Man, Language and Society: Contributions to the Sociology of Language 9783110905281, 9789027921208

Table of contents :
PREFACE
CONTENTS
1. 'LANGUAGE' OR 'DIALECT'?
2. A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING
3. SOCIOLINGUISTICS : AN APPROACH TO FUNCTIONAL GRAMMAR
4. SITUATIONAL MEASURES OF NORMATIVE LANGUAGE VIEWS IN RELATION TO PERSON, PLACE AND TOPIC AMONG PUERTO RICAN BILINGUALS
5. L'ÉVOLUTION ACTUELLE DES RECHERCHES SUR LA PATHOLOGIE DU LANGAGE
6. VÖLKER, SPRACHEN, MUNDARTEN
7. THEME OF INCEST IN NAVAHO FOLKLORE
8. CONCEPT CATEGORIES AS MEASURES OF CULTURAL DISTANCE
9. THE CURRENT STATE OF SOCIOLINGUISTICS IN THE UNITED STATES
BILINGUAL EDUCATION FOR CHILDREN OF LINGUISTIC MINORITIES
11. STRUCTURAL ANALYSIS IN MATERIAL CULTURE
12. REGARDLESS OF FRONTIERS: A CASE STUDY IN LINGUISTIC PERSECUTION
13. COMMUNICATION PERSPECTIVES FOR SOCIOLINGUISTICS
14. PROXEMICS AS NON-VERBAL COMMUNICATION
15. MAN, LANGUAGE AND SOCIETY

Citation preview

JANUA

LINGUARUM

STUDIA

MEMORIAE

NICOLAI VAN WIJK DEDICATA edenda curai C.H. VAN

SCHOONEVELD

Indiana University

Series

Minor,

109

MAN, LANGUAGE A N D SOCIETY CONTRIBUTIONS TO THE SOCIOLOGY OF LANGUAGE

edited by

SAMIR K. GHOSH Indian Institute

of Human

Sciences

1972

MOUTON THE HAGUE • PARIS

© Copyright 1972 in The Netherlands Mouton & Co. N.V., Publishers, The Hague No part of this book may be translated or reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publishers

LIBRARY OF CONGRESS CATALOG CARD NUMBER: 70-169995

Printed in Belgium, by NICI, Printers, Ghent.

Dedicated to: William Bright University of California at Los Angeles Charles A. Ferguson Stanford University John J. Gumperz University of California at Berkeley Dell H. Hymes University of Pennsylvania teachers, friends and colleagues. S.K.G.

PREFACE

Probably no one likes to read a preface, but a writer or an editor often needs to state certain facts or clarify some of information in order to defend his views, and these remarks are most appropriately placed in the Preface. This volume was originally conceived as a joint festschrift for two well-known anthropologists, one in Paris and the other in Chicago. The idea had to be withdrawn for a variety of reasons, the primary one being that the original publisher (a Dutch organization) kept the project in cold storage for almost a year and then almost abandoned it. As a result, some of the papers, concerning social anthropology as well as sociolinguistics, were returned to the contributors with regrets from the editor. Fortunately, at this juncture the editor took a chance and wrote to Mr. Peter de Ridder of Mouton and Co., asking if he would be interested in publishing the volume with only articles about the sociology of language. Mr. de Ridder's action was very prompt; he not only saw the possibility of such an undertaking, but also encouraged its publication. New contributions were quickly sought out again — this time only from sociolinguists. The editor would like to acknowledge, first and foremost, that this volume has come into print ONLY with the help and active interest of Mr. de Ridder and his assistant, Mr. Dick Coutinho. Mr. Coutinho, a trained linguist, has been most helpful throughout. Unfortunately, even with the cooperation of Mouton and Co., the project has been kept in suspension for a period of more than a year and a half.

8

PREFACE

Under normal circumstances, this volume would have appeared in December of 1969. The editor feels that in a growing field, such as the sociology of language, in which new faces and articles appear almost monthly, many of them presenting new studies and proposals, a publication such as this one can lose some of its value and importance by appearing a little too late: time is both money and knowledge. Strictly speaking, this book aims at making a contribution to a promising new field, LANGUAGE SOCIOLOGY. Most of the contributions deal with language sociology (exceptions are Landar and Rohan Csermak); most of the papers are being published for the first time (Bernstein [revised] and Modiano are reprints); and most are written in English (exceptions: Hecean in French and Kloss in German). Naturally, the views of the sociolinguists presently active in the field are extremely diverse, but there are many points about which they agree. The agreements and disputes among the contributors will be apparent to the reader and will not be discussed here. The editor has refrained from annotating the papers a la mode Martin Joos or elaborating their views a la mode Dell Hymes, (this also falls within the Joos law of paraphrasing). The editor had very much hoped to present a survey of language sociology up to 1969, but neither time nor space was permitting. The language sociology bibliography which was prepared by the editor (with the help of Dr. Martha Allen) was rejected by the publishers. Fortunately, the editor is at least permitted to make a few observations. The impetus for studying and advancing the theories in language sociology came from amateur interests as well as from a feeling among scholars that expert knowledge of areas critical to the interests of language-society specialists was lacking. Even now, only a handful of specialists have been attracted to or trained in the lore of language sociology — whether within or outside the university departments of linguistics, sociology, anthropology, or psychology. Consequently, when a scholar has difficulty understanding or analysing social phenomena, he has no one to turn to

PREFACE

9

for discussion or explanation of the problem (and no field can exist without this kind of dialogue). Only recently have some scholars given their full time to teaching and research aimed at overcoming the demonstrated inadequacies of study in this area — and these scholars still form a minority. Language sociology is without a doubt an underrated field and it suffers all the disadvantages of this status. Very few people can, at this time, imagine that language sociology can ever become a discipline in itself; on the contrary, language sociology is only an instrument for motivating and integrating the allied fields of sociology, anthropology, linguistics, psychology and even political science and history. Academicians strive for achievement in their 'major' discipline; the courses offered to them in language sociology (other than their own) are seen as peripherals, unequal to those in their basic discipline. This kind of program means that a linguist has to do all exploring alone, without guidance or background, when he tries to systematically examine the fundamentals of phonology (taxonomic or generative), morphology, syntax, semantics, linguistic theories (Chomskian or Stratificational grammar), acoustics, and even the study of some classical and modern languages, and at best field work in tribal (Amerindian) languages. Only by combining all these disciplines can one expect to have a basis for understanding linguistics (and the same is true of anthropology and sociology). In short, in its present stage teaching and research in language sociology has the task of encouraging the various traditional university departments to offer courses which combine to form an integrated study program, thus insuring students of familiarity with the 'peripheral' disciplines, so that the ultimate aim of permitting them to make full use of the materials their 'major' discipline offers them can be fulfilled. This present state of affairs, i.e., the lack of integrated study, is indeed lamentable. Let me repeat that no student is expected to become a 'language sociologist'; he should become as expert as any other linguist, anthropologist, or sociologist in using the tools of his 'primary trade'. This is made clear to the newcomer in the

10

PREFACE

field, who is soon informed that if he ultimately seeks employment as a teacher (or with the government) he will have to be able to teach in one of the 'primary' fields, whatever his particular one may be. There is another factor which influences the status of language sociology and is equally devastating to its ideas: the more the disciplines improve their tools of analysis, i.e., their technical skill, the more the 'art' of language sociology may lose its appeal. The 'heroes' of these disciplines are no longer the 'observers', such as von Humboldt, Vico, Boas, Sapir, and other notables of earlier days who made their reputations by 'sensing' the direction of language sociology on the basis of massive experience with its phenomena. In today's terms these great names would be 'human computers', in that they took masses of data into their systems, digested it in terms of life-time experience, experimented with similar (or nearly similar) constellations of facts, and finally spewed forth an analysis which seemed sufficiently close to reality to be acceptable to their readers: — this analysis was often a 'perceptive study'. Another disadvantage is that language sociology does not have a 'new hero' in the sense that anthropology has Claude Lévi-Strauss. Furthermore, it suffers from not yet having formed any distinct schools, as is the case in the modern linguistics of Chomsky, Firth, Hjelmslev, Lamb, Pike, etc. (it does not even have a Martinet). Thus, lacking a guru as guide, language sociology lacks followers, adherents or disciples as such. The study of language sociology could be regarded as backwater, for men seeking advancement in their respective professions (except for some cultural anthropologists) shun it; and in considerable measure their criticism is justified. Put in an interesting way, this criticism is due to man's OWN failure to understand the role of language in society, that is, social communication. A growing amount of thought in the social sciences is very much aware that the 'man-animal' cannot be put into proper perspective unless we have real knowledge of the 'language animal' and his interaction with the society he lives in. Furthermore, it is agreed that language sociology should explore the various ways in which 'linguistic'

PREFACE

11

aspects of social life can be studied, that is, language sociology should distinguish the STRICTLY LINGUISTIC from the LINGUISTICALLY RELEVANT and these two in turn from the LINGUISTICALLY CONDITIONED.

