44623773 Lexical Syllabus Willis

8/18/2019 44623773 Lexical Syllabus Willis

1/142

The Lexical SyIlabus

A new approach to language teaching

Dave Willis

COLLINS E.L.T

London and Glasgow

Contents

Introduction iii

Chapter 1 From methodological options to syllabus design l

Chapter 2 Words and structures 15

Chapter 3 The lexical research and the COBUILD project 27

Chapter 4 Syllabus content 39

Chapter 5 Communicative methodology and syllabus specification 57

Chapter 6 Syllabus organisation 74

Chapter 7 Word, structure, function and discourse 91

Chapter 8 A brief review 124

Bibliography 133

Index 134

Collins ELT

8 Grafton Street

LONDON W1X 3LA

COBUILD is a trademark of William Collins sons & co Ltd

First published 1990

10987654321

All rights reserved. No part of this book may he reproduced, stored in a retrieval system, or transmittedin any form or by any means, electronic, mechanical, photocopying, recording or otherwise, withoutthe prior permission in writing of the Publisher.(NOTE: The University of Birmingham has obtained permission from the publishers to make this book,now out of print, available to its students, on the University website. The copyright of the publishers

should be respected in the usual way.)


2/142

Introduction

There is general agreement nowadays that we learn a language best by using it to do

things, to achieve outcomes. Communicative activities involving games playing and

problem solving have become a more and more important part of the language

teacher's stock in trade over the last fifteen years or so. Some writers (see, for example, Maley and Duff, 1978) display great ingenuity in devising such activitiesand there is a wealth of supplementary material which exploits these activities. Yet in

spite of this virtually all coursebooks rely on a linguistic syllabus which 'presents' thelearner with a series of linguistic items.

It seems that communication is good fun and well worthwhile for a bit of variety,

but that the serious business of language learning needs to have a firm grammatical

basis resting on the assumption that the grammar of the language can be broken down

into a series of patterns and reconstructed in a way accessible to the learner. Even

coursebooks based on a notional-functional syllabus specification, which take units of

meaning as syllabus items, still rest on a methodology which 'presents' learners with a

series of patterns. The notionalfunctional syllabus is communicative in that it tried tospecify the syllabus in terms of meaning, in terms of what was to be communicated.

But the methodology which realises the notional-functional syllabus is little differentfrom the methodology which realises the structural syllabus which it seeks to replace.

Both depend on a three part cycle of presentation, practice and production.

My dissatisfaction with this methodology has a theoretical basis but it is strongly

reinforced by experience in the classroom. The theoretical base draws on the work of

people like Prabhu (1987) and Rutherford (1987) both of whom point to the glaring

inadequacy of pedagogical grammars. They argue that we cannot begin to offer

anything like an adequate description of the language on which to base a pedagogical

grammar. Given this, our only recourse is to depend on the innate ability of learners torecreate for themselves the grammar on the basis of the language to which they are

exposed. The conclusion is similar to that drawn by interlanguage theorists like Corder

(1967) and Selinker (1972) and classroom resear~hers like Ellis (1984). Teachers and

researchers have been aware for many years that 'input' does not equal 'intake', that

what teachers claim to be teaching bears only a tenuous relationship to what learners

are actually learning. But in spite of this, coursebook writers continue to act on the

assumption that language can be broken down into a series of patterns which can then

be presented to learners and assimilated by them in a predictable sequence. It does not

seem to worry people a great deal that this assumption flies in the face of our

experience as teachers. My experience in the classroom, like that of all teachers I suppose, has seen both

failures and successes. On the one hand I found that students often failed to learn what

I thought I was teaching them. On the other hand most of them showed an ability to

transcend the limited language which I had so carefully presented to them. It wasclear to me that my efforts to present the grammar of the language met with very

limited success, yet in spite of this mv students' English was improving. It is

encouraging to know that so much learning is taking place in the classroom. It is

sobering to realise just how little control the


3/142

iv The Lexical Syllabus

teacher has over what is being learned. The obvious conclusion to be drawn from this

is that students learn a great deal directly from exposure to language through reading

and listening, without the need for the teacher to impose a description on what is

learnt. One of the most plaintive cries in any staffroom goes along the lines of "I'vetaught them that so many times, and they still get it wrong." There is overwhelming

evidence from my experience as a teacher that teachers have little control over what is

actually learnt and reproduced in spontaneous language use. How many times, for

example, do we teach the distinction between the present simple and the present

continuous before students begin consistently to get it 'right'? It usually takes them a

long, long time. Could this be because it is not the controlled presentation which does

the trick but rather constant exposure over a period of time? Could it be that students

learn in the controlled environment of the language classroom not because language is presented to them, but because they are constantly exposed to language? And if this is

the case should we not be looking to methodologies which maximise meaningfulexposure to and use of language?

Taking meaningful exposure as a starting point it is possible to develop an

approach to language teaching which takes advantage of the learner's natural tendency

to make sense of language and to learn for himself. In order to take full advantage of

this approach, however, two other things must be done. First a methodology must be

defined which encourages the learner's ability to learn. Teachers need to encourage

learners to look critically at language and to recognise the need to develop and refine

their language code in order to achieve their communicative aims. Secondly we needto look carefully at the kind of language to which learners are exposed. Random

exposure is of little value. Exposure must be organised.

What should be aimed at, is exposure that is organised in three ways. First thelanguage that learners are expected to understand and produce should be graded in

some way so that learners do not face such difficulties and complexities at an early

stage that they become demotivated. Secondly the language they are to be exposed to

should be carefully selected so that they are given not random exposure, but exposure

to the commonest patterns and meanings in the language - the patterns and meanings

they are most likely to meet when they begin to use language outside the classroom.

Thirdly there should be some way of itemising the language syllabus so that it should be possible not simply to expose students to language, but also to highlight important

features of their language experience, and to point to what language we mightreasonably expect them to have learned from their experience.

The first of these problems is relatively easy to surmount. It is not too difficult to

design tasks which involve a meaningful use of language but which can still be

handled by learners who have relatively little control of the language - the kind of

learner who is often referred to, somewhat unfortunately in my opinion,as a remedial

or false beginner. Tasks which have a clear outcome and involve the exchange of

highly specific information can be made accessible to false beginners. As I have said,

such tasks have been used as supplementary material for many years. The second and third factors were, until recently, more problematic. When


4/142

Introduction v

my wife Jane and myself were asked by Collins ELT in 1983 to begin to research and

write a series of coursebooks, the Collins COBUlLD English Course, we began to ask

ourselves a number of related questions. How could we identify the commonest

patterns and meanings in English and how could we highlight these for students?Obviously many of them are covered in most elementary courses. The verb be and itsforms and most of its uses would obviously come high on any list as would

prepositions of place. But other equally common forms such as the passive voice and

modal verbs are traditionally left until much later. Also, we discovered as we became

more involved in the research that a number of important words and patterns are often

omitted altogether. Words like problem, solution, idea, argument and thing are

commonly used with a noun clause introduced by that to structure discourse. It is

difficult to get very far in speech or writing without them.

And what about items which seemed to take up far too much time in elementarycourses, items like the present continuous used to talk about what is happening here

and now? Apart from a traditional belief that certain patterns are 'difficult', thereseems to be little objective reasoning behind the selection and ordering of items. We

were soon to find evidence that a syllabus based on these established values was

likely to be highly uneconomical.

But how could we go beyond the traditional approach to itemising and organising

a syllabus? Given the range of language experience which is bound to come from

exposure to a series of tasks which are graded for difficulty but not otherwise

linguistically graded, how would one choose which elements of language to

highlight? How would one decide which items to specify as part of an efficientlearning programme? Perhaps the most convincing attempt in the field so far was the

Council of Europe Threshold and Waystage Syllabus. But this was ultimately a very

subjective piece of work. It took as its basis the intuitions of scholars and teachers. Itdid not rest on an analysis of actual language use.

