Applied Linguistics Volume 1, No.2, 1980

Applied Linguistics Vol. I,

No.2,

Summer 1980

CONTENTS Page A R T I CL E S

The Essentials of a Communicative Curriculum i n Language Teaching. By MICHAEL P. BREEN and CHRISTOPHER N. CANDLIN Sociolinguistic Surveys: the State of the Art . By ROBERT L. cooPER Speech Acts and Second Language Learning. By RICHARD w scHMIDT and JACK C. RICHARDS

89 113 1 29

D I S CU S S I O N

Being interdisciplinary-Some Problems Facing Applied Linguistics. By C. J. BRUMFIT Models and Fictions. By H. G. WIDDOWSON

1 58 1 65

R E V I EW S

Evelyn Hatch, Second Language Acquisition: a Book o f Readings. By H. DOUGLAS BROWN

Betty Wallace Robinett, Teaching English t o Speakers of Other Languages. By cHRISTOPHER BRUMFIT Pierre Macherey, A Theory of Literary Production. By MICHAEL FISCHER

Peter Strevens, New Orientations in the Teaching of English. By ROBERT B. KAPLAN

Malcolm Coulthard, An Introduction to Discourse Analysis. By GILLIAN SANKOFF

Terence Hawkes, Structuralism and Semiotics; Roger Fowler, Linguistics and the Novel; E. L. Epstein, Language and Style. By M. H SHORT

171 1 73 1 75 1 77 1 79 1 80

A new description of intonation ... * takes into account the interactive significance of intonation * relates intonation to an existing description of discourse

structure * discusses the general place of intonation in language teaching

and how this particular description might be taught

Discourse Intonation and Language Teaching David Brazil, Malcolm Coulthard and Catherine Johns 224pp

ISBN 0 582 55366 0

£3.00 net

... and a new rationale for contrastive analysis * highlights bilingualism and language pedagogy * includes a wide range of examples, encouraging readers to

carry out their own analyses * assesses, for the first time, the research potential of

contrastive 'communicative' analysis within the macrolinguistic area of text analysis, discourse and pragmatics.

Contrastive Analysis Car/James 216pp

ISBN 0 582 55370 9

£3.00 net

Two new titles in the Longman Applied linguistics and language

Study Series. For a full list write to: Longman English Teaching Services Longman Group Limited, Burnt Mill, Harlow, Essex CM20 2JE

.. Ill ...

Longman=:

THE ES SENTIALS OF A COMMUNICATIVE CURRICULUM IN LANGUAGE TEACHING M I C H A E L P . B R E E N and C H R I S T O P H E R N . CAN O L I N University ofLancaster

I N T R O D U CT I O N

Applied Linguistics, Vol. I, No. 2

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

AT a time when there i s a recognised need i n language teaching t o give adequate attention to language use as well as language form, various 'notional-functional' or so-called 'communicative approaches' to language teaching are being advocated. In this context, the present paper is offered as a set of proposals in an effort to define the nature of communicative language teaching. Any teaching curriculum is designed in answer to three interrelated questions: What is to be learned? How is the learning to be undertaken and achieved? To what extent is the former appropriate and the latter effective? A communicative curriculum will place language teaching within the framework of this relationship between some specified purposes, the methodology which will be the means towards the achievement of those purposes, and the evaluation procedures which will assess the appropriateness of the initial ' purposes and the effectiveness of the methodology. This paper presents the potential characteristics of communicative language teaching in terms of such a curriculum framework. It also proposes a set of principles on which particular curriculum designs can be based for imple­ mentation in particular situations and circumstances. The diagram summarises the main areas with which this paper will deal. In discussing the purposes of language teaching, we will consider (I) communication as a general purpose, (2) the underlying demands on the learner that such a purpose may imply, and (3) the initial contributions which learners may bring to the curriculum. In dis­ cussing the potential methodology of a communicative curriculum, we will consider (4) the process of teaching and learning, (5) the roles of teacher and learners, and (6) the role of content within the teaching and learning. Finally (7) we will discuss the place of evaluation of learner progress and evaluation of the curriculum itself from a communicative point of view. 1 Inevitably, any statement about the components of the curriculum runs the risk of presenting in linear form a framework which is, in fact, characterised by interdependence and overlap among the components. In taking purposes, methodology, and evaluation in tum, therefore, we ask readers to bear in mind the actual interdependence between them. What follows is a consideration of those minimal requirements on com­ municative language learning and teaching, which, in our view, must now be taken into account in curriculum design and implementation.

90

E S S E N T I A L S OF A C O M M U N I C A T I V E C U R R I C U L U M THE CURRICULUM

4 . The classroom

I. Commumcat1on

process

2.

Demands on the--� learner

3.

Learner's an1t1al contnbut1ons

6.

1.

Ofleamer

Ofcumculum

W H A T I S T H E P U R P O SE O F T H E C U R R I C U L U M ?

The communicative curriculum defines language learning as learning how to communicate as a member of a particular socio-cultural group. The social conventions governing language form and behaviour within the group are, therefore, central to the process of language learning. In any communicative event, individual participants bring with them prior knowledge of meaning and prior knowledge of how such meaning can be realised through the con­ ventions of language form and behaviour. 2 Since communication is primarily interpersonal, these conventions are subject to variation while they are being used. In exploring shared knowledge, participants will be modifying that knowledge. They typically exploit a tension between the conventions that are established and the opportunity to modify these conventions for their par­ ticular communicative purposes . Communicating is not merely a matter of following conventions but also of negotiating through and about the con­ ventions themselves. It is a convention-creating as well as a convention­ following activity. So, in learning how to communicate the learner is confronted by a variable process. In communication, speakers and hearers (and writers and readers) are most often engaged in the work of sharing meanings which are both dependent on the conventions of interpersonal behaviour and created by such behaviour. Similarly, the ideas or concepts which are communicated about contain dif­ ferent potential meanings, and such potential meanings are expressed through and derived from the formal system of text during the process of com­ munication. To understand the conventions which underlie communication, therefore, we not only have to understand a system of ideas or concepts and a system of interpersonal behaviour, we have to understand how these ideas and this interpersonal behaviour can be realised in language-in connected texts. Mastering this unity of ideational, interpersonal and textual knowledge allows us to participate in a creative meaning-making process and to express or inter­ pret the potential meanings within spoken or written text . 3 There is an additional characteristic of this unified system of knowledge. The social or interpersonal nature of communication guarantees that it is

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

7

Role ofcontent

M I CHA E L P. B R E E N A N D CH R I ST O P H E R N . CA N D L I N

91

2.

WHAT U N D E R L I E S T H E U L T IMATE DEMANDS O N T H E LEAR N E R ?

A language teaching curriculum, from a communicative point o f view, will specify its purposes in terms of a particular target repertoire. 4 Different curricula will hopefully select their own particular repertoires from a pool of communicative performance on the basis of a sociolinguistic analysis of the target situation. This does not imply that any one curriculum will be neces­ sarily entirely distinctive in the target repertoire to which it is devoted . At the surface there will be inevitable overlap among different repertoires. However, underlying any selected target repertoire there will. be an implicit target com­ petence. It is this target competence which we may define as the capacity for actual use of the language in the target situation. So, in specifying the purposes of the curriculum, a requirement for the communicative approach would be to make an initial distinction between the target repertoire ultimately demanded of the learner and the target competence which will underlie and generate such a repertoire. How can we characterise this target competence? We have already proposed that learning to communicate involves acquiring a knowledge of the con­ ventions which govern communicative performance. In addition, we have proposed that such communicative knowledge can be seen as a unified system

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

permeated by personal and socio-cultural attitudes, values and emotions. These different affects will determine what we choose to communicate about and how we communicate. The conventions governing ideas or concepts, inter­ personal behaviour, and their realisation in texts all serve and create attitudes, judgements and feelings. Just as communication cannot be affectively neutral, learning to communicate implies that the learner will come to terms with the new learning to the extent that his own affects will be engaged. At that point, the learner's affects become further involved in a process of negotiation with those affects which are embodied within the communicative performance of the target community. So, affective involvement is both the driving-force for learning, and also the motivation behind much everyday communication and the inspiration for the recreation of the conventions which govern such communication. Communication in everyday life synthesises ideational, interpersonal, and textual knowledge-and the affects which are part of such knowledge. But it is also related to and integrated with other forms of human behaviour. In learning how to communicate in a new language, the learner is not confronted by a task which is easily separable from his other psychological and social experiences. The sharing and negotiating of potential meanings in a new language implies the use and refinement of perceptions, concepts and affects. Furthermore, learning the conventions governing communication within a new social group involves the refinement and use of the social roles and the social identity expected by that group of its members. Thus, learning to com­ municate is a socialisation process. In much of his previous experience the learner has seen communication as the basic means whereby human activity and consciousness is shared and reflected upon socially. Therefore, it makes sense for the teacher to see the overall purpose of language teaching as the development of the learner's communicative knowledge in the context of personal and social development.

92

E S S E N T I A L S OF A C O M M U N I CA T I V E C U R R I C U L U M

o f ideational, interpersonal, and textual knowledge, which incorporates a range of affects. We have also suggested that communication and learning how to com­ municate involve the participants in the sharing and negotiating of meanings and conventions. Such sharing and negotiating implies the existence of particular communicative

abilities as an

•

essential part of competence. Therefore,

we may identify within competence both the knowledge systems and the abilities which call upon and act upon that knowledge. These abilities can be distinguished within competence more precisely. In order to share meaning, the individual participant needs to be able to interpret the meanings of others and to express his own meanings. However, such interpretation and expression will most often take place in the contex,t of interpersonal and personal

participants in communication negotiate with one another. But, in en­ deavouring to interpret and express with a new language, the learner will himself negotiate between the communicative competence he already possesses and that which underlies the new learning. s We suggest, therefore, that the communicative abilities of interpretation, expression, and negotiation are the essential or 'primary' abilities within any target competence. It is also likely that these three abilities continually inter­ relate with one another during communicative performance and that they are complex in nature. They will involve psychological processes for the handling of rich and variable data-the attention and memory processes for example­ and they may contain within them a range of secondary abilities. 6 The use of these communicative abilities is manifested in communicative per­ formance through a set of skills. Speaking, listening, reading and writing skills can be seen to serve and depend upon the underlying abilities of interpretation, expression and negotiation. In this way we are suggesting that the skills repre­ sent or realise underlying communicative abilities. The skills are the meeting point between underlying communicative competence and observable com­ municative performance; they are the means through which knowledge and abilities are translated into performance, and vice versa. In selecting any target repertoire, therefore, a communicative curriculum also distinguishes and specifies the target competence on which the er­

p

formance of such a repertoire depends and through which it is achieved. This specification would indicate the ideational, interpersonal and textual con­ ventions-and the affective aspects of such conventions-as a related and underlying system of knowledge which is shared and developed within the target community. The specification would also indicate the demands upon the learner's communicative abilities of interpretation, expression, and negotiation similarly underlying communicative performance in the target community-and the range of skills which manifest these abilities. Such a specification would account for what the learner needs to know, and how the learner needs to be able to use such knowledge. The ultimate demands on the learner in terms of some specific target repertoire will, in our view, derive from and depend upon this underlying competence of communicative knowledge and communicative abilities.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

negotiation. The ability to negotiate operates between participants in com­ munication and within the mind of the individual participant-the latter negotiation is perhaps more conscious during new learning. More obviously,

M I C H A E L P. B R E E N A N D C H R I ST O P H E R N . C A N D L I N

93

3. W H A T A R E T H E L E A R N E R 'S I N I T I A L C O NT R I B U T I O N S ?

A communicative curriculum will focus on the learner from the very beginning by relating the initial contributions of the learner to the ultimate purposes of the curriculum. More precisely, the communicative curriculum seeks relationships between any specific target competence and relevant aspects of the learner's own initial competence. We need to ask: What com­ municative knowledge-and its affective aspects-does the learner already

roots of our objectives can already be discovered in our learners-however beneath the surface of the actual target repertoire these roots may be. We need to try to recognise what the learner knows and can do in communicative per­ formance with the first. language and not assume that the learner's ignorance of the target repertoire implies that the learner is a naive communicator or someone who evaluates communication in only a superficial way. This principle, which seems to require us to credit the learner with a highly relevant initial competence of communicative knowledge and abilities, has often been overlooked or only partially applied in language teaching. In the past, it has seemed easier to somehow separate the learner from the knowledge to be learned-to 'objectify' the target language as something completely unfamiliar to the learner. This objectification of the language in relation to the learner has perhaps been encouraged by a narrow definition of what the object of learning actually is, and by an incomplete view of what the learner has to offer. We have tended to see the target only in terms of 'linguistic competence' or textual knowledge, and we have limited such knowledge to the level of syntax without reference to structure above the sentence. Thus, ideational and interpersonal knowledge, which continually interact with textual knowledge and from which textual knowledge evolves, have tended to be overlooked or neutralised. We have often seen the learner primarily in terms of the frrst language, and we have often assigned to it 'interference' value alone-again taking a narrow textual knowledge as our criterion. More recently, due to developments within sociolinguistics, we have recognised the significance of 'sociolinguistic competence' and also of the 'functional' aspect of language. 7 However, a partial and knowledge-based view of learner competence seems to remain with us and the learner's communicative abilities underlying the initial repertoire still need to be more thoroughly exploited. Rather than just allowing the use of the frrst language in the classroom, we should perhaps be more con­ cerned with activating that which underlies the initial repertoire of the learner, and to evoke and engage what we may describe as the learner's ongoing or process competence. 1 Once we define the object of learning as communication, then we are enabled to perceive the learner in a new light. His initial textual

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

possess and exploit? What communicative abilities-and the skills which manifest them-does the learner already activate and depend upon in using and selecting from his presently established repertoire? Also, can the curriculum build upon features of that performance repertoire which we describe rather narrowly, perhaps, as the learner's first language or mother tongue? Similarly, can the curriculum build upon what the learner may already know of and about the target repertoire-however fragmentary or 'latent' such an awareness may be? A communicative specification of purposes supports the principle that the

94

E S S E N T I A L S OF A CO M M U N I CA T I V E CU R R I CU L U M

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

knowledge is placed in its proper perspective-it is merely the tip o f the ice­ berg. Language teaching need no longer be primarily concerned with 'linguistic competence'. We can begin with the assumption that text is the sur­ face realisation of communicative knowledge and abilities and that text is used and created-and learned-on the basis of them. The communicative curriculum seeks to facilitate-even guarantee-the involvement of the learner's communicative knowledge and abilities from the outset rather than overlook them for the sake of some apparent 'fluency' with text. If we present the learner with language only as an object, as if it was separable from the learner's relevant psychological and social experience, we are almost certainly postponing development of the learner's ability to communicate through the language. We may be divorcing language learning from its essential inter­ personal nature and offering it as a static object to learners who have them­ selves experienced the use of their first language as very much an interpersonal undertaking. The purposes of the communicative curriculum will incorporate that which the learner already knows and can do as a communicator from the very start. However, learners not only contribute prior knowledge and abilities, they also have expectations about the learning of a language. What the curriculum seeks to achieve in terms of any specified purposes must be balanced by what the learner personally expects of the curriculum. Perhaps the current interest in teaching language for 'special purposes' may eventually reveal the chalienge to curriculum designers: that all learners regard themselves as learning a language for some special purpose. We can identify several types of learner expectations and these may, of course, influence one another. We can ask: What is the learner's own view of the nature of language? What is the learner's view of learning a language? (The answers to these questions may lie in the learner's previous formal education, and how he reacted to that experience.) We can also distinguish between, first, how the learner defines his own language learning needs; secondly, what is likely to interest the learner both within the target repertoire and the learning process; and, third, what the learner's motivations are for learning the target repertoire. All these initial expectations are distinct and need to be discovered in some way so that areas of potential match and mis­ match between learner expectations and the selected target repertoire and its underlying competence can be best anticipated. Two important problems need to be identified here in accounting for learner expectations. These expectations are inevitably various and-more signi­ ficantly-they are subject to change over time. So, the curriculum will need to accommodate and allow for a heterogeneity of learner expectations. It will also need to allow for changes in different learners' perceptions of their needs, in what interests different learners, and in the motivations of different learners. In this way, curriculum purposes should account for initial ex­ pectations of learners and anticipate changes in expectations during the learning-teaching process. Such an account and such anticipation may appear to be an impracticable dream when confronted with the variety and fluctuation in the real expectations of learners. That we should try to account for and anti­ cipate these is a further motivation for a communicative curriculum, and­ more particularly-for a communicative methodology (see sections 4ff). How-

'

M I C H A E L P. B R E E N A N D C H R I S T O P H E R N . C A N D L I N

95

ever, there is a second important aspect of learner expectations: expectations can be educated. For this to happen, learners need to be enabled to express their own expectations; to explore them and the sources from which they derive. They also need to be enabled to interpret the expectations which the specific purposes of the curriculum make upon them as learners. They need to interpret-at the start of the learning-teaching process and throughout this process-what the target repertoire and its underlying competence demands of them. However vague a learner's initial interpretation may be, he is not going to learn anything unless he has an idea of what he is trying to achieve. There­ fore, a process of negotiation between the learner's contributions-including expectations-and the target repertoire, and the means by which these two are brought together, is likely to be characteristic of a communicative metho­

these subjective contributions of the learner and, thereby, call upon the genuine intersubjective responsibility of that learner.

4.

H O W A R E T H E C U R R I C U L U M P U R P O S E S TO BE A C H I E V E D ?

4.1 Methodology as a Communicative Process Language learning within a communicative curriculum is most appro­ priately seen as communicative interaction involving all the participants in the learning and including the various material resources on which the learning is exercised. Therefore, language learning may be seen as a process which grows out of the interaction between learners, teachers, texts and activities. This communicative interaction is likely to engage the abilities within the learner's developing competence in an arena of cooperative negotiation, joint interpretation, and the sharing of expression. The communicative classroom can serve as a forum characterised by the activation of these abilities upon the learners' new and developing knowledge. This activation will depend on the provision of a range of different text-types in different media-spoken, written, visual and audio-visual-which the participants can make use of to develop their competence through a variety of activities and tasks. 9 The presence of a range of text-types acknowledges that the use of communicative abilities is not restricted to any one medium of communication. The earlier distinction we saw between underlying abilities and the set of skills which serve and depend on such abilities enables us to perceive that the learner may exploit any selected skill or combination of skills to develop and refine his inter­ pretation, expression and negotiation. The learner need not be restricted to the particular skills performance laid down by the target repertoire. Because communicative abilities permeate each of the skills, they can be seen to under­ lie speaking, hearing, reading and writing and to be independent of any prescribed selection or combination of these skills. Similarly, just as no single communicative ability can really develop independently of the other abilities, so the development of any single skill may well depend on the appropriate development of the other skills. In other words, a refinement of interpretation will contribute to the refinement of expression, and vice-versa; just as a refine­ ment of the skill of reading, for example, will contribute to the refinement of 10

the skill of speaking and vice-versa. Classroom procedures and activities can involve participants in both

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

dology. Curriculum purposes inform and guide methodology, and an account of lea�ner expectations within purposes can enable.methodology to involve

E S S E N T I A L S OF A C O M M U N I C A T I V E C U R R I C U L U M

96

communicating and metacommunicating. We have referred to the charac­ teristics of communicating in section 1 of this paper. By metacommunicating we imply the learner' s activity in analysing, monitoring and evaluating those knowledge systems implicit within the various text-types confronting him during learning. Such metacommunication occurs within the communicative performance of the classroom as a sociolinguistic activity. in its own right. Through this ongoing communication and metacommunication, learners not only become participants in the procedures and activities, they may also become critically sensitised to the potential and richness of the unified system of knowledge, affects and abilities upon which their communication depends.

cipants in a process of communicating through texts and activities, and meta­ communicating about texts, is likely to exploit the productive relationship between using the language and learning the language.

4.2 Methodology as a differentiated process The emphasis given in the previous section to the interactive nature of the communicative curriculum suggests, in turn, the need for a communicative curriculum to be differentiated. A communicative curriculum begins with the principle that we should differentiate within purposes between the target repertoire and the communicative knowledge and abilities which underlie it. A second principle is that the learner's process competence needs to be dif­ ferentiated from the target competence, and that different learners may exploit different process competences as the means towards some particular target. These kinds of distinctions involve differentiation at the curriculum level between purposes and the methodology adopted to achieve such purposes. Within methodology, differentiation is a principle which can be applied to the participants in the learning, the activities they attempt, the text-types with which they choose to work, and the ways they use their abilities. It is worth considering differentiation within these areas in more detail: (a)

Learners' Contributions:

Individual learners bring individual contributions to the language learning process in terms of their initial competence, their various expectations about language learning, and their changing needs, interests and motivations prior to and throughout the language learning process. Also, within the language learning process, different learners develop different process competences­ different degrees of communicative knowledge and different ways of using their abilities in acting upon and with such knowledge. We can recognise that, even in the achievement of some common target competence, different learners may well adopt different means in attempting to achieve such com­ petence.

(b) Routes: The emphasis within a communicative curriculum on the communicative process of language learning, with the consequent emphasis on cooperative learner activities, offers a natural means for differentiation. Different learners need the opportunity of following different routes to the accomplishment of

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Thus, in working together upon the target competence in a context of class­ room communication, learners may constantly refine their own awareness of its potential, make links to their initial competence, and develop and exploit their own process competences. In particular, the involvement of all the parti­

M I C H A E L P. B R E E N A N D C H R I ST O P H E R N . C A N D L I N

97

some individual or common group objectiv-e. Such variation in choice of route typically involves selection among alternative skills or combinations of skills, and hence the choice of alternative media. The variation may be motivated by the need to work at a different pace from other learners, or by the desire to pursue alternative content. This selection among routes can itself be open to joint interpretation, the sharing of expression and cooperative negotiation.

(c) Media:

(d) Abilities:

Whatever the route chosen or the media and text-types selected for com­ municative learning, different learners will have differentiated ways of making use of the abilities within their communicative competence, and will therefore adopt different learning strategies. 11 Such heterogeneity is often seen as problematic for the teacher, but a communicative methodology would take advantage of this differentiation among learning strategies, rather than in­ sisting that all learners exploit the same kinds of strategy.

These four illustrations of the principle of differentiation within a com­ municative methodology imply more than merely offering to individual learners opportunities for differential communication and learning, or acknowledging differences between performance repertoires and the developing competences underlying them. Differentiation demands and authenticates communication in the classroom. The various perspectives offered by alternative media, the accomplishment of shared objectives through

a variety of routes, and the opportunities for exploiting different learning strategies, all facilitate the conditions for authentic communication among the participants in the learning. Differentiation also enables the learner to authen­ ticate his own learning and thereby become involved in genuine com­ munication as a means towards it. Further, if we confront learners with texts and text-types which are alsp authentic, this obliges us to allow for different interpretations and differences in how learners will themselves negotiate with texts. Generally, then, the search for authentication during language lear­ ning-for the individual and the group-and the need for differentiation are complementary.

4.3 Methodology exploits the communicative potential of the learning­

teaching context

We are easily tempted to excuse the classroom as an artificial or synthetic

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

In order to allow for differences in personal interest and ease of access, or to permit the search for alternative perspectives on the content, learners should be offered the possibility of working with one or more of a range of media. Each medium will have its own distinctive set of authentic text-types. For: a text-type to be authentic it must obey the principle of verisimilitude-or truth to the medium. We mean by this that learners would be expected to act upon text-types in the appropriate medium: written texts would be read, spoken ones listened to, visual ones seen. Just as communication is governed by con­ ventions, so we can see that the different media represent and obey con­ ventions specific to themselves. Learning dialogue by reading, for example, may neutralise the authentic conventions of spoken discourse, and we may be asking the learner to become involved in using and applying knowledge in a distorted way.

98

E S S E N T I A L S O F A COM M U N I CAT I V E CUR R I CU L U M

language learning context-as distinct from some natural o r authentic en­ vironment. The communicative curriculum seeks to exploit the classroom in terms of what it

can

realistically offer as a resource for learning. This would

not necessarily mean changing or disguising the classroom in the hope that it will momentarily serve as some kind of 'communicative situation' resembling situations in the outside world. The classroom itself is a unique social environ­ ment with its own human activities and its own conventidns governing these activities. It is cultural reality communicative be overcome or

an environment where a particular social-psychological and is constructed. This uniqueness and this reality implies a potential to be exploited, rather than constraints which have to compensated for. Experimentation within the prior constraints

learning by, on the one hand, the 'formal' language learning contexts of the classroom and, on the other, the 'informal' learning which takes place at any time, anywhere. The classroom can be characterised by the kinds of learning which are best generated in a group context, while 'informal' learning under­ taken beyond the classroom is often an individual commitment, especially in the context of foreign language learning. Thus the 'formal' context is one where the interpersonal relationships of the classroom group have their own potential contribution to make to the overall task. Within the communicative curriculum, the classroom-and the procedures and activities it allows-can serve as the focal point of the learning-teaching process. In adopting a methodology characterised by learning and teaching as a communicative and differentiated process, the classroom no longer needs to be seen as a pale representation of some outside communicative reality. It can become the meeting-place for realistically motivated communication-as-learning, com­ munication about learning, and metacommunication. It can be a forum where knowledge may be jointly offered and sought, reflected upon, and acted upon. The classroom can also crucially serve as the source of feedback on, and refinement of, the individual learner's own process competence. And it can serve as a springboard for the learner's 'personal curriculum' which may be undertaken and developed 'informally' outside the classroom. As a co­ participant in the classroom group, the learner's own progress can be both monitored and potentially sustained by himself on the basis of others' feed­ back and by others within some shared undertaking. To ensure that the special and differing contributions offered by both 'formal' and 'informal' contexts of learning can be fully exploited, a com­ municative methodology has to try to relate the two. The classroom can deal with and explore phenomena which are significant in the experienced 'outside world' of the learner, and it can become an observatory of communication as everyday human behaviour. As well as looking outwards, the classroom has a reflexive role as a laboratory where observations can become the means for the discovery of new knowledge and the development of abilities. A communicative methodology will therefore exploit the classroom as a resource with its own communicative potential. The classroom is only one resource in language teaching, but it is also the meeting-place of all other

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

of any communicative situation is, as we have seen, typical of the nature of communication itself, and the prior constraints of classroom communication need be no exception. We can make a distinction between the different contributions offered to

M IC H A E L P . BREEN AND CHR ISTOPH E R N . CANDLIN

99

resources-learners, teachers, and texts. Each of these has sufficiently hetero­ geneous characteristics to make classroom-based negotiation a necessary undertaking. The classroom is 'artificial' only if we demand of it that which it cannot achieve-if, for example, we treat it as a rehearsal studio where 'actors' learn the lines from some pre-scripted target repertoire for a performance at some later time and place. The authenticity of the classroom lies in its dual role of observatory and laboratory during a communicative learning-teaching process.12

5.

WHAT ARE T H E

R O L E S OF T H E T E A C H E R A N D T H E L E A R N E R S

W I T H I N A CO M M U N I C A T I V E M E T H ODOLO G Y ?

5.1 The Teacher

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Within a communicative methodology the teacher has two main roles. The first role is to facilitate the communicative process between all participants in the classroom, and between these participants and the various activities and texts. The second role is to act as an interdependent participant within the learning-teaching group. This latter role is closely related to the objective of the first role and it arises from it. These roles imply a set of secondary roles for the teacher: first , as an organiser of resources and as a resource himself. Second, as a guide within the classroom procedures and activities. In this role the teacher endeavours to make clear to th _ e learners what they need to do in order to achieve spme specific activity or task, if they indicate that such guidance is necessary. This guidance role is ongoing and largely unpredictable, so the teacher needs to share it with other learners. Related to this, the teacher-and other learners-can offer and seek feedback at appropriate moments in learning-teaching activities. In guiding and monitoring the teacher needs to be a 'seer of potential' with the aim of facilitating and shaping in­ dividual and group knowledge and exploitation of abilities during learning. In this way the teacher will be concentrating on the process competences of the learners. A third role for the teacher is that of researcher and learner-with much to contribute in terms of appropriate knowledge and abilities, actual and ob­ served experience of the nature of learning, and organisational capabilities . As a participant-observer, the teacher has the opportunity to 'step back' and monitor the communicative process of learning-teaching. As an interdependent participant in the process, the teacher needs to actively share the responsibility for learning and teaching with the learners. This sharing can provide the basis for joint negotiation which itself releases the teacher to become a co-participant. Perceiving .the learners as having im­ portant contributions to make-in terms of initial competence and a range of various and changing expectations-can enable the teacher to continually seek potential and exploit it. A requirement on the teacher must be that he distinguish between learning and the performance of what is being learned. The teacher must assume that the performance within any target repertoire is separable from the means to the achievement of that repertoire. Also, he must assume that learners are capable of arriving at a particular objective through diverse routes. The teacher needs to recognise learning as an interpersonal undertaking over which no single person can have full control, and that there will be differences between ongoing learning processes. The teacher has to

100

E S S E N T I A L S OF A C O M M U N I CA T I V E C U R R I C U L U M

accept that different learners learn different things in different ways a t dif­ ferent times,' and he needs to be patiently aware that some learners, for example, will enter periods when it seems that little or no progress is being made and that, sometimes, learning is typified by silent reflection. 13

5.2 The Learner Regardless of the curriculum in which they work and regardless of whether or not they are being taught, all learners of a language are confronted by the task of discovering how to learn the language. All learners will start with dif­ fering expectations about the actual learning, but each individual learner will be required to adapt and continually readapt in the process of relat�ng himself

Within a communicative methodology, the role of learner as negotiator­ between the self, the learning process, and the object of learning-emerges from and interacts with the role of joint negotiator within the group and within the classroom procedures and activities which the group undertakes. The implication for the learner is that he should contribute as much as he gains, and thereby learn in an interdependent way. The learner can achieve interdependence by recognising responsibility for his own learning and by sharing that responsibility with other learners and the teacher. A further implication is that the learner must commit himself to undertake com­ municative and metacommunicative acts while working with other participants in the group, and while working upon activities and texts. This commitment

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

to what is being learned. The knowledge will be redefined as the learner un­ covers it, and, in constructing and reconstructing his own curriculum, the learner may discover that earlier strategies in the use of his abilities need to be replaced by other strategies. Thus, all learners-in their own ways-have to adopt the role of negotiation between themselves, their learning process, and the gradually revealed object of learning. A communicative methodology is characterised by making this negotiative role-this learning how to learn-a public as well as a private undertaking. Within the context of the classroom group, this role is shared and, thereby, made interpersonal. If we recognise that any knowledge which we ourselves have mastered is always shared knowledge and that we always seek con­ firmation that we 'know' something by communicating with other people, we have to conclude that knowledge of anything and the learning of anything is an interpersonal matter. Also, if we recognise that real knowledge is always set in a context and this context is botli' psychological and social-what is known will always be contextualised with other knowledge in our minds and will always �arry with it elements of the social context in which it was experienced-then we also have to conclude that a significant part of our learning is, in fact, socially constructed. These justifications for a genuinely interpersonal methodology are quite independent of the nature of what is to be learned. If the object of learning is itself communication, then the motivation to enable the learner to adopt an interpersonal means to that learning is doubly justified. Quite simply, in order to learn to communicate within a selected target repertoire, the learner must be encouraged to communicate-to communicate about the learning process, and to communicate about the changing object of learning on the basis of accepting that 'learning how to learn' is a problem shared, and solved, by other learners.

