Ways of teaching foreign languages

Yuklangan vaqt:

20.09.2019

Ko'chirishlar soni:

0

Hajmi:

158 KB

Ko'chirib olish

Ways of teaching foreign languages Contents 1. Introduction 1.1 General characteristics of the work 2.1 How to teach foreign languages (general remarks) 2. The Main Part 1.2 Comparing instructed and natural settings for language learning 2.2 Natural and instructional settings 3.2 Classroom comparisons 4.2 Five principles for classroom teaching 5.2. The principle getting right from the beginning 6.2. The principle of saying what you mean and meaning what you say 7.2. The principle of listening 8.2. Teach what is teachable 9.2. Getting right in the end 10.2. Grammar a question: Focusing on past tenses and conditionals (work-out) 11. 2. The implications of classroom research for teaching 3. Conclusion 4. Bibliography

1. Introduction 2.1. How to teach foreign languages (general remarks) Every few years, new foreign language teaching methods arrive on the scene. New textbooks appear far more frequently. They are usually proclaimed to be more effective than those that have gone before, and, in many cases, these methods or textbooks are promoted or even prescribed for immediate use. New methods and textbooks may reflect current developments in linguistic/applied linguistic theory or recent pedagogical trends. Sometimes they are said to be based on recent developments in language acquisition theory and research. For example, one approach to teaching may emphasize the value of having students imitate and practise a set of correct sentences while another emphasizes the importance of encouraging 'natural' communication be tween learners. How is a teacher to evaluate the potential effectiveness of new methods? One important basis for evaluating is, of course, the teacher's own experience with previous successes or disappointments. In addition, teachers who are informed about some of the findings of recent research are better prepared to judge whether the new proposals for language teaching are likely to bring about positive changes in students' learning. Our graduation paper is about how English language can be learned at classrooms on the bases of new pedagogical technologies with having taking into consideration the national aspect, i.e. influencing native Uzbek language and typical mistakes and difficulties in learning English by Uzbek speaking students. First of all we have written it for English language teachers who teach this language to Uzbek students at schools at 5-6 grades, but it could also be useful for afult learners who are only going to learn a wonderful world of English. We believe that information about findings and theoretical views in second language acquisition research can make you a better judge of claims made by textbook writers and proponents of various language teaching methods. Such information,

combined with insights gained from your experience as a language teacher or learner, can help you evaluate proposed changes in classroom methodology 2.The Main Part 1.2. Comparing instructed and natural settings for language learning 1 Most people would agree that learning a second language in a natural acquisition context or 'on the street' is not the same as learning in the class room. Many believe that learning 'on the street' is more effective. This belief may be based on the fact that most successful learners have had exposure to the language outside the classroom. What is special about natural language learning? Can we create the same environment in the classroom? Should we? Or are there essential contributions that only instruction—and not natural exposure—can provide? In this chapter, we will look at five proposals which theorists have made for how second languages should be taught. We will review research on second language learning which has been carried out in classroom settings. This will permit us to explore further the way in which second language research and theory contribute to our understanding of the advantages and the limita tions of different approaches to second language teaching. Before we go further, let us take a moment to reflect on the differences between natural and instructional language learning settings. We will then look at transcripts from two classrooms and try to understand what principles guide the teacher in each case. 2.2. Natural and instructional settings Natural acquisition contexts should be understood as those in which the learner is exposed to the language at work or in social interaction or, if the learner is a child, in a school situation where most of the other children are native speakers of the target language and where the instruction is directed toward native speakers rather than toward learners of the language. 1 Based on the book: Lightbown P., Spada N. How Languages are learnder Oxford University Press Oxford 1993 p.69-111

The traditional instruction environment is one where the language is being taught to a group of second or foreign language learners. In this case, the focus is on the language itself, rather than on information which is carried by the language. The teacher's goal is to see to it that students learn the vocabu lary and grammatical rules of the target language. The goal of learners in such courses is often to pass an examination rather than to use the language for daily communicative interaction. Communicative instruction environments also involve learners whose goal is learning the language itself, but the style of instruction places the emphasis on interaction, conversation, and language use, rather than on learning about the language. The topics which are discussed in the communicative instruction environment are often topics of general interest to the learner, for example, how to reply to a classified advertisement from a newspaper. Alternatively, the focus of a lesson may be on the subject matter, such as his tory or mathematics, which students are learning through the medium of the second language. In these classes, the focus may occasionally be on lan guage itself, but the emphasis is on using the language rather than on talking about it. The language which teachers use for teaching is not selected on the basis of teaching a specific feature of the language, but on teaching learners to use the language in a variety of contexts. Students' success in these courses is often measured in terms of their ability to 'get things done' in the second language, rather than on their accuracy in using certain grammatical features. In the chart below, mark a plus (+) if the characteristic in the left-hand col umn is typical of the learning environment in the three remaining columns. Mark a minus (-) if it is not something you usually find in that context. Write '?' if you are not sure. Table 1: Comparison of natural and instructional settings Characteristics Natural acquisition Traditional instruction Communicative instruction error correction

learning one thing at a time ample time available for learning high ratio of native speakers to learners variety of language and discourse types pressure to speak access to modified input As you look at the pattern of + and - signs you have placed in the chart, you will probably find it matches the following descriptions. In natural acquisition settings - Learners are rarely corrected. If their interlocutors can understand what they are saying, they do not remark on the correctness of the learners' speech. They would probably feel it was rude to do so. - Language is not structured step by step. In communicative interactions, the learner will be exposed to a wide variety of vocabulary and structures. - The learner is surrounded by the language for many hours each day. Some of it is addressed to the learner; much of it is simply 'overheard'. - The learner encounters a number of different people who use the target language proficiently. - The learner observes or participates in many different types of language events: brief greetings, commercial transactions, exchanges of informa tion, arguments, instructions at school or in the workplace. - Learners must often use their limited second language ability to respond to questions or get information. In these situations, the emphasis is on getting

meaning across clearly, and more proficient speakers tend to be tollerant of errors that do not interfere with meaning. - Modified input is available in many one-on-one conversations. In situ ations where many native speakers are involved in the conversation, however, the learner often has difficulty getting access to language he or she can understand. Learners in traditional instruction These differ from natural learners in that: - Errors are frequently corrected. Accuracy tends to be given priority over meaningful interaction. - Input is structurally simplified and sequenced. Linguistic items are pres ented and practised in isolation, one item at a time. - There is limited time for learning (usually only a few hours a week). - There is a small ratio of native speakers to non-native speakers. The teacher is often the only native or proficient speaker the student comes in contact with. - Students experience a limited range of language discourse types (often a chain of 'Teacher asks a question/Student answers/Teacher evaluates response'). - Students often feel great pressure to speak or write the second language and to do so correctly from the very beginning. - When teachers use the target language to give instructions or in other classroom management events, they often modify their language in order to ensure comprehension and compliance. Not all language classrooms are alike. The conditions for learning differ in terms of the physical environment, the age and motivation of the students, the amount of rime available for learning, and many other variables. Class rooms also differ in terms of the principles which guide teachers in their language teaching methods and techniques. The design of communicative language teaching programs has sought to replace some of the characteristics of traditional instruction with those more typical of natural acquisition contexts. Communicative language teaching classrooms Thus, in communicative language teaching classrooms we may find the fol lowing

