Lexis and Structure: Rules, Patterns and Words

Rules, Patterns and Words

Grammar and Lexis in English

Language Teaching

1       What is taught may not be what is learnt: Some preliminary questions

Whenever we do anything in the classroom we are acting on our beliefs about language and language learning. If we ask learners to listen and repeat a particular sentence, we are acting on the belief that such repetition is useful enough to justify the valuable classroom time it takes up, perhaps the belief that it helps rote learning which in turn promotes general language learning. If we give learners grammatical rules or encourage them to discover rules for themselves, we are acting on the belief that rules make a valuable contribution to language description and that this kind of understanding helps promote learning.

Our beliefs about language learning and teaching are shaped by our training, but also by our classroom experience. Unfortunately, learning from experience is not always easy. Teaching is such an absorbing business that it is difficult to stand back and ask appropriate questions about what is happening in the classroom.

My own experience as a language teacher – and also as a learner – suggests to me that learning a language is a much more complex and difficult process than we would like to think. We need to look very carefully at some of the assumptions we make about language learning and about language itself. A first step is to look at what happens in classrooms, and to identify some of the questions that need to be asked.

In the classroom teachers often act on the assumption that language learning is a matter of learning a series of patterns or structures. Learners gradually add to their stock of structures until they have a usable model of the language. They often start with the present tense of be, and soon they are exposed to the definite and indefinite articles. At a later stage we add the passive voice and reported speech, and continue until we reach the dizzy heights of the third conditional. The syllabus is presented to learners in a 'logical' order and the language is built up piece by piece until learners have achieved a usable competence, a form of the language which meets their needs.

As teachers, however, we observe that learning proceeds in a much less predictable manner. What is 'taught' is often not learnt, and learners often 'learn' things which have not been taught at all. Learners often produce sentences such as: I am student or My father is engineer even though they have never been taught this, and even though their conscientious teacher is at pains to point out that the definite article is required here: You are a student; Your father is an engineer. Often learners persist in these errors for a long time, in spite of repeated correction.

This is frustrating for both learners and teachers, but the full picture is even more complicated than this. Learners soon reach a stage at which they produce accurately: I am a student when they are thinking carefully about the language; but when they are producing language spontaneously, or when their attention is drawn to another feature of the language, they continue to produce: I am student. There are, it seems, two kinds of learning. One of them has to do with learning to make sentences. Learners think hard about what they are doing and produce thoughtful, accurate samples of the language. The second kind of learning has to do with learning to produce language spontaneously, without conscious attention to detail. What learners produce spontaneously is often very different from what they produce when they are concentrating on making sentences.

We come up against this phenomenon time and time again in our classroom practice. We constantly observe instances where learners make errors which they are easily able to correct once they are pointed out. And we also observe, time and time again, that the same errors are repeated, even after they have been pointed out. This is one of the central puzzles in language teaching: how is it that learners can know something, in the sense that they are well aware of it when they are making sentences carefully and attentively, but at the same time not know it when they are producing language spontaneously?

In this chapter I will look first at my own experience in a class on question tags: why is it that these tags, which are relatively easy to explain, are so difficult for learners to master? I will then look at question forms in general: why do learners go on getting these wrong for so long even after they have understood the rules for question formation? The way learners go about learning question forms raises questions about learning in general – I will highlight some of these questions and speculate on possible explanations.

1.1 Some questions about tags

My first teaching job was at a secondary school in Ghana, West Africa. My Ghanaian students, who did not share a common first language, were learning English as a second language. They had not acquired English as their first language at their mothers' knees. Most of them had their first contact with English in primary school, and by the time they reached secondary school nearly all of their lessons were taught through the medium of English. Their spoken English, however, was a dialect form which was very different from standard British English. They used this dialect not only in the classroom, but also when speaking to fellow students who came from another language group.

'Sensible' languages have a single form for question tags. French has n'est-ce pas?; Greek has δεν ειναι? (dhen eeneh?); Spanish uses verdad? or no? Unlike these sensible languages English has a wide range of question tags:

We've met before, haven't we?

You'll be there on time, won't you? They can do it, can't they?

But in the dialect of English used by my Ghanaian students there was only one tag, as in French and Greek:

We've met before, isn't it?

You'll be there on time, isn't it? They can do it, isn't it?

This tag is a form which is also often used by learners of English as a foreign language. It is even used by some native speakers of English – We'll see you tomorrow, innit?

