkish secondary schools, Eldridge (1996) found that most code switching was purposeful and related to pedagogical goals, and that higher and lower achievers code switched with about the same frequency. Similar to Nikolov (1999), Swain and Lapkin (2000) were surprised that learners used MT more frequently during communicative tasks than they had anticipated. In a study of French immersion in secondary school, they found that students used MT in around 25% of the turns, but that only around 12% of the turns were off-task. Although Swain and Lapkin noted that teachers tended to view this phenomenon negatively, deeper analysis of transcripts led them to conclude that learners used the rst language for important cognitive and social functions, for example, discussing strategies for how to carry out the task, and that using the MT judiciously could help them to acquire TL vocabulary and syntax.

642

TESOL QUARTERLY

Classroom Management
For teachers dealing with large classes of children, handling noise or indiscipline is a central issue, particularly in those contexts where teacher-fronted activities are the norm and pair or group activities are less universally practised. Tensions arise when facilitative teacher roles in task-based teaching conict with prevailing cultural norms that indicate a more authoritative teacher persona (Carless, 1999; Ellis, 1996; Kramsch & Sullivan, 1996). For example, case studies involving extended classroom observation (Tsui, 2003) illustrate how important discipline is to language teachers in Hong Kong secondary schools. One of the teachers Tsui observed, for instance, had difculty distinguishing between noise that foreshadowed disciplinary problems and noise that indicated high levels of involvement. Willis (1996) claims that the larger the class, the more logical taskbased teaching is for language practice, noting that she has observed successful pair work in classes with as many as 80 students. She also points out that noise levels may often rise because teachers have allotted too much time to a task and advises that allowing all students time to nish is sometimes not feasible. Li (1998), in a discussion based on teacher self-reported data about implementing communicative approaches in South Korea, argues that large classes accentuate noise and indiscipline during group activities. As one respondent in her study observed, when everyone starts to talk, the class can be very noisy. . . . Some kids like to play around during group-work time (p. 692). Similarly, Cameron (2001) notes that teachers must skillfully manage group and pair work to make activities well structured and motivating rather than a noisy time for some and a time to opt out for others. Tomlinson (1988) argues, and this may still remain valid, that communicative approaches have had little impact in state schools because they encourage activity, noise, initiative, and disorder, which usually conict with the normal ethos of schooling.

Target Language Production

The considerable discussion in the literature concerning the quantity and quality of language that students produce during tasks is based primarily on adult ESL learners. School foreign language contexts have received relatively little attention, although notable exceptions are Kowal and Swain (1994), and Swain and Lapkin (2000, 2001). In his critique of task-based interaction, Seedhouse (1999) points out that students often focus on completing the task and that, as a result, they sometimes

ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION

643

produce only the modest linguistic output necessary to complete it. Seedhouse suggests that tasks might actually encourage students to communicate at a level beneath their full potential. An extreme example of lack of language production occurred during the piloting for the current study, when I observed a lesson in which the teacher set up a group activity that required pupils to complete a survey of their group members using the question, How do you come to school? To my surprise, the students completed the task in silence because, as I later discovered, the learners already knew how their friends came to school (mainly on foot) and saw no need to use spoken language to complete the task. The problem seemed to lie partly in the task design, which neglected communicative intent, but also in the students seemingly being unaware that the teacher wanted them not only to complete the task, but also to practise English. A possible solution lies in Lin and Luks (2002) analysis of how a Hong Kong teacher (Tracy) found ways to improve student engagement and participation by making explicit to them the rationale and learning objectives behind the tasks she asked them to do. Task-based learning can also hinder language production because it requires young learners to make or do something. To reduce the likelihood of TL use grinding to a halt under such circumstances, Cameron (2001) suggests a number of strategies; for example, pupils could decide what materials they need for their making and doing and request them from the teacher, the teacher can participate by moving around the classroom and reinforcing key language items, or students can recite or listen to rhymes or songs in the TL to give them background exposure. In summary, past research reports on the extent of MT use, tensions between involving students in tasks while maintaining an orderly classroom, and organizing tasks to maximise students linguistic participation. A number of the studies, however, relied solely on teachers selfreported data, and others involved only limited periods of classroom observation. These limitations in the ndings detail and depth underscore the need for the current studys methodology, which employed sustained classroom observation triangulated with teacher interview data. The research questions guiding the study are (a) What are the teachers attitudes and understandings toward task-based teaching? (b) How are the teachers attempting to implement task-based teaching and what issues emerge from these attempts?