characteristics: - There is a limited amount of error correction, and meaning is emphasized over form. - Input is simplified and made comprehensible by the use of contextual cues, props, and gestures, rather than through structural grading (the pre sentation of one grammatical item at a time, in a sequence of 'simple' to 'complex'). - Learners usually have only limited time for learning. Sometimes, how ever, subject-matter courses taught through the second language can add time for language learning. - Contact with proficient or native speakers of the language is limited. As with traditional instruction, it is often only the teacher who is a proficient speaker. In communicative classrooms, learners have considerable expos ure to the second language speech of other learners. This naturally contains errors which would not be heard in an environment where one's interlocutors are native speakers. - A variety of discourse types are introduced through stories, role playing, the use of 'real-life' materials such as newspapers and television broad casts, and field trips. - There is little pressure to perform at high levels of accuracy, and there is often a greater emphasis on comprehension than on production in the early stages of learning. - Modified input is a defining feature of this approach to instruction. The teacher in these classes makes every effort to speak to students in a level of language they can understand. In addition, other students speak a simpli fied language. 3.2 Classroom comparisons In this activity we are going to look at transcripts from two classrooms, one using a traditional audiolingual, structure-based approach to teaching, and the other a communicative approach. Audiolingual teaching is based on the behaviourist theory of learning which places emphasis on forming habits and practising grammatical structures in isolation. The communicative approach, in contrast, is

based on innatist and interactionist theories of language learning and emphasizes the communication of meaning. Grammatical forms are only focused on in order to clarify meaning. The theory is that learners can and must do the grammatical development on their own. With each transcript, there is a little grid for you to check off whether certain things are happening in the interaction, from the point of view of the teacher and of the students. Before you begin reading the transcripts, study the following definitions of the categories used in the grids: 1 Errors Are there errors in the language of either the teacher or the students? 2 Error correction When grammatical errors are made, are they corrected? By whom? 3 Genuine questions Do teachers and students ask questions to which they don't know the answer in advance? 4 Display questions Do teachers and students ask questions they know the answers to so that learners can display knowledge (or the lack of it)? 5 Negotiation of meaning Do the teachers and students work to under stand what the other speakers are saying? What efforts are made by teacher? By the students? T eacner/student interactions In the following excerpts, T represents the teacher; S represents a student. Classroom A: An audiolingual approach (Students in this class are 15-year-old Uzbek speakers.) Errors Teacher Student Feedback on errors Genuine questions Display questions Negotiation of

T OK, we finished the book - we finished in the book Unit 1, 2, 3. Finished Workbook 1, 2, 3. So today we're going to start with Unit 4. Don't take your books yet, don't take your books. In 1, 2, 3 we worked in what tense? What tense did we work on? OK? S Past T In the past—What auxiliary in the past? S Did T Did (writes on board '1-2-3 Past'). Unit 4, Unit 4, we're going to work in the present, present progressive, present continuous—OK? You don't know what it is? S Yes T Yes? What is it? S Little bit T A little bit S ... . T. Eh? S Uh, present continuous T Present continuous? What's that? S e-n-g T i-n-g S Yes T What does that mean, present continuous? You don't know? OK, fine. What are you doing, Mahmud? S Rien T Nothing? S Rien—nothing T You're not doing anything? You're doing something. S Not doing anything. T You're doing something. S Not doing anything. T You're doing something— Are, are you listening to me? Are you talk ing with

Manzura? What are you doing? S No, no—uh—listen—uh— T Eh? S to you T You're you're listening to me. S Yes T Oh—(writes 'What are you doing? I'm listening to you' on the board) S Je- T What are you—? You're excited. S Yes T You're playing with your eraser—(writes 'I'm playing with my eraser' on the board). Would you close the door please, Bernard? Claude, what is he doing? S Close the door T He is closing the door, (writes 'He's closing the door' on the board) What are you doing, Khamid? S I listen to you. T You're listening to me. S Yes T OK. Are you sleeping or are you listening to me? S I don't – firty-fifty, half and half. T Half and half, half sleeping, half listening. Classroom B: A communicative approach (Students in this class are 10-year-old Native language speakers. In this activity, they are telling their teacher and their classmates what 'bugs' them. They have written 'what bugs them' on a card or paper which they hold while speaking.) Errors Teacher Student Feedback on errors Genuine questions

Display questions Negotiation of meaning S It bugs me when a bee string me. T Oh, when a bee stings me. S Stings me. T Do you get stung often? Does that happen often? The bee stinging many times? S Yeah. T Often? (Teacher turns to students who aren't paying attention) OK. Salima and Bakhrom, you may begin working on a research pro ject, hey? (Teacher turns her attention back to 'What bugs me') S It bugs me (inaudible) and my sister put on my clothes. T Ah! She—borrows your clothes? When you're older, you may ap preciate it because you can switch clothes, maybe. (Teacher turns to check another student's written work) Mahliyo, this is yours, I will check.—OK. It's good. S It bugs me when I'm sick and my brother doesn't help me— my—my brother, 'cause he—me— T OK. You know—when (inaudible) sick, you're sick at home in bed and you say, oh, to your brother or your sister: 'Would you please get me a drink of water?'—'Ah! Drop dead!' you know, 'Go play in the traffic!' You know, it's not very nice. Doniyor! S It bug me to have— T It bugs me. It bugzz me S It bugs me when my brother takes my bicycle. Every day. T Every day? Ah! Doesn't your bro—(inaudible) his bicycle? Could his brother lend his bicycle? Uh, your brother doesn't have a bi cycle? S Yeah! A new bicycle (inaudible) bicycle. T Ah, well. Talk to your mom and dad about it. Maybe negotiate a new bicycle for your brother. S (inaudible)

T He has a new bicycle. But his brother needs a new one too. S Yes! T Hey, whoa, just a minute! Jean? S Martin's brother has— T Martin, who has a new bicycle? You or your brother? S My brother. T And you have an old one. S (inaudible) T And your brother takes your old one? S —clutch—(inaudible) bicycle T His bicycle! Ah! How old is your brother? S March 23. T His birthday? S Yeah! T And how old was he? S Fourteen. T Fourteen. Well, why don't you tell your brother that when he takes your bike you will take his bike. And he may have more scratches than he figures for. OK? Characteristics of input in the two classrooms Classroom A 1 Errors: Very few on the part of the teacher. However the teacher's speech does have some peculiar characteristics typical of this type of teaching, for example, the questions in statement form—often asked with dramatic ris ing intonation (for example, 'You don't know what it is?'). The students don't make many errors because they don't say very much. 2 Error correction: Yes, constantly from the teacher. 3 Genuine questions: Yes, a few, and they are almost always related to class room management. No questions from the students. 4 Display questions: Yes, almost all of the teacher's questions are of this type.