Unfortunately my Ghanaian students were supposed to be learning standard British English. In their examinations they would be tested on standard British English – including the entire range of question tags. And, for some reason best known to themselves, examiners love to test question tags. I knew that my students would be tested in public examinations and that in those examinations, which in those days were in multiple-choice format, question tags would figure largely.

I was determined to eradicate their apparently serious error, and carefully prepared a lesson. This happened back in the 1960s, and, to someone trained in the 1990s, my lesson may have appeared to be oldfashioned in some respects, since it was based initially on grammatical explanation. It began with an explanation and demonstration showing how the auxiliary or modal verb was repeated in the tag, and how an affirmative clause had a negative tag. Then we looked at some sample sentences on the blackboard, until the students were able to supply tags consistently. I called out some statements and the students responded with the appropriate tag. I finished with one half of the class repeating a statement after me, and the other half of the class responding in chorus with the right tag.

We're learning English… aren't we?

We will have English next Monday… won't we? We have English every Monday… don't we?

It all went beautifully. I felt all the warm satisfaction of someone who has achieved his lesson aims. There was one final stage. I asked the students to take out their exercise books so that they could write down a few sample tags to help them remember what they had learned. They all looked a little sheepish. Finally one of them, one of the brightest students in the class, put up his hand and explained the problem: Please, sir, you've got our exercise books… isn't it? My beautifully prepared and highly successful lesson vanished before my eyes. What my students seemed to have learnt turned out not to have been learnt, even by one of the brightest.

Please, sir, you've got our exercise books, isn't it?

In one sense I had done my job. I am sure that, when faced with multiple-choice questions, and given time to think, most of my students would be able to identify the correct tags. But most of them never incorporated these tags into their spontaneous speech. I soon learned that almost all Ghanaians, including those who were fluent, even eloquent in English, used only the all-purpose tag isn't it? – even if they could reproduce the complex system used by speakers of standard British English when asked to do so.

At the time I was simply puzzled and frustrated. I had spent a lot of time teaching something which was difficult and had little practical value. I had taught it so that it could be tested and so that my students might respond appropriately in a test. But it had certainly not become a part of their usable repertoire of English.

1.2 Some questions about questions

We know from research into second language learning that learners have to go through a series of stages before they are able to produce question forms consistently and accurately. This is something that teachers know from bitter experience. It takes a long time, for example, before learners spontaneously produce questions with the 'dummy auxiliary' do, as in: What do you want? Even sentences which they hear over and over again are distorted. On teacher-training courses I refer to this as the 'Please, teacher, what mean X?-syndrome'. Learners may have been endlessly drilled in forms like What do you want? Where do you live? and so on. They will certainly have heard the phrase What does X mean? many, many times. But in class they consistently put up their hands and ask the question Please, teacher, what mean X?

In time, usually a long time, they get past this stage and begin to produce questions with do in the appropriate form, and the teacher breathes a sigh of relief at this evidence of real progress. But later we move on to reported questions: Do you know where they live? Tell me what you want. In these forms there is, of course, no dummy auxiliary do. Students are familiar with the forms … they live and … you want. There should be no real problem with putting these after a WH-word such as what or where to produce: Tell me what you want and Do you know where they live? But what happens? They regularly produce the forms: Do you know where do they live? Tell me what do you want. In a test on reported questions they may be able to produce the appropriate forms, but it takes some time, often a considerable time, before they eliminate the do auxiliary from their reported questions. This process is similar to that observed among L1 learners. The mastery of question forms might appear to be straightforward, but it involves a complex developmental process.

Why should this be the case? It may be that the forms What do …? What did …? and so on have become 'consolidated'. Once students have learned to use direct questions, then a WH-word like what or where automatically triggers an auxiliary, including the dummy auxiliary. What once came to them naturally – Where I live? What you want? – no longer comes naturally to them. The new forms – Tell me what you want; Do you know where they live? – are easily demonstrated, explained and understood, but they are not used spontaneously. To use them spontaneously it seems that learners first have to unlearn their old habits. They have to break the link between a WH-word and the auxiliary which they have acquired with such difficulty in the process of learning direct questions.

1.3 Some questions about learning

Some years ago, on an in-service teacher-training course, I asked teachers to make a list of the ten commonest mistakes made by learners. I asked one half of the group to list the most frequent errors in their first year classes, and the second group to list errors made in third year classes. When the lists were compared the teachers were horrified to see that seven of the mistakes they had listed occurred in both the first year and the third year. Third year students, like their first year counterparts, consistently produced forms like: She want … instead of: She wants … First and third year students seemed to have the same problems with articles, including the production of the forms: I am student and You are teacher, which I referred to above. Third year students still had problems with question forms, particularly the do- auxiliary, and so on.