M I C H A E L P. B R E E N A N D C H R I ST O P H E R N. C A N D L I N

101

6.

WHAT

IS THE

ROLE

OF

CONTENT

WITHIN A COMMUNICATIVE

M ET H ODOLOGY?

Language teaching curricula have often been traditionally defined by their content. Such content has itself been derived from a target repertoire in terms of some selected inventories of items analysed prior to the commencement of the teaching-learning process and often acting as predeterminants of it. Similarly, sets of formal items taken from an analytic gr ammar of the language, or sets of 'functions' taken from some list of semantic categories, have been linked to themes and topics deemed in advance to be appropriate to the expectations of the particular learners. Communicative curricula, on the other hand, do not look exclusively to a

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

can be initiated and supported by a milieu in which the learner's own con­ tributions-interpretations, expressions, and efforts to negotiate-are recognised as valid and valuable. Such a context would be typified by the acceptance of ongoing success and failure as necessary prerequisites towards some ultimate achievement, where it is assumed that learners inevitably bring with them 'mixed abilities' and that such ·a 'mixture' is, in fact, positively useful to the group as a whole. Commitment to communication on the lear­ ner's part need not be regarded as something unattainable or threatening­ even for the 'beginning' learner-because he is expected to rely on and develop that which is familiar: his own process competence and experience of com­ munication. As an interdependent participant in a cooperative milieu where the learner's contributions are valued and used, the individual learner is potentially rewarded by having his own subjective expectations and decisions informed and guided by others. In a context where different contributions and dif­ ferential learning are positively encouraged, the learner is allowed to depend on other learners and on the teacher when the need arises, and also enabled to be independent at appropriate moments of the learning. He can feel free to exploit independent strategies in order to learn, to maintain and develop per­ sonal affective motivations for learning, and to decide on different routes and means which become available during learning. The paradox here, of course, is that genuine independence arises only to the extent that it is interdependently granted and interdependently accepted. Learning seen as totally a personal and subjective matter is seeing learning in a vacuum; indeed we may wonder whether such learning is ever possible. Learners also have an important monitoring role in addition to the degree of monitoring which they may apply subjectively to their own learning. The learner can be a provider of feedback to others concerning his own inter­ pretation of the specific purposes of the curriculum, and the appropriateness of methodology to his own learning experiences and achievements. In ex­ pression and negotiation, the learner adopts the dual role of being, first, a potential teacher for other learners and, second, an informant to the teacher concerning his own learning progress. In this latter role, the learner can offer the teacher and other learners a source for new directions in the learning­ teaching process of the group. Essentially, a communicative methodology would allow both the teacher and the learner to be interdependent participants in a communicative process of learning and teaching.

102

E S S E N T I A L S OF A C O M M U N I C A T I V E C U R R I C U L U M

selected target repertoire as a specifier o f curriculum content, for a number of reasons. First, the emphasis on the process of bringing certain basic abilities to bear on the dynamic conventions of communication precludes any specification of content in terms of a static inventory of language items­ grammatical or 'functional'-to be learned in some prescribed way. Second, the central concern for the development and refinement of underlying com­ petence as a basis for a selected target repertoire requires a distinction between

(a) Focus From what has been proposed in this section so far, it follows that content within communicative methodology is likely to focus upon knowledge-both cognitive and affective-which is personally significant to the learner. Such knowledge would be placed in an interpersonal context which can motivate personal and joint negotiation through the provision of authentic and problem-posing texts. If content is to be sensitive to the process of learning and to the interpersonal concerns of the group, it needs to reflect and support the integration of language with other forms of human experience and behaviour.

(b) Sequence If we accept that the communicative process requires that we deal with dynamic and creative conventions, we cannot assume that any step-by-step or cumulative sequence of content will necessarily be appropriate. In learning, the various and changing routes of the learners crucially affect any ordering of content, so that sequencing derives from the state of the learners rather than from the implicit 'logic' of the content itself. Just as any movement from

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

that target and any content which could be used as a potential means towards it. Third, the importance of the curriculum as a means for the activation and refinement of the process competences of different learners, presupposes dif­ ferentiation, ongoing change, and only short-term predictability in what may be appropriate content. The communicative curriculum would place content within methodology and provide it with the role of servant to the learning-teaching process. Thus, content would not necessarily be prescribed by purposes but selected and organised within the communicative and differentiated process by learners and teachers as participants in that process. Therefore, the learner would use the content of the curriculum as the 'carrier' of his process competence and as the provider of opportunities for communicative experiences through which per­ sonal routes may be selected and explored as a means to the ultimate target competence. From this concern with means rather than ends-with the process of learning-teaching rather than with the product-the communicative curriculum will adopt criteria for the selection and organisation of content which will be subject to, and defined by, communicative learning and teaching. The content of any curriculum can be selected and organised on the basis of some adopted criteria, and these criteria will influence five basic aspects of the content: its focus, its sequence, its subdivision (or breakdown), its continuity, and its direction (or routing). We will now consider the possible criteria for the selection and organisation of content within the communicative curriculum with reference to each of these five aspects in turn:

M I C H A E L P . B REEN A N D C H R I STOP H E R N . CAND L I N

103

'simple' to 'complex' is a very misleading way of perceiving the relationship between any text and its meaning potential-a simple text may realise complex

(c) Subdivision

Traditionally content has been subdivided into serialised categories of structures or 'functions'. A communicative view of content precludes this fragmentation and argues for subdivision in terms of whole frameworks wherein there is interaction between all the various components of the know­ ledge system-ideational, interpersonal and textual-and all the abilities involved in using such knowledge. Content would be subdivided or broken down in terms of activities and tasks to be undertaken, wherein both know­ ledge and abilities would be engaged in the learners' communication and meta­ communication. The various activities and tasks would be related by sharing a holistic 'core' of knowledge and abilities. So, we would not be concerned with 'units' of content, but with 'units' of activity which generate communication and metacommunication.

(d) Continuity

The need to provide continuity for the learner has, in the past, been based upon content. Within a communicative methodology, continuity can be identified within at least four areas. First, continuity can reside in the activities and the tasks within each activity; and from one activity to another and from one task to another. An activity or task sets up its own requirements for its progressive accomplishment, and it is the pursuit of these requirements which can provide tangible continuity for the learner. Second, continuity potentially resides within communicative acts during the learning and teaching: either at the 'macro' level in terms of the whole lesson and its 'micro' sequences of negotiation, or within the structure of discourse in terms of the 'macro' com­ municative act with its own coherent sequence of utterances. 14 Third, con­ tinuity is provided through the ideational system which can also be seen in terms of 'macro' and 'micro' levels. At the 'macro' level the learner may have access to continuity of theme, while at the 'micro' level the learner can have access to conceptual or notional continuity. Because ideational continuity is realised through a refinement of textual knowledge-the refinement of a concept, for example, can imply a refinement of its linguistic expression, and

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

meaning, and vice-versa-so it may be wrong to assume that what may be 'simple' for any one learner is likely to be 'simple' for all the learners. Sequencing in communicative content is therefore likely to be a cyclic process where learners are continually developing related frameworks or aggregations of knowledge and ability use, rather than accumulating separable blocks of 'static' knowledge or a sequence of ordered skills� Learners would typically move from global to particular perspectives-and vice-versa-in their negotiation with the content. Thus, content becomes something which learners move into and out from, and to which they return in a process of finer analysis and refined synthesis. Curriculum designers cannot, therefore, predict with any certainty the 'levels' of content on which learners will decide to evolve their own sequencing in learning. All such designers can do is to anticipate a range of content which will richly activate process competences so that the ultimate target repertoire becomes accessible and its specific demands recognised by the learner.

1 04

E S S E N T I A L S O F A C O M M U N I C AT I V E C U R R I C U L U M

(e) Direction Traditionally, learners have been expected to follow the direction implicit in some prescribed content. Typically an emphasis on content led the learner from the beginning, through the middle, to the end. From what has been indicated so far, a communicative methodology would not exploit content as some pre-determined route with specific entry and exit points. In a com­ municative methodology, content ceases to become some external control over learning-teaching procedures. Choosing directions becomes a part of the curriculum itself, and involves negotiation between learners and learners, learners and teachers, and learners and text. Who or what directs content becomes a justification for communication about the selection and organisation of content with methodology, and about the various routes to be adopted by the learners through any agreed content. Content can be predicted within methodology only to the extent that it serves the communicative learning process of the participants in the group. It might well be that the teacher, in negotiation with learners, will propose the adoption of aspects of the target repertoire as appropriate content. However, the teacher would recognise that the central objective of developing underlying communicative knowledge and abilities can be achieved through a range of alternative con­ tent, not necessarily including aspects of the target repertoire. Such 'carrier' content can be as diverse as the different routes learners may take towards a common target: perhaps content can be more various and more variable. Also, the teacher would remain free to build upon the contributions of learners­ their initial competences and expectations-and exploit the inevitably different ways in which learners may attain the ultimate target. Further, the teacher would not regard any surface performance as synonymous with its underlying knowledge and abilities, and he will avoid transferring this possible confusion to the learners.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

vice-versa-there i s a parallel continuity of ideation and text. Fourth, and finally, continuity can reside within a skills repertoire or a cycle of skill-use during an activity. For example, there could be a progression from reading to note-taking to speaking for the achievement of a particular activity. A com­ municative methodology would exploit each of these areas of continuity as clusters of potential continuities, rather than exploit any one alone. All can be inherent in a single activity. These kinds of continuity offer two important advantages. They can serve the full process competences of learners-know­ ledge systems and abilities-and they can allow for differentiation. Learners need to be enabled to seek and achieve their own continuity and, therefore, the criteria for their own progress. In the process of accomplishing some im­ mediate activity, learners will impose their own personal and interpersonal order and continuity upon that activity, the communication which the activity generates, the interpersonal, ideational and textual data which they act upon, and on the skills they need to use in the activity's achievement. As a result, the progressive refinement of the learner's own process competence can provide an overall /earning continuity. Once the teacher can accept that each of these areas provides potential continuity for different learners, it ceases to be a problem if different learners pursue several routes or progress at different rates.

M I C H A E L P. B R E E N A N D C H R I S TO P H E R N . C A N D L I N

7.

105

H O W IS T H E C U R R I C U L U M P R O C E S S T O BE E V A L U AT E D ?

Criteria for eventual success-in some particular task-could be initially negotiated, achievement of the task could be related to these agreed criteria, and degrees of success or failure could be themselves further negotiated on the basis of the original criteria. Evaluative criteria, therefore, would be established and applied in a three-stage process: (i) What might 'success' mean? (ii) Is the learner's performance of the task successful? (iii) If so, how successful is it? Each stage would be a matter for communication. Instead of the teacher being obliged to teach towards some externally imposed criteria­ manifested most o ften by some external examination or standardised test-he can exploit the interpretation of these external or standardised criteria as part of the joint negotiation within the classroom. The group's discovery of the criteria inherent in such end-of-course or summative assessment would be one means for the establishment of the group's own negotiated criteria and, crucially, for the sharing of responsibilities during the learning-teaching process. In a communicative curriculum we are dealing with an interdependence of the curriculum components of purposes, methodology, and evaluation. It follows that any evaluation within the curriculum also involves an evaluation of the curriculum itself. Any joint negotiation among the various participants within the curriculum may obviously deal with the initial purposes and on­ going methodology which have been adopted . Indeed, communicative evaluation may well lead to adaptation of initial purposes, of methodology, and of the agreed criteria of evaluation themselves. Evaluation within and of the curriculum can be a powerful and guiding force. Judgements are a crucial part of knowledge, learning, and any educational process. By applying judgements to the curriculum itself, evaluation by the users of that curriculum can be brought into the classroom in an immediate and practical sense. Once within the classroom, evaluation can be made to serve as a basis for new directions in the process of teaching and learning. A genuinely communicative use of evaluation will lead towards an emphasis on formative or ongoing evaluation, rather than summative or end-of-course

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

The communicative curriculum insists that evaluation i s a highly significant part of communicative interaction itself. We judge 'grammaticality', 'ap­ propriateness' , 'intelligibility', and 'coherence' in communicative per­ formance on the basis of shared, negotiated, and changing conventions . Evaluation within the curriculum can exploit this 'judging' element of every­ day communicative behaviour in the assessment of learners' communication and metacommunication. The highly evaluative aspect of communication can be adopted as the evaluation procedure of the curriculum. If so, the essentially intersubjective nature of evaluation can be seen as a· strong point rather than , possibly, a weakness. How might we evaluate learner progress? Evaluation of oneself, evalua­ tion of others, and evaluation of self by others is intersubjective. In this way, evaluation need not be regarded as external to the purposes of the curriculum or external to the actual process of learning and teaching. In recognising that relative success or failure in the sharing of meaning, or in the achievement of some particular task, is most often an intersubjective matter, the com­ municative curriculum would rely on shared and negotiated evaluation .

106

E S S E N T I A L S OF A CO M M U N I C A T I V E C U R R I C U L U M

evaluation which may b e based on some prescribed criteria. I f evaluation is accepted as an immediate and ongoing activity inherent in communication and in learning to communicate, then it is very likely to be genuinely formative for both the individual learner and the whole curriculum. That is, it can shape and guide learning and guide decisions within the curriculum process. Any shared and negotiated evaluation within the classroom will generate potentially formative feedback for and between learners and between learners and the teacher. Formative evaluation may not only indicate the relative successes and failures of both learner and curriculum, it can also indicate new and different directions in which both can move and develop. Such evaluation can be an almost inevitable product of a differentiated methodology, just as it can act as a motivating force for differentiation.

8.

A C H I E V I N G CO M M U N I C A T I V E L A N G U A G E T E A C H I N G

We emphasised at the outset of this paper that any curriculum framework for language teaching and learning · necessarily involves designers, materials writers, teachers and learners in a process of relating the three components of purpose, methodology and evaluation. Even so, we need to acknowledge that any curriculum-including a communicative curriculum-cannot strictly be designed as a whole from the start. We can only deduce and propose the principles on which a variety of communicative; curricula may be based. Any curriculum is a personal and social arena. A communicative curriculum in particular, with its emphasis on the learning and teaching of communication, highlights a communicative process whereby the interrelating curriculum components are themselves open to negotiation and change. From this it follows that the communicative curriculum-no more than any other-can never be one uniquely identifiable language teaching curriculum. In a real sense there can be no such thing as an ideal and uniquely applicable

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

This placing of evaluation within the communicative process as a formative activity in itself does not necessarily invalidate the place of summative evaluation. Sumniative evaluation becomes valuable if it can reveal the learners' relative achievement of a particular target repertoire. However, we have already proposed that any target repertoire needs to be seen as the tip of an iceberg. Therefore, an essential requirement on any summative evaluation would be that it can adequately account for the learner's progress in the refine­ ment of a particular underlying competence-the communicative knowledge and abilities which provide the capacity for the use of a target repertoire. Sum­ mative evaluation, in other words, needs to be sensitive to differential com­ petences which may underlie some common target. As such, summative evaluation within a communicative curriculum needs to focus on the assess­ ment of the learner's developing communicative knowledge and abilities as well as on his actual performance within the target repertoire. Summative evaluation has still to account for developed knowledge and abilities in ad­ dition to the surface manifestations of such development. Therefore, the essential characteristics of evaluation within a com­ municative curriculum would be that such evaluation is itself incorporated within the communicative process of teaching and learning, that it serves the dual role of evaluating learner progress and the ongoing curriculum, and that it is likely to be formative in the achievement of this dual role.

M I C H A E L P. B R E E N A N D C H R I ST O P H E R N . C A N D L I N

107

language teaching curriculum since any realisation of the curriculum must reflect a realistic analysis of the actual situation within which the language teaching will take place. To cope with this requirement of appropriateness to situation, the communicative curriculum has to be proposed as a flexible and practical set of basic principles which underlie a whole range of potential com­ municative curricula. It is this set of principles which we have tried to present in this paper, in the knowledge that such proposals need to be translated into action in the classroom in order to test their own validity. This is, after all, the only means by which curriculum theory and practice can develop. Even though the curriculum designer may have taken account of the actual language teaching situation, he has to recognise that from design to implementation is itself a communicative process . J . M. Stephens (1 967) identified this process when he said:

While Stephens, in talking about stimuli, does not emphasise transactions as a two-way process, he clearly implies that the translation from principles through design to implementation is most often a process of reinterpretation of the curriculum, and a process of negotiation between the curriculum and its users. If adopted within the design and implementation procedure, the con­ ditions or minimal requirements on any communicative curriculum must take account of those situational constraints which are unchangeable. However, such minimal requirements should also serve as the general criteria against which any situational constraints will be tested in order to assess whether or not the constraint is genuinely immutable or whether it may be overcome. If a curriculum based upon the principles which we have examined here is not implementable within a particular situation, then it may be that a genuinely communicative curriculum is simply not viable. It may be the case that curriculum designers and teachers in such a situation need to consider whether the achievement of language learning as communication is ap­ propriate. Communicative curricula need-through time and according to situation­ to be open and subject to ongoing developments in theory, research, and practical classroom experience. Communicative curricula are essentially the means of capturing variability. Variability will exist in selected purposes, methods, and evaluation procedures, but variability must also be seen as inherent in human communication and in the ways it is variously achieved by different learners and teachers. The classroom-its social-psychological reality, its procedures and activities-is potentially a communicative environ­ ment where the effort to pull together such variability is undertaken. The

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

The curricular reforms emanating from the conference room will be effective only insofar as they become incorporated into the concerns that the teacher is led to express. Any statements or decisions coming from the curriculum committee will not be transported intact into the lives of pupils. Such statements must work through a complex chain of interactions. The original statements of the com­ mittee will act as stimuli for one set of people such as subject-matter supervisors. These people, in turn, will react to the stimuli, possibly merely mirroring what they receive, more likely, incorporating much of themselves into the reaction. Their reactions will then act as stimuli for a second set of people who will also react in their own way. After a number of such intermediary transactions some­ one, the teacher, will apply some stimuli to the pupil himself. (pp. 1 2- 1 3)

E S S E N T I A L S O F A COM M U N I CA T I V E CU R R I CU L U M

108

learning-teaching process in the classroom is the meeting-point of all curriculum components and it is the place where their coherence is continually tested . The learning-teaching process in the classroom is also the catalyst for the development and refinement of those minimal requirements which will underlie future curricula.

(Received September 1979)

NOTES 'Curriculum' can b e distinguished from 'syllabus' i n that a syllabus i s typically a specification

Content and its organisation is subsumed within a curriculum as part of methodology (Section 6 of this paper). A syllabus 1s therefore only part of the overall curriculum within which it operates. For mteresting discussions of curriculum theory and design see, inter alia, Lawton, 1973; Stenhouse, 1 975; Golby et al. , 1975. 1

Here the term 'conventions' may be more appropriate than the term 'rules' for they are

socially and personally agreed, acted upon and recreated rat her than merely followed or obeyed as i f external to social and personal influence. Such conventions would embrace lingu1suc rules, sociolinguistic rules and psycholinguisuc constraints and the interrelationships between these systems underlying communication. ' This distinction between the three knowledge systems denves from Halliday ( 1973). Halliday descnbes the three as 'macro-functions' of language, and proposes that any linguistic unit

IS

the

simultaneous realisation of the ideational, interpersonal and textual. In this paper we take the v1ew that text, seen both in terms of ind1v1dual sentences and structures above the sentence, acts as the embodiment of, and resource for, particular ideas·and interpersonal behaviour. Discourse can be seen as the process of deriving and creating meanings-ideational and interpersonal-through text (Widdowson, 1 978). In adopting th1s definition, we are, of course, aware that the term discourse is also sometimes used for suprasentential structure, i.e., the ways in which paragraphs and other large units are organised (Halhday aqd Hasan, 1975). Such structures and their organisation remam, in our view, within text, and knowledge of the rules wh1ch govern these make up textual k nowledge. •

The concept of 'target repertoire' is preferred to the notion of 'target language' because the

latter narrowly emphasises form, while being unspec1fic as to the inevitable selection that any curriculum makes from a language. The concept of 'repertoire' denves from sociolinguistics (Gumperz, 1964) and captures the fact that a speaker/hearer typically controls a selection of language varieties. ' This negotiative interaction w1thin the learner between pnor knowledge and the new learning has been a concern within psychology for many years. The work of Piaget ( 1953) on 'assim1Iauon' and 'accommodation' and the proposals of Jerome Bruner ( 1 973), on how learners go beyond the informauon given, exemplify this interest. More recently, Ulric Neisser ( 1976) develops Tolman's ( 1 948) theory of 'cognitive maps' and proposes-in an mteresung and provocative way-the nature of interaction in perception, both between the perceiver and the object of perception and within the perceiver. The present authors believe the possible relationships between com­ muni cative negotiation between people and negotiative interaction between the mind and the external world offer a rich area for psycho-socio-linguistic research. ' Such secondary abilities would include, for example, 'coding' (Bernstein, 1 97 1 ) or 'repre­ sentation' (Bruner et al., 1 966); the ability to 'code-switch' (Hymes, 197 1 ) including 'style­ shifting' (Labov, 1970); and the ability to JUdge commumcative performance m a var1ety of ways against a variety of criteria (Labov, 1966; Lambert, 1967; Hymes, 197 1 ; Giles and Powesland,

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

1

of the content of the teaching and learmng and the organisation and sequencing of the content.

M I C H A E L P. B R E E N A N D C H R I ST O P H E R N . CA N D L I N

109

1975). It seems that we can regard the primary abilities of mterpretation, expression, and negotiation as themselves complexes of secondary abilities which mterrelate with one another. Further we can assume that the primary abilities may not necessarily be specific to linguistic communicauon. 1

'Sociolinguistic competence' (knowledge of the rules governing speech evems) was originally

proposed as 'communicative competence' by Dell Hymes ( 1 97 1 ) but the present authors would want to regard it as one aspect of interpersonal knowledge. The 'functional' aspect of language seems, at first sight, a more reaiiSiic basis for language teaching. However, It seems to us that recent efforts to incorporate 'Functions' into language teaching materials are based upon inap­ propriate and quite misleading assumptions. First, that Functions are 'items' like categories of gramf!lar or rules of syntax; second, that an utterance 1s hkely to be associated with a single Function; and, third, that there is a predictable relationship between a Function and its symactic or textual realisation. These assumptions seem to denve from the tendency to apply textual criteria alone to every aspect of language use. For a more detailed examination of the components of the knowledge systems underlying commumcatlve competence, see Breen and Candlin (forthcoming) 1

Th1s 'process competence' is the learner 's changing and developing commumcative knowledge

and abihties as the learner moves from initial competence towards the target competence. It is partly revealed through a senes of ' lmerlanguages' (Selinker 1 972, Tarone 1 977, Corder 1 978). It is also revealed by the commumcative strategies (particular use of abilities) of learners in discovering and exploiung the meaning potential of texts. Sections 4 and 5 of this paper consider this 'process competence' in more detaiL •

Within the spoken medium, for example, different 'text-types' would include newscasts, con­

versations, radio commentaries, public speeches, announcements over public address systems, etc. For a discussion of the term text-type, and a classificauon of possible text-types m language learnmg, see Protoko/1 9 of the Bundesarbeitsgemeinschaft Engllsch an Gesamtschulen 1976, (*Federal Working Party for English in Comprehensive Schools) which is avrulable from the Hessisches l nstitut fur Lehrerfortbildung, 3501 Fuldatal, Federal Republic of Germany. For an inventory of communicatively-oriented exercises using a ·variety of text-types see Bundesarbeits­ gemeinschaft Englisch an Gesamtschulen (eds) Kommumkativer Englischunterrtcht: Prinztpien und Obungstypologie, (*Communicative English Teachmg: Principles and an Exercise Typology)

Munich, Langenscheidt-Longman 1978. 1 0 Regardless of the ability used or the medium through which the ab1hty works, what seems to matter for the user, and the learner, is meaning-making or the search for meanmg potemial (see,

for example: Johnson-Laird, 1974 and Smith, 1971). Further, it is very likely that the ability to interpret, for instance, exists m an interdependent relationship with the ability to express (see, for example: Clark et al. , 1 974 and Ingram 1974). Similarly, the skills of speaking and listening are themselves mterdependent, and the acquisition of literacy skills have, in turn, a dependence on these. Once basic literacy IS mastered the four skills conunually interact. 11

Learning strategies are how learners use their abilities in order to learn: therefore they are one

defining feature of process competence. These strategies are revealed by learners' 'interlanguages' and

learners'

tendencies

towards

simplification,

over-generalisation,

and

'idiosyncratic'

regularisation of the target language (Siobin, 1973; Dulay and Burt , 1975; Richards, 1975; Gat­ bouton, 1978 offer some of the evidence). Learning strategies can exemplify more general and di ffering 'learning styles' (Marton, 1975 and Krashen, 1977 discuss interesting variables, whilst Rubin, 1975 and Naiman et a!, 1978 reveal strategies wh1ch typify the 'good' language learner). 1

2

On the d1fferemial contributions of the ' formal' and 'informal' language learning en­

vironments, Krashen ( 1 976) makes a number of very interesting observatiOns on the advantages of a ' formal' environment in terms of, specifically, the self-monitoring activity of the learner. A communicative methodology would, of course, exploit monitoring as both an individual and shared undertaking when appropriate to an activity and to pamcipants in an activity. Krashen 's 'monitoring' relates closely to what we have described as communication-about-learning and meta-communication (see Sections 4. 1 and 7). I I So me readers may fee) that the main rOles Of facilitatOr Of and interdependent partiCipant

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Ch . I and other sources there referred to.

1 10

E S S E N T I A L S O F A COM M U N I CA T I V E CU R R I CU L U M

within a communicative methodology present teachers with unaccustomed challenges. However, if we more closely consider the problems for the teacher within more traditional methodologies (see for example: Rogers, 1969; Allwnght, 1978), we can d1scover quite separate motivations for the roles suggested here. " On lessons seen in terms of 'macro' and 'micro' levels see Sinclair and Coulthard ( 1 975). On discourse

as

'macro speech acts' and ' m1cro speech acts' see van D1jk ( 1977).

R E F E R E N CE S

Bernstein, B., 1 97 1 . Class, Codes and Control, Volume I : Theoretical Studies towards a Sociology ofLanguage. London: Routledge and Kegan Paul. Breen, M . P. and Candlin, C. N., forthcoming. The Communicative Curriculum in Language Teaching. London: Longman. Bruner, J. S., Olver, R. and Greenfield, P . , 1 966. Studies in Cognitive Growth . New York: John Wiley & Sons. Bruner, J. S . , 1 973. Beyond the Information Given. London: George Allen & Unwin. Clark, R . , Hutcheson, S . , Van Buren, P . , 1 974. ' Comprehension and production in language acquisition' Journal of Linguistics, 1 0, 1 974. Corder, S. P . , 1 978. 'Error analysis, interlanguage and second language acquisition' in Kinsella, V. (ed.) Language Teaching and Linguistics: Surveys. Cambridge University Press. Dulay, H . and Burt, M . , 1 975. 'A new approach to discovering universal strategies of child second language acquisition' in Dato, D. (ed.) Developmental Psycho­ linguistics. Georgetown University Press. Gatbouton, E . , 1 978. ' Patterned phonetic variability in second language speech' Canadian Modern Language Review, 34, 3: 1 978. Giles, H . and Powesland, P . , 1 975. Speech Style and Social Valuation. London: Academic Press. Golby, M . , Greenwald, J. and West, R . , (eds.) 1 97 5 . Curriculum Design . London: Croom Helm in association with the Open University Press. Gumperz, J. J . , 1 964. 'Linguistic and social interaction in two communities' in Gumperz, J . J. and Hymes, D. (eds.) A merican A nthropologist 66 (6 ii): 1 964. Johnson-Laird, P . , 1 974. Psychology 25 , 1 974.

'Experimental psycholinguistics' Annual Review of

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Allwright, R . L. 1 978. 'Abdication and responsibility i n teaching' . Paper presented at the Berne Colloquium on Applied Linguistics to be published in a special issue of Stud1es in Second Language AcquiSition. Indiana: Indiana University Press.

M I C H A E L P . B R E E N A N D C H R I ST O P H E R N . C A N D L I N

Ill

Halliday, M . A . K., 1 97 3 . 'The functional basis of language' in Bernstein, B. (ed.) Class, Codes and Control, Volume 2: Applied Studies towards a Sociology of Language. London: Routledge and Kegan Paul. Hymes, D., 1 97 1 . 'On communicative competence' in Pride, J. and Holmes, J . , 1 972. Sociolinguistics. Harmondsworth: Penguin Books. Ingram, D. , 1 974. 'The relationship between comprehension and production' in Schiefeldbusch, R. and Lloyd, L. (eds.) Language Perspectives: Acquisition, Retardation and Intervention. London: Macmillan. Krashen, S . , 1976. ' Formal and informal linguistic environments in language acquisition and language learning' TESOL Quarterly, 10, 1 976.

Lambert, W. E . , 1 967 . 'A social psychology of bilingualism' in Macnamara, V. (ed.) 'Problems of Bilingualism' Journal of Social Issues 23, 1 967. Labov, W., 1 966. Social Stratification of English in New York City. Washington D.C.: Center for Applied Linguistics. Labov, W . , 1970. 'The study of language in its social context' Studium Generale 23, 1 970. Lawton, D., 1 973. Social Change, Educational Theory and Curriculum Planning. London: University of London Press. Marton, F., 1 975. 'What does it take to learn?' Paper presented for Council of Europe Symposium 'Strategies for Research and Development in Higher Education' . University of Gothenburg, September 1 97 5 .

·

Naiman, N . , Frohlich, M . , Stern, H . H . and Todesco, A . , 1 978. The Good Language Learner. Toronto: Ontario Institute for Studies in Education. Neisser, U., 1 976. Cognition and Reality. San Francisco: W. H. Freeman & Co. Piaget, J . , 1 953. The Origins of Intelligence in the Child. London: Routledge and Kegan Paul. Richards, J . , 1975. 'Simplification: a strategy in adult acquisition of a foreign language' . Language Learning 25 , 1 975. Rogers, C., 1 969. Freedom to Learn. Columbus: Ohio, Merrill. Rubin, J . , 1 975. 'What the "good language learner" can teach us' . TESOL Quarterly 9, 1 975. Selinker, L., 1 972. ' lnterlanguage' . IRAL 10: 3 , 1 972. Sinclair, J . McH. and Coulthard, M . , 1 975. Towards an Analysis of Discourse. London: Oxford University Press.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Krashen, S., 1 977. 'The monitor model for adult second language performance' in Burt, M . , Dulay, H. and Finnochiaro, M . (eds). Viewpoints on the Acquisition of English as a Second Language. New York: Regents.

