'Along the Routes to Power' Explorations of Empowerment through Language 9783110185997, 3110185997, 2006004858

Table of contents :
Preface
Acknowledgements
Introduction: Along the routes to power
Section 1. Theoretical perspectives: Linguistic empowerment and language choices
Sociolinguistics: More power(s) to you! (On the explicit study of power in sociolinguistic research)
The power of language, the language of power
Language endangerment, the construction of indigenous languages and world English
The power to choose and its sociolinguistic implications
How codeswitching as an available option empowers bilinguals
Section 2. Language policy and language planning: Empowering speakrs of minority languages in communities and institutions
Language policy failures
Empowerment through the community language – A challenge
Pidgins and Creoles between endangerment and empowerment: A dynamic view of empowerment in the growth and the decline of contact languages, especially in the Pacific
Lost in transculturation: The case of bilingual education in New York City
Language policies in Spain: Accomodation or alteration?
The potential of parliaments for the empowerment of linguistic minorities: Experiences from Scotland and Norway
The dominance of languages and language communities in the European Union (EU) and the consequences
Section 3. The language empowerment discourse: Case studies of language policy and language planning in Africa
Socio-political factors in the evolution of language policy in post-Apartheid South Africa
Marginalisation and empowerment through educational medium: The case of the linguistically disadvantaged groups of Botswana and Tanzania
Language policy, cultural rights and the law in Botswana
We speak Otjiherero but we write in English - Disempowerment through language use in participatory extension work
Empowerment through English – A realistic view of the educational promotion of English in post-colonial contexts: The case of Nigeria
Life in a Tower of Babel without a language policy
JK Nyerere of Tanzania and the empowerment of Swahili
Living on borrowed tongues? A view from within
Index

Citation preview

'Along the Routes to Power'

W G DE

Contributions to the Sociology of Language 92

Editor

Joshua A. Fishman

Mouton de Gruyter Berlin · New York

'Along the Routes to Power' Explorations of Empowerment through Language

edited by

Martin Pütz Joshua A. Fishman Jo Anne Neff-van Aertselaer

Mouton de Gruyter Berlin · New York

Mouton de Gruyter (formerly Mouton, The Hague) is a Division of Walter de Gruyter G m b H & Co. KG, Berlin.

© Printed on acid-free paper which falls within the guidelines of the A N S I to ensure permanence and durability.

Library of Congress Cataloging-in-Publication

Data

Along the routes to power : explorations of empowerment through language / edited by Martin Pütz, Joshua A. Fishman, JoAnne Neffvan Aertselaer. p. cm. — (Contributions to the sociology of language ; 92) Includes bibliographical references and index. ISBN-13: 978-3-11-018599-7 (hardcover : alk. paper) ISBN-10: 3-11-018599-7 (hardcover : alk. paper) 1. Sociolinguistics. 2. Sociolinguistics — Africa. 3. Power (Social sciences) 4. Language policy. 5. Language planning. I. Pütz, Martin, 1955— II. Fishman, Joshua A. III. Neff-van Aertselaer, JoAnne. IV. Series. P40.A38 2006 306.44—dc22 2006004858

ISBN-13: 978-3-11-018599-7 ISBN-10: 3-11-018599-7 ISSN 1861-0676 Bibliographic

information published by Die Deutsche

Bibliothek

Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the Internet at < h t t p : / / d n b . d d b . d e > .

© Copyright 2006 by Walter de Gruyter G m b H & Co. K G , D-10785 Berlin All rights reserved, including those of translation into foreign languages. N o part of this book may be reproduced in any f o r m or by any means, electronic or mechanical, including photocopy, recording or any information storage and retrieval system, without permission in writing f r o m the publisher. Printed in Germany.

Preface This book contains powerful and persuasive writing from a team of top international scholars. It is ground-breaking by impressively focusing on how power and empowerment are foundational to the health and prospect of all the world's languages. This book announces the centrality of the 'sociolinguistics of power' by confronting researchers, teachers and planners across various language and cultural disciplines with 'power and empowerment' as crucial to any modern understanding of languages. Colin Baker University of Wales, Bangor

Acknowledgements This collection of papers is the result of the 31 st International LAUD Symposium, which was held at the University of Koblenz-Landau (Campus Landau) on April 19-22, 2004. The conference theme, "Empowerment through Language" concentrated on various aspects of the theory and application of the relation between 'language and power' seen from an interdisciplinary perspective: the various discussions centered around the linguistic, social, psychological and educational issues involved in multilingual and multicultural settings. We would like to thank a large number of people who have contributed to this volume. First of all, many thanks are due to the participants in the symposium for the stimulating discussions in a very pleasant atmosphere, and to the contributors to the present volume, who have responded with alacrity and professionalism to all the requests that have been made of them. The contributions were selected for inclusion only after a lengthy process of refereeing and, in some cases, extensive revising. In this regard we would also like to express a great debt of gratitude to the expertise of the many scholars who acted as our referees: Efurosibina Adegbija t, Dick Baldauf, Enrique Bernärdez, Pauline Bryant, Peter Chilton, Rene Dirven, Angela Downing, Rosalyn Finlayson, Carmen Fought, Herbert Igboanusi, Nkoko Kamwangamalu, Bob Kaplan, Karsten Legere, Marilyn MartinJones, Andreas Musolff, Tove Skutnabb-Kangas, and Christa van der Walt. We are also grateful to the organising staff of the symposium, in particular Heike Ramsauer, Sandra Balzert, and Holger Schmitt whose enthusiasm, experience and readiness were an enormous support which contributed to the success of the symposium. Special thanks are due to our secretary Ms. Marion Liithe who did an outstanding job in word processing and in designing the layout of the book. Thanks are also due to Heidi Kissel for proofreading major parts of the manuscript. For generous funding, we express our thanks to the German Research Foundation (DFG), the University of Koblenz-Landau, the Faculty of Cultural and Social Sciences (Campus Landau), the Ministry of Education and Culture of the Rhineland-Palatinate, the Paul and Yvonne Gillet Foundation (Edenkoben) and the Friends and Supporters of the University of Koblenz-Landau (Campus Landau).

viii Acknowledgements Finally, we would like to thank Rebecca Walter (Mouton de Gruyter) for her kind assistance and cooperation with this venture.

Martin Pütz University of Koblenz-Landau Joshua A. Fishman Stanford University, New York University and City University of New York JoAnne Neff-van Aertselaer Universidad Complutense de Madrid

Contents Preface

ν

Acknowledgements

vii

Introduction: Along the routes to power Martin Pütz, Joshua A. Fishman and JoAnne Neff-van Aertselaer

x m

Section 1. Theoretical perspectives: Linguistic empowerment and language choices Sociolinguistics: More power(s) to you! (On the explicit study of power in sociolinguistic research) Joshua A. Fishman

3

The power of language, the language of power John Edwards

13

Language endangerment, the construction of indigenous languages and world English Janina Brutt-Griffler

35

The power to choose and its sociolinguistic implications Florian Coulmas

55

How codeswitching as an available option empowers bilinguals Carol Myers-Scotton

73

Section 2. Language policy and language planning: Empowering speakers of minority languages in communities and institutions Language policy failures Bernard Spolsky

87

χ

Contents

Empowerment through the community language - A challenge Michael Clyne

107

Pidgins and Creoles between endangerment and empowerment: A dynamic view of empowerment in the growth and the decline of contact languages, especially in the Pacific Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

129

Lost in transculturation: The case of bilingual education in New York City Ofelia Garcia

157

Language policies in Spain: Accomodation or alteration? JoAnne Neff-van Aertselaer

179

The potential of parliaments for the empowerment of linguistic minorities: Experiences from Scotland and Norway Heiko F. Marten

199

The dominance of languages and language communities in the European Union (EU) and the consequences Ulrich Ammon

217

Section 3. The language empowerment discourse: Case studies of language policy and language planning in Africa Socio-political factors in the evolution of language policy in postApartheid South Africa Neville Alexander

241

Marginalisation and empowerment through educational medium: The case of the linguistically disadvantaged groups of Botswana and Tanzania Herman M. Batibo

261

Language policy, cultural rights and the law in Botswana Lydia Nyati-Ramahobo

285

Contents xi We speak Otjiherero but we write in English - Disempowerment through language use in participatory extension work Rose Marie Beck

305

Empowerment through English - A realistic view of the educational promotion of English in post-colonial contexts: The case of Nigeria Hans-Georg Wolf and Herbert Igboanusi

333

Life in a Tower of Babel without a language policy Augustin Simo Bobda

357

JK Nyerere of Tanzania and the empowerment of Swahili Karsten Legere

373

Living on borrowed tongues? A view from within Paulin G. Djite

405

Index

421

Introduction: Along the routes to power Martin Pütz, Joshua A. Fishman and JoAnne Neff-van Aertselaer

It is with a modicum of justifiable pride that the editors introduce the present collection of papers concerning the relevance of "power" to the sociology of language. Just a little over 40 years ago, in 1964, sociolinguistics itself (a major portion of the total sociology of language "pie") was launched. It has by now become such a robust and productive area of international specialization that it is no longer possible for an individual scholar, no matter how broad his or her interests may be, to keep fully abreast of its total reach. Nevertheless, the human mind refuses to be bound by its own physical or structural limitations and new sub-specializations have continued to appear endlessly, as have the absolutely essential abstracting, retrieving and distributing tools required for their cultivation, so that their future depends only on their own provocative vitality to researchers, teachers and students alike. And now we stand at the launching of yet another topical area of possible concentration within the total sociology of language enterprise! Its purview is implied by the concept of "power", a folk-concept for sure, like so many others that have been investigated, precisioned and incorporated into the sociology of language. This collection of papers, stemming from the 30th LAUD Symposium held at the University of Koblenz-Landau (April 2004), clearly reveals, even had we wished to hide or disguise this, how much remains to be done for a new topic to really "take off'. We are quite aware that many of the papers that follow use differing definitions of "power" and, that some, indeed, invoke or pursue no definition whatsoever. To some extent (hopefully minor) this may always be so with "power", one of those words that we seemingly understand so easily in everyday informal language use. This will not always continue to be the case in sociolinguistic theory and empirical research on "language and power" if this volume can persuade many of its own contributors, as well as at least some of the many others who will examine it closely, to stay with it and not to

xiv

Martin Pütz, Joshua A. Fishman and JoAnne Neff-van Aertselaer

abandon it in its infancy because of the growing-up pains that it is very likely to have and to cause. Section 1 Theoretical Perspectives: Linguistic Empowerment and Language Choices concentrates on theoretical issues pertaining to the relationship between language and power, linguistic empowerment and the verbal repertoire as an inventory of behaviors and language choices. It includes contributions by Fishman, Edwards, Brutt-Griffler, Coulmas and MyersScotton. In his contribution Sociolinguistics: More power(s) to you!, Joshua Fishman introduces the general topic of language and power as it has been traditionally dealt with in sociolinguistic research by observing that very few researchers have appropriately defined this field as their area of major concentration. Fishman advocates a type of sociolinguistics which is concerned with situational and demographic variation as the major independent variables of the discipline. Furthermore, he points out that examples of power differences between various languages at different times in world history can be revealing of the gains to be expected by pursuing this line of inquiry. For Fishman, the formal "sociolinguistics of power" is about to be born. In his article The power of language, the language of power, John Edwards broadens the discussion from general considerations of power to matters of concrete linguistic empowerment. He argues against the idea that the process of empowerment coexists with 'knowledge' and rather sees 'knowledge' as a component of power or one which has to be transformed into power. His implication is to cast some shadow over the claims that e.g. the subordinate position of ethnolinguistic minorities can be substantially affected by heightened awareness alone; or more concretely that socially disadvantaged children will have improved life-chances if their language varieties are "empowered" in the classroom. Similarly, Janina Brutt-Griffler argues that the choice between a dominant language such as English or a "mother tongue" as the medium of instruction presents educational complexities that have been insufficiently considered. In Language endangerment, the construction of indigenous languages and world English she goes even so far as to say that mother tongue education can represent a socio-political policy of fostering ethnic divisions similar to that pursued by the apartheid regime in South Africa to facilitate white rule. The "language endangerment discourse" is thus brought into the discussion and pursued further in a number of controversial empirical studies dealing with language planning issues in Africa (see Section 3).

Introduction: Along the routes to power

xv

The remaining two articles in Section 1 deal with the notion of variability and language choice as an important aspect in sociolinguistics in general and bilingualism per se. In The power to choose and its sociolinguistic implications, Florian Coulmas argues that every aspect of every utterance is the result of the speaker's choice. Because choice is a notion that is central to our self-understanding, it offers itself as a focal point for the integration of the various subfields of sociolinguistics such as conversational analysis and politeness, social dialectology as well as codeswitching and diglossia in multilingual surroundings. Accordingly, Coulmas explores the notion of power by making reference to the observation that language choices are subject to restrictions and that whenever a common language exists it thereby limits its speakers' choices. The problematic issues of power and its relation to language choices as witnessed in the speech of bilinguals is taken up by Carol Myers-Scotton when answering the question How codeswitching as an available option empowers bilinguals. The article discusses codeswitching (CS), the use of more than one language in the same conversational context as it figures in negotiations of power differentials between individuals or groups. When the social elite engage in CS between a widely known local language and an elite language (official or international), they present themselves as different from nonelites. Interestingly enough, Myers Scotton shows by way of studies in Nigeria and Sri Lanka that through certain types of CS, the non-elite can circumvent elite closure. Consequently, CS plays a role in negotiating relationships that can restrict, but can also provide, access to power. Section 2 Language Policy and Language Planning: Empowering Speakers of Minority Languages in Communities and Institutions focuses on the central elements of language policy and current strategies, methods and goals of planning practice in several regions such as Australia, the Pacific, USA (with a focus on New York City) and Europe. Social elites are in positions of political, social, economic and educational power and may be able to control language planning processes for their own advantage and to the detriment of language minority speakers. The question then to be asked will be: in whose interests is language planned - individuals, the state, or agencies and organisations? The section includes papers by Spolsky, Clyne, Ehrhart/Mair/Mühlhäusler, Garcia, Neff-van Aertselaer, Marten, and Ammon. In his article Language policy failures Bernard Spolsky deplores the fact that bilingual education is usually maintained to increase the power

xvi

Martin Pütz, Joshua A. Fishman and JoAnne Neff-van Aertselaer

and acceptance of a powerful ethnic or local or national variety (Irish, French in Quebec, Catalan or Basque in the autonomous regions, Maori in New Zealand etc.), or to speed transition to an international language (Russian in the Soviet republics, English in Singapore) but seldom to ease the way into school for speakers of unfavoured languages. Spolsky makes a number of suggestions to bridge the gap between what appears to be accepted scientific theory and actual educational practice. As a case in point Michael Clyne in his paper Empowerment through the community language - A challenge explores to what extent it is possible to empower immigrants and their children by raising the status of their 'minority' languages through status planning, especially in the education system of Australia. Impediments to such empowerment and strategies for the sharing of community languages to the common good are researched in regard to current practices in Australia and on the basis of an action research project in four Melbourne schools. The issue is discussed in the context of deepseated monolingual/monocultural attitudes in the community and an environment in which disabling changes have outweighed enabling changes, and especially where there has been a move away from comprehensive, cohesive language policy development. An interesting account of the "majority-minority language" debate refers to the situation of European-lexifier Pidgins and Creoles, sometimes seen as aggressors in precarious linguistic ecologies. In their article Pidgins and Creoles between endangerment and empowerment: A dynamic view of empowerment in the growth and the decline of contact languages, especially in the Pacific region, Sabine Ehrhart, Christian Mair and Peter Mühlhäusler present data from two endangered English-related Creoloids from the South Pacific, i.e. Norfolk and Palmerston. They convincingly show that in addition to the risks threatening all languages with small communities of speakers it is, paradoxically, their relation to English and respective language planning policies which pose an additional threat to their survival. Concentrating on a much smaller but cosmopolitan speech community, Ofelia Garcia examines the language situation of Latino immigrants in New York City, one of the most multilingual communities in the world where Caribbean Spanish has an enormous presence. Her paper entitled Lost in transculturation: The case of bilingual education in New York City focuses on the role that schools and educational programs have to empower and sometimes disempower Spanishspeaking New Yorkers. Through an analysis of this dis/empowerment through English/Spanish bilingualism, the paper makes evident the language ideologies in the United States today.