This, of course, raised serious questions about what was happening in these classes. Had teachers really taken a full two years of teaching to eliminate only three mistakes? Were their third year students really not much better than their first year students? How could we account for this appalling failure?

Although the teachers accepted that they had been conspicuously unsuccessful in eradicating common errors, they still insisted that third year students had a much better command of English than first year students. They pointed out that third year students had a much wider vocabulary than the first years. They used English with greater fluency and confidence. Some of them were able to produce several consecutive sentences, albeit littered with errors. This was quite beyond their first year counterparts. The third years could understand and produce language that was quite beyond a first year student and, as part and parcel of this, they could make lots of mistakes that the first years could not even dream of.

The conclusion we reached was this: if it is the teacher's role to eliminate error, then these teachers had been remarkably unsuccessful – even though most of them were, by all reasonable standards, very good teachers. But if it is the teacher's role to help students develop enhanced performance and confidence, then all the teachers could claim genuine success. Their third year students spoke more English than their first year students, and they spoke it with greater fluency and confidence.

This, however, still left us looking for an explanation as to why the teachers' efforts to eliminate error had met with so little success. One teacher asked me if I had been any more successful in my days as a classroom teacher. Remembering my lesson on question tags, and countless other similar experiences, I had to admit that I had not. I had no simple answer to the question why some aspects of language are so resistant to teaching, and I certainly had no simple solution as to what might be done about this.

One possible explanation for this phenomenon is that learners are simply careless. They know that they should add s to the third person singular of the present simple tense, and they know how to form questions with the auxiliary do, but they are simply too careless to apply this knowledge when they are using the language spontaneously. But second language acquisition research, as well as our experience as teachers, tell us that these are stages that almost all learners go through. We can hardly dismiss all learners as careless. It seems much more likely that the processes we have described are a necessary part of learning, that learners have to go through a process which involves making mistakes before they can produce appropriate forms spontaneously and without conscious attention.

There is, then, plenty of evidence that learners do not move immediately from an understanding of new language forms to the spontaneous production of those forms. They go through a stage at which they can produce the form only when they are paying careful attention. They cannot produce the form when they are using language spontaneously, when they are thinking about getting meaning across rather than producing accurate sentences. In spontaneous language use there are conflicting priorities. The learners' main priority is to get their message across with appropriate speed and fluency; they may also be keen to produce language which is accurate – but speed and fluency conflict with accuracy.

1.4 Learning processes

It seems, then, that there is no direct and straightforward connection between teaching and learning. We cannot determine or predict what learners will make a part of their spontaneous language behaviour. However, our experience as teachers and the experience of the teachers in training reported above suggest that classroom instruction does help learners, and this is reinforced by second language acquisition research (see, for example, Long, 1983, 1988) which appears to show that learners develop more quickly and go on learning for longer if they are supported by instruction.

It is possible that teaching makes learners more aware of a particular form, it makes the form more noticeable. Until their attention is drawn to it, learners may not even notice the structure of do-questions. Perhaps they simply identify these forms as questions through their intonation patterns without paying attention to their form. Once the structure has been pointed out to them they begin to notice it when they come across it. Over time this repeated noticing enables them to incorporate the acceptable forms into their spontaneous language production. It is also possible that teaching helps learners form hypotheses about the language which they then go on to test and to refine. Yet another possibility is that classroom procedures encourage learners to think carefully about the language for themselves, and help to make them more independent learners.

It is worth looking at a number of processes which might contribute to learning, and following on from that we can go on to consider ways in which teachers might assist learning. Let us begin by postulating three language learning processes which I will refer to as Recognition, System building and Exploration. Let us look at these processes one by one.

Recognition: The first stage in learning probably involves recognising what it is that is to be learnt. Whether or not something is recognised is subject to a number of influences. It is subject, for example, to salience, how much it stands out from its background. This can be annoying for teachers, because strange and unusual words and phrases often stick in students' minds. On the other hand, syntactic markers, such as articles and auxiliary verbs, are far from salient. We need to draw attention to such items quite explicitly, and to encourage learners to look for them in future input.