1 12

E S S E N T I A L S OF A C O M M U N I C A T I V E C U R R I C U L U M

Smith, F . , 1 97 1 . Understanding Reading. New York: Holt, Rinehart & Winston Inc. Stenhouse, L . , 1 975. An Introduction to Curriculum Research and Development. London: Heinemann. Stephens, J. M . , 1 967. The Process of Schooling: A Psychological Examination. New York: Holt, Rinehart & Winston. Tarone, E . , 1 977. 'Conscious communication strategies in inter-language: a progress report' . Paper presented at the 1 1 th TESOL Convention, Miami, Fl. 1 977. Tolman, E . C . , 1 948. 'Cognitive maps in rats and men' Psychological Review 55, 1 948 . van Dij k, T. A . , 1 977. Text and Context: Explorations in the Semantics and Pragmatics ofDiscourse. London: Longman. Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Widdowson, H. G . , 1 978. Teaching Language as Communication. Oxford: Oxford University Press.

S O CIOLINGUIS TIC S URVEYS : THE STATE OF THE ART ROBERT L . COOPER Hebrew Umversity ofJerusalem

T H E N A T U R E OF S U R V EY R E S E A R C H

Varying grounds have been used to distinguish survey research from other types of social science research. Survey research is sometimes characterized as exploratory in nature, as being especially useful at the preliminary stages of an investigation, when the investigator is o ften ignorant of the relevant variables. While sociolinguistic surveys can be preliminary in nature, more often they are not. Scotton's (1972) sociolinguistic survey of Kampala, for example, was carried out after extensive fieldwork had determined the categories to be embodied in her interview schedule. Similarly, Labov's ( 1 966) survey of New York's Lower East Side was undertaken after . the phonological and social variables to be studied had been isolated on the basis of previous fieldwork. Applied Linguisllcs, Vol. I , No. 2

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

S O C I A L science survey methods, as old as the social sciences themselves, have been continuously developed and refined . Any paper on the state of the art of sociolinguistic surveys, therefore, invites the accusation of disciplinary in­ sularity. There is, after all, an extensive literature on social science survey methods. Those who wish to learn about survey design , instrument con­ struction, sampling, data collection, and data analysis can choose among dozens of books _for instruction (see, for example, Jahoda, Deutsch, and Cook 195 1 ; Lindzey and Aronson 1969; Moser 1 958; Warwick and Lininger 1 975). While it is true that sociolinguistic surveys adapt the technology found in social science surveys more generally, one can justify their separate treatment on several grounds. For one thing, sociolinguistic surveys, like surveys under­ taken from any disciplinary perspective, reflect the theoretical orientations and empirical concerns of their practitioners. Thus the aims of sociolinguistic surveys can be distinguished from those of other types of survey. Second, some of the techniques associated with sociolinguistic surveys, if not originating with such surveys, have been developed in connection with them. Third, unlike most social science surveyers, many of the practitioners of socio­ linguistic surveys are essentially self-taught with respect to social science research methods. Finally, widespread interest in sociolinguistic surveys is new, paralleling and stemming from the relatively recent interest in socio­ linguistics itself. Just as many sociolinguistic studies antedate the early 1 960's, when recognition of sociolinguistics as a field of inquiry began to grow, so sociolinguistic surveys are not a new phenomenon (Jernudd 1 975). Nonethe­ less, almost all of the sociolinguistic surveys reported in the literature have been carried out since the mid- 1960's, and the number of new surveys seems to grow at a geometric rate. Thus, while the technology of sociolinguistic surveys derives from social science surveys more generally, they have features which justify their separate consideration.

1 14

S O C IO L I N G U I S T I C S U R VEYS

GOALS

The goals of sociolinguistic surveys can b e analyzed i n at least two ways . First, we can categorize the behaviors which such surveys are designed to assess. Second, we can classi fy the reasons for assessing such behaviors .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Sometimes survey research is distinguished from experimental research. The former is said to be correlational, i.e. , associations but not causal relationships are established among variables . Experimental research, in contrast, establishes causal relationships by comparing the effects of differing treat­ ments, either on subgroups which are equivalent with respect to all relevant characteristics, except exposure to the treatments of interest, or on res­ pondents who are exposed to all treatments and thus serve as their own control. It is probably true that most sociolinguistic surveys are correlational in nature, although inferences of causality can sometimes be drawn from correlational data (see, for example, Cooper and Horvath 1 973; Lieberson 1 970; Weinreich 1 957). However, some sociolinguistic surveys are experi­ mental in character. Thus in surveys carried out by Labov ( 1 966) and by Ma and Herasimchuk ( 1 97 1 ), the effects of differing elicitation procedures on the production of phonological variables were described and compared. The distinction between survey and other types of research which will be followed here is based on the obtained results ' generalizability to a target population. Survey research is defined here as research carried out with respect to an entire population, whether as small as a hundred neighboring households (Fishman et a/. 1 97 1 ) or as large as a nation (Ladafoged et a/. 1 97 1 ) . Thus censuses are surveys as are studies whose samples are drawn so as to represent a defined population. Because of the constraints imposed by limitations of time and money, there is often a trade-off between the number of cases that can be observed and the number of questions which can be asked about each case. Inasmuch as studies of whole populations, whether by complete enumeration or by sample survey, typically involve large numbers of cases, the number of questions asked about each case is often smaller than the number which can be asked by other types of study. Thus a survey is sometimes what the literal meaning of the word suggests, an 'overview', an examination which obtains breadth of observation at the expense of depth . That this is not a necessary characteristic of survey research, however, can be seen from the surveys conducted by Das Gupta et a/. ( 1 972), Fishman et a/. ( 1 97 1 ) , Labov ( 1 966), and Le Page ( 1 972, 1 975), among others, which obtained both minutely detailed and voluminous data. Survey research, therefore, is defined here not according to the degree of detail it obtains, nor according to its correlational or experimental nature, nor according to the stage of in­ vestigation it represents , but rather it is defined in terms of the generalizability of its results to a specified population. Sociolinguistic surveys gather information about the social organization of language behavior and behavior toward language in specified populations. Because such behaviors encompass multitudinous phenomena and because populations vary in size, complexity, and basis of organization, sociolinguistic surveys vary widely in the questions which they ask, the ways in which they answer them, and the uses for which their results are intended. This paper out­ lines some of the major goals and procedures of such surveys.

R O B E RT L. C O O P E R

l iS

Surveys designed to measure the same behavior might be differently motivated and conversely surveys motivated by the same purpose might assess different behaviors. Thus for example information with respect to a population' s

second language proficiency i n various languages can b e collected t o help formulate language policy, e.g., the language varieties to be chosen for use in radio broadcasting, or to help test a hypothesis, e.g . , with respect to language spread. Conversely a survey carried out for the purpose of formulating language policy might gather information about second-language proficiency or it might collect data with respect to language attitudes. We shall first classify the behaviors assessed by sociolinguistic surveys and then we shall categorize the purposes for collecting such data. B E H A V IORS ASSESSED

p

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

A convenient dichotomy for discussing sociolinguistic behavior has been proposed by Agheyisi and Fishman ( 1 970), namely the distinction between language behavior and behavior toward language. The language behaviors assessed by sociolinguistic surveys can be classified into three major categories: proficiency, acquisition, and usage. Proficiency refers to what has been learned; acquisition refers to the sequence of learning, particularly what has been learned first; and usage refers to what is typically done. Each of these broad classes of behavior can be examined at either of two levels of ab­ straction. At the 'microsociolinguistic' level of observation surveys are con­ cerned with particular features of language. With respect to the variants of a phonological, lexical, or syntactic variable, for example, a survey might describe which variants different groups of speakers can use (proficiency) , which they learned first (acquisition), and which they typically use in different types of speech situation . or communicative context. At the 'macrosocio­ linguistic' level of observation, surveys describe what languages or language varieties different groups of speakers know (proficiency), the order in which they learned them (acquisition), and the contexts in which they use them (usage). The intersection of three ty es of behavior with two levels of ob­ servation creates six cells into which the language behaviors assessed by socio­ linguistic surveys can be classified. With respect to behavior toward language, it can be classified as either attitudinal or implementational. Language is an attitudinal object concerning which people have opinions and feelings and towards which they may be pre­ pared to act. While attitudes are unobservable constructs whose characteristics must be inferred on the basis of observable behavior, implementational behavior is itself observable. That is to say, people act overtly toward language. For example, they may voluntarily enroll their children in schools in which the language of instruction differs from that used in the majority of a community's schools; they may participate in organizations whose purpose it is to accomplish spelling reform; they may struggle to maintain or to change the status of their mother tongue by participating in organizations devoted to such purposes or by lobbying or petitioning legislators; they may correct another speaker's use of forms which they consider to be nonstandard or improper; they may work in language planning agencies or participate in activities sponsored by such agencies; or they may be actively involved in language policy decisions.

SO C I O L I N G U I S T I C S U R V E Y S

1 16

Both attitudinal and implementational behaviors can b e directed toward microsociolinguistic as well as macrosociolinguistic objects, thus creating four cells for the classification of behaviors toward language. If we add these to the six cells specified above, we have a taxonomy of ten cells into which the behaviors assessed by sociolinguistic surveys can be categorized (see Figure 1) .

Figure 1: Behaviors Assessed by Sociolinguistic Surveys Behavior

Level of observation Macro Micro

Language behavior

Behavior toward language Attitudinal Implementational

Although sociolinguistic surveys sometimes are directed to behaviors which fall within only one of these ten cells, more commonly such surveys measure behaviors from each of several cells. One example of a 'multicelled' survey is the International Research Project on Language Planning Processes (Das Gupta et at. , 1 972; Fishman 1 975). Among the data gathered by this survey was information collected from high school and university students and their teachers in India, Indonesia, and Israel. Information was sought with respect to the languages the respondents could speak and read (macro-proficiency), the languages they spoke first as children (macro-acquisition) and the languages they used for various specified contexts (macro-usage). In addition, respondents were tested with respect to their knowledge of certain technical terms in the standard (planned) language (micro-proficiency); they were asked if they know any alternative terms and if so when they learned them (micro­ acquisition); and their responses to word-naming tasks were scored for the proportion of language academy-approved terms they employed (micro­ usage). The project also assessed behaviors toward language. Respondents were asked their opinions about a sample of language academy-produced words in the fields of chemistry, civics, and language and literature (micro­ attitudinal) and about various languages and language planning (macro­ attitudinal), and they were asked about their participation in language planning organizations (particularly corpus planning agencies) or in the ac­ tivities sponsored by these organizations (macro-implementational). In ad­ dition, textbook writers were asked about the source and degree of control to which they were subjected with respect to their use of technical terminology (micro-implementational) . Thus the International Research Project on Language Planning Processes tapped behaviors from all ten cells of the taxonomy. This survey was unusually comprehensive in the behaviors it assessed. None­ theless, sociolinguistic surveys often encompass more than one cell. That they

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Proficiency Acquisition Usage

ROBERT L . COOPER

117

P U R POSES

Sociolinguistic surveys have been undertaken for a variety o f reasons, and often a survey is motivated by more than one of them. Some of the major motivations are described below.

Language Policy Decisions

In linguistically heterogeneous polities, it often becomes necessary to decide what languages or language varieties will be used for various purppses. In making decisions of this sort, it is useful to know what segments of the population can speak and understand each of the languages or language varieties among which a choice is to be made and the attitudes of the public towards their use for various purposes. While language policy decisions are often made on irrational grounds, decisions presumably are sounder if they are based on accurate and relevant information (Ferguson 1 966). An example of the use of sociolinguistic surveys for language policy decisions can be found in the work of the Summer Institute of Linguistics in Mexico (Casad 1 974). The S.I.L. was faced with the problem o f choosing among various vernaculars for Bible translations and literacy campaigns. The extent to which various language varieties can be understood by speakers of related varieties is one factor which can appropriately be considered when deciding what varieties

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

do so is not surprising. Just as the order o f language acquisition, language proficiency, and language usage are related to one another, so language behavior and behavior toward language are intertwined. Th� problems or issues which motivate the description of behaviors from one cell therefore often motivate the description of behaviors from another. Because the relationships among these behaviors is often weak, it is sometimes the case that whichever of the behaviors is of crucial interest can be best explained in terms of a combination of the others. This can be seen most clearly in multiple regression analysis, inasmuch as cumulative prediction succeeds to the extent that the predictor (independent) variables are strongly related to the criterion (dependent) variable but weakly related to each other. Thus a survey of a Puerto Rican neighborhood in Jersey City found that proficiency criterion scores could be successfully predicted by a combination of variables which were maximally independent of one another (Fishman et a/. 1971). For example, almost 75o/o of the variation in respondents' Spanish repertoire range (the number and fluency of speech styles observed in Spanish, as rated by two linguists who based their j udgments on tape-recorded samples) was explained by a combination of seven predictors. Some of these were microsocio­ linguistic, e.g., the frequency of particular phonological variants observed, others were macrosociolinguistic, e.g. , literacy in Spanish. Some were pro­ ficiency scores, e.g. , number of words produced in a Spanish word-naming task, and others were usage scores, e.g., self-report ratings of the frequency with which specified words dealing with the work sphere were encountered. When such predictor items, which had low correlations with each other, were combined they accounted for an unusually high proportion of the criterion's variability. Thus the measuremen� of behaviors from several cells of the behavioral taxonomy is likely to improve the survey's contriblltion to the issues or problems which motivated it.

1 18

S O C I O L I N G U I ST I C S U R V E Y S

Program Planning

Once language policy has been formulated (whether or not on the basis of relevant data) , it must be implemented, and the planning of the imple­ mentation often requires basic information about the language situation among the people with whom the policy is to be carried out. The Work­ Oriented Adult Literacy Project in Ethiopia provides one such example. Sponsored by the United Nations Special Fund and by the Ethiopian Ministry of Education and Fine Arts, the project's job was to determine the potential effectiveness of adult literacy campaigns based on vocationally-oriented teaching materials. Accordingly, several trial campaigns had to be set up. In order to plan these campaigns, it was necessary to determine what proportion of the population in these areas to be served by the campaigns would be likely to enroll in literacy classes. Thus the planners needed to know what proportion of the population in these areas was not literate; what proportion of the illiterate adults would attend classes if the times and location of classes were convenient; how many hours a week potential students were willing to attend class; and the amount of time potential students were willing to spend walking to and from class. In addition, because the medium of instruction was to be Amharic, the program planners had to know what proportion of the potential learners could understand Amharic. This information was obtained by means of sample surveys in the trial campaign areas (Cooper et a/. , 1 976; Cooper and Fasil, 1 976). Another example is the national survey to- be carried out by the U.S. Bureau of the Census of 1 976 for the purpose of estimating the proportion of the population which has ' limited English speaking ability' , particularly with respect to schooling in which the medium of instruction is English. Such in-

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

should b e used s o a s to reach the largest number of people at a given cost. Accordingly the S.I.L. undertook an extensive set of intelligibility studies among the local varieties of each of many Indian languages spoken in Mexico. Other factors, particularly language attitude and political pressure, may prove to be of greater importance (the S.I.L. surveys included some language at­ titude data), but intelligibility was clearly relevant to the language policy decisions to be made. A second example of a sociolinguistic survey undertaken to facilitate language policy decisions is the Survey of Language Use and Attitudes towards Language in the Philippines (Sibayan 1 975; Sibayan, Gonzalez, and Otanes 1 975; Tucker 1 975). Among the persons interviewed were over 2,300 ho�se­ holders, in communities scattered throughout the country, who were asked questions about their language background, opinions and preferences with respect to language use, attitudes towards variation in language, and pre­ ferences concerning the use of language in the schools. In addition, a similar number of teachers were interviewed with respect to their use of languages in teaching and the languages they preferred or discouraged for particular subjects . The Board of National Education, the nation's education policy­ making body, endorsed the survey and resolved to use the results as guidelines for policy decisions regarding language. Among the changes made on the basis of the survey's results was the Bureau of Public Schools Director's order per­ mitting the use of Pilipino as the medium of instruction in the first grades of schools located in non-Tagalog speaking provinces.

ROB ERT L . COOPER

1 19

formation can help the Congress, which mandated the survey under Section 73 1 (c) (1) (A) of Title VII ESEA, plan remedial educational programs.

Evaluation ofLanguage Policy

study of a large number of Jordanians from varying educational and oc­ cupational backgrounds. In this study, the relationships among a number of variables were examined, including demographic characteristics, self-reported English proficiency, self-reported English usage in specified contexts, and attitudes towards English and English instruction. The survey' s authors proposed a number of recommendations, based on their findings, with respect to English-language policy, curriculum change, teacher training, and research. Although, according to one observer, some of the recommendations are unlikely to be implemented and others are likely to have been implemented even if there had been no survey, some changes were made as a direct result of the survey (Ibrahim 1 975). It can be pointed out in passing that the Irish and Jordanian studies illus­ trate the fact that language policy surveys need not be confined to linguistically heterogeneous settings. From the point of view of mother-tongue distribution, both countries are quite homogeneous, although the situation in Jordan is complicated by the presence of a classical variety. The examples cited thus far demonstrate that sociolinguistic surveys which yield data that can be used for one of the three purposes listed above (language policy decisions, implementation planning, and evaluation) can often be used for one or both of the others as well. Thus information collected for purposes of policy evaluation can serve as guidelines for the formulation of new policies or the modification of old ones. Information gathered for purposes of language policy formulation can sometimes be used for planning the im-

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

The formulation o f language policy and the planning of its implementation are two aspects of language planning. A third aspect is the evaluation of the effectiveness of language policy (Rubin 1 97 1 ; Gorman 1 975). One of the best examples of a survey designed to evaluate the outcomes of language policy is Macnamara's study of Irish children's achievement in school. His primary objectives were to determine (1) the effect on arithmetic achievement of teaching arithmetic via the medium of Irish to children whose first language is English and (2) the effect on the level of English achievement of the program to revive Irish through its promotion in the schools. Among his conclusions were the following: (1) teaching arithmetic through the medium of Irish was associated with retardation in problem (but not mechanical) arithmetic; (2) native-English speakers in Ireland who had spent 42% of their primary-school years learning Irish did not achieve as well in written English as their mono­ lingual British peers; and (3) the written Irish attainment of the native-English speakers was below that of the native-Irish speakers. His results, then, strongly suggest that the policy of promoting Irish through its study and use in the schools was an expensive one from the point of view of educational achieve­ ment, and his conclusions were probably influential in the subsequent weakening of this policy. The English-Language Policy Survey of Jordan (Harrison et a/. , 1 975; Ibrahim 1 975; Tucker 1 975) serves as another example of a sociolinguistic survey conducted for purposes of policy evaluation. Here the effectiveness of the Jordanian English-language program was assessed by means of a field

1 20

SO C I O L I N G U I S T I C S U R V E Y S

plementation of policy. Similarly, data gathered i n order t o plan the im­ plementation of policy may lead to the modification of policy if the data in­ dicate that the original policy is unworkable or too expensive. Again, in­ formation collected for the purpose of determining language policy or planning its implementation can serve to evaluate the impact of past policies. Thus the first three categories proposed for the characterization of socio­ linguistic survey uses are overlapping and complementary.

Hypothesis Testing

Language Promotion Sometimes a survey is carried out not only for the data to be collected there­ by but also for the opportunity to promote the surveyed language. An out-

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Not all sociolinguistic surveys are conducted with immediate practical concerns in view. Some are conducted in order to test hypotheses about the relationships between social structure and language use. Labov's ( 1 966) work is a notable example of this type of survey. By means of unobtrusive ob­ servations and interviews, he tested the hypothesis that variations in the pronunciation of New York City English which heretofore had been viewed as ' free, random or unexplained' , was in fact a function of social stratification and the formality or casualness of the speech situation. His data clearly supported this hypothesis. Ma and Herasimchuk ( 1 97 1 ) extended Labov's techniques in a survey of a bilingual Puerto Rican community near New York City. Their hypothesis that demographic variables and contextual formality would account for phono­ logical variation in both English and Spanish was only partially supported by their data. Contextual formality accounted for variation in Spanish but not in English, and demographic variables accounted for variation in English but not in Spanish. The demographic variables, which were related to the opportunity to learn English and thus to proficiency in English, were not related to the opportunity to learn Spanish, and social stratification was not great enough to be reflected in the pronunciation of Spanish. Another example of a hypothesis-testing survey can be found in a study of language usage in Ethiopian markets (Cooper and Carpenter 1 969). One of the purposes of this study was to determine whether a lingua franca for trade had developed. It was hypothesized that either Amharic ·or Arabic would be ser­ ving this function. Amharic was predicted because of its widespread dispersion throughout the towns of the country. Arabic was predicted because in Ethiopia Muslims have traditionally dominated small retail trade and because one might expect Ethiopian Muslims to have some knowledge of Arabic through attendance at Quranic schools. Besides, there is a widespread belief in Ethiopia that Arabic is used as a trade lingua franca between Muslims. A survey of 23 markets in 8 towns suggested, however, that neither Amharic nor Arabic nor any other language was serving as a trade lingua franca. Thus the hypothesis was not supported. In the markets studied, it appeared that the seller accommodated himself to the buyer by speaking the latter's language. This suggested that languages can spread through selected routes. Amharic, which appears to be spreading through urbanization, industrialization, and education, does not seem to be spreading through trade, although trade is a classic avenue for the spread of a lingua franca.

R O B E R T L. C O O P E R

121

O AT A - C O L L E C T I O N T E C H N I Q U E S

Sociolinguistic surveys have employed all of the techniques used by surveys in the social sciences more generally: interviews, tests, questionnaires, rating scales, non-reactive observations of ongoing behavior, and content analyses of printed materials. With respect to many of the techniques which are used in sociolinguistic surveys, there is little to distinguish them from those used in other types of social science survey, except, of course, for the subject matter involved. Others, however, if not unique to sociolinguistic surveys, have been developed in connection with them. The contribution of sociolinguistics to survey methodology has been clearest for interviewing, testing, attitude scales, and non-reactive observations . Some of these contributions are outlined below.

Interviewing One problem encountered in interviewing, from the point of view of linguistic fieldwork, is that respondents tend to interpret the situation as a formal one and accordingly use speech appropriate to formal situations. One of Labov's methodological contributions is what we may call the ' socio­ linguistic interview' . Here elicitation techniques are employed which are designed to obtain speech which varies along a continuum of carefulness or casualness. Thus respondents are asked, for example, to read word lists and paragraphs; they are asked questions calculated to encourage them to speak in a measured, careful manner; and questions are asked or topics introduced

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

standing example of such a survey is the sociolinguistic survey of Barcelona by Antoni Badia i Margarit, carried out in 1 964-65 (Robinson 1975). Badia sent questionnaires to almost 22,000 people or 1 .40Jo of the city's population as reported in the 1960 census records, and about 3 ,400 completed questionnaires were returned. One goal of this survey was to obtain information on the status of Catalan in Barcelona. Accordingly respondents were asked about their knowledge and usage of Catalan and about their attitude towards its use. A second goal of the survey was to stimulate public interest in Catalan. One of Badia's reasons for selecting such a large initial sample was 'the desire that his questionnaire enter as many homes, and hence inspire as much commentary, as possible' (Robinson 1975). To promote interest in Catalan, he has lectured and published frequently on topics related to the survey. In addition, the results are being published in Catalan, the first volume having appeared in 1 969 (La Llengua dels Barcelonins: Resultats d'una enquesta sociologico­ lingiiistica). Although, as Robinson points out, publication in Catalan limited the book's message to those who needed it the least, the use of Catalan has had symbolic value, demonstrating that Catalan is an appropriate written medium for scientific discourse. Badia's work is remarkable when one considers not only that he was the sole investigator in this large-scale study and that he had had no previous experience with social science research methods, but also that he worked without any encouragement or support from the governmental and educational establishments. The official climate under which he worked was most unfavorable. A� the time, Catalan was not used in public life. His survey and his publicizing of it were directed towards a change in this situation, and the situation has in fact begun to change in the direction of greater public use.

1 22

SOCIOLINGU ISTIC SURVEYS

which are designed t o make the respondent forget the constraints imposed by the interview situation and to speak freely. Thus a much more representative sample of the respondent's verbal repertoire is obtained than through con­ ventional linguistic elicitation techniques.

Testing

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

There are at least two types of test which have been developed in connection with sociolinguistic surveys. These are intelligibility tests and contextualized proficiency tests. Intelligibility tests are designed to determine the extent to which speakers of one language or language variety can understand speakers of another related language or language variety. After Voegelin and Harris ( 195 1 ) suggested that objective intelligibility tests were possible, linguists began to construct in­ creasingly sophisticated intelligibility measures. But intelligibility testing reached its fullest development within the context of sociolinguistic surveys, particularly in the Mexican surveys of the Summer Institute of Linguistics in Mexico, described above, and the Survey of Language Use and Language Teaching in Eastern Africa. The design developed for the Ugandan survey (Ladefoged et a/., 1 97 1 ) was particularly elegant. A problem in all previous intelligibility testing was that different passages were employed to represent different languages. Thus, for example, a respondent would hear language A in passage 1 , language B in passage 2, language C in passage 3 , etc. For this reason it was never clear to what extent differences between intelligibility scores were due to differences associated with the languages and to what extent they were due to differences associated with the content of the passages . The Ugandan survey solved this problem by translating the n languages studied into each of n passages, creating n 2 texts. These texts were then distributed among n test tapes in such a way that each test tape contained an example of each language and each passage. Respondents were then randomly divided into n subgroups, each subgroup hearing a different test tape. Since intelligibility scores were determined for the group as a whole, the average score of each language was based on the same n passages, not on one passage alone. Thus differences in the content of the passages could not contribute to differences in the average score obtained for each language. Contextualized proficiency tests were developed for use in sociolinguistic surveys after Fishman ( l %8) pointed out that global tests of bilingual proficiency may mask differences that exist for some communicative functions or contexts, particularly in situations of stable societal bilingualism. A bilingual word-naming task developed by Lambert ( 1 955) for example, was modified to reflect proficiency in each of five hypothesized domains-family, , neighborhood, religion, education, and work (Cooper 1 969) . Administered to Spanish-English bilinguals representing a Jersey City neighborhood, the average score, totaled across all five domains, was the same in English as it was in Spanish. There was, however, a statistically significant interaction between domain and language so that for some domains there were more words produced in Spanish than in English whereas the reverse was true for other domains. Such differences would have been missed if only an undifferentiated, global score had been obtained for each language.

R O B E R T L. C O O P E R

123

Attitude Scales

Non-Reactive Procedures Public resistance to being surveyed is as old as surveys themselves. The Bible, for example, reports opposition to David's census (Second Samuel 24), probably because the enumeration was to serve as a basis for taxation and military levies. Respondents' willingness to answer direct questions honestly varies, of course, with the nature of the questions and the respondents' per­ ception of them. Even if willing, however, respondents are sometimes unable to report their own behavior accurately. This is probably the case for most phonological variation, for example. Because of limitations in respondents' willingness and ability to answer questions accurately, non-reactive pro­ cedures-techniques which gather data without people realizing they are being observed as part of an investigation-are quite useful. Another reason for the usefulness of such techniques is that they typically enable the investigator to gather an exceptionally large number of observations at minimum cost. Two types of non-reactive measures have been developed for use in sociolinguistic surveys. These are staged encounters and transaction counts. One of the best-known examples of a staged encounter is that invented by Labov ( 1 966) for use in his study of the social stratification of post-vocalic (r) in New York City department stores. He selected thf'ee stores each of which concentrated on goods at a different part of the price continuum and whose

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Because language attitude is a central concern in sociolinguistics, language attitude measures have become associated with sociolinguistic surveys. (For a methodological and substantive review of language attitude studies, see Agheyisi and Fishman 1970; for collections of language attitude studies, see Cooper 1974; Shuy and Fasold 1973 .) Perhaps the most distinctive of these measures are those involving the use of aural stimuli. Such measures were first developed by Lambert and his colleagues (Lambert 1 967) as the 'matched­ guise' technique. In its classic form, this technique presents a series of tape­ recorded voices reading a standard passage, half the passages in one language and half in another. Respondents are asked to rate the personality of each speaker, typically on semantic-differential scales, and to use voice cues only as the basis for judgment. Unknown to the respondents, all speakers are heard in both languages, so that differences in the average ratings for each language cannot be attributed to differences between speakers. Between-language dif­ ferences presumably stem from attitudinal differences, either towards the languages or towards the groups represented by the languages. Variations of this technique have been developed for use in surveys in, among other places, Kampala (Scotton 1 972), Jersey City (Fishman et a/. , 1 97 1 ), and Jerusalem (Cooper and Fishman 1 974) . Another use of aural stimuli to measure language attitude is seen in Labov's use of subjective reaction tests. Here, evaluations were obtained in response to tape-recorded sentences embodying particular variants of given phonological variables. Respondents were asked to indicate the type of job for which the speech heard on the tape would be acceptable. Since the stimulus tapes were chosen so as to vary principally in the speakers' treatment of particular phono­ logical variables, the average ratings indicated which variants were perceived as superior and which were stigmatized.

124

SO C I O L I N G U I STIC S U RV E Y S

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

customers and staff presumably reflected similar differences along a con­ tinuum of social stratification. At each store, he asked employees for the location of an item which was sold on the fourth floor. The answer, fourth floor, contained two examples of postvocalic (r), one before a consonant and the other at the end of a word. He asked each respondent twice, the first time casually and the second time as if to suggest that he had not heard the first reply. Thus from most respondents, Labov obtained four examples of post­ vocalic (r) , two casually produced, in response to the first query, and two emphatically produced, in response to the second. As Labov points out, a customer's request for directions is such a familiar occurrence that the en­ counter barely enters the consciousness of the employee, who can scarcely have realized that he or she was a subject in an investigation. Another example of a staged encounter is one developed by Rosenbaum et a/. ( 1 977). In an effort to determine the percentage of pedestrians on a busy street in Jerusalem who could speak English, the investigators posed as tourists asking for directions in English. Half the time pedestrians were asked to give directions to an address written in English and half the time they were asked the address orally. Again, the respondents were not aware that they were participating in an investigation. Whereas staged encounters require an interaction, however minimal , between the investigator and the person observed, transaction counts do not require the investigator to participate in the observed interactions . The trans­ action-count procedure (Bender et a/. , 1 973) enumerates transactions, en­ counters, conversations, or responses observed in each of various specified languages under specified conditions . Thus for example in each of the Ethiopian markets studied in a survey of language usage (Cooper and Car­ penter 1 969), transactions were observed in each of the market's major com­ modities and the language in which each transaction was conducted and was recorded. Similarly in a survey of language usage in Ethiopian law courts (Cooper and Fasil 1976), in each court studied the languages used by the various participants in each court case were recorded (e.g., judge to witness , judge to plaintiff). A third example is the study by Rosenbaum et a/. , men­ tioned above in connection with staged encounters, who also recorded the languages overheard on the street, as used by pedestrians with one another, as well as the languages overheard in the shops which lined the street, as used by customers with one another and as used by customers with shopkeepers and shopkeepers with one another. All measurement procedures contain their own sources of error. Thus sole reliance on any one type is dangerous. To the extent that agreement exists among measures obtained from different types of procedure, we can be more confident in our results. The use o f non-reactive measures is relatively un­ common in social science surveys, so there are not many examples of the combined use of reactive and non-reactive measures. This is also true in socio­ linguistic surveys. However, in two of the studies cited here, both types were used and the results obtained from one type of measure were consistent with the results obtained from the other. Thus Rosenbaum et a/. , asked shop­ keepers what foreign language was most commonly heard in their shops and on the street outside, and the most frequently given answer (English) agreed with the results obtained from the transaction counts . Similarly, the results

R O B E RT L . C O O P E R

1 25

obtained by Labov from his staged encounters in department stores agreed with those he obtained from his interviews on the Lower East Side. C O N C L U D I N G R EM A RKS

(Received July 1979)

R E F E R E N CES

Agheyisi, Rebecca and Fishman, Joshua A., 1 970. ' Language attitude studies: a brief survey of methodological approaches' . Anthropological Linguistics, 12, 1 37- 1 57 . Bender, M. Lionel, Cooper, Robert L. and Ferguson, Charles A., 1 975. 'Language in Ethiopia: implications of a survey for sociolinguistic theory and method' . Lflnguage in Society, 1972, I , 2 1 5-23 3 . Also in Ohannessian Sirarpi, Ferguson, Charles A., and Polome, Edgar, (eds.), Language Surveys in Developing Nations: Papers and Reports on Sociolinguistic Surveys. Washington, D.C.: Center for Applied Linguistics, pp. 1 9 1 -208. Casad, Eugene H . , 1 974. Dialect Intelligibility Testing. Norman: Summer Institute of Linguistics of the University of Oklahoma. Cooper, Robert L . , 1 969. 'Two contextualized measures of degree of bilingualism' . Modem Language Journal, 5 3 , 1 72- 1 78 .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Sociolinguistic surveys vary in their technical sophistication, which is not surprising in view of the fact that many of the investigators who have under­ taken them have had little or no training in social science research methods, particularly in the collection, processing, and analysis of mass data. Such shortcomings that result from lack of training can be and usually are overcome with experience, but the investigator who must learn for himself what could have been conveniently acquired as part of his professional training subjects himself to additional delays and frustrations. This comment is intended as a criticism not of the investigator but of the parochialism of professional training programs . Indeed, the commonly heard injunction for inter­ disciplinary awareness has rarely been followed as completely as it has been in the case of investigators who undertake sociolinguistic surveys. Many sociolinguistic surveys, of course, are already carried out at very accomplished levels with respect to design, data collection, and data analysis. It must be remembered, however, that technique exists for the service of ideas. Without content, technical virtuosity is barren. Sociolinguistic surveys can be useful not only to the extent thauheir techniques are appropriate and sound but, more importantly, to the extent that the information they are designed to collect is worth gathering, i.e., the hypotheses they test are worth testing, the problems which motivate them are worth solving, and the answers they seek are relevant to the issues at hand. Good questions are harder to ask than to answer. Good hypotheses and problems are harder to formulate than are techniques. The state of the art of sociolinguistic surveys therefore depends ultimately on the state of sociolinguistics itself.