Introduction: Along the routes to power xvii

The remaining papers of Section 2 deal with case studies of language planning in European countries (Spain, Scotland, Norway) and the European Union as a parliamentary institution. JoAnne Neff-van Aertselaer's paper Language policies in Spain: Accommodation or alteration? deals with the role that language policies have played in shaping the regional and cultural diversity of Spain during the early 1990s up to the 2004 national elections. The paper specifically discusses the way in which attempted state intervention via two educational laws, one concerning primary and secondary education and the other concerning tertiary education, have sought to curtail processes of linguistic change, brought about by the devolution processes, in three historical-institutional contexts: the Basque Country, Catalonia and Galicia. In his article The potential of parliaments for the empowerment of linguistic minorities: Experiences from Scotland and Norway Heiko F. Marten takes up the notion of yet another aspect of politics and language, i.e. the decentralization of parliamentary power for the benefit of minority languages. Two case studies deal with the relationship between the Scottish Parliament and the Gaelic language on the one hand, and between the Norwegian Sämi Parliament (the Sameting), and the Sämi language on the other hand. The underlying idea is to empower speakers of minority languages to have as many instances of language choice as possible. Finally, Ulrich Amnion discusses The dominance of languages and language communities in the European Union (EU) and the consequences for individual speakers and the entire communities of languages involved. The EU's linguistic reality appears such that its most striking feature is the ever greater predominance of just one powerful language, English. This fact is often felt to be in conspicuous contrast with the EU's language-policy objectives. Ammon, however, makes several suggestions as to how the dominance of English and multilingualism could be at least partially compatible in terms of planning languages in the EU. Section 3 discusses the major concepts of the Language Empowerment Discourse such as language diversity; maintenance, shift and loss; as well as linguistic human rights with a focus on empowering speakers educationally, economically and socially through language. Generally, all papers refer to case studies of language contact and conflict in Africa where the status and use of the autochthonous languages (endogenous) alongside dominant, mostly European (exogenous) or pan-African languages (lingua francae), has led to differing language planning efforts in various parts of Africa. The papers discuss possible solutions to the problems of educa-

xviii

Martin Pütz, Joshua A. Fishman and JoAnne Nejf-van Aertselaer

tional failure and loss of linguistic and cultural identity; they partly propose to revalorize the use of African languages by developing and promoting them, but they also imply critical stances as to an overestimation of minority or African languages in this regard. The section includes contributions by Alexander, Batibo, Nyati-Ramahobo, Beck, Wolf and Igboanusi, Simo Bobda, Legere and Djite. Neville Alexander considers the Socio - political factors in the evol ution of language policy in post-Apartheid South Africa that have favoured or impeded language planning efforts in the country since 1994. A major challenge of the dynamics of language policy evolution and language practices in South Africa refers to the danger of stagnation, i.e. the refusal on the part of state authorities and other decision makers to tackle the language issue as a matter of urgent social policy. Such inertia prevents or retards the intellectualisation of the African languages and condemns the majority of the people to permanent dependence on an English-knowing elite. Instead, Alexander advocates an anti-ethnicist language policy of genuine regard for multilingualism and pluralist alternatives which are to be propagated and implemented. The multilingual situations of a neighbouring country of South Africa, i.e. Botswana and its eastern counterpart, i.e. Tanzania, represent the topical foci in a paper by Herman Batibo titled Marginalisation and empowerment through educational medium: The case of the linguistically disadvantaged groups of Botswana and Tanzania. Both, Botswana and Tanzania, have gone through different socio-political and linguistic experiences since their independence. In the cases of Botswana and Tanzania the speakers of the minority, African languages or those non-fluent in the adopted lingua francae, i.e. Swahili (Tanzania) and Setswana (Botswana), suffered many linguistic and cultural disadvantages due to the use of their mother-tongue. Batibo discusses in greater detail the issue of dis-empowerment of minority speakers such as the adverse educational outcomes of these conflict situations, particularly in the area of school performance, student drop-out and related aspects. A more 'activist' approach referring to the legal and cultural rights of minority speakers in Botswana is put forward by Lydia Nyati-Ramahobo in her article Language policy, cultural rights and the law in Botswana. The prestigious status of the lingua franca Setswana and the territorial and economic power of the Tswana-speaking inhabitants has permitted sociocultural discrimination along linguistic, ethnic and economic lines, rendering individuals belonging to non-Tswana communities practically invisible. Nyati-Ramahobo presents the impact of current language policy on linguistic and cultural

Introduction: Along the routes to power xix human rights and the long-standing agitation of the unrecognized groups for such rights which include parliamentary motions, the formation of linguistic associations, litigations and the engagement of United Nations procedures. Based on these language policy efforts, Nyati-Ramahobo offers a somewhat optimistic view of language revitalization leading to linguistic and cultural democratization. Indirectly, the notion of linguistic human rights and development is advanced further by Rose Marie Beck in her article We speak Otjiherero but we write in English - Disempowerment through language use in participatory extension work. As a case in point, Beck cites the multilingual situation in Namibia where linguistic fragmentation is commonly seen as a barrier to development. Focussing on a Herero-speaking community, the author intends to show how a policy of linguistic exclusion can be observed at grassroots levels and what repercussions it has on participation and developmental efforts. Specifically the author shows how the use of written English in a predominantly Herero-speaking context is a way of 'doing elite'. The situation encountered in the study is thus a fine example of empowerment of the few and disempowerment of those who are targeted as beneficiaries of a development project. Taking up the notion of choice, Hans-Georg Wolf and Herbert Igboanusi refer to the Foucaultian perspective stating that linguistic powerlessness can be seen as equivalent to a limitation of linguistic choice. In their article Empowerment through English - A realistic view of the educational promotion of English in post-colonial contexts: The example of Nigeria, the authors critically review accounts of linguistic imperialism, linguistic human rights and the power of English by pointing out that such critics fail to take into account the linguistic realities of some multilingual nations. As a case in point they refer to the linguistically extremely complex situation in Nigeria where concentrating on the local language to the detriment of English will perpetuate a restricted linguistic potential and thus the disenfranchisement and seclusion of the rural population. They argue for the further promotion of an acculturated variety of English in educational settings of Nigeria and other multilingual African countries. Another highly multilingual country is neighbouring Cameroon which, compared to Nigeria, bears the extra burden of two exoglossic official languages in addition to a greater concentration of languages over a specific area. In his paper Life in a tower of Babel without a language policy, Augustin Simo Bobda partially deplores the fact that Cameroon is paralleled by a conspicuous absence of language policy. The paper discusses the manifestations of this situation, some non-governmental attempts to give a

xx

Martin Pütz, Joshua A. Fishman and JoAnne Nejf-van Aertselaer

direction to language use, and some advantages and disadvantages of the lack of language planning in Cameroon. The disadvantages, for example, include the gradual death of some of the local languages and the impossibility of communicating with the local communities in their own languages, thus disempowering their speakers. The long history, status and function of an African national language and lingua franca, i.e. Swahili, and its relation to language planning issues is discussed by Karsten Legere in his article JK Nyerere of Tanzania and the empowerment of Swahili. Legere refers to the personal role of the late Nyerere who, as the country's first president for many years, made a substantial contribution to spreading, promoting, developing and thus empowering Swahili. Although strongly supporting the role of Swahili as the language of political discourse, social and economic interaction and as an expression of a Tanzanian identity, Nyerere was also somewhat biased as far as the status of English in Tanzania was concerned, the other major topic discussed in the paper. The last paper of Section 3 refers to the multilingual situations of French Western Africa. In his paper Living on borrowed tongues? A view from within, Paulin G. Djite discusses the colonial discourse of French including its subjugation of all indigenous languages as being 'uncivilized' versus the construction of a 'beautiful' and 'rational' language of French. The aim of the paper is to explore the socio-political factors behind the maintenance of French in language-in-education planning. It proposes insights from within to the questions as to whether French is still the lever of upward social mobility in French-speaking Africa and whether there are really dramatic socio-linguistic discontinuities between one's pre-school cognitive categories and those taught in primary schools where French is the sole medium of instruction.

Conclusion The collection of papers included in this volume offers insight into the variety of perspectives from which the social evolution of languages may be studied. Covering a diversity of geographical contexts, the underlying issues discussed in the papers range from the individual's choices of language behaviour to those of the group and the repercussion these choices may have regarding social power.

Introduction: Along the routes to power xxi Thus, while each of the papers may define or allude to power in slightly different ways, according to the issues addressed, all of the papers refer to dynamic processes in which power is construed as the ability, whether by overt agency or not, to affect the actions or ideas of others. We hesitate to add Weber's "despite resistance" to our conception of power because it is not at all clear that a relation involving power needs to be conceived of as "coercive" (see Fishman's reference to the power of kinship feelings, of love, of friendship, etc.). Another issue, also bringing us to the questions posed by Fishman in chapter 1, is how power can be measured, either at a macro- or microlevel. We hope that this volume will provide opportunities for continued debate both on the concept of power and on the variables which may be examined regarding language behaviours.

Section 1. Theoretical perspectives: Linguistic empowerment and language choices

Sociolinguistics: More power(s) to you! (On the explicit study of power in sociolinguistic research)*

Joshua A. Fishman

Sociolinguistics has too seldom paid adequate attention to the centrality of "power" in human behavior and to "social power" more explicitly in its study of language in society. This is not to say, of course, that "power" has been totally overlooked, but it has rarely been explicitly defined, measured nor theoretically related to the socially patterned variation in language use and in language attitudes that constitute the very heart of sociolinguistic concerns. On the other hand, I am fully aware of the fact that American sociolinguistics/sociology of language came into being in the mid-60s after at least four power-related studies that have, since then, come to be considered as classics in our field.

1.

Early studies

I have in mind for this distinction certain well-known early studies by Clifford Geertz (1960), Charles Ferguson (1959), Roger Brown and Albert Gilman (1960) and Bill Labov (1964). As you may remember, Geertz studied verbal interaction under conditions of strict and seemingly immutable social stratification, such that each interlocutor's choice of any part of speech (nouns, verbs, adjectives, etc.) was determined by both correctly estimating the "other's" social class position relative to one's own, on the one hand, and the formality of the context (commercial, religious, educational, etc.) then obtaining. Clearly, certain social classes had more power than others, something that is implicit in the very notion of social class, and certain situational (cultural) settings elicited more social class differentiation than did others. Unfortunately, Geertz's findings are based entirely on non-quantitative ethnographic data so that neither he nor we can examine the question of whether social class power or the situational facilitation

4

Joshua A. Fishman

or inhibition of that power is the greater of the two main effects being investigated. Ferguson's famous discussion of "diglossia" dealt somewhat more directly with power, but it too was entirely non-quantitative in nature. Once more we are in a context of both class and situational influences and it is not entirely clear which one of the two is the stronger (e.g., initial class differences or a common "near death" experience), although generally speaking Ferguson emphasizes the former rather than the latter. Brown and Gilman's study of reciprocal and non-reciprocal pronouns and verb-forms ("formal" and "informal") was the first to reveal (in a quasi-addendum) that the theoretical differences posited by class and situational differences do not always obtain in quantified practice, suggesting that there may well be other contributory variables (societal, ideological and personal) rather than merely the two that are obviously at play. Finally we come to Labov's famous study which quantitatively reveals the response variance which the other three investigators had probably noted but couldn't analyze or present to us. Like the others, Labov too deals with social class and with situational modifiers of power (ingeniously represented by different elicitation techniques). Although he does not compare the two directly, any investigator interested in doing so can tease this comparison out of Labov's tables. The social class curves are a lot further apart under the most formal elicitation procedures than under the informal ones that capture hurried and spontaneous speech. There is a wonderful (but unrealized) opportunity to also study the "interaction" between these two sources of variance (e.g., given that the upper class is always more frequently formal than the lower classes, are there any situations in which this is most particularly so or reversed?). Labov adumbrates this issue in his discussion of "hyper-correction" by the lower-middle class but it would have been greatly preferable to quantify this unexpected finding and to test its significance.

2.

The post-classical study of "power": a growth industry

Only one of the early studies that we have mentioned above (Brown and Gilman 1960) explicitly mentions "power" (it is entitled "Pronouns of power and solidarity") and although each of them became a model for future research, "power" per se was not often mentioned (or not mentioned nearly as much as the language-use variation that is of interest to sociolin-

Sociolinguistics: More power(s) to you!

5

guists). The discipline was still new and there were many, many other topics that called (and begged) for exploration. There was also no marked attempt outside of "variationist" circles to quantify "power" and even there the trend was to focus on (or at least to utilize) social class as the be all and end all of power. Today, the research climate is much different with respect to the sociolinguistic study of "power". The LLBA bibliographic retrieval system reveals over 8000 items that have been classified as power-related in their "keywords" or in their broad topical foci. It has come to be realized that studies relating language use to race, ethnicity, gender, politics, ideology and the societal assignment of languages to functions in multilingual societies are all essentially studies of language and power. At a narrower topical level, the pursuit of and the resistance to "power" should now be recognized as an essential dimension in studies of literacy campaigns, the spread of English, language death, language planning.

3.

Defining "power"

Although this newfound growth and diversity in the study of "power" may help us escape the clutches of vulgar Marxism, it nevertheless contains the seeds of many new problems that need to be considered. The first of these is arriving at a consensual definition of "power" that will enable us to be sure that we are all talking about essentially the same social phenomenon. After all, there are all kinds of power (e.g., physical strength, monetary power [or the possession of items easily converted, recognized and accepted as legal tender], natural resources [coal, iron, oil, clean air and water], military power, social status, authority ["by the authority vested in me, I hereby pronounce you...."], social approval, legal power, brain power and, last but not least, manpower and fire-power). By extension, any languages or varieties recognized as associated with more of any of the foregoing, have more "power" than do those associated with less. However, this great diversity leads me to suggest parsimoniously that "power" be simply defined as "control over scarce resources", i.e., resources that are not available equally nor to the degrees that they are desired by the various parties to any interaction. I am certainly not suggesting that the various kinds of power named above are equivalent. They most certainly are not (because they are not equally useful for the resolution of any particular social problem that may be at hand), but it would

6

Joshua A. Fishman

certainly be welcome to compare (separately and then combined) the efficacy of two or more of them within the same population at any given time. In this connection, I would like to call attention to the study by Bilaniuk (2003) in which the independent variables were language of message (Ukrainian, Russian or English speakers), ethnicity of judge (Russian or Ukrainian) and gender of judge, while the dependent variable was the height or degree of the favorability ratings (re social power, income and friendliness) allotted by judges to each apparently native speaker after listening to his/her message. Among the main findings are that, first of all, female judges rate speakers higher in Russian than in Ukrainian, as well as higher in English than in Ukrainian and, secondly, that Russian raters of either sex rate English higher than Russian on all favorability attributes. These attributes were the scarce social desirability resources that the judges were distributing to speakers after hearing each one's presentation. Of course, additional scarce resources could have been utilized as dependent variables (e.g., the age of speakers, the education of speakers, etc.), but, as it is, Bilaniuk's study is already enviably richer than any of the classical studies mentioned earlier (not to mention being indicative of gender-of-judge differences that are interpretable within the research tradition that has repeatedly found that women are more judgmental of dialectstandard differences than are men). Far from perfect though it may be, Bilaniuk's study deals with clearly definable and individually measurable criteria of power, does not stop with "social power" alone and avoids using or discussing the concept of "social class" entirely. Of the 8000 language-related studies touching upon "power" that LLBA lists under the keyword "power" I would judge that only a minority (perhaps fewer than 10%) possess these desirable features.

4.

Resisting the simplification of human social behavior

I have already mentioned that defining "power" as the control of scarce resources, as useful as that may be, can also be problem-fostering rather than only problem-solving. It can easily lead to a preference for material and easily quantifiable resources, i.e., to resources that are neither sufficiently contextualized nor viewed comparatively to other criteria. Humans and their societies display a vast array of subtle values that reach far above and beyond material resources alone. Gella and I seriously wanted our

Sociolinguistics: More power(s) to you!

