Task-based Learning - A new Approach

Seminar Paper

Task-based Learning:

A new Approach

Introduction

This seminar paper is intended to give an overview of a new method of language teaching, which came into fashion over the last few years. This new approach is called "task-based learning" and is based on the assumption that languages are better learnt through communicative tasks, where an outcome must be achieved, than through isolated excercises which only serve to practise one aspect of the language. In contrast to traditional teaching, grammar is not analysed before working on the task, but rather after having completed the assignment.

Ellis points out that "the view that communcation serves as one of the primary ways in which learners obtain data with which to construct their interlanguages is now well established" (Ellis, 43). Allwright even argued that interaction is "the fundamental fact of pedagogy" and that "successful pedagogy involves the successful management of classroom interaction" (Allwright, 1984, quoted in: Ellis, 43).

According to Widdowson, it is not surprising why teaching approaches based on communication are in current fashion:

They bring the means of learning into alignment with its eventual ends - the achievement of an ability to use language to communicative effect. Furthermore, at the same time, they represent the language to be learned as the same sort of natural phenomenon as the language the learners already know, and so allow them to draw on their own experience in the process of learning.

(Widdowson, 160)

Although a communicative approach has its obvious attractions, it has also its problems. One of them is the natural learning problem. Widdowson states that learners do not very readily infer knowledge of the language system from their communicative activities and the grammar then proves elusive (Widdowson, 161). Very often the situation arises where learners acquire a fairly imperfect repertoire of performance which is not supported by an underlying competence.

Of course, there has to be a certain structure in the learning process, as natural it might be called. Teachers cannot expect from learners to derive all necessary knowledge of the language system from communication tasks and listening activities. Analysing language and understanding grammar is an essential part in the language learning process. Task-based learning takes this part of language teaching into consideration and does not deny it, as other communicative approaches might do. The difference to traditional teaching, however, is that language focus here comes after the communcation tasks and is based on the language that occured in the authentic input and naturally by the learners in the language exposure phase and the task phase.

Task-Based Learning

2.1 A new approach

It is possible to learn to speak a language quite fluently without any teaching at all. We all have learnt our mother-tongues by having been exposed to natural language which we then imitated and analysed in order to understand how it works. Some learners aquired a foreign language in that way. They may not always be totally accurate, but their level of language ability is entirely adequate for their needs. Learners having managed to learn a second language in the way most people learn their first, are usually very motivated, have received a lot of exposure to authentic language and have had many opportunities to speak and experiment with the language, as Willis states (Willis, 4).

Task-based learning tries to set up classroom and learning conditions that language learning can take place in the most natural way possible. The situations with which the learners are confronted are not designed to practise a specific part of the language but to give learners the opportunity for using language not in terms of form, but rather in terms of meaning.

Tasks are always activities where the target language is used by the learners for a communicative purpose in order to achieve an outcome. The emphasis is on understanding and conveying meaning in order to complete the task successfully. To achieve the demanded outcome, learners would be focusing first on meaning and then on the best ways to express that meaning linguistically. So, learners are free to choose whatever language forms they want to use in order to fulfil the task goals. When the need for communication is strongly felt, they will find a way of getting round words or forms they do not yet know or cannot remember.

Skehan gives a number of examples for task-based activities:

completing one another's family trees;

agreeing on advice to give to the writer of a letter to an agony aunt;

discovering whether one's paths will cross [] in the next week;

solving a riddle;

leaving a message on someone's answering machine.

(Skehan, 95 f.)

In contrast, a number of classroom activities would not count as tasks:

completing a transformation excercise;

(most) question and answer activities with the teacher;

inductive learning activities where pre-selected material is conducive to the generation of language rules.

(Skehan, 96)

The teacher can monitor the activities in class from a distance and should encourage all attempts to communicate in the target language. He should, however, not correct learners' mistakes or give advice. Fluency in communication is what counts. The emphasis is on meaning and not on practising language forms correctly.