Interestingly, however, the students sometimes interpret display questions as genuine questions (T: What are you doing, Khamid? S: Nothing.) 5 Negotiation of meaning: Very little, learners have no need to paraphrase or request clarifications, and no opportunity to determine the direction of the discourse; the teacher is only focused on the formal aspects of the lear ners' language. Classroom B 1 Errors: Yes, when students speak but hardly ever when the teacher does. Nevertheless, the teacher's speech also contains incomplete sentences, simplified ways of speaking, and an informal speech style. 2 Error correction: Yes, sometimes the teacher repeats what the student has said with the correct form (for example, 'he bugjszme'—pointing out the third person singular). However, this correction is not consistent or in trusive as intrustive as the focus is primarily on letting students express their meanings. 3 Genuine questions: Yes, almost all of the teacher's questions are focused on getting information from the students. The students are not asking questions in this exchange. 4 Display questions: No, because there is a focus on meaning rather than on accuracy in grammatical form. 5 Negotiation of meaning: Yes, from the teacher's side, especially in the long exchange about who has a bicycle! Summary of the two classroom excerpts You have no doubt noticed how strikingly different these transcripts from the two classrooms are, even though the activities are both teacher-centred. In the transcript from Classroom A, the focus is on form (i.e. grammar) and in Classroom B, it is on meaning. In Classroom A, the only purpose of the interaction is to practise the present continuous. Although the teacher uses real classroom events and some humour to accomplish this, there is no doubt about what really matters here. There is no real interest in what stu dents 'are doing', but rather in their ability to say it. There is a primary focus on correct grammar, display questions, and error

correction in the transcript from Classroom A. In the transcript from Classroom B, the primary focus is on meaning, con - versational interaction, and genuine questions, although there are some brief references to grammatical accuracy when the teacher feels it is necessary. 4.2 Five principles for classroom teaching The teaching methodologies in Classrooms A and B differ because they reflect opposing theoretical views concerning the most effective way to learn a second language in classroom settings. Theories have been proposed for the best way to learn a second language in the classroom and teaching methods have been developed to implement them. But the only way to answer the question 'Which theoretical proposal holds the greatest promise for improving language learning in classroom set tings?' is through research which specifically investigates relationships between teaching and learning. Both formal and informal research are needed. Formal research involves careful control of the factors which may affect learning. It often uses large numbers of teachers and learners in order to try to limit the possibility that the unusual behaviour of one or two individuals might create a misleading impression about what one would expect in general. Researchers doing this kind of work must sometimes sacrifice naturalness in order to ensure that only those factors under investigation are different in the groups being compared. Informal research often involves small numbers, perhaps only one class with one teacher, and the emphasis here is not on what is most general but rather on what is particular about this group or this teacher. While formal research may add strength to theoretical proposals, informal research, including that carried out by teachers in their own classrooms, is also essential. It is hardly necessary to tell experienced teachers that what 'works' in one context may fail in another. In the section below, we will examine five proposals relating to this issue, provide

examples from classroom interaction to illustrate how the proposals get translated into classroom practice, and discuss how the findings from some of the formal research in SLA fit them. For each proposal, a few relev ant studies will be presented, discussed, and compared with one another. The labels we have given these proposals are: 1 Get it right from the beginning 2 Say what you mean and mean what you say 3 Just listen 4 Teach what is teachable 5 Get it right in the end 5.2. The principle getting right from the beginning The 'Get it right from the beginning' proposal for second language teaching best describes the underlying theory behind the teaching practices observed in Classroom A. Indeed, it is the proposal which probably best de scribes the way in which most of us were taught a second language in school. It reflects the behaviourist view of language acquisition in assuming that learners need to build up their language knowledge gradually by practising only correct forms. Teachers avoid letting beginning learners speak freely because this would allow them to make errors. The errors, it is said, could become habits. So it is better to prevent these bad habits before they happen. Here are some more examples from classes based on this approach. Example 1 (The teacher and students from Classroom A. This time the exercise in based on the simple present of English verbs.) S1 And uh, in the afternoon, uh, I come home and uh, uh, I uh, wash ing my dog. T I wash. S1 My dog. T Every day you wash your dog? S1 No. S2 He doesn't have a dog!

S1 No, but we can say it! Clearly, in this case, the student's real experience with his dog (or even the fact that he did or did not have a dog) was irrelevant. What mattered was the correct use of the simple present verb. Example 2 (A group of 12-year-old learners of English as a foreign language.) T Repeat after me. Is there any butter in the refrigerator? Group Is there any butter in the refrigerator? T There's very little, Mom. Group There's very little, Mom. T Are there any tomatoes in the refrigerator? Group Are there any tomatoes in the refrigerator? T There are very few, Mom. Group There are very few, Mom. (etc.) Pure repetition. The students have no reason to get involved or to think about what they are saying. Indeed, some students who have no idea what the sentences mean will successfully repeat them anyway, while their minds wander off to other things. Research findings There is little classroom research to support this proposal. In fact, it was the frequent failure of traditional grammar-based methods to produce fluency and accuracy in second language learners which led to the development of more communicative approaches to teaching in the first place. Supporters of communicative language teaching have argued that language is not learned by the gradual accumulation of one item after another. They suggest that errors are a natural and valuable part of the language learning process. Furthermore, they believe that the motivation of learners is often stifled by an insistence on correctness in the earliest stages of second language learning. These opponents of the 'Get it right from the beginning' proposal argue that it is better to encourage learners to develop 'fluency' before 'accuracy'. Recently, some researchers and educators have reacted to the trend toward

communicative language teaching and have revived the concern that allowing learners too much 'freedom' without correction and explicit instruction will lead to early fossilization of errors. Once again we hear the call for making sure learners 'get it right from the beginning'. Unfortunately, little research has been carried out to test the hypothesis that an early and exclusive emphasis on form will, in the long run, lead to higher levels of linguistic performance and knowledge than an early and exclusive emphasis on meaning. The widespread adoption of communicative language teaching in recent years has meant that researchers in some settings have not been able to find classrooms which are exclusively form-oriented in order to make direct comparisons with classrooms that are exclusively meaning-oriented. None the less, there are findings from second language classroom research which are relevant to this issue. These include descriptive studies of the interlanguage development of second language learners in audiolingual programs (Study 1), and studies of the development of second language proficiency in classroom learners who have received different amounts of form- and meaning-based instruction (Studies 2 and 3). Study 1: Audiolingual pattern drill In the late 1970s, Patsy Lightbown and her colleagues in Quebec, Canada, carried out a series of longitudinal and cross-sectional investigations into the effect of audiolingual instruction on the second language interlanguage development of francophone ESL learners, aged eleven to sixteen 1 (Lightbown 1983, 1987). Students in these programs typically participated in the types of rote repetition and pattern practice drill we saw in Classroom A. The researchers compared aspects of the learners' acquisition of English grammatical morphemes (such as plural –s and the progressive -ing) with the 'natural' order of acquisition by uninstructed second language learners. The results indicated several differences between the 'natural order' and the order in which these classroom learners produced them. The findings also suggested that the type 1 See: J. Glidden, J. John, and C. Therien Comprehension-based Second Language. Univ. of Ottawa Press, pp. 353- 70.