1 26

SOCIOLINGU ISTIC SURVEYS

Cooper, Robert. L . , 1 974, (eel.) 'Language Attitudes I' (Special issue of the International Journal of the Sociology of Language, 1 (3).) Cooper, Robert L. and Carpenter, Susan, 1 969. 'Linguistic diversity in the Ethiopian Market ' . Journal of African Languages, 8 (Part 3), 1 60- 1 68. Cooper, Robert L . and Nahum, Fasil, 1 976. ' Language in the court' . In Bender, M. L., et al. Language in Ethiopia. London: Oxford University Press. Cooper, Robert L. and Fishman, Joshua, A . , 1 974. 'The study of language attitudes'. International Journal of the Sociology of Language, 1 (3). Cooper, Robert L. and Horvath , Ronald, J . , 1 97 3 . ' Language, migration, and urbanization in Ethiopia. Anthropological Linguistics, 1 5 , 221 -243.

Cooper, Robert L . , Singh, B. N. and Abraha Ghermazion, 1 976. 'Mother tongue and other tongue in Kefa and Arusi' . In Bender, M. L . , et al. Language in Ethiopia. London: Oxford University Press, pp. 2 1 3-243. Das Gupta, Jyotirindra, Ferguson, Charles A . , Fishman, Joshua A . , Jernudd, Bjorn, Rubin, Joan, et a/. 1 972. Draft Report of International Research Project on Language Planning Processes. Mimeographed. Ferguson, Charles A . , 1 966. 'On sociolinguistical1y oriented language surveys' . The Linguistic Reporter: Newsletter of the Center for Applied Linguistics, 8 (4), 1-3. Also in Ohannessian Sirarpi, Ferguson, Charles A. and Polome, Edgar (eds.) Language Surveys in Developing Nations: Papers and Reports on SociolinguistiC Surveys. Washington, D.C.: Center for Applied Linguistics, pp. 1-5. Fishman, Joshua A., 1 968. ' Sociolinguistic perspective on the study of bilingualism' . Linguistics, 39, 2 1 -50. Fishman, Joshua A., 1 975. ' Some implications of "The International Research Project on Language Planning Processes (IRPLPP)" for sociolinguistic surveys' . In Sirarpi Ohannessian, Ferguson, Charles A. and Polome, Edgar (eds.) Language Surveys in Developing Nations: Papers and Reports on Sociolinguistic Surveys. Washington, D.C.: Center for Applied Linguistics, pp. 209-220. Fishman, Joshua A . , Cooper, Robert L . , Ma, Roxana et a/. , 1 97 1 . Bilingualism in the Barrio. Bloomington: Research Center for the Language Sciences, Indiana University. Gorman, Thomas P., 1 97 5 . 'Introductory essay: approaches to the study of educational language policy in developing nations ' . In Harrison, William, Prator, Clifford and Tucker, G. Richard (eds.) English-Language Policy Survey of Jordan: A Case Study in Language Planning. Washington, D.C.: Center for Applied Linguistics, pp. xi-xii. Harrison, William, Prator, Clifford and Tucker, G. Richard (eds.), 1 97 5 . English­ Language Policy Survey(' of Jordan: A Case Study in Language Planning. Washington, D.C.: Center for Applied Linguistics.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Cooper Robert L. and Singh, B. N . , 1 976. ' Language and factory workers ' . In Bender, M. L . , et al. Language in Ethwpia. London: Oxford University Press.

R O B E RT L. C O O P E R

127

Ibrahim, Muhammad H . , 1 97 5 . 'The implementation and impact of the English­ Language Policy Survey of Jordan' . Paper prepared for the International Conference on the Methodology of Sociolinguistic Surveys, Montrea l, May 1 9-21 . Jahoda, Marie, Deutsch, Morton and Cook, Stuart W., 195 1 . Research Methods in Social Relations with Especial Reference to Prejudice. New York: Dryden, Two volumes. Jernudd , Bjorn H . , 1 975. 'How unique are sociolinguistic surveys?' Comment prepared for the International Conference on the Methodology of Sociolinguistic Surveys, Montreal, May 1 9-2 1 . Labov, William, 1 966. The Social Stratification of English in New York City. Washington, D.C.: Center for Applied Linguistics.

Lambert, Wallace E., 1 955. 'Measurement of the linguistic dominance of bilinguals' . Journal ofAbnormal and Social Psychology, 50, 1 97-200. Lambert, Wallace E., 1967. 'A social psychology of bilingualism '. Journal of Social Issues, 23 (2), 9 1 - 1 09.

Le Page, R. B., 1 972. ' Preliminary report on the sociolinguistic survey of Cayo District, British H� nduras'. Language in Society, I , 1 55-172. Le Page, R. B., 1 975. ' "Projection, focussing, diffusion" , or steps towards a sociolinguistic theory of language, ' illustrated from the Sociolinguistic Survey of Multilingual Communities, Stages 1 : Cayo District, Belize (formerly British Honduras) and I I : St. Lucia. Paper prepared for the International Conference on the Methodology of Sociolinguistic Surveys, Montreal, May 19-2 1 . Lieberson, Stanley, 1 970 .. Language and Ethnic Relations in Canada. New York: John Wiley. Lindzey, Gardner and Aronson, Elliot (eds .), 1 969. The Handbook of Social Psychology, volume 2, Research Methods. Reading: Addison-Wesley. Ma, Roxana and Herasimchuk, Eleanor, 1 97 1 . 'The linguistic dimensions of a bilingual neighborhood' . In Fishman, Joshua A., et al. Bilingualism in the Barrio. Bloomington: Research Center for the Language Sciences, Indiana University. Macnamara, John, 1 966. Bilingualism and Primary Education: a Study of Irish Experience. Edinburgh: Edinburgh University Press. Moser, C. A., 1 958. Survey Methods in Social Investigation. New York: Macmillan. Robinson, Joy, 1 975. 'Catalan in Barcelona: La Llengua dels Barcelonins of Antoni Badia i Margarit' . Paper prepared for the International Conference on the Methodology of Sociolinguistic Surveys, Montreal, May 1 9-2 1 . Rosenbaum, Yehudit, Nadel, Elizabeth, Cooper, Robert L . and Fishman, Joshua A., 1 977. 'English on Keren Kayemet Street' . In Fishman, Joshua A., et a/. The Spread

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Ladefoged, Peter, Glick, Ruth and Criper, Clive, 1 97 1 . Language in Uganda. Nairobi: Oxford University Press.

1 28

SOCIO L I N G U I S T I C S U R V E Y S

of English: Th e Sociology of English as an Additional Language. Rowley: Newbury House, Mass.

Rubin, Joan, 1 97 1 . 'Evaluation and language planning' . In Rubin, Joan and Jernudd, Bjorn H . (eds.) Can Language Be Planned? Sociolinguistic Theory and Practice for Developing Nations. Honolulu: East-West Center, pp. 2 1 7-252. Scotton, Carol Myers, 1 972. Choosing a Lingua Franca in an African Capital. Edmonton and Champaign: Linguistic Research. Shuy, Roger W. and Fasold, Ralph W . (eds.), 1 97 3 . Language Attitudes: Current Trends and Prospects. Washington, D.C.: Georgetown University Press.

Sibayan, Bonifacio P . , Gonzalez, Andrew and Otanes, Fe T . , 1975. 'Organization and logistics of a sociolinguistic survey, the Philippine experience in part ' . Paper prepared for the International Conference on the Methodology of Sociolinguistic Surveys, Montreal, May 1 9-21 . Tucker, G. Richard, 1 975. 'Methodological aspects of data collection' . Paper prepared for the International Conference on the Methodology of Sociolinguistic Surveys, Montreal, May 1 9-2 1 . Voegelin, C . F . and Harris, Z . S. , 1 95 1 . 'Methods for determining intelligibility among dialects of natural languages' . Proceedings of the American Philosophical Society, 95 , 3 22-329. Warwick , Donald P. and Lininger, Charles A . , 1 97 5 . The Sample Survey: Theory and Practice. New York: McGraw Hill. Weinreich, Uriel, 1 957. 'Functional aspects of I ndian bilingualism'. Word, 1 3 , 203-233 .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Sibayan, Bonifacio P . , 1 975. 'Survey of Language Use and Attitudes towards Language in the Philippines'. In Ohannessian Sirarpi, Ferguson, Charles A. and Polome, Edgar (eds.) Language Surveys in Developing Countries: Papers and Reports on Sociolinguistic Surveys. Washington, D.C . : Center for Applied Linguistics, pp. 1 1 5-143 .

S P E E CH ACTS AND S E CON D LANGUAGE LEARNING R I C H A R D W . S C H M I DT and J A C K C . R I C H A R D S University of Hawaii and Chinese Universzty, Hong Kong

I NT R O D U CT I O N SEVERAL

1.

W H A T IS A S P E E C H ACT?

Speech act theory has to do with the functions and uses of language, so in the broadest sense we might say that speech acts are all the acts we perform through speaking, all the things we do when we speak. Such a definition is too broad for most purposes, however, for the uses to which we put speech en­ compass most human activities. We use language to build bridges, to conApplied Linguistics, Vol. I, No. 2

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

new paradigms have emerged within applied linguistics i n recent years. The Chomskyan paradigm has had a marked influence on theories of language and language learning. The goal of language learning within the Chomskyan approach is identified with the acquisition of underlying linguistic categories and systems, from which surface forms are derived through the application of transformational and other rules and processes of a universal type. Despite the addition of a philosophical framework for the theory, and . while the Chomskyan concept of language knowledge is quite different in its own terms from the concept of language knowledge implicit in pre­ Chomskyan theory, it is only a partial account of the knowledge required to use a language. This paper considers other areas of knowledge which con­ stitute an equally important dimension of the task of learning a language, with particular reference to second and foreign language learning. Sociolinguists and others have long acknowledged the limitations in the Chomskyan formulation of competence, and stressed the need to include knowledge of the rules of use and communicatively appropriate performance. Bruner, writing of first language learning, has argued that mother tongue acquisition should be looked at not as a solo flight by the child in search of dis­ embodied rules of grammar, but as a problem-solving transaction. The essential problems to be solved by mother and infant have to do with ' how to make our intentions known to others, how to communicate what we have in our consciousness, what we want done on our behalf, how we wish to relate to others, and what in this or other worlds is possible' (Bruner 1 978). In this paper we will consider second language acquisition from a similar perspective to that advocated by Bruner for first language acquisition, focussing on the development of communicative rather than linguistic or grammatical com­ petence. While communicative competence theory covers a range of different dimensions of language behaviour in the individual and in the speech com­ munity, we will focus on one aspect of communicative competence, namely, speech acts, and consider the contribution of speech act theory to our under­ standing of second language acquisition.

130

S P E E C H A C T S AND S E C O N D L A N G U A G E L E A R N I N G

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

solidate political regimes, to carry on arguments, t o convey information from one person to another, to entertain, in short to communicate. We use speech in ceremonies, games, recipes, and lectures. On some occasions, e.g . , social · gatherings, we use language successively to introduce one person to another, carry on conversations, tell jokes, criticize and praise third parties both present and absent, expound on favourite topics, seduce or attempt to seduce, and say farewell. We could extend such lists indefinitely, but as Halliday ( 1 973 : 1 8 , 28) has pointed out, such lists do not by themselves tell us very much, for the innumerable social purposes for which adults use language are not represented directly, one to one, in the language system. Hymes (1 972) has }:>roposed a useful distinction between speech situations, speech events, and speech acts. Within a community one finds many situations associated with speech, such as fights, hunts, meals, parties, etc. But it is not profitable to convert such situations into part of a sociolinguistic description by simply relabelling them in terms of speech, for such situations are not in themselves governed by consistent rules throughout. The term speech event can be restricted to activities that are directly governed by rules or norms for the use of speech, events such as two party conversations (face-to-face or on the telephone), lectures, introductions, religious rites, and the like. This notion of speech event is related to the traditional concept of genre, though Hymes argues that the two must be treated as analytically independent, and a great deal of empirical research is needed to clarify the relationship between the terms. Speech acts (in a narrow sense now) are the minimal terms of the set: speech situation/event/act. When we speak we perform acts such as giving reports, making statements, asking questions, giving warnings, making promises, approving, regretting, and apologizing. Sinclair and Coulthard (1975), who have analyzed classroom transcripts, also propose a 'top-down' analysis, beginning with the social occasion (the lesson) as the outermost analytic frame and successively dividing and sub­ dividing the sequence of discourse down to the smallest unit, the act, which they define as the minimal unit of speaking which can be said to have a func­ tion. Acts are labelled according to discourse function, e.g . , elicitation, question, etc. In this paper we will be focussing primarily on individual speech acts. How­ ever, it is necessary to look somewhat beyond the isolated act represented by the individual sentence, primarily the verb. Austin (1 962) pointed out that there are a great number of speech acts (illocutionary acts, in his terminology) and in English there are a great number of verbs which refer to them. Consider for example just the related set : ask, request, direct, require, order, command, suggest, beg, plead, implore, pray. Austin claimed that there are over a thousand such verbs in English. But while English verbs provide a useful initial taxonomy for speech acts , the acts are not in fact equivalent to the verbs which frequently name them. Searle (1976) points out that many verbs are not markers of illocutionary force, but of some other feature of the speech act. Insist and suggest, for example, mark degree of intensity, but do not mark separate speech act functions or illocutionary points. Both may be used with directive function ('I suggest/insist that we go to the movies') or with repre­ sentative function ('I suggest/insist that the answer is found on page 16'). We need to recognize also that speech acts are not identifiable with the sentence, or

R I C H A R D W. S C H M I DT A N D J A C K C. R I C H A R D S

131

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

any other level of grammatical description. Hymes' (1 972) position is that the level of speech acts mediates between the usual levels of grammar and the rest of a speech event in that it implicates both linguistic form and social norm. Whether or not a particular utterance has the status of a request, for example, may depend upon a conventional linguistic formula ('How about picking me up early this afternoon? '), but it may also depend upon the social relationship between speaker and hearer. It needs to be recogitized too that speech acts occur within discourse, and that the interpretation and negotiation of speech act force is often dependent on the discourse or transactional context. As a minimum, we need to consider the fact that talk is often organized into two-part exchanges. As Goffman ( 1 976) points out, this organizing principle follows from very fundamental requirements of talk as a communication system. A speaker needs to know whether his message has been received and understood; a recipient. needs to show that he has received and understood the message. We therefore must recognize such 'adjacency pairs' as summons-answer (Schegloff, 1 968), statement-reply (Goffman, 1976), question-answer, request-refusal of request, and the like. An investigation of speech acts therefore leads naturally into questions of act sequencing (events) and contexts (speech settings or situations). Rehbein and Ehlich, quoted in Candlin ( 1 978), list the different operations that may take place inside a restaurant when the activity is ordering a meal: entering, looking around, judging, taking a seat, wanting the menu, asking for the menu, wanting information, asking for information, consulting, deciding, ordering, transmission, production, delivery, serving, consuming, wanting to pay, asking for the bill, drawing up the account, getting/presenting the bill, accepting the bill, paying, leaving. Norms of linguistic behaviour identify various parts of the sequence. Different participants have different amounts of talking to do and different types of talking, as well as different topics to talk about. Within speech events there are norms for opening and closing sequences, sequencing rules, and distribution frequencies and probabilities for particular speech acts. 'Assigning the value command to any of a range of possible utterances ("hot dog", "that one", "please bring me X" , a deictic gesture) is a function of recognizing the social world of the restaurant with the rights, duties and social relationship between the participants, as well as that of being aware of the discoursal position of the "act of commanding" within the transactional process. ' (Candlin, 1978, p. 17 .) Both speech acts and speech events have been studied extensively in recent years and have constituted topical foci for scholars from a great number of disciplines. Speech events have been investigated by anthropologists and ethnographers (Albert, 1964, Gumperz and Hymes, 1972, Sanches and Blount, 1 975), folklorists (Abrahams, 1 962, Dundes et al. , 1972), literary critics (Pratt, 1 977), and sociologists (Allen and Guy, 1974). The most detailed and perhaps the most provocative analyses of speech events have been provided by those sociologists who work within the area of sociology termed ethnomethodology, the primary goal of which is to give rigorous sociological formulation to the interactional basis of the things people say and do in the settings of everyday life. Working primarily from transcripts of natural conversations, charac­ terizations have been developed for a variety of conversational activities : turn

S P E E C H ACTS A N D SECOND L A N G UAGE LEARN I NG

1 32

taking (Sacks et a/. , 1 974), story telling and identity negotiations (Sacks, 1 972), opening and closing conversations (Schegloff and Sacks, 1 973), telephone conversations (Schegloff, 1 968), and many other aspects o f the establishment and management of social relations through conversational roles (Sudnow , 1 972, Schenkein, 1978, Garfinkel, 1 967, Goffman, 1 972 and 1 976). Speech acts, on the other hand , have been studied primarily by philosophers of language (Austin, 1 962, Searle, 1969 and 1976, Grice, 1 968 and 1 975) and linguists (Ross, 1 970, Gordon and Lakoff, 1 97 1 , Cole and Morgan, 1 975).

2.

T HEORETICAL QUESTIONS

The following are some o f the major theoretical issues discussed i n the speech act literature. For linguistic analysis, the units of concern are sentences . Contrasts between well-formed and ill-formed (ungrammatical) sentences are primary data. While the grammatical paradigm has been followed by many linguists who have dealt with issues in speech act theory (see most of the papers in Cole and Morgan, 1 975) and while basic semantic differences are indeed likely to have . syntactic consequences (Searle, 1 976), speech acts are in essence acts, not sentences. Speech acts cannot be equated with utterances either, for we often perform more than one act (e. g . , inform and request) with a single utterance ' I ' m hungry' . Finally, speech acts cannot be equated with the notion of turn as an interactional unit, as it may take several speaker turns to accomplish a single act, or, conversely, several acts may be performed within a single speaker turn. So far we have presented only a very vague description of what speech acts are. Perhaps the notion is best clarified by examples, with some effort to group together illocutionary acts into major types. Searle ( 1 976) presents the clearest taxonomy. For Searle, the basis for classification is 'illocutionary point' or purpose of the act, from the speaker's perspective. According to Searle, speech acts can be grouped into a small number of basic types based on speaker intentions : One of the basic things we do with language is tell people how things are. We assert, claim, say, report, and the like. The point or purpose of this class of representatives is to commit the speaker in varying degrees (suggest, doubt, and deny are members of this class also) to the truth of something. One test of a representative is whether it can be characterized as true or false.

Representatives.

When we use language, we do not j ust refer to the world and make statements about it. Among our most important uses for language is trying to get people to do things. The class of directives includes all speech acts whose primary point is that they count as attempts on the part of the speaker to get the hearer to do something. Suggestions, requests and commands are all directives. They differ in the force of the attempt, but are all attempts by the speaker to get the hearer to do something.

Directives.

Commissives.

Commissives are those illocutionary acts whose point is to

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

2. 1 Units and categories

R I C H A R D W. S C H M I D T A N D J AC K C. R I C H A R D S

1 33

commit the speaker to do something. Promises and threats both fall into this category, the difference between them being the speaker' s assumption about whether or not the promised action is desired by the hearer. Searle makes the interesting point that there is a difference in the direction of fit between the words of a speech act and the state of affairs in the world when comparing representatives with directives and commissives. With repre­ sentatives the direction of fit is words-to-world, i.e., what is at issue is whether the words uttered ('The world is flat') match the world. With both repre­ sentatives and commissives the direction of fit is world-to-words. Future actions are to be done in accordance with words previously uttered. The basic distinction between requests and commissives is that hearer actions are the point of requests and other directives, while speaker actions are the issue with promises and other commissives.

actions, regret, thank, welcome, etc. With expressives there is no direction of fit, but the state of affairs specified in the following proposition is simply assumed to be true. Note also that while representatives, directives and com­ missives are all associated with a consistent psychological dimension (belief, by expressives wish and intent, respectively), the psychological states expressed · are extremely varied.

Declarations. Some speech acts bring about changes in the world simply through their successful execution. 'You're fired,' says the boss, and the employee must start the search for a new position. 'I do,' say the bride and groom, and after the presiding official (secular or clerical) says his part the marriage has taken place. The defining characteristic of this class is that the performance brings about the correspondence between the words and the world. This class is closest to Austin's ( 1962) original notion of a per­ formative, an act of doing something in the world rather than an act of saying alone.

Other classes, major and minor. Several taxonomies have been proposed in addition to that of Searle. Fraser ( 1 975) adds a few categories. In addition to acts of asserting ( = Searle's representatives), he includes acts of evaluating, the point of which is to express the speaker's assessment of the truth of a proposition and the basis of the judgement, e.g., analyse, conclude, hypothesize. In addition to acts of requesting ( = Searle's directives), Fraser has a category of acts of suggesting, e.g., recommend, suggest, urge. Acts of stipulating express a speaker's desire for the acceptance of a naming con­ vention expressed by the proposition, e.g., call, classify, designate. Hancher ( 1 979) has suggested two additional kinds of acts, those that combine com­ missive with directive illocutionary force (e.g. , offering, inviting, challenging) and those that require two participants (e.g., giving, selling, contracting). While the great majority of speech acts can probably be analysed as examples of Searle's major classes, or Fraser's somewhat longer list, there are doubtless some speech acts which are outside these particular taxonomies. Greetings and farewells, for example, constitute a small category (or categories) of acts which are not generalizable as major classes, but which

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Expressives. The point of this class is to express feelings and attitudes about states of affairs. We apologize for things we have done, deplore other people's

1 34

S P E E C H ACTS AND SECO N D L A N G U A G E L EA R N I N G

deserve attention. I t is also useful t o mention such acts as

refusal of a request,

although utterances which fall into such a category will in most cases be already classifiable in terms of the basic act types: 'I'm sorry, but I can't' expressive + representative; ' I ' ll be able to see you tomorrow' (not today) commissive; ' Do it yourself directive. =

=

=

2.2 How to perform a speech act Searle ( 1 965) has attempted to provide analyses of various illocutionary acts, asking what conditions are necessary and sufficient for a particular act to have been performed by the uttering of a particular sentence. For promises, the conditions are identified as follows:

Normal input and output conditions obtain, i.e., the speaker and hearer are not insane, they are not play acting, etc. A speaker expresses a sentence, the propositional '
Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Performative verbs. From Austin's original notion of a performative come the current and important terms performative verb and explicit performative (sentence or utterance). These are verbs (sentences, utterances) which explicitly name the acts being performed, e.g . , 'I promise to be there ' , an explicit per­ formative which can be contrasted with the implicit 'I'll be there ' . There are certain syntactic requirements generally assumed to hold for a verb to function performatively, such as the requirement that the subject (if expressed) be first person, the addressee (if expressed) second person and the requirement that the verb be in the present tense. Thus ' I promise you that I'll be there' is ex­ plicitly performative, while ' He promised that he'd be there' is not-in fact it is not a promise (commissive) at all, but rather a report (representative). While most authors see the performative as a sentence type with such syntactic requirements, Fraser ( 1 975) demurs, arguing that strict syntactic requirements cannot be proved, favouring instead a distinction between strongly per­ formative examples (those which are easily seen as counting as the act denoted by the verb) and weakly performative examples.

R I C H A R D W. S C H M I D T A N D J A C K C . R I C H A R D S

135

to see that how an interchange unfolds will depend somewhat on the type of speech act involved, but that an attempt must be made 'to uncover the prin� ciples which account for whatever contrast is found on a particular occasion between what is said, what is usually meant by this, and what in fact is meant on that particular occasion of use' . Searle (1975) talks of inferential strategies and suggests how the second of the following statements could be taken as a rejection of the proposal made in the first statement. Student X: Let's go to the movies tonight . Student Y : I have t o study for a n exam.

'Step 1 : I have made a proposal to Y, and in response he has made a statement to the effect that he has to study for an exam (facts about the con­ versation). Step 2: I assume that Y is cooperating in the conversation and that therefore his remark is intended to be relevant (principles of conversational cooperation). Step 3 : A relevant response must be one of acceptance, rejection, counter­ proposal, further discussion, etc. (theory of speech acts). Step 4: But his literal utterance was not one of these, and so was not a relevant response (inference from Steps 1 and 3). Step 5: Therefore, he probably means more than he says. Assuming that his remark is relevant, his primary illocutionary point must differ from his literal one (inference from Steps 2 and 4). �tep 6: I know that studying for an exam normally takes a large amount of time relative to a single evening, and I know that going to the movies normally takes a large amount of time relative to a single evening (factual background information). Step 7: Therefore he probably cannot both go to the movies and study for an exam in one evening (inference from Step 6). Step 8: A preparatory condition on the acceptance of a proposal, or on any other commissive, is the ability to perform the act predicated in the propositional content condition (theory of speech acts). Step 9: Therefore, I know that he has said something that has the consequence that he probably cannot consistently accept the proposal (inference from Steps 1 , 7 and 8). Step 1 0: Therefore, his primary illocutionary point is probably to reject the proposal (inference from Steps 5 and 9). '

Grice's (1975) 'general principles of co-operative behaviour' likewise at­ tempt to identify presuppositions that enable the participants in a speech event to assign appropriate illocutionary value to utterances. Grice refers to four maxims: Maxim of Quantity: Make your contribution (just) as informative as is required. Maxim of Quality: Make your contribution one that is true. Maxim of Relation: Be relevant. Maxim of Manner: Avoid obscurity and ambiguity; be brief and orderly.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Searle ( 1 975, p. 63) reconstructs the steps necessary to derive the intended meaning in the following way (without proposing that these are conscious operations).

1 36

S P E E C H ACTS A N D S E CO N D L A N G U AGE L E A R N I NG

Grice gives the following example.

'Suppose that A and 8 are talking about a mutual friend C, who is now working in a bank. A asks 8 how C is getting on in his job, and 8 replies: "Oh, quite well, I think; he likes his colleagues, and he hasn't been to prison yet " . At this point A might well enquire what 8 was implying, what he was suggesting, or even what he meant by saying that C had not yet been to prison . . . in a suitable setting A might reason as follows: "(I) 8 has apparently violated the maxim "Be relevant" and so may be regarded as having flouted one of the maxims con­ joining perspicuity; yet I have no reason to suppose that he is opting out from the operation of the Cooperative Principle; (2) given the circumstances I can regard his irrelevance as only apparent if and only if I suppose him to think that C is potentially dishonest; (3) 8 knows that I am capable of working out step (2). So 8 implicates that C is potentially dishonest' . One of the most controversial aspects of speech act theory has to do with whether illocutionary point is part of the 'meaning' of a sentence and whether that aspect of meaning ought to be represented in the grammar of a language, in the deep structure.

2.3 . 1 The performative analysis. In traditional school grammars of English , there is an assumed fit between sentence type and illocutionary point, to wit: declarative sentences (a grammatical sentence type) are used for making assertions (a speech act category); imperative sentences are used for orders; interrogative sentences are used for asking questions (requests for verbal responses). The 'performative analysis ' is essentially an attempt to capture this relationship, by positing for all imperative sentences, for example, a highest performative clause 'I order you' in the deep structure. Ross (1 970) has claimed that declarative sentences must be derived from deep structures containing an explicitly represented performative 'I say (assert, state, etc.) to you X'. Ross presents a large number of syntactic arguments to support the existence of both pronouns in the higher clause, such as pseudo­ reflexives in sentences like, 'This paper was written by Ann and myself' . Ross does not attempt to prove that the highest performative is a specific English verb, like say or state, but simply asserts that it must be [ + performative] , [ + communication] , [ + linguistic], and [ + declarative] . In its simplest form, the performative analysis does not take us very far in understanding the relationship between linguistic form and illocutionary point. Ross's syntactic arguments have been strongly criticized (see Matthews, 1 972), and there are obvious problems with the assumed fit between sentence type and illocutionary force on semantic grounds . Declarative sentences are not always assertions, but can function as questions (when the hearer rather than the speaker is assumed to have knowledge about the proposition-Labov, 1 972), or as orders ('No one will leave this room, and that means you ! '). Syntactic imperatives may function as other speech acts than orders, e.g . , in a sentence like 'Spare the rod and spoil the chil d ' . In general the fit between sentence type and function is only typical , not absolute (Bolinger, 1 967). Sadock ( 1 970) first tackled the problem of what he called 'whimperatives ' , sentences which have imperative force but question form , e.g . , 'Will you close the door please? ' Sadock analysed such constructions as conjunctions o f

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

2 . 3 Meaning, deep structure, and surface structure

R I C H A R D W . S C H M I DT A N D J A C K C . R I C H A R D S

137

questions and imperatives. Other analyses are possible. Whimperatives could be analysed as ordinary questions (thus failing to take any account of the imperative force, but leaving this to pragmatic, extragrammatical ex­ planations), "r one could analyse them as identical in deep structure to im­ peratives (Heringer, 1 972). One could claim that forms like 'Will you shut up? ' are merely simple imperatives ('Shut up') to which tags have been added and then preposed (Green, 1 975).