Recognition takes place at a number of levels. We might, for example, encourage learners to recognise a general phenomenon, such as the behaviour of uncountable nouns in English, nouns which are not found in the plural nor with the indefinite article. We might do this at first by drawing attention to a number of frequently occurring nouns which refer to items of food and drink: bread, food, rice, water etc. Later we might go on to make the same point about other substances such as oil, gas, iron and wood. Once learners are aware that some nouns in English behave in this way they may immediately make links with similar nouns in their own language, and as a result go on to generalise that abstract ideas (beauty, bravery, death etc.) and activities (help, travel, sleep etc.) behave in the same way. If the learners' first language does not offer this kind of support, they may need more help with recognition. Even if their own language is similar to English in its general classification and treatment of uncountable nouns, the teacher might still usefully provide help with some very frequent nouns which are uncountable in English but not in most other languages, words like advice, furniture, homework and equipment.

Thus, teachers can help learners with recognition by explanation, by showing students how to recognise uncountable nouns. They can reinforce this by pointing out specific examples of these nouns as they occur in the language which learners experience in the classroom, and later by encouraging learners to identify these nouns for themselves. They can go on to exemplify and list uncountable nouns.

With some vocabulary items learning proceeds largely by recognition. If a word has an obvious referent in the outside world, it can be learnt as an individual item. I have an impressive restaurant vocabulary in Spanish even though my competence in Spanish conversation is very limited. I acquired my restaurant vocabulary mainly by studying restaurant menus and lists of words in a Spanish phrase book. As a result I can work my way through a menu and find what I want, even though I cannot engage a waiter in a productive discussion of how the food has been prepared. There are a number of lexical fields which lend themselves to this kind of learning, but we do need to be wary of rote learning. Even a simple word like foot can cause problems. For a speaker of Greek, for example, the word ποδι is the closest equivalent to foot, but ποδι refers not simply to the foot, but to the entire leg below the knee. This can occasionally cause problems for Greek learners of English as well as for English learners of Greek.

Depending on which is the student's first language, some grammatical items in English may also be assimilated without too much trouble once they have been recognised. Most European languages have words which are almost exact counterparts of the English direct and indirect articles, for example. Speakers of those languages can acquire the article system as if the, a and an were straightforward lexical items, without worrying about complex differences in use. For speakers of Greek, for example, the basic distinction is clear, but there are difficulties with proper names which in Greek always take a definite article. The way proper nouns are handled in English is inconsistent. In general we do not use the definite article with names but it is used with the names of seas and oceans, for example, although not with lakes. There is no logical reason why English should talk of Lake Geneva and Lake Superior, but insist on the Atlantic Ocean and the North Sea. French is similar to English in that it operates an inconsistent system, but the inconsistencies in French are different from those in English. For example French uses the definite article for the names of countries (la France, la Grande Bretagne), but not for towns or cities; it often uses the definite article for days of the week, but not for the months of the year. There is, therefore, a certain amount of 'tidying up' to do for all learners, but for many, including speakers of most European languages, the basic distinction between the definite and indefinite articles is straightforward, and the article system can be assimilated without too much difficulty.

Teachers can assist learners with recognition by providing lists of words organised into useful groups and by encouraging rote learning. They can identify grammatical systems which can usefully be transferred from the students' first language. As we have seen, one example for most European learners of English is the article system. In the same way, for French learners of English, the going to future can simply be transferred from the French.

System building: Language learning involves conscious processes which are familiar to all who have learnt a second language. Learners begin to form hypotheses about how grammatical systems work and teachers can help them do so. A good example is the relationship between continuous and simple tenses in English. In most elementary English courses learners begin by recognising the difference in meaning between the present simple and the present continuous. Without help and direction from the teacher it would be very difficult for learners to make the generalisation that the present simple is generally used for habitual actions or ongoing states:

I usually go to church on Sunday.

We live just outside Birmingham.

whereas the present continuous is generally used for something which is happening at the time of utterance:

Wait a minute, I'm listening to the radio. Dad's watching the football on TV.

Without further help from the teacher it is even more difficult for learners to recognise that the present continuous can also be used for habitual actions or ongoing states if these actions or states are regarded as temporary:

She's in her sixties but she's still playing tennis regularly. We are living in Selly Oak for the time being.

Teachers can provide useful rules of thumb to help learners work out the grammar, and they can support these rules with carefully chosen examples as well as by asking learners to find examples for themselves in the language they experience. They can supplement this by setting exercises which will require learners to apply the rules in order to produce language.

In the early stages of learning learners may practise routines which contribute to system building at a later stage. At the elementary level, for example, students may be introduced to a vocabulary building game which also incorporates insights into the use of the definite and indefinite articles in English. One such game, What's in the bag?, involves taking into the classroom a bag filled with objects that are familiar to the learners:

Teacher: What do you think I've got in my bag?

Student: A pencil.

Teacher: Yes, I've got a pencil. Here it is. (puts the pencil on her desk) Where is the pencil?