of instruction provided, a regular diet of isolated pattern practice drills, contributed to the alterations in the learners' natural interlanguage development. For example, while learners were able to produce a particular form (for example, the -ing form) with a high degree of accuracy during the time that their instruction focused on it, the same form was produced with considerably less accuracy (and frequency) when it was no longer being practised in class. These findings provided evidence that an exclusive emphasis on accuracy and practice of particular grammatical forms does not mean that learners will be able to use the forms. Not surprisingly, this type of instruction did not seem to favour the development of fluency and communicative abilities either. Study 2: Grammar plus communicative practice Sandra Savignon 1 (1972) studied the linguistic and communicative skills of 48 college students enrolled in Native language language courses at an American university. The students were divided into three groups, all of which received the same number of hours per week of audiolingual instruction where the focus was on the practice and manipulation of grammatical forms. However, the 'communicative group' had an additional hour per week devoted to communicative tasks in an effort to encourage practice in using Native language in meaningful, creative, and spontaneous ways; the 'cultural group' had an additional hour devoted to activities, conducted in English, which were designed to 'foster an awareness of the Native language language and culture through films, music and art'; and the control group had an additional hour in the language laboratory doing grammar and pronunciation drills similar to those which they did in their regular class periods. Tests to measure learners' linguistic and communicative abilities were administered before and after instruction to see if there were any significant differences between groups on these measures. The tests of 'linguistic competence' included a variety of grammar tests, teachers' evaluations of speaking skills, and course grades. The tests of'communicative competence' included measures of fluency and of the ability to understand and transmit information in a variety of 1 See: Savignon, S. 1972. Communicative Competence: An Experiment in Foreign-language Teaching. Philadelphia, Pa

tasks, which included: (1) discussion with a native speaker of Native language, (2) interviewing a native speaker of Native language, (3) the reporting of facts about oneself or one's recent activities, and (4) a description of ongoing activities. The results revealed no significant differences between groups on the lin guistic competence measures. However, the 'communicative group' scored significantly higher than the other two groups on the four communicative tests developed for the study. Savignon interprets these results as support for the argument that second language programs which focus only on accuracy and form do not give students sufficient opportunity to develop communicative abilities in a second language. Study 3: Grammar plus communicative practice In a similar study, Carol Montgomery and Miriam Eisenstein (1985) followed a group of adult learners receiving an additional communicative component to their regular, grammar-based instruction 1 . This group was compared to a control group which received only the grammar course. The researchers reported that beginner and intermediate level ESL learners engaging in communicative activities in addition to their regular, required grammar course made greater improvements in accent, vocabulary, grammar, and comprehension than did learners who received only the required grammar course. Somewhat unexpectedly, the area of greatest improvement for the group getting 'real world' communicative practice was in grammatical accuracy. Interpreting the research The studies reviewed above provide evidence to support the intuitions of teachers and learners that the 'Get it right from the beginning' proposal is not a very effective way to provide second language instruction. Learners receiving audiolingual instruction or more traditional grammar-based approaches have not benefited from this instruction in a way that permits them to communicate their messages and intentions effectively in a second language. Experience has also shown that primarily or exclusively grammar-based approaches to teaching do not guarantee that learners develop high levels of accuracy and linguistic knowledge. 1 Borrowed from: Montgomery, C. and M. Eisenstein. 1985. 'Reality revisited: An experimental communicative course in ESL.' TESOL Quarterly 19: 317—34.

In fact, it is often very difficult to determine what such learners know about the target language; the classroom emphasis on accuracy usually results in learners who are inhibited and will not 'take chances' in using their knowledge for communication. The results from these studies support the claim that learners require opportunities for communicative practice. It is important to emphasize that in the Savignon and the Montgomery and Eisenstein studies, all subjects received their regular, grammar-focused instruction and differed only in terms of the presence or absence of an additional communicative practice component. In other words, these studies offer support for the hypothesis that meaning-based instruction is advantageous, not that form-based instruction is not. The contributions of communicative practice and grammar- focused instruction will be discussed in more detail in relationship to the 'Teach what is teachable' and 'Get it right in the end' proposals. 6.2 The principle of saying what you mean and meaning what you say This is the theoretical view underlying the teacher-student behaviour in the transcript from Classroom B. Based on the interactionists' hypothesis, advocates of'Say what you mean and mean what you say' emphasize the necessity for learners to have access to meaningful and comprehensible input through conversational interactions with teachers and other students. They have argued that when learners are given the opportunity to engage in conversations, they are compelled to 'negotiate meaning', that is, to express and clarify their intentions, thoughts, opinions, etc., in a way which permits them to arrive at a mutual understanding. The negotiation, in turn, leads learners to acquire the language forms—the words and the grammatical structures—which carry the meaning. Negotiation of meaning is accomplished through a variety of modifications which naturally arise in conversational interaction. For example, learners will ask each other or their teacher for clarification, confirmation, repetition, and other kinds of information as they attempt to negotiate meaning. This can be seen in the

transcripts from Classroom B. The claim is that as learners, in interaction with other learners and teachers, work toward a mutual understanding in the negotiation process, language acquisition is facilitated. Advocates of interactionism argue quite simply that learners will learn by 'saying what they mean and meaning what they say' in conversations which encourage them to do so. Look for cases of negotiation for meaning in the examples below and com pare this with the examples given for the 'Get it right from the beginning' proposal. Example 3 (The teacher and students from Classroom B. Students are checking answers on a written task.) S Me and Josee, we don't have the same as her. T That's fine. Yeah, because there'll be different answers. S Why... uh, we do that with a partner? T Simply so you can consult. (In Examples 4, 5, and 6, a group of 12-year-old students are discussing with their teacher a questionnaire about their pets.) Example 4 S The fish is difficult to wash? T Fish is difficult to wash? S Yes. T Fish... Oh, not so difficult. Fish are difficult to wash?!? What's your uh... [question]? S Do you have an animal? Yes, I do. Do you ever feed it? Yes, r— T Do you know what 'feed' means? S Ah, no. It's uh...? T To give food to it. Example 5 T How often do you walk your dog? S Never. T Why? S Because I don't have a dog.