I 'd like you to go now. Could you be a little quieter? Well, are you going to help me?

asserts speaker-based sincerity con­ dition S wants H to do A questions hearer-based preparatory condition H is able to do A questions hearer-based preparatory con­ dition. It is not obvious to both S and H that H will do A in the normal course of events

It is clear that the conversational postulates are not quite as neat as Gordon and Lakoff suggest. For example, one can convey requests by asserting hearer­ based conditions as well as by questioning them, e.g., ' You could be a little quieter, you know', 'From now on, when I say jump you will jump' . But as Clark and Clark ( 1 977) have pointed out, it is an extraordinary cor­ respondence when speakers make indirect requests by making u se of the social conventions that cover the proper use of requests. 2.3.3 Surface structures and contexts. Ervin-Tripp ( 1 976) has proposed a strikingly different analysis of English directives . Ervin-Tripp argues that although native speakers' understanding can be treated as inferences from literal interpretations, social factors are what determine the actual choice of directive type. Based on a number of empirical studies, Ervin-Tripp reports that need statements ('I need a match') occur between persons differing in

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

2.3.2 Conversational postulates. An entirely different approach to the analysis of whimperatives and other indirect speech acts has been proposed by Gordon and Lakoff (197 1). Following Grice ( 1 968), they argue that sentences n;1ay convey more than their literal meaning. The sentence ' It's cold in here' , when spoken by a superior to a subordinate, may convey the meaning of 'Close the window' , but that does not mean that the analysis of ' It's cold in here' should include positing an imperative force-indicating device in the deep structure. Gordon and Lakoff propose that speakers and hearers interpret such sentences by reference to conversational postulates . Thus whimperatives are to be analysed grammatically as simple questions, but interpreted as im­ peratives by means of a conversational postulate or entailment rule, such as: a speaker can convey a request by asking if the hearer intends to do the act , as in 'Will you close the door?' The conversational postulates proposed by Gordon and Lakoff are both highly predictive and intuitively satisfying. They directly relate the philosophical analyses of what is involved in certain speech acts with the forms of language. For requests, the full form of the conversational postulate is that one can convey a request by either asserting a speaker-based condition or questioning a hearer-based condition. Thus we have the following forms :

138

S P E E C H A CT S A N D S E C O N D L A N G U A G E L E A R N I N G

2.4 Universals For the purpose of investigating speech acts in the context of second language learning, perhaps the most important question is whether and to what extent the various aspects of speech acts discussed so far are universal. Consider first the basic units. Can it be safely asserted that essentially the same classes of speech events (conversations, lectures, discussions, debates etc.) and the same taxonomy of speech acts (i .e., representatives, directives, commissives, etc.) hold for all languages and speech communities? Most researchers assume that the answer to this question is yes , but in fact there has been no ethnographic research carried out to confirm or disprove the assumption. It is probably not true that all languages name the same speech acts with illocutionary verbs (does every language recognise a suggest: insist distinction?), but again, no research has been reported . The universality of the strategies for performing speech acts, particularly indirect speech acts, has been discussed in the literature. Gordon and Lakoff say that they have checked with a number of speakers of widely divergent languages and would not be surprised to find that the conversational postulates they propose were universals. Fraser ( 1 978) has recently claimed that the strategies for performing illocutionary acts are essentially the same

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

rank. Permission directives, sentences which look like requests for permission but in fact require action on the part of the hearer ('Can I have my record back?'), are usually directed upward in rank, when the hearer controls resources. Hints, which do not include a literal expression o f the act desired, are frequent in families and communal groups. The social variables which affect directive choice include age, rank, familiarity, presence of outsiders, territorial location, the seriousness of the service asked and many others . Moreover, Ervin-Tripp claims that directives do not require inference from literal interpretations. Where knowledge of obligations_ and prohibitions is shared, simple interpretation rules allow prompt understanding. Reviewing the linguistic debate over the incorporation of illocutionary point in the analysis of sentences, Sadock (1 975) suggests two methods of removing arbitrariness from current descriptions. One would . eliminate all trans­ derivational constraints that state an interaction between logic and language; the other would require a logicogrammatical treatment wherever it is possible to provide one. Sadock recognizes that the result would be two very different interpretations of sentence meaning, one very shallow and one very deep, but states that ' I am not sure that anything at all rides on this difference'. Does the difference matter for our view of the teaching/learning process? For teaching purposes, especially the preparation of materials, both the deep structure analysis of Gordon and Lakoff and the surface structure oriented analysis of Ervin-Tripp provide valuable source material . But when we consider the implications of the different models for our view of language learning, there does appear to be a difference. The logicogrammatical 'deep ' model would force us to view the acquisition of grammatical competence and the acquisition of communicative competence as essentially the same thing, while the 'shallow' model would allow us to consider the development of grammatical forms quite distinctly from the pragmatic ability to match linguistic forms with appropriate social contexts.

R I C H A R D W . S C H M I DT A N D J AC K C. R I C H A R D S

139

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

across languages . Comparing request strategies in fourteen different languages, Fraser found that the same basic strategies were available in each language . If this is correct, then Fraser is correct in claiming that acquiring social competence in a new language does not involve substantially new concepts concerning how language is organized and what types of devices serve what social functions, but only new (social) attitudes about which strategies may be used appropriately in a given context . Goffman (1 976) draws a distinction between 'system constraints' (those which follow from the requirements of any communication system), which he suggests are pancultural, and ' ritual constraints' (such as constraints regarding how each individual ought to handle himself with respect to others), which can be expected to vary markedly from society to society. System constraints - include norms such as those identified by Grice : be relevant, be informative, etc. Ochs-Keenan (1 976) has attempted to assess the status of some of Grice's maxims cross-culturally and has found that the maxim 'Be informative' does not hold in Malagasy society. Interlocutors regularly violate the maxim by providing less information than is required by their conversational partner, even though they have the required information. However, it can be argued that the maxims are universal, but that deviations from the norm force us to attempt to uncover additional maxims, motives and strategies to account for departures from an 'ideal' communication system. Perhaps the most persuasive (and most detailed) argument for the uni­ versality of speech act strategies has been put forth by Brown and Levinson ( 1 978). They point out that most speech acts are in some way threatening to either the speaker or the hearer, either by imposing on one party's freedom of action, as with acts of requesting (an attempt to restrict the freedom of the hearer) or by damaging the positive self-image of one of the parties, as with criticisms (hearer's face is damaged) and apologies (speaker's face is damaged). Brown and Levinson argue that speakers compute the level of threat involved, considering such factors as social distance, degree of power that one party may have over the other and the ranking of impositions within a particular culture, and then select a strategy for doing the act. Very threatening actions may not be done at all, and minimally threatening actions are usually done directly and explicitly. It is the great area in between which is most complex. Speakers may select a strategy of 'positive politeness' , one which minimizes the threatening action by reassuring the hearer that he or she is valued by the speaker, that somehow the speaker wants what the hearer wants, that they are members of the same in-group, etc. Or a speaker may select a strategy of 'negative politeness', redressing the threat to basic claims of territory and self-determination, for example by apologizing or being indirect and formal. Thus a request for forgiveness might be expressed in a positively polite form as 'Gimme a break, Sweetheart' or in a negatively polite form as ' I hope you'll b e able t o excuse my error' . . Brown and Levinson describe a great number of positive and negative speech act strategies and investigate their use in three languages (English, Tamil and Tzeltal). They report that they find a fine-grained parallelism in the expression of politeness in these unrelated languages, often including the minutiae of linguistic forms. They argue that interactional systematics, the basis for linguistic realizations, are based largely on universal principles.

1 40

S P E E C H A C T S A N D S E C O N D L A N G U A G E L EA R N I N G

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

There is sufficient evidence t o argue, however, that speech act strategies will be found to be universal only if they are phrased in extremely general terms. All languages have some verbs which name performative acts, for example, and some of these may be used to issue directives, but this does not mean that all such request forms in English have literal translations which function the same way in all languages. Consider the distinction between ' I request that . . . ' and 'I hereby request that . . . ' , where the 'hereby' not only makes the request yet more explicit but also lends a quasi-legal flavour to the sen­ tence. In French a similar distinction may be conveyed through quite different linguistic means, such as the use of an elaborated verb form in preference to a simple one, e.g., 'Je vous prie' as against ' Je vous prie de bien vouloir' . I t i s possible that 'hedges' o n illocutionary force may be a universal strategy or negative politeness, but while this operation may be carried out by the use of tag questions ('It was amazing, wasn't it?') or by intonation in some languages (including English}, in other languages the parallel operation may involve other devices, such as the Japanese particle ne (Brown and Levinson , 1 978, p. 1 52}. It appears that other speech act strategies can also be considered universal only if they are phrased very generally. It is perhaps the case that one can make a request in any language by referring to the hearer's ability to perform the action, but again exact translations of English sentences often fail to carry identical implied force. Searle (1975} points out that while 'Can you hand me that book?' can be translated literally into Czech, the resulting sentence will sound extremely odd to a Czech speaker. English 'can', 'could' , and 'able' when indicating a request can only be translated into Cantonese as hoyih; other modals usually translated into English as 'can' refer specifically to physical ability and do not imply directive force. A sentence like 'Can you reach the book on the top shelf? ' , if translated with the modal naahnggau ('able'}, would be answered with 'yes' or 'no ' , with no attempt to get the book by the hearer (Marcus, 1 978}. Green ( 1 975} observes that conditional forms equivalent to English 'would' ('Would you leave it on my desk when you finish, please') cannot carry imperative force in Spanish, Hebrew or Japanese, although they can in English, German and Finnish. In English, we can make requests with non-literal let 's ('Let's all think before we raise our hands'), but Cole ( 1 975} reports that in both Swahili and Yiddish such constructions are ungrammatical. Searle has argued that the mechanisms (strategies} for indirect speech are general, not peculiar to this language or that, but within this framework certain standard forms tend to become conventionally established as the standard idiomatic forms. The standard forms for one language may not maintain their indirect speech act potential when translated into another language because (a) the translation may not be idiomatic in the second language and (b) even if idiomatic, the resulting forms may not be those which are conventionally selected as devices for indirect speech acts (Searle, 1 975). Even if speech act strategies are to a certain extent universal, therefore, learners of new languages still need to learn several important things . They need to learn the particular conventionalized forms in the new language, particular applications of general principles which vary systematically among cultures and groups (and to a certain extent among individuals}. They need to

R I C H A R D W . S C H M I D T A N D J A C K C. R I C H A R D S

141

3.

I M P L ICATIONS

FOR

LANGUAGE

LEARN ING

RESEARCH

AND

T H EO R Y

The above account o f speech events and speech acts reviews the major contributions to speech act theory that have been made by linguists , philosophers, ethnomethodologists and others. We now consider in what ways speech act theory can contribute to our understanding of second language acquisition. A major contribution of speech act theory is in its clarification of dimensions of communicative competence. While the concept of com­ municative competence is not new, much remains to be done to substantiate the concept empirically, and the study of the role of speech acts in second language learning could make a useful contribution to our knowledge of how second and foreign languages are acquired. Until recently, theories of second language learning have followed, rather narrowly, models developed in linguistic theory. Thus it was widely assumed that transformational-generative grammar could serve both as a general model for language and as an explanatory model for second language learning. Within much L2 theory and research the primacy of syntax has been taken for

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

learn the general 'ethos' of the new speech community, whether the inter­ actional style in general is stiff and formal or relaxed and open. They need to learn which speech acts are particularly threatening in a particular culture. One culture might place particular emphasis on modesty and circumspection in the expression of speaker beliefs, for example, while in another community requests (or criticisms) might be especially threatening. Learners need to learn the social relationships of the community, the networks of relationships and responsibility which obtain, the kinds of acts which can be directed towards which persons, etc. (Brown and Levinson, 1 978). Learners also need to learn some very specific contexts which call for particular speech acts , which vary from society to society. Apte ( 1 974) has identified the contexts which call for 'thank you' in South Indian languages (very restricted), as opposed to American English (extensive). Ueda ( 1 974) has discussed refusals in Japanese, the situations that permit saying 'no ' , and the ways to refuse a request when a direct refusal is not possible. Candlin argues that 'interethnic and intercultural variation among mother tongues, domains of language use, inter-language attitudes and language learning purposes lead to misunderstanding, and that such misunderstanding can be understood through the study of discourse patterning. He stresses that the performance of speech acts depends on 'culturally specific appropriateness criteria' (Candlin, 1 978). Clyne ( 1 975) discusses communication breakdown (where an intention is misunderstood) and communication conflict (where a misunderstanding leads to friction between speakers) and suggests that both can often be attributed to cross-cultural (interlingual or dialectal), social (sociolectal) or individual (ideolectal) differences in communicative com­ petence rules, e.g., different rules for the realization of particular speech acts . This suggests that a fruitful area of research in second language acquisition is the contrastive analysis of norms for the realization of speech events and speech acts in different speech communities, which could usefully complement contrastive analysis , error analysis, performance analysis and related ap­ proaches.

142

S P E E C H A C T S A N D S E C O N D L A N G U AGE L E A R N I N G

3 . 1 First Language Learning Halliday ( 1 975), Dore (1975, 1977) and Bruner ( 1 975 , 1 978) have examined the development of speech acts in young children (before one year of age) and concluded that knowledge of communicative function precedes true language. Dore ( 1 975) in particular argues that illocutionary force is a language uni­ versal, that the speech act is the basic unit of linguistic communication, and that early language development consists of the child's pragmatic intentions gradually becoming grammaticalized. Bruner ( 1 978) has characterized the empiricist associationist view of language learning as 'impossible' and the nativist view as 'miraculous' and suggests that a speech act viewpoint is more explanatory than either. Bruner argues that mother tongue acquisition is a problem solving transactional enterprise, involving an active language learner and an equally active language teacher. Bruner stresses the importance o f mother-child interaction and finds this related t o the progression i n the kinds of requests made by children. First requests are directed at nearby objects, usually held by the mother, and the mother's main job seems to be to establish the sincerity of the request. A second type of request is related to shared ac­ tivity in games, e.g., 'Mummy read ' , in the context of reading together. These requests-and the mother's responses-are tied to the development of turn­ taking, the assignment of roles, and agency. The last type of request to develop, emerging at 1 5-16 months with Bruner's children, is for supportive action , such as persuading the mother to get a toy telephone from the cup­ board so that the child may play with it. While in both the earliest and the latest requests in the sequence what is desired may be an object, the later request forms are more sophisticated because they involve a goal and a means o f getting to it . A similar distinction has been drawn by Halliday between the instrumental ('I want') and the regulatory ('Do as I tell you') functions of early language. Clark and Clark ( 1 977) report that at the two-word stage children use mainly two types of speech acts , assertions (representatives) and requests (directives) . They d o not promise things o r use declarations; these are not added t o their repertoire of speech acts until they are considerably older, but they elaborate the kinds of directives and representatives they make as soon as they begin to produce longer utterances.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

granted an d the syntactic paradigm has been dominant. While phonology and other areas have not been ignored, second language learning has largely been described as a continuum of gradually complexifying syntactic systems. The bulk of the empirical research of recent years has been on such issues as morpheme development, error analysis, developmental study of L2 syntax, and these have been related to the concept of proficiency in a second or foreign language. Speech act theory on the other hand, defining proficiency with reference to communicative rather than linguistic competence, looks beyond · the level of the sentence to the question of what sentences do and how they do it when language is used. It thus broadens the scope of enquiry to include the study of how second language learners use sentences to perform speech acts and to participate in speech events. In first language acquisition, the acquisition of speech act routines has recently been considered of primary importance.

R I C H A R D W. S C H M I D T A N D J A C K C. R I C H A R DS

143

Reviewing studies by Halliday, Bates ( 1 976), Dore (1 975) and Garvey ( 1 975) on early request forms, Ervin-Tripp concludes : From a very early age (children) have a rich system of alternations in form that is systematically related to social features . They sensitively identify social contrasts signalled by tag modals, polite forms, address terms, modal embeddings. What they gradually learn to do is conceal their purposes. Wnile they use diverse syntactic forms, they still refer explicitly to their desires and goals, when they are not obvious from the context. So the major differences between adults and young children is not diversity of structure, not diversity of social features­ though the rules may increase in number of variables and in complexity with age-but systematic, regular, unmarked requests, which do not refer to what the speaker wants. Wide use of tactful deviousness is a late accomplishment.

3.2. 1 Second Language Learning. In reviewing research on second and foreign language learning, Swain ( 1 977) proposes a four part model of second language learning, isolating four areas of relevant research: I. Input factors refers to input to the learning process or situation, and includes

both linguistic and extralinguistic variabfes. 2. Learner factors refers to the contribution of learner variables (age, attitude, motivation, etc.) to the learning process. 3. Learning factors refers to strategies and processes used by the learner to learn elements of the target language-generalization, imitation, transfer, analogy, inference etc. 4. Learned factors refers to the particular feature of the target language being acquired by the learner (question forms, auxiliaries, negatives, phonology, etc.).

We will consider speech act theory with reference to two of the factors discussed by Swain, namely input factors and learning factors, and discuss how speech act theory contributes to our understanding of the nature. of the input to the learning process and to the strategies used by the learner in learning or using a second langu�ge. 3 .2.2 Input. A theory of second language acquisition must take account of the input to the learning process. The study of speech events and speech acts allows for focus on the typical speech settings encountered by second language learners and the identification of discourse structure and norms for the speech events encountered . This includes opening and closing sequences, turn taking

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Mitchell-Kernan and Kernan ( 1 977) have looked at the choice of directive type among older children (7- 1 2 years) and found that both requests and refusals are in some cases peculiar to children 's culture in the way they are elaborated. Children so often use directives to define and test status relation­ ships and obligations that they react testily to directive forms which, on the surface at least, seem perfectly appropriate. While requests that have little cost are usually honoured by adults, Mitchell-Kernan and Kernan found that their children did not honour them and frequently insisted on courtesy phrases (e.g . , pretty please), and even if these were used did not always comply. The frequent use of challenges ('Who do you think you are? ') indicates that the children are constantly on guard to preserve their rights and to defend them­ selves against challenges to their status.

1 44

S P E E C H A C T S A N D S E C O N D L A N G U AG E L E A R N I N G

T: P: T: T:

Initiation: What did we call this picture? Response: Piece of paper. Follow-up: A piece of paper. Yes. Initiation: What did we call this?

The speech function of orders is likewise frequent in classroom language, with a wide range of linguistic realizations. Learning within a classroom context must therefore be understood in relation to the highly structured and selective type of language which typifies classroom language and teaching situations. Second language learners may encounter other situations as input to the learning process which show particular discourse structuring. Candlin et a/. 1 976 have studied speech events within the context of doctor-patient com­ munications, with a view to identifying the structure of relevant speech events, to clarify the difficulties encountered by foreign doctors working in British hospitals. Such learners have to acquire rules for the speech event of the consultation. Speech acts identified as typical within the consultation speech event include greet, elicit, interrogate, question, make sure, extend, action­ inform, diagnose inform, progress inform etc. 'Investigation of a wide range o f consultations revealed that casualty consultation discourse is highly structured, in that there are significant probabilities to the occurrence of the above functions and to their distribution' . (Candlin, 1978, p. 1 5 . ) 3 .2.3 Learning Factors. Under this heading Swain lists a number of second language learning strategies and processes. The following seem to apply to research into the acquisition of speech act rules in a second or foreign language: (a) Inference, (b) Transfer, (c) Generalization, (d) Transfer of training. The category communication strategies identified in the L2 literature with reference to the acquisition of syntax, would appear to be redundant with reference to speech acts, since all the examples discussed here can be regarded as instances o f communication strategies .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

rules, sequencing rules, presupposition, role marking, as well as speech acts (Coulthard, 1 977). In the study of language input to second language learning the structure of speech events within the language teaching classroom is particularly important. The structure of classroom language can be defined with reference to its discourse characteristics (Holmes, 1978) . Turn taking is controlled by the teacher in typical classroom settings , and the amount of talking is likewise weighted in the teacher's favour. Classroom talk is largely teacher talk. Delamont, quoted by Holmes, notes that of teacher talk, 500Jo is made up of the speech events of teaching or lecturing and the other 50% in­ cludes 'explicit disciplinary and management moves and . . . reactions to pupil's contributions' . The speech function of questioning is frequent in class­ rooms, but it is typically a closed question from the teacher where only one acceptable answer is required, and not an open question where several dif­ ferent answers are possible . Coulthard ( 1 977, p. 81) reports the observation that teachers typically ask questions not to find out answers, but to find out if pupils know the answers, and thus once a pupil has produced the answer he needs to know whether it was the right one. The follow-up move, referring back and commenting on the answer, allows for the need .

R I C H A R D W . S C H M I D T A N D J A C K C. R I C H A R DS

145

·

Sales Clerk I : Sales Clerk 2: Customer:

But Korean Airline won't endorse the ticket, I don't think. (Looking directly at customer) You can call them and ask. OK . . . would you do that please? Would you phone them and ask?

Sales Clerk 2 meant her remarks as a suggestion to the customer that she phone . The customer either thought that the suggestion was directed at Sales Clerk 1 , or misinterpreted the utterance as an offer to make the call or as a general statement of possibility (i.e., as meaning 'One could call . . ') or chose to interpret the utterance in one of these ways. The casual observer cannot tell in this case. On questioning the clerk as to how she would analyse the exchange she later said 'She's not as dumb as she pretends' . Non-fluent language users would thus appear to be more dependent on .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

(a) Inference. lnferencing is defined as the process by which the learner derives a hypothesis or conclusion about language based on the evidence presented to him. It is the means by which the learner forms hypotheses about the target language. Candlin refers to ' interpretive strategies' which enable the speaker/hearer to retrieve discourse value from speech situations to arrive at an interpretation whereby the hearer's (reader's) interpretations match those of the speaker (or writer) . Candlin emphasizes that discourse value is not a constant but varies according to the type of discourse, the relations between the participants, and the influence of the setting and the topic. Thus for example the sentence Is the cook new? said in the kitchen of a restaurant by a waiter on noticing an unfamiliar face in the kitchen, may be interpreted as a Yes/No question asking for information. The same question, said by a client in the restaurant to a waiter on receiving a poorly prepared meal, would have the illocutionary force of a complaint. The nature of inferencing or interpretive strategies in speech act theory remains problematic even for native speakers. But instances of communication breakdown and misunderstanding among non-fluent language users suggest they frequently operate primarily at the surface structure level, identifying propositional content where it is marked directly by lexis or grammar, but often missing indirectly marked speech acts and functions. Thus will might be understood as a marker of future tense, for example, and modal overtones missed. In one case, a Japanese woman who had lived in the United States for about a year did not respond to indirect request forms such as 'can you' and 'will you' , but only to the explicit request marker 'please' . She later recognized the directive intent of such indirect forms, but still misinterpreted them, thinking that such forms could be interpreted (and used) only in sales-clerk/customer or other service interactions (Honda, 1 977) . Austin ( 1 962) refers to uptake, i .e. , the interpretation of the illocutionary force of a sentence by the hearer, which may differ from the intended uptake of the speaker. The following exchange between a customer in an airline office (a Korean woman) and two sales clerks illustrates the contrast between in­ tended uptake and uptake, and also demonstrates the practical difficulties of determining inferencing strategies. The customer was trying to change flights from one airline to another. Business was slack and a second sales clerk was occasionally joining in the transactional discussion.

1 46

S P E E C H ACTS A N D SECO N D L A N G UAGE L E A R N I N G

Thus the Anang value speech highly and the young are trained in the arts of speech, while for the Wolof, speech , especially in quantity, is dangerous and demeaning. French children are encouraged to be silent when visitors are present at dinner; Russian children are encouraged to talk. Among the Arucanian there are different expectations of men and women, men being encouraged to talk on all occasions, women to be silent-a new wife is not permitted to speak for several months {Coulthard, 1 977, p. 49).

Particular speech .events such as telephone conversations have also been compared from a cross cultural perspective, showing how transference of rules and expectations from one language to another may create confusion or misunderstanding. In Japanese, callers rather than answerers generally speak first on the telephone. In France, the fact that telephone calls are generally regarded as impositions on answerers may account for the fact that there are restrictions on caller behaviour which do not hold in English speaking countries (Godard, 1 977). In Egypt, there is an expectation that many calls will result in wrong numbers and callers frequently demand to know the identity of answerers; this seems rude to foreigners resident in Egypt, who often conclude that there are no rules at all for 'polite' telephone behaviour in the country (Schmidt, 1 975). Clyne 1 975, in a study of immigrants in Australia, discussed 'pragmatic transfer' , based on transfer of speech act rules from one language to another, which ,can lead to communication breakdown or communication conflict. Transfer may operate with respect to a number of dimensions . (i) Difference in opening or closing formulae for speech events. Speech events in a given language may have differing opening or closing formulae, which when transferred to the target language lead to incongruence. For example

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

contextual o r linguistic clues in inferencing. This in tum shapes the discourse directed to them by native speakers. Foreigner talk would appear to contain more explicit performatives than speech directed to fluent language users. Thus a teacher's opening to a joke addressed to a class of L2 learners began ' Let me tell you a joke . . . I'm going to tell you a joke . . . OK' . Such direct marking of the illocutionary value of the speech event would not be necessary with fluent language users who would be expected to infer the intended uptake from perhaps 'Did you hear the one about . . . ?' Candlin notes that for foreign university students to derive the intended uptake from university lectures they need to be aware of the careful and close integration of the visual , paralinguistic element with the spoken word, if they were going to understand the constant interplay in lectures between 'the main and the subsidiary planes of discourse-the essential argument and the audience-directed subsidiary comment ' (Candlin, 1 978:22). (b) Transfer. While the concept of transfer or inference has often been applied to the explanation of L2 performance at the phonological and syn­ tactic.level, little attention has been given to the effect of transfer operating at the level of discourse rules, and to its effects on speech event and speech act realizations in second language performance. There is evidence however to suggest that rules governing speech events may differ substantially from one language group to another, thus leading to different rules and norms for tum taking, amount of talking, speech act realizations etc .

R I C H A R D W. S C H M I DT A N D J A C K C. R I C H A R D S

147

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

with regard to meal talk, French and Malay begin with 'bon appetit' , and 'selamat makan' , respectively, which when transferred to English as good eating or good appetite appear unusual. When languages have similar formulae, ritualistic or markedness con­ siderations may be at variance. Greetings in many (perhaps all) speech com­ munities may include questions about the addressee's health, e.g . ' How are you?' In English, Hindi, Spanish, French and many other languages, such questions are largely ritualistic and need not be answered sincerely. In English , ' How are you?' is often not answered at all . In Arabic, on the other hand, the question must be answered and in almost all contexts the only appropriate answer is the ritual response formula 'ilhamdulillah' ('praise to God'). In Thai, however, 'sabaaj dii ryy?' ('How are you?') is a non-ritualistic, marked greeting, generally used only if one person has not seen the other for a long time and/or is sincerely concerned about his or her health . The unmarked greeting form in Thai is 'Paj naj ? ' ('Where are you going'). Transfer of un­ marked formulas could well lead to English speakers judging Thais to be far too curious about the other's whereabouts, while Thais may wonder why English speakers are so concerned about health problems (J. Fieg, personal communication) . (ii) Formulae used to realize a speech act have different meanings in two languages. A common transferable· formula may exist, but with quite different uptake in the native compared with the target language. An offer of a cigarette for example is declined in German or Indonesian with the equivalent of thank you, but accepted with thank you in English . Indonesians frequently cause confusion by declining offers with thank you. Their interlocuters, if native speakers of English, have been heard to respond with 'Do you mean thank you or no thank you?' Likewise a native speaker of English who responds to an offer of something when speaking Indonesian, with the Indonesian equivalent of thank you, may be taken as having declined. Silence is particularly ambiguous and difficult to interpret cross-culturally. Silence after a request may be taken as either assent or refusal in a great many cultures, but the non-native speaker will have great difficulty deciding which meaning is meant in unfamiliar contexts. Formulae which are realizations of the same communicative or politeness strategy but which are only parallel and not identical in form and use may cause particular difficulty. A general strategy of negative politeness is to at­ tempt to minimize the imposition on the hearer. In English, this can be done by using such expressions as 'just' or 'a little' (e.g. , 'I just want to ask you a little favour') or euphemisms such as 'borrow' for 'take' ('Can I borrow a cigarette? ') or 'a second' or 'a minute' for 'a few minutes' (e.g., 'I 'll be with you in just a second'). Exactly the same strategy and similar (but not identical) linguistic realizations are involved in the Arab's or Persian 's or Indian's or Mexican's use of such sentences as 'This will be ready tomorrow', meaning 'in a few days' (Brown and Levinson, 1 978). However, the native speaker of English generally will take 'tomorrow' only in its literal sense, will be angry when the goods are not provided on time, and will be tempted to make extreme generalizations about the character and sense of time of the people in the new culture. (iii) Different social conventions associated with realizations of speech acts.

1 48

S P E E C H ACTS AND S E C O N D L A N G UAGE L E A R N I NG

Two men are speaking. One wants something from the other (a loan, a service, his company in going somewhere) and both know it. The petitioner does not want to put his petition directly for fear of angering the petitioned; and the petitioned does not want to state his refusal directly for fear of frustrating the petitioner too severely . Both are very concerned with the other's emotional reactions because ultimately they will effect their own. As a result they go through a long series of formal speech patterns, courtesy forms, complex in­ directions, and mutual protestations of purity of motive, arriving only slowly at the point of the conversation so that no one is taken by surprise.

Clyne discusses culturally specific routines for the realization of such speech acts such as persuading and apologizing. 'Persuasion may be done through speech acts like the promise of a bribe, a threat of complaint to a higher of­ ficial, flattery or self-eulogy, or by overstating the case' (Clyne, 1 975:4). Transference of routines from one culture to the other may lead to the inter­ pretation that the speaker is aggressive, impolite, uncouth etc. (c) Generalization. This term includes 'regularization' , 'overgener­ alization' , analogy, and related concepts referred to in the literature. In the second language learning literature •it refers to the extension of something known in the L2 to a new context. With reference to speech act rules we will apply the term to the extension of speech act and speech event rules to inap­ propriate contexts. (i) Opening or closing sentences for speech events. Consider the following exchange, made by a non-native speaker to an office colleague on en­ countering him in the corridor. Non-native speaker: Native speaker:

How do you do? Oh hi.

Here the phrase How do you do has been extended beyond its boundaries in English-a greeting said on a first encounter in a formal-semiformal situation-to become a generalized greeting said on encountering friends. The appropriate greeting is of course How are you or some such phrase .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Here a number of different dimensions may be subject to transfer. We need to consider at least the following: Appropriateness of topic. Here we are concerned with what for example can one request in one language compared with another. Which requests can safely be declined? What can be denied or disagreed with and how safely can one transfer such choices across languages? What topics can one ask about on a first encounter with a stranger ( 1 ) of equal status (2) of higher status (3) of lower status (4) of same sex (5) of different sex etc.? Thus common questions from Asians on first encounters are Are you married? Ho w old are you? What is your salary? The Arabic question which most annoys non-Arabs is 'How much did it cost?' Such questions violate culturally specific speech act con­ ventions in English. Degrees of directness of realization of a speech act. A particular speech act such as refusal may be expressed differently in two languages. Geertz, 1 960 for example, discusses how refusal is communicated indirectly in Javanese. He describes a typical situation that his language teachers would use in the model conversations they used to teach him Javanese .