According to Nunan, communication tasks have been defined as tasks that "involve learners in comprehending, manipulating, producing or interacting in the target language while their attention is principally focused on meaning rather than form" (Nunan, 1989, quoted in: Ellis, 79). The pedagogic rationale for the use of communcation tasks rests partly on the claim that they will help to develop learners' communicative skills and partly on the claim that they will contribute incidentally to their linguistic development. Hence, communication tasks are important for both fluency and accuracy.

It is, obviously, necessary to adapt the task dimension and task difficulty to the learners concerned. Skehan proposes a three-way distinction for the analysis of tasks which distinguishes "code complexity", "cognitive complexity" and the category of "communcative stress" which includes factors such as time limits and time pressure, speed of presentation or the length of texts used (Skehan, 99).

2.2 Conditions for language learning

As Willis writes, for anyone to learn a language with reasonable efficiency, three essential conditions must be met (Willis, 11).

Firstly, there has to be much exposure to the target language, a rich but comprehensible input of real spoken and written language in use. Modification of the target language in spoken discourse by the teacher in order to help the learners understand is done quite unconsciously and is beneficial as long as it is not carried too far. Simplifying texts, however, can deprive learners of the opportunity to become familiar with original forms, which may occur frequently in the target language. Besides, it does not necessarily make the task of comprehension easier.

As well as input, also output is generally considered essential for language learning. Learners need opportunities to communicate what they want to say and express what they feel or think. They need the experience of communication in a variety of situations, e.g. in different size groups or for different audiences. In a positive, supportive and low stress atmosphere created by the teacher, creativity is most encouraged.

The third essential condition learners need is motivation to learn; motivation to process the exposure they receive and motivation to use the target language as often as possible. Learners' motivation may be intergrative, that is to identify with the target language and culture, or mere instrumental, that is to see the target language as a means to achieve a specific aim. Success and satisfaction are, however, the key factors in sustaining motivation. The more learners can be motivated efficiently to seek out opportunities for exposure and use of the target language inside and outside the classroom, the better.

The role instruction plays in the language learning process is somewhat unclear amongst scholars. It is generally accepted that instruction focused on language form can speed up the rate of language development and raise the level of the learners' attainment. What instruction does not seem to do, still, is to change the learners' developmental sequence. Learners will not necessarily learn what is taught to them. A syllabus reflecting the natural acquisition order is not the ultimate solution to this problem. Firstly, studies about the natural aquisition order are restricted to certain morphemes like -ing, -ed and -s endings. But language involves much more than this and we have little idea of the order in which other features may be acquired. Secondly, we cannot certainly know which learners are at which stage. Even within a class there may be major differences amongst learners. Instruction can certainly help learners notice specific features of the target language. It can give them the opportunity to process grammatical and lexical patterns and to form hypotheses about their use and meaning. Instruction should, however, not become more important than the exposure and use of the language.

2.3 Task-based learning and traditional methods of language teaching

As Skehan writes, the most influential traditional method of organising laguage teaching is referred to as PPP - presentation, practice and production (Skehan, 93). This traditional method is used all over the world by a lot of teachers. The aim of a PPP lesson is to teach a specific language form, e.g. a grammatical structure or the realisation of a particular function. The language is tightly controlled and the emphasis is on getting the new form correct.

In the presentation stage, the teacher begins by presenting an item of language in a context or situation to clarify its meaning. In the practice stage, the learners repeat target items and practice sentences until they can say or write them correctly. In the production stage, finally, learners are expected to produce the items they have just learnt, in a "free" situation.

Although this way of teaching new items seems logical and efficient, there are nevertheless several problems with this paradigm. Sometimes learners manage to pass the production stage without using the target form at all. Sometimes they tend to overuse the target form and produce very unnatural pieces of conversation. Besides, PPP gives an illusion of mastery since learners can often produce the required forms confidently in the classroom, but once outside, they either do not produce them at all or use them incorrectly.