Example 6 S And what is 'feed'—? T Feed? To feed the dog? S Yes, but when I don't have a ... T If you don't have a dog, you skip the question. Example 7 (Students from Classroom B, doing a morning warm-up activity.) T How are you doing this morning? S1 I'm mad! S2 Why? T Oh boy. Yeah, wKy? S1 Because this morning, my father say no have job this morning— T Your father has no more job this morning? Or you have no job? S1 My father. How different these examples are from the essentially meaningless interaction often observed in classrooms where communication and form-focus are separated from each other. Such genuine exchanges of information must surely enhance students' motivation to participate in language learning activities. Research findings There have been no studies which have directly examined the effects of either the number or type of interaction opportunities on second language acquisition. Most of the research has been descriptive in nature, focusing on such issues as: How does negotiation which takes place in classrooms differ from that observed in natural settings? Do task types contribute to different kinds of interactional modifications? How does teacher- versus student-centred instruction contribute to differences in classroom interaction? Some research has examined relationships between modifications in conversational interaction and comprehension. Here are a few studies relevant to the interactionist proposal. Study 4: Group work and learner language One of the earliest studies to measure the different types of interaction patterns in

second language settings was carried out by Michael Long and his colleagues (1976). In their study, differences in the quantity and quality of student language in group work versus teacher-centred activities were investigated. They found that the students produced not only a greater quantity but also a g 1 reater variety of speech in group work than in teacher-centred activities. Not surprisingly, in the teacher- centred activities, the students primarily responded to teachers' questions and rarely initiated speech on their own. In contrast, learner language in group work activity was filled with questions and responses and many more occasions where learners took the initiative to speak spontaneously. In addition, the learner-centred activities led to a much greater variety of language functions (for example, disagreeing, hypothesizing, requesting, clarifying, and defining). Although this study was small, involving only two pairs of learners and two 40- minute lessons, it was one of the first studies to suggest how opportun ities for more group work interaction may be beneficial for second language learning. Study 5: Learners talking to learners Patricia Porter examined the language produced by adult learners per forming a task in pairs. There were eighteen subjects in the study: twelve non-native speakers of English whose first language was Spanish, and six native English speakers. The non-native speakers were intermediate or advanced learners of English. Each subject was asked to participate in separate discussions with a speaker from each of the three levels. For example, an intermediate-level speaker had a conversation with another intermediate-level speaker, with an advanced-level speaker, and with a native speaker of English. The investigator wanted to compare the speech of native and non-native speakers in conversations as well as to compare differences across proficiency levels in these conversation pairs. Learners talked more with other learners than they did with native speakers. Also, learners produced more talk with advanced-level than with intermediate-level partners, partly because the conversations with advanced learners lasted longer. 1 See: Long, M. H., L. Adams, M. McLean, and F. Castanos. 1976. 'Doing things with words—verbal interaction in lockstep and small group classroom situ ations' TESOL 76. Washington, D .C. PP . 137-53.

Porter examined the number of grammatical and vocabulary errors and false starts and found that learner speech showed no differences across contexts. That is, intermediate-level learners did not make any more errors with another intermediate-level speaker than they did with an advanced or native speaker. This is a particularly interesting finding because it calls into question the argument that learners need to be exposed to a native-speaking model (i.e. teacher) at all times if we are to ensure that they produce fewer errors. Overall, Porter concluded that although learners cannot provide each other with the accurate grammatical input that native speakers can, learners can offer each other genuine communicative practice which includes negotiation of meaning. Supporters of the 'Say what you mean and mean what you say' proposal argue that it is precisely this negotiation of meaning which is essential for language acquisition 1 . Study 6: Interaction and comprehensibility In one of the few studies which has directly investigated the effects of different input conditions on comprehension, Teresa Pica, Richard Young, and Catherine Doughty (1987) found that modifications in interaction led to higher levels of comprehension than modifications in input 2 . In their study, the sixteen learners were asked to follow instructions and complete a task under either of two different conditions. In the first condition, the students listened to a script read by a native speaker. The script had been simplified in a number of ways to facilitate comprehension. For example, there were repetition and paraphrasing, simple grammatical constructions and vocabulary, and so on. In the second condition, the learners listened to a script which contained the same information, but which had not been simplified in any way. Instead, as learners listened to the script being read, they were encour aged to ask questions and seek verbal assistance when they had any difficulty following the directions. 1 See Long, M. and P. Porter. 1985. 'Group work, interlanguage talk, and second language acquisition'. TESOL Quarterly 19: 207-28. 2 See: Pica, T. R. Young, C. Doughty. 1987. "The impact of interaction on comprehension'. TESOL Quarterly 21: 737-59.

The results indicated that learners who had the opportunity to ask clarification questions, and check their comprehension as they were listening to the instructions, comprehended much more than the students who received a simplified set of instructions to do the task but had no opportunity to interact while completing it. Study 7: Learner language and proficiency level George Yule and Doris Macdonald 1 (1990) investigated whether the role that different proficiency-level learners play in two-way communication tasks led to differences in their interactive behaviour. In order to do this they set up a task which required two learners to communicate information about the location of different buildings on a map and the route to get there. One learner, referred to as the 'sender', had a map with a delivery route on it and this speaker's job was to describe the delivery route to the other learner so that he or she could draw the delivery route on an incomplete map. To determine whether there would be any difference in the nature of the interactions according to the relative proficiency of the 40 adult participants, different types of learners were paired together: one group which consisted of high-proficiency learners in the 'sender' role and low-proficiency learners in the 'receiver' role, and another group with low-proficiency 'senders' paired with high- proficiency 'receivers'. The results showed that when low-proficiency learners were in the 'sender' role, the interactions were considerably longer and more varied than when high-proficiency learners were the 'senders'. The explanation provided for this was that high- proficiency 'senders' tended to act as if the lower-proficiency 'receiver' had very little importance and contribution to make in the completion of the task. As a result, the lower-proficiency 'receivers' were almost forced to play a very passive role and said very little in order to complete the task. When low-proficiency level learners were in the 'sender' role, however, much more negotiation of meaning and a greater variety of interactions between the two speakers took place. Based on these findings, the researchers argue that teachers should place more advanced 1 See: G. Yule, D. Macdonald. 1990. 'Resolving referential conflicts in L2 interaction: The effect of proficiency and interactive role. Language Learn ing: 539-56.

students in less dominant roles in paired activities with lower-proficiency-level learners. Interpreting the research The research described above (and other related research) investigating the factors which contribute to the quality and quantity of interactions between second language learners has provided some very useful information for teaching. Certainly, the early work of Long and his colleagues and the more recent findings of Porter and Yule and MacDonald have contributed to a better understanding of how to organize group and pair work more effectively in the classroom. 1 As indicated above, the difficulty with this line of research is that it is based on the not yet fully tested assumption that specific kinds of interactive behaviours lead to more successful second language acquisition. Although the Pica, Young, and Doughty study is important in this regard because it is one of the first to provide support for the claim that specific types of interactive behaviours lead to greater comprehension, more research is needed to directly test the hypothesis that better comprehension leads to more successful acquisition. 2 7.2 The principle of listening This proposal is based on the assumption that it is not necessary to drill and memorize language forms in order to learn them. However, unlike the interactionists' emphasis on providing opportunities for interaction of the kind we saw in some of the excerpts in the 'Say what you mean and mean what you say' proposal, the emphasis here is on providing comprehensible input through listening and/or reading activities. Read the classroom example below to get a feel for how this theory of class room second language learning can be implemented in classroom practice. Example 8 It is the English period at a primary school in a Native language-speaking area of 1 See: G. Yule, D. Macdonald. 1990. 'Resolving referential conflicts in L2 interaction: The effect of proficiency and interactive role. Language Learn ing: 539-56. 2 For futher reading see: Pica, T. R. Young, C. Doughty. 1987. "The impact of interaction on comprehension'. pp. 737-59.