7

children to do well in their public school studies (all of which involved the English language) and we helped them overcome difficulties in that connection in their homework. However, as was the case in my own childhood home, we permitted no English in our home that was unrelated to schoolwork or guests. In our home to this very day, even though there are no children or grandchildren under foot to socialize and although Yiddish offers precious few material benefits to us at home or in the world at large, it would be easy for an observer to conclude that Yiddish shows many signs of being a stronger language than English. It would require a very sensitive recognition of "varieties of power" and their functional contextualization in order to adequately explain how the language that is much weaker monetarily, nevertheless successfully maintains itself in the repertoire of many families that know and use both for different functions. Use of quantitative and materialistic criteria of power alone would impoverish both our sociolinguistic knowledge of language and our appreciation of humankind insofar as their true complexity, subtlety and contextualization are concerned. Indeed, we might accidentally overlook the power of beliefs and convictions to lead "fanatics" to prize them more than they prize life itself. This is true not only for Ultra-Orthodox Islam but for each one of us who can honestly admit that our beliefs are our most precious resources, precisely because they force us to constantly realize anew that the reduction of all power to dollars and cents alone is also a reduction of humanity per se. The power of kinship feelings, loyalties and obligations is also difficult to quantify, monetarily or otherwise. Aristotle realized that the Greeks could defeat the Persians if they took their most intimate family members along with them into battle with that infinitely more numerous enemy. Theories of ethnicity have benefited from recognizing the power of kinship (Fishman 1965). The power of love and friendship is also exemplified by the risks to his own life and to his future throne that Jonathan (the son of King Saul) accepted because of his deep friendship for David. The power of altruism must be related to such human tendencies. The curative power of laughter has been documented in connection with fostering cures and the moving power of tears to persuade is known to husbands (whether or not they know the Rabbinic injunction to husbands to speak softly to one's wife because a woman's tears are always close to her eyes). The power of tradition and of historical memory are exemplified by Gella's and my need to wait approximately 55 years after the Holocaust before visiting Germany today. The power of sociocultural norms to channel and

8

Joshua A. Fishman

control behavior is so great that Dürkheim considered such conventionenforcers to be essentially religious in nature, because of their constant interpretation of what was good and bad. Even though modern western cultures do not recognize the power of politeness and the use of honorifics as much as do various cultures of Asia and the Pacific, I have a former student (now a colleague) who still cannot bring herself to call me anything but "Professor Fishman" even after more than a quarter century. The power of language (both generally and specifically in terms of particular languages) is also revealed by the many authoritative efforts to control the use of one or another of them. "Life and death are determined by language" the ancient rabbis opined. And, of course, we must not overlook the interlocking nature of various forms of social power, such as the power of literacy and numeracy (and of much of formal education en toto) which facilitate access to yet other kinds of power (higher order power in government, technology and the professions). Mahatma Gandhi never tired of reminding his followers of the power of weakness. This is not a contradiction in terms if the contributions of a number of followers and the approbation of world opinion are included as predictors of the goals Gandhi pursued, not to mention the power of the righteous example ("turning the other cheek") of which Christianity speaks. The purpose of this long recitation of kinds of "power" is both to remind us of the many varieties of "power" that we have not yet begun to adequately consider in sociolinguistic research and, also, to reexamine the issue of the adequacy of our initial attempt to define "power". Does our original definitional attempt (power derives from "the control of scarce resources") apply to all of the various kinds of power that we have briefly listed above? It seems to do so quite well, although more complex formulations and criteria are often required. The two-step flow in connection with the power of education is repeated in many other contexts as well. The power of moral and ethical beliefs and practices to heal is attributable in many cultures to physicians who must themselves be in a state of grace, i.e. the physician calls upon the help of God (presumably the supreme controller of resources that are rarely in adequate supply) in order to be an effective healer. The power of social approval often leads to selfconfidence, which, in turn, facilitates job access and job-success. The relative power of learning specific languages is revealed through income differentials, which in turn, are powerfully related to more advantageous marriages, which in turn are related to the establishment of family control

Sociolinguistics: More power (s) to you!

9

over particular markets. From this we can conclude that the number of interlinked scarce resources may be fairly endless and, therefore, that the determination of all of the different kinds of power relative to the control of any particular resource may constitute a super-human task. Fortunately, the analysis of variance enables us to know what proportion of the variance in the dependent variable can be accounted for by any particular set of predictors. We must be ready for some disappointments along the way, since some of above mentioned kinds of power may, on closer examination, turn out to be either redundant or merely non-scientific figures of speech. Nevertheless, it is healthy for a discipline to cultivate research methods that enable it to find out such things empirically (and rapidly), rather than to continue "in the dark" about them.

5.

Measuring predictors of "power" (dependent variables) and the degree of control of scarce resources (independent variable)

I do not want to enter into a dispute as to whether quantification is always possible or preferable. Even the great psychologist Edward Thorndike who held that "if a thing exists at all, it must exist to some degree and if it exists to some degree, it can be measured", nevertheless excused his daughter's suitor from answering the question "how much do you love her?". Nevertheless where valid quantification is possible it would be a pity not to use it for it alone allows us to tackle certain fundamental questions: the amount of variance that remains unexplained and the order of importance of various independent explanatory variables. We need to be able to answer questions such as whether numeracy or literacy is more important (and whether personality is of any importance at all) in accounting for social success. The role of positive and negative predictors must be disentangled, both visa-vis each other and vis-a-vis the success criterion being investigated.

6.

An all-purpose term is a no-purpose term

The importance of demographic predictors of "power" cannot be overestimated since they reveal large scale societal biases (e.g., gender, race, ethnicity or religion) that greatly influence access to power. Such large scale preferences exist in every society and should be documented at the same

10

Joshua A. Fishman

time that investigators focus on more nearly individual predictors such as attitudes, aspirations, interests and goals. Also, "power" must not become an empty figure of speech such that the less we know about it concretely the more fashionable it becomes to use it. Something of this sort happened to the concept of "language prestige", when it became clear that less prestigious vernaculars repeatedly displaced more prestigious classicals in function after function throughout Europe, via movements for the democratization of the ballot, for free public education and for a literate popular culture. Many individuals of great wealth and high social standing not only joined but spearheaded efforts on behalf of bettering the condition of the lower classes (viz. DeValera, Neheru, the narodniki in 19th century Czarist Russia, etc.), efforts that could succeed only in the vernaculars. Empowering the lower classes may well require the "de-powerment" of earlier elites, but the quest for power is not the only or even the major force today in human destiny.

7.

Conclusions

In order to advance sociolinguistic research on "power" and to keep it from becoming "just about anything to everyone" (and, therefore, something a little different in each study) it would be highly desirable to specify in each case the rare resource the control of which leads to power. It would also be desirable to specify the outcomes of power, be they material or non-material. Furthermore, it would seem to be obvious that sociolinguistics would introduce situational and demographic variation into its study-designs, these being the major independent variables of the discipline. Finally, comparative studies should also be undertaken across different kinds of power (dependent variables) in order to arrive at both constant and differential conclusions, e.g., are age, gender and family-ties similarly predictive of both financial power and of ritual power? Although there are now a great number of language-related-studies of power, there are very few (if any) researchers who have defined this field as their area of major concentration, to be returned to time and again and to be developed cumulatively along theoretical lines. It has been my hope that this paper might help somewhat to rectify this problem, especially in connection with clarifying how a variety of types of powers relate to both language choice and other language behaviors (including behavior toward language) in different social settings.

Sociolinguistics: More power (s) to you! 11

Note * Paper prepared for the biennial LAUD Symposium, held at the University of Koblenz - Landau (Campus Landau), Germany (April 2004). I would like to express my appreciation to Professor Martin Pütz for his invitation to participate as well as for his sensitive hospitality during the entire visit.

References Bilaniuk, L. 2003

Gender, language attitudes, and language status in Ukraine. Language in Society 32: 47-78.

Brown, R. and A. Gilman 1960 The pronouns of power and solidarity. In Style in Language, T. A. Sebeok (ed.), 253-276. Cambridge Mass.: MIT Press. Ferguson, C. A. 1959 Diglossia. Word 15: 325-340. Fishman, J. A. 1965

Geertz, C. 1960 Labov, W. 1964

Varieties of ethnicity and language consciousness. Georgetown University Monograph Series in Languages and Linguistics 18: 69-79. Linguistic etiquette. In The Religion of Java, C. Geertz, 286-290. Glencoe: Free Press. Phonological correlates of social stratification. American Anthropologist 66: 164-176.

The power of language, the language of power John Edwards

A language is a dialect that has an army and a navy Max Weinreich (1894-1969)

1.

Introduction

Society has never distributed its blessings fairly or equitably, so it is no surprise that manifestations of linguistic access and recognition have historically favoured some groups more than others. "Only before God and the linguist," said Bill Mackey (1978: 7), "are all languages equal." He elaborated his pithy observation by noting that "everyone knows that you can go further with some languages than you can with others." This rather Orwellian qualification of equality has been widely accepted for a very long time, even within liberal cloisters. For example, John Stuart Mill ([1861] 1964: 359-362) is generally considered to be an early proponent of the view that the "common sympathies" - encapsulated in such markers as shared history, religion and language that unite people under a national banner are both natural and positive qualities of the human condition. Thus, his famous remark: "it is in general a necessary condition of free institutions that the boundaries of governments should coincide in the main with those of nationalities." This is a sentiment which, in one form or another, is widely endorsed today among nationalists and their intellectual supporters; indeed, what could come more easily to the liberal perspective than the idea that those with "common sympathies" ought to rule themselves? And many would take things further, and argue that such an arrangement is one of right rather than preference. So, Mill's apparent support for nationalist organisation and its (potential) linguistic underpinnings can be seen to buttress some contemporary understandings of political rights and, more specifically, of language rights - a clear reflection of the power of language or, at least, of the power that language ought to have (see Edwards 2003).

14 John Edwards The qualifiers, however, that Mill uses in his observation here ("in general" and "in the main") are very important, for his approval of ethnolinguistic and nationalist aspiration did not prevent him from noting the following: Nobody can suppose that it is not more beneficial to a Breton, or a Basque of French Navarre, to be brought into the current of the ideas and feelings of a highly civilised and cultivated people - to be a member of the French nationality ... than to sulk on his own rocks, the half-savage relic of past times ... the same remark [he added] applies to the Welshman or the Scottish Highlander as members of the British nation.1

If we put this parenthesis against the earlier remark - and, indeed, against other indications of Mill's approval of the coincidence of linguistic and political units - it would seem that the language-power equation is compromised: either it ought not, in principle, to carry much weight for certain subaltern groups; or it cannot, in practice. Either interpretation disturbs contemporary adherents of Mill's first, and more inclusive, remark. The principled reduction of linguistic power in certain circumstances is rejected as unfair, and an abrogation of liberal-democratic ideals; and the de facto relative weakness of some varieties is something to be remedied (typically by way of educational provision and legislative arrangement). Points of view here are of course supported by modern analyses that convincingly demonstrate the intrinsic worthiness of all linguistic varieties. Linguistic power is now (correctly) seen to rest upon social bases, and is not really a matter of language at all; this is an insight that was generally foreign to intellectuals before modern times - even to those who, like Mill, were not ill-disposed in principle to "small" languages and communities. We are inexorably brought back, too, to Mackey's point: educators, officials and - more importantly in the long term - the public-at-large are rarely either linguists or immortals, and this has powerful implications for the power of language and, more specifically, for any realignment or redistribution of the sociocultural and sociolinguistic resources that are seen as unfair to certain groups. From general considerations of power, then, we move very quickly to matters of empowerment. Power and, more particularly, the process of empowerment, is often said to coexist with knowledge. In his famous essay on heresy, Francis Bacon (1597) wrote nam et ipsa scientia potestas est - knowledge itself is power - and this has been uncritically accepted in many quarters ever since. It is clearly, however, a false equation. Knowledge is, rather, "a kind of fuel; it needs the motor of understanding to convert it into power"

The power of language, the language ofpower

15

(Wyndham 1960: 176) - and exactly the same sentiment has been expressed in more generalised fashion, most recently perhaps by the philosopher, Vilem Flusser (2003: 95): "knowledge has to be transformed into power." So, it would be more accurate to see knowledge as a component of power. While it is of course possible to have power without knowledge, the latter is generally a necessary but not sufficient element of the former - especially where power is to be more than fleeting. The particular implication here, for matters of language, is to cast some shadow over the breathless claims that - for example - the subordinate position of ethnolinguistic minorities can be substantially affected by heightened awareness alone; or that minority-group or socially disadvantaged children will have improved life-chances if their varieties are "empowered" in the schoolroom.

2.

Power and its manifestations

This is hardly the place for a discussion of either the philosophy or the sociopolitics of power, but it is important to note that, on this topic at least, the two perspectives are appropriately closely entwined. This is made clear in the work of Weber, for instance, whose insights - most conveniently consulted in his Economy and Society (1968) - still provide a framework for much sociological thinking 2 Power, Weber pointed out, is the ability to achieve desired ends despite resistance or, more generally, the exercise of one's will over others. He then subdivided it into the components of coercion, influence and authority. In the first two instances, power resides in force - force that can vary from blunt imposition to more democratic debate. Under the heading of authority we recognise the institutionalisation and legitimation of power, and it is easy to see, then, that it logically contains coercion and influence. In fact, it might be appropriate to slightly amend Weber's insight: there is a continuum of force along which the wielding of power can be situated, and the most important face of power, regardless of strength, is that of authority. Weber returns to categorisation when noting that there are types of authority: he calls them traditional, legal (or rational) and charismatic. That is, we may respond to directives because we understand them to emanate from accepted sources of long standing; or, we may obey because we see the rational value behind them; or, we may be influenced by particularly appealing qualities of the source. Power - at least in its most psychologically and sociologically interesting forms - involves relationships, particularly ones of inequality. As al-

16

John Edwards

ready suggested, its range extends from mild suggestion, to influence (perhaps the commonest manifestation of power), to authoritarian fiat, and depending upon the source and the degree of potency - the feelings and the reactions of those who are subject to it will vary in both strength and practical relevance. If there is a gun to one's head, for instance, any attempt to exchange points of view may be at best fruitless, and at worse perilous; but if a teacher is attempting to encourage or shape a student's thinking, then we like to believe that such exchange is not only acceptable but desirable. Acquiescence to influence is an important quantity in many individual and social relationships, which implies greater reciprocity than the word "power" may typically suggest. But when matters are in a "rawer" or more bluntly coercive state, acquiescence "amounts to no more than that he [the person being affected] continues to value whatever is being used as a lever against him - an acquiescence that only the stoic, perhaps, would seriously regard as a matter of choice" (Benn 1972: 425). Benn's useful observation suggests two points: first, the person influenced is a captive of his or her own desires in regard to what is controlled by another; second, if one were able to renounce or "devalue" these desires, the force of coercion would collapse. But sometimes such devaluation is impossible - if what is controlled bears directly upon survival, for instance. Similarly, in the political arena, in the world of cultural contact, in matters of linguistic dominance and subordination, individuals or groups often find that the controlled quantity cannot be ignored. It may be seen as impossible or, at the very least, highly traumatic to accept linguistic or cultural assimilation, for instance. In such situations, it is hard to imagine that even Benn's stoic would perceive many real options here; choices are essentially of Hobson's variety. Nonetheless, some power may still reside in acquiescence, even in the most authoritarian of relationships, because the stable exercise of power is hard pressed to rely solely upon coercion and enforcement - it must be able to count upon a large measure of ovine passivity. Or, as Ortega y Gasset (1967: 127) more colourfully reminds us: The truth is that there is no ruling with janissaries. As Talleyrand said to Napoleon: "You can do everything with bayonets, Sire, except sit on them." And to rule is not the gesture of snatching at power, but the tranquil exercise of it. In a word, to rule is to sit down, be it on the throne, curule chair, front bench or bishop's seat. Well, one can rule with janissaries, of course - but for how long, and to what effect? So acquiescence, even in the most passive forms, theoretically remains a potentially mobilisable force, or counter-force.

