The Language Teacher
02 - 2000
A Holistic Approach to Task-Based Course Design
Jane Willis
Aston University, UK
After outlining three basic principles relating to language learning, I then describe and illustrate a holistic process of course and materials design which takes these principles into account.
1 Three principles for materials designers
Principle 1: Language is a meaning system.
Halliday (1973) emphasises that learning a second language involves the acquisition of a new system for realizing familiar meanings. In natural SLA circumstances, we begin by wanting to mean, (and understand what others mean) and then go on to seek or notice wordings that express those meanings. Language does not exist in a vacuum, and it does not develop in a vacuum. This is why in classroom circumstances lists of words and sample patterns taught as single items very rarely become part of the learners' deployable system. Language develops in response to the need to mean and to understand what others mean (Halliday, 1973; Willis & Willis, forthcoming). It follows that materials we offer learners should allow them to focus first on meanings in contexts and then go on to look at the wordings that realize the meanings.
From this, we can argue that any pedagogical process which supports natural acquisitional processes should therefore lead from meanings to wordings.
This is a major principle behind a task-based approach to course design. In setting learners a task to achieve (e.g., a problem to solve), the emphasis is first on learners' exchanging meanings to complete the task, using whatever language they can recall. Then they examine the language that fluent speakers or writers used to do the same task and focus on typical words, phrases and patterns (i.e., wordings) that occurred (Willis, 1996b, 1998a, 1999).
Principle 2. Exposure to the target language in use is vital.
To acquire a new language system, learners need exposure to the kinds of language that they will need (Krashen, 1987). It follows, then, that whatever learners hear and read as part of their course needs to reflect, as far as is possible, the typical features of the language of the learners' target discourse communities.
For example, if learners need to understand spontaneous informal interaction in English, they must be exposed to typical samples of spontaneous talk. If learners of business English need to be able to respond to letters of complaint, they need to have read and studied the language of typical responses to letters of complaint.
If learners don't know why they are learning English, they need exposure to a broad and varied selection of materials that will encourage them to go on using their English and learning outside class on their own and to gain a solid foundation they can build on, once they discover what their language needs will be.
In all cases, the choice of language data, both recordings of spoken language and written texts, is of vital pedagogical importance. Course designers should aim to choose a representative set of target text types from accessible real life sources-samples that reflect the typical language features of genres from the learners' present or future discourse communities. This is a major principle behind corpus-based approaches to language syllabus design and data-driven learning.
Principle 3. Some focus on language form is desirable.
Although many people acquire a new language with no formal tuition (through exposure to the target language and opportunities to use it to express their own meanings, there is now some evidence that learners do better if, at some point, their attention is drawn to typical features of language form (Skehan, 1994). This can be done in two ways:
- Through consciousness-raising exercises highlighting frequently used language items, to help learners perceive patterns (Long, 1988; Schmidt, 1990), and systematize what they know.
- By challenging learners to communicate in circumstances where accuracy matters (e.g. making a public presentation of their ideas or findings), so they feel the need, at a prior planning stage, to organize their ideas clearly and to check that their lexical choices, their grammar and pronunciation are accurate. (A similar need for a prestige variety was identified in research by Labov, 1970.) The cycle of Task -> Planning -> Report, which forms the central part of a TBL framework, caters to this (Willis, 1996a, 1998a).
2 From principles to practice
We have said that to acquire the target language effectively, learners need to engage actively in processing the meanings of whatever they hear and read. A variety of communication tasks can be designed which will motivate and give learners a purpose for doing this. These tasks should also give learners practice in the skills they will need, for example, reading a text quickly to extract specific information, taking part in spontaneous spoken interaction, or giving planned oral or written presentations. Subsequent exercises can then be devised which draw attention to the language of the texts and recordings, to the words, phrases, and patterns typical of that genre or topic.
Thus, materials designers have three distinct responsibilities: (a) providing appropriate language data for the course, (b) designing meaning-focused communication tasks arising out of those data that engage learners in meaning and that encourage genuine use of language, (c) designing form-focused language study exercises that raise learners' awareness of typical and useful formal features of language.