New Brunswick, Canada. Students (aged nine to ten) enter the classroom, which looks very much like a miniature language lab, with small carrels arranged around the perimeter of the room. They go to the shelves containing books and audio- cassettes and select the material which they wish to read and listen to during the next 30 minutes. For some of the time the teacher is walking around the classroom, checking that the machines are running smoothly. She does not interact with the students concerning what they are doing. Some of the students are listening with closed eyes; others read actively, pro nouncing the words silently. The classroom is almost silent except for the sound of tapes being inserted or removed or chairs scraping as students go to the shelves to select new tapes and books. Just listen' is one of the most influential—and most controversial— approaches to second language teaching because it not only holds that second language learners need not drill and practise language in order to learn it, but also that they do not need to speak at all, except to get other people to speak to them. According to this view, it is enough to hear and understand the target language. And, as you saw in the classroom description above, one way to do this is to provide learners with a steady diet of listening and reading comprehension activities with no (or very few) opportunities to speak or interact with the teacher or other learners in the classroom. The material which the students read and listen to is not graded in any rigid way according to a sequence of linguistic simplicity. Rather, the program planners grade materials on the basis of what they consider intuitively to be at an appropriate level for the different groups of learners, because a given text has shorter sentences, clearer illustrations, or is based on a theme or topic that is familiar to the learners. The individual whose name is most closely associated with this proposal is Stephen Krashen, particularly with his hypothesis that the crucial requirement for second language acquisition is the availability of comprehensible input. Research findings Several studies which are relevant to this proposal include: (1) research in

experimental comprehension-based ESI. programs in Canada; (2) research investigating the effects of the 'Total physical response' method of second language teaching; and (3) research in Canadian Native language immersion programs. Study 8: Comprehension-based instruction for children Example 8 was a description of a real program which was developed in experimental classes in a Native language-speaking region in Canada. From the begin ning of their instruction in grade 3 (age eight years), these francophone students only listen and read during their daily 30-minute ESL period. There is no oral practice or interaction in English at all. Teachers do not 'teach' but provide organizational and technical support. Thus, learners re ceive a steady diet of native- speaker input but virtually no interaction with the teacher or other learners. Patsy Lightbown and Randall Halter 1 have investigated the second language development of hundreds of children in this program and have compared these findings with the second language development of those in the regular, aural-oral ESL program at the same grade level. Their results have revealed that learners in the comprehension-based program learn English as well as (and in some cases better than) learners in the regular program (Lightbown 1992). This is true not only for their comprehension skills but also for their speaking skills. This comes as something of a surprise since the learners in the innovative programs never practise spoken English in their classes. Study 9: Total physical response One of the best-known examples of the 'Just listen' proposal is the second language teaching approach called 'Total physical response' (TPR). In TPR classes, students —children or adults—participate in activities in which they hear a series of commands in the target language, for example: 'stand up', 'sit down', 'pick up the book', 'put the book on the table', 'walk to the door'. For a substantial number of hours of instruction, students are not required to say anything. They simply listen and show their comprehension by their actions. This instruction differs from the 1 See: Harley, B. and M. Swain. 1984. 'The interlanguage of immersion students and its implications for second language teaching' in A. Davies, C. Griper, and A. Howatt (eds.): Interlanguage. Edinburgh: Edinburgh University Press, pp. 291-311.

comprehension-based instruction described in Study 8 and from Krashen's theoretical version of' 'Just listen' in an important way: the vocabulary and structures which learners are exposed to are carefully graded and organized so that learners deal with material which gradually increases in complexity and each new lesson builds on the ones before. TPR was developed by James Asher, whose research has shown that students can develop quite advanced levels of comprehension in the language without engaging in oral practice (Asher 1972) 1 . When students begin to speak, they take over the role of the teacher and give commands as well as following them. It is clear that there are limitations on the kind of language students can learn in such an environment. Nevertheless, the evidence seems to show that, for beginners, this kind of active involvement gives learners a good start. It allows them to build up a considerable knowledge of the language without feeling the nervousness that often accompanies the first attempts to speak the new language. Study 10: Native language immersion programs 2 in Canada Other research which is often cited as relevant to the 'Just listen' proposal comes from Canadian Native language immersion programs, which have been described by Krashen as communicative language teaching 'par excellence'. The reason for this is that the focus in Native language immersion is on meaning through subject- matter instruction and the provision of rich, comprehensible input. In many ways, Krashen could not have asked for a better laboratory to test his theory. What have the studies shown? First, there is little doubt that the overall findings provide convincing evidence that these programs are among the most successful large-scale second language programs in existence. Learners develop fluency, functional abilities, and confidence in using their second language. There is, however, a growing awareness that Native language immersion learners still fail to achieve high levels of 1 See: Asher, J. 1972. 'Children's first language as a model for second language learning.' Modern Language Journal '56: pp. 133-9 2 Borrowed from: Krashen, S. 1982. Principles and Practice in Second Language Acquisition. Oxford: Pergamon. P. 23

performance in some aspects of Native language grammar even after several years in these programs 1 (Harley and Swain 1984). There are several possible explanations for this. Some researchers believe that the learners engage in too little language production because the classes are largely teacher-centred and students are not required to give extended answers (Swain 1985). This permits students to operate successfully with their incomplete knowledge of the language because they are rarely pushed to be more precise or more accurate. Communication between students and between teacher and students is quite satisfactory in spite of numerous errors in the students' speech. Other observers have suggested that the students need more form-focused instruction. This is based partly on experimental studies in which the addition of form-focused instruction has been shown to benefit learners 2 . It has also been observed that certain linguistic features rarely or never appear in the language of the teacher or the students in these content-based instructional environments. Furthermore, the presence in the classroom of other learners whose interlanguages are influenced by the same first language, the same learning environment, and the same limited contact with the target language outside the classroom, make it difficult for an individual learner to work out how his or her own use of the language differs from the target language. Interpreting the research The results of the Native language immersion research confirm the importance of comprehensible input in that the students develop not only good compre hension (in reading and listening), but also confidence and fluency in Native language. However, research does not support the argument that an exclusive focus on meaning and comprehensible input is enough to bring learners to mastery levels of performance in their second language. Indeed, the fact that Native language 1 See: Harley, B. and M. Swain. 1984. 'The interlanguage of immersion students and its implications for second language teaching' in A. Davies, C. Griper, and A. Howatt (eds.): Interlanguage. Edinburgh: Edinburgh University Press, pp. 291 2 See: Stu dies 14-17 under the 'Get it right in the end' proposal, pages 97-102.