According to Willis, the PPP cycle derives from the "behaviourist view of learning, which rests on the principle that repitition helps to 'automate' responses, and that practice makes perfect" (Willis, 135). Yet, language learning rarely happens in an additive way, with bits of language being learnt one after the other. Language learners rather take advantage of their exposure to the target language in use and grapple with meaning, try to understand underlying structures and observe how others express the meanings they want to be able to convey at the same time. We cannot predict what learners are going to learn at any given stage. Instruction, as stated above, does help but it cannot guarantee when something is learnt. Rich and well-chosen exposure to the target language helps language develop gradually and organically, out of the learner's own experience. The PPP cycle, still, restricts the learner's experience of language by focusing on a single item. Therefore, the goal of the last stage, the production phase, namely free production, is often not achieved.

The task-based learning framework offers far more opportunities for free language use. All components of the learning cycle are free of language control. Learners rely upon the linguistic resources they have already acquired and always use language for a genuine purpose. The task supplies a genuine need to use language to communicate, and the other components follow naturally from the task. The process of consciousness raising used in language focus activities encourages learners to think and to analyse, not simply to repeat, manipulate and apply.

The framework solves another language-teaching problem; that of providing a context for grammar teaching and form-focused activities. In a PPP cycle, where the presentation of the target language comes first, this context has to be invented. In task-based learning, the context is already established by the task itself.

Here, the concept of focused and unfocused communication tasks has to be considered. Ellis makes a distinction between these two. In the case of unfocused communication tasks,

no effort is made in the design of execution of a task to give prominence to any particular linguistic feature. The language used to perform the task is natural and only very broadly determined by the content of the task.

(Ellis, 82)

A focused communication task, in contrast,

does result in some linguistic feature being made prominent, although not in a way that causes the learner to pay more attention to form than meaning.

(Ellis, 83)

The teacher's roles and approach to lesson planning also differ. In a PPP lesson, teachers are in the centre of the class, having everything under control. In task-based learning, however, teachers have to learn to set things up and then hold back, intervening only when needed and reviewing each phase at the end.

So far, we have mainly been concerned with oral communication in the task cycle. Of course, also the skill of writing should be trained within this framework. Yet, the approach towards this also differs from the traditional PPP approach. In real life, only a small proportion of people do more than writing personal letters or filling out forms, even in their first language. Language learners, of course, need to write for other reasons. It is well known that writing is in itself a learning process. It helps people clarifying ideas and creating new ones.

In real life, we only write in order to communicate something to someone. Foreign language writing is traditionally often done for display, so it can be graded rather than for any real communication purpose. To make a change, to give learners a real sense of purpose and to raise motivation, task-based learning promotes the idea of thinking of audiences that might benefit by reading something learners have written. Guide books to the village or town where the school is set or brochures about local activities could be useful for tourists and visitors, a brief school history might be interesting for parents and a class magazine or a diary of a holiday course could be read by other classes or learners.

Summing up, a PPP cycle, according to Willis, "leads from accuracy to fluency; a task-based learning cycle leads from fluency to accuracy" (Willis, 137). It begins by providing learners with a holistic experience in language and then helps them to analyse this language in order to learn more efficiently. PPP, on the other hand, provides specific language items in a vacuum and then looks for some activity to practise them.

2.4 Micro-evaluation: Does a task work?

To say that a task "works" can mean a number of rather different things.

Teachers reflecting on a task might feel that it has worked if they have evidence that the learners found it enjoyable and useful. Yet, it is perfectly possible that learners enjoy doing a task and give it positive ratings in a questionnaire and yet fail to perform it successfully or learn nothing from it.

Whereas the criteria for the evaluation of a closed task are embedded within the task itself, the criteria required for evaluating an open task are not. They are, as Ellis states, "external to the task, and because they are usually not specified by the person who devised the task, they place a considerable burden on the teacher" (Ellis, 102).

A full evaluation of a task calls for an "external evaluation". This could be carried out theoretically by determining to what extend it meets explicitly defined criteria. Such an evalution is predictive in nature. Alternatively, a retrospective evaluation could be carried out. Ellis proposes three different kinds of retrospective evaluations:

A student-based evaluation provides information about how interesting and useful learners perceive a task to be. A response-based evaluation is internal in nature because it simply addresses the question "Was the students' response the one intended by the designer of the task?" A learning-based evaluation is external in nature because it goes beyond the task itself by trying to determine whether the task actually contributed to the learners' second language proficiency.