Student: It's on the desk?

Teacher: Good. It's on the desk. What else have I got in my bag? Student: A pen.

Teacher: Yes, I've got a pen … etc.

As well as building vocabulary this game provides exposure to a number of useful phrases: What have I got? I've got … What else?, and at least one useful pattern N + is + prepositional phrase. It also provides a number of possible insights into the use of the referential system in English: it introduces the indefinite article a(n); it illustrates the use of the pronoun it to refer back to something which has been introduced; it shows the use of the definite article to refer to something specific.

However, if it is learnt at all, it is learnt only as a routine and leaves many questions unanswered. The fact that the teacher says: It's a pencil rather than: It's the pencil may appear to contradict the 'rule' that the first mention of a noun uses the indefinite article, while subsequent mentions use either the definite article or a pronoun like it. Why is the pencil described as being on the desk, rather than a desk ? A command of routines such as these does not mean that students have mastered these elements of the system; it simply provides them with samples of language which they can perhaps draw on as the system develops.

Although we have discussed words on the one hand and grammar on the other, it is often quite impossible to separate the two. This will become apparent as soon as we look at some of the words in English which are associated with complex grammatical patterns. The word agreement is a good example. In fact there are two words for agreement:

there is a countable form of the word, which is found in sentences like:

We made an agreement to meet the following week.

while the uncountable agreement is found in sentences like:

We failed to reach agreement on the outstanding issues.

This uncountable agreement occurs in a number of fixed phrases such as in agreement or by agreement. In order to use this word effectively, a learner needs to know a good deal about the patterns in which it occurs. There are a number of collocational restrictions: we do not talk of doing an agreement; we normally reach or come to an agreement; we talk about general agreement or broad agreement, but not wide agreement. The word is also postmodified in particular ways: we talk about agreement on a particular issue, or agreement on a course of action; we frequently talk about agreement to do something; we say that there is general agreement that … . Before learners can make productive use of the word agreement they need to be aware of these patterns, and of common collocations and collocational restrictions.

Knowing the meaning of the word and its first language equivalent or equivalents is a matter of recognition, and this provides an important starting point. But if learners are to make the word a useful part of their vocabulary, recognition can only be the first stage in a more complex learning process which involves system building. System building related to the word agreement links the word to other nouns formed from verbs. We not only talk about an agreement to do something – the words decision, plan and arrangement are used in exactly the same way. So nouns denoting the outcome of negotiation or planning are followed by the to-infinitive. Similarly nouns related to reporting verbs are often followed by a that-clause – nouns such as belief, claim and suggestion. We talk about reaching or coming to agreement. We also talk about reaching or coming to an arrangement, a decision or a conclusion. So the behaviour of a word like agreement is systematic. Learners will begin to use the word quickly and effectively if they are able to link it systematically to other words in the language.

Exploration: A lot of learning takes place by exploration. As they are exposed to language, learners find things out for themselves and begin to develop systems without even being aware that they are doing so. Foreign language learning in a natural environment involves a lot of exploration. If we are living in a foreign language environment we begin to make sense of the language we hear, and to develop grammatical systems without even thinking about it. We produce language because it feels right. There are at least two good reasons why discovery is an important and a necessary process, not only in the natural environment, but in classroom language learning too.

Learning a language is a huge task. Firstly, there is simply not enough time for a teacher to provide guidance on every aspect of language. As we pointed out above, the word agreement relates to a group of other words in a number of different ways. It belongs to various different networks. There are so many networks and so many words that we cannot help learners understand all of them. There is so much to learn that it cannot all be covered explicitly with rules and explanations.

Secondly, even if we wanted to, we cannot always provide learners with the guidance they need. For example, Hughes and McCarthy (1998) show how the generally accepted pedagogic rule, 'that the past perfect tense is used for an event that happened in a past time before another past time …', enables learners to make well-formed sentences such as: I spoke to Lisa Knox yesterday for the first time. I had met her 10 years ago but had not spoken to her. But, as Hughes and McCarthy go on to point out, this rule does not show 'that the two sentences would be equally well formed if the second were in the past simple', although the emphasis would be different. What Hughes and McCarthy do not show is that a careful application of the rule would lead learners to produce some forms like: I opened the door when the postman had knocked, which are distinctly odd, if not ungrammatical. It is virtually impossible to frame a rule which will enable learners to make appropriate choices between the past simple and past perfect in these contexts. Hughes and McCarthy go on to draw the conclusion that:

The rule therefore … does not offer sufficiently precise guidelines to generate the choice when appropriate. In situations such as this our proposal is to look at the choices that real speakers and writers have made in real contexts and consider the contextual features that apparently motivated one choice or the other.