3 A holistic approach
3.1 Initial stages: gathering a pedagogical corpus
A holistic approach to course and materials design means that you begin by assembling a set of language data on topics of interest to your learners: e.g., written texts and recorded spoken extracts, roughly sufficient to form the basis for your course. This forms the raw material for your pedagogic corpus (Willis & Willis, 1996:67), which may later be refined. You then organize and sequence this material into pedagogical units, keeping notes of any gaps or imbalance of material.
The next step is to analyze this provisional corpus to identify language features that your learners would benefit from studying, and to check coverage of other useful features.
This process so far can be summarized thus:
|
|
Examples for three different courses follow.
For a general course, the pedagogic corpus might consist of authentic written texts on suitable topics from a variety of sources and recordings of natural speech--both planned talk and spontaneous interaction--with transcriptions. Sources of written data could include text books, resource books, magazines or newspapers and for spoken data, extracts from the BBC World Service (with permission), recordings of interviews, or communication tasks on specified topics carried out by fluent speakers of the target language.
For an ESP course for medical students, you might use extracts from introductory medical texts books, from popular WWW sites, for example medical notice boards, from popular magazines which have a doctor's advice column, or extracts from radio programs with medical themes. In Japan, you might interview doctors who commonly treat the expatriate community, with a view to obtaining a series of two-minute interviews, each covering a different medical topic. For more advanced students, you might use academic research articles, specialist medical journals, and recordings of mini-lectures.
In addition to appropriate story books and other texts written for children, your data for a Young Learners course could include transcripts of typical primary school activities, such as children at play in English and stories told to audiences of children.
Whether your course is ESP, EAP, or of a more general nature, you would obviously need to take into account length, complexity, and conceptual and cultural accessibility of the text or recording.
At the end of this initial design process, it is important to check that you have a good variety of text types and interactions and to identify any gaps. For example, you might find you have sufficient written texts but too few spontaneous spoken interactions with transcripts. The next step is to refine this corpus.
3.2 Task design and making recordings
For each topic or text, you can design tasks that your learners would find engaging to do themselves and that would generate the type of interaction they may need to take part in. Vary the types of task to offer different degrees of cognitive challenge. Tasks can engage learners in listing, classifying, matching, comparing, problem-solving, sharing experiences, and anecdote telling. Some can have more creative goals, such as writing a story or designing a brochure (see Willis, 1996a: 149-154; Willis, 1999).
Task instructions can then be adapted to provide opportunities for practice of the different skills your learners need: e.g., beginning with spontaneous exploratory interaction or writing individual notes or reading a text prior to doing the task, and then planning an oral (or written) public presentation of the task outcome. Different interaction patterns can be stimulated by having learners work as a team (with or without individual roles), in pairs, or as individuals.
Make recordings of fluent speakers doing these tasks (set a time limit of 1, 2 or 3 minutes) and transcribe the tasks that worked well (see Willis, 1996a: 87-98 for advice). You could also record short interviews with fluent speakers on one aspect of the topic of the unit.
The transcriptions of the tasks and interviews can then be added to the pedagogic corpus, to increase the amount of spoken data available for the course and to facilitate the study of typical features of spontaneous interaction. You should by now have a well balanced collection of texts and transcripts: Your pedagogic corpus is now ready for the next stage of course design.
3.3 Organizing the material
You will now need to sequence these raw materials into suitable pedagogical units, taking into account local constraints such as length of course and learners' willingness to read, prepare, or write up outside class hours. For example, each unit might contain one or two texts and recordings on a common theme.
Each unit will also need language-focused activities to follow the task cycles. But before these can be produced, it is better to gain an overall picture of the linguistic content of your pedagogic corpus, so that you can share out linguistic features in a balanced way among units, with free space in some units for recycling.
3.4 Two tips
- At this stage, keep back a few parallel texts and task or interview transcripts to use for class tests.
- Learners can often help in the design process by suggesting topics, supplying texts, recording interviews, and identifying skills they may need.
4 Preparation of language-focused activities
4.1 Identifying useful language features
The next step is to carry out a liguistic analysis of the pedagogic corpus, to identify the most frequent and typical language features which will be of most use to your learners.
There are two ways of doing this-either by computer or by hand.