The power of language, the language ofpower

17

The relevance of the foregoing for contexts of linguistic and cultural contact can be summarised easily enough. First, the fact that language shift is such a common occurrence, coupled with the reasonable assumption that - if it were at all possible - people would prefer not to have to shift, reinforces the idea that knowledge by itself is not power. Second, matters of empowerment in the linguistic domain are unlikely to be ones in which unwanted pressure can be removed via "devaluation" of what is being influenced. Third, the power against which subordinated or minority groups have to struggle is the power of authority. That is, the "large" linguistic and cultural forces threatening "smaller" varieties generally have legitimated and institutional status; this is so even when - as in most instances such status derives more from strong unofficial pressures than from more formally statutory ones. Force emanating from broad social currents, rather than from consciously legislated ones (although, of course, the latter may be put in place to formalise the former), suggests an influence that is both wide and deep. Relatedly, an implication for "small" varieties may be that investing too many of their hopes in official measures may be naive or shortsighted, or both. It is also useful to consider the most common types of authority here: if we return to Weber's categorisation, we see that the institutionalised influences with which endangered languages and cultures have to contend are, above all, legal/rational - either statutes exist which make it difficult or impossible to live a full life through the original variety, or (more likely, as just noted) the pressures are "rational", which in this context implies the calculation of the social economist rather than the logic of the mathematician.3 As well, however, these "rational" pressures are often complemented by Weber's other two elements. The big language that is casting shadows over smaller ones usually possesses "traditional" force - and in two senses: first, it typically has the weight of a long internal tradition of emerging and consolidated power; second, its current dominance over others begins to take on, itself, traditional qualities. (We observe this process, for instance, with English vis-ä-vis other varieties - a long period during which English itself moved from a language much less important than those of its continental neighbours, to one of global significance, and now to one that is the stereotypical linguistic ogre.) There is also ample evidence that Weber's third element - charisma - is salient and, in many cases, a more or less inevitable accompaniment to the other two. Again, one has only to consider the extra-communicative and symbolic appeal of English; or, more specifically, its ability - abetted by

18 John Edwards its hosts - to burrow its way into other varieties; or, more specifically still, its potency in advertising. I have in front of me, for instance, three wrappers from Japanese sweets, picked up during a very recent visit to that country. Each has an abundance of Japanese-language product information, lists of ingredients, place of manufacture, notes of special offers, and so on - but each also displays English messages. We read, for example, that "Meltykiss Chocolate" is a sweet that "gently melts in your mouth like a snowflake available only in winter" (original absence of punctuation). Then there is "Chelsea Yogurt Scotch" - advertised on the packet as "the taste of old Scotland ... the candy with traditional Scottish flavour. Please enjoy its superior taste." And the third wrapper encloses "Men's Pocky Chocolate", described as "crispy pretzel dipped in dark chocolate for the type of person who enjoys the finer points of life." To focus on the oddity of usage here is to miss the point - which is that any such oddity is meaningless. Very few of the purchasers of these sweets will decipher these English words, nor will they either know or care that "yogurt" and "Scotland", and "Chelsea" and "Scotland", are not immediately associated in the anglophone mind, nor that "meltykiss" may have rather unintended connotations, nor that the idea of "men's chocolate" - particularly if it is "pocky" - would not be a strong selling point in London or New York. The powerful connotations here lie not with the words but with the choice of language itself, and the more or less constant marketing surveys imposed upon the Japanese consumer, as well (of course) as actual buying patterns, reveal the positive associations with this choice.4 Finally here, it is important to remember that all the manifestations of "authoritative" linguistic power are, themselves, by-products of cultural contact and conflict - very often, in fact, they are secondary and hardlyconsidered varieties of fall-out subsequent to the interaction of unequal sociopolitical entities. This has many implications, the most important of which is that language shift - considered as a consequence of some languages-in-contact scenario - is symptomatic and cannot, therefore, be understood as a matter that could be dealt with (or eased, or reversed) as if it had an independent existence. (This, I think, is a curious point: on the one hand, it is hard to imagine any informed sociolinguistic observer denying the complex tangles of social fibrillae; but, on the other, many discussions of language shift - especially those motivated by desires to reverse it proceed as if such independence were in fact the case.)

The power of language, the language of power 3.

19

Is it better to give or to receive?

It seems to me that any discussion of social empowerment and, therefore, any consideration of its implementation, ought to bear in mind that, beneath the specifics of language (or religious observance, or cultural custom, or any other such "marker") lies the psychologically deeper matter of identity. While linguistic instrumentality is usually the immediate or patent bone of contention - and while, indeed, instrumentality may, in some instances, be the only such bone - the point at which people more typically become exercised over language matters occurs when they feel that more than instrumentality is at stake. If groups go to the barricades over the language of school or of work, there usually exist more latent worries, too. And this is not particularly mysterious, because the symbolic and the instrumental (or communicative) aspects of language - while separable - often coexist. More importantly, perhaps, in "at-risk" settings in which such separation is seen to be in train, users desire that they should coexist - and this may well lead to actual or attempted interventions. While it is undoubtedly true that the symbolic value of language can (and commonly does) outlast communicative language shift, and thus continue to provide some psychosocial anchor for groups and individuals (a typical context here involves old-world immigrants in new-world settings) and while, indeed, such continuity is by far the most common scenario in most situations in which minorities and majorities must rub along together, it should be remembered that, in a long-term historical sense, the communicative aspect has been a prerequisite for the symbolic. And, since most psychological matters having to do with identity typically rest upon - whether consciously articulated or not - such a long-term perspective (as the use of words like "heritage", "tradition" and "authenticity" makes clear), it is no wonder that linguistic instrumentality is to be buttressed wherever possible - for it is unlikely that the symbolic half of the picture will persist forever without the reinforcement of ordinary usage. Related to this first general point is that language empowerment is commonly seen as a compensatory device. That is, action on behalf of ethnocultural or sociocultural groups seen to be "at risk" is seldom initiated before some rationale is perceived; and, in many real-life settings, past and present, this perception often seems (to some, at least) a rather delayed reaction. Regardless, however, of whether or not intervention is seen to be too little or too late, considering empowerment within a compensatory framework surely has some telling implications. First, the very word "em-

20

John Edwards

powerment" has come to have compensatory connotations, to suggest an exchange between unequal partners. In its original senses, to empower meant to license, to authorise: it already had, then, the implication of inequality - since, clearly, the one authorising is the more senior participant but it is a delimited sort of inequality, and one that does not immediately conjure images of strength on the one hand and frailty on the other; the authorisation is for a specific undertaking, commission, investigation, and so on. Today, however, the sense has expanded considerably, and any dictionary definition will associate empowerment with verbs like "giving" or "enabling" or "equipping". The old, limited meaning has been replaced with one that clearly suggests the relief of broad insufficiency through the intervention of general power. To combine the two preceding points, then: contemporary understandings and treatments of cultural and linguistic empowerment typically assume that important psychosocial factors provide the most central rationale for intervention, and that such intervention is justified on the basis of longstanding oppression, prejudice, conquest, colonisation, and so on - historical or systemic unfairness, in a word. The sense is that what was stripped away ought now to be restored, and that it is eminently reasonable that those who inappropriately benefitted in previous contact settings should now be the ones to make restitution. But this sense is surely inseparable from another: that the individuals or groups currently endangered are unable to resolve their linguistic and cultural dilemmas by themselves. All of which leads to the observation that empowerment inevitably involves the bestowal of something by the strong - which, I suggest, is very different from the taking of power, so historically characteristic of those who are currently strong. More pointedly, the suggestion here is that empowerment as it is commonly understood logically implies, and actually reinforces, a continuation of group inequalities. The school is the most obvious arena for exercises in empowerment aimed at the two most salient groups: speakers of nonstandard varieties of the national language, and members of ethnonational minority groups whose identity is considered to be at risk of assimilation. Indeed, the school is often seen as the place in which to act on behalf of linguistic and cultural maintenance or revival. It is an obvious place, for now very familiar reasons. First, school is a powerful institution which, while generally an arm of mainstream society, is nonetheless committed to the development of all who attend - in that sense, it is one of the most "neutral" territories that people are likely to find themselves in. Second, school benefits from hav-

The power of language, the language of power

21

ing a captive audience; all children must attend from an early age - and it is significant that going to school represents the first and most important "break" from the home - and they must remain at school for a considerable period of time. Third, the educational context is a natural one in which cultural contact and its consequences might be discussed, understood, treated, and so on. A less familiar - but not less important - reason for school to be a common site of attempted empowerment is that it encapsulates that idea of empowerment-as-compensation noted above: school is at once a compensating mechanism and a continuingly strong arm of that society whose dominance creates the need for compensation in the first place. The potential conflicts here are not, of course, lost upon the participants. Internal contradictions can arise here even where one may assume a general goodwill towards the "others" within the school precincts. If, as is sometimes the case (and was often the case), school is a rather less generously-minded institution, then it is easy to see that any "compensation" on offer was or would be severely restricted, in line with the perceived requirements of the mainstream. Historical accounts of workingclass education - where it existed at all - illustrate that the general aim was more to facilitate the continued production of workers than it was to assist in upward social mobility. Prior to the mid-nineteenth century, at any rate, "disadvantaged" pupils who were helped to rise by education were usually inadvertent beneficiaries of the workings of some Smithian "invisible hand". Even allowing for excellent intentions, dedicated teachers and adequate resources, however, there are strong grounds for thinking that reliance upon school as an empowering agent is naive - and, as with naivete in other settings, innocence coupled with ignorance can exacerbate matters. Since I have already discussed the difficulties here, and given present limitations of time and space, I will leap to the summary of these difficulties: school cannot compensate for society (for fuller details, see Edwards 1985, 1989). In line with earlier remarks about the expansion of the notion of "empowerment" itself, however, it may be useful to consider briefly a very recent treatment. In a short section of their well-known encyclopedia, Baker and Jones (1998) note that linguistic and cultural empowerment in the classroom is a catalyst for success and that, where parents can also be involved, the benefits may extend into the local community; they suggest that such success may perhaps occur via the intermediate step of enhanced

22

John Edwards

self-esteem, "positive, valued and honoured identities", and the production of "a sense of ownership" of education. Now, "self-esteem" and its concomitants have become a sacred cow and, like the real animal, often a barrier to movement - but arguments are now increasingly raised that too much emphasis on the alleged fragility of the young psyche, and too little on more traditional concerns for learning, is not a good thing for children in general, and particularly harmful for the "disadvantaged". 5 This is not, of course, to deny the very real value of multicultural awareness in education - in fact, I have noted elsewhere that all good education worthy of the name must be multicultural. Equally, a well-structured personal and group identity is presumably a good thing, and there is evidence to suggest that the very exercise of empowerment can, itself, be empowering in this regard - specifically, with regard to perceptions of dialect legitimacy. Heightened teacher sensitivity to dialect variation in the classroom and the reproduction of nonstandard language in printed form may be psychosocial^ useful. But if this is "empowerment", it is of a mild form, there is no evidence linking it directly to enhanced achievement (notably in reading, of course) and, most importantly, there is traditionally very weak support for it among the intended beneficiaries themselves (see Edwards 1989). But apart from the supposed linkage between empowerment and selfesteem, Baker and Jones go on to discuss other matters. They note, for example, that traditional testing procedures "tend by their very nature to locate problems in the individual student ... [and] may fail to locate the root of the problem in the social, economic or educational system"; improved assessment procedures, based upon "advocacy", will focus upon the "system"; and so on (p. 542). These matters, while not necessarily connected with considerations of "self-esteem", typically accompany them in discussions of this sort. Educational "empowerment" is thus seen to involve criticism of "the system". This is a reasonable corollary, it seems to me, but it is typically undercut (as it is in this case) by an immediate refocussing upon too narrow a context. That is, just as discussions of language shift often take place in curiously disembodied fashion (see above), so here we find a ritual nod in the direction of "changing the system" but a quick return to a decontextualised accent upon the school alone. The ritual nod often takes the form, as in Baker and Jones's article, of some sort of observation that the school can't go it alone: "empowerment can be furthered by education, but also needs to be realized in legal, social, cultural and particularly economic and political events" (p. 542). This glib rounding up of

The power of language, the language of power

23

the usual suspects is worse than not mentioning them, for the implication is that, after all, they collectively constitute only some sort of minor extension to the real arena of action. Nothing, of course, could be more inverted. The other difficulty - and, again, it is a regrettably typical one - revealed in the foregoing citations also involves "the system." While, as I have just noted, it is quite plausible to suggest that empowerment requires some examination of the system, it is not so reasonable to proceed on the assumption that the social context is to be blamed for problems illuminated through testing. This is, admittedly, a delicate matter, for it is a truism that a prejudiced or oppressive society extends a generally deadening hand over its disadvantaged segments. But at the same time, it can be all too easy to inaccurately lay all problems at the social, and not the individual, foot. The delicacy of the point here is sharpened when one considers that the attempt to lessen the burden of the disadvantaged individual in this way often proceeds from good intentions - but it is, once again, an example of empowerment-as-compensation. As such, it can be seen as condescending and demeaning, as actually reinforcing a debilitating scenario of victimisation and passivity, as failing to accomplish any real playing-field levelling, as (at best) a delaying tactic rather than a solution - and, overall, as an example of attitudes and behaviour increasingly rejected by those most directly affected (see Ν 5).

4.

An argument for "real" empowerment - and the question of "will"

Bestowing power - like enforcing tolerance or granting freedom - has a curious ring to it, because the action of the verb is antithetical to the meaning of the noun; curious, too, because the action is essentially "top-down" and might therefore be reneged upon. But taking or assuming power, besides presumably being a better boost to "self-esteem", is a firmer bulwark. The problem, of course, is how subaltern groups can reasonably expect to take power - after all, relative impotence is their defining quality. An answer might lie in two terms encountered above: acquiescence and will. It will be recalled that a fairly wide-spread passive acquiescence, on the part of the population-at-large, is a minimum requirement for efficient and stable government. Ruling elites can, of course, actively and continuously impose their authority on restive groups, but it is naturally some-

24

John Edwards

might be more profitably placed elsewhere. An understanding of this is the classic motivation for resistance and guerrilla movements and, more generally, it reflects the latent power of the masses. (This will remind some of Richard Nixon's appeal to the "silent majority" of American voters ... but that ought not put us off!) In authoritarian regimes there is little likelihood that this passive quantity can be transformed into something more dynamic without a heavy, and often impossible, price. But in liberal democracies - obliged by their deepest principles to attend and respond to cultural pluralism and social heterogeneity - the price drops considerably. There are, of course, practical impediments to change, and the most central one here has to do with the minority status that characterises the groups under discussion. But, as Erik Allardt once pointed out (1984: 203), some minority groups are more minor (sont plus minorises) than others, and a corollary must be that some are stronger. And, as the current American experience demonstrates, social disadvantage can actually be more marked among larger minorities: each currently numbering about 35 million persons, the Black and the Hispanic groups together comprise about 30% of the total population - and, whether considering numbers or percentages, it is clear that they are significant minorities. It is also obvious that their presence has had, and continues to have, important ramifications in daily life, particularly of course where, beyond numbers and percentages, a third factor is also strong: concentrations of population. It is clear, then, that for some traditionally disadvantaged minority groups, numbers and concentrations are sufficient, in liberal-democratic societies, for passive acquiescence to be transformed into a more socially aggressive stance. And this is where we recall that second term: will. The argument, then, is quite simple: given sufficient group will, power can be assumed, not just received. An obvious demonstration of this, in capitalist America, is the accommodating response of the market-place to speakers of nonstandard or non-English varieties. And if the power of the purse, why not the power of the ballot-box, of the educational amendment, of the Californian proposition? But of course things are not quite this simple. The sufficiency of will required for the assumption and subsequent exercise of power has been more evident in some areas than in others - and, to bring things squarely back to matters of language and culture, those groups whose clout is evident in, say, economic areas seem not to have been as demanding in others. Why not? There is, of course, simple inertia, an inherent problem wherever passivity is to be galvanised into action. There are obvious rea-

The power of language, the language of power

25

sons for this, most of them having to do with lack of sufficient awareness, coupled with the economic and pragmatic imperatives that affect ordinary life (for everyone, of course, but probably more centrally for those who are of subordinate or disadvantaged status). So, it often proves difficult or impossible to translate a rather inert goodwill into something more dynamic. (Language revival efforts, for instance, are characterised by a small group of activists nervously glancing over their shoulders to see how many of their alleged adherents are following them.6) It is also possible that populations have been "taken in", as it were, by mainstream groups, so that they no longer know or trust their own linguistic and cultural instincts. Many years ago, for example, Wally Lambert and his associates - drawing upon earlier social-psychological work showing how negative, authoritarian and prejudiced evaluations of stigmatised social and religious groups were sometimes replicated within these groups themselves - described a "minority-group reaction" by which "small" linguistic communities may come to accept that their language has less favourable connotations than that of some larger surrounding population (Lambert et al. 1960; see also Gardner and Lambert 1972). While these sorts of explanations imply a group inadequacy that action in other arenas makes unlikely, it is the case that "the lack of will to stop shrinking is an intrinsic characteristic of a shrinking language community" (Fennell 1981: 30). An acquired frailty of will is perhaps a more general manifestation of Lambert's finding (above); it is certainly a deeper and more subtle manifestation than any superficial listlessness and, even if it is only restricted to some areas of psychosocial life, it presents a gritty problem. It reflects, in fact, powerful factors already noted here, notably the contact between unequal groups, communities or systems, and the socioeconomic changes set in train by this contact. In an earlier summary, I noted that, among virtually all groups, there is (Edwards 1995: 115): a powerful concern for linguistic practicality, communicative efficiency, social mobility and economic advancement. These are the greatest advantages possessed by dominant varieties and the greatest disadvantages of smaller ones ... a decline in the existence and attractions of traditional life styles also inexorably entails a decline in languages associated with them. If "will" might, however, be galvanised in some circumstances, how important can it really be? At the 2004 LAUD symposium, for example, it was argued that it was an imprecise concept, and that examples of its explanatory power would therefore be hard to delineate. I pointed out that the Quebec case was perhaps an apposite one. After enduring long years