644

TESOL QUARTERLY

METHOD
The data presented in this article are part of a larger study that investigated how three Cantonese native-speaking English teachers in different primary school settings implemented the task-based innovation TOC in their classrooms. I used case study as an investigative technique because it enabled me to probe deeply into the teachers viewpoints and actions, thereby helping me to understand the implementation from their perspectives. Pupils perspectives on task-based learning, although immensely valuable, were not a focus of the current study. Case study data facilitate working hypotheses (Lincoln & Guba, 2000) or fuzzy generalisations (Bassey, 2001, p. 5) that can be transferred to other contexts. I selected the informants for the study based on my perceptions of the following relevant attributes: young and capable teachers in their late 20s or early 30s, open-minded in responding to change and therefore willing to engage with the innovation, condent enough in their teaching to be observed in the classroom, professionally motivated to take part in the study, cognisant of the studys demands and willing to participate on that basis, and comfortable interacting in English, the medium of the interviews. Data for the study were collected using classroom observation, focused interviews, and an attitude scale. Each classroom was observed for ve or six consecutive English lessons in three observation cycles during the school year. Seventeen lessons were observed for each teacher, making a total of 51 lesson observations. This method permitted me to observe each classroom over an extended period and to observe lessons on different topics, which enabled me to develop a rounded impression of what was going on in the classrooms. It also helped to mitigate, as far as possible, observer paradox problems or atypical, one-off display lessons. All lessons were audiotaped on a single tape recorder, which was sufcient because it was not intended to record the specic discourse of pairs or groups of pupils. Instead, I circulated around the classroom, listened in to what pupils were doing and recorded observations in my eld notes. This procedure limits what can be said about the kind of language used during tasks. I focused on collecting qualitative data using eld notes and lesson transcriptions, although I did collect quantitative data using a tailor-made classroom observation schedule (not discussed in this article). I conducted a series of six semistructured interviews, lasting between 40 and 60 minutes, with each of the three teachers, totalling around 15 hours of interview data, which I transcribed verbatim to deeply immerse myself in the data. The arrangement and focus of the interviews is outlined in Table 1.
ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION 645

TABLE 1 Arrangement and Focus of Interviews Interview Baseline interview Three postobservation interviews Summative interview Postanalysis interview Focus Collect relevant background information about teacher and school Seek views on relevant issues emerging from the lessons just observed Probe the main issues arising from the ongoing data analysis Seek respondent validation or disconrmation of the data analysis

Note. A 5-point Likert attitude scale was developed to measure the respondents orientation to English language teaching and TOC. An overall orientation toward TOC was computed for the three teachers and a wider sample of primary school English teachers. The attitude scale is not a focus of this article, but I refer to it briey when discussing the teachers orientations toward task-based teaching.

I began analysing the qualitative data when hunches emerged during the data collection process and later more systematically analysed the data using reduction and synthesis. Data from the 51 lesson observations was reduced using focused summaries pertaining to the research questions and other emerging issues. Next, codes were assigned to the interview transcripts and the classroom observation summaries. From these codes a number of themes were identied; for example, all three teachers mentioned discipline tensions when carrying out task-based teaching. Once such a theme was identied, all data touching on it from the different research tools was pooled and further analysed. Pooling all data on a given theme helped me to identify evidence that might support or disconrm a particular line of argument. Making sense of the data through an organised construction (Holliday, 2002, p. 100), I moved repeatedly from data to emergent ndings and compared informants interpretations with their classroom actions. To enhance trustworthiness, I asked the respondents and other associates both in Hong Kong and overseas to give feedback on the data analysis and the emerging propositions. The drafting and redrafting process of writing was itself also a method of inquiry, discovery, and analysis (Richardson, 2000). In view of the unavoidable subjectivity in qualitative (and other forms of) research (Peshkin, 1988), I endeavour to distinguish the primary data from my own interpretation so that readers can judge the arguments for themselves, mindful that my constructions are among the many possible interpretations of reality (Holliday, 2002).
646 TESOL QUARTERLY

THE CONTEXT AND TEACHERS

Primary schooling in Hong Kong, where the study took place, starts at age 6 and lasts for 6 years. Chinese, mathematics, and English, the three main primary school subjects, dominate the curriculum. Teaching is generally traditional and textbook oriented, and innovative pedagogies, often derived from Western models, are usually viewed as unnecessary or impractical (Adamson & Morris, 1998). English is the sole foreign language in the Hong Kong primary school curriculum, and it is usually taught for between seven and nine 35-minute lessons per week (Carless & Wong, 2000). It is the only primary school subject taught using English, although the extent of TL use can vary according to the teachers and pupils English language abilities (Lin, 1990, 2000; Pennington, 1997). Concerns about perceived declining English language standards among teachers and students have led to controversial benchmark tests of teacher language prociency (Glenwright, 2002).

Teachers Experience
Table 2 introduces the teachers who participated in the study. Alice (all names are pseudonyms) is a trained and experienced teacher in a middle management position in her school, responsible for developing TOC for the subject of English. She describes one of TOCs aims as to let pupils participate actively, to use activities to arouse their interest. . . . To make learning more like real life, to involve them in learning by doing (Month 3). She describes task as follows: Task is an activity; in the task, pupils should have the chance to use the language meaningfully, to interact with each other, not just read after the teachers or repeat something. The most important thing is to let them use the language (Month 3). She seems to have developed her own personal understanding of task-based teaching and has identied a number of task features (active participation, real-life relationship, learning by doing, putting language and meaning into use). Betty was in her third year of teaching, and because she had only just begun formal training, she had not yet been exposed to methodologies such as communicative approaches. She says that one of her main roles as a teacher is to provide input to the pupils, such as teaching vocabulary and grammar. She also believes in the need for student discipline: The discipline should be settled before the lesson starts. . . . If the discipline is lost then I think the lesson cannot be continued (Month 4). This emphasis on traditional input and discipline prompted some hesitations about carrying out activities or tasks that might require her to relinquish disciplinary control. Turning to her perspective on
ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION 647

TABLE 2 Teachers Qualications and Teaching Specications Teaching experience Standard of pupils Above average

Teacher Alice

Qualications/training a

Class taught P1 class (Year 1), 33 pupils, aged 67 P2 class (Year 2), 31 pupils, aged 78 P1 class (Year 1), 26 pupils, aged 67

Teaching certicate (English); 8 years completed 16 week full-time inservice course on communicative approaches BA (music) untrained; undergoing part-time in-service training Teaching certicate (English); B.Ed (UK); started M.Ed 2 years