Though language sociology is a relatively new field of specialization, it is developing with extraordinary speed. Some remarkable papers and anthologies discussing the problems in language sociology have already been published. The present work complements these publications and provides both students and teachers throughout the world with description and direction for research programs and aims, at a time when they are relatively new. The purpose of this book is thus to ENCOURAGE the individual to THINK about these issues, to question stereotypes and to be aware of the future and fate of language sociology, which is being created daily by actions and decisions by language animals everywhere on earth. Like most new fields of specialization, language sociology's theoretical structure is still very malleable, which gives rise to much theoretical contention. These theories are diverse and often uncharted, consequently we will not discuss their integration here. To solve these problems, language sociologists will not only have to engage in continuous interdisciplinary discussions with anthropologists, linguists, psychologists, historians, and phenomenologists, but will also have to formulate new theories of social action and structural time and delve into the sociology of knowledge. The topics which should engage their immediate attention are many: among them are the scope and content of language sociology, its foremost ideas and conceptual framework, its relation to other social sciences and history, as well as to macro- and micro-sociolinguistics, individual, social, and national aspects of language behavior, languages' role in the structure and performance of traditional (pre-colonial and pre-industrial) as well as colonial and post-colonial (for Japan, and for Europe, postfeudal) societies as they relate to development, ecology and modernization, and lastly, methods and models for viewing language in society.

12

PREFACE

May we remind ourselves of an important point by quoting Raymond Aron: With language, that form of communication by symbols which makes possible the preservation of accumulated experience and an accompanying sense of values—in short, with awareness and moral consciousness—a new dimension of history appears, a dimension that includes the plurality of more or less incompatible cultures and the search for existential truth. (Progress and Disillusion, p. 2Î3J

The editor believes that language sociology is an American contribution to the body of human enquiry. Therefore, he has the honor of dedicating this volume to four leading American sociolinguists. He is perhaps most indebted to them and other acclaimed sociolinguists for being able to present this volume, however needless to say, they are not responsible for what he has done with their ideas. The Institut zur Erforschung der UdSSR in Munich has provided me with time and leisure not usually available to a university teacher. I am grateful to its personnel and to my colleagues on the staff there, particularly Dr. Heinrich Schulz, Director, and Mr. Edward Crowley, Deputy Director. The editor would like to take this opportunity to put in writing his sincere respect for his mother, Sm. Sailabala Ghosh, for her patience, encouragement and understanding of his work. He would also like to thank Dr. Martha Allen (Geneva), Prof. Joshua Fishman (Yeshiva), Frau Inge Hais (Greiz), Dr. David Nissman (Munich), Prof. Frederick Rose (Berlin), Mile Marie-Claude St.-Hilary (Paris), Miss Jamuna Subrahmaniyam (Delhi), and Dr. Tan Wen (Santiniketan). Reviewers may find defects in this volume, but if it serves as an apéritif for future and further research in language sociology, the editor will be amply rewarded. He feels that the major need in language sociology is to generate a new will — the will to think and work, to tax ourselves until we are able to meet the vital needs of the sociolinguistic community. Charaibeti !

PREFACE

13

In closing the editor wishes to repeat his sincere thanks to the staff of Mouton & Co. for their helpfullness and efficient assistance. October 2, 1970 Munich, West Germany Samir K. Ghosh

CONTENTS

1. OLGA AKHMANOVA

'Language'or'Dialect'?

17

2 . BASIL BERNSTEIN

A Sociolinguistic Approach to Social Learning

23

3 . MARSHALL DURBIN

Sociolinguistics : An Approach to Functional Grammar .

51

4 . LAWRENCE GREENFIELD AND JOSHUA A . FISHMAN

Situational Measures of Normative Language Views in Relation to Person, Place and Topic Among Puerto Rican Bilinguals

64

5 . H . HÉCAEN

L'évolution actuelle des recherches sur la pathologie du langage

87

6 . HEINZ KLOSS

Vôlker, Sprachen, Mundarten

103

7 . HERBERT LANDAR

Theme of Incest in Navaho Folklore

118

8 . WILLIAM F . MACKEY

Concept Categories as Measures of Cultural Distance . .

134

16

CONTENTS

9 . MADELEINE MATHIOT

The Current State of Sociolinguistics in the United States

169

10. NANCY MODIANO

Bilingual Education for Children of Linguistic Minorities

179

11. G . DE ROHAN-CSERMAK

Structural Analysis in Material Culture

190

12. VICTOR SADLER a n d ULRICH LINS

Regardless of frontiers: A Case Study in Linguistic Persecution

206

13. HARVEY SARLES, GAIL BENJAMIN a n d JEANNE L . SWOPE

Communication Perspectives for Sociolinguistics . . . .

216

14. O . MICHAEL WATSON

Proxemics As Non-Verbal Communication 15. SAMIR K .

224

GHOSH

Man, Language and Society

232

1

'LANGUAGE' OR 'DIALECT'?

OLGA AKHMANOVA

The reason why it still remains a vexed question is that natural human languages are not 'closed' semiotic systems or codes, they are not 'immanent', and they do not function merely as sets of underlying rules to be applied in the construction of messages. Their ontology can be properly understood only if it is fully realized that they exist because there exist groups of human beings, 'speech-communities', who keep communicating with each other by means of certain signs, the latter being what they are because they are so employed. If it had not been for the continuous exchange and the ensuing mutual influence and mutual dependence of the innumerable speech events, there would be no languages. The unity of a language as a semiotic system evolves because the various particular linguistic messages interact not only on the content, but on the expression plane as well. It should be added that the interlocutors' using the same language is a direct indication of their belonging to the same social group, while the existence of variant forms within the same language points to the existence of social subgroups, ethnical subdivisions within the larger or overall speech-community. The different subgroups differ widely as to the place they occupy within the total social structure. It follows naturally that the relative weight of the particular realizations of the overall linguistic unity will vary in accordance with the hierarchy of social relations.

18

OLGA AKHMANOVA

is normally apprehended by children not merely as one of the possible or many acceptable variants, but as the variant, because to the child his parents are not members or representatives of a social group; they are representatives of the community as a whole. They are idealised as carriers of the whole semiotic system in question. The tables may be turned of course as when, for instance, a grown-up son returns to his native village after a period in town or Army service. The way he speaks now may well affect the parents' ideolects, for now he may have become the bearer of a more influential form of speech — a speaker of a prestige dialect even! In a complex modern society the complexity of social relations is so great that the mutual relationship between the different realisations of a given linguistic system are not only extremely complicated, but often contradictory as well. It is nevertheless possible to abstract from the innumerable particular realisations that which is basic and general to the language, whose objective existence is assured so long as there are people to communicate and interact linguistically (both the communication and the interaction do not have to be always direct and immediate, for there exist a large number of intermediate links, such as writing, the press, the radio, etc.) It goes without saying that nowadays the size of a speech community may be very great; it may even stretch over great distances — what with the different modern systems of telecommunication. A very important factor in the creation of linguistic unity is intensity of communication, based on closeness of social ties, the degree of mutual dependence, determined by the economic and cultural development of the community in question, its political organisation, social attachment, etc. This is the reason why the unity, the fullness of integration, varies so widely both synchronically and diachronically for different languages. A larger social group is normally subdivided into varying numbers of subgroups. This usually manifests itself in the emergence of stable varieties of speech, more or less easily recognisable as distinct from one another. The tendency thus becomes centrif-

'language' o r 'dialect' ?

19

ugal. Instead of increased mutual influence between the different realisations and a persistent levelling out of discrepancies, there coexists in the consciousness of one individual two or more potential realisations (di- and poly-glossy) to choose from: to suit the manifold requirements of place and purpose of communication. Linguistic change being a comparatively slow process, the emergence of new linguistic unities often falls behind, cannot keep pace with the requirements of the new social conditions. Hence the fact of linguistic situations no longer corresponding to the actual socialeconomic and political organisation of Society. Thus, for instance, in many of the great European countries, the linguistic state still reflects — more or less faithfully — the feudal stage in the development of these countries. These relics, however, are by no means a sort of museum pieces! They persist as living linguistic facts; they interact with the emerging issues and are transformed in the process of further social development. However great the variety of social and political systems and the dissimilarity of the underlying historical and other factors may be, there are certain basic features and divisions which are equally applicable to them all. The most important of these is the division into the independent vs. the dependent natural semiotic systems. A 'language' is independent, a 'dialect' dependent. A language exists only in and through its different realisations or manifestations in speech. But as a whole it is independent of them, it is not one of the realisations which together form the overall or supersystem in question. In contrast with a language, a dialect is not independent. The specific realisations (vosproizvedenija v reci) in and through which it exists are mutually correlated, mutually connected (or interconnected) with other dialects. In other words the ontology of a language is fully accounted for by its reproduction in a given ensemble of various manifestations. A s for a dialect there is always a frayed edge, a specific relationship with the adjacent semiotic systems, with the rest of the realisations of the 'language' as the overall semiotic system. The confines of a language are clearly marked off by interruption of interdependence and mutual influence of different