In the mid-1980s a number of things began to come together. After years of

teaching English as a foreign language, a period of work as a teacher and teacher

trainer in the second language environment of Singapore had forced me to look more

closely at methodological issues, particularly the relationship between accuracy and

fluency (Willis and Willis 1987). This helped to formalise a communicative approach

to ELT and to identify some of its important components. The writing of the CollinsCOBUILD English Course provided us with the opportunity to put these

methodological insights to work. The coursebooks were to be a part of the COBUILD research project in lexical

development, a major computing and publishing venture involving cooperation

between Collins and the English Language Research Department at Birmingham

University.

The first part of this project had involved the assembly on computer and

subsequent analysis of a 7.3 million word corpus (later extended to over 20 million

words) of spoken and written English. It was proposed by John Sinclair, Professor of

Modern English Language at Birmingham and Editor-in-chief of the COBUILD project, that this computational analysis should provide the basis for a new

coursebook syllabus, a lexical syllabus. Sinclair advanced a number of arguments in

favour of the lexical syllabus, but the underlying argument was to do with utility and

with the power of the most frequent words of English.


5/142

vi The Lexical Syllabus

The 700 most frequent words of English account for around 70% of all English text.

That is to say around 70% of the English we speak and hear, read and write is madeup of the 700 commonest words in the language. The most frequent 1,500 words

account for around 76% of text and the most frequent 2,500 for 80%. Given this, wedecided that word frequency would determine the contents of our course. Level I

would aim to cover the most frequent 700 words together with their common patterns

and uses. Level 2 would recycle these words and go on to cover the next 800 to bring

us up to the 1,500 level, and Level 3 would recycle those 1,500 and add a further

1,000. We would of course inevitably cover many other words in the texts to which

students were exposed, but we would highlight first the most frequent 700, then 1,500

and finally 2,500 words in the language.

In one way this took us back to the pioneering work in the analysis of lexis of scholars like West and Thorndike in the 30s and 40s. But the computer would be able

to afford a much more thorough and efficient analysis than had been possible in those

days. The database assembled at Birmingham would provide us with detailedinformation about the commonest words and patterns in English and the meanings and

use of those words and patterns. At first we had doubts about the practicality of the

lexical syllabus. But the more we worked with the information supplied by the

COBUILD research team the more we became convinced that the syllabus which

emerged was highly practical, entirely realistic and vastly more efficient than

anything we had worked with before.

I have already pointed to words like problem, solution and idea which are omittedfrom most language courses, even though they play a vital function in structuring the

way we speak and write. A particularly striking example is the word way, the third

commonest noun in English after time and people. The word way in its commonest

meaning has a complex grammar. It is associated with patterns like:

. . . different ways of cooking fish.

A pushchair is a handy way to take a young child shopping.

What emerges very strongly once one looks at natural language, is the way the

commonest words in the language occur with the commonest patterns. In this case theword way occurs with of and the ring form of the verb and also with the to infinitive.

It is also extremely common with a defining relative clause:

I don't like the way he talks.

The lexical syllabus does not identify simply the commonest words of the language.

Inevitably it focuses on the commonest patterns too. Most important of all it focuses

on these patterns in their most natural environment. Because of this, the lexicalsyllabus not only subsumes a structural syllabus, it also indicates how the 'structures'

which make up that syllabus should be exemplified. It does this by emphasising theimportance of natural language.

As we began work on the course design, therefore, a number of basic principles

were agreed:

-

The methodology employed would be based entirely on activities involving real


6/142

Introduction vii

language use.

- Learners would be exposed almost entirely to authentic native speaker language. They would not betaught through the medium of'TEFLese'- a language designed to illustrate the workings of a

simplified grammatical system and bearing a beguiling but ultimately quite false similarity to realEnglish.

- Spoken material recorded specially for use in the course would not be scripted and rehearsed. Itwould be spontaneous speech and would therefore contain many linguistic features normallyidealised out of language teaching material.

- We would not 'present' learners with language but would encourage them to analyse for themselvesthe language to which they were exposed and thus to learn from their own experience of language.We would not say to learners 'I, the teacher, will exemplify for you the important features of English,

and you, the learner, will thereby build up a description of the language in the way that I havedetermined'. We would say instead 'You, the learner, already have valuable experience of thelanguage. We will help you to examine that experience and learn from it'.

In effect what we planned to do was create a learners' corpus and encourage

learners to examine that corpus and generalise from it. I have already referred to theCOBUILD corpus of 20 million words. By studying this corpus in great detail,lexicographers were able to make valid and useful generalisations about the meanings

and uses of the words in the language. For Level 1 of our course we intended to create

a corpus which would contextualise the 700 most frequent words of English and their

meanings and uses. We would then highlight those words with their meanings and

uses to provide learners with valuable exposure and experience. We would then

devise exercises to encourage learners to analyse that experience of language and to

learn from it. Levels 2 and 3 would go on to do the same at the 1,500 and 2,500 word

frequency levels.

We set about designing tasks for use in the classroom. Some of these were based

on written and some on spoken texts. All of the spoken tasks designed to be performed by learners were carried out by native speakers and recorded. This gave us

a bank of texts, both spoken and written which we could use to provide learners with

balanced exposure to the language. The balance was determined by the original

COBUILD research. We identified from that research the important features of

language we wished to illustrate and then constructed our corpus by selecting texts

which would indeed illustrate those features of language. This was a long and

time-consuming process. All the texts we used had to be closely analysed and many

of them were rejected on the grounds that they did not afford us economical coverage.

What we finished with was a small corpus of language which presented the learner

with a microcosm of the 20 million COBUILD corpus. In becoming familiar with thiscorpus, the learner would become familiar with the language as a whole since the

corpus contained all the important features of the words which make up 80% of

language use.

The lexical syllabus, therefore, affords the learner a coherent learning opportunity.

It does not dictate what will be learned and in what order. It offers the learner

experience of a tiny but balanced corpus of natural language from which it is possible

to make generalisations about the language as a whole. It then provides the learner

with the stimulus to examine that mini-corpus in order to make those productivegeneralisations.


7/142

viii The Lexical Syllabus

The process of syllabus design involves itemising language to identify what is to be

learned. Communicative methodology involves exposure to natural language use toenable learners to apply their innate faculties to recreate language systems. There is an

obvious contradiction between the two. An approach which itemises language seemsto imply that items can be learned discretely, and that the language can be built up

from an accretion of these items. Communicative methodology is holistic in that it

relies on the ability of learners to abstract from the language to which they are

exposed, in order to recreate a picture of the target language.

The lexical syllabus attempts to reconcile these contradictions. It does itemise

language. It itemises language minutely, resting on a large body of reseach into

natural language. On the basis of this research it makes realistic and economical

statements about what is to be learned. But the methodology associated with thelexical syllabus does not depend on itemisation. It allows learners to experience

language items in natural contexts and to learn from their experience. It relies

crucially on the concept of the learners' corpus. It is the concept of the learners' corpuswhich reconciles the contradiction between syllabus specification and methodology.

Once we had come to this realisation the concept of the learners' corpus was simple.

The processes by which we came to this concept, and the procedures which realised it

are far from simple. It is those processes and procedures which are described in this

book.