(Ellis, 103 f.)

To carry out a micro-evaluation, which does not have to follow any formal conventions and rules but can consist of easy and straight-forward questions to find out if the intended goal has been achieved, can contribute effectively to the teacher's own development. He can easily realise which parts of the task went well and which were too difficult. This can help him a lot in finding ways of helping the learners to fully appreciate the task and to cope with predictable problems.

Task evaluations engage teachers as insider researchers and reflective practitioners, thus helping them to examine and reconstruct their own pedagogical values and develop their own personal theories of language teaching. The task-based learning framework is flexible enough to account for all findings and possible necessary changes and adaptations.

The components of the framework

The task-based learning framwork basically consists of three phases; pre-task, task cycle and language focus. The pre-task phase introduces the class to the topic and the task by activating topic-related words and phrases. The task cycle offers learners the chance to use whatever language they already know in order to carry out the task, and then to improve that language. Exposure to language can be provided at different stages, depending on the type of task. The last phase in the framework, language focus, allows a closer study of some of the specific features naturally occuring in the language used during the task cycle. It includes analysis and practice components and serves the purpose most PPP concepts rely on - explicit study of language form.

The components within each phase of the framework provide a naturally flowing sequence, each one preparing the ground for the next. Let us now take a closer look at each of the individual steps of a task-based learning cycle.

3.1 The pre-task phase

The pre-task phase is usually the shortest stage in the framework. It can last between two and twenty minutes, depending on the learners' familiarity with the topic and the type of task.

At first, learners have to be given a definition of the topic area. They may, especially if they come from other cultures, hold different views on what some topics are about. To make the learners then ready for the task, words and phrases that might be useful have to be recalled. This can be done in a number of ways.

According to Willis, pre-task activities to explore topic language "should actively involve all learners, giving them relevant exposure, and [] create interest in doing a task on this topic" (Willis, 43). One way of doing this is by classifying words and phrases connected with the topic. Playing "Odd one out", where an item that does not fit in a set of related words or phrases has to be found, or matching phrases to pictures are also useful techniques. In some classes, drawing mind-maps might help learners to become familiar with the topic area.

The third step in the pre-task phase is to ensure that all learners understand what the task involves, what its goals are and what outcome is required. Apart from mere explaining the task, the teacher can show the class what previous learners have achieved or demonstrate the task with a good learner.

3.2 The task-cycle

After working hard to set the scene in the introduction phase, the teacher now in the task stage acts as an observer monitoring what is going on in the classroom and acts as a time keeper. He should make sure that all groups are doing the right task and that really all the learners take part. As a passive observer, he ought to be forgiving about errors of form and should only interrupt and help if there is a major communication break-down.

The task component helps learners to develop fluency in the target language and strategies for communcation. The main focus lays on the meaning which has to be conveyed. Through tasks, learners may well become better communicators and learn new words and phrases but it is often argued that this does not necessarily streches the learners' language development or help with internalisation of grammar.

This is supplied by the report stage, where learners naturally aim for accuracy and fluency. It gives them a natural stimulus to upgrade and improve their language. In fact, it is a real linguistic challenge - namely to communicate clearly and in accurate language appropriate to the circumstances.

In the task phase, when they speak in real time, learners just tack words and phrases together in a more or less improvised fashion. In planning their report, in contrast, they have to create a comprehensive and compact summary of what has happened with the support of their group, the teacher, dictionaries and grammar books. The teacher's main role now is that of a language adviser, helping learners to shape their meanings and to express exactly what they want to say. He ought to comment on good points and creative use of language and should, if learners ask to be corrected, point out errors selectively - most important are those which obscure the meaning. For other errors of form, learners should try to correct each other.