R I C H A R D W . S C H M I DT A N D J A C K C. R I C H A R D S

1st encounter How do you do.

149

Subsequent encounter How are you? Hi etc.

Leave-taking formulae may also be generalized to speech events where they are not appropriate. The following exchange is between an office boy delivering a consignment of books to an office. Office boy: Addressee: Office boy:

Where shall I put these books please? Put them on the table. (some minutes later). I 'II be making a move now.

Do you have a car? Yes, I do.

A request, however, cannot be answered in the same way. Can you pass.me the milk? Yes, I can.

Barkin and Reinhart ( 1 978, p. 58) discuss second language learners' dif­ ficulties with the phrases excuse me and I'm sorry. A typical mistake is to use these for inappropriate speech acts, as in the following example where the non­ native speaker declines an invitation to the movies. Excuse me. I'd like to go but I don't have time.

Homer, a five year old Iranian child, generalized the English formula 'What's ' this?' to numerous contexts beyond simple NP identification (Wagner-Gough, 1 975). 1.

Identification a. What this IS Elmer. ( This is Elmer). b. What this is? ( What is this?) 2. Advice or help a. What this is? ( What should I do now?) 3. a. What is it tunnel. ( Stop pushing sand in my tunnel .) b. What this is Homer. ( I'm Homer and you can't tell me what to do.) c. What is this this it. ( Give me that truck.) =

=

=

=

=

=

Of course, 'what's this?' is multifunctional in adult native English (compare 'what's this' said scornfully, curiously, hintingly, etc.) so that some of Homer's generalizations may be functionally appropriate while others are not; it is difficult to evaluate functional appropriateness in this case when gram­ matical relationships remain unclear. (d) Transfer of training. This category refers to features of the learners interlanguage which are traceable to teaching procedures used or to the particular textbook or teaching materials from which the learner has studied a language. Here are two examples of what was interpreted as inappropriate

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

(ii) Speech act routine generalized to inappropriate context. Some errors that on first sight would be attributed to stylistic inappropriateness or mistakes of lexis may turn out to be instances of a routine, appropriate to a particular speech act, generalized to a different type of speech act where it is no longer appropriate. A Yes/No question, for example, which functions as a request for information, can be answered with Yes/No plus verb repetition.

I SO

S P E E C H ACTS A N D SECOND LANG UAGE L E A R N I NG

directives and which are probably traceable to transfer of training. They were noted by the spouse of a non-native speaker. Example 1 . Example 2.

Mother (a non-native speaker) to her son . So after supper you will do your homework. The wife to her husband. Tomorrow we will go to see the movie, al/right?

4.

I M P L I C AT I O N S F O R R E S E A R C H A N D T E A C H I N G

The review presented here raises a number of questions which require empirical investigation and further study before conclusive statements can be made. We have focussed primarily on proficiency in the realization and inter­ pretation of speech acts, rather than the acquisition of speech act rules. The acquisition of pragmatic or communicative competence is an emerging interest in language acquisition studies. Pragmatic rather than grammatical constraints are seen as crucial in accounting for both the structuring of child language utterances and interlanguages, in the work of Peters, Wagner Gough , Hatch and others. Peters (1 977) has distinguished two styles of first language acquisition : an analytic style, one word at a time, and a 'gestalt' style, an attempt to use whole utterances in socially appropriate situations. Many in- ' vestigators of second language learning (e.g., Wagner-Gough and Hatch, 1 975) have reported that second language learners are apt to use a gestalt style even more than first language learners, using prefabricated routines and

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Even given that husbands and wives who speak the same language are often at odds over the choice of directive forms used in the family, something more appears to be going on here. The first sentence, addressed to the NNS's son, would be perfectly appropriate if homework were an issue in the family. How­ ever, it is not, and the NNS reports that she meant to suggest and did not intend to be or sound imperious. The native speaking spouse suggested that in both these examples can , or even better, c 'n, would have been a better choice of modal to convey the reported intentions of the speaker . But this NNS never uses can when reference is to future time, even though this is possible in the native language. She was taught that it is extremely important to indicate time reference in English , and she was taught (contrary to fact) that uncontracted forms are always more polite and proper than contracted forms. In general this speaker pays careful attention to literal meanings. The relationship be­ tween her forms and her social meanings could be defined in terms of con­ versational postulates, though the details of the rules for use of these · postulates would differ somewhat from those of a native speaker. Transfer of training may interact with the other learning factors, such as transfer and generalization, as well as attitudes towards languages, leading to inappropriate language. In Japanese, for example, a great deal rests on control of a highly complex system of honorifics. When the Japanese learns English, he finds nothing very similar, nothing that can be directly transferred. In addition, he generally believes and is probably taught (in accordance with the prevailing stereotype) that while Japanese is a very 'polite' language, English is 'logical', 'direct' , and not very polite. The Japanese learner of English may therefore be insensitive to the nuances of English politeness, which are not concentrated in one sub-system of the language.

R I C H A R D W. S C H M I DT A N D J A C K C. R I C H A R D S

lSI

patterns (which may include speech act formulas) i n an attempt t o com­ municate in a socially appropriate way beyond their linguistic competence. One issue of current concern is the degree to which such formulaic language is crucial to the overall development of language. Fillmore ( 1 976) argues that the use of such formulaic speech, motivated by the learner's need to establish social contact, gradually evolves into creative language. Krashen and Scarcella ( 1 978), on the other hand, maintain that routines and patterns play only a minor role in second language acquisition, with the creative construction process evolving in an essentially independent manner. Study of the acquisition of speech acts by non-native speakers should enable us to clarify of these and other issues. Possible issues for further research are the following:

·

The relevance of speech act theory and research to language teaching is through its contribution to the theory of communicative language teaching. Writers on communicative syllabus design such as Munby and Wilkins, make use of speech act and speech event theory in their accounts of notional and communicative syllabuses for language teaching, as have various other writers on communicative teaching. (Allen, 1 977: Stratton, 1977; Paulston 1974; Holmes and Brown, 1977; Widdowson, 1 978). The central issue is to what degree successful second language learning can be identified with acquiring rules for speech act realization and interpretation. An emphasis on strategies for speech act realization as a central goal for intermediate and advanced language teaching, would lead to a focus on learning as a process rather than on what is learned as a product. Another basic issue concerns translating the concepts of speech act theory and discourse analysis into units which can be realized within a language teaching programme, i.e., which can be operationalized for teaching purposes. Candlin (1978) warns: 'We know

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

1 . Descriptive studies of the types of speech acts encountered in specific settings for second language use and learning, according to such factors as age of speakers (e.g., adult-child; child-child) roles: (e.g., teacher-student; friend-friend; parent-child)-in particular settings, (classroom; work domain) and for specific speech events (e.g., interviews; conversations). 2. Studies relating stages of grammatical development to speech act realizations in inter languages of different types of learners. 3. Studies of acquisition of rules for the realization and interpretation of speech acts over time among interlanguage users. 4. Cross linguistic comparisons aimed at determining whether different languages make use of the same classes of speech acts and similar strategies for realizing and interpreting speech acts. 5 . Studies of the effects of speech act realization on the discourse patterning and conversational structure of non-native language users in different types of discourse. 6. Study of pragmatic errors in non-native discourse, e.g., the failure to code or inter­ pret speech acts appropriately or to recognize or assign appropriate illocutionary force to utterances of native speakers. 7. Studies of the attitudes of native speakers to violation of native speaker rules for speech act realization, and the contribution of such violation to communication conflict or breakdown. 8. Studies relating strategies for the performance and acquisition of speech act rules to language learning processes in general, e.g., to what degree do such factors as transfer, inferencing, and overgeneralization also apply to pragmatic dimensions of language learning?

1 52

SPEECH ACTS AND S E C O N D LANGUAGE L E A R N I N G

1 . To what degree should realization and interpretation strategies for speech acts be taught explicitly in a language teaching programme? 2. How useful are contrastive statements of such coding procedures for learners whose mother tongues adopt different coding strategies? 3 . What speech acts are basic and can speech acts and discourse rules be ordered for the purpose of teaching? 4. Are techniques for the teaching of other areas of the target language (dialogues, drills etc.) also appropriate for teaching discourse rules, or are speech act rules acquired as a biproduct of communication? 5 . How do we choose for the purpose of teaching, the forms for the realization of speech acts? 6. To what degree should the emphasis on ability to perform and interpret speech acts take priority over the ability to code sentences grammatically within a teaching programme?

(Received October 1979)

REFERENCES

Abrahams, R . , 1 962. 'Playing the Dozens' . Journal of American Folklore 7 5 , 209-220.

Albert, E . , 1 964. ' ' 'Rhetoric" , ''logic' ' , and ' 'poetics' ' in Burundi : cultural patterning of speech behavior' . American Anthropologist 66.6. Revised and reprinted in Gumperz and Hymes, 1 972. Allen, D. and Guy R . , 1 974. Conversational Analysis: the sociology of talk. The Hague: Mouton. Allen, Patrick. , 1 977. 'Structural and Functional Models in Language Teaching'. TESL TALK, 8, 1 , 5 - 1 5 .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

enough, however, t o realize methodologically that w e must avoid latter day 'structuralism' of concepts and site utterances firmly within connected dis­ course. Furthermore ways of teaching should shift from teacher-telling to learner interpreting within a syllabus whose prime goal is the development o f strategies for discourse processing, rather than as an assembly of items' . Candlin's work on speech events within doctor-patient communications begins from 'detailed functional description of native speakers' interaction, and attempts not simply to teach single functions but to show doctors how to open and close interviews, how to participate in other types of exchange, how to build exchanges into longer sequences, how to manipulate the turn-taking system' . (Coulthard 1 977, p. 146). The progression from grammatical to communicative competence within a formal language teaching programme is thus a movement towards the organization of learning and teaching in terms o f creating contexts for the realization and interpretation of speech acts within a framework of discourse rules. Questions which require further consideration from this perspective include:

R I C H A R D W. S C H M I DT A N D J A C K C. R I C H A R D S

153

Apte, M., 1974. "'Thank you" and south Asian languages: a comparative sociolinguistic study' . In Cooper, R . , ed . , Language Attitudes I, International Journal of the Sociology of Language 3, 67-69. Austin, J ., 1 962. How to Do Things With Words. Oxford: Oxford University Press. Bates, E . , 1 976. Language and Context: the acquisition of pragmatics. New York: Academic Press. Bolinger, D., 1 967. 'The imperative in English' . In To Honor Roman Jakobson: essays on the occasion of his 70th birthday. The Hague: Mouton, 335-62. Borkin Ann and Reinhart, Susan M . , 1 978. 'Excuse Me and I'm Sorry' . TESOL Quarterly 12, I , 57-69.

Bruner, J . 1 -20.

1975 . 'The ontogenesis of speech acts ' . Journal of Child Language 2 . 1 ,

Bruner, J ., 1978. ' Learning the mother tongue' . Human Nature 1 .9 , 42-49. Candlin, C . , 1 978. ' Discoursal Patterning and the Equalizing of Interpretive Opportunity'. Paper presented at the Conference on English as an International Auxiliary Language. East West Center. Honolulu. April l 978. Candlin, C., Leather, Jonathan H. and Bruton, Clive J., 1976. ' Doctors in Casualty : Applying communicative competence to components of specialist course design' . IRAL, Vol . 14/3, 245-272. Clark, H . and Clark E . , 1 977. Psychology and Language: an introduction to psycholinguistics. New York: Harcourt Brace Jovanovich. Clark, H. and Lucy, P . , 1 975. ' Understanding what is meant from what is said; a study in conversationally conveyed requests' . Journal of Verbal Learning and Verbal Behavior 14, 56-72. Clyne, Michael . , 1 97 5 . ' Intercultural Communication Breakdown and munication Conflict: Towards a linguistic model and its exemplification' .

Com­

Cole, P . , 1 975. 'The synchronic and diachronic status o f conversational implicature'. In Cole and Morgan, 1 975. Cole, P. and Morgan, J . (eds .), 1 975 . Syntax and Semantics, Volume 3: Speech Acts. New York: Academic Press. Coulthard, R. M . , 1 97 5 . 'Discourse Analysis in English-A Short Review of the Literature'. Language Teaching and Linguistics Abstracts, Vol. 8, 2, April 1 975, 73-89. Coulthard, R. M . , 1 977. An Introduction to Discourse Analysis. London: Longman .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Brown, P. and Levinson, S., 1 978. ' Universals in language usage: politeness phenomena ' . In Goody, Esther N . , ed. Questions and Politeness: strategies in social interaction. (Cambridge Papers in Social Anthropology). Cambridge: Cambridge University Press, 56-3 10.

1 54

S P E E C H ACTS AND SECOND L A N G UAGE L E A R N I N G

Dore, J . , 1 97 5 . ' Holophrases, speech acts, and language universals' . Journal of Child

Language 2 , 2 1 -44. Dore, J., 1 977. ' "Oh them Sheriff" : a pragmatic analysis of children 's response to questions' . In Ervin-Tripp, S. and Mitchell-Kernan, C . 1 977. Dundes, A., Leach, J . and Ozkok, B . , 1972. 'The strategy of Turkish boys' verbal duelling rhymes ' . In Gumperz and Hymes, 1 972. Ervin-Tripp, S . , 1 977. 'Wait for me, Roller Skate' . In Ervin-Tripp, S . and Mitchell­ Kernan, C . , 1 977. Ervin-Tripp, S. and Mitchell-Kernan, C. (eds .), 1 977. Child Discourse. New York: Academic Press.

Fillmore, L. W., 1 976. 'Cognitive and social strategies in language acquisition' . Ph.D. Dissertation, Stanford University. Fraser, B . , 1 97 5 . 'Hedged performatives ' . In Cole and Morgan, 1 975. Fraser, B., 1 978. 'Acquiring social competence in a second language'. RELC Journal 9/2, 1 -26. Garfinkel, H . , 1 967. Studies in Ethnomethodology. Englewood Cliffs: Prentice-Hall. Garvey, C . , 1 975 . 'Requests and responses in children's speech ' . Journal of Child Language 2 , 4 1 -63. Geertz, Clifford, 1 960. The Religion of Java. New York: The Free Press. Godard , D . , 1 977. 'Same setting , different norms: phone call beginnings in France and the United States ' . Language in Society 6.2, 209-2 1 9 . Goffman , E . , 1 972. Relations m Public. New York: Harper and Row. Goffman, E . , 1 976. 'Replies and Responses ' . Language in Society 5 . 3 , 257- 3 1 4 . Gordon, D. and Lakoff, G . , 1 97 1 . 'Conversational postulates ' . Papers from the Seventh Regional Meeting of the Chicago Linguistic Society, 63-84. Green, G . , 1 97 5 . 'How to get people to do things with words' . In Cole and Morgan, 1 977. Grice,

H.,

1968.

'Utterer's

meaning,

sentence-meaning

and

word-meaning'.

Foundations ofLanguage 4, 225-242. Grice, H . , 1 975 . ' Logic and conversation ' . In Cole and Morgan , Speech acts, 4 1 -58 1 975. Gumperz,

1 . and Hymes, D. (eds .), 1 972. Directions in Sociolinguistics: the ethnography of communication. New York: Holt, Rinehart and Winston .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Ferguson, C . , 1 976. 'The structure and use of politeness formulas ' . Language. in Society 5 . 2 , 1 37- 1 52.

155

R I C H A R D W. S C H M I D T A N D J A C K C: R I C H A R D S

Halliday, M . , 1 97 3 . Explorations in the Functions of Language. London: Edward Arnold. Halliday, M., 1975 . Learning how to Mean: exploratiOns in the develooment of language. London : Edward Arnold. Hancher, M . , 1 979. 'The classification of cooperative illocutionary acts ' . Language in

Society 8 . 1 , 1 - 14. Hatch ,

Evelyn,

1 978.

'Discourse analysis,

speech

acts , and second

language

acquisition ' . In Ritch1e, William C. (ed. ) Second Language AcquiSition Research: Issues and Implications. New York: Academic Press. Heringer , J ., 1 972. 'Some grammatical correlates of felicity conditions and presuppositions ' . Workmg Papers in Lmguistics (The Ohio State University) I I ,

Holmes, Janet and Brown, Dorothy F . , 1 977. ' Sociolingmstic Competence and Second Language Learning ' . Topics in Culture Learmng. August 1 977. Holmes, Janet, 1 978. 'Sociolinguistic Competence in the Classroom ' . I n Richards, Jack C. (ed .) Understandmg Second and Foreign Language Learning. Rowley, Mass . : Newbury House. Honda, Gary T . , 1 977. 'Directives ' . Unpublished paper, University of Hawaii. Hymes, D . , 1 967. 'Models of the interaction of language and social setting'. Journal of Social Issues 23.2, 8-28. Revised and reprinted in Gumperz and Hymes, 1 972. Hymes, D., 1 972. 'On Communicative Competence' . In Sociolinguistics, Pride, J . B. and Holmes, J . (eds.) 269-293 . Harmondsworth: Penguin. Krashen, S. and Scarcella, R . , 1 978. 'On routines and patterns in language acquisition and performance' . Language Learning 28.2, 283-299. Labov , W . , 1 972. ' Rules for ritual insult' . I n Sudnow, D. (ed.) Studies in Social Interaction. New York: The Free Press. Labov , W. and Waletsky, J . , 1967. ' Narrative analysis : oral versions of personal experience' . In Essays on the Verbal and Visual Arts (Proceedings of the 1 966 spring meeting, American Technological Society). Washington: University of Washington Press. Lyons , John, 1977. Semantics 2. Cambridge: Cambridge University Press. Marcus, L . , 1978. 'Strategies for requests and other directives in Cantonese' . Unpublished term paper, University o f Hawaii. Matthews, P . ,

1972. ' Review of Jacobs and Rosenbaum, Readings in English Transformational Grammar' . Journal ofLingutstics 8, 1 25- 1 3 7 .

Mitcheli-Kernan, C . and Kernan, K . , 1 977. ' Pragmatics o f directive choice among children ' . In Ervin-Tripp, S. and Mitcheii-Kernan, C. Child Discourse, 1 89-208.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

1 - 1 10.

1 56

Munby,

S PE E C H ACTS AND SECO N D L A N GUAGE L E A R N I N G

John ,

1 978.

Communicative Syllabus Design. Cambridge: Cambridge

University Press. Nunes, S . , 1 976. ' Bar tal k : a study of order-giving at a Waikiki drink call station ' . Unpublished term paper, University of Hawaii. Ochs-Keenan, E . , 1 976. 'The universality of conversational postulates ' . Language in Society 5 . 1 , 67-80. Paulston,

Christina Bratt . ,

y

1974.

' Linguistic and Communicative Competence' .

TESOL Quarterl 8.4, 347-362. Peters, A . , 1 977. ' Language learning strategies: Does the whole equal the sum of the parts ' . Language 5 3 , 560-573.

Ross, J . , 1 970. 'On declarative sentences ' . In Jacobs, R. and Rosenbaum, P. (eds . ) Readings in English Transformational Grammar. Waltham , Mass . : Blaisdell. Sacks, H . , 1 972. 'On the analyzability of stories by children' . In Gumperz, J. and Hymes, D . , 1 972. Sacks, H . , Scheg1off, E . and Jefferson, G . , 1 974. 'A simplest systematics for the organization of turn-taking for conversation' . Language 50.4, 696-725 . Sadock , J . , 1 970. 'Whimperatives ' . In Sadock, J. and Vanek, A. (eds.) Studies Presented to Robert K. Lees by his students, Linguistic Research, Inc . , 223-238. Sadock , J . , 1 972. ' Speech act idioms' . Papers from the Eighth Regional Meeting of the Chicago Linguistic Society, 329-339. Sadock , J ., 1 97 5 . 'The soft, interpretive underbelly of generative semantics' . In Cole and Morgan, 1 977. Sanches , M . and Blount, B . , 1 97 5 . Sociocultural Dimensions of Language Use. New York: Academic Press . Schegloff, E . , 1 968. 'Sequencing in conversational openings ' . American Anthro­ pologist 70, 1 075-95. Schegloff, E . and Sacks, H . , 1 973. 'Opening up closings ' . Semiotics 8 . 4, 289-327. Schenkein, J . , 1 978. Studies in the Organization of Conversational Interaction. New York: Academic Press. Searle, J . , 1 965. ' What is a speech act?' In Black, M. (ed .) Philosophy in A merica, Allen and Unwin and Cornell University Press, 2 2 1 -239. Searle, J . , 1 969. Speech Acts. Cambridge: Cambridge University Press. Searle, J . , 1 97 5 . 'Indirect speech acts ' . In Cole and Morgan, 1 977.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Pratt, M . , 1 977. Toward a Speech Act Theory of Literary Discourse. Bloomington: Indiana University Press.

R I C H A R D W . S C H M I D T A N D J A C K C. R I C H A R D S

157

Searle, J . , 1 976. 'The classification of illocutionary acts ' . Language in Society 5 . 1 , 1 -24. Sinclair, J . and Coulthard, R . , 1 975. Towards an Analysis of Discourse. London: Oxford University Press. Stratton, F. S . , 1 977. 'Putting the Communicative Syllabus in 1ts Place'. TESOL Quarterly 1 1 , 2 , 1 3 1 - 1 4 1 . Sudno� . D . , 1 972. Studies in Soc1al lnteraction. New York: Free Press. Swain, Merrill, 1977. 'Future Directions in Second Language Research ' . In Carol Henning, (ed .) Proceedings of the Los Angeles Second Language Research Forum. UCLA: Los Angeles.

Ueda, K.. , 1 974. 'Sixteen ways to avoid saying "no" in Japan ' . In Condon, J. and Saito, M . Intercultural Encounters with Japan: communication-contact and

conflict. Tokyo: The Simul Press, 1 85-92. Urmston Philips, L . , 1 976. 'Some sources of cultural variability in the regulation of tal k ' . Language in Society 5 . 1 , 8-95. Wagner-Gough, J . and Hatch, E., 1 975. 'The importance of input data in second language acquisition studies ' . Language Learning 25 .2, 297-308 . Widdowson, H. G . , 1 978. Teaching Language University Press.

as

Communication. London: Oxford

Wilkins, D. A . , 1 976. Notional Syllabuses. London: Oxford University Press.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Tarone, Elaine E . , 1 977. 'Conscious Communication Strategies in Interlanguage: A Progress Report' . In On TESOL 77. Brown, H. Douglas , Yorio, Carlos and Crymes, Ruth (eds .) Washington: TESOL.

D IS CUS S I ON The BAAL Annual Meeting, at Manchester Polytechnic in September 1 979, addressed itself to the theme 'Applied Linguistics, or Linguistics Applied?' Two papers from the meeting are included below in the hope that they might provoke further discussion .

C. J. B R U M F I T Umversuy ofLondon

In this paper I hope to consider a problem which faces any applied activity­ that of relating insights drawn from a variety of different disciplines to the solution of specific practical difficulties. In doing this, I shall be forced to attempt a definition of applied linguistics. This problem is particularly acute for applied linguistics, for language problems are by their nature complex and intimately bound up with human needs and behaviour. Any abstraction or idealisation of linguistic performance which may be methodologically necessary for the descriptive linguist (Lyons, 1 972) will be confused by social and psychological factors as soon as the connection is renewed, as it must be as soon as we consider classrooms, or literary traditions, or medical services. There have of course been attempts to incorporate linguistic procedures into teaching in fairly undiluted forms, but even when textbooks demonstrate a close relation between theoretical assumptions and classroom practice (e.g., in very di fferent ways, Kennedy 1 930 and Roberts, 1964), the intervention of the teacher will certainly reflect pedagogic as much as descriptive requirements. And, of course, educationists are always ready to complain that the concerns of theorists have too little to say for teachers (Rosen, 1978). The difficulty arises partly because language is simultaneously used to express what Halliday ( 1 975) has called 'mathetic' acts of meaning ('an act of meaning that is self-sufficient, calls for no response,. and functions as an explicit reality-creating device' Halliday, 1 979, p. 83), and pragmatic acts which are essentially interactional. There is thus a conceptual dimension to be explored as well as one incorporating strategies for social interaction. However, the complexity and availability of language poses its own problems, for the system can be isolated and idealised to give rise to complex and fascinating analysis and speculation-witness the whole history of linguistics itself. Language is thus a system which is operated and negotiated socially, a system which enables an individual to comprehend and cla,ssify through his Applied Lmguisucs, Vol. I , No. 2.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

BEING INTERDISCIPLINARY -SOME PROBLEMS FACING APPLIED LINGUISTICS

C. J. BRUMFIT

159

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

own experience, and a system which is capable of being abstracted, reified and examined in isolation from both individuals and society. The first two systems interact, to increase the complication, and furthermore carry tremendous extra weight, for the social-linguistic system operates, by definition, cross­ culturally, involving structures of meaning which have scarcely yet been ex­ plored in which we ourselves participate and which we cannot therefore ob­ jectify (see, for example, Bernstein 1 975), and the psycholinguistic system, both in learning and use, involves considerations of motivation and cognitive depth (see Stevick 1 976) as well as the whole weight of individual past ex­ perience. Yet these two systems are the material we actually work with, while the idealised one is no more than a methodological device-albeit a necessary one-to enable us to come to grips with the other two systems. Language is not of course the only system of interaction, nor the only symbolic system, but with no other systems is the power to generate complex and subtle symbols so closely related to the specifiable features of the in­ teractive process: our idealised linguistic system appears to give us a metalanguage to explore the means of communication and conceptualisation. Stated like this, it might appear that the suspicion of linguistics shown by psychologists and sociologists in the past had good reason. A major con­ tribution of linguistics, however, has been to demonstrate the complexity of th� language process. The linguistic system, even in an idealised and stan­ dardised form, defies simple explanation�, and tendencies to oversimplify result in apparently naive discussion of the role of language in society or in learning processes. The most widespread naivety in this direction centres on the relationship between language use and the concepts associated with particular examples of usage. Much discussion of Bernstein's early work ( 1 971), adequately criticised by Stubbs ( 1 976), Trudgill (1975) and Rogers (1976), assumed a close and clearly specifiable relationship between language and concept formation, and such a relationship is assumed by many-perhaps most-sociologists when writing about language. It is also implicit, of course, in the discussion of language associated with sexist and racist attitudes. The central difficulty is that meaning is both conventional and arbitrary. The descriptive terms of one decade may be the insults of another ('negro' , ' fascist'), but there is always an element o f human intention-'we are not the prisoners of our cultural semiotic' (Halliday, 1 975 : 1 40)-unless, that is, enough of us choose to be and thus change the convention. When someone says 'The sun rises in the east' , he does not demonstrate himself to believe in a pre-Copernican universe, and I have not demonstrated myself by my use of 'he' and 'himself in this sentence to be a believer in an all-male human race­ but the two cases are distinct nonetheless, for the sexist issue is live where the other is dead and we can only speak effectively the language of those we speak to. It may be true that sociologists have frequently written as if language determined reality, to the posthumous delight of Whorf, and it may also be true that psychologists have seen language as primarily about naming, at least until fairly recently. Such differences in attitude may have been partially responsible for the greater enthusiasm with which psychologists regarded the linguistic advances of the various chimpanzees which have been taught 'language' (Linden, 1 975). What about the linguists themselves, however? How has their view of language been partial? Certainly, over-emphasis on the

1 60

D ISCUSSION

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

centrality of syntax can lead t o a distorted view o f language work i n schools, particularly if such a view is coupled with a learning theory which ignores cognitive and contextual requirements (Stevick, 1976) . The greatest risk lies more in the confusion of procedures which are in essence investigatory and descriptive with those that are developmental and pedagogical. Attempts to develop a particular person's ability in a language demand quite separate procedures from attempts to develop the understanding of the human race about the nature of language, or indeed the nature of a particular language. The former is intri�sically a subjective operation, the latter an objective one, for the former can be judged by results without reference to the means-which may be hidden-and with the latter the means, the explicitness of the procedures used, is the process. The goal for the language learner is deter­ mined by the conditions for effective operation in the community of language users to which he aspires. The goal for the scholar cannot be this, since the 'truth' cannot be final. He can hope for no more than to present an approach so clearly that the criticisms made of that approach lead to further and better analyses, and improved explanatory models. Now the differences in attitude to language reflect primarily the differing interests of the practitioners of the various disciplines. This is obvious. But the person who applies insights to a particular problem has to operate in a world where the differing concerns of different disciplines become confusing. Language, as was pointed out above, operates simultaneously in several directions at once: the network is so intricate that no one variable can be held constant without many others being distorted. In cases of severe malfunction drastic actions (i.e . , those based on a grossly over-simplified view of the working of language) may be necessary. But most applications of linguistic insight& are concerned not with severe malfunction but with normal situations. This is true particularly of the application of linguistics to language teaching. Applications to language teaching of views of language restricted to any one discipline will be inappropriate in that they will limit language to too restricted a set of features. Language cannot be seen purely as a creator of ideology because we can break its meaning if we choose to; it cannot be seen solely as a device for establishing categories or merely as exemplification of particular syntactic patterns because in the former case language in use characteristically does other things as well or because in the latter case we are not primarily teaching the process of linguistic description. An over-emphasis on language for 'making sense of the world' may lead to too insensitive an emphasis on creative writing at the expense of working towards social norms which have to be recognised for effective communication to occur, or on the mathetic at the expense of the pragmatic. An over-emphasis on the pragmatic for foreign learners may lead to neglect of mathetic aspects essential for genuine in­ ternalisation. But we are remaining at the moment the prisoner of our own categorisations. The dimensions of language we want to explore can be readily fitted in to the patterns of the various intellectual disciplines, since we recognise the boundaries imposed by these disciplines. In what sense can problems of application be similarly fitted in to discipline boundaries? In practice, applied linguists expect their students to look at the linguistic en­ vironment variously from sociological, psychological, or pedagogic points of view and to marry together their conclusions as best they can. Readers on