If all your texts and transcripts can be typed or scanned into a computer, and if you have some concordance software (listed in references), you will be able to get a list of words in frequency order and a list of the top X number of words in alphabetical order; you can also call up concordances of key words sorted alphabetically to the right and to the left, which will enable you to identify frequently-occurring word combinations-i.e., lexical chunks. For more ideas on this, see Tribble & Jones (1990) or Barnbrook (1996).
A program like Wordsmith Tools, created by Mike Scott (University of Liver pool, OUP) will give you (and your learners) the chance to build lists of the most common 2-, 3-, or 4-word (or more) chunks. Notice in which texts and in which units these occur. It may be that your data will yield enough examples to allow you to focus on common time phrases or phrases of location or quantity in early units, as these are naturally so plentiful. Some phrases are topic-specific and can be highlighted together. Interactional phrases, e.g., Know what I mean? and vague language chunks, e.g., and that kind of thing are ubiquitous; they could be collected from early units and focused on in a later unit where several of them occur together in the spoken data.
With no computer, you can do a simple text-by-text analysis by hand, using published frequency lists to help you identify common words and notice useful collocations, chunks, and patterns that occur with them. Willis (1996a: 171-172) gives lists of the top 200 words (spoken and written separately).
You can also look out for patterns in discourse and how these are signaled and different ways of expressing particular meanings, notions and functions. These are only identifiable by hand.
The linguistic features thus identified can become the focus of language study exercises and shared out among the units.
4.2 Designing language study materials
This is the final stage in TBL course design. Language study exercises will be based largely on examples occurring in the data, occasionally supplemented with dictionary examples. Materials can range from traditional practice exercises to learner-centered, consciousness-raising activities which involve different kinds of operations, including identifying patterns or usages classifying hypothesis building and checking, cross-language exploration, deconstruction and reconstruction of text, recall, and reference activities (Willis & Willis, 1996: 63-76).
5 Summary and example
We can now add to our first figure to show the process in more detail.
As an example of this process, let us take the topic of Conversation And Cultural Norms for learners who want conversational English.
Text selection: for an advanced class, a short extract from a book giving advice to business people socialising with their Japanese counterparts for the first time.
For a lower level class, a recorded anecdote about how a British couple in England felt very embarrassed when an eminent Japanese professor, on arrival at their house for dinner, asked them how much they had paid for their house.
| Selection of topics. Assemble texts & recorded data, check coverage, identify gaps. Design tasks. Supplement with recordings of tasks & interviews. Final selection of data. |
Identify useful language features, allot to units, design language focus activities. |
We needed some spontaneous interaction, so we planned a two-minute task: "What don't you talk about in your culture? Make a list of things you would avoid asking about and say why." We recorded two British people doing this and transcribed the first half of their discussion.
To provide a written text parallel to the spoken interaction, we asked a fluent English-speaking Japanese woman to write some advice for visitors to Japan--in response to the same task instructions.
We sequenced these within the unit, using the spoken anecdote at pre-task stage to introduce the topic. The learners would then do the task themselves in pairs and report their list and reasons to the whole class.
As a second task, they would read the written advice and compare this with the content of their own lists. For homework, they could be encouraged to revise their lists and reasons, and record themselves giving advice as if to an English person.
So far, the emphasis has been primarily on exchanging meanings. There is also a focus on accuracy, which arises naturally when learners are asked to report their list to the whole class and later, when recording their advice, as both these lead up to public "performances."
A more specific focus on form is introduced in the subsequent Language Focus activities. The instructions for one activity were:
- Identify, from the transcript and the written text, two or three ways of introducing advice on cultural matters.
The examples in the spoken transcript were:
"You don't normally go and ask people personal questions, do you?"
"Like, I wouldn't ask you how much you paid for your sweater or what have you."
"The other day somebody said that English people don't like talking about religion, money, sex, . . . er . . . politics--what was the other one?"
The teacher here could point out that when giving advice, reporting what other people have said or done in the past is very typical in social interaction in Britain. Written examples included
"Avoid talking about politics and religion."
"Avoid starting discussions or arguments."