20

OLGA AKHMANOVA

speech-events. In the case of dialects, the different realisations spill over the borderlines without loss of continuity. This blurring of outlines is the reason why dialects may be regarded as 'parts' of language. This does not, however, mean to say that together they form a simple entity — dialects are not fragments but unities within the greater unity of language. Each dialect is the language in question, but such a form of it which is neither realised nor realisable in all its actual or potential manifestations. Thus, for instance, when certain Danish dialects in the second millennium of our era lost their former allegiance and became dialects of English, this did not result in mutilation of the parent language its «dismemberment», although some important changes or did take place in the overall systems of both English and Danish. It should be noted, in this connection, that the relationship may be further complicated by the fact that the dialect of one language may be raised to the status of a language when confronted with a different linguistic system. This was, for example, the case with Anglo-French. The fact that it was this particular dialect of French that was responsible for the 'frenchification' of English had, of course, far-reaching historical and sociolinguistic consequences — the French element was assimilated not only in a specific territorial-dialectal shape, but in a sociodialectal one as well, insofar as it was the dialect of the upper strata of the feudal society of the time. The sociolinguistic aspect of all the above relationships and processes is of the greatest significance — so much so that there is every reason in general to regard the whole question as a sociolinguistic one. It has already been stated at the beginning of this article that the existence of a language is in itself a proof of social unity. But there are always not only dialects, but ideolects as well. No two people, even when they belong to the same social group, speak in exactly the same or identical way. How do they then react to the inevitable absence of complete identity? By trying to minimize and belittle the dissimilarities. If this is the case, then the mutual influence, the mutual dependence of the respective realisations takes the form of adaptation. In other cases the influence will

'language' o r 'dialect'?

21

have to be measured in terms of relative social position and prestige. Between the dialects of a language there is normally a considerable similarity. This however is no criterion in itself, for the similarity of two different languages may be no less than that of two dialects of one and the same language. Thus for instance, the Anglo-Saxon language was no less similar to continental Saxon than some of its Alemanic dialects. It was, nevertheless, a separate language — a fact amply proved by its subsequent historical development. It is therefore essential to draw a distinct line between unity and uniformity, i.e., the mutual dependence, the interconnection of different realisations vs. their outward similarity. The latter may be due to extralinguistic factors, including parallel development, or remain as a mere relic of a past state. To establish the unity of a linguistic system, a statement of mere uniformity is never sufficient. It may prove useful only provided the history of the synchronic state of affairs, the underlying historical, and social and linguistic factors, has been fully elucidated first. From what has been expounded above it might appear that the relationships between different natural semiotic systems are of three kinds: (1) between dialects of the same language (2) between two different languages and (3) between a language and one of its dialects. But this is not so, for (3) is not a relationship between two semiotic systems. It is a relationship within the same unity, within one semiotic whole. To sum up: a language is a global semiotic unity which is independent of all the linguistic realisations, of all speech events which do not belong to the particular system or ensemble of utterances interacting within its confines. In contrast with a language, a dialect is a dependent semiotic system. It achieves unity only through condensation, within the larger unity of language, of certain specific forms of linguistic expression. The borderlines between dialects are fuzzy and set up by the weakening of mutual influence and interaction between the different realisations of the overall system. Both the weakening and strengthening are conditioned by sociolinguistic factors. If a social group is becoming iso-

22

OLGA AKHMANOVA

lated, the dialectal differences are enhanced. If, on the other hand, twoor more social groups merge into one the dialectal differences may be dropped altogether. Department of Linguistics Moscow State University, 1968

2

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING • BASIL BERNSTEIN

1.1

This paper is concerned with: (1) The neglect of the study of speech by sociologists; (2) The role of speech as a major aspect of culture and the means of its transmission ; (3) The relations between forms of speech and forms of social relation ; (4) The social and educational consequences of differential access to forms of speech. The reader may well think that the early discussion bears little relation to education. It is important, however, although the argument is a complex one. Perhaps one of the most important events that has taken place in scientific endeavour in the twentieth century is the convergence of both the natural and social sciences upon the study of linguistic aspects of communication. The consequences of this convergence and the new relations between the disciplines which it has brought about may well be worthy of a chapter in the next book on the sociology of knowledge. Through the study of language the link between biological and socio-cultural orders is gradually being established. The clarification of this link and the resultant theories may well have consequences for control as exciting as the progress in our understanding of the genetic code. This is not the place to * Reprinted with permission from: Social Science Survey, ed. J. Gould (Penguin Book, 1965).

24

BASIL BERNSTEIN

discuss the trends in separate disciplines which have led to this convergence, but a number of works may serve as guides for the reader. 1 What is a little odd is the negligible contribution of sociology to the study of language. The textbooks celebrate the fact of man's symbolic possibilities in chapters on culture and socialisation and then the consequences are systematically ignored. One might go as far as saying that the only time one is made aware that humans speak in the writings of contemporary sociologists is incidentally through the statistical relations induced from social-survey inquiries.2 And here all that is required is that the subjects can read: speech confounds the later arithmetic. Even when what a person says is considered to be relevant, what is actually said is rarely, in itself, singled out as worthy of systematic study. The origins and consequences of forms of saying, linguistic forms, their conditions, formal patterning, regulative functions, their history and change are not included in the sociologist's analysis. And yet long ago both Durkheim and Weber drew attention to the social significance of language. In its struggle for recognition, sociology has continuously insisted upon the fact that there exists an order of relations, arising out of 1

O.S. Akhmanova et al., Exact Methods in Linguistic Research (University of California Press, 1963); C. Cherry, On Human Communication (McGrawHill, New York, 1957); M. Cohen, Pour une sociologie de langage (AlbinMichel, Paris, 1956); E. Gellner, "The Crisis in the Humanities and the Mainstream of Philosophy", in J.H. Plumb, ed., Crisis in the Humanities (Penguin Books, 1964); J.O. Hertzler, "Towards a Sociology of Language", Social Forces 32 (1953) pp. 109-19: D. Hymes, "Linguistic Aspects of Cross-Cultural Personality Study", in B. Kaplan, ed., Studying Personality Cross-Culturally (Row Peterson & Co., New York, 1961); D. Hymes, "The Ethnography of Speaking", in T. Gladwin and W.C. Sturtevant, eds., Anthropology and Human Behaviour (Anthropological Society of Washington, D.C. [A.S.W. Smithsonian Institution], 1962); D. Hymes, Ed., "The Ethnography of Communication", American Anthropologist (Special Issue, December, 1964); G.H. Mead, Mind, Self and Society: From the Standpoint of a Social Behaviorist (University of Chicago Press, 1936); G. Miller, Language and Communication (McGraw-Hill, New York, 1951); S. Saportaed., Psycholinguistics: A Book of Readings (Holt, Rinehart & Winston, New York, 1961). 2 There are, of course, exceptions, forexample,L.Schatzman,andA.L. Strauss, "Social Class and Modes of Communication", American Journal of Sociology 60 (1955) pp. 329-38, and J.H.S. Bossard, "Family Modes of Expression", American Sociological Review 1945, pp. 226-37.

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING

25

the inter-actions of members of a society, which constrains and directs behaviour independent of the unique characteristics of its members. Sociologists have been concerned to explain the nature of this order, in particular the processes making for its diversity and change, and to develop on a formal level a grammar or syntax which controls the conceptualising of this order. They have studied the major complexes of social forms which shape the social order, their inter-relations, and the factors responsible for their change. Language is seen as an integrating or divisive phenomenon; as the major process through which a culture is transmitted; the bearer of social genes. However, this has rarely given rise to a study of language as a social institution comparable to the analyses made of say the family, religion, etc. As far as speech is concerned this has been viewed as a datum, taken for granted, and not as an object of special inquiry. It is, of course, true that through the writings of George Mead the role of language, really the role of speech, has been explicitly recognised in the formation of a distinctly social self. And yet, in the study of socialisation, it is not possible to find an empirical study which systematically examines the role of speech as the process by which a child comes to acquire a specific social identity. In fact, in the numerous studies of child-rearing with the exception of a very few, there is no account of the patterning of the linguistic environment.3 Groups are studied, their formal ordering elegantly discussed, but the implications and consequences of linguistic aspects of their communications seem to be unworthy of sociological consideration. Graduates are trained to conduct surveys, to construct questionnaires, to interview, without, at least, in England, any explicit and systematic training in what Dell Hymes has called the "ethnography of speaking" — although there is an intuitive or unsystematic recognition of differences in the patterning and consequences of speech events in various sub-cultures. Sociologists, who focus upon social dynamics as these are expressed through changes in the major institutional forms, have 3

I am ignoring here the many studies limited to the development of speech in children.