8/142

The Lexical Syllabus; Dave WillisOriginally published by Collins ELT, 1990

CHAPTER 1: From methodological options to syllabus

design

Syllabus and methodology

It is tempting to see syllabus design and methodology as discrete options. The

syllabus specifies what is to be learned and the methodology tells us how it is to be

learned. It seems that there need be no conflict between the two. We can specify a

syllabus in whatever way seems sensible, and can then use whatever methodology we

want in order to transmit our syllabus content. Unfortunately, in recent years there has been conflict between syllabus and methodology. The failure to recognise this conflict

has on occasions led to a good deal of confusion. There is general agreement nowadays that people learn a language best by actually

using the language to achieve real meanings and achieve real outcomes. This belief

has brought into the classroom a wide range of activities designed to promote

language use - role play, games playing and problem solving activities for example.

These activities are contrasted with activities which involve manipulating language in

ways which do not involve any exchange of meaning. Listening and repeating,

transformation exercises and controlled pattern practice are activities which involve

the production of language but not the use of language. This emphasis on language use has obliged us to look carefully at the content of

language courses in terms of topics and activities. The best way to ensure that learners

really use language is to put them in a situation which makes them want to uselanguage. We must catch their interest in some way, or present them with a challenge

they feel motivated to meet. They will then be predisposed to use language in order to

communicate their interest or to meet the challenge of a game or problem.

There are, then, at least two kinds of language production as part of the learning

process in the classroom. At times people produce language in order to communicate.

At other times they produce language simply in order to practise correct forms, or to

demonstrate that they can produce a correct form. This may seem to be astraightforward distinction, but at times it can cause confusion in the classroom.

Here is an example from some actual classroom data (J R Willis 1981). Theteacher has worked very hard to set up a situation in which students are to practise a

number of verbs followed by the gerund form -ing. She tells one student:

Antonio, ask Socoop if he likes being a father.

Antonio says:

Socoop, do you like being a father?


9/142

2 The Lexical Syllabus

Socoop replies:

Yes, I erm . . . I am father of four children.

By standards operating outside the classroom this is a perfectly reasonable reply. It isalso, as it happens, an acceptable sentence of English. The teacher, however, is not

satisfied with this reply. She says:

Yes, all right, listen to the question though.

Socoop listens to the question and then tries a series of replies without real success

until the teacher resolves the issue by answering for him:

Yes, I do. I like being a father .

The learners do not challenge the truth of the teacher's utterance, even though the

teacher is a woman, because they know it is not a real statement intended to

communicate something about the teacher's attitude to parenthood. It is simply the

teacher correcting Socoop and giving him a model of the target pattern. Socoop's

mistake, of course, was to behave as if the question he was asked was a real question,

and as if he really was expected to explain to the class his feelings about fatherhood.

McTear (1975) gives a similar example:

Teacher: Where are you from?Students: We're from Venezuela . . .

Teacher: No. Say the sentence: Where are you from?Students: Where are you from?

Here again the students imagine that the teacher is asking a real question whereas in

fact the teacher is simply giving them a model to follow. The literature on classroom

research is full of misunderstandings of this sort, where an utterance is taken as

having some communicative value, when in fact it is simply intended as a sample of

language for the learners to copy or manipulate in some way, usually by repeating

word for word or by producing another sentence incorporating a similar pattern.

Unfortunately, it is not only learners who are sometimes confused about the nature

and purpose of language produced in the classroom.

Most teachers nowadays would claim that their approach to teaching rests, as Ihave already said, on the belief that we best learn a language by using that language

rather than simply by producing samples of it for the teacher's inspection and

correction. Broadly speaking such an approach is referred to as communicative, since

it is based on the use of language to communicate. Even if a language programme is based on a grammatical syllabus, it may be described as communicative on the

grounds that it rests on a communicative methodology. But what if there is, as I have

claimed, a conflict between syllabus and methodology?

I believe that such a conflict is revealed in attempts to harness a communicative

methodology to a grammatical or structural syllabus. The syllabus aims are expressed

as a series of language patterns with their associated meanings. The aim of each unit

is that by the end the learner should have mastered the prescribed pattern or patterns.One methodology which might realise such a


10/142


11/142

4

The Lexical Syllabus

'right' answer to:

What are you doing tomorrow?

is not:

I don't know. I might play tennis.

or: I'll probably play tennis if the weather's okay.

or:

I'm going to play tennis.

The 'right' answer is:

I'm playing tennis.

This is because the focus of the activity is not really on the content of the language,

the meanings that are being exchanged. The real focus is on the form of the utterancesused to realise those meanings.

Presumably, then, it is at the production stage that learners are involved in real

language use. The first two stages have an enabling role. They provide students with

the language they will need in the production stage. But what is it that is to be

produced? If, as the label implies, the purpose of this stage is to produce the target

form, then what we have is yet another form-focused activity. The intention may bethat the production of the target form is subordinate to some other activity, a role play

or problem solving exercise for example. But if learners are predisposed to producespecific forms of the language, then it is difficult to escape the conclusion that the

activity is one which focuses on form and on formal accuracy. During the presentationand practice stages, learners have been encouraged to give the 'correct' response to the

question - correct in that it incorporates the form under study. In the same way during

the production stage learners will be strongly predisposed to produce the target form.

In the example I have given they will be predisposed to make statements about the

future not by using the modal will or might or by using going to, but by using the

present continuous, irrespective of the meaning they wish to convey. In other words

learners will have a mental set such that form takes priority over meaning. When itcomes to talking about the future in a classroom context, the focus of the productionstage is very much on form.

Sometimes this predisposition on the part of the learners is reinforced by theteacher and the materials used. Learners may be encouraged to 'try to use phrases like

these'. They will then only be regarded as having performed successfully if they do

indeed produce the forms which have just been presented and practised and if they fail

initially to do this the teacher will intervene to ensure conformity. Socoop made the

mistake of assuming that he was being asked a real question, and he had to be

corrected by the teacher. In the same way a learner who fails to produce the present

continuous after the kind of presentation and practice stages we have described may

be 'corrected' by the teacher. It is easy to drift into a situation in which the main purpose of the pro-


12/142

From Methodology to Syllabus 5

duction stage is not to exchange meanings but to produce the target form.

In spite of this, there is sometimes a claim that this kind of methodology is in some

way communicative. Littlewood (1981) outlines a sequence based very much on

presentation, practice and production in which he subsumes presentation and practiceunder 'pre-communicative activities', leading up to 'communicative activities'corresponding to the production stage:

Through pre-communicative activities, the teacher isolates specific elements of knowledge or skill which compose communicative ability, and provides the learners withopportunities to practise them separately . . . In communicative activities, the learner has to

activate and integrate his pre-communicative knowledge and skills, in order to use them for the communication of meanings. He is now engaged in practising the total skill of

communication. (Littlewood 1981)

Littlewood suggests that the normal sequencing will be for teachers to provide input

in the form of a form-focused pre-communicative activity, and to follow this with a

communicative activity 'during which the learners can use the new language they

have acquired and the teacher can monitor their progress'. But if the purpose of theso-called communicative activities is for students to demonstrate control of the newly

introduced language forms, how does the teacher 'monitor their progress'? Presumably

by listening to see if they do indeed incorporate the target form, and additionally to

see if they produce it accurately.

It is difficult to see how such activities can be regarded as truly communicative if

the learners' main object is not to achieve some outcome through the use of language,

but to demonstrate to the teacher their control of a target form. True communication

involves the achievement of some outcome through the use of language, and demandsthat the language used should be determined by the attempt to achieve that outcome.

In the kind of communication described by Littlewood, the so-called communicativeactivity is simply an opportunity to use a particular form and the language used is

conditioned by this.