The report stage, Willis points out, then probably presents "slightly less of a learning opportunity than the planning stage" (Willis, 58). But without the report, the learning process of planning, drafting and rehearsing would not happen. Learners naturally feel curious what their colleagues have achieved during the task and actively join in the report stage. A report might last as little as 30 secondes or up to two minutes. Of course, also the reports are bound to strange wordings and grammatical errors. What must be taken into consideration, however, is that learners here offer the best language they can achieve at that moment, given the linguistic resources and time available. During the report stage, the teacher acts as a chairperson, introducing the presentations, deciding who speaks next and summing up at the end. He ought to keep an eye on the time and stop the report stage when it becomes repetetive.

Giving reports can be done orally or by writing. Audio and video presentations can be included and a number of media should be used to make the reports as interesting and vital as possible. If feedback is given by the teacher, it should be tactfully and positively. Whenever possible, learners ought to be encouraged to find out mistakes by themselves. This can be done by little quizzes and guessing games, noting the respective phrases on the blackboard but leaving a gap where the mistake occured. Learners then should complete the phrase in order to make it correct.

3.3 Language focus

Within the task-based learning framework, tasks and texts give learners a rich exposure to language and also opportunities to use it themselves. In addition to that, they also benefit from instruction focused on language form. This is not necessarily teacher-led, although the teacher mostly introduces the activities, is on hand while learners do them and reviews them in the end.

The activites mentioned above are sometimes called "consciousness raising activities" or "meta-communicative tasks". These are tasks that focus explicitly on language form and use, an aspect that is normally covered first in traditional language teaching. To avoid a PPP situation, analysis activities should not, as Willis writes, "consist of decontextualised presentation and practice of language items in isolation" (Willis, 102). By following the task cycle, they rather involve learners in a study of those language forms they actually used and needed during the cycle. Analysis activities give learners time to systematise and build on the grammar they know already, to make and test assumptions about the grammar and to increase their repertoire of useful lexical items.

While learners test their own hypothesis and make their own discoveries, the teacher should hold back but ought to be ready to handle individual questions. In reviewing the material they have been exposed to and the language they have used, learners not necessarily notice the same aspects as the teacher but rather pick out things that are new to them and they can fit into their own developing picture of the target language.

There are three main starting points for analysis activites: semantic concepts, words or parts of a word and categories of meaning or use. Of course, the teacher has to set certain guidelines where the learners' investigation should be leading to. Starting points that will catch the right kind of samples to stimulate a deeper investigation into grammar and meaning have been proven useful. Looking for had in a text, for example, will lead learners to verb phrases with had and help them explore the use and meaning of the past perfect.

The main themes in a text or transcript are revealed in the lexis. In analysing semantic concepts, identifiying the theme words and phrases helps learners to notice lexical repetition and how this can form cohesive through the text. These words or phrases can also be used for categorising, for exploring shades of meaning and finally building up lexical sets.

Analysis tasks staring from words or parts of words can involve learners in classification according to grammatical function, exploring the meaning and effects of alternative choices of form, exploring collocation or classification according to meaning and use. Learners might also want to collect similar examples from their previous knowledge or from a dictionary.

Working on categories of meaning or use mostly consists of concordance analysis. Learners are asked to find phrases or verbs with a specific form that serve a specific function. They might be asked, for example, to find all phrases with verbs ending in -ing, which describe someone or something, which follow is/was/are/were or which follow verbs like stop and start. If there are any constructions left over, learners could try to classify them as well.

Once most learners have finished the activity, the results are discussed in class. When presenting their findings, learners should be asked to explain their reasons for classifying an example in a particular way. When the review is completed, further examples that fit in these categories can be added to the list. The teacher may also focus on other useful words or collocations that occur, always based on the linguistic material provided by the task cycle.

In the course of the analysis activities, learners practise saying target words and phrases and hear them repeated in different contexts. Practise activities can combine naturally with analysis work. On their own, they are unlikely to give learners deeper insights into the meaning and use of grammatical patterns or speed up their acquisition of these patterns. In connection with analysis work, however, they serve a valuable function and provide confidence and a sense of security.