Betty

Average

Connie

5 years

Slightly below average

The teaching profession in Hong Kong does not yet exhibit a high degree of professionalism in training and qualications. According to the Education and Manpower Bureau (2002), at the primary school level, 52.4% of teachers do not hold degrees and 9% are untrained, and many subjects are taught by teachers not trained in that particular subject: 41% of English teachers are not trained to teach English.

tasks, her stated denition is that task mainly has objectives and it can link the pupil ability of understanding, conceptualising, that kind of communication (Month 4). In my view, this denition is vague and does not distinguish tasks from other approaches that would also contain objectives. In none of the interviews was she able to clearly dene tasks nor to distinguish between a task and what under TOC would be regarded as an exercise. Bettys preliminary understanding was consistent with her recent entry into the teaching profession and lack of training, which meant she did not have the background in communicative approaches shared by the other teachers. Connie is well qualied academically, has been trained in the theory and practice of communicative approaches, and holds the position of panel chair for English (similar to head of department). Her perspective on task-based teaching is illustrated by the following quotation:
I think we should try to motivate the pupils, try to increase their interest in learning. I think putting the knowledge in use is quite important in TOC. I think it should be more lively, not just a classroom situation, not just learn this but know that it is useful and they can use it. (Month 5)

I interpret Connie as having developed a sound personal understanding and interpretation of some of the innovations key elements. Although, like Alice, Connie did not use TOC terminology directly, she touched on some relevant elements, for example, knowledge in use (cf. purpose in TOC), not just a classroom situation (cf. context in TOC).
648 TESOL QUARTERLY

Teachers Knowledge and Attitudes

In addition to their understandings, the interview data and the attitude scale also gauged the teachers attitudes toward task-based teaching. The attitude scale data showed that of the three teachers, Connie was the most positively oriented (very positive) toward task-based approaches, Betty the least positively oriented (neutral stance), and Alice somewhere in between (quite positive).

FINDINGS
Classroom episodes were chosen to illustrate the key issues already identied: using MT, managing the classroom, and producing TL in the lessons. Additional interview and classroom data provide the teachers perspectives on those issues.

Classroom Episodes
The following classroom episodes present critical incidents (Tripp, 1993) that illustrate the main challenges for implementing task-based teaching in the classrooms under discussion. The episodes were selected because they contained lessons from which challenges identied by the teachers emerged most prominently.

Episode A
This episode was taken from a double lesson lasting 70 minutes during the second cycle of observing Alices classroom. It illustrates discipline problems, widespread use of MT, and lack of English language production. Alice took some plates of fruits to the classroom and invited selected students to carry out an open pair work dialogue in front of the class with the following main components:
Smell this. Is it nice? Yes it is / No it isnt. What is it? It is a [name of fruit].

To create an information gap, the student smelling the fruit was blindfolded. While the students were carrying out this open pair work, the teacher told them that they were too noisy and delivered a number of disciplinary warnings and exhortations both in English and in MT. In an
ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION 649

attempt to maintain order, one strategy she used was, If you are noisy, I wont choose you [to come and smell the fruit]. Four pairs of pupils were chosen to demonstrate the activity at the front of the class, then pupils worked in six groups (ve or six students per group). Three groups carried out the task with the plates of fruit, while three groups did a written workbook exercise, and after 8 minutes they transposed their roles. With respect to implementing this task, I contemporaneously recorded the following comment in my eld notes:
The activity provokes a lot of noise and excitement, particularly from the boys. Somewhat disappointingly, it generates hardly any English3 at all from the groups near to me. Some of the other groups appear to be using a bit more English, but the overall level of noise makes it difcult to hear what is being said. Somewhat surprisingly the three groups chosen to do the workbook exercise rst, do it willingly and show no resentment at being forced to wait their turn for the fun part. Perhaps, they quite like getting down to some solid written work. (Field notes, Book 2, p. 29)

In the postlesson interview, Alice identied as a task4 interviewing their friends to ask about smells and other feelings. She expressed satisfaction with the lesson: I like the part of smelling something and asking their friends is it nice and what is it? She also highlighted two issues related to this lesson. First, she stated that she was unhappy about the extent of noise generated but wished to provide the opportunity for pupils to put language into use. With respect to this issue she stated, I am quite liberal. I can accept the noise during activities more easily than most of my colleagues. They think I am rather Westernised, less traditionally Chinese. Second, regarding pupils use of MT she said, I dont like them using Cantonese and encourage them to use English, but some are just lazy and will prefer to use their mother tongue (postobservation interview, Month 7).

Episode B
This episode was taken from the third cycle of observing Bettys classroom. It shows the teachers concerns about discipline problems and pupils use of MT. The language focus for this lesson was responding to the questions:
3 My eld notes indicated that noise and excitement were frequently correlated with using MT, irrespective of the pupils English language prociency. 4 The extent to which the tasks in these episodes meet the TOC denition could be critiqued, for example, in terms of purposefulness or contextualization (see Littlewood, 1993). I return to this issue in the conclusion.

650

TESOL QUARTERLY

Where is s/he? Where are they? What is s/he doing? What are they doing?