There is, therefore, a tension, perhaps a basic contradiction, between a

grammatical or structural syllabus and a communicative methodology. A grammatical

syllabus demands a methodology which focuses on the correct production of target

forms. It is form-focused. A communicative methodology, if it involves real

communication, demands that learners use whatever language best achieves the

desired outcome of the communicative activity. There is no real sense in which the

presentation and practice stages described above can be called communicative, because they restrict the freedom to use whatever forms best realise communicativeintent. Learners are not able to choose whatever forms best realise the meanings they

want to realise, but rather have to use the forms that have been identified and

prescribed for them by their teacher.

At the production stage teacher and learner have two options. The purpose of the

stage may be for learners to produce the target form. If this is the case then

communication has been subordinated to the primary goal, which is to rehearse the

use of a particular form. The other option is for them to see this last stage as free.

Learners use whatever language they want in order to achieve the desired

communicative outcome or intention. But if they do this we can


13/142


hardly speak of a 'production' stage, because we can no longer say what it is that is to

be produced and we can no longer point to a link between the activity and the

syllabus.

Sometimes a claim to a 'communicative' approach rests on syllabus specificationrather than methodology. Many language teaching programmes take anotional-functional or 'communicative' syllabus as their starting point. Such a

syllabus, like the Council of Europe Threshold Syllabus, is seen as communicative

because it consists of an inventory of units of communication rather than an inventory

of sentence patterns. It has units entitled 'Making Requests' and 'Cause and Effect', so

it is concerned with what is to be communicated rather than with how it is to be

communicated. In this case one would expect to match the syllabus statement with a

communicative methodology.

But the communicative syllabus based on specified notions and functions does notreally consist of such communicative units. Those units are the abstract categories on

which the syllabus is based, but these categories are realised by a set of sentence patterns. The real syllabus is an inventory of such patterns. Thus the unit on requests

may cover the models would and could in patterns like:

Could you open the window please.

The methodology which is usually associated with such a syllabus is a presentation

methodology of the kind I have described above. It depends on learners producing thetarget pattern rather than encoding requests in whatever way seems to be most

appropriate. It is not 'communicative' because it does not involve learning through

use.

It seems, therefore, that syllabus and methodology are not discrete options. If wechoose a syllabus which specifies an inventory of language forms, it is difficult to see

how we can achieve this syllabus by means of a communicative methodology. And if

we want to use a communicative methodology in which learners use language freely,

it is difficult to see how we can then specify what language forms will be covered bythis methodology.

One response to this conflict is to adopt an eclectic approach. For example thesyllabus may be defined linguistically as an inventory of language structures, and

realised through a presentation and practice methodology. This methodology may

then be supplemented by giving learners ample opportunity to use language freely to

enable them to consolidate and extend what they have been taught. This is what

underlies Brumfit's (1984) recommendation that a language learning programme

should offer a balance of activities, some of which focus on accuracy and some on

fluency. There is a focus on accuracy when learners are concerned with the form of

the language they produce, and on fluency when they are concerned with exchangingmeanings and achieving outcomes.

One could achieve this double focus by operating with a structural syllabusrealised through a presentation and practice methodology, and by having in parallel a

series of activities which encourage learners to use language. This would not be a

production stage but a discrete series of activities, so that learners did not feel

constrained to 'produce' any particular form, but simply to communicate as best they

could with whatever language they could command.


14/142


However, an eclectic approach of this kind skirts around the problem of reconciling a

syllabus specified in linguistic terms with a methodology based on language use.

There is no serious attempt to ensure that there is a real relationship between the

language syllabus, realised by controlled activities, and the communicative activities.Presumably one would hope that there would be a good chance that in thecommunicative activities learners would use at least some of the language that had

been presented and practised, even if one did not judge success simply in terms of

what language was produced and how accurately. But this does not provide more than

a tenuous link between syllabus and fluency activities.

I have argued up to now that there are potential conflicts between the way we

specify a syllabus and the way we realise that syllabus. I have argued that there is a

basic dichotomy in the language classroom between activities which focus on form

and activities which focus on outcome and the exchange of meanings. I havesuggested that we need to be clear about the relationship between syllabus

specification and methodology. I have also suggested that the choices involved areconcerned crucially with the ,way language is used in classrooms - whether the focus

is primarily on language form or on language as a means of communication.

Grammar in the classroom

Rutherford (1987) is highly critical of the view of language enshrined in a

presentqtion methodology. He argues that this approach to language learning regards

the process as one of 'accumulated entities'. According to this view learners graduallyamass a sequence of parts. At intervals their proficiency in the language is measured

by determining what parts and how many parts they have accumulated. Rutherford

argues that most commercially produced foreign language textbooks reflect this view,an indication that it is a view widely held in the language teaching profession. The

associated methodology is based on:

. . . the discovery of a target language whose structure has been analysed into its putativeconstituent parts, the separate parts thus serving as units of pedagogical content, focus,

practice and eventual mastery. (Rutherford 1987)

The danger with an approach of this kind is that it trivialises grammar, and trivialises

language description in general. It does give recipes for the construction of some

clauses and sentences, and for the production of samples of language. But the

grammar of a language is not a set of clauses and sentences. It is the systematicrelationship between meaning and form which underlies the production of

grammatical clauses and sentences. It is useful to acquire samples of language only inso far as those samples lead us towards insights into the underlying system.

Language behaviour is highly systematic. We produce language in accordance

with a complex system of rules. Most people, even though they are successful

language users, are quite unable to give anything but the most rudimentary statements

about how that system works. They can make statements about whether or not

something 'sounds all right', but find it very difficult to


15/142


explain these decisions. Most native speakers would have no doubts that the

sentences: John is being silly.

and: John is being careful.are grammatically acceptable. Equally most native speakers would have doubts about

the sentences:John is being tall.

and:John is being handsome.

Silly and careful belong to a class of adjectives often referred to as dynamic. They are

used to describe someone's behaviour rather than their inherent attributes. Dynamic

adjectives, such as awkward, mischievous, kind and cruel, are regularly used with

the present continuous, whereas other adjectives which are stative are not. This is a

rule which all native speakers operate but very few would be able to explain. There is a difference then, between a user's grammar and a grammaticaldescription. A user's grammar is an internalised system, the operational system

underlying our language behaviour. We normally operate the system unconsciously

and are quite unable to explain it. A grammatical description is an attempt to

characterise that behaviour, and to identify the categories and concepts on which it

rests (categories like adjective, dynamic and stative).

Prabhu (1987) argues, like Rutherford, that most approaches to language teaching

are based on 'internalisation of the grammatical system through planned progression,

pre-selection and form-focused activity'. In other words there is a description of thelanguage which is communicated to the learners bit by bit by revealing to them

samples of the language in a predetermined order. Prabhu claims that such anapproach is based on a number of quite false assumptions.

The most basic of these assumptions is that we have a description of grammar

which is adequate to this task. The user's grammar is and always will be, more

complex than any descriptive grammar. Indeed attempts to describe the grammar

simply showed:

. . . that the internal grammatical system operated subconsciously by fluent speak ers was vastly more complex than was reflected by or could be incorporated intoany grammatical syllabus- so complex and inaccessible to consciousness in fact,that no grammar yet constructed by linguists was able to account for it fully.

However much we may wish to, we simply cannot give the learner a description of

the language which works. It must follow that if our pedagogic description of the

language is inadequate, then in order to learn the language the learner must operate

learning strategies which do not depend on a grammatical description of the language.

There must be important and subtle insights into the structure of language which

learners are able to make quite unaided. A look at most coursebooks will confirm that the number of patterns actually

brought directly to the attention of learners does not go very far towards a com-


16/142


prehensive grammar of English. Fortunately learners are able to transcend or, perhaps

more accurately, to by-pass the grammar that is presented to them and to go beyond it.