(Hughes & McCarthy, 1988: 268)

This is an interesting proposal, but it is impossible to carry out. The distinctions are simply too subtle and complex to demonstrate and explain. Although my explicit grammar of English is much more complete than that of most learners, and although I have spent a good deal of my professional life working on grammatical description, I am quite unable to provide a satisfactory explanation why I opened the door when the postman had knocked is a most unlikely sentence of English whereas I opened the door when the postman had gone seems perfectly reasonable. This means that I am able to operate grammatical systems which are much more subtle than anything I am able to explain. In assessing whether something is or is not grammatical we often act on feel, and are quite unable to explain our intuitions. The sentence I opened the door when the postman had knocked is a case in point.

Much learning depends on something subtler than the conscious application of rules, even if those rules attempt to take account of contextual features. As learners are more exposed to language, they begin to refine the systems they have consciously built, and to develop systems that they are not even aware of. This is largely an unconscious process, but it is a process that can be sharpened and informed by instruction. We can provide learners with useful hints – like the rule about the past perfect cited by Hughes and McCarthy – but this is simply the beginning of a process of exploration. Learners must be encouraged to go on working with texts and gradually refining their own model of the verb system.

To stimulate the process of exploration we need to encourage learners to focus carefully on the wording of texts. To help with this, teachers can design consciousness-raising activities designed to encourage learners to search input for clues to assist language development, and to help them learn more independently. These activities can be quite straightforward, simply drawing learners' attention to text and requiring them to look carefully at the language they have processed. But one thing is certain: unless learners process language unconsciously to refine the systems they have built by conscious effort, they will not develop a model of the language which even begins to approach that of the native speaker.

I would like also to draw attention to a fourth element which I will call rehearsal. This is an activity rather than a process, and generally comes between recognition and system building.

Rehearsal: Learners work consciously to develop routines, and are assisted in this by teacher-led activities. Often a routine may consist of no more than a single utterance. Learners repeat and manipulate patterns and phrases which they believe will be particularly valuable: Would you like …? Would you mind ___ing …? So do I. etc. When learning a language in the outside world, we sometimes rehearse whole encounters. Before going to the shops and using a foreign language which I do not speak very well I go over possible encounters in my mind, trying to predict the language I will hear and the language I will need to produce.

Rehearsal seems to contribute to learning in the early stages. Teachers organise and orchestrate repetition of individual utterances on an individual and a class basis. They encourage learners to repeat samples of a form they want learners to master. Activities of this kind certainly seem to reinforce learners' motivation. They may assist recall and use, certainly for basic vocabulary, such as my Spanish menu items. It is much less likely to be the case with complex grammatical systems like the tense system. Paradoxically it does not seem to help a great deal with the terminal -s and with question forms, which would seem to be ideal candidates for this kind of learning. Current research simply does not tell us how this kind of controlled repetition contributes to learning, although this does not mean that we should ignore it entirely. If it is sensibly contextualised within various learning processes, it may well be useful. It does mean, however, that we should not make it the basis of a methodology. Learning is a complex developmental process; it is tempting to think that we can offer a quick fix, but it is a temptation which we should resist.

We have now looked at three main processes which contribute to learning. The first of these, recognition, can be directly assisted by teacher intervention, drawing students' attention to aspects of language form. The second process, system building, is a conscious process whereby learners try to work out rules, speculating on the systems of the language and how they relate to one another. This too can be assisted by teacher intervention: teachers can either provide input in a way which helps learners to formulate rules for themselves, or they can intervene by providing rules for learners. Finally we have exploration. This is an unconscious process whereby learners discover or refine the language for themselves. Teachers cannot assist this process by direct intervention, but they can devise activities which will encourage learners to look carefully at language in ways that are likely to prompt discovery.

We need, then, to design classroom activities which will promote recognition and conscious system building. We need also to design activities which will encourage learners to discover language for themselves, to explore the relationship between meaning and form. Activities appropriate to different learning processes will be illustrated throughout the following chapters. But, as we have shown, learning is of little use unless what is learnt becomes a part of the learner's spontaneous language production. We also need to provide learners with plenty of opportunities to use the language, so they can gradually begin to put into practice what they have learnt. Before we begin to consider language use in the classroom we will look briefly at how language is used in the outside world.