The learners practised these structures using a text from the course book about a family going to the beach. Later in the same lesson, the teacher showed some authentic photographs of various people carrying out activities, for example, Betty playing the piano in church, some pupils and teachers having a picnic, and a group of teachers having a meal in a restaurant. After some teacher-fronted, whole class discussion of one of the photos, the teacher distributed more photos so that the pupils could try in groups to answer the four designated questions for the photos they had been given. My eld notes record that the pupils were stimulated by the authentic photos and became quite animated, so not surprisingly, in addition to using English, a number of pupils commented in Cantonese. I thought this activity succeeded because pupils were motivated, involved in the activity, and challenged to generate their own English interpretations of the photos. Betty perceived that using the given photos to answer their partners question was a task. She noted two contrasting aspects of the pupils response to this activity. On the one hand, she was pleased that the photos catch the students attention and they love to see the pictures, but, on the other hand, she interpreted the lively response somewhat negatively in her comment that discipline problems were caused when the teacher showed the photos. Sometimes doing the group work is difcult because they just use the opportunity to talk in Cantonese (postobservation interview, Month 7).

Episode C
This episode is taken from the second cycle of observing Connies classroom in a double lesson lasting 70 minutes. It illustrates the uneven distribution of turns in the oral part and the lack of language production during the time-consuming drawing activity. The language focus was
What can you see? I can see a ______.

The students then carried out the following task. The class divided into groups of ve or six pupils, and the group leaders looked out of the classroom window while their classmates remained in their normal seats. The group leaders then rejoined their groups and reported what they had observed, in response to the groups repeatedly asking, What can you see? After the students had completed this oral activity, they made a drawing of what the group leaders had seen from the window and labelled it with an English caption.
ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION 651

My eld notes identied the activity as successfully involving students in language use, although for the oral part, I noted that only certain individuals used English; the group leaders were active but the other pupils were restricted to repeating the single utterance, What can you see?, and in practice some of them did not contribute any English speech at all. Although I was aware that learning might be taking place without students actually saying anything, the extent of pupil involvement in the task did seem to me to run counter to the teachers stated aims of language practice and communication. For the drawing part of the activity, which lasted for a lengthy 32 minutes, my eld notes identied a variety of pupil responses:
A few of the pupils are discussing (mainly in English) with the teacher, one or two are discussing what they can see with me in English, some others are writing in English what they can see. The majority are still drawing/colouring or chatting with their neighbours in Cantonese. (Field notes, Book 2, p. 106)

Connie stated that describing the things which could be seen from the window was a task. She expressed satisfaction with the oral part: Some of the pupils can really ask and answer the questions during the activity. She expressed some reservations about the drawing activity: They like doing the drawing and colouring, but it takes so much time and I wonder if they are really learning anything (postobservation interview, Month 9).

Interviews and Additional Classroom Data

Teachers perspectives help to illuminate the three issues under analysis. The teachers held differing views on the value of using MT in the classroom, but they all noted the tensions they felt between their desire to carry out activities that engaged the learners interest and to maintain an orderly classroom. All teachers saw tasks as a means of getting students to use English and were frustrated when they felt that not all students were beneting from the opportunity.

Use of MT
In their postanalysis interviews, all three teachers expressed views on how often pupils used MT during tasks, both as a general phenomenon and specically in relation to the episodes already described. Alice was very much against pupils using MT because she believed that her high ability pupils should be able to cope with tasks in English. The other two teachers, with lower ability pupils, stated that pupils could use MT under certain circumstances, such as when they were working out what to do
652 TESOL QUARTERLY

(cf. Pinter, 1999). In their open-ended interview responses, the three teachers noted a number of reasons pupils use MT. Alice stated that some are just lazy, but others do have difculties in using English (Month 16); Betty believed that they use Cantonese because they nd it facilitates the activities (Month 14); Connie argued that it depends on the activity type. If the rules or instructions are too complicated, they will use Cantonese to solve their problems and to make themselves clear (Month 15). Overall, based on their interview responses, I categorised the teachers views on recourse to MT as follows: Because of limited English prociency (Alice, Betty, Connie) To facilitate the activity (Alice, Betty, Connie) Because the activity was too complicated (Betty, Connie) Expressing their feelings, arguing or complaining (Connie) Laziness (Alice) or lack of initiative (Betty) Although all the teachers tended to discourage pupils from using MT, they did note some constructive uses for it. For example, Alice noted that more able students would use Cantonese to teach the less able ones through, for example, explanations, reminders, or corrections. Connie suggested that pupils use MT to reduce anxiety, particularly Primary 1 pupils, whom she perceives as being quite nervous. She also suggests that the less able students could ask the teacher questions in Cantonese, and the teacher would then try to respond in English. Connie was the only teacher who used a systematic reward system to encourage TL use (and good behaviour) during activities. The teachers own models of TL/MT use in the classroom also seemed to inuence the students. Alice, with high ability pupils, used English for instruction but not for disciplining students; Betty frequently used MT for explanations, as a social lubricant, and for humorous purposes; Connie believes strongly that English should be taught using English, and even though her pupils were of quite low ability, she rarely used MT during whole-class interaction. Overall, the teachers provided no evidence that they had explicitly discussed with pupils when they might usefully employ MT, and the teachers general strategy appeared to be gently discouraging MT use and occasionally turning a blind eye or being preoccupied with other classroom matters.