They begin, for example, to use the present perfect tense with reference to future time

in sentences like: Please let me know as soon as you havefixed your travel plans. I'll come round later if I've finished what I have to do.

even though this particular use is hardly ever presented in coursebooks. They learn, as

we shall see later, that the word any and its compounds are used in affirmative

sentences like: Anything you can do I can do better. Come round any time.

even though they may have been taught quite explicitly that any is used only in

negative and interrogative sentences. We should ask very seriously how it is that

learners are able to go beyond what they are taught in this way. An obvious possibility

is that they learn a good deal for themselves from the language that they read andhear. They do not need to be taught, because they have an innate ability to generalise

from the language they read and hear in order to build up and refine a workable

grammatical system. It is also difficult to see how the learner can move from an inventory of discrete

patterns towards important generalisations about the grammar of the language. Wehave already pointed out that there is much more to language than a series of

structures which can be presented to a learner. We can present, for example, the

pattern which is commonly, though misleadingly, called the first conditional: If it rains we will get wet.

This pattern is regarded as difficult, and therefore worth presenting to students,

because the use of the present simple tense with reference to future time causes

particular problems. But this is a feature of the so-called present tense, not simply of the first conditional. The present tense is commonly used with future reference in

temporal clauses: It'll be quite late when we arrive.

and after verbs like hope: I hope somebody is there to meet you when you arrive.

The same use is common in other subordinate clauses: There will be a prize for the one who finishes first.

The present tense is an option when the future is already fixed or arranged. I recently

had a conversation trying to arrange a meeting involving a number of people. One of

the participants turned down a proposed date saying:

I'm sorry, I'm in Bhutan.This was obviously not a reference to present time since we were in a British

university at the time. It was a reference to future time and was acknowledged


17/142


by another participant:

Oh, yes. When do you go?

Drilling or repeating the first conditional pattern may show a learner that this is anacceptable pattern of English, and the pattern may eventually be incorporated in the

learner's language. But it tells the learner nothing of great value about the grammar of

the present tense. Indeed, by implying that there is something unusual about this use

of the tense and that this unusual feature is associated with the conditional, it is

actually getting in the way of learners developing a more complete description of the

present tense and realising that 'the present simple tense is neither present nor simple'

(Lewis 1989).

There is also an assumption on the part of those who present language to the

learner that the learner is actually in a position to receive what is presented, that we

can specify what will be learned and in what order. This again flies in the face of our experience as teachers. We know very well that it will be a long long time before

learners distinguish consistently between, for example, the present perfect and past

simple forms of the verb. We may 'present' some version of this distinction but it willnot immediately become part of the learner's language behaviour. A learner may

ignore the distinction altogether or may operate it only in a few instances. It will be along time before the learner has any control of this part of the verb system of English.

We cannot realistically hope to present the learner with usable information in this

way. All we can realistically do is attempt to make the learner aware that these

concepts and these distinctions are a part of the grammar of English. Whether and at

what point the learner will be able to act on that information is beyond our control.

If we are to help learners to acquire the grammar of the language in the sense of an operating system, we must begin by acknowledging that we can only do this

indirectly. We may be able to offer useful hints, but we cannot begin to offer a fulldescription of the language. We may be able to devise activities which will help

learners internalise the grammar of the language for themselves, but we cannot present them with usable chunks of language. A methodology should take account of

the fact that any pedagogic grammar will be inadequate, that what is presented will

not necessarily be received and, most important of all, that the crucial participant in

the attempt to internalise a grammar is not the teacher or the materials but the learner.

Use and usage

Even if we were able to teach the grammatical system effectively, there is noguarantee that this would be translated into an ability to use the target language.

Widdowson (1978) argues that a methodology which focuses simply on languageform is deficient in that it is concerned simply with enabling students to produce

correct sentences. He feels that the ability to use language involves more than just the

ability to produce grammatical sentences.

Someone knowing a language knows more than how to understand, speak, readand write sentences. He also knows how sentences are used to communicative

effect. (Widdowson 1978)


18/142


At first sight this may seem to be a highly artificial distinction. How can someone

know how to 'understand, speak, read and write sentences' without being able to use

these sentences to communicative effect?

It seems to me that there are two ways in which this can happen. The first is probably familiar to very many of us who have learned a foreign language. We canwork out the meaning of a spoken sentence and perhaps reply to that sentence, but

only if we are given time to process the language involved. Given time we can do a

lot with the language, but under the kind of time pressure which usually accompanies

language use we just cannot get things together. There is a sense in which we know

the language in that we know what the forms mean and we know what forms we want

to produce. But there is another sense in which we do not know the language. We

cannot get things together with sufficient speed and confidence to use the language

when we are required to do so. We have a knowledge of the forms, but we do nothave the kind of fluent control demanded in real communication.

There is a second sense in which we may be said to 'know' the language and at thesame time not to know it. We can produce and understand acceptable sentences in the

target language, but we are not sure in what circumstances these sentences would be

appropriate as tokens of communication, and we are not sure how we would deploy

them in communicative discourse. This is what Widdowson has in mind.

Take, for example, an English speaker who has a good knowledge of French

grammar and lexis and who then tries to put this knowledge to use in writing a

business letter in French. The letter would be unlikely to create a favourable

impression in a French reader. The conventions of letter writing in French are quitedifferent from those in English, and if the words and phrases are translated directly

into English they sound elaborate and ornate to an English ear. Similarly, the direct

equivalent of an English letter might sound abrupt and dismissive to a French speaker.We all have to learn the conventions of certain types of communication in our own

language, even though we have a sophisticated knowledge of the grammar and lexis.

We have to do the same in a foreign language.

We must also learn how to deploy sentences in discourse. There is a phrase in

English which seems to have become very common in recent years. The phrase is

'Having said that...', and it is used to introduce some modification or something which

partly contradicts what has just been said. There is nothing in the meaning of the phrase 'Having said that. ..' which can be gleaned from its Iexis and grammar to give

us any indication of its use. We have to know what value it has in discourse, how it isused to structure what follows.

Widdowson develops a distinction between language as a lexico-grammatical

system, which he refers to as language usage, and language as used for

communication, which he refers to as language use. One of his conclusions is that we

need to take much more account in our teaching strategies of language use. But the

problem here is that we know very little about language use. We do not, he argues,

have any adequate description of language use. We do not know enough about the

conventions of communication and about the way phrases, clauses and sentencescome to have a value quite separate from that of their component parts.


19/142


It must be stressed that the study of language in use is still in its early stages: we know verylittle at present about such matters as the way discourse develops and the way differentrhetorical activities are to be characterised. There is no source of reference for the teaching of use as there is for the teaching of usage. In these circumstances it is prudent not to be too

positive in one's recommendation.(Widdowson 1978)

This is true in the sense that we do not have an accepted model for the analysis and

description of discourse or for the classification and characterisation of rhetorical

activities. But we can still look at language in use and encourage learners to makegeneralisations about it. One thing, however, is sure. If we are to study language in

use, then what we must study is language in use. This is a tautology, but it is onewhich is often brushed aside:

. . . there has been for many years in English teaching a loss of respect for the natural patternsof a language. Because of the difficulty of analysing language that occurs in everyday

contexts, teachers have got in the way of accepting all sorts of invented or adapted texts.These are grimly defended by some, but there is no virtue in them; they were only made up because it was not practicable to harness real language. (Sinclair 1988)

Approaches which focus primarily on the form of a simplified and idealised

'language' are indeed unlikely to take us anywhere near the study of language in use.

If we are to study language in use then we must study real language designed to serve

some communicative purpose, rather than language simply designed to illustrate

aspects of usage. But a methodology based on presentation and practice is not

equipped to handle problems of use. As we have seen, the language involved is not

being used. Socoop's teacher, for example, when she says:

Yes, I do. I like being a father.

is not seen as expressing how she feels about fatherhood. Presentation and practice are

concerned purely and simply with usage.