"It's rude to cut into someone's talk. If you have to interrupt someone, it's better to start by saying something like 'I'm sorry to interrupt but . . ."'
The teacher here could point out these patterns and ask learners to find more examples of each. Other consciousness-raising activities could include the following:
- Find, in the spoken transcript, eight phrases containing the word I and try to classify them according to their function in the interaction.
- Find seven phrases containing words ending in -Iy. Write down any five phrases but miss out the -Iy words. See if your partner can read your phrases out loud putting back the right -ly word. E.g., That's funny because the other day... (the word actually is missing).
- Find, in the written text, phrases with one verb followed by another verb in +ing, e.g., ..it's better to stop talking. Can you find in the text five different verbs which come before a verb ending in -ing? Can you think of two more?
These activities illustrate three of the consciousness raising activity types listed in 4.2 above. Can you identify which? (This is a classifying task!)
7 Conclusion
Some approaches to materials design begin with lists of specific language forms and skills, and then try to find (or concoct) texts and tasks which illustrate their use. These I would call additive in approach, as they lead from wordings to meanings.
This paper has described a holistic course design process which begins with whole texts and activities that involve processing meanings in contexts, and which then leads on to a focus on wordings and form.
Basing materials on a well-selected pedagogic corpus means that recycling of common and typical language features will happen naturally inside the classroom, and that learners will be far better prepared for whatever English they meet and need to use outside the classroom.
References
Barnbrook, G. (1996). Language and computers: A practical introduction to a computational analysis of language. Edinburgh: University of Edinburgh.
Halliday, M. (1975). Learning how to mean.. London: Edward Arnold.
Krashen, S. D. (1985). Principles and practice in second language acquisition.. Oxford: Pergamon.
Labov, W. (1970). The study of language in its social context. In P. Giglioli (Ed.), Language and social context. Harmondsworth: Penguin.
Long, M. (1988). Instructed interlanguage development. In L. Beebe (Ed.), Issues in second language acquisition: Multiple perspectives. Rowley, Mass: Newbury House.
Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics 11, 2 (129-158).
Scott, M. (1998). Wordsmith tools (concordance software) available only through OUP's Internet site: http://www.oup.co.uk/elt
Skehan, P. (1994). Interlanguage development and task-based learning. In M. Bygate, A. Tonkyn, & E. Williams (Eds.), Grammar and the language teacher (pp.175-199). Hemel Hempstead: Prentice Hall.
Tribble, C. & Jones, G. (1990). Concordancing in the Classroom. Harlow: Longman.
Willis, D. & Willis, J. (forthcoming). Six propositions in search of a methodology: Applying linguistics to task-based learning. In I. de la Cruz, C. Santamaria, C. Tejedor, C. Valero (Eds.), Estudios de linguistica aplicada a finales de siglo. Alcala: Universidad de Alcala.
Willis, J. (1996a). A framework for task-based learning. Harlow: Longman.
Willis, J. (1996b). Consciousness-raising activities in the language classroom. In J. Willis & D. Willis (Eds.), Challenge and change in language teaching. Oxford: Heinemann ELT.
Willis, J. (1997). Exploring spoken language: Analysis activities for trainers and trainees. In I. McGrath (Ed.), Learning to train. Englewood Cliffs, NJ: Prentice Hall International.
Willis, J. (1998a). Task-based learning--What kind of adventure?" The Language Teacher 22, 7 (17-18).
Willis, J. (1998b). Concordancing in the classroom without a computer. In B. Tomlinson (Ed.), Materials development in second language teaching. Cambridge: Cambridge University Press.
Willis, J. (1999). Designing and using tasks to promote optimum language development. In A. Barfield, R. Betts, J. Cunningham, N. Dunn, H. Katsura, K. Kobayashi, N. Padden, N. Parry, & M. Watanabe (Eds.), Focus on the classroom: Interpretations. Proceedings of the 24th JALT international conference (pp. 119-126). Tokyo: The Japan Association for Language Teaching.
Willis, J. & Willis, D. (Eds.). (1996). Challenge and change in language teaching. Oxford: Heinemann ELT.
All materials on this site are copyright © by JALT and their respective authors.
For more information on JALT, visit the JALT National Website