Discipline Challenges
In the discussion of Episodes A and B, I noted tensions between the desire to carry out activities and a wish to maintain a quiet, orderly classroom. All three teachers referred to such dilemmas in the interviews. For Alice, a recurrent theme throughout the 17 lesson observations
ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION 653

was difculties in controlling the classroom discipline, particularly during activities. In all three postobservation interviews, she expressed dissatisfaction with students behaviour, especially those naughty boys. She did, however, reiterate her belief in motivating pupils to put language into use and commented on the tensions: There is a bit of conict between using the language and discipline. Its unavoidable having noise when they are talking in groups; some teachers and principals may not accept that (summative interview, Month 10). Tensions between activities and discipline also occurred in Bettys lessons, albeit from a contrasting perspective. Alice emphasised activities somewhat at the expense of discipline, while Betty stressed discipline somewhat at the expense of activities. Betty also referred to the difculty of monitoring performance during activities, observing that it is hard for the teacher to monitor the students, especially as there are many pupils in the class. It is hard to tell whether they actually do the activity or not, especially for pair work or group work (postobservation interview, Month 9). In the interview data, Betty repeatedly referred to a need to focus on discipline, when to my eyes the class seemed very well behaved. Bettys doubts about providing less controlled activities that allow pupils more freedom seem to resonate with the Korean teachers perceptions that Li (1998) discusses. Connie expressed satisfaction that pupils could behave acceptably but also involve themselves actively in the lessons. I always believe that students should be quiet in class, but in carrying out activities you have to tolerate some noise (postanalysis interview, Month 15). She noted some conict between the cultural desire for a quiet classroom and the necessity to tolerate some noise. In fact, she highlighted her belief that the teachers tolerance for noise during activities is important and specically identied developing this tolerance as something that had improved her teaching.

TL Production
Task-based teachings main aim, according to these teachers, is to require pupils to use English. Tensions arose when students did not meet these expectations to the teachers satisfaction. Alice was satised when pupils were communicating through English. Under two circumstances, however, she saw limitations in language production. In some tasks, only certain individuals used English; in Episode A, for example, only eight pupils did the open pair work, although more would have liked to do so. On other occasions, the nature of tasks limited language production, as when pupils constructed a zoo from cardboard cutouts, for example. Alice commented on this task: Sometimes the activities are quite time654 TESOL QUARTERLY

consuming without much English language learning, but they do learn other things, some concepts like which animals go together, and also they learn to cooperate in problem solving (postobservation interview, Month 10). Betty observed that during stages of some tasks, not all pupils could participate, and if the activity was motivating, the less involved pupils could nd it frustrating: Sometimes it is difcult for everyone to participate in an activity since it is basically done by one or two students in the group (postobservation interview, Month 9). Betty also identied a task asking students to combine watercolours as being interesting and enjoyable but not leading to much TL production. She also reected on whether she sometimes discouraged language production: Often I dominate the activities. I speak too much and could do more to get the pupils to speak (postobservation interview, Month 9). Connie noted that through TOC tasks, pupils had developed their oral skills: In TOC approach, they are already accustomed to talking a lot in English in class and this helps improve their results in oral examinations (postobservation interview, Month 10). She did, however, identify tasks that limited language production, such as the task discussed in Episode C, where only group leaders were producing language. Table 3 summarises the main teacher orientations toward the implementation issues; where applicable, indicates a largely positive process or outcome and indicates a negative one.

TABLE 3 Summary of Teacher Views on Implementation Issues Teacher Alice MT use Facilitates peer support; facilitates activities. Denotes laziness. Facilitates activities; social lubricant. Denotes lack of initiative. Facilitates activities; reduces anxiety. Used for arguing or complaining. Discipline Willing to tolerate some indiscipline. Language production Some communication through English. Some tasks time-consuming and not producing much language. Beginning to reect on the issue. Some tasks dominated by teacher or a few students. Notes improvement in oral skills. Some domination by individuals or other lack of language production.

Betty

Desires to maintain tight control.

Connie

Seeks balance between good behaviour and noisy activities.

ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION

655

DISCUSSION
The ndings on these three issues need to be considered in view of broader professional perspectives. With respect to MT use, my interpretation of the data tentatively supports Swain and Lapkins (2000) conclusion that MT use is a complex issue. Allowing students to use MT during tasks may affect their conduct positively (e.g., working out how to do the task) or negatively (e.g., irrelevant chitchat). Swain and Lapkin duly recommend that teachers neither encourage nor discourage pupils from using MT. The three teachers have adopted a similar approach, making essentially pragmatic judgements about when they would tolerate or gently discourage students from using MT. To go beyond ad hoc decision making in the young learner classroom, however, these issues merit further exploration: What kinds of MT dialogue generated by young learners might facilitate second language development (Swain & Lapkin, 2000, 2001)? What guidance might teachers provide to pupils on the extent and desirability of using MT during tasks (Eldridge, 1996)? What tasks might involve, for example, language analysis in which the teacher might recommend that students use MT? Although these questions need to be investigated, frictions clearly remain between viewpoints that see MT use as a barrier to developing TL and those that see it as a potential support. The second issue, classroom management and discipline, has been underexplored in the literature on task-based teaching, but it is important for many teachers, particularly those who teach in more traditional contexts. The study showed tensions between maintaining a quiet, orderly classroom, according to local school cultures, and a wish to carry out oral English tasks. I interpreted the data as showing that concerns over noise and discipline inhibited implementing task-based teaching. In traditional contexts, teachers often handle teacher-fronted instruction better than they handle communicative tasks (Li, 1998), a challenge that further exacerbates the difculties of implementing task-based teaching. This issue may become increasingly important as English language teaching in primary school contexts expands in China, Japan, South Korea, and Taiwan (Ho & Wong, 2003). In such contexts, teachers should learn to tolerate constructive noise, as Connie hinted. For example, Tsui (2003) shows how an expert Hong Kong secondary school teacher deals with discipline by distinguishing between off-task noise (not permitted) and on-task noise (permitted). In this case, the distinction facilitates achieving instructional objectives rather than simply reinforcing the teachers authority. Good teaching is characterised not by establishing routines, such as discipline, but by possessing the judgement that informs executing the routines with some exibility (Tripp, 1994; Tsui, 2003). Senior teachers and school princi656 TESOL QUARTERLY