The production stage following presentation and practice is also concerned

primarily with usage. When learners produce the present continuous with future

reference, their decision to use this form is not based on criteria of use. They do not

choose this form because it is the form which best expresses the meaning they want toexpress. They produce the form to demonstrate their familiarity with the aspect of

usage which is the focus of that particular lesson. Once learning targets have beenspecified in terms of form learners are predisposed to usage rather than use.

Use and usage in the classroom context

To a large extent the presentation methodology I have described above has replaced

the old grammar-translation approach. Grammar-translation was characterised by a

good deal of explanatory talk in the learner's first language,


20/142


with relatively little production of the target language on the part of either teacher or

learners. One of the features of presentation, practice and production is that there is agreat deal of the target language produced in the classroom, and perhaps this is the

reason for its relative success. If you observe very carefully a lesson based on presentation, or, even better, if you

look at transcripts of such lessons, you will probably notice two rather surprising

things. You will probably see that there is a lot of language produced in addition to

the language that presents and practises the target form. One reason is that teachers

use a lot of language to organise the lesson. They are constantly giving instructions

and explanations to give structure to the lesson and make sure that learners know what

is expected of them. Another reason is that a language lesson is a social event. There

is more to it than simply learning a language. Teachers and learners greet each other,tell stories, make jokes, get to know one another and do all the other things that

contribute to an easy social atmosphere.

Another thing you will notice about the language in a classroom, particularly in anelementary classroom, is that teachers produce a lot of language which is beyond the

level the learners are supposed to have reached. They do not, indeed they cannot,

restrict themselves to the very limited language which has already been presented. A

teacher might well begin a lesson, even at the elementary stage, by saying:

Okay, Unit 6. Could you turn to Unit 6? Right, Andreas, what about the first picture? What's in the first picture?

This would be quite unremarkable teacher behaviour even if learners have not yet

'done' the modal could or the phrase 'What about . . . ?'

By the same token, learners manage to get across meanings which are beyond

their target language resources. In the lesson featuring Socoop (see page 1) one of Socoop's classmates wanted to make the point that women often do jobs which are

traditionally regarded as a man's prerogative:

Victoria: (A woman) . . . He works, he . . . she works . . .

Teacher: Yes. Victoria: in sever(?) for her husband. Teacher: Mm? Victoria: He works teacher or, er engineering or many jobs, er, the sever in a man.

Teacher: The same. Victoria: The same

Teacher: As a man. (J R Willis 1981 )

One of the important things about the way a presentation methodology is realised by a

sensitive teacher is that it is language rich. Learners are involved in a lot of languageuse. But, paradoxically, this is not a deliberate part of the methodology. It is very

much a by-product of the methodology. But it would help to explain how learners

learn a lot of language which has not been presented to them. It would also help to

explain how in some cases, as in the case of any cited above (page 9), they may learn

something very different from what has been presented to them.


21/142


It is also important to remember that presentation and practice are only part of what

happens in a language teaching programme. The eclectic approach referred to earlier

(page 6) brings into the classroom activities involving listening and reading skills

which give much more, and much more varied, exposure to language than does a wellorganised and controlled presentation-based lesson. It is also the case that suchactivities are much less likely to have an overt language focus in the sense of being

targeted at a particular language form. In recent years such lessons have often been

referred to as 'skills-based' lessons. Perhaps this is an acknowledgement of the fact

that language use is a skill rather than a body of knowledge, and that the best way of

acquiring a skill is by practising that skill. This is, in fact, another way of asserting the

basic principle behind communicative methodology, that the best way to learn a

language is by using it to communicate.

It is certainly true that language use in its various manifestations involves theapplication of skills. But those skills operate on language. If, for example, learners are

being encouraged to predict the development of a text, they can, in the final event, dothis only on the basis of their knowledge and experience of what words and phrases in

texts are predictive and how they are predictive. To take an example quoted earlier,

when they hear a speaker produce the phrase 'Having said that . . . ', they are alerted to

the fact that what follows is likely to introduce some contradiction or modification of

what has been established so far.

It is likely, therefore, that the effectiveness of a skills-based approach to learning

would be considerably enhanced if we could identify the linguistic knowledge on

which particular skills operate. This takes us back to the need for a linguistic syllabus,and back to the contradiction that a linguistic syllabus is likely to lead to a focus on

form rather than use, whereas the strength of skills-based activities is that they are

based firmly in use. I am arguing that the presentation of language forms does not provide sufficient

input for learning a language. The grammatical system is much more complicated

than we can possibly reflect in a methodology which claims to rely on the

presentation of a very limited set of discrete patterns. In spite of this, a presentation

methodology seems to work tolerably well. I am suggesting that this is partly because

it is language rich. In spite of the fact that the methodology is based on presentation of

samples of usage, the methodology succeeds because it provides a context in whichthere is a great deal of language use.

This brings us back to the uneasy relationship between syllabus specification andmethodology. The syllabus specification must, directly or indirectly, consist of an

inventory of language forms. I have suggested, however, that a successful

methodology must rest on language use. The problem for a materials writer is to

produce a specification of those language items which a learner is likely to need and

then to match this with a methodology which involves a predominance of language

use. We must look for a methodology which aims quite deliberately at language use

rather than a methodology which offers language use as a by-product. We should try

to devise a methodology which is based on using language in the classroom toexchange meanings and which also offers a focus on language form, rather than a

methodology which focuses on language form and which only incidentally focuses on

use.


22/142

The Lexical Syllabus; Dave Willis 15Originally published by Collins ELT, 1990

CHAPTER 2: Words and structures

The Collins COBUILD English Course

In 1983 my wife, Jane, and myself were commissioned by Collins to write a new

English course to be called the Collins COBUILD English Course. Once a design had

been agreed we were to have overall responsibility for writing the course, but we were

not to be entirely free agents in drawing up the syllabus which would form the basis

of the course. A decision had already been taken that the syllabus would be lexically

based. Instead of specifying an inventory of grammatical structures or a set of

functions, each stage of the course would be built round a lexical syllabus. This would

specify words, their meanings, and the common phrases in which they were used.

Initially, the notion of a lexical syllabus gave us two grounds for concern. We both had firm ideas on the kind of methodology we would like to incorporate in anEFL course. It would be a task-based methodology firmly based in language use. We

were, however, for the reasons outlined in Chapter 1, far from sure that our ideas on

methodology would be compatible with a linguistically specified syllabus.

Secondly we had at that stage no real idea what a lexical syllabus would look like.

We were familiar with the idea of a syllabus built round grammatical patterns and

notions, and we were equally familiar with the idea of a functionally based syllabus.

We could not understand at first how a list of words with their meanings and common

phrases would be significantly different. It was only when we began to look at the

grammar of English very much from a lexical viewpoint that we began to see real

possibilities. We felt that a lexical approach might answer at least some of the doubtswe had so far entertained about structure-based pedagogical grammars, and about the

syllabus as an inventory of structures.

Priority and difficulty

Very often one of the striking features of ELT materials is the lack of balance in the

treatment of grammar. I have already suggested that the number of patterns presented

in most coursebooks gives a very restricted picture of the grammar of English. Most

courses spend a great deal of time on the verb phrase and on a limited set of clause

and sentence structures. Relatively little time is spent on some areas of English whichformal grammars find extremely difficult to handle, such as transitivity and the

structure of the noun phrase.