Language practise activites start with mere repetion and listen-and-complete exercises and can reach up to memory challenge tasks and concordance and dictionary exercises. The teacher's creativity here is, of course, unlimited.

Different types of tasks

The kind of language learners are exposed to during the task circle can come from a number of sources. In the following section, the most important ones will be introduced.

4.1 Text-based tasks

Text-based tasks require learners to process a text, any piece of spoken or written continuous speech, for meaning in order to achieve the goals of the task. This primarily involves reading, listening to or viewing the text with some kind of "communicative purpose", as Willis points out (Willis, 68), and may also involve talking about the text or perhaps writing notes.

Many teachers use supplementary materials in their lessons since the language found in some coursebooks is restricted and simplified. These have to be chosen, however, with due regard both for the language and the learner. Most learners have their favourite topics or specialist areas. Extra motivation can be achieved when the teacher chooses supplementary material reflecting the learners' interests.

Texts can be found in a variety of sources.

Continuous spoken language used in the classroom would normally come from professionally made radio or TV-programmes. Some sources, like the BBC World Service, are aware of the fact that their audiences are non-native speakers of English and adapt the language they use in a natural way. According to Willis, such an adaptation can still be called "authentic", because it has not been produced "with a specific-language teaching purpose in mind, but mainly to communicate, inform and/or entertain" (Willis, 69).

When turning to written language, one has to distinguish between published and unpublished sources. Published sources are books, newspapers or magazines, unpublished sources would include for example letters from pen-friends and data collected by learners doing specialist project work. Nowadays, also the Internet is becoming a useful resource. A whole range of text-types is available, most of the material being spontaneous, unedited and available without charge. Yet, the Internet is definitely not teachers' paradise, as section 4.2 will show.

As always in task-based learning, it is also important in the case of reading tasks to give learners a specific purpose for what they are doing. Unless learners are given a specific purpose for reading, they see the text as a decontextualised learning device and read one word at a time, looking up all the words they do not understand. They should, in contrast, work out which words belong together and form units of meaning. This can only be achieved when reading for meaning is promoted. Learners have to get used to the idea that reading for partial or approximate comprehension is much more useful than aiming at perfect understanding each time.

Listening to the radio or watching TV is slightly different from reading in the sense that these activities have to be done in real time and in sequence. This can be a problem in lessons since some learners tend to panic, then get left behind and finally give up. Carefully designed tasks on well-chosen texts can prevent this happening. It is important to encourage learners to listen to the source, predict and make guesses about meanings without penalising wrong ones.

Text-based tasks cover a variety of different tasks.

In predicting tasks, for example, learners predict or attempt to reconstruct the content of a text on the basis of given clues from part of it, without having read, heard or seen the whole. In jumbles, learners are confronted with sections of parts of a complete text, but in the wrong order. It is the learners' job to rearrange these sections. Another useful excercise are restoration tasks, where learners replace words or phrases that have been omitted from a text, or identify an extra sentence or paragraph that has been put in. In jigsaw tasks, the aim is for learners to make a whole from different parts of a text, each being held by a different person or taken from a different source. Comparison tasks, finally, invite learners to compare two or more similar texts to spot factual or attitudinal differences, or to find two points in common.

Learners might have difficulties with certain texts. Factors which are likely to cause problems are unknown words or phrases, unusual metaphors and complex phrase- or clause structures. Teachers ought to consider these items when preparing their classes for the task. Generally, if a text is linguistically complex, an easy task should be set. If, on the other hand, a text is easy, more challenging tasks can be set.

4.2 The Internet - unlimited archives for teaching material?

The democratic structure of the Internet, that gives every user the opportunity to contribute his thoughts, allows a totally different view of our society and makes the Internet also interesting for education and teaching. The speed in which information is provided and can be downloaded makes it absolutely unrivalled.

Governments and governmental organisations all over the world put a great effort into making people fit for the Internet since this is the medium of the future. Yet, the view of the Internet which is promoted by various people because of various intentions is not always what the Internet really is. It offers such fantastic opportunities that very soon a commercialisation of the Internet could be noticed. This is fair enough as long as the intention is clear and as long as it makes things easier for the consumer. Nowadays, all large companies have got their own homepage which they use mainly as an up-to-date and cheap means of advertising.