pals can support teachers by reassuring them that a noisy classroom will not, in itself, be viewed unfavourably. Another factor that may prompt discipline problems is how students perceive activities or tasks. Students, I suggest, may view activities as a lull from serious instruction and an opportunity to take a break. Barkhuizens (1998) study of South African secondary school language students revealed that they preferred mechanical learning activities over communicative-type activities, a nding that surprised their teachers. This nding recalls Episode A, when groups of pupils willingly carried out a traditional workbook exercise while their classmates did the fruit-tasting task. Are certain types of learner more disposed toward learning through tasks or group work, while other learners are more oriented toward traditional classrooms (Barkhuizen, 1998)? Indeed, a traditional educational context may precondition pupils somewhat to resist task-based approaches. How might teachers resolve the tensions between societal or cultural expectations of classroom roles and the need to involve pupils actively in language use (Carless, 1999; Kramsch & Sullivan, 1996; Tsui, 2003)? All three teachers grappled with this issue, and, to my mind, only Connie resolved it satisfactorily by tolerating some noise while maintaining vigilance toward pupils drifting off-task. The third issue relates to TL production. All three teachers saw taskbased teaching as providing the opportunity for pupils to practise using English, and they expressed concerns over activities that enabled only a few students to use language. In my view, tasks in which some pupils may remain silent have limitations, although as Breen (2001) points out, nonparticipating learners can learn from others participation. Another issue affecting TL production was spending too much time on nonlinguistic activities, such as making things, colouring, or drawing. Although a certain amount of drawing or colouring is generally accepted as good primary practice (Rixon, 1991), tasks must be carefully designed and implemented so that nonlinguistic elements such as drawing do not outweigh English language aspects. The studys research design limits more detailed analysis of the language produced. I would reafrm Seedhouses (1999) call for more transcripts of task-based interaction, particularly those focused on young learners. Issues for further exploration include the following: How can teachers help young learners to optimise their language production when completing tasks (Seedhouse, 1999)? If some learners are not actively producing TL, to what extent is that a problem? How can teachers maximise language use in activities that primarily involve drawing, colouring, or making something (Cameron, 2001)? Teachers might have different expectations for language production from young learners than from more mature learners.

ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION

657

CONCLUSION
Through TOC and recent English syllabi (Curriculum Development Council, 1997, 2002), the Hong Kong government has put in place approaches that it refers to variously as task-based (1997, p. 50) or task-based learning (1997, p. 49; 2002, p. 95). Using extensive eldwork and interviews, this article has shown how teachers have ltered and interpreted the innovation, with particular emphasis on three issues that challenged implementing it. The teachers ltering process included reconciling task-based approaches with their own understandings of tasks, their young pupils limited language resources, which made reversion to MT or lack of TL production difcult to avoid, and their belief that discipline or order are important, which may deter non teacher-fronted teaching. I characterise the teachers perspectives on tasks as follows: A task is an activity that involves using language, communicating within a situation that is authentic rather than just based in the classroom. But certain activities challenge implementing tasks by causing students to become noisy, usually speaking in MT, meaning that students may produce little TL. This off-task noise causes some tensions for the teachers and their colleagues, particularly because their traditional school context does not expect noise. One of the issues foregrounded at the outset was the nature of task. Given the weakly developed interlanguage systems of 67-year-olds, the young learners in the classrooms under discussion were given principally language practice activities. If English language structures are not pretaught, then beginning learners will probably not have sufcient English language to use during tasks and may have no alternative other than to complete the task using MT. Following from this observation, I suggest that, for the Hong Kong context, the weaker form of task-based teaching that the three teachers practised is more feasible than the stronger task-based learning, where the syllabus would focus on the tasks themselves, and the TL would emerge from the tasks. In the more traditional Hong Kong context, the teachers had difculty operationalising the concept of task in TOC. A large-scale survey of TOC provides wider evidence on implementing task-based approaches in the Hong Kong primary context that crossvalidates the cases discussed in this article. Using questionnaire surveys, interviews, and classroom observations, Clark, Lo, Hui, Kam, Carless, and Wong (1999) found that teachers had difculty interpreting and

658

TESOL QUARTERLY

implementing tasks. Questionnaire data from their study indicated that a slight majority of teachers favoured traditional as opposed to task-based teaching; interview data revealed widespread misunderstandings about the nature of TOC tasks; and classroom observations indicated that teachers rarely implemented activities that met the TOC denition of task. The task-based approach in the Hong Kong context is, I believe, incongruent with generally accepted understandings of task and more closely conforms to what readers in other contexts may refer to as language practice activities, communicative teaching, or communicative tasks. In view of this potential mismatch, if the Hong Kong government insists on using the term task to denote communicative language practice, then a more suitable term might be task-supported teaching. Task-supported teaching refers to a weak version of task-based instruction that enables students to practice communicating using language items that the teacher has introduced in a traditional way (Ellis, 2003). This term seems to better characterise what I observed in the classrooms than the term task-based learning used in the government documentation (Curriculum Development Council, 1997, 2002). The fact that the teachers reinterpreted the task-based innovation underscores a point from the literature on change management: Teachers knowledge and experiences are central to the change process, but they are often neglected (Cohen, 2002). Teachers mould innovations to their own abilities, beliefs, and experiences; the immediate school context; and the wider sociocultural environment. At a time when English language teaching is expanding in primary schools in the Asia Pacic region (Nunan, 2003), more studies are needed of how educational practices are reinterpreted according to learners age and prociency levels, the sociocultural realities of schooling, and the teachers training and beliefs.
ACKNOWLEDGMENTS
I would like to thank Bill Littlewood, Paul Stapleton, Martin Bygate, Jill Bell, and the two anonymous reviewers for their very helpful engagement with previous versions of this article.