26

BASIL BERNSTEIN

thrown a shadow on problems implicit in the work of the great nineteenth century theorists. Weber, for example, discusses various types of rationality, and their associated institutional orders and forms of authority. Complex societies involve various forms of rationality which may be differentially distributed among their members. Weber's typology of rationality bears some resemblance to cultural themes which determine modes of action. How does an individual come to acquire a particular form of rationality? Weber's concept of rationality requires an explicit formulation of the interrelations between institutional and cultural orders and of the process whereby individual experience manifests itself in special modes of social action. Durkheim's analysis of the origins and consequences of mechanical and organic solidarity pre-suppose the same problem.4 The concept of the individual in Durkheim is reduced to an unstable state of appetities — an instinct-system tending towards disintegration in conditions where the energies are not subordinate to a normative order of a particular kind. His formulation has the distinct merit of stating the problem of the relationship between biological and socio-cultural orders. A major attempt to relate biological, institutional and cultural orders has been made with the use of the writings of Freud. Indeed, much work on socialisation, on the relation between culture and personality, both in anthropology and sociology, implicitly or explicitly attempts a solution of Durkheim's problem in these terms. However, this approach precludes the study of language and speech. As a result of working with the Freudian theory certain elements within the theory limited interest in linguistic phenomena. The gains of this approach are partly outweighed by the tendency to reduce the social to the psychological by means of a theory of unconscious motivation giving rise to an affective theory of learning. Although the ego in psychoanalytic theory is essentially a linguistically differentiated organisation, speech tends to be regarded epi-phenomenally as a process shaped by the patterning of the mechanisms of defence. It is, of course, true that 4

Durkheim tends to leap from types of social integration to the quality of a series of individual acts.

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING

27

in this theory reality-testing is accomplished essentially through verbal procedures, but the patterning of speech is accorded no independence in this theory nor in the behaviour which the theory illuminates. 5 As a result, anthropologists and sociologists who used Freudian theory in their attempts to understand the transformation of the psychic into the social paid little attention to either language or speech, and so carried over into their work the dichotomy between thought and feeling implicit in Freud. Further, the institutional and cultural order are often interpreted in terms of projections of unconscious formations within the individual. It would seem then that sociologists, because of their emphasis on changes in the major institutional forms in industrial society, have tended to neglect until very recently the study of the transmission of culture. Where this has been attempted, for example in the study of socialisation, the influence of Freud has diverted attention from the linguistic environment. The influence of George Mead, who stressed the role of speech in the formation of a distinct social identity, assisted the rise of what has been called interaction theory, but paradoxically not to any special study of the medium of inter-action, i.e., speech. The net effect of these movements has been to weaken the possibility of connexion between sociology and linguistics and the cross-fertilisation of theories and methods between the two disciplines. This neglect of the study of language and speech in sociology has certainly not been typical of a school of anthropologists who have firmly and boldly stated a controversial relation between language and the interpretation of reality. Wilhelm von Humboldt's statement in 1848 that "man lives with the world about him principally indeed... exclusively as language presents it" was echoed by Boas 6 who claimed that a purely linguistic analysis "would provide the data for a thorough investigation of the 5 The major interest has been concerned with symbolism. It is important to note work done in the area of schizophrenic thought disorder and the stress on communication emphasised by the existential school. 6 F. Boas, Handbook of American Indian Languages, Part I (Washington D.C., Government Printing Office, 1911).

28

BASIL BERNSTEIN

psychology of the peoples of the world". However, it was with Sapir, a student of Boas, that a new elegance, clarity, subtlety, and originality was introduced into the discussion of the inter-relations between language, culture, and personality and which has deeply affected all work in this area. Language, according to Sapir7, "does not as a matter of fact stand apart from or run parallel to direct experience but completely interpenetrates it". Hoijer8 succinctly stated Sapir's thesis as follows: Peoples speaking different languages may be said to live in different "worlds of reality" in the sense that the languages they speak affect to a considerable degree both their sensory perceptions and their habitual modes of thought. Sapir writes: Language is a guide to 'social reality'. Though language is not ordinarily thought of as of essential interest to the students of social science, it powerfully conditions all our thinking about social problems and processes... It is quite an illusion to imagine that one adjusts to reality essentially without the use of language and that language is merely an incidental means of solving specific problems of communication or reflection. The fact of the matter is that the real world is to a large extent unconsciously built up on the language habits of the group... We see and hear and otherwise experience very largely as we do because the language habits of our community predispose certain choices of interpretation. 9 Whorf 10 , a student of Sapir, went further and attempted to derive from the morphological syntactic and lexical features of Hopi the "habitual thought" or "thought world" of the people. The thought world is "the microcosm that each man carries about inside himself by which he measures and understands what he can 7

E. Sapir, Encyclopedia of the Social Sciences, vol. 9 (1933) pp. 155-69. H. Hoijer, "The Relation of Language to Culture", in S. Tax, ed., Anthropology Today: Selections (University of Chicago Press, Phoenix Books, 1962). 9 E. Sapir, "The Status of Linguistics as a Science", in D.G. Mandelbaum, ed., Selected Writings of Edward Sapir (University of California Press, 1949). 10 B.L. Whorf, "The Relation of Habitual Thought and Behaviour to Language", in L. Spier, ed., Language, Culture and Personality: Essays in Memory of Edward Sapir (Menasha, Wisconsin Sapir Memorial Publication Fund, 1941); also in J.B. Carroll, ed., Language, Thought and Reality: Selected Writings of Benjamin Lee Whorf (John Wiley & Sons Inc., New York, 1956) • 8

A SOCIOLINGU1STIC APPROACH TO SOCIAL LEARNING

29

of the macrocosm". Hoijer 11 , one of the major interpreters of Whorf, states that the fashions of speaking peculiar to a people, like other aspects of their culture, are indicative of a view of life, a metaphysics of their culture, compounded of unquestioned and mainly unstated premises which define the nature of the universe and man's position within it. This is not the place to follow the many twists and turns of the controversy these writings give rise to, or to examine the empirical support for the theory, but the reader will find in the footnote a guide to this literature. 12 This thesis had repercussions for psychology and has been an important factor in bringing about a relationship between linguistics and psychology. One of the many difficulties associated with it is that it focuses upon universal features of the formal patterning of language. Although Whorf 1 3 insists that the influence of language upon habitual thought and behaviour does not depend so much on any one system (e.g., tense or nouns) within the grammar as upon ways of analysing and reporting experience which have become fixed in the language as integrated 'fashions of speaking' which cut across the typical grammatical classifications, so that a 'fashion' may include lexical, morphological, syntactic, and otherwise systematically diverse means co-ordinated in a certain frame of consistency. These fashions of speaking, the frames of consistency, are not related to an institutional order, nor are they seen as emerging from the structure of social relations. On the contrary, they are seen as determiners of social relations through their role in shaping the culture. In Whorf's later writings, and in the writings of his 11

See Footnote 8. Roger Brown, Words and Things (Free Press of Glencoe, New York, 1958); J.A. Fishman, "A Systématisation of the Whorfian Hypothesis", Behavioral Science 5 (University of Michigan publication 1960) pp. 323-39; H. Hoijer, ed., Language in Culture ( = American Anthropological Association Memoir, no. 79, 1954) (also published by University of Chicago Press); H.C. Triandis, "The Influence of Culture on Cognitive Processes", in L. Berkowitz, ed., Advances in Experimental Social Psychology (Academic Press, New York, 1964). 13 See Footnote 10. 12

30

BASIL BERNSTEIN

followers, it is certain that morphological and syntactic features of the language are made psychologically active through the fashions of speaking which elicit habitual and characteristic behaviour in the speakers. In other words, the link between language, culture and habitual thought is not mediated through the social structure. The view to be taken here is different in that it will be argued that a number of fashions of speaking, frames of consistency, are possible in any given language and that these fashions of speaking, linguistic forms, or codes, are themselves a function of the form social relations take. According to this view, the form of the social relation or more generally, the social structure generates distinct linguistic forms or codes and these codes essentially transmit the culture and so constrain behaviour. This thesis is different from that of Whorf. It has more in common with some of the writings of Mead, Sapir, Malinowski, and Firth. Whorf's psychology was influenced by the writings of the Gestalt school of psychology whereas the thesis to be put forward here rests on the work of Vygotsky14 and Luria 15 . In a sense the Whorfian theory is more general and more challenging; although, perhaps, it is less open to empirical confirmation, for it asserts that owing to the differential rates of change of culture and language the latter determines the former. The thesis to be developed here places the emphasis on changes in the social structure as major factors in shaping or changing a given culture through their effect on the consequences of fashions of speaking. It shares with Whorf the controlling influence on experience ascribed to "frames of consistency" involved in fashions of speaking. It differs and perhaps relativizes Whorf, by asserting that, in the context of a common language in the sense of a general code, there will arise distinct linguistic forms, fashions of speaking, which induce in their speakers different ways of relating to objects and persons. 14

L.S. Vygotsky, Thought and Language (John Wiley & Sons Inc., New York 1962). 15 A.R. Luria and F.I. Yudovich, Speech and the Development of Mental Processes in the Child (Staples Press, 1959); A.R. Luria, The Role of Speech in the Regulation of Normal and Abnormal Behaviour (Pergamon, 1961).