1.5 Some questions about language

Up to now in this chapter we have taken it more or less for granted that learning a language means learning to produce appropriate sentences in that language. This is certainly the traditional view of learning: success or failure is normally measured in terms of this ability to produce appropriate sentences. When our students produce accurate question tags, we feel we and they have succeeded. When they fail to do so, we feel that we have failed. Unfortunately, if we measure success in this way, then language programmes are usually characterised by failure rather than success. But there is another way of looking at language and language learning, and that way may lead us to a very different view of success and failure.

In 1975 Michael Halliday published a book describing how his young son, Nigel, learnt his first language, English. Normally we think of children as learning how to talk. When a child reaches the age of two, we say things like: She can talk quite a lot now or She can say a lot of things now. Halliday, however, looked at language in a rather different way. We can see this from the title of his book. He called it 'Learning How to Mean'. For Halliday the important thing about language is the capacity to mean. What a child has to acquire is the ability to interact with others in a way which produces desired outcomes. Clearly the ability to achieve meanings is related to the ability to make sentences, but they are not the same thing. By the age of two children are able to realise a range of meanings, but they rarely utter a sentence which would be considered grammatical in terms of the adult language system.

It is not always easy to work out what children want to mean. At an early age children communicate by putting words together and relying on someone else, usually their mother, to work out the meaning with the help of the context. At this stage children don't bother with the little words which are so frequent in the language of the adult speaker: articles (the, a(n)), prepositions (in, on, at and so on) and the forms of the verb BE. Such words are often called structural words as opposed to lexical words: nouns, adjectives, verbs and adverbs. A child will say, for example: Dolly chair. If this is accompanied by a pointing gesture, it probably means: The doll is on that chair. If it is accompanied by a gesture handing the doll to an adult, it may mean: Please put the doll on a chair. So the child puts together a string of lexical words supported by gesture and context, and depends on the adult's willingness to work out meanings and act on them.

Children rapidly add to their stock of words, and as they do so the grammar gradually develops. It is some time before they begin to build in structural words, and some time before they begin to use anything but a very rudimentary verb system. Nevertheless, in the early stages, children manage to communicate a lot, even though their language is very limited and consists mainly of strings of vocabulary items. As children grow older their developing intellectual capacity demands more and more complex meanings, making more and more demands on the child's grammar. The child responds by developing a grammar to meet the new demands.

We should be careful not to overestimate the similarities between first and second language learning. Unlike a child, the adult has reached a high level of intellectual development. An adult learner already speaks at least one language fluently and is able to use that language as a resource to help with the learning of a new language. In spite of this, few adults master a second language to anything like native speaker level, whereas almost all children successfully acquire their first language to the extent that they can speak it fluently and accurately. So there must be marked differences in the learning processes.

It seems, however, that learners acquiring language outside the classroom, where there is a premium on effective communication, will work like children acquiring their first language, and attempt to build up a meaning system. Like children they are content in the early stages of learning to rely on stringing words together, using a minimal grammar. One way the grammar develops is that learners begin to acquire new grammatical forms, which they could not previously use. But learners also have other ways of increasing their ability to mean.

The teaching process normally encourages learners to increase their stock of language by learning new words and patterns, but learners may also increase their capacity to mean by making better use of the language they already have. Thus, a learner who does not have adequate control of negative verbs forms, but who knows the word no, will produce forms like I no want …; I no like …, and so on. Resourceful learners will make use of their first language to create new forms even if these forms are not a part of standard English. A speaker of French who does not have control of the present perfect tense, for example, may use the present simple: I already tell you this. A third way of learning to mean is by generalising from what we know to generate something we don't know: a learner who does not know the past tense of the verb run will sensibly offer the form runned. A fourth way is by using alternative means of achieving a given end: a learner who does not have command of question forms may rely on intonation and a puzzled facial expression to mark questions. All of these are legitimate ways of extending the meaning system; they are all legitimate meaning expansion strategies and should be encouraged in the classroom.

But there is more to meaning than simply getting a basic meaning across – we need to get meanings across in a way that can be readily and easily processed by a listener. My Ghanaian students, for example, had a complex and efficient dialect of English which they used successfully and effectively in communicating with other Ghanaians from different language groups. But this system was sometimes enormously difficult to use for another speaker who did not share that particular dialect. My job was to offer them a form of English which would be more widely negotiable, which would be understand by an international community, a community which did not have access to the Ghanaian dialect.

We need to have a form of English which can be readily processed by a wide range of other users, an internationally negotiable meaning system. But there is a clash of priorities between teaching and learning. Learners of English at the intermediate level face a difficult choice. Should they produce English fluently so that they can take part in a conversation in a way which other speakers do not find irritatingly slow? If they choose to do this, they will certainly produce many grammatical errors and may, at times, make themselves difficult to understand.