Looking at homepages without a commercial background, one can find a quite good reflection of our world and of society. Thanks to free offers for homepages and webspace from advertising-financed webservice providers, everybody can nowadays easily publish their views and opinions on the Internet. This is positive on the one hand, since information can be made available to the world quite fast and unbureaucratically. It is negative, on the other hand, because there is no guarantee, not even a hint for the quality of the offered information. This is not of relevance in personal homepages presenting its author. It is, however, of great significance, when one considers the educational value of the Internet.

When we go to a library to inform ourselves about a certain topic, we can assume that the books we find there, are of rather high quality. Especially in science and studies, the points of view of the various scholars may differ enormously, but at least we can proceed on the assumption that what finally was published is well-researched and well underpinned. This is guaranteed to us by the readers and the publishers of the publishing house. On the Internet, everyone is their own publisher and nobody else than the author himself decides what to make available for the public and what not. Speaking of science and studies, information that has been insufficiently researched or that has yet not been thought through to the end can be easily found on the Internet. Not only unknown or even anonymous authors or sources which are not well-known but also traditional and well-known publishing houses offer insufficient information that is sometimes not reliable. Their data is limited just to avoid to compete the books they are publishing. Very likely, this unreliable data makes up the majority of information offered on the Internet. It is, therefore, quite risky to fully rely on the Internet when looking for important data.

4.3 Exposure to spontaneous speech

Spontaneous speech and spoken interaction in the target language are important sources of exposure for learners. Yet, this is the most difficult type of language to bring into the classrooms for teaching purposes.

Apart from teacher-led conversations, typical samples of real-time interaction are generally all too rarely heard in the language classroom. But what learners need are the kinds of words and phrases that sustain the interaction and link ideas without sounding awkward. This can be achieved by exposing learners to pieces of recorded speech showing them how fluent speakers manage the organisation of a conversation.

One practical solution for finding comprehensible material is to make one's own recordings of fluent target language speakers doing the same tasks as the learners. This exposure to samples of real-time talk is immediately relevant to the learners' learning situations.

Learners in this case have a reason for listening and get to know what the task goals are. Additionally, they get used to listening for specific things and hear "how speakers negotiate opening moves, sustain the interaction, evaluate progress and bring things to a close" (Willis, 89). Most learners find it useful using a transcript accompanying the listening task.

We can distinguish between "closed tasks" and "open tasks" here. If the task consists of a problem or a puzzle to solve, it would obviously be of no use to play the whole recording first. In this case, learners would better do the closed task first, then hear the recording afterwards. They can compare the strategies speakers used in the recording with their own strategies.

If, however, the task consists of comparing personal experiences, then it might be useful for learners to listen before they do it. In this case there is no solution - just a range of different perspectives.

While listening to task recordings, it is important for learners to feel they are managing to understand quite a lot for themselves. They should, however, not be expected to understand everything. Setting a different purpose each time they listen, each slightly more challenging than the last, is a way of grading the activity.

When teaching in an environment where fluent speakers of the target language are easily accessible, it might also be possible to get groups of learners to record interviews to bring back into the classroom or to simply bring them into contact with native speakers.

Basically, most people are willing to talk about things the learners are interested in. And learners, even if feeling a little nervous in the beginning, find that they can speak enough to hold interviews, and bring back to class something unique, personal and satisfying. Any written documentation accompanying the interviews might serve as background information for the class.

It might be a good idea to plan a series of interviews starting with people learners are familiar with, then progressing to people outside the school. Finally, it will be a challenge to find local personalities who learners do not know.

Thirdly, teachers can exploit the recorded interviews in textbooks and resource books. Although they are rarely natural and spontaneous, as samples of a certain type of spoken interaction, they are always useful.