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING

31

It leaves open the question whether there are features of the common culture which all members of a society share which are determined by the specific nature of the general code or language at its syntactic and morphological levels. It is, finally, more distinctly sociological in its emphasis on the system of social relations.

2.1. ELABORATED AND RESTRICTED CODES

A general outline of the argument will be given first. This will be followed by a detailed analysis of two linguistic forms or codes and their variants. The discussion will be linked to the problem of educability as this is conceived in industrial societies.

2.2.

INTRODUCTION

To begin with, a distinction must be made between language and speech. Dell Hymes (1961) 16 writes: Typically one refers to the act or process of speech, but to the structure, pattern or system of language. Speech is a message, language is a code. Linguists have been preoccupied with inferring the constants of the language code. The code which the linguist invents in order to explain speech events is capable of generating n number of speech codes, and there is no reason for believing that any one language or general code is in this respect better than another, whether it is English or Hopi. On this argument language is a set of rules to which all speech codes must comply, but which speech codes are generated is a function of the system of social relations. The particular form a social relation takes acts selectively on what is said, when it is said, and how it is said. The form of the social relation regulates the options which speakers take u p at both syntactic and lexical levels. For example, if an adult is talking 16

See Footnote 1.

32

BASIL BERNSTEIN

to a child he or she will use a speech form in which both the syntax and the vocabulary is simple. Put in another way, the consequences of the form the social relation takes are often transmitted in terms of certain syntactic and lexical selections. In as much as a social relation does this, then it may establish for speakers principles of choice, so that a certain syntax and a certain lexical range is chosen rather than another. The specific principles of choice which regulate these selections entail from the point of view of both speaker and listener planning procedures which guide the speaker in the preparation of his speech and which also guide the listener in its reception. Changes in the form of certain social relations, it is argued, act selectively upon the principles controlling the selection of both syntactic and lexical options. Changes in the form of the social relation affect the planning procedures used in the preparation of speech and the orientation of the listener. The speech used by members of an army combat unit on manoeuvres will be somewhat different from the same members' speech at a padre's evening. Different forms of social relations can generate quite different speech-systems or linguistic codes by affecting the planning procedures. These different speech-systems or codes create for their speakers different orders of relevance and relation. The experience of the speakers may then be transformed by what is made significant or relevant by the different speech-systems. This is a sociological argument, because the speech-system is taken as a consequence of the form of the social relation, or to put it more generally, is a quality of the social structure. As the child learns his speech or, in the terms used here, learns specific codes which regulate his verbal acts, he learns the requirements of his social structure. The experience of the child is transformed by the learning which is generated by his own apparently voluntary acts of speech. The social structure becomes the substratum of his experience essentially through the consequences of the linguistic process. From this point of view, every time the child speaks or listens the social structure of which he is a part is reinforced in him and his social identity is constrained. The social

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING

33

structure becomes the developing child's psychological reality by the shaping of his acts of speech. Underlying the general pattern of his speech are, it is held, critical sets of choices, preferences for some alternatives rather than others, which develop and are stabilised through time and which eventually come to play an important role in the regulation of intellectual, social, and affective orientations. The same process can be put rather more formally. Individuals come to learn their roles through the process of communication. A role from this point of view is a constellation of shared learned meanings, through which an individual is able to enter into persistent, consistent, and recognised forms of interaction with others. A role is thus a complex coding activity controlling the creation and organisation of specific meanings and the conditions for their transmission and reception. Now, if it is the case that the communication system which defines a given role behaviourally is essentially that of speech, it should be possible to distinguish critical roles in terms of the speech forms they regulate. The consequences of specific speech forms or codes will transform the environs into a matrix of particular meanings which becomes part of psychic reality through acts of speech. As a person learns to subordinate his behaviour to a linguistic code, which is the expression of the role, different orders of relation are made available to him. The complex of meanings which a role-system transmits reverberates developmentally in an individual to inform his general conduct. On this argument it is the linguistic transformation of the role which is the major bearer of meanings: it is through specific linguistic codes that relevance is created, experience given a particular form, and social identity constrained. Children who have access to different speech-systems (i.e., learn different roles by virtue of their status position in a given social structure) may adopt quite different social and intellectual procedures despite a common potential.

34

BASIL BERNSTEIN

2.3. ELABORATED AND RESTRICTED CODES: DEFINITIONS AND BRIEF DESCRIPTION

Two general types of code can be distinguished: elaborated and restricted. They can be defined, on a linguistic level, in terms of the probability of predicting for any one speaker which syntactic elements will be used to organise meaning across a representative range of speech. In the case of an elaborated code, the speaker will select from a relatively extensive range of alternatives and the probability of predicting the organising elements is considerably reduced. In the case of a restricted code the number of these alternatives is often severely limited and the probability of predicting the elements is greatly increased. On a psychological level the codes may be distinguished by the extent to which each facilitates (elaborated code) or inhibits (restricted code) an orientation to symbolise intent in a verbally explicit form. Behaviour processed by these codes will, it is proposed, develop different modes of self-regulation and so different forms of orientation. The codes themselves are functions of a particular form of social relationship, or more generally, qualities of social structures. A distinction will be made between verbal or linguistic, and extra-verbal or para-linguistic components of a communication. The linguistic or verbal component refers to messages where meaning is mediated by words: their selection, combination, and organisation. The para-linguistic or extra-verbal component refers to meanings mediated through expressive associates of words (rhythm, stress, pitch, etc.) or through gesture, physical set and facial modification.

2.4. RESTRICTED CODE (LEXICAL PREDICTION)

The pure form of a restricted code would be one where all the words and hence the organising structure irrespective of its degree of complexity are wholly predictable for speakers and listeners. Examples of this pure form would be ritualistic modes of commu-

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING

35

nication: relationships regulated by protocol, types of religious services, cocktail-party routines, some story-telling situations. In these relations individual difference cannot be signalled through the verbal channel except in so far as the choice of sequence or routine exists. It is transmitted essentially through variations in extra-verbal signals. Consider the case of a mother telling her child stories which they both know by heart. " A n d little Red Riding Hood went into the wood", (ritualistic pause). " A n d what do you think happened ?" (rhetorical question). If the mother wishes to transmit her discrete experience, her uniqueness, she is unable to do this by varying her words. She can only do it by varying the signals transmitted through extra-verbal channels; through changes in intonation, pitch, speech rhythm, facial set, gesture, or even through changes in muscular tension, if she is holding the child. The code defines the channels through which new information (i.e., learning) can be made available. The discrete intents of mother and child, inter-personal aspects of the relation, can only be transmitted extra-verbally. Given the selection of the sequence, new information will be made available through the extra-verbal channels, and these channels are likely to become the object of special perceptual activity. The code defines the form of the social relationship by restricting the verbal signalling of individual differences. Individuals relate to each other essentially through the social position or status they are occupying. Societies differ in terms of the use made of this code and the conditions which elicit it. It is suggested that where there is an exchange of verbal message of maximal predictability, such as social routines, the context will be one where the participants have low predictability about each other's individual attributes. The code offers here the possibility of deferred commitment to the relationship. Decisions about its future form will be based upon the significance given to the exchange of extra-verbal messages. Consider a cocktail party. T w o people are introduced who have never met before. A social routine is likely to develop. This

36

BASIL BERNSTEIN

establishes mutual predictability and so the basis of a social relation. What is said is impersonal in that the verbal messages are all previously organised. The individuals will be highly sensitive to extra-verbal signals and so these signals are likely to become the object of special perceptual activity. How the social relation will develop initially depends upon the choice of social routine and the significance accorded to extra-verbal signals. Here, orientation is towards the extra-verbal channels: there is a minimal level of planning involved in the preparation of speech; the exchange of verbal sequences pre-supposes a shared cultural heritage which controls the verbal communications offered by the occupants of this cocktail-party status. It is important to note that: (1) The status or positional aspect of the social relationship is important. (2) Orientation is likely to be towards the extra-verbal channels as new information will pass through these channels. (3) Specifically verbal planning is confined to choice of sequence, rather than involving the selection and organisation of the sequence. (4) The code restricts the verbal signalling of individual difference.