If we are to judge priorities by the amount of time afforded different features of

English, then tense, aspect and voice are seen by most coursebook writers as being of

overwhelming importance. In addition to this, a number of sentence patterns feature

heavily and take up a good deal of the learner's time. Among these are the three

conditionals. Another item which takes up a lot of time is


23/142


24/142

Words and Structures 17

Similarly frightened:

13 He was frightened of snakes.

.: is descriptive or stative. But:

14 He was frightened by a snake.

is dynamic.

But it is not only past participles that can be either stative or dynamic, with some

having the potential to be either. As we have already seen, the same is true of

adjectives:

Stative and dynamic adjectives differ in a number of ways. For example, a stative

adjective such as tall cannot be used with the progressive aspect or with the imperative: *He's being tall, *Be tall. On the other hand we can use careful as adynamic adjective: He's being careful, Be careful. (Quirk et al. 1972)

A Grammar of Contemporary English goes on to list well over fifty adjectives some

of them such as kind and nice extremely common - which can be used dynamically. It seems, therefore~that the only real distinguishing feature of the passive is the

use of by with a noun phrase to mark an agent. Rather than pick out the passive for

special treatment, an economical teaching strategy will allow the past participle to be

treated adjectivally. One of the consequences of this is that the collocation of be with

-ed forms is noted but not given undue prominence:

5 + -ed / -en

Your father's called John? and your mother's called Pat? (19)

It was built in 1890. (55)It was built for William Randolph Hearst. (55)This street is called Montague Street Precinct. (67). . .teenage girls who are interested in fashion. . . (95)

Are you tired?Wally is awakened by the phone ringing. (91)

. . .so that I can make sure that you are properly looked after. (193)

Listen for the words that are stressed. (103)

Once this is put together with:

by (111)

1 who / what did it

Wally is awakened by the phone ringing. (91 )

Handicrafts made by people in the Third World. (104)

Is that a magazine published by Macmillan? (146)

the learner has all that is needed to produce the passive. But the greatest prob


25/142

18 Thc Lexical Syllabus

lem with the passive is not form but use. Again, the teaching strategy proposed here

seems more likely to be effective than a transformational approach which relates the

passive closely to the active. If the participle is treated adjectivally it will quite

naturally be used when the focus of attention is on the subject of the passive verb. Thedifficulty is not with the sentence structure. This is no different from sentencestructure with adjectives. The difficulty lies in understanding that the past participle is

passive in meaning.

The second conditional

The COBUILD main corpus which was analysed to produce the Collins COBUILDEnglish Language Dictionary contains just under 15,000 occurrences of the word

would. It is the forty-fourth most frequent word in the COBUILD corpus, more

frequent than will, for example, which has 8,800 occurrences. In around half of its

15,000 occurrences would is described as 'used to talk of events which are of ahypothetical nature at the time of being mentioned, either because they are in the

future or because they depend on other events which may or may not occur'.

Examples include:The people of South Vietnam would receive their conquerors with relief / I think The Tempest

would make a wonderful film / I suspect that the West Germans would still be a little bitcautious.

In these examples a condition has been established earlier in the text, or is implied in

the word would. This use accounts for around 7,500 of the occurrences in the

COBUILD corpus. A sub-category of this, accounting for a further 1,200 occurrences,

is would used in explicitly conditional sentences:It would surprise me very much if sterling strengthened. / If he wasn't such a reactionary

I'd feel sorry for him.

In fact although many ELT grammars and coursebooks talk about the three

conditionals: 1 If it rains we'll get wet. 2 If it rained we would get wet. 3 If it had rained we would have got wet.

everyone is well aware that there are actually a very large number of possible

conditional patterns: 4 You can always explore the neighbourhood if you have half an hour to spare. 5. Even if I had the time I feel too tired.

6 If it got out it might kill someone. 7 If it's all right by you we could start now.

Why then does ELT practice isolate three patterns for special treatment?

All of the models, not only will and would, are common in conditional sentences.

Most of these models are taught lexically. Students learn that might and could, for example, are used for possibility. It is not thought necessary to teach a fourth and fifth

conditional like 6 and 7 above. Provided learners know what if means and they know

what might and could mean, it is assumed that they are capable of creating for

themselves sentences like 6 and 7. In exactly the same


26/142


way, if would is taught lexically with its main meaning of hypothesis, learners will be

well able to generate for themselves sentences like 2.

The strategy of highlighting word meanings is a much more productive one than

the strategy of teaching structural patterns. If the second conditional is taught as ameans of introducing learners to the most important meaning of would this seems tome to be an economical teaching strategy. Learners may then be led to the

generalisation that would also occurs in all kinds of environments without if . But this

is not what generally happens. The second conditional is taught as though it had some

life of its own, as if there was something unique about this combination of the past

tense and the modal would. But both these elements carry the meaning of hypothesis

quite independently of the second conditional. In fact would in conditionals is no

more difficult than might or could in conditionals. It is simply more common. This

again stems from its meaning, since conditional sentences are very much concernedwith hypotheses.

The Collins COBUILD English Course (CCEC) Level 2 includes a sectionentitled 'Your favourite cheap meal'.

This Language Study exercise simply draws learners' attention to the use of the

past tense and of would to express a hypothesis. It also makes the point that would is

preferred to will for an unreal hypothesis. Knowing the second conditional is not a

matter of being able to recite a particular pattern of words: it is a matter of knowing

the meaning of would and the meaning of this use of the past simple tense.

89 Your favourite cheap meal

a Jenny asked the others whae they would cook

for eheir favourite cheap meai for four people.David chose baked poraroes with a fiiiing of cheese and Jenny said she would do scrambledeggs on oast. Danny said he wouidn t cook

any~hing himself. He would go out for some pieand mashed potatoes. Jenny then asked them howmuch i'wouid cost to cook these things at home

and how much it would cost if they went out to acafe or restaurant.

89a Make notes about how much each meziwould cost. Compare your notes with a friend.

• Tell the class.

•

89b b Listen and see if you were right.

c What would members of your group cook andhow much would their mesis cost?

• Tell the class. Whose dish would be the bestvalue for money? Take a vote.

90 Language study

Would

a Look at the verbs in colour. What tense are theyin? Do they refer to past time?

JV: Are we ready? Yes. Erm now what would each of you cook if someone dropped in unexpectedly and stayed for a meal inthe evening?

JV: What would you cook David? DF: Whatever vegetables

happened to be there.JV: Supposing they arrived after the restaurants had shut.JV: But er and if you’d made it at home. . .

Why are they in the past tense?

b Look at these sentences. What does would mean? Whyis it would not will?

We asked Jenny Bridget David and Danny what they would cook for an unexpected guest.JV: What would you do Danny? DL: Would I have to cook them something. because I d

prefer to take them ouUor a meal.

JV It says here What would each ot you cook? . DL: Emm... JV: So. to summarise. Bndget would cook sausage and beans

Danny would cook an omelette David would cook

something exotic that he'd rustled up trom bits in the fridgeand I would cook a cheese flan.


27/142


Reported statements

It is a fact of the English language that the tense we select is liable to change if wetake a different standpoint in time. If George says'l'm tired' end I report this as 'George

said he was tired' I can choose the past tense because George's being tired occurred in

the past, rather than because the verb said is past tense. Even if George is still tired, I

may nevertheless choose to say 'George said he was tired.' But if George is still tired

and I want to make this clear I can choose to report what he said by saying 'George

said he's tired' or even 'George says he's tired.' So the choice between past and present

does not simply indicate when something happened. It may also indicate whether or

not I think the happening is still relevant. The fact that we sometimes have a choice between past and present tenses is not

simply a feature of reported speech. I might talk about something which happened in

the past by saying 'We stayed in the Grand Hotel. It was an awful place.' If the hotelstill exists and is still awful I can nevertheless choose to use the past tense if I do not

think my statement has any relevance to the present. On the other hand I can choose

to give my assessment some present relevance by selecting the present tense: 'We

stayed in the Grand Hotel. It's an awful place. You certainly shouldn't stay there.'