C. J . BRUMFIT

161

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

applied linguistics and introductory books (Wilkins 1972a, Allen & Corder 1 973, Corder 1 973) no less than books on language for teachers (Wilkinson, 1 97 1 , 1975, Arthur 1 973) put together information drawn from research sources in various disciplines as if there is no problem of compatibility. And indeed, while talking loosely and suggestively about the solution of practical difficulties, there is very little problem. The major difficulty arises when we want to ask that applied linguistics shoul� justify itself as a term by showing itself to be something other than discussion of practical problems encountered in (mainly) language teaching. It has, indeed, been suggested recently by Bernard Spolsky that it would be more appropriate to use the term 'educational linguistics', by analogy with educational psychology or educational philosophy for the application of linguistic insights to teaching problems (Spolsky, 1 978) . And there is no doubt that a great deal remains to be learnt from linguists, in various forms, working in education. Part of this discussion appears to lead to an argument about terminology which it would be easy to trivialise. Put at its starkest, if applied linguistics were to be considered merely the application of linguistics to anything to which it could be applied, then it would be no more than a mirror for linguists to peer into-for the only issues which linguistics can confront alone are linguistic issues, not applied ones. If real problems are to be confronted, as was argued above, the issues will not be solely linguistic. Whether we call the intelligent solution of problems iri which language plays a major part, applied linguistics or not is unimportant, providing the use of the term does not obscure im­ portant truths about the activity. However, the argument about terminology may conceal a more fun­ damental difficulty, revealed not in the connection with linguistics, but in the formality of the term 'applied linguistics' . fit Corder asks pertinently, 'Can we say that any of the approaches to language as knowledge, as behaviour , as skill, as habit, as an event or an object can safely be disregarded by the language teacher?' (Corder 1 973: 2 1 ) . Yet by using the term 'applied linguistics' we imply that there is something, some procedure perhaps, some discipline, some unique body of knowledge not found in any of the feeder disciplines, which performs the act of integration between these approaches, something more than a context only. If we simply say, 'Let us talk about our problems, using whatever ideas and supporting information that comes to hand' , then there will be no implicit claims being made. But once we give the activity a name-and indeed once we set up organisations under the name, once we establish a profession-we imply something with status, something which can be described as uniting the various areas of human activity to which linguistics can contribute: in short we claim academic respectability and recognition. But, like the analogous discipline of education, applied linguistics has yet to demonstrate that it is anything other than an activity which draws people of similar interests together to discuss common problems as carefully as they can. Now it may be, of course, that the careful discussion of common problems is all we can realistically demand, and that to ask for more is to misunderstand the nature of our various fields of interest. If this is so, however, it would be worth asking exactly what speech therapists or stylisticians find they can learn from the discussions o f language teachers. It is perfectly possible-education is perhaps a case in point-for careful discussion of common problems to be

162

DISCUSSION

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

conducted as rigorously as possible, without implying that all questions are capable of the same degree of formalisation or rigour in their answering. Indeed, in educational circles a great deal of harm has ·been done by the en­ thusiasm of practitioners for inappropriate statistically-based experimental work, when discussion of a synthetic rather than analytic nature may have much greater value: there are academic dangers in formalism and practical risks in the adoption of inappropriate ritual. But it seems to me that applied linguistics is not in exactly the same position as education, that it is not simply an activity, but that it has within it the seeds of an integrated view of language applied to the world which should underly the work of all applied linguists. These seeds have not yet borne fruit; there are major areas of language ac­ tivity-mother tongue teaching is perhaps the most notable-in which the major preoccupations of applied linguists have been seen as either irrelevant or dangerous. If applied linguistics is to be perceived as a fundamentally im­ portant discipline, this will only be when it produces ideas of sufficient generality to affect any language-using situation-when it performs the task of integrating all the various attitudes to language of researchers in other fields, and produces an account of language in use which is both convincing and readily comprehensible. Where will such an account come from? There are, I think, some pointers. Over the last few years it has become increasingly apparent that there are intimate connections between psychological and sociological approaches to language. Studies of language acquisition have, as they have moved towards investigating the development of meaning, been forced to take a fuller account of situation (Halliday, 1975) and it is increasingly apparent that any account of the development of syntax (and even perhaps of phonology) cannot be isolated from the interaction of speaker and social environment. And this applies whether one is talking about language acquisition with mother tongue or second language. Furthermore, interesting similarities are being perceived between what used to be called language acquisition (mother tongue) and language learning (formally-based foreign or second language) (Corder, 1978) and these similarities do not only extend to the behaviour of the learners, or acquirers, themselves, but to the interactional behaviour of those who talk with them, whether teachers, parents, or peoP.le speaking to foreigners. More interesting still, these features of simplified language may be important in relation to the development, historically and geographically, of individual languages, in the relationships between languages, geographical dialects, pidgins, creoles and various kinds of functional dialect which have been discussed as registers of one sort or another (Schumann, 1 978). What is beginning to emerge, in other words, is an account of language which is not static or idealised (and therefore only indirectly applicable to the needs of workers of any kind with real problems), but an account which is dynamic, fluid, and increasingly motivated by reference to interaction, to active learning and using strategies associated with learners' responses to social demands. This is an account, in short, of users' application of language to the problems of the world and being in it. Such an integrated account is overdue, and in a number of spheres has been much in demand in reaction against too-simple transfers of the assumptions of traditional core linguistics to, for example, the classroom. The whole 'communicative' movement in language teaching has been a reaction, in a variety of ways, to too narrow and 'syntactic' an

C. J . BRUMFIT

163

R E F E R E N CE S

Allen, J . P . B. and S . P . Corder ( 1 973), The Edinburgh Course in Applied Linguistics, Vol. 1 Oxford: Oxford University Press. Arthur, B. ( 1 973) Teaching English to Speakers of English New York: Harcourt, Brace Jovanovich. Bernstein, B. ( 1 97 1 ), Class, Codes and Control, Vol. 1 London: Routledge and Kegan Paul. Bernstein, B. ( 1 975), Class, Codes and Control, Vol. 3 London: Routledge and Kegan Paul. Corder, S. P. ( 1 973), Introducing Applied Linguistics Harmondsworth : Penguin. Corder, S . P . ( 1 978), Learner Language and Teacher Talk, A VLJ, 1 6, 1 , 1 978, 5 - 1 3 . Curran, C. ( 1 976), Counseling-learning in Second Languages Illinois: Apple River Press.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

interpretation of language teaching (Wilkins 1 972b, 1 976; Widdowson, 1 978, 1979). The so-called 'humanistic' approaches to language teaching, popular particularly in the United States (Gattegno, 1 976; Curran, 1 976; Stevick, 1 976) are attempts to increase the social and personal value of the language learning process. Such responses, coming from experienced teachers have been received sympathetically on a wide front, but there is no systematic approach to language in society by which they can readily be explained or to which they can readily be referred. When such an account is fully developed (and it cannot be too far away) it will be an applied linguistic account, not a linguistic one, for at every point the emphasis will have to be on the interaction between language use (in relation to personal and group needs and the social environment) and language development, for the process of using and the process of developing (both of language by individuals and of languages by the uses they are put to by individuals) will be seen to be intimately related. Such an account cannot fail to have direct relevance to all those whose concern is the teaching of languages, the remediation of linguistic malfunction, the explanation of specific examples of linguistic performance, or any other application of linguistics. An account such as I have outlined is, I suggest, a reasonable prediction. It is concerned, however, with a level of generality somewhat higher than the immediate concerns of many who call themselves applied linguists. Yet a detailed and systematic account of the nature of language development and change, linking individual motivated and unmotivated change with the historical development of languages, will provide a framework within which attempts to cause changes in linguistic behaviour can be described and evaluated. At the same time, such an account will allow activities such as teaching or speech therapy to be defined more positively in their relations to language use, for language will be seen as changing by its nature and by the uses to which it is put: the process of change will not be seen as deviant when the nature of change is more fully understood.

1 64

DISCUSSION

Gattegno, C . ( 1 976), The Common Sense of Teaching Foreign Languages New York: Educational Solutions . Halliday, M . A . K. ( 1 975), Learning How to Mean London: Edward Arnold. Halliday, M . A. K. ( 1 979), Development of texture in child language, in Myers 1 979. Hinde, R. ( 1 972), Non-verbal Communication Cambridge: Cambridge University Press. Kennedy, B. H. ( 1 930), Revised Latin Primer London: Longman, revised edition. Linden E. ( 1 975), Apes, Men and Language Harmondsworth: Pelican. Lyons, J. ( 1 972), 'Human language', in Hinde, 1 972.

Rogers, S. (ed .) ( 1 976), They Don't Speak Our Language London: Edward Arnold. Roberts, P. ( 1 964), English Syntax New York: Harcourt Brace & World. Rosen, H. ( 1 978), 'Signing On' , The New Review, February 1 978, reprinted in BAAL Newsletter no. 7, June 1 979. Schumann, John H . ( 1 978), The Pidginization Process Rowley Mass: Newbury House. Spolsky, B. ( 1 978), Educational Linguistics Rowley Mass: Newbury House. Stevick, E. ( 1 976), Memory, Meaning & Method, Rowley Mass: Newbury House. Stubbs, M . ( 1 976), Language, Schools and Classrooms London: Methuen. Trudgill, P. ( 1 975), Accent, Dialect and School London: Edward Arnold. Widdowson , H. G. ( 1 978), Teaching Language as Communication Oxford:· Oxford University Press. Widdowson, H. G. ( 1 979), Explorations in Applied Lznguistics, Oxford: Oxford University Press. Wilkins, D. A. ( 1 972a), LinguistiCS in Language Teaching London: Edward Arnold. Wilkins, D. A. ( 1 972b) Grammatical, situational and notional syllabuses ' , Proceedings of third international Congress of Applied Linguistics Heidelberg: Julius Groos Verlag. Wilkins, D. A. ( 1 976), Notional Syllabuses Oxford: Oxford University Press. Wilkinson, A. ( 1 975), Language and Education, Oxford: Oxford University Press.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Myers, T. (ed.) ( 1 979), The Development of Conversation and Discourse, Edinburgh : Edinburgh University Press.

MODELS AND FICTIONS H. G. W I D D O W S O N University ofLondon

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

How are we to name the activity we are engaged in when we attempt to work out a principled approach to the solution of practical problems in the acquisition and use of language? This is not a trivial question. One does not have to embrace extreme Whorfian doctrine to recognize that how a thing is called can have a critical effect on how it is conceived. And this can have important consequences on how we conduct our enquiries and how they appear to potential sources of research funds. Our activity has commonly been called applied linguistics. A metathetic alternative has been suggested: linguistics applied. What do we understand by these two terms and which of them provides the most appropriate designation for what we do? I assume that, semantically, the difference between the two expressions is that in the case of applied linguistics we have a type classification, what Bolinger refers to as a 'characterization' (Bolinger 1 952) whereas with linguistics applied we do not. We might compare Lost paradise with Paradise lost. On the one hand a type of paradise, a palm-fringed beach perhaps, thronged with lotus eaters; on the other hand, the fall of man, the un_ique loss of Eden. Thus applied linguistics can be understood as a kind of linguistics, like historical linguistics or folk linguistics. This presumably allows its practitioners to define an independent perspective on the general phenomena of language and to establish principles of enquiry without necessary reference to those which inform linguistics tout court. With linguistics applied we do not have this option. Whatever we do with linguistics, however we apply it, the informing principles which define this area of enquiry, already pre­ established, must remain intact. Any other principles we invoke must be auxiliary operating principles and have to do not with theory as such but with the technology of application. It seems to me, then, that with linguistics applied the theory of language and the models of description deriving from it must be those of linguistics. As an activity, therefore, it is essentially conformist. Applied linguistics, on the other hand, can develop its own non-conformist theory, its own relevant models of description. Both lines of approach have their dangers. The tendency of linguistics applied will be to dance attendance to whatever tune is currently in theoretical fashion. The tendency of applied linguistics will be to dance around in circles to no tune at all. For linguistics applied, therefore, the question of central concern is: how far can existing models of description in linguistics be used to resolve the practical problems of language use we are concerned with . For applied linguistics, the central question is: how can relevant models of language description be devised, what are . the factors which will determine ' their effectiveness. This first question takes us at once to the theoretical status of linguistic models. Linguistics claims to be a science, like physics. It must, therefore, conform to general principles of scientific enquiry, and in particular it must avoid intuitive contamination in the interpretation of its findings. And it is here that we come against a major problem. Linguists have a way of claiming

1 66

DISCUSSION

mind of E. M. Forster's remark about Henry James: 'Most of human life has to disappear before he can do us a novel' (Forster 1927: 147). This reference to literature is not gratuitous and I shall have more to say about fiction presently. Sampson argues that linguistics is a science, like physics. Other people are not so sure. Here is Peter Matthews: Is there a true analogy . . . . between a 'science' like linguistics and a 'science' like physics? The problem is not that linguistics is at this stage 'less advanced' : will the descriptive linguist ever b e able t o propose an ordinary mathematical theory for a language, will he ever be able to falsify a 'theory' by one ob­ servation, will he ever be able to construct an absolute test of his predictions? Surely we delude ourselves if we imagine that linguistics will be 'like physics' sometime in the future . . . . (The problem) is rather that the extent of our data is in principle not precise. Languages change, and language interacts con­ tinuously with other forms of social behaviour . . . . We cannot even be ex­ pected to account for 'all of the data' , simply because we do not know what 'all o f the data' means! (Matthews 1 972:77 quoted in Bolinger 1 975:55 1 )

W e must surely accept the possibility that human knowledge and behaviour are just not the same kind of phenomena as those which are independent of human agency, that they are intrinsically participant, first and second person phenomena which cannot be detached to be third person data of an objective kind and are not therefore susceptible to scientific enquiry. But if this is so, what then is the status of statements in linguistics? This is not the place, and I am not the person, to enter into the complexities of epistemology. But I would like to make a lay observation or two which I think might have a bearing on the question. Central to the issue is the matter of the privileged position of the enquirer. In the physical sciences, the enquirer is trained in detached analysis, in putting the phenomena he wishes to investigate at a remove from the immediacy of perception and intuition. In this way he aims at correcting the distortions of the human factor. What this involves, in

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

validity for solutions which seem outlandish and make no appeal whatever to ordinary intuition on the grounds that language, like God, moves in a mysterious way and outside the range of the common man's awareness. On the other hand, this does not inhibit them from commending solutions because they seem 'intuitively correct' and disapproving of those which seem to be 'counter-intuitive'-these expressions are of high frequency in the literature. So it is that the linguist finds a way of both having his cake and eating it too. For although intuition may serve as a source of hypotheses, it cannot scien­ tifically be used to test them or to provide them with a mark of approval. Geoffrey Sampson, for one, is very insistent on this point . In his discussion of the evidence for linguistic theories he says: ' We do not need to use intuition in justifying our grammars, and, as scientists, we must not use intuition in this way (Sampson 1 975 :70) . Sampson argues that linguistics, like any other science, is, and must be, empirical. But it seems that in order to meet this requirement, linguistics has to restrict its attention to a fairly narrow strip of language phenomenon: most of what is of human interest about language behaviour is left out of account. This, of course, is Hockett's objection to Chomskyan grammar: it leaves out of account, he says 'just those properties of real language that are most important' (Hockett 1968 : 1 0) . This puts me in

H . G . WI DDOWSON

1 67

a raid on the inarticulate with shabby equipment always deteriorating And every attempt is a wholly new start and a different kind of failure. (East Coker)

The linguist presents his view of the world, his model of reality. Shift the perspective and you get a different view, and a different reality (for further discussion see Widdowson 1979). The linguist's model has no privileged patent on the truth, it is open to investigation by another, that model by yet another, and so on. Now, of course, it can be very interesting and enlightening to see

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

effect, is the acquisition of a secondary culture of a non-participant kind defined by the philosophy of science and apart from the popular beliefs and values of participant primary culture. From the vantage point of this secon­ dary culture, the scientist can reveal a different, third person reality, which will often run counter to that which is popularly accepted. He can, fur­ thermore, demonstrate its truth, though until the respectability of such a secondary culture was established and its practical consequences made clear this truth was commonly condemned as heresy. But now how does all this apply to the linguist as scientist? Can he detach himself and take up a privileged observer's position in the same way? I do not think so. The linguist is of necessity a participant: like other social scientists he enacts the data he analyses so that there are no separable facts out there, as it were, open to objective validation. We get a model of behaviour (that of the linguist) used to analyse another model of behaviour (that of the common man), one set of cultural assumptions (which are secondary and superposed) used to interpret another set of cultural assumptions (which are primary and acquired through ordinary socialization). So it is that the linguist is not a representative speaker/listener. He comes from a different culture and so is able to analyse in detachment. The data he produces is devised to demonstrate the general underlying rules of language he has inferred. The sentences that linguists present have the same sort of status as the components of kinship analysis presented by anthropologists: devices for the design of ideal systems. But what cognitive reality do they have for people who use language and enact kinship relations in the ordinary business of social life? What is their empirical status? As soon as a language user is presented with a sentence he quite naturally converts it into an utterance. His culture prevents him doing otherwise: he will associate the isolated linguistic expression with· a likely context of occurrence and interpret it accordingly. If you train him to do otherwise, he loses his innocence and at the same time his usefulness as an informant. He becomes tainted by foreign influence. The linguistic analyst, in conformity with the secondary cultural principles he embraces, represents language as a system in a steady state, with each term in its place in peaceful co-existence with others-a picture of classical harmony and well-formed order. An artefact, a cultural construct. For really there is conflict, an intrinsic instability-with rules moving into dominance or decline. And every use of language is a resolution of this conflict, the ex tempore exploitation of the meaning resources available for expressing the variety and contradictions of human experience; in short, as T. S. Eliot puts it:

1 68

DISCUSSION

how one set of cultural values interprets another and the exercise can yield all kinds of insights, but there is no way, as far as I can see, of establishing which is true and which is not. What I am saying is that the linguist, in common with other social scientists, simply fashions his data to fit a particular set of cultural assumptions. Further, he will be inclined to interpret behaviour according to his own predispositions. Liam Hudson has this to say about psychologists and his remarks apply, mutatis mutandis, to linguists as well:

We return to fiction. It seems to me that models of human behaviour in the social sciences are comparable in status and function to the representations of human behaviour in novels or plays or any other art form. Both depend upon idealization procedures which in effect yield archetypes of a kind which we can set into correspondence with actual and non-idealized reality. There is not, and cannot be, any direct empirical link between either of them and the external world. Descriptive models and fictional representations create archetypal norms of human behaviour which we can accept as a plausible pattern against which actuality can be compared. Their function is not to be correct but convincing, to serve as a means towards a more perceptive awareness of what we do and who we are. Hamlet and Heathcliff do not carry conviction as characters because they have empirical counterparts among our acquain­ tances. If they did, they would diminish into insignificance. So it is with models of linguistic description. Their value, as a kind of art form, depends on idealization and a detachment from empirical villidation. Linguistic theories, I suggest then, are fictions. This is a perfectly respec­ table thing to be. As Harre points out the fictionalist view of theories has a long history. He describes it like this: As an aid to thinking a plausible theory may be a very powerful tool, and by using known laws of nature in describing the behaviour and nature of the entities with which it deals it gains plausibility. But the entities themselves have no more reality than the characters of fiction, and the terms which are used to describe and particularly to refer to them are like the names.and addresses of characters in novels. (Harre 1 972: 8 1 )

I find the notion of a theory as a n aid t o thinking an appealing one. You do not apply a theory which is a fiction, any more than you apply a novel. What you do is to use it to develop your awareness of what a useful model of behaviour might be for your particular purposes. In this view, a linguistic model is of value only to the extent that it can help us towards a definition of what an applied linguistic model should be. So we come to the second question I posed at the beginning. What would a relevant model of language in applied linguistics look like: what assumptions,

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

The truth is that, to a remarkable extent , psychologists do research in their own image. One notices this daily. Psychologists with high I .Q.s do research that reflects well on those with high I . Q.s. Neurotic introverts show that neurotic introverts do well at school and university. Convergers do research that bodes well for convergers; divergers, for divergers . . . . Just as novelists draw on their experience, so too do psychologists. We would both be cut off, otherwise, from the springs of our intellectual vitality. (Hudson 1 972: 129)

H . G . W I D DOWSON

169

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

beliefs, values would it need to satisfy? What would be its cultural character, if you like? It would, I think, have to be congruent with the knowledge and attitudes of language users, and be developed, therefore, along the sort of line that ethnome�hodologists have made familiar. It would attempt to record the language user's intuitive, imaginative as well as rational awareness of the nature of language. It naturally follows that applied linguistic models of language would vary according to the kind of language user concerned. They would be consumer based. Halliday 1 964 (severely savaged in certain purist quarters) with its argument in favour of relating adequacy to specific relevance indicates an appropriate rationale for the applied linguist. Again, Halliday 1 969, con­ cerned as it is with relevant models of language points us, I think, in the direction we might go. Thus the relevant model of language for teachers of literature will take into account the cultural perspectives of literary criticism and aesthetics. The relevant model for students of physics doing a course in ESP will take into account the philosophy and methodology of science. The relevant model for learners in a particular cultural context will take into ac­ count the traditions, values, beliefs and customary practices of that culture and the way language is used to express them. In all such cases, the starting point is the need and purpose that the model of language must provide for. You do not start with a model as given and then cast about for ways in which it might come in handy. You start instead with a characterization of the learner and his circumstances: his behaviour and beliefs, his habitual patterns of thinking, his conventional attitudes to language, the extent and influence of literacy and other social patterns of cultural transmission and control, and so on. All these, and more, are factors which might effect the model of languages to be presented. The relevance of the learner's first language has long been recognized and interesting work has been done recently on the matter of language distance (see Corder 1 979, Kellerman 1979). But questions of social and psychological distance, as discussed, for example, in Schumann 1 978, are of equal importance in the design of effective applied linguistic models. We are concerned with language as the rightful property of language users, not as the special preserve of the linguist. The issue of directionality is critical here, as it is for sociolinguistic studies (see Cooper's paper in this issue). Linguistics applied works in one direction and yields descriptions which are projections of linguistic theory which exploit the data of actual language as illustration. Applied linguistics, as I have characterized it here, works in the opposite way and yields descriptions which are projections of actual language which exploit linguistic theory as illumination. My own belief is that it is only by preferring applied linguistics to linguistics applied that we shall achieve something which is relevant and ac­ countable in terms of usefulness and avoid the kind of ethnocentrism and cultural imposition that has marked so much of language study and teaching in the past. There are signs of this preference emerging in recent work. But we have a long way to go and there will be many an alluring linguistic Will-o' -the­ wisp to lead us astray from the path of our intentions.

1 70

DISCUSSION

REFER E NCES

Bolinger, D. L . 1 952, ' Linear modification', PMAL 61 1 1 1 7- 1 144. Bolinger, D. L . ( 1 975), Aspects of Language (2nd Edition) New York: Harcourt, Brace Jovanovich. Corder, S. P. ( 1 979), ' Language distance and the magnitude of the language learning task', in Studies in Second Language A cquisition Vol. 2 No. 1 . Forster, E . M . ( 1 927), Aspects of the Novel London: Edward Arnold . Harre, R . ( 1 972), Th e Philosophy of Science Oxford: Oxford University Press.

Halliday, M . A. K. ( 1 969), 'The relevant models of language' in Educational Review University of Birmingham 22. 1 pp. 26-37, reprinted in Explorations in the Functions ofLanguage London: Edward Arnold 1 973 . Hockett, C. F. ( 1 968), The State of The Art The Hague: Mouton. Hudson, L. ( 1 972), The Cult of The Fact London: Jonathan Cape. Kellerman, E. ( 1 979), 'Transfer and non-transfer: where are we now' in Studies in Second Language Acquisition Vol. 2 No. 1 . Matthews, P . H . ( 1 972), Review o f Paul Garvin (ed.) Method and Theory in Linguistics The Hague: Mouton 1 970 In Lingua 29: 67-7 7 . Sampson, G . ( 1 975), The Form ofLanguage London: Weidenfeld and Nicolson. Schumann, J. H . ( 1 978), 'Social and psychological factors in second language acquisition' in Richards J. C. (ed.) Understanding Second and Foreign Language

Learning Rowley Mass . : Newbury House. Widdowson, H. G . ( 1 979), 'The partiality and relevance of linguistic descriptions' in Explorations in Applied Linguistics Oxford: Oxford University Press.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

Halliday, M . A . K. ( 1 964), 'Syntax and the Consumer' in C . l . J . M. Stuart (ed.) Report of the 1 5th Annual Round Table Meeting on Linguistics and Language Study. Washington D.C . : Georgetown Monographs 17 pp. 1 1 -24.

RE V I E W S EVELYN HATCH, Second Language Acquisition: A

Book ofReadings

Rowley, Mass. : Newbury House, l 978. Pp. 483

Applied Linguistics, Vol. I, No. 2

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

The field of second language (LJ acquisition has grown rapidly in the past decade. Ten years ago, a one-volume anthology of L 2 research articles might actually have been a plausible overview of the field. Today, both the variability and depth of research in L 2 acquisition exceed the possibilities of adequate representation i n a single collection o f papers. Some might be misled b y the title of Professor Evelyn Hatch's volume into believing that the material provides comprehensive coverage of the field. However, Hatch herself recognizes that Second Language Acquisition: A Book of Readings (hereafter SLA) is not at all exhaustive; its purpose 'is to acquaint the reader with the em­ pirical studies of second language acquisition in natural environments-that is, language acquisition outside the classroom' (p. 1 7). According to Hatch, the volume is an answer to a 35-year-old plea by Werner Leopold (to whom she dedicates the volume) 'to add sorely needed case histories of infant bilingualism and infant language to the available material, as indispensable spade work for the higher purposes of linguistics' (p. 1 1 ). Hatch has given us a unique selection of articles. While a few of the twenty-six ar­ ticles in the volume are reprinted from previous publications, most of them have not appeared heretofore in standard publications. These selections are from conference presentations (for example, Hatch's own introduction is the substance of her plenary address at the Second Language Research Forum at UCLA in February of 1 978), Master's theses by UCLA graduate students in teaching English as a second language, and other material available usually only through the 'academic underground' (mimeo­ graphed, dittoed, and xeroxed papers that manage to get passed around from scholar to scholar). Even among the previously published material, Hatch has provided some selections which we would be unlikely to consult in our everyday browsing through journals-Imedadze's article on Soviet research, or Leopold's and Burling's articles printed over two decades ago. Another refreshing feature of the selection of articles in SLA is the range of languages which are represented. At least eight languages are featured: Chinese, English, French, Garo, German, Japanese, Norwegian, and Spanish. At the end of the book Hatch has included ninety-eight abstracts of research papers which fell into the same category as the twenty-six which were actually included in SLA. These papers were presented in abstracted form 'since they were either too lengthy to include, not published in English, or were widely available in child language collections' (p. 1 1). The abstracts do indeed 'make the picture more epmplete' (p. 1 1) and provide the researcher with a significant collection of research findings. However, their presentation in alphabetized, rather than subject-categorized, order makes them difficult to refer to. The reader would need to become thoroughly familiar with all ninety-eight abstracts to be able to find a subset of studies in a particular area. A helpful feature of SLA is the introductory essay preceding each article. Hatch has in each case carefully emphasized the significance of the selection in developing an under­ standing of research in the field. The reader can thereby gain a picture both of the rationale for including each article and of the editor's overall understanding of the field. SLA begins with an excellent introduction by Hatch in which she offers a historical summary of research on untutored L2 acquisition, an exhaustive list of research on natural L 2 acquisition in the last decade, and comments on some 'myths' about L 2 acquisition. The first major section of SLA consists o f seventeen papers o n case (observational) studies of L 2 acquisition. This section is divided into three subsections according to the age of acquisition: (l) infant bilingualism, or simultaneous acquisition

1 72

R E V I E WS

(Received A ugust 1979)

Reviewed by H . DOUGLAS BROWN University of Illinois

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

of two languages; (2) L 2 acquisition in 'young' children; and (3) L2 acquisition in 'older' learners. The second major section of SLA provides seven experimental research papers, including three papers representing the now widely-known mor­ pheme-acquisition research. The third section of the book contains two articles on dis­ course analysis in L1 acquisition; one of these articles is Hatch's own-previously available only in the 'underground' . I t i s of course clear t o all that a n anthology of papers will never provide an adequately cohesive exposition of a field of study. Hatch's introductory notes, combined with her own introduction, come close to providing the kind of thread that one needs in order to d4.scern history and development and interrelationships. However, the reader must simply remember that it is not the purpose of an anthology to produce the kind of comprehensive statement of a field which perhaps only a singly­ authored textbook can provide. A somewhat disturbing aspect of SLA is that it is never made clear who the book is intended for. It appears to be primarily, if not exclusively, a handbook of sorts for researchers . In spite of its rather forbidding length (483 pages), it has potential for a textbook for a course on L2 acquisition, but there is no indication from the editor on how it might be used as such. A rather simple paragraph or two in the introduction could have indicated practical uses of the volume. And the addition of discussion questions after each selection would have served classroom purposes. The most serious shortcoming of SLA is a careless job of referencing. An anthology of research ought to be meticulously referenced, so that the serious user can locate material that is cited or referred to in the selections. A nine-page bibliography at the end of the volume combines references from all twenty-six papers. But literally dozens of references are missing ! In article after article, important references are omitted from the bibliography. One wonders what criteria were used for including references, since so many are omitted that it is hardly a matter of a few 'slip-ups' . Not only are references omitted, but in many cases (too many to cite) incorrect information is given-dates, pages, and an inexcusable number of misspelled names. In some of Hatch's introductory notes, names of researchers are 'dropped' without reference to specific sources. We can only hope that a revised edition of SLA may provide the kind of impeccable referencing expected of research publications. And an index would cer­ tainly be an important addition to SLA. Finally, while the articles included in the volume are indeed both interesting and soundly conceived, one cannot help but wonder if the inclusion of no less than twelve articles by 'local' UCLA students and faculty gives us a somewhat jaded picture of the state of the art. It is an editor's prerogative to bias the selection of material , but in the interest of representing both the theoretical and methodological variability in tbe field, surely quite a number of the ninety-eight abstracted studies might have easily sub­ stituted for some of the UCLA selections. These weaknesses need not detract from the contribution which Hatch has made in giving us a collection of papers that we would not otherwise have encountered. In so doing, she has carefully reflected a growing and important subfield of study in L 1 acquisition. The selected articles are evidence of a field o f study that has a substantive body of research undergirding it and supporting the formulation of theoretical stances within it. Hatch has shown us that current research in natural L2 acquisition is methodologically sophisticated, a great deal of which research Hatch herself is responsible for stimulating and guiding. Perhaps we can now look forward to other volumes which represent perceptively defined subfields of research on L 2 acquisition, and therewith be aided in the consolidation of the mushrooming of knowledge about how and why people learn second languages.

REVIEWS

1 73

BETTY WALLACE ROBINEIT, Teaching English to Speakers of Other Languages: Substance and Technique

University of Minnesota Press, 1 979. Pp. xiv + 32 1 .