THE AUTHOR
David Carless is a senior lecturer in the English Department at the Hong Kong Institute of Education. His main research interests are classroom-based research, the management of educational change, and assessment for learning.

ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION

659

REFERENCES
Adamson, B., & Morris, P. (1998). Primary schooling in Hong Kong. In J. Moyles & L. Hargreaves (Eds.), The primary curriculum: Learning from international perspectives (pp. 181204). London: Routledge. Barkhuizen, G. (1998). Discovering learners perceptions of ESL classroom teaching/learning activities in a South African context. TESOL Quarterly, 32, 85108. Bassey, M. (2001). A solution to the problem of generalisation in educational research: Fuzzy prediction. Oxford Review of Education, 27, 522. Breen, M. (1987). Learner contributions to task design. In C. Candlin & D. Murphy (Eds.), Language learning tasks (pp. 2346). Englewood Cliffs, NJ: Prentice Hall. Breen, M. (2001). Overt participation and covert acquisition in the language classroom. In M. Breen (Ed.), Learner contributions to language learning: New directions in research (pp. 112140). Harlow, England: Longman. Bygate, M. (2000). Introduction to special issue on tasks in language pedagogy. Language Teaching Research, 4, 185192. Bygate, M., Skehan, P., & Swain, M. (2001). Introduction. In M. Bygate, P. Skehan, & M. Swain (Eds.), Researching pedagogic tasks: Second language learning, teaching and testing (pp. 120). Harlow, England: Pearson. Cameron, L. (2001). Teaching languages to young learners. Cambridge, England: Cambridge University Press. Candlin, C. (1987). Language learning tasks. In C. Candlin & D. Murphy (Eds.), Language learning tasks (pp. 522). Englewood Cliffs, NJ: Prentice Hall. Candlin, C. (2001). Afterword: Taking the curriculum to task. In M. Bygate, P. Skehan, & M. Swain (Eds.), Researching pedagogic tasks: Second language learning, teaching and testing (pp. 229243). Harlow, England: Longman. Carless, D. (1997). Managing systemic curriculum change: A critical analysis of Hong Kongs Target-Oriented Curriculum initiative. International Review of Education, 43, 349366. Carless, D. (1999). Perspectives on the cultural appropriacy of Hong Kongs TargetOriented Curriculum (TOC) initiative. Language, Culture and Curriculum, 12, 238 254. Carless, D. (2002). Implementing task-based learning with young learners. ELT Journal, 56, 389396. Carless, D. (2003). Factors in the implementation of task-based teaching in primary schools. System, 31, 485500. Carless, D., & Gordon, A. (1997). Primary English teachers attitudes towards the implementation of task-based teaching. Journal of Basic Education, 7, 139159. Carless, D., & Wong, P. M. (2000). Teaching English to young learners in Hong Kong. In M. Nikolov & H. Curtain (Eds.), An early start: Young learners and modern languages in Europe and beyond (pp. 209222). Graz, Austria: European Centre for Modern Languages. Clark, J., Lo, Y. C., Hui, M. F., Kam, M., Carless, D., & Wong, P. M. (1999). An investigation into the development and implementation of the TOC initiative with special reference to professional competencies, professional development, and resources: Final report. Hong Kong SAR, China: Hong Kong Institute of Education. Clark, J., Scarino, A., & Brownell, J. (1994). Improving the quality of learning: A framework for Target-Oriented Curriculum renewal. Hong Kong SAR, China: Institute of Language in Education. Cohen, R. (2002). Schools our teachers deserve: A proposal for teacher-centred reform. Phi Delta Kappan, 83, 532537. Curriculum Development Council. (1997). Syllabuses for primary schools: English language primary 16. Hong Kong SAR, China: Government Printer.
660 TESOL QUARTERLY