26

John Edwards

of sociopolitical and religious paternalism, the francophone population experienced a revolution tranquille, transformed and modernised itself, and assumed the provincial mastery that its inherent strength had always promised; an important corollary of the transformation was linguistic engineering on behalf of a French language considered to be at risk. Another symposium speaker, Bernard Spolsky, has commented upon this situation in a new book on language policy: he writes of francophones beginning to become "conscious of English dominance" (2004: 196). Elsewhere in the book he uses terms like "commitment" and "ideological support" (p. 205), states bluntly that "language policy is about choice" (p. 217) and emphasises the importance of the "perception" of sociolinguistic situations (p. 219). These usages are not all, or always, synonymous with "will", of course, but they all suggest how important convictions, attitudes and perceptions are in matters of language and power.7 Returning to "will" itself, one or two even more immediately relevant examples suggest themselves. At the LAUD symposium, for instance, Joshua Fishman - one of the critics of the utility of the concept - presented a chatty plenary address entitled "Sociolinguistics: More power(s) to you!" (this volume), part of which consisted of a list of many intangible aspects of power and its operation - virtually all of which could just as easily have been described in terms of the operation of will. More pointedly, he made reference to the decision in his own family to create and maintain a Yiddish-speaking home. This is quite obviously an illustration of conscious will-power at work, of a decision taken on grounds of conviction rather than practical necessity - and a personal reflection of that broader and often-expressed argument about the importance of regular family transmission of languages from one generation to the next. If one family can make certain language choices, then others might do so as well, and to be able to extrapolate from the family to the community would clearly be of the greatest impact in the life of "threatened" varieties. (Indeed, as Spolsky [2004: 40] has recently pointed out, while application of any language policy requires a community, that community can be "of whatever size ... ranging from a family ... [to] nation state or regional alliance"). Fishman has also made frequent reference to will in his writings; throughout his (1991) volume on the reversal of language shift, for example, he returns repeatedly to the theme in one form or another. Thus, he observes that the success of the re-vernacularisation of Hebrew rested upon "the rare and largely fortuitous co-occurrence of language-andnationality ideology, disciplined collective will, and sufficient social dis-

The power of language, the language ofpower

27

location" (1991: 291). In discussing efforts on behalf of Frisian, he says that "the basic problem seems to be in activating this [passive] goodwill" (p. 180). Bemoaning the disregard of the "moral and spiritual dimensions of modern life" (p. 387) in his treatment of language-shift reversal, Fishman sees the movement as helping to re-establish "local meaning" in the face of a "mechanistic and fatalistic" outlook (p. 35). Successful reversal involves involvement in "the qualitative emphases" (p. 8) of contemporary life. And, in an earlier piece, Fishman wrote approvingly (in a discussion of Herder) of those peoples who have not "capitulated to the massive blandishments of western materialism, who experience life and nature in deeply poetic and collectively meaningful ways" (1982: 8; see also Edwards 1993). The invocation of the concept of "will" is surely also accurate when we consider the actions of those strongly committed to the maintenance and revival of at-risk languages - the nationalists, activists and enthusiasts are typically few in numbers but fiercely committed to their linguistic cause. Consider the Cornish and Manx revivalists, or those native anglophones who move to the gaehachtai of Ireland and Scotland, or those who carry the banners for Gaelic in Cape Breton Island, and so on - there are a great many apposite cases here. The other side of this coin - and the one that often gives the activities of revivalists their poignancy - must obviously be the will of those who choose not to move to minority-speaking enclaves, or to bring up their children in some threatened medium, or to otherwise encourage it. Now, it might be thought that this second category is not particularly interesting or illuminating, representing merely passivity, non-exercise of will, or a decision to not make a decision but, rather, to drift along in some prevailing "mainstream". In fact, however, there are contexts in which conscious decisions unfavourable to minority languages, on the part of potentially important players, are equally illustrative of the power of active will. It is a testament to the depth and sensitivity of the 2004 LAUD symposium that one of the most important of these contexts - the post-colonial setting - was extensively discussed, notably by Africans and Africanists. It was frequently pointed out, for example, that one consequence of colonialism is that the elites in newly-independent countries have typically been educated abroad - most often, of course, in the language of the former colonisers - and continue to value that language more highly than indigenous varieties. When it comes, then, to encouraging local vernaculars and their development, or opting for the mediums of education, the mindset of

28

John Edwards

those in power is - or so it is alleged - still stuck in a linguistic rut; the operation of their will stifles local languages. Given the great divides that often exist between the rulers and the ruled, the implication is that a change in that mindset, a recalibration of that will, could have profound consequences for those large numbers who are linguistically and educationally excluded from the corridors of power, whose languages remain widely used but unfairly reined in. It is, of course, of the greatest significance that the exercisers of will in these circumstances are indigenous individuals themselves - elite maybe, and socioeconomically far removed from the vast majority of the citizens but, nonetheless, unquestionably of the place. These are not callous outsiders whose language policies, however reprehensible, are understandable in the traditional colonial context; they are people of whom more might have been expected and, indeed, people who have often fulfilled the social and political expectations of them in other matters. The poignancy of all this has been eloquently discussed in a number of essays by Ngügi wa Thiong'o (1986, 1993). His own conscious decision to write in Giküyü, and not in English, is the personal background to his many impassioned pleas for the linguistic and cultural "decolonising" of the African mind - and to his indictment of those in power whose minds apparently remained colonised. The eminent Nigerian writer, Chinua Achebe, spoke for many African authors when he noted (cited in Ngügi wa Thiong'o 1986: 7): Is it right that a man should abandon his mother tongue for someone else's? It looks like dreadful betrayal and produces a guilty feeling. But for me there is no other choice. I have been given the language and I intend to use it. For Ngügi wa Thiong'o, however, Achebe's position is morally untenable: he bluntly notes that "African literature can only be written in African languages" (p. 27 - and he does not accept the argument, by the way, that European varieties have become African ones). As a writer himself, NgügT wa Thiong'o is particularly concerned with "decolonisation" in literary and dramatic contexts, but he touches upon political ones as well. He mentions Leopold Senghor, for instance, who admitted that French had been forced upon him but who yet remained "lyrical in his subservience" to the language (1986: 19). Or Hastings Banda, who created an elite Englishlanguage academy in Malawi, expressly designed to encourage able students to be sent to "universities like Harvard, Chicago, Oxford, Cam-

The power of language, the language ofpower

29

bridge and Edinburgh" (p. 19); all the teachers were recruited from Britain.8 To summarise: while I certainly do not endorse all the sentiments noted here concerning linguistic and cultural "will", the fact of its importance and the importance of other, related psychosocial factors - can hardly be denied. The importance rests, ultimately, upon matters of perception, the demonstration of whose centrality to social life is the single greatest insight of modern psychology. And matters of perception are not necessarily matters of fact. We are dealing here with intangibles, but what is intangible is often the strongest and the most resistant to change - or to cultural adversity or endangerment.

5.

Conclusion

The essence of cultural and linguistic "empowerment" is no longer (at least not in democratic societies - and the reflexes of others do not figure here) a matter of authoritarian or legislated denials of some desired social alteration. Rather, it is the scope and the depth of that desire itself So, even if we discard most of the existing conceptions of empowerment, on the grounds that they are patronising and often half-hearted attempts to compensate for the inequities of a system which is not about to change, and even if we were to support and encourage a more direct or aggressive accession to power, we would still be naive to assume that major alterations in the linguistic and cultural landscape would be forthcoming. For such an assumption would ignore the subtle interweavings of a sociohistorical fabric that has many more colours than simple black and white. It would ignore the fact that those who argue for change typically want only some selected reweaving of this fabric - not a wholesale reworking - and any such change inevitably involves all the connecting threads. And it would ignore, above all, the evidence of the world around us, evidence that suggests that the social choices of relatively disadvantaged groups are made on exactly the same principles used by others. To say that these are often Hobson's choices, that the action of choosing means little if one horn of a dilemma is obviously sharper than the other - well, this is only to describe a very common consequence of social, economic, political and other pressures, and certainly not one unique to any particular community.

30

John Edwards

Notes 1. A similar sentiment was expressed later by Ortega y Gasset (1967). In a world where the old order is changing in response to the inexorable rise of the "masses", Ortega argued that bizarre and frivolous agitations can be expected: "de aqui el vibriönico panorama de «nacionalismos» que se nos ofrece por todas partes" (1967/1930: 218). (Interested readers may like to investigate the unusual word which Ortega puts in italics here.) 2. Almost all of the important philosophers have of course discussed power notable here are Hobbes, Spinoza and Hume. For important post-Weberian, social-scientific considerations, see the writings of Carl Friedrich (1963) and Harold Lasswell (Lasswell and Cleveland 1962; and Lasswell and Kaplan 1965). 3. We are obviously not very far here from matters treated more fully under headings like "rational-choice" or "instrumentalist" theories of nationalism and ethnocultural accommodation (see several recent key articles in Leoussi 2001). 4. This symbolic English potency is not restricted to the written word. Cheshire and Moser (1994: 451-452), for example, discuss Haarmann's (1989) finding that the use of English in Japanese television advertising involves "no expectation that viewers will understand what they see or hear" (my italics). More generally, they point out - as I have done here - that English is working "not as a system of signs, but as a sign itself." In some contexts - Murray (2003) illustrates the Swiss case - the use of English in advertising avoids having to make invidious choices among indigenous varieties. This is a minor manifestation of the way in which "big" languages acquire further muscle by acting - bizarre though this seems to many - as a relatively neutral medium in linguistically heterogeneous settings, even those in which such varieties had earlier been colonial impositions. 5. Twenty-five years ago, in fact, the West Indian scholar, Maureen Stone, was arguing that "multiracial education" was "conceptually unsound ... while at the same time creating for teachers, both radical and liberal, the illusion that they are doing something special for a particularly disadvantaged group" (1981: 100). In a similar vein were the comments cited by Sally Tomlinson (1984: 149): "Black parents don't want black studies or multicultural education for their children ... black pupils need to be good at science, history, geography ..." 6. Further notes about language-revival difficulties - problems of a generalisable nature - can be found in my Multilingualism (1995: see particularly Chapter 4, "Languages in conflict"). 7. It was Joshua Fishman who, at the conclusion of the talk upon which this chapter is based, raised the points about the imprecision of "will" and its consequent lack of explanatory value. It is a bit ironic, then, that some of the examples I go on to illustrate in the present chapter are drawn from Fishman himself. The discussion of "choice" was prominent at the LAUD symposium. While it

The power of language, the language of power

31

was given its most immediate emphasis in Florian Coulmas's plenary address, "The power to choose and its sociolinguistic implications" (this volume), the concept - like "will" - was frequently referred to (directly or indirectly) throughout the conference. 8. At the LAUD meeting, Karsten Legere discussed the impact of "J. K. Nyerere and the empowerment of Swahili of Tanzania". Even this more-than-usually committed statesman was apparently unable to bring himself to endorse Swahili as a medium for post-primary education. Neville Alexander, in his "Sociopolitical factors in the evolution of language policy in post-apartheid South Africa" (this volume), remarked that altering political mindsets - hegemonic mentalities produced by longstanding exogenous pressures - even in a dynamic and reinvigorated setting, usually takes a considerable (and, to many, a frustrating) amount of time. And Paulin Djite, in his "Living on borrowed tongues - a view from within" (this volume), expressly noted the lack of sufficient African will to think beyond European varieties.

References Alexander, N. this volume Sociopolitical factors in the evolution of language policy in postApartheid South Africa. Allardt, E. 1984 Bacon, F. 1597

What constitutes a language minority? Journal of Multilingual and Multicultural Development 5: 195-205. Essays: Religious Meditations. Places of Persuasion Disswasion . Seene and Allowed. London: Hooper.

and

Baker, C. and S. Jones 1998 Encyclopedia of Bilingualism and Bilingual Education. Clevedon: Multilingual Matters. Benn, S. 1972

Power. In Encyclopedia of Philosophy. Volume VI, P. Edwards (ed.), 424-427. New York: Macmillan.

Cheshire, J. and L.-M. Moser 1994 English as a cultural symbol: The case of advertisements in French-speaking Switzerland. Journal of Multilingual and Multicultural Development 15: 451-469.

32

John Edwards

Coulmas, F. this volume

The power to choose and its socio-political implications.

Djite, P. this volume

Living on borrowed tongues? A view from within.

Edwards, J. 1985

Language, Society and Identity. Oxford: Blackwell.

1989

Language and Disadvantage. 2nd ed. London: Cole & Whurr.

1993

Language revival: Specifics and generalities. Studies in Second Language Acquisition 15: 107-113.

1995

Multilingualism. London: Penguin.

2003

Contextualizing language rights. Journal of Human Rights 2: 551-571.

Fennell, D. 1981

Fishman, J. 1982

Can a shrinking linguistic minority be saved? In Minority Languages Today, E. Haugen, J. McClure and D. Thomson (eds.), 32-39. Edinburgh: Edinburgh University Press. Whorfianism of the third kind. Language in Society 11: 1-14.

1991

Reversing Language Shift. Clevedon: Multilingual Matters.

this volume

Sociolinguistics. More powers to you! (On the explicit study of power in sociolinguistic research).

Flusser, V. 2003 Friedrich, C. 1963

The Freedom of the Migrant: Objections to Nationalism. Urbana: University of Illinois Press. Man and his Government: An Empirical Theory of Politics. New York: McGraw-Hill.

Gardner, R. and W. Lambert 1972 Attitudes and Motivation in Second-Language Learning. Rowley, Massachusetts: Newbury House. Haarmann, Η. 1989

Symbolic Values of Foreign Language Use: From the Japanese Case to a General Sociolinguistic Perspective. Berlin: Mouton de Gruyter.

The power of language, the language of power

33

Lambert, W., R. Hodgson, R. Gardner, and S. Fillenbaum 1960 Evaluational reactions to spoken languages. Journal of Abnormal and Social Psychology 60: 44-51. Lasswell, Η. and Η. Cleveland (eds.) 1962 The Ethic of Power: The Interplay of Religion, Philosophy and Politics. New York: Harper. Lasswell, Η. and A. Kaplan 1965 Power and Society·. A Framework for Political Inquiry. New Haven, Connecticut: Yale University Press. Legere, Κ. this volume

J.K. Nyerere of Tanzania and the empowerment of Swahili.

Leoussi, A. (ed.) 2001 Encyclopaedia Transaction. Mackey, W. 1978

Mill, J. S. 1964 Murray, H. 2003

of Nationalism. New Brunswick, New Jersey:

The importation of bilingual education models. International Dimensions of Bilingual Education, J. Alatis (ed.), 1-18. Washington: Georgetown University Press. Considerations on Representative Government. London: Dent. Original edition, London. Parker & Bourn, 1861. The status of English in Switzerland. In Living with Languages: The Contemporary Swiss Model, J. Charnley and M. Pender (eds.), 48-71. Berne: Lang.

NgugTwa Thiong'o 1986 Decolonising the Mind: The Politics of Language in African Literature. London: Currey. 1993

Moving the Centre: The Struggle for Cultural Freedoms. London: Currey.

Ortega y Gasset, J. 1967 The Revolt of the Masses. New York: Norton. Original edition, Madrid: Revista de Occidente, 1930. Spolsky, B. 2004

Language Policy. Cambridge: Cambridge University Press.

34

John Edwards

Stone, M. 1981 Tomlinson, S. 1984 Weber, M. 1968 Weinreich, Μ. 1980 Wyndham, J. 1960

The Education of the Black Child in Britain: The Myth of Multiracial Education. London: Fontana. Home, school and community. In Education and Cultural Pluralism, M. Craft (ed.), 66-82. London: Falmer. Economy and Society: An Outline of Interpretive Sociology. New York: Bedminster. History of the Yiddish Language. Chicago: University of Chicago Press. The Midwich Cuckoos. London: Penguin.

Language endangerment, the construction of indigenous languages and world English Janina Brutt-Griffler

1.