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

This book is a general introduction to what is described (p. xiv) as 'the art of teaching English to speakers of other languages' and is intended to be 'only a beginning ' . It is a fairly substantial beginning, nevertheless, and is clearly intended to cover most of the basic groundwork towards equipping the reader to be a competent teacher. It assumes a mainly American audience, which in itself makes it desirable that European readers should look at it, but it will be reviewed here specifically from the other side of the Atlantic, for it raises in its approach a number of interesting differences between British and American traditions. British general introductions to EFL/ESL teaching have tended to be less weighty than this and perhaps to have had a more direct concern with classroom practice (Billows, 1 96 1 ; Bright and McGregor, 1 970; Harrison, 1 973). Indeed, for most of the past decade Bright and McGregor has provided the only really substantial work in this area, and it has been used far outside the setting, secondary schools in post-colonial Africa, for which it was originally written. More recently, Broughton, et a/. ( 1 978) have produced a more general overview, but this too by its attitude to the servicing dis­ ciplines, particularly linguistics, and to classroom practicalities, places itself firmly within a pragmatic, British tradition. The book is subtitled 'substance and technique' , and the introduction remarks (p. xi) that 'an effective teacher must have a solid grasp of the substance of the subject to be taught and, especially, the techniques for teaching it'. In fact, the sutistance seetion of the book deals in some detail with the grammatical, sound and vocabulary systems of English, but with nothing else. The reader may be forgiven for thinking that the subject to be taught is really a description' of English of a somewhat traditional type. The second part of the book is also rather confusingly presented. As in most American dis­ cussions of methodology, the teaching of culture has a more prominent position than in other traditions. Various current trends (linguistics and psychology, contrastive and error analysis, a note on Gattegno and Curran via Stevick , 1 976, and a cursory reference to Wilkins) are mentioned , second language acquisition is surveyed, there are over seventy pages on the teaching process itself, and a final short chapter on testing. It is true that a detailed reading of the book reveals that there is a great deal more to English teaching than simply the application of principles derived from descriptive linguistics, but there is very little sense of integration. A general account of research is offered, but whereas in the British books mentioned, or among American books, in Finocchiaro ( 1 969) or Rivers ( 1 968) there is a feeling of constant reference to the total process of classroom teaching, here we seem to be considering a great deal of material simply because it is there. At the end the book just stops, and the naive reader may legitimately ask , 'What is it all for?' No-one really explains. Part of the lack of excitement springs from the lack of contact with many of the most exciting recent developments on the fringes of core linguistics. There are few references to discourse analysis, interlanguage, the work of Krashen, Fanselow, or on this side of the Atlantic, Corder, Widdowson, Sinclair and Coulthard and many others whose work has been much discussed in the last few years. Although Wilkins is briefly mentioned, I cannot recall any reference to the work of the Council of Europe. It may be objected that none of this work is appropriate to a beginner's book, but this is a substantial volume, and a great deal of less exciting work is discussed in some detail. Nearly fifty pages are devoted to the sound system of English-all of it material which is available elsewhere, and the preface remarks, curiously, that 'relatively little attention is paid to this aspect of language teaching'. On the other hand, the chapter on the vocabulary system is useful, and deals with an area which is truly neglected. The difficulty is that little of the 'substance' material presented is new, while 'technique' is

REVIEWS

1 74

Reviewed by CHRISTOPHER BRUMFIT University of London Institute of Education

(Received October 1979)

R EFER ENCES

Billows, F. L., 1961. The Techniques ofLanguage Teaching. London: Longman.

Bright, J. A., and McGregor, G. P . , 1 970. Teaching English as a Second Language. London: Longman. Broughton, G. et al. , 1 978. Teaching English as a Foreign Language. London: Routledge & Kegan Paul. Finocchiaro, M . , 1 969. Teaching English as a Second Language, 2nd ed., 1 974. New York: Harper and Row. Harrison, B., 1973. English Arnold.

as

a Second and Foreign Language. London: Edward

Rivers, W . , 1968. Teaching Foreign-language Skills. Chicago, Illinois: University of Chicago Press. Stevick, E . , 1 976. Memory, Meaning and Method. Rowley, Mass. : Newbury House.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

all too often reduced t o further analysis of the material or activities i n teaching, for example the extensive discussion of drills on pp. 204ff. There are important issues to be raised, even with beginners, over creating communication in classroom contexts, and drills provide a convenient starting point for such discussion, but the kind of drill labelled 'communicative' on p. 209 in no sense offers students the chance to produce their own language in a remotely realistic situation. Drills such as this certainly have some place in teaching, but there are many other major methodological principles to be explored before they can be related to communicative procedures. Altogether, aspects such as classroom management, classroom therapy, classroom drama are dealt with better elsewhere. This book's strongest moments are when it is closest to technique and furthest away from education. The analysis is partial. Perhaps it seems strange thatin an applied linguistics journal a book such as this, centred so strongly on linguistics, should be attacked for precisely that. But there is an important issue at stake, for applied linguistics as well as for methodology. No ap­ plication can be pure, if it is a genuine application. The British tradition in methodology has been if anything too sceptical of linguistics, and indeed of theorizing in general, but the best methodological books, British or American, see teaching as a pragmatic and a principled activity, in which the principles derive from many non­ linguistic sources. This book uses linguistics for teaching, and will no doubt help many students, but it risks alienating those who are most sensitive to the fact that the sub­ stance of TESOL is a great deal more than knowledge of English language and culture.

REVIEWS PIERRE MACHEREY, A

1 75

Theory ofLiterary Production

London: Routledge and Kegan Paul, 1 978. Pp. ix + 326.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

In an early essay, T. S. Eliot distinguished between two kinds of literary indebtedness. 'Immature poets,' he noted, 'imitate,' while 'mature poets steal'; 'the good poet welds his theft into a whole of feeling which is unique, utterly different from that from which it was torn; the bad poet throws it into something which has no cohesion . ' For better or for worse, much the same can be said about interdisciplinary borrowings. The 'mature' literary critic or historian, for example, 'welds' insights from other disciplines into the distinctive 'whole' that defines his or her subject. Appropriations from other fields thus generally do not change a discipline but reinforce already established tendencies. The application of linguistics to literary criticism illustrates this point. Out of the diversity of contemporary linguistics, literary critics have usually selected those ideas most compatible with the formalism that has dominated modern critical theory. Even before the recent explosion of interest in structural linguistics, formalist critics (like the American New Critics) assigned special importance to language, arguing that literary works call attention to their medium rather than represent an external referent. Familiarity with modern linguistics, especially with the theories of Ferdinand de Saussure, has furthered disenchantment with mimesis by encouraging critics to enclose even reality itself within fictive discourse. All language, not just literary language, now seems to many critics ' fundamentally, constitutively unrealistic, ' to use Roland Barthes' words; no referent escapes prior constitution by some order of signifiers. In deflecting attention from 'reality' to their own verbal devices, literary works remind us what other texts try to suppress, namely, that in all discourse 'sign and meaning, ' text and referent, can 'never coincide' (Paul de Man). ' Far from being an analogical copy of reality , ' Barthes concludes in Critical Essays, 'literature is on the contrary the very consciousness of the unreality of language' (his emphasis). Until recently, Marxist critics have turned to economics or sociology, not linguistics, to support their critical theories. Rather than correct the obsession with form that dis­ tinguisheP their 'bourgeois' adversaries, Marxist critics in the 1 930s and '40s tended to leap to the other extreme. Concern for the formal features of a literary work gave way to analysis of what the work said, or, more precisely, what the economic infrastructure of history disclosed through the text . If 'vulgar' Marxists thus grounded literature in economics or history, more recent ones have felt this ground give way or at least move out from under the dialectical schema of classical Marxist thought. Loss of confidence in Marx's overview of history has in turn released interest in the autonomy of literature, in the ways in which literary works shape rather than copy the external world. Contemporary Marxist criticism, especially the works of Fredric Jameson, Terry Eagleton, and Lucien Goldmann, has tried to open Marxism to formalism, using its insights to complicate but not sever the relations between literature and its historical milieu. Pierre Macherey's A Theory of Literary Production (first published in France in 1 966 as Pour une theorie de Ia production litteraire) extends this interest of Marxist criticism in formalist theory, especially structuralism. The highly _repetitious chapters which open the book sketch a view of literature indebted to contemporary structuralist thought. Macherey argues, for example, that literary works do not imitate some pre­ existent truth but create what they say: 'the real, as it is formulated in the discourse of the work, is always arbitrary because it depends entirely on the unfolding of this dis­ course. ' Rather than measure the work against 'an extrinsic or concealed truth , ' the critic consequently seeks to describe the 'conditions' of the text's autonomous 'production.' The text, in other words, allows for a 'multiplicity of readings' rather than communicates an already known external truth; the 'line of its discourse is thickened by reminiscences, alterations, revivals , and absences' because 'the object of this discourse is multiple, a thousand separate, hostile and discontinuous realities . ' For

176

R E V I E WS

all its multiplicity and randomness, the text still seems 'necessary, ' held together by 'a specifically poetic-rather than logical-rigour which unites form and content. ' The critic's task, then, is to uncover the internal, freely invented principles responsible for this formal 'rigour. ' After freeing literature from external reality, Macherey struggles vainly to reunite the two. Admittedly 'autonomous,' the text is not 'independent' ; though obviously not the same as 'a theoretical knowledge' (by ' knowledge, ' Macherey, followingAithusser, means the 'scientific analysis' formulated by 'the Marxist party'), the 'writer's version of reality' is nonetheless more than illusion. Literature offers 'the analogy of a knowledge' : 'This does not mean that the book i s able to become its own criticism: i t gives an im­ plicit critique of its ideological content, if only because it resists being incorporated into the flow of ideology in order to give a determinate representation of it. . . . Fiction is not truer than illusion; indeed, it cannot usurp the place of know­ ledge. But it can set illusion in motion by penetrating its insufficiency , by transforming our relationship to ideology. '

'The structure of the work, which makes i t available to knowledge, i s this internal dis­ placement, this caesura, by which it shows without reflecting. The literary work gives the measure of a difference, reveals a determinate absence, resorts to an eloquent silence. ' Chapters on Tolstoy, Verne, Borges, and Balzac never clear up the ambiguity of this passage. Drawing on Freud, Macherey tries to show that history is the ' unconscious' which the very 'absences' of these texts reveal. Like patients, the works display contra­ dictions unknown to themselves (and to most readers), which 'set in motion' an analysis that terminates in 'Marxist science.' Literature 'criticizes' ideology in much the same way that a murder victim 'criticizes' his assailant; in each case, a battered form embodies truths that await articulation in the discourse of a superior observer. Such condescension toward literature mars Macherey's argument and results from his initial decision to separate 'illusion, fiction, theory, ' allowing for 'no bridge, no unbroken path, from one order of discourse to another. ' He accepts, in short , a for­ malist definition of literature but tries to enlist it in the service of Marxist theory, stopping short of questioning the scientific claims of the latter. Between separating fiction and theory, and identifying the two as fictive writing, lies a third alternative that Macherey does not consider, namely, distinguishing, but not dividing, literature from genuine knowledge. In demonstrating the structural affinities between literary and extra-literary writing, linguistics has already given critics grounds for exploring this third alternative. But theorists like Barthes have used this insight to demonstrate the 'unreality' of all writing, not the continuity of literary and cognitive discourse. Macherey's theory contests this reduction of fiction to illusion, but only by ac­ centuating its subservience to Marxist theory . Failing to challenge the 'inevitable' remoteness of 'the project of writing a novel from that of telling the truth , ' he reduces fiction to a mute, yet formally 'eloquent, ' appendage of Marxist 'science.' In his book at least, Marxism continues to offset inflated tributes to the autonomy of literature with equally exaggerated assertions of its dependency. (Received A ugust 1979)

Reviewed by MICHAEL FISCHER Department ofEnglish University of New Mexico

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

The torturous argument here recalls New Critical attempts to claim some kind of 'knowledge' for literature after releasing it from accountability to the preexistent world. In each case, the desire to differentiate literature from illusion conflicts with the unwillingness to make literature, in Macherey's words, 'answerable to any external meaning or reality.' For Macherey, the very form of a literary work allows it to 'set illusion in motion' ; the silences, gaps and contradictions in the text transform the ideology that it 'mirrors . '

REVIEWS PETER STREVENS, New

1 77

Orientations in the Teaching ofEnglish

Oxford : Oxford. University Press, 1 977. Pp. xii + 183.

'The definition [of applied linguistics] is framed as a set of five propositions:

1 . Applied linguistics has a basis in theory and principle. . . .

2.

These bases are multiple . . . .

3 . Applied linguistics is not restricted to an interest in the learning and teaching of languages . . . .

4. Applied linguistics re-defines itself afresh for each task . . . . 5 . Applied linguistics is dynamic, not static' (pp. 37-40). In brief, Strevens is trying to show that language teaching/learning is a concern of applied linguists but that no single theory derived from formal linguistics or formal psychology, or any other root discipline can provide the answers necessary in language teaching/learning but rather that it is necessary for the applied linguist and the language teacher to abstract from a variety of sources the elements which will ultimately produce both a coherent method of teaching and learning languages and a formal model to account for the principles underlying the method . This is a position with which most applied linguists would find themselves in sympathy. The book itself is organized into five parts : Principle and Theory in Language Teaching, Methodology and Teacher Training, Special Problems in ELT, the Language We Teach (varieties of English) and Some Technical Questions (which deals with instructional technology and with cost-effectiveness). Each of the five parts is divided into smaller sections, so that there are, in fact; 14 sections to the book, plus a bibliography and an index . Nine of the 1 4 parts are reworkings of articles previously

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

'Men learn languages for the ordinary intercourse of society and communication of thoughts in common life, without any further design in the use of them . And for this purpose, the original way of learning a language by conversation not only serves well enough , but it is to be preferred as the most expedite, proper and natural. Therefore, to this use of language one may answer, that grammar is not necessary . . . ' 1 It is a brave author who can use the phrase 'new orientations' in the title of a book addressed to language teachers in this third quarter of the twentieth century. As ' Strevens points out in his first chapter, ' . . . there is no accepted, well-articulated theory taking account of all the complex elements entailed in the organized teaching and learning of languages; the history of language teaching during the past fifty years describes, chiefly, a search for the single most effective "method" of optimizing learning while standardizing and , hopefully, minimizing teaching, together with a quantity of experimentation whose results have often been ambiguous or too specific to lend themselves to generalization (p. 3 ) . ' Indeed, the history of language teaching i n recent years has entailed a variety of attempts to adapt models from formal linguistics or from formal learning theory to the practical needs of language teachers and language learners. The attempts have not been enormously successful, and they have been made despite warnings from theorists that their formal m�dels were not applicable to the needs of the language teaching profession. Strevens uses an eclectic approach . In this book, he has not ventured a 'full' definition of 'applied linguistics' ; he has, however, done so in other instances: notably in his paper entitled 'On Defining Applied Linguistics > � and in another slightly longer paper which will appear in On the Scope of Applied Linguistics . l Here, he contents himself with a more general statement : ' In the past decade there have emerged a number of important issues and concerns in applied linguistics, language-teaching methodology and the teaching of English . . . (p. xi) . ' There i s , however, a brief chapter ' O n Defining Applied Linguistics , ' i n which he writes:

1 78

REVIEWS

(Received October 1979)

Reviewed by

R. B.

KAPLAN

University of Southern California

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

published i n various sources during the decade between 1 966 and 1 976. The bibliography cites approximately 120 items of which more than half are from the decade between 1 964 and 1 974 and of which the great preponderance are British (rather than American). While there is no question that the book is intended for teachers, it seems unlikely that it would be equally successful on both sides of the Atlantic. The section which compares British and American methodologies in ELT gives the impression that there may be -greater weaknesses on the American side; for example, in summarizing current British practice, Strevens claims that the British approach is 'pragmatic, not dogmatic' (p. 59), implying that the American approach is therefore dogmatic. There are in the following pages a number of comparable implications, all of which would tend to make American teachers somewhat uncom fortable with the boo k . It is, of course, not surprising that Strevens should see greater strength on the British side; after all, he has spent a great part of his career working with British teachers or British-trained teachers in domestic and foreign situations in which the British variety of EFL/ESL was being taught. His experience in the United States is somewhat narrower, while his global experience is indeed impressive. In general, much of what Strevens has to say is informative, interesting, useful to teachers, and sensible. The possible exception lies in the part of the book devoted to varieties of English. While there is no question that teachers of English should have some clear notion of the differences among the major dialects of English, and while the distinction he draws between dialect and accent is a useful one for teachers, there may be just too much discussion of varieties of English for the sort of audience for whom the book seems in all other instances to be intended. Though Strevens worked on the preparation of English 901 , derived from English 900, which required thorough awareness of differences between the two dialects, and though he has been involved in a number of other scholarly differentiations between the two dialects, • some of the highlighted differences in section 12 (pp. 148-1 54) seem trivial while some differences (particularly those characteristic of less formal speech) do not get mentioned; e.g. , the use of go in all of its inflected forms as a synonym for say in the less formal speech o f many Americans under 25 years of age-'I threw the chalk, and then my teacher goes, "Helen, you stop that ! " '-a form that seems not to occur in any dialect of British English. There is also a tendency to discuss American educated English as though it were a monolith; in fact, there are several varieties of geographic and social 'accents ' in standard American English, and there may be more dialect variety even in the standard than at first appears possible. But the book is a most useful one, despite these minor weaknesses. Its approach is intelligent and humane. Its eclecticism is reasoned; it does draw on theory and prin­ ciple, though it draws on no one theory . It derives the best available information from psycholinguistic and sociolinguistic theory as well as from educational learning theory without being dogmatic. It succeeds in being pragmatic without being chaotic. Indeed, it is an entirely rational book. There is no question that there has been a long-standing need for such a book. It will be useful not only in the United Kingdom but in all those places in the world where the British metropolitan model is taught and where teachers are trained to teach it. But I remain unconvinced that the orientation is in fact new. Finally, not only does this book present aJucid orientation to the teaching of English as a second/foreign language, but the presentation itself is lucid. Here is a difference between British and American activity. There are a number o f useful books in the United States, but (as an unfair generalization) they are s o overweighted with educational and linguistic jargon that they are not always lucid. It is indeed a pleasure to find a book the reading of which is so effortless.

REVIEWS

179

REFERENCES ' John Locke, Some Thoughts Concerning Education. 1 692.

' Proceedings of the Fourth International Congress of Applied Linguists, Vol. I, 1 976, Stuttgan: Hochschul Verlag. 1 '

Ed. by Kaplan, Robert B . , Rowley, Mass. : Newbury House, Forthcoming. British andAmerican English, New York: Collier Macmillan, 1972.

MALCOLM COULTHARD, An Introduction

London : Longman, 1 977. Pp. 195 .

to Discourse Analysis

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

I wouldn 't have thought i t possible, i n a slim little volume like this, t o review most o f the major current lines o f work in discourse analysis. Much less t o relate i t all t o the teaching of English as a foreign or second language (TEFL or TESL, depending on which side of the Atlantic you are on). But Malcolm Coulthard has done a good job of it, simply and without much fuss. Following a bi-ief introduction, there are three chapters dealing respectively with work on speech acts, the ethnography of speaking, and conversational analysis . These chapters are organised in terms of a particular logic, which is finally made explicit in the next chapter (on classroom interaction), to wit, Sinclair's f-our criteria for adequate linguistic descriptions: ( 1 ) 'the descriptive apparatus should be finite ' ; (2) 'the whole of the data should be describable ' ; (3) 'there must be at least one impossible combination of symbols' ; (4)_'the symbols in the descriptive app_ara!us shmlld be precisely relatable to their exponents in the data . . . if we call some phenomena a "noun" or a "repair strategy" . . . we must establish exactly what constitutes the class;;,. i th that label' (pp. 98-99). The failings Coulthard identifies in each of the 'schools' described in the first three chapters thus turn out to revolve around one or other of these criteria. The speech act philosophers and those linguists who have followed them closely are criticised for using only invented examples and for lacking a finite set of categories; the ethnographers of speaking are faulted for treating mainly special, marked cases of language use rather than everyday interaction; and the conversational analysts, though much lauded for their contribution, are twitted for a 'lack of explicitness and for­ malisation' (p. 92), i.e., for not operationally defining things like 'repair strategies ' . This kind o f criticism, though certainly relevant for anyone wanting to apply the work of these scholars to an actual analysis of, say, what goes on in a classroom, is not completely fair, since such ready application was never a goal envisaged by most of those whose work is discussed. In particular, many of them would surely reject the idea that their work attempts to provide the kind of descriptive apparatus that Sinclair's criteria are meant to evaluate. Nor am I convinced that progress would have been made by these people had they followed Sinclair's strictures. Vices, by these lights, may well be virtues. Rather, each group of workers has been trying to solve particular and fairly distinct problems having to do with the interface between speech, language and social action. This is not to say that Coulthard is unaware of these different goals or of the difficulty of the problems. Indeed he frequently stresses both. But it may be inherent in any short treatment of discourse analysis to make it seem too straightforward. And any slim book that reviews speech acts, the ethnography of speaking, and conversational analysis in eighty pages is probably doomed to mislead as to the unity of these efforts or their sufficiency to constitute a coherent • field ' . Having clearly adopted descriptive-analytical goals, the two succeeding chapters more explicitly introduce topics in terms of tools for analysis . Thus with chapter five, which presents Sinclair's schema for the analysis of classroom interaction, and chapter six on intonation. The last three chapters have somewhat the status of 'substantive topics ' , though they are well integrated into the whole. Chapter seven, which deals with

REVIEWS

180

Reviewed by GILLIAN SANKOFF University of Pensylvania

(Received September 1979)

TERENCE HAWKES,

Structuralism and Semiotics

London: Methuen, 1 977. Pp. 1 92.

ROGER FOWLER, Linguistics and the Novel

London: Methuen, 1 977. Pp. xiii

+

1 45 .

E . L . EPSTEIN, Language and Style

London: Methuen, 1 978. Pp. x + 92. The books under discussion are the first three in the Methuen New Accents series. Their purpose according to the general editor's preface is to widen the boundaries of literary study by making its readers aware of new ways of approaching literature, many of which spring from linguistics . The readership would thus seem to be students and critics with an interest in, but little specific knowledge of, linguistics and related fields. Structuralism and Semiotics is mainly about structuralism. It begins with an in-

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

discourse analysis and language teaching, i s the most clearly 'applied' chapter, but Coulthard succeeds in maintaining the theoretical concerns of earlier chapters by reintroducing the problem of language functions. Function is discussed again in chap.ter eight, on the acquisition of discourse, but this important problem is, to my mind, nowhere completely resolved. Coulthard had introduced the question at the very beginning, claiming that 'the final problem for discourse analysis is to show how the functional categories are realised by formal items-what is the relationship between "request" or "question" and the grammatical options open to the speaker' (p. 8). Perhaps this problem escapes resolution partly because the prior question, that of validating whatever functional categories are chosen, is so difficult. A short, provocative, whimsical chapter on literary discourse is the last in the volume. Happily, it has no long, recapitulative conclusion, which would have been unnecessary in a book each of whose chapters is so succinct. Certainly, researchers in any one of the areas dealt with may find slight inaccuracies or infelicities. For example, not all of those mentioned on p. 52 as ethnographers of speaking would likely identify themselves as such. The work, however, is planned as an introduction , and it does an admirable job of making some complex concepts ac­ cessible to the beginner. There are a few terms in a few places that may give problems, e.g., 'occasionally constructed data' on p. 8, well in advance of the chapter on con­ versational interaction; 'paralinguistic overloudness' on'p. 62, prior to the chapter on intonation; and 'primary delicacy' on p. 25, not explained until p. 1 05 . (None of these occur in the index, so the reader just has to wait.) The other major irritation to a beginning student reading this book will surely be the incompleteness of the bibliography. At least a dozen of the works referred to in the text are not to be found in the bibliography. The index, however, is in somewhat better shape, containing only two or three author omissions, and the text as a whole is remarkably free of printing errors. In sum, Coulthard succeeds in covering a remarkable amount of material. Even specialist readers will probably find some information that is new to them . The book is coherent, well organized, and successfully unites practical and theoretical concerns. ' But by doing such a good job, it seems destined to be widely used, and as such may be courting another danger. In selecting particular lines of work for inclusion as part of a survey of discourse analysis, it constitutes that particular. combination of sources into the legitimate 'field' . Doing this early on in such a rapidly growing and changing area may fix the directions too soon, making it harder for people to innovate or to seek out non-legitimated sources.

REVIEWS

180

Reviewed by GILLIAN SANKOFF University of Pensylvania

(Received September 1979)

TERENCE HAWKES,

Structuralism and Semiotics

London: Methuen, 1 977. Pp. 1 92.

ROGER FOWLER, Linguistics and the Novel

London: Methuen, 1 977. Pp. xiii

+

1 45 .

E . L . EPSTEIN, Language and Style

London: Methuen, 1 978. Pp. x + 92. The books under discussion are the first three in the Methuen New Accents series. Their purpose according to the general editor's preface is to widen the boundaries of literary study by making its readers aware of new ways of approaching literature, many of which spring from linguistics . The readership would thus seem to be students and critics with an interest in, but little specific knowledge of, linguistics and related fields. Structuralism and Semiotics is mainly about structuralism. It begins with an in-

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

discourse analysis and language teaching, i s the most clearly 'applied' chapter, but Coulthard succeeds in maintaining the theoretical concerns of earlier chapters by reintroducing the problem of language functions. Function is discussed again in chap.ter eight, on the acquisition of discourse, but this important problem is, to my mind, nowhere completely resolved. Coulthard had introduced the question at the very beginning, claiming that 'the final problem for discourse analysis is to show how the functional categories are realised by formal items-what is the relationship between "request" or "question" and the grammatical options open to the speaker' (p. 8). Perhaps this problem escapes resolution partly because the prior question, that of validating whatever functional categories are chosen, is so difficult. A short, provocative, whimsical chapter on literary discourse is the last in the volume. Happily, it has no long, recapitulative conclusion, which would have been unnecessary in a book each of whose chapters is so succinct. Certainly, researchers in any one of the areas dealt with may find slight inaccuracies or infelicities. For example, not all of those mentioned on p. 52 as ethnographers of speaking would likely identify themselves as such. The work, however, is planned as an introduction , and it does an admirable job of making some complex concepts ac­ cessible to the beginner. There are a few terms in a few places that may give problems, e.g., 'occasionally constructed data' on p. 8, well in advance of the chapter on con­ versational interaction; 'paralinguistic overloudness' on'p. 62, prior to the chapter on intonation; and 'primary delicacy' on p. 25, not explained until p. 1 05 . (None of these occur in the index, so the reader just has to wait.) The other major irritation to a beginning student reading this book will surely be the incompleteness of the bibliography. At least a dozen of the works referred to in the text are not to be found in the bibliography. The index, however, is in somewhat better shape, containing only two or three author omissions, and the text as a whole is remarkably free of printing errors. In sum, Coulthard succeeds in covering a remarkable amount of material. Even specialist readers will probably find some information that is new to them . The book is coherent, well organized, and successfully unites practical and theoretical concerns. ' But by doing such a good job, it seems destined to be widely used, and as such may be courting another danger. In selecting particular lines of work for inclusion as part of a survey of discourse analysis, it constitutes that particular. combination of sources into the legitimate 'field' . Doing this early on in such a rapidly growing and changing area may fix the directions too soon, making it harder for people to innovate or to seek out non-legitimated sources.

REVIEWS

181

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

formative and reasonably accurate account of the way i n which structuralism grew out of linguistics by way of anthropology. This is followed by a long middle chapter on the structures of literature. There is a general 27 page chapter on semiotics before the . conclusion. As a general introduction to structuralism related to literature this book is pretty good. It is lucidly written and has an extensive bibliography, and I will certainly recommend it (with some reservations of detail) to students as a good general survey. However, as Hawkes himself acknowledges, there are a number of good introductory books on �he market already, notably Culler ( 1 975), Jameson ( 1 972) and Scholes ( 1 974), all of which he draws on . Given this fact I hoped for some extended application of structuralist analysis to literary texts, but found very little. Moreover, Hawkes's treatment could usefully have been more critical than it is. For example, he gives a clear account of Levi-Strauss's plot analysis of the Oedipus myth, but does not notice that in order to complete the four-term homology which his theory needs, Levi-Strauss in­ troduces character names as well as plot elements into the analysis. This lack of criticism becomes more apparent as the book moves from relatively concrete ap­ proaches, like that of Greimas and Todorov, which stand some reasonable chance of being applied by other analysts, to the inherently vague work of writers like Barthes. Fowler's Linguistics and the Novel is the most interesting of the books I have examined. It does not pretend to be a general account of linguistic approaches to prose fiction (for example it explicitly omits a treatment of speech presentation). Instead, it takes the deep/surface distinction from transformational linguistics as its starting point and suggests that as agency, proposition type and modality are assigned at deep structure, the examination of the-deep-structure of sentences can provide an- insight into the particular world which is being portrayed (world view) and also into the 'person' (e.g., character, narrator) through whose eyes one looks into that world (mind style). An obvious example is the difference between John rang the bell and The bell rang, where the former specifies a human agent but where the latter allows an inter­ pretation where the bell rang itself. An examination of the transformational distortion of underlying structure into different surface structure alternatives, on the other hand, gives an insight into what Fowler calls perspective on meaning (cf., the way in which the active/passive choice allows a writer to change the topic and focus structure of the same proposition). More interestingly, Fowler also suggests that the use of trans­ formational alternatives changes connotational or implicational meamng. For example, in discussing a passage from James's The Ambassadors he notes that the nominalisation of action verbs 'is a standard convention in discourse contexts like this, and has obvious connotations of inactivity, repression of the agency function, reduction in the strength of the will in those human characters to whom this style is applied' (p. 1 1 1 ). Observations of this kind will of course be familiar to stylisticians . They have been made in informal discussion for some time. But Fowler's is a welcome introductory treatment in an area where even the more advanced literature is rather sparse. Moreover, he does analyse extended passag-es in some detail and with a deal of insight. Perhaps his most interesting new contribution is his development of the notions of mind style and world view from Halliday's ( 1 9 7 1 ) work on The Inheritors. My main criticism of Fowler is that he bites off more than he can chew. Linguistics and the Novel is only 1 45 pages long. But besides the work in areas already discussed he includes a final chapter on the novelist, the reader and the community, and spends some time looking at the structuralist 'speculative extension' of applying sentential analysis to whole novels analogically, for example by suggesting that componential lexical analysis might be applied to character types and that particular characters might, a Ia Todorov, be generally felt to be, say, 'active' or 'passive' by virtue of the fact that they typically turn up in the appropriate case slot in the underlying structure of the sentences of the relevant text. There may well be something in this, but it would need a very detailed treatment to be convincing, detail not possible in the scope of such a short book. As this discussion partly overlaps with Hawkes's account of Todorov

REVIEWS

1 82

(Received August 1979)

Reviewed by M. H. SHORT University of Lancaster

R EFERENCES ' Culler, Jonathan, 1 975. Structuralist Poetics: structuralism, linguiStics and the study of literature. London: Routledge and Kegan Paul. 2 Jameson, Frederic, 1 972. The Random House of Language. Princeton and London: Princeton University Press. 1

Scholes, Robert, 1974. Structuralism in Literature: An Introduction. New Haven and London:

Yale University Press.

Downloaded from applij.oxfordjournals.org by guest on January 1, 2011

and Greimas, one wonders whether Fowler might have been wiser t o use his space for more detailed exemplification and discussion of the issues referred to above. The reason for the smallness of the volumes in this series is presumably because of a wish to produce relatively cheap books for student consumption. However, Epstein's Language and Style is only 92 pages long, including index etc . , and seems overpriced at £2.25, particularly when one reads the opening chapter on style, which is so confusing and confused that I do not have the space to untangle it here. Fortunately, chapter 3 , ' Playing the Literature Game' stands u p without the introductory remarks. Here, by the use of detailed analysis of examples, Epstein demonstrates at successive linguistic levels how structure can imitate and embody the sense of what is said. There are a number of points with which one could quarrel; for example some of the phonetic correlates with the meaning of poetic lines which Epstein suggests do not work quite so well when transferred from his American to an English accent. However, he does show that the iconic role of linguistic structure deserves more attention than it has received of late. Overall, New Accents is an interesting and at times stimulating series. But it would be unwise to set one's students or critic friends to reading without allowing plenty of discussion time for clearing up disagreements and filling out accounts which have been telescoped because of space limitations .