immersion learners continue to make the same linguistic errors after years of exposure to the second language in classrooms which provide a great deal of comprehensible input is a challenge to the claim that language will take care of itself as long as meaningful comprehensible input is provided. The results of the research on comprehension-based ESL also appear to pro vide support for Krashen's comprehensible input hypothesis. It is important to keep in mind, however, that the learners in the comprehension-based studies are beginner- level learners and it is far too early to know how their second language skills will continue to develop. It is certainly possible (indeed probable) that learners in comprehension-based programs, like the Native language immersion learners, will have considerable gaps in their linguistic knowledge and performance over time. And, like the Native language immersion learners, they too will probably need and benefit from opportunities to use the language interactively as well as from some careful form-focused intervention later in their development. The TPR results also show great benefits for learners in the early stages of development. Krashen says of TPR that it prepares learners to go out into the target language community to get more comprehensible input which, he says, will carry their language acquisition further. In summary, comprehension-based programs appear to be beneficial in the development of basic comprehension and communicative performance in the early stages of learning (particularly in situations where learners have no other contact with the target language apart from in classroom situations). But they may not be sufficient in getting learners to continue to develop their second language abilities to advanced levels. 8.2 Teach what is teacheable The proposal referred to as 'Teach what is teachable' is one which has received increasing attention in second language acquisition research in recent years. The researcher most closely associated with this view is Manfred Pienemann. He and

his associates are concerned with being able to explain why it often seems that some things can be taught successfully whereas other things, even after extensive or intensive teaching, seem to remain unac-quired. They claim that their research provides evidence that some linguistic structures, for example, basic sentence word order (both simple and complex) develops along a particular developmental path. Thus, for example, any attempt to teach a word order pattern that is a 'Stage 4' pattern to learners at 'Stage 1' will not work because learners have to pass through 'Stage 2' and get to 'Stage 3' before they are ready to acquire what is at 'Stage 4'. The underlying cause of the stages has not been fully explained, but there has been considerable research showing that they may be based at least in part on learners' developing ability to process (unconsciously analyse and or ganize) certain elements in the stream of speech they hear. Researchers supporting this view also claim that certain other aspects of lan guage —vocabulary, some grammatical features—can be taught at any time. A learner's success in learning these variational features will depend on factors such as motivation, intelligence, and the quality of instruction. While this line of research has the potential to inform classroom teachers about which aspects of language acquisition are 'developmental' (and thus teachable only in a given sequence) and which are Variational' (and thus teachable at various points in learner language development), there is much work to be done before the findings of this research can lead to recommendations about whether particular forms can be taught and when. In Examples 9 and 10 below, we see a teacher trying to help students with question formation. The students seem to know what they mean, but the level of language the teacher is offering them is beyond their current stage of development. The students react by simply answering the question or accepting the teacher's formulation. Example 9 (A group of twelve-year-old students, interviewing each other as they play the roles of imaginary people.)

S1What's your nationality? S2 I am Russian. S1 What old, um, do you, uh, have—? T 'How old' dear. 'How old' were you—? S1 How old do you have... No, never mind. T How old were you when you came here? S1 Uh,yeah. Example 10 (The same group of students, asking fellow students questions about award poster which they had recently received.) S1 Mavluda, where you put your 'Kid of the Week' poster? T Where didyou put your poster when you got it? S2 In my room. (2 minutes later) S3 Mashhura where you put your 'Kid of the Week' poster? T Where did you put your poster? S4 My poster was on my wall and it fell down. In Example 11 below, the student is using a 'fronting' strategy which is typical of Stage 3 learners. That is, the student simply places an auxiliary verl (in this case 'is') at the beginning of the sentence but does not change the res of the sentence. (Note that if the student had fronted 'does', the sentencl would have been correct, but we would not have been able to see how the student thought question formation worked.) In this case, the teacher's correction leads the student to produce a Stage 4 question. In Example 12, same situation appears. This time, however, the correction leads not to reformulation of the question, but simply to an answer. Example 11 (Examples 11, 12, 13, and 14 are from a group of twelve-year-old Uzbek speakers learning English as a foreign language.) ('Famous person' interviews) S1 Is your mother play piano? T 'Is your mother play piano?' OK. Well, can you say 'Is your mother play piano?'

or 'Is your mother a piano player?' S1 'Is your mother a piano player?' S2 No. Example 12 (interviewing each other about house preferences) S1 Is your favourite house is a split-level? S2 Yes. T You're saying 'is' two times dear. 'Is your favourite house a split-level?' S1 A split-level. T OK. Example 13 ('Hide and seek' game) S Where the teacher books are? T Where are the teacher's books? S Where are the tea—the teacher books? Here the student asks a Stage 3 question, the teacher provides a Stage 4 correction, and the student is able to make the change. Note, however, that the student still doesn't change the possessive 's, something which Uzbek speakers find very difficult. Research findings The 'Teach what is teachable' view is one which claims that while some features of the language can be taught successfully at various points in the learner's development, other features develop according to the learner's internal schedule and that no amount of instruction can change the 'natural' developmental course. Let us examine a few of the studies which have tested this hypothesis. Study 11: Ready to learn In a study of the acquisition of German as a second language, Manfred Pienemann 1 (1988) investigated whether instruction permitted learners to 'skip' a stage in the natural sequence of development. Two groups of learners who were at Stage 2 in 1 See: Pienemann, M. 1985. 'Learnability and syllabus construction' in K. Hyl-tenstam and M. Pienemann (eds.): Modelling and Assessing Second Language Acquisition. Clevedon, UK: Multilingual Matters, pp. 23-75.

their acquisition of German word order were taught the rules associated with Stage 3 and Stage 4 respectively. The instruction took place over two weeks and during this time, learners were provided with explicit grammatical rules and exercises for Stage 4 constructions. The results showed that the learners who received instruction on Stage 3 rules moved easily into this stage from Stage 2. However, those learners who received instruction on Stage 4 rules did not move into this stage. They either continued to use Stage 2 behaviours or they moved into Stage 3. That is, they were not able to 'skip' a stage in the 'natural route'. Pienemann interprets his results as support for the hypothesis that for some linguistic structures, learners cannot be taught what they are not 'developmentally ready' to learn. Study 12: Teaching when the time is right Catherine Doughty 1 (1991) examined whether particular aspects of relative clause formation would benefit from instruction at a time when learners were developmentally 'ready' to learn them. Twenty subjects were divided into three groups: two experimental and one control. All groups received exposure to relative clauses over a period often days through a series of computer-delivered reading lessons. During these lessons all learners were asked to read the passages and answer a variety of comprehension questions which focused on reading skills such as skimming and scanning. For the experimental groups, two instructional techniques were added to the reading comprehension exercises. These were presented to the learners by means of an additional 'window' on the learners’ computer screens. One experimental group received instruction which focused on meaning-orientated techniques. This included both vocabulary help and paraphrases of sentences in the reading comprehension texts. The other experimental group received instruction which focused on rules. This included instruction on relative clause formation through a combination of explicit grammatical rules and on-screen sentence manipulation. All learners were pre-tested immediately before the instructional treatment and 1 See: Doughty, C. 1991. 'Second language instruction does make a difference: Evidence from an empirical study of SL relativization.' Studies in Second Language Acquisition 13/4: pp. 431-69.