Conclusion

It is clear that we cannot find a universal language learning methodology fitting on any teacher type, learner type, cultural background and personal preference. Each concept has its advantages, otherwise it would not have been created, but also its disadvantages. Concepts appreciated by one learner might be rejected by the other. All that methodology can do is to try to detect certain trends in society and then combine them with what research finds is good for most effective language learning.

Task-based learning, I think, is a typical approach for our time where two ideas are very important: naturalness and communication. These ideas lead like a red thread through all aspects of our lives. It is naturalness in style, in food, in behaviour, and learning; it is communication in business, in public, in privacy, and learning. The combination of the two must be the key to most effecient language learning: Let us all become native-like communication experts of any language. Naturalness stands here not only for the learning process, but also for the achieved ability for language use.

I do not think that it is impossible to speak a second language as fluent as native speakers do. Spending a number of years in the target country can help a lot. But I doubt that language learners can achieve the ability to use the respective language as natural as native speakers. Language knows so many different shades and tones, one simply has to grow up with it to be able to express them all.

Although the exposure to most natural and authentic language will not produce clones of native speakers, it can though help a lot. Although or maybe just because we cannot turn language learners into native speakers, we have to aim at the highest degree of natural language use possible. Very often, there is a major difference between the kind of English taught in schools and the kind of English spoken in Great Britain or the United States. Language forms are used inadequately, vocabulary is learnt in inappropriate contexts and the pronunciation is sometimes very similar to the first language. This is probably due to a methodology interested more in the structure of the language than in its use and to teachers who illustrate theory with invented examples adapted to the respective structure.

Task-based learning tries to avoid such an unnatural understanding of language by presenting pieces of authentic language from which the learners can derive theories about the structure of the language and which function as a model for authentic language use. Making learners to communicate themselves as much as they can supports communication abilities. The main emphasis always lays on authenticity which just means that pieces of language are not produced to serve the function to illustrate grammatical theories but to communicate certain contents.

Obviously, learners cannot find out about the structure of language just by themselves, no matter how comprehensible and appropriate the input might by. There are a number of concepts which exist in most languages and therefore can be easily discovered but certain grammatical ideas differ enormously from the learners' first language and are not as obvious. It is a central matter of task-based learning that the teacher steps back and acts as an observer in the background. This definitely supports the learners' independence and may also increase motivation but I envision the teacher being sometimes more active in the language focus stage than described in most works about task-based learning. He does not need to prescribe what learners have to think, as this is the case in traditional approaches to language learning, but it would be, according to my view, quite helpful if he would describe grammatical concepts that are beyond the learners' capability more actively.

Task-based learning is an interesting concept which tries to combine modern findings of second language acquisition research with a traditional, structural approach. Additionally, it is a highly flexible framework; its components can be easily adapted to fit any learning situation. The ideas according to which task-based learning is designed are innovative on the one hand but not really revolutionary on the other. Innovation is inherently threatening, as Prabhu has pointed out:

A new perception in pedagogy, implying a different pattern of classroom activity, is an intruder into teachers' mental frames - an unsettling one, because there is a conflict of mismatch between old and new perceptions and, more seriously, a threat to prevailing routines and to the sense of security dependent on them.

(Prabhu, 1987, quoted in: Ellis, 25)

With task-based learning, however, teachers and learners should not have major problems getting used to the new method. Communcation stands in the centre of attention but structure comes right after. So, the gap between "communication" and "structure", the two opposing concepts in language learning, is being narrowed. It can, I think, never be completely closed but a balanced compromise between the two ideas is the most effecient way anyway.

Bibliography

Ellis, R. (1997): SLA Research and Language Teaching. Oxford University Press.

James, C. (1998): Errors in Language Learning and Use. Longman.

Lightbown, P. and N. Spada (1998): How Languages are Learned. Oxford University Press.

McLaughlin, B. (1987): Theories of Second Language Learning. Edward Arnold.

Skehan, P. (1998): A Cognitive Approach to Language Learning. Oxford University Press.

Widdowson, H. G. (1990): Aspects of Language Teaching. Oxford University Press.

Willis, J. (1996): A Framework for Task-Based Learning. Longman.