2.5. RESTRICTED CODE (SYNTACTIC PREDICTION)

What is more often found is a restricted code, where prediction is only possible at the syntactic level.17 The lexicon will vary from one case to another, but in all cases it is drawn from a narrow range. It is necessary to point out that because a lexicon is drawn from a narrow range this is no criterion for classifying the code as a restricted one. The most general condition for the emergence of this code is a social relationship based upon a common, extensive set of closely-shared identifications and expectations self-consciously held by the members.18 It follows that the social relation17

Prediction here refers to an ability of a special observer not of the speakers. Restricted codes will arise in prisons, combat units of the armed forces, in the peer group of children and adolescents, etc. 18

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING

37

ship will be one of an inclusive kind. The speech is here refracted through a common cultural identity which reduces the need to verbalise intent so that it becomes explicit, with the consequence that the structure of the speech is simplified, and the lexicon will be drawn from a narrow range. The extra-verbal component of the communication will become a major channel for transmitting individual qualifications and so individual difference. The speech will tend to be impersonal in that it will not be specially prepared to fit a given referent. How things are said, when they are said, rather than what is said, becomes important. The intent of the listener is likely to be taken for granted. The meanings are likely to be concrete, descriptive or narrative rather than analytical or abstract. In certain areas meanings will be highly condensed. The speech in these social relations is likely to be fast and fluent, articulatory clues arc reduced; some meanings are likely to be dislocated, condensed, and local; there will be a low level of vocabulary and syntactic selection; the unique meaning of the individual is likely to be implicit. Restricted codes are not necessarily linked to social class. They are used by all members of a society at some time. The major function of this code is to define and reinforce the form of the social relationship by restricting the verbal signalling of individual experience.19

2.6.

ELABORATED CODE (LOW SYNTACTIC PREDICTION)

An elaborated code, where prediction is much less possible at the syntactic level, is likely to arise in a social relationship which raises the tension in its members to select from their linguistic resources a verbal arrangement which closely fits specific referents. This situation will arise where the intent of the other person cannot be taken for granted, with the consequence that meanings will have to be expanded and raised to the level of verbal explicitness. The 19 A restricted code does not necessarily affect the amount of speech, only its form.

38

BASIL BERNSTEIN

verbal planning here, unlike the case of a restricted code, promotes a higher level of syntactic organisation and lexical selection. The preparation and delivery of relatively explicit meaning is the major function of this code. This does not mean that these meanings are necessarily abstract, but abstraction inheres in the possibilities. The code will facilitate the verbal transmission and elaboration of the individual's unique experience. The condition of the listener, unlike that in the case of a restricted code, will not be taken for granted, as the speaker is likely to modify his speech in the light of the special conditions and attributes of the listener. This is not to say that such modifications will always occur, but that this possibility exists. If a restricted code facilitates the construction and exchange of communalised symbols, then an elaborated code facilitates the verbal construction and exchange of individualised or personal symbols. An elaborated code, through its regulation, induces in its speakers a sensitivity to the implications of separateness and differences and points to the possibilities inherent in a complex conceptual hierarchy for the organisation of experience. An example at this point is necessary to show how these various codes control social relations. Imagine a man is at a party where he finds a large number of people whom he has never met before. He goes up to a girl. He will then use, initially, a restricted code (lexicon prediction), which will provide the basis for the social relation. He will attempt to improve upon his understanding of her specific attributes by the meaning he gives to her presence and extra-verbal transmissions. He is then likely to move towards an elaborated code (if he possesses one) so that they may both have a means for elaborating verbally their distinctive experience. The possibility of discovering common ground is in this way increased, and the man may then move into a restricted code (syntactic prediction). The quality of the relationship at this point has shifted, and the girl may then regard this as slightly presumptuous and so force the man back to an elaborated code, or, if he is very unfortunate, to a restricted code (lexicon prediction). On the other hand she may accept the change in the social relation. The important points here are that the codes are induced by the social

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING

39

relation, are expressing it, and are regulating it. The ability to switch codes controls the ability to switch roles. This is a very simple example but it illustrates all the points made earlier.

2.7. FORMAL SOCIOLOGICAL CONDITIONS FOR THE EMERGENCE OF THE TWO CODES

It is possible to state the formal sociological conditions for the emergence of the two codes by distinguishing between the generality of the meanings controlled by the codes and the availability of the speech models from whom they are learned. To the extent that meanings are made explicit and are conventionalised through language, meanings may be called universalistic whilst if they are implicit and relatively less conventionalised through language, meanings can be called particularistic. Similarly, if the speech models are potentially generally available, such models can be called universalistic, whilst if the speech models are much less available they can be called particularistic. Using these concepts, a restricted code is particularistic with reference to its meaning and so to the social structure which it pre-supposes. However, it is universalistic with reference to its models, as such models are generally available. It is important to note here that the concern is with the availability of a special syntax. An elaborated code is universalistic with reference to its meanings and so to the social structure which it presupposes. However, it is likely that the speech models for this code will be particularistic. This does not mean that the origin of this code is to be sought in the psychological qualities of the models but that the models are encumbants of specialised social positions located in the system of social stratification. In principle this is not necessary, but it is likely to be empirically the case. Thus, because a restricted code is universalistic with reference to its models, all people have access to its special syntax and to various systems of local condensed meanings; but because an elaborated code is very likely to be particularistic with respect to

40

BASIL BERNSTEIN

its models, only some people will have access to its syntax and to the universalistic character of its meanings. Following this argument, the use of an elaborated code or an orientation to its use will depend not on the psychological properties of a speaker but upon access to specialised social positions, by virtue of which a particular type of speech model is made available. Normally, but not inevitably, such social positions will coincide with a stratum seeking or already possessing access to the major decision-making areas of the society. In terms of learning the codes, the codes are different. The syntax of a restricted code may be learned informally and readily. The greater range of and selection from, the syntactic alternatives of an elaborated code normally requires a much longer period of formal and informal learning. These distinctions are useful in isolating the general conditions for a special case of a restricted code (syntactic prediction). This is where the speech model is particularistic and the meaning is also particularistic. In this situation the individual is wholly constrained by the code. He has access to no other. The consequences of this are thought to be relevant to the problem of educability in developed or emergent industrialised societies. The sociological conditions may be summarised as follows: Restricted Code (lexical prediction): Ritualistic components of status or positional relationships Restricted Code (high syntactic prediction): 1. Model: universalistic; meaning: particularistic. 2. Model: particularistic; meaning: particularistic. Elaborated Code (low syntactic prediction): Model: particularistic; meaning: universalistic. The concept code refers to the control on the verbal realisation of the deep meaning structure of a culture or sub-culture — the core meaning structure. This involves a distinction similar to the distinction which linguists make between surface and deep grammar. Thus sentences which look superficially different can be shown to be generated from the same rules. In the same way, although the linguistic choices involved in a summary will be

A SOCIOL1NGUISTIC APPROACH TO SOCIAL LEARNING

41

markedly different from the linguistic choices in an analysis of physical or moral principles or different again from the linguistic realisations of forms of inter-personal control, they may all under certain conditions point to the underlying regulation of restricted or elaborated codes. It is of crucial importance to distinguish between speech variants and codes. A special variant is a pattern of linguistic choices which is specific to a particular context. For example, when one talks to children; a policeman giving evidence in court, talking to friends whom one knows well, the ritual of cocktail parties or train encounters. Because a code is restricted it does not mean that no linguistically elaborated speech variants will be realised, neither does it mean that if a code is elaborated no linguistically restricted speech variants will not be realised. Elaborated codes are media for the realisation of universalistic orders of meaning in which principles and operations are made verbally explicit. Restricted codes are media for the realisation of particularistic orders of meaning in which principles and operations are less verbally explicit. Thus elaborated codes are relatively less tied to a particular social structure or subculture and they contain potentially within themselves changes in principles and operations. Whereas in the case of restricted codes the meaning system to which it gives rise is relatively closely tied to a particular social structure or subculture. In this case the code does not contain within itself the potential of change in principles. Here change is likely to come from the outside.

2.8.

VERBAL PLANNING, LINGUISTIC CODES, AND SOCIAL STRUCTURES

The codes have now been defined, briefly described, and their formal sociological determinants specified. It is necessary to show how these codes may become established on a psychological level and this will be done by looking more closely at the process called verbal planning. When one person talks to another it is suggested that the

42

BASIL BERNSTEIN

following processes at different levels occur in the listener before he is able to produce a sequential reply. Orientation: the listener first scans the communication for a pattern of dominant signals. Not all the words and extra-verbal signals will carry the same value; some will carry greater significance than others for the listener. Selection: There will be associations to the patterns of dominant signals which will control the selections the listener makes from his potential stock of words, sequences and extra-verbal signals. Organisation: The listener will then have to fit the selected words and sequences into a grammatical frame and integrate them with the extra-verbal signals. On a psychological level, codes are generated by specific kinds of verbal planning. It follows that restricted and elaborated codes will establish different kinds of regulation which crystallize in the nature of verbal planning. The originating determinant of the kind of orientation, selection, and organisation, is the form of the social relation or, more generally, it is a quality of the social structure. The codes, linguistic translations of the meanings of the social structure, are nothing more than verbal planning activities at the psychological level and only at this level can they be said to exist. The consequences of the form of the social relationship are transmitted and sustained by codes which at the individual level consist of verbal planning processes. Particular orders of relationship to objects and persons inhere in linguistic codes. These orders of relation are then spontaneously generated by the individual as the verbal planning processes become stabilised. Following this argument, changes in the social structure, in the organisation of forms of social relation, modify speech systems or linguistic codes. These in turn, by virtue of verbal planning procedures, change the order of significance which individuals spontaneously create as a consequence of their acts of speech and which in their creation transform them. Clearly not all aspects of social structure are translated into elements of the linguistic code, but it is considered that the major aspects are so translated.