Language endangerment and its discursive practices

Over the past few years, there has been an attempt to draw linguists into the political arena to defend the objects of their study - languages. The recent upsurge of interest in 'endangered' languages has had as its aim to call the attention of linguists, scholars, policy makers, and the public at large to the potential disappearance of a large proportion of the world's linguistic heritage (Bradley and Bradley 2002; Crystal 2000; Mühlhäusler 1996; Nettle and Romaine 2000). Conveyed in discourse often borrowed from the environmentalist movement, linguistic preservationism seeks to change the linguistics profession from its traditional academic to the less familiar activist role of helping to safeguard language resources. There is, of course, need for linguists, like all scholars, to play a role in public policy and discourse, as Pratt (2003) has argued. But there are potential pitfalls as well. In entering the political arena, in which finer points of theory may appear remote and obscure, the complexities of these processes can easily be lost sight of in the attempt to fit the case of language endangerment into the established categories of environmentalist political discourse. Consider for example, the claim put forward in the introduction to a recent volume entitled Language Endangerment and Language Maintenance: "The globalisation of English and the spread of other national languages are not so different from the spread of new genetically modified plant varieties controlled by multinational companies" (Bradley and Bradley 2002: xii). It is clear why a movement patterning itself after environmentalism would want to theorize World English in such terms. Attacking corporations as the prime threats to the environment and the public health has proven to be an effective strategy, given the abundance of evidence supporting that notion. In conceptualizing the terms of the struggle as identical at the same time it appropriates environmentalist discourse, linguistic preservationism attempts to reap the environmentalist movement's widespread credibility as well - although the former lacks the same body of

36

Janina Brutt-Griffler

evidence buttressing its contentions. It is a strategy all the more likely to succeed because it is the sort of explanation of the process of language endangerment we would tend to come up with if we were asked to construct an explanation before we had actually investigated it. After all, the rise of World English and the decline of a large proportion of the world's languages have largely coincided. Why not make the link direct? Nettle and Romaine (2000: 5) do just that in Vanishing Voices: "Some have used the terms 'language murder' and 'language suicide,' suggesting that languages do not die natural deaths. They are instead murdered. English, as Glanville Price puts it, is a 'killer language.'" To bolster this point, they cite the case of Africa, in which English spread, they assert, "is leading to the top-down displacement of numerous other tongues" (Nettle and Romaine 2000: 144). The political campaign represented by the linguistic preservation movement has relied on constructing an untenable theorization of World English to fit its political narrative. The focus within linguistics on endangered languages has fostered a scholarly and language policy discourse predicated on an unproblematic dichotomy of dominant and indigenous languages. In that conception, language policy is most often projected as an instrument of ensuring the survival of 'indigenous' languages from the encroachment of 'dominant' languages, principally English, thereby safeguarding the rights of ethnic communities and preserving their sociolinguistic diversity. This narrative of language endangerment and the theorization of World English on which it depends do not at all adequately describe what is happening in Africa and other parts of the world today. Scholars such as Mufwene (2002) and Mazrui (2004) have shown that indigenous languages like Swahili - and not English - have posed and continue to pose the greatest threat to indigenous languages in nations throughout Africa. This paper argues that, all the more unexpected according to the dominant narrative in the field - in Southern Africa it is not the spread of English but the steps taken in the colonial period and today to halt its spread that constitute the greatest challenge to the indigenous languages of the region. A consideration of this process offers important insight into the complexities of bilingual speech communities and language change that get lost in the political discourse for public consumption.

Language endangerment and world English 2.

37

The construction of indigenous languages: The sociohistorical context

The argument that 'indigenous' languages are in need of special protection explicitly or implicitly invokes the notion that they constitute the authentic products of indigenous cultures. Recent work has shown that such seemingly common sense assumptions require substantial modification in light of the complex sociolinguistic legacy of colonialism. In southern Africa, investigations of such languages as Tswana, Zulu, Xhosa, Shona, and Ndau reveal a far more ambiguous picture. Scholars who have closely investigated the history of the region's languages have concluded that these 'indigenous' languages as we know them are inextricably linked with the European colonial project (Brutt-Griffler 2002; MacGonagle 2001; Makoni 1998; Mufwene 2002). Their view is succinctly summarized by Lelyveld's (1993) conclusion with respect to India: "The very languages that are called 'native' are products of an intricate dialectic between colonial projects of knowledge and the formation of distinctive group identities" (in Breckenridge and van der Veer 1993: 6). To understand how that might be, it is necessary to expose the myth that has been widely disseminated by the school of linguistic imperialism that the language policy goal of the British empire was to spread English for purposes of an ideological agenda (Phillipson 1992). Brutt-Griffler (2002) demonstrates that the basis of colonial language policy was actually quite different. It involved the limitation of the spread of English that inevitably accompanied British political and economic rule within the narrow bounds of the indigenous elite of African and Asian colonies necessary for the functioning of colonialism. Far from encouraging the spread of English, British colonial policy was designed to prevent it. Toward this end, colonial education was based on the use of non-English linguistic mediums which were neither necessarily indigenous nor mother tongues. The purpose was clear and unambiguous: to educate the majority within colonized Africa and Asia for a life of manual labor, producing the raw materials on which metropolitan European industry depended. Specifically, it was to ensure that middle class branches of employment, which necessarily required a knowledge of English, were beyond the grasp of the vast majority of the population. This containment policy was directed toward the end of keeping the non-English speaking majority separate from the Englishspeaking elite by a system of linguistic exclusionism.

38

Janina Brutt-Griffler

As part of this colonial language project, the British regimes of Southern Africa undertook to manufacture national languages in some places to correspond with colonial boundaries (Brutt-Griffler 2002). It was a prodigious task. Given that such colonial borders were generally arbitrarily drawn, they grouped together large numbers of ethnic groups that spoke many languages. The persons charged with the task of creating standard languages out of these heterogeneous parts were not linguists, nor often even colonial officials, but missionaries, whose proficiency in the languages in question could be virtually nonexistent. One of the missionaries who took part in the work of laying the foundations of modern Shona, for example, relied on an English-speaking Mozambican teacher who knew Zulu, who in turn relied on a few children as informants (Chimhundu 1992). The direct sources of the constructed language were therefore, at best, second language speakers of the language they were recording. Ironically, the native speakers of Southern African languages were displaced and rendered irrelevant to the process of codifying 'their own' languages, the alleged repositories of their cultural authenticity. 2.1.

Specific cases of colonial construction of indigenous languages

Faced with a complex linguistic environment featuring continua that defied easy categorization, European missionaries and colonial authorities differentiated or grouped together what they saw as 'dialects', often based on rudimentary understanding and superficial investigation of the actual linguistic situations. For example, in Southern Rhodesia (present-day Zimbabwe), a colonial commission decreed that the "four or five languages" spoken in the part of the colony they named Mashonaland were really "four or five dialects of one language." The committee of missionaries and colonial officers, under the guidance of a South African-born linguist who lacked any proficiency in the languages in question, created a vocabulary, choosing representative words from each of the different languages, and also worked out a standard grammar (Brutt-Griffler 2002: 81-82). The colonial commission opted to call the language it had created Shona. It is interesting to note their own impression of how the choice of name was received: It has been widely felt that the name 'Shona' is inaccurate and unworthy, that it is not the true name of any of the peoples whom we propose to group under the term' Shona-speaking people' and further, that it lies under strong suspicion of being a name given in contempt by the enemies of the tribes. It is pretty certainly a foreign name, and as such is very likely to

Language endangerment and world English

39

be uncomplimentary. It is true that the name 'Mashona' is not pleasing to the natives, but that may simply be because it is a group name imposed from without, and ignoring all true tribal distinctions. Certainly no people in the country claim the name Mashona as their tribal name, and each would prefer to be described by the proper name of his particular group (quoted in BruttGriffler 2002: 83). A similar process took place in numerous cases throughout Southern Africa and the British empire. Harries (1998) has shown that missionaries in South Africa's eastern Transvaal forged Tsonga, Ronga and other languages out of diverse speech forms of a refugee population in southern Africa, as in the case of Shona, creating both new linguistic and political identities. In Nigeria, missionaries from Sierra Leone codified a group of languages under the name Yoruba, itself a Hausa word not familiar to those whose language it purportedly described. On the other hand, colonial officials opted to make separate languages of Zulu and Xhosa, though the two were as close as the languages "unified" in other cases (Irvine 1993: 28). The same process has been noted in India (Lelyveld 1993). 2.2.

The implications of a socially constructed indigenous language on its speakers

The emergence of varieties of colonial indigenous languages as a consequence of colonialism was not lost on the Africans for whom they were to henceforth serve as mother tongues, as evident in the names they gave to the languages they were being assigned to speak. For example, the variety of Shona associated with evangelical work in Zimbabwe was generically referred to as 'chibaba '— the language of the priests (Jeater 2001), or more specifically, depending on the geographical region in which the missionaries were working, 'Church Manyika Language' (Chimhundu 1992). In the Republic of the Congo, the Kituba associated with the colonial state was referred to as 'kikongo ya leta' (The kikongo of the rock) and 'kikongo ya bula matari' (The kikongo of the stonebreaker) (Mufwene 2001: 176). Important to their eventual entrenchment as written standards in Southern Africa, these languages were taught to colonial authorities and used by them in their interaction with Africans. The construction of European versions of indigenous languages represented a systematic effort to develop command over language in order to develop a language of command, as one South African Police officer aptly put it in a 1969 interview. He articulated the powerful political reasons for seeking to exercise control over

40

Janina Brutt-Griffler

languages at the time of his joining the police in 1907: "Unless you can talk the person's language, how can you tell him what to do, and interpret the law to him?" (Jeater 2001: 451). There was, for instance, a two-tiered salary system depending on whether the civil servant had passed examinations in 'indigenous' languages. Missionaries, too, were required to demonstrate proficiency in Southern African languages, as demonstrated by a 1909 dispatch from the United American Board of Missions stipulating that "every new missionary, missionary wives included, should not be allowed voting rights until they had passed a detailed examination in the local vernacular" (Jeater 2001: 455). Civil servants in the British Empire were also required to pass such examinations. Interestingly, some of the earliest tests in southern African languages in the 1890s were those for colonial agents. Not surprisingly in light of their conspicuous exclusion from this process of codification, there was, if any, only the most limited input by Africans in the construction of the language tests in 'their' languages. Whatever input they might have was 'mediated' by the European linguists whose knowledge of the speech forms they were describing was limited (cf. Makoni 1998). The 'indigenous' languages produced by colonial administrations throughout Southern Africa and other parts of Africa were designed to constrain fluid identities within colonial contexts to facilitate European rule, creating both new linguistic and political identities. The political implications would be profound. These written languages - produced as much by colonial agency as by Southern African, bearing at times little resemblance to the spoken language of the region's peoples-became, in effect, "mother tongues in search of speakers" (Makoni and Brutt-Griffler forthcoming). The languages that emerged out of this process, named, codified and standardized by imperial agents, were then assigned to Southern Africans in what could amount to rather arbitrary ways in pursuit of the British Empire's containment policy of limiting the spread of English. Since this colonial language policy was carried out largely through what was euphemistically then and now called 'mother tongue' education, these 'vernaculars' became the basis of the primary school education that was alone available to the vast majority of Africans. While the set of assumptions on which the linguistic preservationist operates includes the notion of the imposition of English as part of a European policy of linguistic imperialism, the colonial reality in Southern Africa was often one of the European imposition of these European generated versions of African languages.

Language endangerment and world English

3.

41

Contemporary language contact phenomena and their relevance to language endangerment discourse

If we ended the story here, it might appear as nothing more than a rather curious episode from the colonial past. After all, every language has its history. And interpretations of the significance of this colonial legacy of the most widely-spoken of Southern Africa's indigenous languages are varied - from the contention that they constitute European inventions that need to be disinvented (Makoni 1998) to the view that the process, aside from the unusual characteristic of exogenous direction, did not differ that greatly from the standardization of languages in other parts of the world, and thus, whatever their past, linguists must strive to preserve them as they are today. And so we return to the political narrative of linguistic preservationism, often couched in language that implies the hegemony of English and other 'dominant' languages over 'indigenous' languages (Nettle and Romaine 2000; Skutnabb-Kangas 2004). It prescribes measures to address this imbalance, such as those taken in post-apartheid South Africa, where the constitution prescribes eleven official languages, nine 'indigenous' languages alongside English and Afrikaans. Although sometimes cited as a model of the kind of needed recognition of indigenous languages necessary to their preservation (e.g., Phillipson 1999), this language policy bears the unmistakable traces of the colonial project of finding speakers for the nation's nine purported 'indigenous' mother tongues. Widdowson (2003: 57) has cogently remarked, "The fate of languages is a function of socio-economic and political circumstances, and to change their fate is to change these circumstances." To understand what have increasingly come to be called language ecologies (Mufwene 2001; Mühlhäusler 1996) is to do so in the context of these socioeconomic and political conditions. In Southern Africa, it means first and foremost analyzing the linguistic consequences of the political economy of migration. In 1990, according to World Bank statistics, 69% of the population of the fifty poorest nations worked in agricultural production (World Bank 1997). Amin (1990) has accurately described the rural locations in which such production takes place as the Fourth World, because of the complete absence of the infrastructure we associate with the first three. In South Africa, the nation with the world's second greatest wealth disparity, the Fourth World conditions of the rural semi-arid regions into which apartheid forced most Black South Africans stand in stark geographical juxtapo-

42

Janina Brutt-Griffler

sition to the First World conditions found in the previously all-white suburbs of its major cities. In a desperate attempt to escape their poverty, millions of South Africans have migrated to urban centers. The percentage of the South African population living in urban areas has increased from 50% in 1991 to 58% in 2001 (World Bank World Development Indicators Database). It is projected to increase to 75% by 2020 (Human Rights Watch World Report 2003). The same phenomenon can be found throughout the southern African region, the continent, and the so-called developing world. Making the transition to an urban environment, however, does not mean moving to the white suburbs. With the repeal of apartheid legislation, migrating Black South Africans have met a barrier just as impenetrable: socioeconomic class. Migrants end up in one of two places, the townships, the formal mass-built settlements created under apartheid to house urban workers, and what are called informal settlements, essentially shantytowns. Migrants to these centers encounter not only the predictable socioeconomic problems of poverty, but in a rapidly urbanizing multilingual nation with no existing indigenous lingua franca, they also encounter communication problems. Apartheid policy restricted Black South Africans to ethnically bounded rural space it euphemistically called homelands. Even in urban areas, it attempted to segregate Black South Africans along ethnic lines, once it found that when ethnicity broke down in the townships it tended to promote the black nationalism so dangerous to white rule. More than the repeal of the apartheid era laws, the mass migrations to the cities over the past decade have increasingly broken down such ethnic enclaves in urban South Africa. Thus languages of wider communication have become urgent necessities for urban dwellers. Few dispute that the language most desired by Black South Africans themselves to fill that role is and has long been English. The student uprisings in the 1970s in Soweto were over the apartheid government's insistence that Black South Africans learn Afrikaans rather than English. Language policy in South Africa, however, continues to disfavor the acquisition of English by the non-elite. Schools where English can be learned to a high level of proficiency are limited to those who can afford to pay the high fees they charge. In the township and rural schools that they can afford, English is offered as an 'additional language,' generally at such a low level that few who enter those schools come out with sufficient knowledge of the language to make it useful to them.

Language endangerment and world English

3.1.

43

Pan-ethnic argots

The linguistic preservationist movement continues to assume that English and other European languages pose the gravest threat to indigenous languages in Africa (cf. Nettle and Romaine 2000), and seem to take it for granted that halting the spread of English and other European languages would deprive Africans of the incentive to abandon their mother tongues. They did not anticipate, and so far seem oblivious to, the condition that the lack of access to English for working class Africans would become a central impetus for the development as second languages via macroacquisition (Brutt-Griffler 2002) of lingua francas known as pan-ethnic argots or urban vernaculars that are replacing the languages they seek to preserve. Viewed historically, the urban vernaculars as lingua francas or pan ethnic argots are intimately bound up with the colonial containment policy and its more recent echoes. South Africa's largest pan ethnic argot, Isicamtho, has spread in the period since the crushing of the student uprisings in Soweto in the 1970s following the apartheid government's insistence that Black South Africans learn Afrikaans rather than English. Its origins bear the unmistakable imprint of colonialism: having drawn from and replaced an earlier Afrikaans-based mixed language, Tsotsitaal (Ethnologue.org). Having only recently become the object of scholarly attention (Childs 1997), these new languages are increasingly accomplishing what linguistic preservationists wrongly fear English is doing: encroaching on and eroding the institutionally recognized indigenous languages at an accelerating rate as post-apartheid South Africa continues to urbanize rapidly. These languages, including non-standard urban varieties of languages like Zulu and mixed languages like Isicamtho, are perhaps on the way to becoming the most widely spoken languages in modern urban South Africa. 3.2.

Language use in urban public transport

One illustration of the process by which these urban vernaculars emerge can be found in urban commuter encounters. Public language use that takes place in urban commuter encounters is of particular importance for two major reasons: First, it allows us to better understand language use in a fluid multilingual and multiethnic community. Second, it provides us with an insider look into language contact phenomena. In Zimbabwe, for example, some 60 to 80 percent of Zimbabweans commute daily to work by small minivans or buses (Mashiri 2002: 21). Such daily commute practice

44

Janina Brutt-Griffler

brings together speakers of a variety of languages and ethnic backgrounds, including Afrikaans, Chewa, English, Ndebele, Shona, Portuguese, and Sotho. Encounters between drivers and passengers provide us with the opportunity to explore the dynamics of communication, and more specifically the use of pan-ethnic argots in a clearly delineated social and linguistic space. The following interaction takes place between a bus conductor (BC) and passengers (P) in urban Harare. ChiShona forms are underlined, and English components are in bold. 1. 2. 3. 4. 5.