While preparing the CCEC materials we asked someone to rewrite a story as a

radio script. The story included this passage:

'What part of London are you headed for?' I asked him. 'I'm going right through London and out the other side,' he said. 'I'm going to Epsom, for the races. It's Derby Day today.' 'So it is,' l said. 'I wish I were going with you. I love betting on horses.' 'I never bet on horses,' he said. 'I don't even watch them run. That's a stupid silly

business.' 'Then why do you go?' I asked. He didn't seem to like that question. His little ratty face went absolutely blank and

he sat there staring straight ahead at the road saying nothing. 'I expect you help to work the betting machines or something like that,' I said. 'That's even sillier,' he answered . . . (Roald Dahl, The Hitch-hiker)

This summary was produced:

The other day I picked up a hitch-hiker who was heading for London and then going on to Epsomfor the Derby. I got very curious about him because it transpired that although he was going to the

Derby he didn't like horses or racing, he didn't bet on races and he didn't seem to have any kind of job at the race track.

The interesting thing about this is that although the second version reports what was

said there are no verbs of saying. There is no past tense verb like said to trigger atense change. The report is in the past tense because the reported events happened in

the past. There is nothing difficult about tense in reported speech in English. The logic it

follows is the same as for the rest of the language. In spite of this, many coursebooks

insist on regarding reported statement as a structure of some kind which has a system

of rules to itself. Instead of looking for broad generalisations about the language, there

is an attempt to cordon off sections and treat


28/142


them as if they were in some way unique. Reported speech, particularly the use of

tense, is treated in this way and is seen as creating great difficulties for learners, even

at quite an advanced level. One practice book for the Cambridge First Certificate, for example, solemnly liststhe 'rules' for reported speech. It explains that changes have to be made to certain

items with the result that this becomes the .. . or that, today becomes that day and I becomes he or she. To complicate the issue further, it is explained that if the reporting

verb is in the past tense then all the senses 'go one step backwards in time'. These

backwards steps are then listed. Present simple becomes past simple, present perfect

and past simple become past perfect and so on.

This is all totally unnecessary. These differences in person and in phrases of time

and place occur because we are taking a different standpoint from the original writer or speaker. It would be stupid to refer to something as happening today if I am well

aware that it happened several days ago. Similarly it would be silly if someone askedme the question:

Do you think I'll be late?

to reply by saying:

Yes I probably will.

We are constantly changing reference to person, time and place to accommodate thestandpoint of a different speaker at a different time. This is a feature of language as a

whole, not simply a feature of reported speech. It is a confusing and uneconomical

teaching strategy to single out reported statements and treat them as if they wereunique in some way.

In fact it is difficult to sustain the argument that reported statement is a useful

grammatical category at all. An analysis of noun clauses introduced by that in the

texts for CCEC Level 3 produced examples like these:

1 Cecil Sharp felt that the old songs of England might disappear for ever.

2 If it's a job interview try to show that you're interested in the job. 3 The government brought in a rule that children under thirteen werentt allowed to work. 4 The unsuccessful artist decided that his prayer had been answered.

5 The monkey said that there was no such thing as food, only fruit. 6 A long time ago there was this theory that women always passed first time.

Altogether in the texts which make up CCEC Level 3 there were 212 occurrences of

that used to introduce a noun clause. Of these 212 occurrences:

87 are introduced by verbs of thinking: think, feel, assume, decide, realise, understand, conclude,

believe, know, wish, recall, remember.40 by verbs of saying: say, tell, demand, report, explain, suggest, point out, assure, argue.

38 by nouns: rule, fact, idea, theory, problem, situation, thing, information, implication, promise, belief, impression, assurance, grounds, speculation, claim, announcement, signs, concern,conclusion, feeling, case, background.


29/142


13 by adjectives: glad, clear, sure, likely, incredulous, satisfied, convinced.34 by miscellaneous other words: show, see, it, except, mean, imply, turn out, hear, notice, pretend, reveal.

This tells us a number of things. First of all, comparatively few of the 212 occurrencescould accurately be described as reported speech. Reported thought is much more

common than reported speech. But reported thought does not figure in pedagogic

grammars with anything like the same inevitability as does reported speech. Secondly,

a large number of the occurrences, such as 2, 3 and 6 above, could not be described as

reports at all. Thirdly, noun clauses are by no means always dependent on a verb.

What, then, does the learner need to know about noun clauses of this kind? As I

have pointed out, many pedagogic grammars imply that the difficulty lies particularly

with tense, and with the changes in time and place reference. But I have argued thatthere is nothing unique about tense or about time and place in these noun clauses. I

would suggest that, as with the passive the most important thing about noun clauses isnot how they are formed but how they are used. They are used, for example, in the

way I have used them earlier in this paragraph with words like argue and suggest to

help develop an argument. They are used with nouns like thing, problem, situation

and theory to help define and develop ideas. In particular they have an important

function in identifying and highlighting a notion that is going to be developed in the

text:

thing . . . problem . . . The situation is(that)

theory . . . difficulty . . .

Once we begin to look at the uses of noun clauses, we begin to look at the words with

which they are associated, and to ask how those words function in text. In asking what

it is that the learner needs to know, and what it is that should be highlighted, we

acknowledge the importance of the noun clause, but we also come back to the

importance of the word as a unit of syllabus design.

English as a lexical language

I have suggested that three of the items traditionally regarded as difficult for thelearner are not in fact difficult in the way they are generally believed to be. They are

generally regarded as being difficult structures. I have argued in effect that the

passive and the conditionals do not need to be presented as 'structures', since they can

readily be created by learners for themselves, provided they have an understanding of

word meaning. This does not mean that they will necessarily be easily acquired by

learners. Even a rule as straightforward as the subject-verb concord in 'he rues' is not

easily acquired. It is a long time before it becomes a consistent part of the learner's

production. We do not know why this should be. Perhaps because it is heavilyredundant. We can never be sure when, or even whether, input will become part of the

learner’s behaviour.


30/142


31/142


This may or may not be the case with other languages, but it certainly seems to be

the case with English. It is perhaps particularly unfortunate that English has for so

long been described in terms of a Latinate grammar derived from a highly inflected

language, when English itself is quite different, a minimally inflected language.Obviously I would not claim that there is nothing more to English than word meaning,

but it does seem that word meaning and word order are central to English in a way

that may not hold true for other languages.

Difficulty in EFL - a re-assessment

Some of the grammatical systems of the language seem to operate a logic to which it

is very difficult for the learner to gain access. Perfective and progressive aspect in

English are notoriously difficult. A lot of time in elementary and intermediate coursesis spent contrasting the present and past simple, and the present and past continuous

tenses, and equally on contrasting the present perfect and the past simple. Another notorious area of difficulty in English is the system of determiners, particularly the

definite and indefinite article. This again is an area which receives a good deal of

attention in most courses. But the vexing thing about grammatical systems like these

is that they are conspicuously resistant to teaching. However hard teacher and learners

may try, some language systems take a long, long time to learn.

A number of theories have been put forward to account for this. It may be that

there is a fixed order of acquisition which is broadly speaking common to all learners.

There is some, though not conclusive, evidence for this view. Prabhu (1987) arguesthat any relationship between the grammatical systems as we describe them and

grammatical systems as they are subconsciously conceptualised by the learner

(between descriptive and operational s

44623773 Lexical Syllabus Willis

Documents

Transcript of 44623773 Lexical Syllabus Willis