A SOCIOLINGUISTIC APPROACH TO SOCIAL LEARNING

43

The following diagram 20 might be helpful in distinguishing the levels of analysis. Level I (Code A

Determining)' B

Social Structure (Complex of Meanings)

C

» Forms of Social Relation (Specific Orders of Meaning)

Level II (Code

generating) '

Verbal Planning Functions

> Created Orders of Meaning

>-Linguistic codes 0 which gives us the following surface structure: Sentence Verb

Objective case marker

close

0

noun phrase the

door

60

MARSHALL DURBIN

SOCIOLINGUISTICS: AN APPROACH TO FUNCTIONAL GRAMMAR

61

If we illustrate the context according to the cases mentioned above we have the following situation: Context

Proposition

Modality

Manipulative

Action

Factitive

Dative

Locative

Agentive

Prevent

entrance

to John

to the room

by parent

The following rule applies to the structure: Manipulative + action (prevent, stop, keep away, etc.) => Manipulative + action (prevent) + Instrumental which is to say that the verb prevent, along with several other verbs are sub-categorized such that the Instrumental case must be adjoined when it occurs in the environment of the Manipulative modality. The result of this transformation is to adjoin the Instrumental case6 to the structure and to embed the sentence structure as that case. The resulting structure is shown on page 60. The meaning of the entire context with its verbal instrument can be phrased as a reading of the output in the following way: The parent agentive

8

manipulatively

prevents I John's

entrance

modality (manipulative)

action

dative

factitive

to close

the door

to the room

by commanding

John

location

modality (imperative)

Agentive

verb

objective

The instrument in this case need not have been a sentence. It could have been paralinguistic in nature where the parent moaned and the child responded; it could have been physical where the parent closed the door or removed the child from the situation entirely. On the other hand, it could have been verbal such that the parent said, "John, don't go in there."

62

MARSHALL DURBIN

There are undoubtedly many problems which have not been dealt with here such as the problem of when to and how to embed, the application of the rules, the problem of linearity versus simultaneity, and the subcategorization of verbs and nouns among others. The point, however, which must be kept in mind is that the context of language and language itself can be viewed through one system, and by a system, in fact, which apparently does comprise, as quoted above, a set of universal, presumably innate concepts which identify certain types of judgments human beings are capable of making about who did it, who it happened to, how it was done, and what got changed. Department of Anthropology Washington University, St. Louis

BIBLIOGRAPHY Bach, Emmon and Robert T. Harms 1968 Universal in Linguistic Theory (Holt, Rinehart and Winston). Beames, John 1872-79 A Comparative Grammar of the Modern Aryan Languages of India: to Wit, Hindi, Panjabi, Sindhi, Gujarati, Marathi, Oriya and Bangali (Trubner & Co). Chomsky, Noam 1957 Syntactic Structures (Mouton and Co). 1965 Aspects of the Theory of Syntax (M.I.T. Press). Durbin, Marshall in press "Linguistics and Writing Systems", Estudios de Cultura Maya, Vol. 7. Fillmore, Charles J. 1968 "The Case for Case", In Emmon Bach and Robert T. Harms, eds., Universals of Linguistic Theory (Harper and Row) pp. 1-88. Firth, J. R. 1957 Papers in Linguistics 1934-51 (Oxford University Press). Gardiner, Sir Alan Henderson 1950 Egyptian Grammar; Being an Introduction to the Study of Hieroglyphs. 2nd. ed., Fully Rev. (Oxford Univ. Press). Hall, Robert A., Jr. 1969 "Some Recent Developments in American Linguistics", Neuphilologische Mitteilungen 70: 192-227.

SOCIOLINGUISTICS : AN APPROACH TO FUNCTIONAL GRAMMAR

63

Hoenigswald, Henry 1966 "A Proposal for the Study of Folk-Linguistics", In William Bright, ed., Sociolinguistics (Mouton and Co.) pp. 16-26. Hymes, D. H. 1964 Language in Culture and Society (Harper and Row). Jesperson, Otto 1909-48 A Modern English Grammar on Historical Principles (C. Winter). Joos, Martin 1958 Readings in Linguistics (American Council of Learned Societies). Lévi-Strauss, Claude 1966 The Savage Mind (The University of Chicago Press). 1958 Antropologie Structurale (Plön). Postal, Paul 1966 "Review article of André Martinet's Elements of General Linguistics", Foundations of Language 2: 151-86. Saint-Jacques B. 1967 "Review of a Review", Foundations of Language 3 : 294-98.

4

SITUATIONAL MEASURES OF NORMATIVE LANGUAGE VIEWS IN RELATION TO PERSON, PLACE AND TOPIC AMONG PUERTO RICAN BILINGUALS * LAWRENCE GREENFIELD AND JOSHUA A. FISHMAN 1

In recent years, several studies have reported on the relationship between verbal behavior and a variety of psychological and social factors, such as the setting, the roles of the participants, the topics of conversation, the functions of interaction, and the views of interlocutors concerning each of the foregoing (Ervin-Tripp, 1964). Labov (1964), for example, found a series of phonological alternates in New York English speech which covaried with elicitation methods (that implied varying situational contexts of verbal interaction) and the socioeconomic status of the speaker. Fischer (1958), who studied the alternation between the use of the suffixes 'in' and 'ing' by New England children found that 'in' was used to a greater degree than 'ing' by boys than by girls, by children of lower than of higher socioeconomic backgrounds, in informal than in formal portions of the interview, and with informal verbs, such as 'chewin' and 'hittin' than with formal ones, such as 'correcting' and 'reading'. Brown and Gilman (1960) found that the use of the This article was first published in Anthropos 65. 1970 : 602-618. It is reprinted by permission of the publishers. The research reported herein was supported under DHEW Contract No. OEC-1-7-062817-0297, "The Measurement and Description of Language Dominance in Bilinguals", Joshua A. Fishman, Project Director. Data analysis was made possible by a grant from the College Entrance Examination Board. The preparation of the current presentation was facilitated by the Institute of Advanced Projects, East-West Center, University of Hawaii, where the Project Director spent the 1968-69 academic year as a Senior Specialist. 1 The authors wish to thank Dr. R. L. Cooper for his advice and encouragement during all stages of the work reported here.

SITUATIONAL MEASURES OF NORMATIVE LANGUAGE VIEWS

65

pronouns 'tu' or 'vous' (and their corresponding verb forms) in several Romance languages depended on relationships of power and solidarity existing between interlocutors. The social and psychological factors that are signaled linguistically by stylistic variation within a single language are frequently expressed by a complete SWITCH IN CODE in some bilingual settings (Ervin-Tripp, 1964; Hymes, 1966; Gumperz, 1964a). Rubin (1962) found that factors such as intimacy and informality were useful in describing the use of Spanish and Guarani in Paraguay. Thus, for example, young men used Spanish when first starting to court their sweethearts but as intimacy developed shifted to Guarani. Gumperz (1964b) and Blom and Gumperz (1966) reported that the use of the local dialect and national standard in a small Norwegian community was predictable from the social background of the interlocutors, the types of networks they formed and the topics discussed. Fishman has proposed the concept of DOMAIN in order to specify the larger institutional role-contexts within which habitual language use occurs in multilingual settings (Fishman, 1964, 1965a, 1965b, 1968). In gathering data appropriate to a given domain the investigator abstracts from or samples social situations at the level of face-to-face interaction involving DOMAIN APPROPRIATE places, role-relationships and topics. For example, in studying habitual language use in the family domain the investigator collects data regarding interactions between such domain appropriate interlocutors as husband—wife, parent—child, grandparent—grandchild, in such domain appropriate locales as 'home', concerning suchdo main appropriate topics as 'proper behavior of children'. Relevant domains for describing language use in many relatively complex multilingual societies would probably include family, friendship, religion, education, work sphere, and government (Fishman, 1968). Using this concept, Fishman has suggested that it is possible to differentiate between stable bilingual societies in which diglossia obtains (Ferguson, 1959), and unstable bilingual societies. In the former, languages tend to be reserved for different domains of life