BC ·.pindatiende muface (Get in so that we may leave, my acquaintance) BC:pindatiende sisi (Get in so that we may leave, my sister) BC: handei kuback seat (Let us go to the back seat) P(female):/zapa«a kwekuenda bhazi rakazara (There is no more place to go to because the bus is full). BC: One asara'.pinda tiende shasha. Ngatisebedzane bigaz (One left behind, get in so that we may leave, pal) (Mashiri 2002: 122)

This interaction illustrates language use in a public domain that takes the form of an amalgam of chiShona and English, a variety of language that is referred to as chiHarare. The meaning of English words in such a panethnic argot at times radically differs from their meaning in Standard English. For example, the noun phrase muface in line 1 combines the chiShona prefix mu + English nounface. In this context, it is used to mean my acquaintance. The pattern of the chiShona prefix plus the English noun is repeated quite frequently. For example, the phrase kuback seat in line 3 combines the chiShona locative £«+English noun back seat. Another striking feature of the combined chiShona and English discourse is the deployment of kinship and terms suggestive of intimate relations by the crew when referring to passengers. For example, we observe this tendency in the bus conductor's use of terms such as muface (line 1), sisi (sister in line 2), bigaz (literally meaning the senior one in line 5) (Makoni, Brutt-Griffler, and Mashiri, forthcoming). ChiHarare, though it has emerged in domains of intense contact among people representing great sociolinguistic complexity (Childs 1997), is not

Language endangerment and world English

45

tied to urban locales but due to speakers' migration in and out of the city context. As such, it extends its reach to rural places, an important circumstance given the assumption of many language preservationists that such locations hold the key in the preservation of 'indigenous' languages (Nettle and Romaine 2000). Such theories fail to reflect the realities of the political economy of migration that dominates much of the world. 3.3.

Urban vernaculars as a sign of urban identities

Deprived of access to English and the socioeconomic rewards that a knowledge of it entail, many black South Africans manifest a disinclination to use the indigenous languages that are supposed to be their protected mother tongues. This language shift from the recognized indigenous languages to the new ones stems in part from the desire to suppress ethnic affiliations (Makoni and Mashiri, Brutt-Griffler forthcoming, Spitulnik 1998). The emergence of such a tendency is also reported, for instance, in studies into urban language use in West Africa, particularly in Dakar in Senegal (McLaughlin 2001: 170). Both lack of proficiency in urban vernaculars and exclusive use of the identifiable indigenous languages can be interpreted in urban settings as a sign of rural origin and recent migration, one that can determine an economic place at the bottom of the socioeconomic ladder. The implications of this association are far-reaching. Ethnographic research in Southern Africa (Makoni, Brutt-Griffler, and Mashiri, forthcoming) is showing that the so-called urban vernaculars are rapidly spreading to rural areas. The rise of Isicamtho reflects these trends, expressing two of the most central social processes shaping South Africa over the past three decades: urbanization and the defeat of apartheid. South African poet Ike Mboneni Wangu Muila describes the social origins of this urban vernacular that embodies these origins: Isicamtho (so called Tsotsi-taal) is a township communication fire works lingo which originated from the backyard of the then Sophiatown to Emzini, the present day Pimville. It spread all over the townships of Soweto, as far as the present-day suburban areas, as a language of identity amongst the city dwellers and people from rural areas. One could always realise who was a person from rural areas amongst city dwellers by their use of the lingo, which was discovered by the Spoilers Gangsters and the Msomi Gangsters (emphasis added, http://www.cccp-online.org/archive/cccp 12/page_29.html, last accessed December 2003).

46

Janina Brutt-Griffler

Muila notes that this language expresses the coming together of different ethnic groups held separate by apartheid policy. In my bloodstream fountain pen isicamtho set me free from petty language barriers. Isicamtho is a language on its own which draws from and brings together all South African languages that kept people apart.

He maintains that through Isicamtho "we eradicated the DFMalan [the architect of apartheid] policy of racial segregation." Finally, he emphasizes the fluidity and changing quality of urban vernaculars like Isicamtho. He says from it, "a new profound language will emerge, a language of identity, a language of an ordinary person in the street, unity in diversity." Interestingly, conforming the notion that urban vernaculars represent at least in part attempts to escape from rural identities, Muila notes that Isicamtho originally arose among recent migrants from rural locations. The point reinforces the need to understand the language ecology of southern Africa within the context of the political economy of migration. Lexically, Isicamtho draws on African languages such as Zulu, Urban Zulu, Afrikaans, and English. Table 1 illustrates its complex morphological and lexical system. Table 1.

If you rob a bank in the townships, is that [really fighting] for survival?

If u-roof-a

I-bank

ya-se-lokishi-ni

Ε Z-A-Z

Z-E

z-z-uz-z

z-z-c

If pre-roof-sui If you-rob

Pre-bank bank

pre-loc-/oca//o«-loc in-locations

pre-ta -zabalaza you-struggle

u-ya-tabalaza?

('E'=English; 'Z'=Zulu, 'C'=Isicamtho, 'A'=Afrikaans, 'UZ'=Urban Zulu) Childs (1997: 350)

The first line gives the transcription of the sentence as spoken, the second the source languages, the third a partial morphemic analysis, the fourth a gloss of each morpheme, and the fifth a translation. Note that the Isicamtho word tabalaza is derived from the Zulu verb zabalaza ('struggle'), following a phonological rule that has been hypothesized to consist in altering initial consonants to disguise or obscure the word's meaning. Other urban varieties also vary significantly from the standard languages they are derived from. Table 1 and 2 illustrate lexical borrowings in Urban Shona from English, Ndebele, Chewa, and Afrikaans.

Language endangerment and world English Table 2.

47

Lexical Borrowings in Urban Shona from English English borrowings in Urban Shona

bigaz Geliza mafella Table 3.

'big' 'girl' 'fellows'

taimi Tonaz Dhombi

'time' 'town' 'dollar'

finaz kulazi pini

'funeral' 'cool' 'pain'

Borrowings from other African languages kugula ukugula kumala kumala vhati vaat

Ndebele Chewa Afrikaans

'be sick' 'to die' 'water'

(Mawadza 2000: 97)

Some of the words which are borrowed from other languages conform to the Consonant Vowel syllable structure typical of Shona but a 'new' syllable structure is beginning to emerge in which the syllables end in a consonant, for example,finaz introducing more phonological variation in urban vernaculars. There are also morphological differences between standardized and urban vernaculars. Shona is morphologically rich but it is going through a process of morphological reduction like other African urban vernaculars (Childs 1997), most noticeable at word edges. For example, Sekuru becomes truncated to kulez (grandmother), while amaiguru is reduced to gulez (mother's elder sister or elder brother's wife) (Mawadza 2000: 97). Table 4.

Semantic Modification in Urban Shona

Standardized Shona

English Gloss

Urban Vernaculars

chidhina chimbeva madziro panze munda kusimuka

bick mouse wall outside field stand up

cell-phone an expensive cell-phone automatic teller machine automatic teller machine workplace death

(Mawadza 2002: 97)

48

Janina Brutt-Griffler

4.

Conclusion

The development throughout Africa of these urban languages (recognizing that the name disguises their rapid spread to rural areas as well) represents one of the most interesting and linguistically important cases of language change in the world today, fraught with implications for the main models of language contact and theoretical discussions of language endangerment. It illustrates that the assumptions that underlie the linguistic preservationist movement of how language change, shift, and death occur do not find broad empirical substantiation, and need to be revised. Above all, it has been assumed that English poses the greatest threat to the world's linguistic diversity because of its global power. For such a notion to hold, it would, however, need to be shown that large numbers of monolingual English speakers are emerging throughout places like Africaand that those Africans who are shifting from their mother tongues are doing so in favor of English (cf. Mufwene 2002: 170). Though the linguistic preservationist movement has found a ready acceptance of such a notion, there is actually no evidence of the rise of the monolingual English speaker as a significant factor in Africa, where English speakers remain bilinguals. On the contrary, absent the alternative of English as a lingua franca, indigenous lingua francas have developed to fill the void - but they have done so at a cost to the goal of linguistic diversity. Pan-ethnic argots or urban vernaculars, although formed as second languages, are becoming transformed into first languages with potentially devastating effects for other African languages. The analytical framework of language endangerment that often consists in dividing all languages into "indigenous" and "dominant" actually serves to obscure the process that is underway. In the most comprehensive survey to date, Mufwene (2002) concludes, "We should not overrate the importance of European languages regarding language endangerment." He writes, "The new indigenous lingua francas (such as Wolof, Swahili and Lingala) have gained economic power and prestige, and have gradually displaced (other) ancestral ethnic languages. It is these that can be said to have endangered indigenous languages, to the extent that some rural populations have been shifting to the urban vernaculars, abandoning some of their traditional cultural values for those practised in the city" (175-176). In the face of empirical evidence that there is significant language shift ongoing within the category of indigenous languages that, in fact, such language shift dominates in Africa the dominant-encroaching-on-

Language endangerment and world English

49

indigenous paradigm seizes on the admixture of lexis and, less often, structural elements of English within the emerging languages or existing languages as evidence of the contention that in all cases of language shift the directionality is indigenous to dominant. First, that overstates and overvalues the actual influence of English on indigenous languages, subsuming even cases in which English influence is non-existent or minimal under the category of English dominance. Such an approach actually trivializes the "indigenous" languages that are said to be the objects of preservation efforts, since it regards differences among them as unimportant in comparison to their overarching distinction from "exogenous" languages. To say that language shift from one "indigenous" African language to another which is lexically and structurally English-influenced amounts to a shift from an indigenous language to English would be akin to arguing that English is really linguistically subsumed under the French that exerted an enormous impact on it - certainly far greater than that of English on African languages. Speakers of English-influenced vernaculars in Africa are no more Anglophone than English speakers are francophone. Such arguments seem to construct the object of language preservationism in terms of maintaining (an impossible and nonexistent) indigenous purity rather than variety, as though the real object were to protect indigenous languages from contamination rather than to preserve as many as possible (cf. Mufwene 2002). In such terms, those speakers who shift to vernaculars are already lost to cultural purity, and so are really no longer indigenous at all, a selfcontradictoiy argument which, if anything, only points to the crudely reductionist nature of the categories employed. If Europeans arbitrarily and artificially create "standard" versions of indigenous languages, that does not negate their character as authentic representations of "indigenous" society and culture that must be preserved at the cost of ecological disaster. If, on the other hand, language contact does in Africa what it has done everywhere else leave its mark on the lexis and structure of languages then the languages so affected are dangerous to the linguistic processes (or language ecology) that produced them along with every other natural language. The point of this article is certainly not to question the social status of English. On the contrary, it is to point to a hitherto largely ignored and rather unexpected effect of withholding it from speakers of African languages: that it accelerates language shift, one disguised by the employment of political, rather than linguistic, analytical tools and categories. The effects are not, however, limited to the realm of language endangerment; they

50

Janina Brutt-Griffler

are more significant and tangible for the speakers involved. Even so seemingly simple a principle as mother tongue medium education is complicated by this circumstance. It leads to the ironic phenomenon in many parts of Africa, so far entirely ignored by the language rights movement, that 'mother tongues' as they are used in schools are less and less the home languages of the students educated through them. It is strange that language rights advocates have been as yet completely silent on this violation of the basic tenets of mother tongue education. To be sure, to criticize it would require the admission that, not English, but the pan-ethnic argots, represent the real threat to the indigenous languages they seek to preserve. They would then have to essentially alter the way they have theorized indigenous versus dominant languages. They would have to give greater attention to bilingual speech communities, and the tendency in multilingual settings toward language change. When the complexities of such linguistic contexts are restored, the neat and tidy political narrative of language endangerment gives way. There is some evidence, then, that language use in Southern Africa is undergoing a process of transformation from a function of ethnic affiliation to one of class affiliation, with a concomitant reconfiguration of identities along those class lines. If that is true, it follows that such a process will erode indigenous languages far faster than the scenarios put forward by models of linguistic imperialism. As Widdowson (2003: 57) observes, "The conservation of languages comes at a cost, which people may, or may not, be prepared to pay." The problem is that the costs, social, cultural, political, economic, and otherwise, of conserving languages have been reckoned according to political assumptions rather than conclusions based on scholarly investigation. Many within the language endangerment movement consider it sufficient to theorize indigenous language solely through a political lens. Thus, the editors of the volume Language Endangerment and Language Maintenance that I cited to begin this presentation write, "We still must contend with community attitudes which do not value 'small' languages and view the global spread of English and other dominant languages as progress." Rather than encasing the analysis in these political terms, it is urgently important that the theorization of indigenous language be tied to conceptions of bilingualism and multicompetence that is, to bilingual speech communities and the language change they initiate. Only by doing so can we take the first step toward an enlightened public discourse on language endangerment, one founded on sound scholarship.

Language endangerment and world English

51

References Amin, S. 1990

Maldevelopment: Anatomy of a Global Failure. London: United Nations University Press.

Bradley, D. and M. Bradley (eds.) 2002 Language Endangerment and Language Maintenance. London: Routledge Curzon. Breckenridge, C.A. and P. van der Veer (eds.) 1993 Orientalism and the Postcolonial Predicament. University of Pennsylvania Press.

Philadelphia:

Brutt-Griffler, J. 2002 World English. A Study of its Development. Clevedon: Multilingual Matters Press. Childs, G. T. 1997

Chimhundu, H. 1992

The status of Isicamtho, an Nguni-based urban variety of Soweto. In The Structure and Status of Pidgins and Creoles, A. K. Spears and D. Winford (eds.), 341-370. Philadelphia: John Benjamins Publishing Company. Early missionaries and the ethno-linguistic factor during the invention of tribalism in Zimbabwe. Journal of African History 33: 255-64.

Crystal, D. 2000

Language Death. New York: Cambridge University Press.

Harris, R. 1998

Introduction to Integrational Linguistics. London: Pergamon.

Irvine, J. 1993 Jeater, D. 2001 Lelyveld, D. 1993

Mastering African languages: The politics of linguistics in nineteenth-century Senegal. Social Analysis 33: 27-44. Speaking like a native. Journal of African History. 450-467. Colonial knowledge and the fate of Hindustani. Journal of Comparative Studies in Society and History 35 (4): 665-82.

MacGonagle, E. 2001 Mightier than the sword: The Portuguese pen in Ndau history. History in Africa 28: 169-186.

52

Janina Brutt-Griffler

Makoni, S. 1998

African languages as European scripts: The shaping of communal memory. In Negotiating the Past: The Making of Memory in South Africa, S. Nuttall and C. Coetzee (eds.), 242-248. Oxford: Oxford University Press.

Makoni, S., Brutt-Griffler, J. and P. Mashiri forthcoming Mother Tongues in Search of Speakers: Crossing Ethnic and Linguistic Boundaries in Southern Africa. Mashiri, P. 2002

Mawadza, A. 2000

Managing 'face' in urban public transport: Polite request strategies in commuter omnibus discourse in Harare. In Africa and Applied Linguistics, Makoni, S. and U. H. Meinhof (eds.), 120-126. AILA. Harare Shona slang: A linguistic study. Zambezia XXVII.

Mazrui, A. A. and A. M. Mazrui 1998 The Power of Babel: Language and Governance in the African Experience. Chicago: University of Chicago Press. McLaughlin, F. 2001 Dakar Wolof and the configuration of an urban identity. Journal of African Cultural Studies 14: 153-172. Muila, I. M. W. "Isicamtho"(http ://www. cccponline.org/archive/cccp 12/ page_29.html, last accessed December 2003) Mufwene, S. S. 2001 2002

The Ecology of Language Evolution. Cambridge: Cambridge University Press. Colonisation, globalisation, and the future of languages in the twenty-first century. International Journal on Multicultural Societies, Vol. 4, (2): 165-197.

Mühlhäusler, P. 1996 Linguistic Ecology: Language Change and Linguistic Imperialism in the Pacific Region. London and New York: Routledge. Nettle, D. and S. Romaine 2000 Vanishing Voices: The Extinction of the World's New York: Oxford UP. Phillipson, R. 1992

Languages.

Linguistic Imperialism. Oxford: Oxford University Press.

Language endangerment and world English 1999 Pratt, Mary L. 2003

53

Political science. In Language and Ethnic Identity, J. Fishman (ed.), 94-108. New York: Oxford University Press. Building a new public idea about language. Profession 2003: 110-119.