post-tested after the ten days of the exposure/instruction with regard to relative clauses. The results revealed a clear advantage for the experimental groups. That is, learners who had received the additional instruction in relative clause formation— regardless of whether it was meaning-orientated or rule-orientated outperformed the control group learners who had received only exposure to relative clauses through the reading comprehension texts. Doughty concludes that instruction on relative clauses made a difference when it was provided at the time when learners were 'developmentally ready' to learn them. Study 13: Can question forms be taught? Rod Ellis 1 (1984) studied the effects of instruction on the acquisition of ques tion forms by thirteen child ESL learners. In this study, learners were also given instruction at a time when they were considered to be 'develop-mentally ready' to acquire wh-question inversion rules. The learners received three hours of instruction. In the first hour the teacher asked a series of wh-questions while referring to a wall poster, and students were asked to respond. In the second hour, the students asked questions (again referring to the wall poster), and the teacher corrected them when they made errors. In the third hour, the teacher 'fired questions at the pupils' about the wall poster. The group results revealed little effect for instruction on the learners' development of question forms, although some individual learners did improve substantially. Interpreting the research The conflicting results of these studies present an obvious problem for assessing the 'Teach what is teachable' proposal. A closer look at some of the procedural problems in one of the studies should shed some light on these seemingly contradictory findings. If one compares the amount of instruction provided, it seems possible that the three hours provided in the Ellis study were not enough to cause changes in the learners' interlanguage systems. Further, there is the possibility that the type of instruction was not sufficiently form-focused. In the 1 See: Ellis, R. 1984. 'Can syntax be taught?' Applied Linguistics 5: 138-55.

limited description of the type of instruction provided in Ellis 1 ’ study, it seems that the learners had more exposure to w//-questions in the teacher's modelling than they did opportunities to produce questions themselves and to receive feedback on their errors, either through correction and/or explicit rule teaching. In this way, the group in Ellis' study may have been more similar to the control group in Doughty 2 's study—the one which received increased 'exposure' but not so much 'instruction' and in the end did not perform as well as those learners who received more focused instruction. It seems reasonable to conclude that because the instruction provided in the Doughty and Pienemann studies was more explicit, carefully controlled, and of a longer duration, their studies provide a more reliable test of the 'Teach what is teachable' proposal. Nonetheless, it is important to note some of the weaknesses in these studies as well. For example, in Doughty's study, no direct comparison was made between learners who were not'devel- opmentally ready' to learn relative clauses and those who were. Further, in both studies, only the short-term effects of instruction were measured. Because of this, there is no way of knowing whether instruction had any permanent or long-term effects on the learners' developing interlanguage systems. In Pienemann's study, results were reported for only a small group of learners. In later studies, however, similar results were reported with other learners. 9.2 Getting right in the end Get it right in the end' is similar to the 'Teach what is teachable' proposal. Its proponents recognize a role for instruction, but also assume that not everything has to be taught. That is, they assume that much will be acquired naturally, through the use of language for communication. They also agree that some things cannot be taught if the timing of the teaching fails to take the student's readiness (stage of 1 See: Ellis, R. 1984. 'Can syntax be taught?' Applied Linguistics 5: 140 2 See: Doughty, C. 1991 . 'Second language instruction does make a difference: Evidence from an empirical study of SL relativization.' Studies in Second Language Acquisition 13/4: p. 431

development) into account. This proposal differs from the 'Teach what is teachable' proposal, however, in that it emphasizes the idea that some aspects of language mustbe taught. For example, when an error learners make is the result of transfer from their first language, and when all the learners in a group tend to make the same error, it will be virtually impossible for learners to discover this error on their own. We can see this in Example 14, where francophone learners of English are having dif ficulties with adverb placement. 'Get it right in the end' also differs from 'Just listen' in that it is assumed that learners will need some guidance in learning some specific features of the target language. Furthermore, it is assumed that what learners learn when they are focusing on language itself can lead to changes in their interlanguage systems, not just to an appearance of change brought about by conscious attention to a few details of form. On the other hand, the supporters of this proposal do not claim that teaching particular language points will prevent learners from making errors. Nor do they assume that learners will be able to begin using a form or structure with complete accuracy as soon as it is taught. Furthermore, they do not argue that the focused teaching must be done in a way which involves explicit explanations of the point or that learners need to be able to explain why something is right or wrong. Rather, they claim that the learners' attention must be focused on the fact that their language use differs from that of a more proficient speaker. As we will see in the examples below, teachers must look for the right moment to create increased awareness on the part of the learner—ideally, at a time when the learner is motivated to say something and wants to say it as clearly and correctly as possible. Proponents of' 'Get it right in the end' argue that it is sometimes necessary to draw learners' attention to their errors and to focus on certain linguistic (vocabulary or grammar) points. The difference between this proposal and the 'Get it right from the beginning' proposal is that it acknowledges that it is appropriate for learners to engage in meaningful language use from the very beginning of their exposure to the second language. They assume that much of language acquisition will develop

$naturally out of such language use, without formal instruction which focuses on the language itself. The difference between this proposal and the 'Just listen' and 'Say what you mean and mean what you say' proposals is that it is not assumed that comprehensible input and meaningful interaction will be enough to bring learners to high levels of accuracy as well as fluency. Researchers who support this proposal argue that learners can benefit from, and sometimes require, explicit focus on the language. Example 14 (Examples 14, 15, and 16 are taken from a classroom where a group of twelve- year-olds are learning English. In Example 14, they are engaged in an activity where scrambled sentences are re-ordered to form sensible ones. The following sentence has been placed on the board: 'Sometimes my mother makes good cakes.') T Another place to put our adverb? S1 After makes\ T After makes. S2 Before good?. T My mother makes sometimes good cakes. S3 No. T No, we can't do that. It sounds yucky. S3 Yucky! T Disgusting. Horrible. Right? S4 Horrible! This is hardly a typical grammar lesson! And yet the students' attention is be ing drawn to an error virtually all of them (native speakers of Uzbek ) make in English. Example 15 (The students are practising following instructions; one student instructs, others colour.) S1 Make her shoes brown. T Now, her shoes. Are those Mom's shoes or Dad's shoes? S2 Mom's. T Mom's. How do you know it's Mom's?$