66

LAWRENCE GREENFIELD AND JOSHUA A. FISHMAN

in the community, i.e., language varieties tend to have definitive institutional associations which exist over and above their more fleeting metaphorical functions. In the latter, domain separation in language use vanishes and the 'other' tongue comes to be used alternatively with the 'mother' tongue, particularly in the family and friendship domains. The institutional separation of codes first becomes secondary to their metaphorical functions and then is re-established along quite different institutional lines. In general, unstable intragroup bilingualism has occurred in cases of immigrant languages in the context of rapid industrialization, urbanization, or other rapid social change, as for example, in the cases of Yiddish, Ukrainian, Hungarian and German in the United States (Fishman, 1965a). Examples of more stable intragroup bilingual speech communities have been described by Barker (1948); Blom and Gumperz(1966); Ferguson(1959); Fishman(1965b); Rubin (1962); and Weinreich (1951). Recently Fishman (1966) has developed a sociolinguistic model which suggests that in diglossia situations there generally exist two major clusters of complementary community values, called L and H, respectively, each of which is realized in a different speech variety or language, ¿-related values are usually those of intimacy, solidarity, spontaneity and informality, while //-related values usually involve an emphasis on status differences, ritual and formality. Furthermore, those members of the community who identify with or accept these two cultural value clusters tend to utilize the culturally approved speech variety or language in their domain appropriate behavior. Typically, the ¿-variety or language is used in domains such as family and friendship, while the Hvariety is reserved for domains such as education, occupation and religion. Moreover, when two individuals interact in a locale or discuss a topic that is incongruent with their usual role-relationship, they struggle toward a redefinition of the situation and tend to use the speech variety or language which is congruent with such a redefinition. For example, a professor and student who are engaged in mountain-climbing may no longer view themselves as professorstudent but as individuals interacting in some other role-relation-

SITUATIONAL MEASURES OF NORMATIVE LANGUAGE VIEWS

67

ship. Under such circumstances, the variety used would be appropriate to the perceived social relationship and to the re-defined total situation of which that relationship is a part. The present paper reports on two experiments which were designed to examine the possibility of incorporating into the measurement of normative language views via self-report instruments some of the interactional sensitivity derived from microsociolinguistic theory and ethnography. Studies of normative language views have thus far been inter-situational at best. The current paper reports two attempts to study the intra-situational normative views of a bilingual population. Several studies have suggested the possibility that unlike most previous immigrant groups in the United States, the Puerto Rican community in New York has many of the features that Fishman describes in his model of diglossic speech community. One factor that has been mentioned in favor of this possibility is that while adapting to life in the United States, the Puerto Ricans in New York continue to maintain close physical ties with their homeland and as a result, come to identify with the values prevalent in both countries (Padilla, 1958; Senior, 1965; Hoffman, 1968). According to these studies, Puerto Ricans learn from the U.S. the importance of social and economic advancement and from their Puerto Rican heritage the importance of maintaining close contact with family members and friends. Therefore, it is hypothesized that in the Puerto Rican community in New York Spanish may have come to be associated with values such as intimacy and solidarity and to be used primarily in domains such as family and friendship, while English may have come to be associated with values such as status differentiation and to be used primarily in domains such as religion, education and employment. Within each of these domains we further hypothesize that Spanish and English will be normatively viewed as polarized in the directions of intimacy and statusstressing respectively.

68

LAWRENCE GREENFIELD AND JOSHUA A. FISHMAN

1. EXPERIMENT 1

1.1. 1.1.1.

Method

Technique

The technique used in the first experiment was derived from studies of the structure of conversations which were conducted by Hershkowitz and Krause (1965) and by Blass (1965). In these studies, lists of persons, places and topics were ranked by groups of American college students along the dimensions of intimatedistant, private-public and personal-impersonal, respectively. The students were asked to imagine themselves in a number of conversations of which two components were supplied by E and the third was to be filled in by them. When E supplied a pair of elements which were of the identical scale position (congruent), the Ss invariably selected the third one from the same end of the scale as the others. When the two provided elements were from opposite ends of the scale (thus being incongruent), there was a tendency for S to re-define one of them so as to be congruent with the other and then to select as the completing element one which was congruent with the perceived position of the first two. For example, when presented with the situation of talking to a friend (intimacy-distance rank no. 1 or no. 2) in the park (private-public rank no. 6 or no. 7), some of the Ss explained the situation by saying that "he wasn't really a good friend", while others explained it by saying that "no one was around". In selecting the third element, the former Ss tended to choose relatively impersonal topics, while the latter ones tended to select personal ones. 1.1.2.

Subjects

The Ss included in this study were a group of boys and girls of Puerto Rican descent who belonged to a Puerto Rican youth organization, Aspira, which sponsors clubs (conducted primarily in English) in New York City high schools. This organization is a private educational agency designed on the one hand, to build

SITUATIONAL MEASURES OF NORMATIVE LANGUAGE VIEWS

69

career opportunities and leadership roles for Puerto Rican youth and on the other, to develop in them a positive self-image by strengthening their Puerto Rican identification. Accordingly, this group was used as a basis for securing subjects who were most likely to identify with the two major value clusters in the community. 1.1.3.

Procedure

Since domains are a higher order generalization derived from CONGRUENT SITUATIONS (i.e., from situations in which individuals interacting in societally appropriate role-relationships with each other, and in the societally appropriate locales for these rolerelationships, discuss topics that are considered societally appropriate to these role-relationships and locales), it was first necessary to test intuitive and rather clinical estimates of the congruencies that were felt to obtain in the Puerto Rican community of New York City. After more than a year of general participant observation as well as focused interviews and discussions with native informants it seemed to the authors that five domains could be generalized from the innumerable situations that they had encountered, namely, "family", "friendship", "religion", "education", and "employment". 2 As a means of collecting self-report data on normative language use, a situation was selected which seemed to be typical of each domain. As indicated below each of these situations consisted of a seemingly congruent situational interlocutor, situational place and situational topic. Domain

Interlocutor

Place

Topic

Family

Parent

Home

Friendship Religion Education Employment

Friend Priest Teacher Employer

Beach Church School Workplace

How to be a good son or daughter How to play a game How to be a good Christian How to solve a math problem How to do your job in the most efficient way

2

For a complete protocol of a validating interview with a native informant see Fishman, Cooper, Ma et al. (1968), Chapter III-3-b (Instrument construction tiyout: Tape A, Informant P2).

70

LAWRENCE GREENFIELD AND JOSHUA A. FISHMAN

An instrument was constructed which required S to complete a number of situations in which two of the three components were provided by E.3 Specifically, S was requested (a) to select a third component (from among five alternatives) in order to complete the situation and (b) to indicate the amount of Spanish and English they would be likely to use if they were involved in such a situation and if they and their Puerto Rican interlocutors knew Spanish and English equally well. For each situation, amount of each language used was to be rated on a five-point scale in which 1 = all in Spanish, 2 = more Spanish than English, 3 = roughly equal amounts of Spanish and English, 4 = more English than Spanish, and 5 = all in English. In some of the situations the components which were provided by E were seemingly congruent, as they appeared to belong to the same domains, and in others they were seemingly incongruent, as one of them appeared to belong to either the family or friendship domains (intimacy value cluster) and the other to the domains of religion, education, or employment (the status value cluster). In accord with our hypothesis concerning the domains which existed in the community and the persons, places and topics that were congruent with these domains, it was expected that where the two components provided by E were congruent with each other the component selected by S would come from the same domain as both of those which were provided by E. Where the two components provided by E were incongruent with each other it was expected that the component selected by E would come from the same domain as one of the two provided components. Furthermore, in accord with the hypothesis that the community studied was a diglossic speech community, it was expected that a preference for Spanish would be reported when the third component chosen by S was appropriate to either the family or friendship domains; conversely, it was expected that a preference for English would be 3

The layout of instrument 1 and the full scoring procedures utilized in connection with it are shown in Appendix VIII-2, Fishman, Cooper, Ma et at., (1968).

SITUATIONAL MEASURES OF NORMATIVE LANGUAGE VIEWS

71

indicated when the component selected by E was appropriate to the domains of religion, education or employment. The data gathering instrument was entirely in English and consisted of three sections in each of which the situations described constantly lacked a given component, namely, either the person, place or topic. The sections were randomly distributed among the Ss who were tested in groups at the conclusion of their club meetings. 1.2. 1.2.1.

Results

Choice of the third component

Table 1 shows the percent of Ss who for each of five seemingly congruent situations, selected the hypothesized domain-appropriate third component as the completing element. In the situation comprising "friend" and the friendship topic, the hypothesized friendship locale, beach, was chosen by only 40 % of the 5s. In each of the remaining seemingly congruent situations, however, the component which was hypothesized to be congruent with those provided by E was selected by at least 80 % of the Ss. Table 2 shows for each of the seemingly incongruent situations the number of 5s who chose a component which was hypothesized TABLE 1

Percent of Ss Selecting 3rd Components Congruent with Two Others Presented by E and Derived from Given Domains Congruent Component Selected Domain

Family Friendship Religion Education Employment

Person (n=16)

Place (n=16)

Topic (n=18)

81 94 81 81 88

100 40 100 93 100

89 100 83 100 100

72

LAWRENCE GREENFIELD AND JOSHUA A. FISHMAN

o «

o ——

H E "8 2

I I

— — OfSrJ-

J; rt «

-H n

O

c «s ^ s f ^ s f e e E Ij'OBIi'OBUi-Ki.ii.

S "O

( ^ u u c ^ u u h b h b h U H U o. E E G

B

&

?I

5 3 3 g i i ^

£

s a >> e u

e!

§ 1

v° ££ tì 'S o .ti ¡2 S C •g s •g a Js « S ö

«

n N H H O N N T f r n - o o o ^ n 1-H tN —