Curriculum Development Council. (2001). Learning to learn: The way forward in curriculum development. Hong Kong SAR, China: Government Printer. Curriculum Development Council. (2002). English language education: Key learning area curriculum guide (primary 1secondary 3). Hong Kong SAR, China: Government Printer. Education Commission. (1994). Report of the working group on language prociency. Hong Kong SAR, China: Government Printer. Education Department. (1994). General introduction to target-oriented curriculum. Hong Kong SAR, China: Author. Education and Manpower Bureau. (2002). Teacher survey. Hong Kong SAR, China: Government Printer. Eldridge, J. (1996). Code-switching in a Turkish secondary school. ELT Journal, 50, 303311. Ellis, G. (1996). How culturally appropriate is the communicative approach? ELT Journal, 50, 213218. Ellis, R. (2000). Task-based research and language pedagogy. Language Teaching Research, 4, 193200. Ellis, R. (2003). Task-based language learning and teaching. Oxford, England: Oxford University Press. Evans, S. (1996). The context of English language education: The case of Hong Kong. RELC Journal, 27, 3055. Fullan, M. (1999). Change forces: The sequel. London: Falmer. Glenwright, P. (2002). Language prociency assessment for teachers: The effects of benchmarking on writing assessment in Hong Kong schools. Assessing Writing, 8, 84109. Ho, W. K., & Wong, R. Y. L. (Eds.). (2003). English language teaching in East Asia today: Changing policies and practices. Singapore: Eastern Universities Press. Holliday, A. (2002). Doing and writing qualitative research. London: Sage. Kowal, M., & Swain, M. (1994). Using collaborative language production tasks to promote students language awareness. Language Awareness, 3, 7393. Kramsch, C., & Sullivan, P. (1996). Appropriate pedagogy. ELT Journal, 50, 199212. Lai, C. (1993). Communication failure in the language classroom: An exploration of causes (Research Report No. 25). Hong Kong SAR, China: City University of Hong Kong. Li, D. (1998). Its always more difcult than you plan and imagine: Teachers perceived difculties in introducing the communicative approach in South Korea. TESOL Quarterly, 32, 677697. Lin, M. Y. A. (1990). Teaching in two tongues: Language alternation in foreign language classrooms (Research Report No. 3). Hong Kong SAR, China: City Polytechnic of Hong Kong. Lin, M. Y. A. (2000). Deconstructing mixed code. In D. S. C. Li, A. Lin, & W. K. Tsang (Eds.), Language and education in postcolonial Hong Kong (pp. 179194). Hong Kong SAR, China: Linguistic Society of Hong Kong. Lin, M. Y. A., & Luk, J. (2002). Beyond progressive liberalism and cultural relativism: Towards critical postmodernist, sociohistorically situated perspectives in classroom studies. Canadian Modern Language Review, 59, 97124. Lincoln, Y., & Guba, E. (2000). The only generalisation is: There is no generalisation. In R. Gomm, M. Hammersley, & P. Foster (Eds.), Case study method (pp. 2744). London: Sage. Littlewood, W. (1993). Cognitive principles underlying task-centred TL learning. In N. Bird & M. Ingham (Eds.), Language and content (pp. 3955). Hong Kong SAR, China: Institute of Language in Education. Littlewood, W. (2004). The task-based approach: Some questions and suggestions. ELT Journal, 58, 319326.
ISSUES IN TEACHERS REINTERPRETATION OF INNOVATION 661

Markee, N. (1997). Managing curricular innovation. Cambridge, England: Cambridge University Press. Morris, P. (2000). The commissioning and decommissioning of curriculum reforms: The career of the Target-Oriented Curriculum. In B. Adamson, T. Kwan, & K. K. Chan (Eds.), Changing the curriculum: The impact of reform on primary schooling in Hong Kong (pp. 2140). Hong Kong SAR, China: Hong Kong University Press. Nikolov, M. (1999). Natural born speakers of English: Code switching in pair- and group-work in Hungarian primary classrooms. In S. Rixon (Ed.), Young learners of English: Some research perspectives (pp. 7288). Harlow, England: Longman & the British Council. Nunan, D. (1999). Second language teaching and learning. Boston: Heinle & Heinle. Nunan, D. (2003). The impact of English as a global language on educational policies and practices in the Asia-Pacic region. TESOL Quarterly, 37, 589613. Pennington, M. (1997). Projecting classroom language use in a group of bilingual graduates of a BATESL course. Language, Culture and Curriculum, 10, 222235. Peshkin, A. (1988). In search of subjectivityones own. Educational Researcher, 17, 1722. Pinter, A. (1999). Investigations into task-related strategy use with young learners of English. In S. Rixon, (Ed.), Young learners of English: Some research perspectives (pp. 117). Harlow, England: Longman & the British Council. Prabhu, N. (1987). Second language pedagogy. Oxford, England: Oxford University Press. Richardson, L. (2000). Writing: A method of inquiry. In N. Denzin & Y. Lincoln (Eds.), Handbook of qualitative research (pp. 923948). Thousand Oaks, CA: Sage. Rixon, S. (1991). The role of fun and games activities in teaching young learners. In C. Brumt, J. Moon, & R. Tongue (Eds.), Teaching English to children: From practice to principle (pp. 3348). Harlow, England: Longman. Seedhouse, P. (1999). Task-based interaction. ELT Journal, 53, 149156. Skehan, P. (1996). A framework for the implementation of task-based instruction. Applied Linguistics, 17, 3862. Skehan, P. (1998). Task-based instruction. Annual Review of Applied Linguistics, 18, 268286. Skehan, P. (2003). Task-based instruction. Language Teaching, 36, 114. Swain, M., & Lapkin, S. (2000). Task-based second language learning: The uses of the rst language. Language Teaching Research, 4, 251274. Swain, M., & Lapkin, S. (2001). Focus on form through collaborative dialogue: Exploring task effects. In M. Bygate, P. Skehan, & M. Swain (Eds.), Researching pedagogic tasks: Second language learning, teaching and testing (pp. 251274). Harlow, England: Pearson. Tomlinson, B. (1988). Conicts in TEFL: Reasons for failure in secondary schools. Institute of Language in Education Journal, 4, 103110. Tripp, D. (1993). Critical incidents in teaching: Developing professional judgement. London: Routledge. Tripp, D. (1994). Teachers lives, critical incidents, and professional practice. International Journal of Qualitative Studies, 7, 6576. Tsui, A. B. M. (2003). Understanding expertise in teaching: Case studies of ESL teachers. Cambridge, England: Cambridge University Press. Widdowson, H. (1998). Skills, abilities and contexts of reality. Annual Review of Applied Linguistics, 18, 323333. Willis, J. (1996). A framework for task-based learning. Harlow, England: Longman.

662

TESOL QUARTERLY