The alternative – and this is the preferred alternative for many learners, particularly adult learners – is to concentrate on producing language with a high level of formal accuracy. This requires careful attention to the language they produce. As a result their production will probably be so slow and hesitant that other speakers may find it irritating and frustrating – and it will still be sprinkled with inaccuracies. Whether the emphasis is on acceptable speed and fluency, or on acceptable formal accuracy, depends on the circumstances of use and on the personality and age of the learner. It is not a simple question, and the answer will nearly always involve a compromise of some sort.

We also need a form of English which will enable us to present ourselves to other users of English in a favourable light. It is a fact of life that we make judgements of others on the basis of the language they use and the way they use it. I often refer to this problem as the 'Tarzan' problem. Tarzan was a popular character in the films of my youth. He was a man who had been raised by animals in the jungle, like Mowgli, the wolf child. He was entirely at home in the jungle. In every episode of his story he would face down lions and wrestle crocodiles with heroic panache. But his language was very limited – he would introduce himself by slapping his chest and proclaiming: Me Tarzan. Now if you have just disposed of a crocodile in heroic manner, this may be an entirely appropriate way to introduce yourself to the admiring onlookers, but if you want to make your way in polite society, you probably need a quite different form. At the very least you need to say: Hi there. I'm Tarzan, the well-known king of the jungle. Perhaps this is making too much of yourself. A more modest introduction would be: Good afternoon. I don't think we've been introduced. I'm Tarzan. I live here in the jungle. So we need a range of language forms which will enable us to choose whatever we see as appropriate to the circumstances and the way we wish to present ourselves.

I have a good friend called Fabienne, a French woman who is an expert in Old English. She can tell me things about the derivation of words in my own language which I find endlessly fascinating. Her English is rapid and fluent, but it is also quite obviously the English of a French speaker. Her accent is such that you need to hear only a single sentence to identify her as a French woman. She is entirely happy with this. She wants to be taken for what she is – a woman who is proud to be French, but who speaks remarkably good English. She has no desire to be taken for a native English speaker.

Decisions of this kind should be taken by learners themselves. What sort of English do they want to speak? Do they want to be able to speak English at a basic level, like Tarzan? Or do they want to speak English with a fluent command of a native-like grammar and vocabulary while retaining their non-Englishness, like Fabienne? Or do they want to be taken for a native speaker of English and sound entirely British or American or Australian or whatever? Language is a system of meanings, but the meanings it carries do more than tell people things and ask them to do things – these meanings also tell people about how we view ourselves and how we view them: these meanings 'make an impression' on people.

Sometimes learners will quite deliberately produce forms which they believe to be non-grammatical. In English there is a broad generalisation that longer is politer:

Open the door.

Please, open the door.

Would you open the door, please.

Please, would you mind opening the door?

The same phenomenon may exist in a learner's own language. Imagine then a learner who wants to be polite but who does not have access to the modal would and the phrase would you mind. Such a learner may well produce something like:

Please, I like you will open the door, please.

Politeness is achieved by the length of the request, but in order to achieve that length learners may quite deliberately produce language which they know to be ungrammatical.

So we can think of language as a meaning system, but we need to think of it as a negotiable meaning system, one that has to be used with a range of other speakers of the language. We also need to think of it as a system which allows us to present ourselves appropriately in a range of situations. And, finally, what we want is a system which enables us to present ourselves to others in a way which we find acceptable. But we need to recognise one important fact: the ability to mean is not directly related to the ability to produce accurate sentences in standard British or American English, or any other standardised form. Learners may accept the production of non-standard forms as the price they have to pay to enable them to speak rapidly and fluently. They may accept nonstandard forms because they have no wish to be taken for a native speaker of English. Thus, we can speak of complementary purposes in producing language:

Basic message → Concern for reader/listener → Presentation of self

Learners are concerned first to get their message across with acceptable speed and fluency in real time. Secondly they will want to structure and mark their message in such a way as to make it readily comprehensible and acceptable to their reader or listener. Finally they will want to carry their message in a way that presents them as they wish to be seen. The language they produce will vary according to the circumstances in which it is produced and according to the learner's communicative priorities. Those familiar with Hallidayan functional grammar (see Halliday, 1978 and 1994) will recognise that these complementary purposes relate very closely to Halliday's metafunctions: ideational, textual and interpersonal.

Let us look at two extreme types of language use in the classroom and go on to consider the implications for learning. We will consider first…

 

Credit:  Dave Willis