Skutnabb-Kangas, T. 2004 Do not cut my tongue, let me live and die with my language. The Journal of Language Identity and Education, Vol 3, (2): 127134. Spitulnik, D. 1998

Mediating unity and diversity: The production of language ideologies in Zambian broadcasting. In Language Ideologies: Practice and Theory, B. Bambi, K. Schieffelin, A.Woolard and P.V. Kroskrity (eds.), 163-189.

Widdowson, H. G. 2003 Defining Issues in English Language Teaching. Oxford: Oxford University Press.

The power to choose and its sociolinguistic implications Florian Coulmas

1.

Introduction

This paper argues two main points: one, language is an artefact, and two, every aspect of every utterance is the result of the speaker's choice. Upon reflection it will be noted that both points are closely related if not, in fact, two sides of the same coin. There are things we can change and others we cannot. However, it is far from clear that the difference between both is a constant. Rather, it seems to be a human primitive to defy fixed limits. As human beings we are able to change our behaviour. The idea that we act as free agents is fundamental to our self-conception. Every word we say reinforces this conviction, for whenever we speak we make choices. The ability to consider alternatives and opt for one is basic to intelligent life. This ability is restricted by our physical nature, the many things we cannot choose, such as, the colour of our eyes, our IQ, or whether we are beautiful or ugly. All this may change soon, as the human species gets ready to do with itself what it has done with other species for a long time: interfere with nature's course, select, breed, grow, and manipulate their genetic makeup. The life-sciences have made spectacular progress over the past several decades, constantly expanding the realm of culture - that which we control - at the expense of nature - that which controls us. No longer confined to science fiction novels, anthropotechnology has crossed the threshold into the real world and become a vital concern of legislation, the paradigm of deliberate regulation of behaviour. The prospects are tempting. Before long, we are told, we will be able to safeguard our offspring against congenital diseases, if not secure immortality for ourselves. At the same time, we are confronted with new challenges. We will have to decide whether to go down every pathway science opens up or to erect occasional warning signs, STOP HERE, at critical junctures. In short, at the present time, we are forced to rethink our place in the universe, the confines of nature, and our own nature.

56 2.

Florian Coulmas Language: Biological and social

Language plays a peculiar role in this regard. It is no coincidence that the scientific study of language has been thoroughly impressed by, and, some would claim, contributed to, the revolutionary changes in the life-sciences. For language is seen as an evolutionary adaptation to communicate information. It is what most distinguishes us from other beasts, chatty chimps and brainy dolphins notwithstanding. The exploration of language, therefore, is indispensable if we want to understand our own nature. Language is innate and common to the species. Those who, following Noam Chomsky, call their field "Biolinguistics" clearly have a point. According to Chomsky (2000: 4), the main task of linguistics is to elucidate the 'faculty of language' which he defines as follows: "The faculty of language can reasonably be regarded as a 'language organ' in the sense in which scientists speak of the visual system, or immune system, or circulatory systems, as organs of the body." The language organ helps us survive. But does Italian, or Dutch, or Bengali? If the faculty of language is part of our genetic heritage and an organ of the body, why are languages so much more different than lungs and adrenal glands? There is no convincing answer to this question unless we open our eyes to the other side of language, the social one. Language is a social fact in that every language is a collective product, an artefact created by its speakers which enables higher forms of social planning and cooperation to evolve. Society is built on language. There is no human society that does not speak and use language as its central instrument of organization. Like every utterance, every language could be different from what it actually is. We know this because we know that today's languages were different in the past, that they have changed and will continue to change. Social facts are historical facts with many contingent features. We experience language as a stable system that works and tend to think of different languages as distinct systems. Adaptation and change happen largely unnoticed. Yet, the fact of language change forces us to look at instability, deviation, and loss of comprehension across generations and dialects. The existence of different languages is a historical fact, a result of language change, a result of choice. Every language is transmitted from one generation to the next by learning and has its unique history. These two facts go a long way to explaining

The power to choose and its sociolinguistic implications 57 linguistic diversity. Diversity means two things, the multiplicity of human languages; and the enormous variety of coexisting forms in every language.

3.

Migration and diversity

Where a substantial body of population moves out of one territory and into another, it will take its language with it, but after some time this language ceases to be the same as that spoken in its original territory. From a theoretical point of view, this is remarkable because it means that social factors are involved in language change. If language change were deterministic, thrust towards a goal and governed entirely by quasi-natural laws inherent in the language system, as in the past historical linguists have claimed,1 we should expect it to be unaffected by migration. In the event, English, French, Spanish, Portuguese, and Dutch should continue to change along the same lines on both sides of the Atlantic. But as it turns out, once a group splits into two, language change is no longer synchronized. Since the two groups are stripped of the opportunity to adjust their speech to each other, the transmission and recreation of their language is propelled onto different trajectories. 3.1.

Desires and norms

Migration usually induces language change, but a speech community's spatial contiguity and temporal continuity are no guarantee for maintaining linguistic homogeneity or a sufficient condition for bringing uniformity about. For, as John Gumperz noted a long time ago, variation serves important social functions (1967: 228; emphasis added): In highly stratified societies such as the caste societies of India, it is quite possible for people to be in constant and regular communication over long periods of time without adopting each other's speech patterns. It would seem that communication leads to uniformity only when there is both the possibility and the desire for social assimilation. Where social norms put a premium on social distinctness, linguistic symbols of such distinctness tend to be maintained. It is not at all rare that linguistic distinctions withstand ostensible forces of homogenization. Efficiency of communication would seem to call for a reduction, if not elimination, of potentially disruptive distinctions in the speech of individuals and groups. Yet, such distinctions persist. Why?

58

Florian Coulmas

Two important notions in the passage quoted above are "the desire for social assimilation" and "social norms [that] put a premium on social distinctness." Both are invoked as causal factors in the process of language change. In what sense an assemblage of individual desires can be understood as a collective desire is a difficult question. For the present purposes the important thing to note is that desires and the willingness to adhere to, or breach, social norms make a difference, since it is by virtue of its members having desires and preferences that the speech community creates and perpetuates its language. Speakers, rather than just being the bearers of abstract structures removed from conscious reflection constraining their speech behaviour are active knowledgeable purposeful agents who make choices whenever they use language. The ability to do so is at the heart of the nexus between language and society. In the remainder of this paper I will explore some of the consequences of this obvious fact. It holds for every level of language, structural and stylistic, and beyond that for the registers and languages used by different groups and in different domains of society. Every language represents a choice of the potential held by universal grammar, and every individual's language represents a choice of his or her collectivity's language. Social norms are restrictions on individual choices making deviations that imperil communication unacceptable, if not impossible.

4.

Choice

Since human bodies consist of particles, their behaviour should be explained in terms of particles and the laws governing their movements. There is no room for a mind with a free will. At the same time, our everyday experience is that our reasonings and choices govern our behaviour, to a significant extent at least. This is the mind-body problem in a nutshell. I will not attempt here to solve it, but we cannot altogether sidestep it, for language has both a physical and a mental side, and these are not always easily kept apart. Sociolinguistics is the linguistics of choice, and, if only for that reason, we have to come to grips with the relationship of freedom of the will,2 human action and language, for choice is a notion which presupposes an agent rather than an automaton. The intricacy of the problem has been pinpointed by two scholars representing as it were the two sides, the philosopher Karl Popper and the neurologist John Eccles. Interestingly,

The power to choose and its sociolinguistic implications

59

they see the very origin of language indissolubly linked with choice. This is what they say (Popper and Eccles 1977: 13): We could say that in choosing to speak, and to take interest in speech, man has chosen to evolve his brain and his mind; that language, once created, exerted the selection pressure under which emerged the human brain and the consciousness of self. Choosing to speak before you know what to say, let alone know what language is, seems quite a feat, but, on reflection, it may be quite common. However that may be, making choices is a central part of the human condition. Interestingly, Popper and Eccles' notion of choice does not require full control and foresight. This is important, for, clearly, babies make choices; every line we draw to show where intentionality begins is arbitrary. Our choices are subject to restrictions of various kinds from birth. The division of labour in the language sciences can be understood in terms of the restrictions on possible linguistic choices. Physical and cognitive restrictions are the field of biolinguistics and cognitive science; social restrictions on linguistic choices are for sociolinguists to investigate. Although most speakers usually leave their speech performance to the automatic pilot, every speaker has the ability to change the way he or she speaks. The choices speakers make are not made in a vacuum but are constrained in many ways. While every speaker's every speech act is the manifestation of choice, the individual act of choice does not reveal the social nature of language. That only becomes apparent if we can show how individual choices add up to form collective choices.

5.

Cooperation and choice

Choice is based on judgment, but it is not fully controlled. More typically, judgment results from the interpretation of incomplete information and an individual's beliefs and preferences. "It is seldom possible for an individual to accurately describe his or her judgment process." (Dhir and Savage 2002: 11) This is particularly true of judgments about one's own speech. Moreover, human judgement is often inaccurately reported and inconsistent. These limitations must caution us not to rely too much on self-reports if we want to find out why people speak the way they do. We can benefit here from certain concepts developed by psychologists and political scientists interested in cooperation. For language is a means of cooperation on two levels: (1) it enables its speakers to exchange infor-

60

Florian Coulmas

mation in order to request, announce and promise certain actions; and (2), it works as long as its speakers use it in such a way that they understand each other. In normal speech behaviour this is taken for granted. That is, speech behaviour is fundamentally cooperative. What, then, is cooperation? Following Robert Axelrod (1984) I favour a wide notion of cooperation that includes, for example, patterns of behaviour found in various nonhuman biological systems. Mutually beneficial interaction between plants and other lower organisms shows that foresight is not necessary for cooperation to evolve. It is also very doubtful that ants have foresight, yet, they clearly cooperate. An approach that does not presuppose foresight is more suitable to language than other decision theories, especially rational choice theory which assumes rational and consistent actors with foresight and full control. For example, the behavioural decision theory (Edwards 1992) is based on the concept of expected subjective utility and attempts to prescribe rational decisions. By contrast, Axelrod's theory does not assume that interactants are rational. "Their strategies may simply reflect standard operating procedures, rules of thumb, instincts, habits, or imitation." (Axelrod 1984: 18). It moreover does not assume that the actions taken by the interactants are necessarily conscious choices. Notice that even in economics where rational and fully informed decision makers used to be assumed for the sake of modelling, this conception of humanity is giving way to one of agents who have emotions and lack complete information. 3 For language choices the idea of cooperation being achieved even in the absence of fully informed rationality is highly relevant, because whenever communication proceeds smoothly stable mutual cooperation has been established. Stable mutual cooperation often follows a treaded path, it is 'path dependent', to borrow a notion much used by psychologists and economists. This notion captures well the nature of choice restricted by habit and inertia. You can leave the treaded path, but doing so takes an effort and bears a risk. We can think of a language as a path. New members of a speech community get set onto this path which under normal circumstances they will never try to deviate from. Staying on the path appears natural, not a matter of choice. Usually, it is tantamount to cooperative behaviour. But it is the result of choice, all the same. This is apparent from instances of uncooperative behaviour. For example, francophone Swiss often complain about being addressed, by their compatriots in German-speaking Switzerland, in Swiss-German. The Francophones learn standard German at school but have difficulties understanding Swiss-German. Since the Swiss-

The power to choose and its sociolinguistic implications

61

Germans control both varieties, their choice of Swiss-German is perceived as not being cooperative. Cooperation is the default case in speech behaviour and vital for preserving the functionality of the language. It can be explained in terms of the (perceived) probability that the speakers concerned may interact again. In a generalized way this is what keeps a speech community together and its language intact. As the theory of public choice emphasizes, reciprocity is essential for maintaining cooperation. Linguistic communication is a cooperative game, not a zero-sum game. Both players - assuming there are just two - benefit if they cooperate. They should not regard the other player as someone who is out to defeat them. The payoffs are for doing what everyone else does. The classic example is traffic: It does not matter which side of the road we drive on but it is extremely important that we all drive on the same side. A common language is usually taken for granted, since locations, institutions and situations are associated with a certain language or variety as the default choice required by a cooperative game. By the population concerned, the default is not usually perceived as a choice at all. But it is a choice in the sense that alternative courses of action are possible, however unlikely. Many actions are carried out routinely, such as walking or shifting into gear when driving a car. A great deal in language behaviour, too, remains on a subconscious level. When we employ routines and enact default choices without thinking we allow our movements to follow treaded paths, as a matter of convenience. Default choices are choices, although they go largely unnoticed. Societies and social groups define themselves and are recognizable in terms of their default choices. For example, in largely monolingual environments, multilingual conversations tend to be perceived as uncooperative, while people in multilingual environments are more familiar with situations where speakers expect each other to understand several languages but use the one they speak best. In the event, for speakers to use a language that is not the same as that in which they have been spoken to may be readily tolerated not being perceived as uncooperative. Uncooperative behaviour is sanctioned, the most effective sanction being the refusal to understand a speaker. Speakers are active creative agents, able to choose their verbal means, but their choices are limited by the need to cooperate and the threat of sanctions. Let us consider then some of the choices that are made in language and the social factors acting on them. A general division is between micro

62

Florian Coulmas

choices and macro choices. Micro choices concern structural features of language, macro choices entire codes.

6.

Micro choices

6.1.

Dialects

Dialect differentiation is known in all speech communities of any size. In industrialized urban societies horizontal geographical dialect differentiation has been reinterpreted as being ordered along a vertical axis of social stratification. People's social position influence the way they speak, whether they choose to use their local dialect or a speech form closer to the standard. Clearly, a standard variety is an artifact, but so is any other recognizable variety a speaker can choose. What the study of choice of dialectal features teaches us is that cooperation goes beyond establishing intercommunicability. Trudgill (1986: 39) notes that "in face-to-face interaction [...] speakers accommodate to each other linguistically by reducing the dissimilarities between their speech patterns and adopting features from each other's speech." The features in question are phonetic, morphological and syntactic. In urban, highly mobile societies mutual accommodation may lead to dialect levelling, but there is a lot of evidence to indicate that in the process dialect boundaries are redrawn rather than abolished. Homogenization is not the preferred choice. 6.2.

Gendered speech

Next, sex and gender are equally well-known as determinants of linguistic variation, and like regional distinctions they are subject to speakers' choice. To mention but one example that is closer to the natural sex side than to the social gender side, voice pitch. Ohara (1997) recorded natural conversations and reading sentences in Japanese and English by the same bilingual speakers. She found that women speak with a higher pitch in Japanese than in English, while men's pitch was the same for both languages. How deep a male voice is and how high a female one thus is to some extent subject to variation and socially conditioned choice. The natural is subject to cultural modulation. More generally, sex-specific and sex-preferential speech forms are artifacts. In a number of Asian languages such as Japanese and Korean these

The power to choose and its sociolinguistic implications

63

distinctions are so pronounced as to form "women's and men's languages". Their function is explained by Ide and Yoshida (1999: 477) as follows: "The choice of one linguistic form over another reflects a perception of the structure of cultural understandings and represents the speaker's identity as a member of the society.[...] The appropriate linguistic choice has the effect of avoiding conflict or misunderstandings in the interaction." The point to stress is that, while the avoidance of conflict and misunderstanding is undoubtedly a general principle of cooperation, there may be situations where speakers have reasons to breach this principle in the interest of pursuing other goals such as indicating non-acceptance of sex-differentiating conventions. Deliberate deviation from the treaded path is the main theme of the feminist critique of linguistic usage that has swept through many societies in recent decades. In every particular case, the complex relationship between sex and speech behaviour involves a language that has been formed by many generations of speakers, as well as the ideological formation of that community's ethnolinguistic tradition. On both levels speakers make choices that reproduce or alter existing conventions, unwittingly or, in some cases, deliberately. The feminist critique of prevalent usage has been directed against accepting sexist attitudes underlying certain expressions. The changes that were enacted were very conspicuous and the speakers who first tried them out, who said he or she or used they as a pronoun with an antecedent in the singular, came across as uncooperative because they intentionally violated existing default choices. But eventually, and within a relatively short time, the campaign to reduce sexism in language in the English speaking world was very successful. The general point that follows from this example is this: Languages can be profoundly affected by deliberate choices of their speakers. 6.3.

Age differentiation

Speech modulates across generations. Age-differentiated language use begins in infancy. In most societies that have been studied, babies are taught from the start that speech varies with speaker age. The means by which this message is communicated is parentese. While parentese does reflect certain features of early language, it also paves the way for agedifferentiated speech in general. It has been demonstrated that young children are quite aware of, and capable of reproducing, the differences between the speech addressed to them and to adults (Steinberg, Hiroshi, and

64

Florian Coulmas

Aline 2001). Parentese is conducive to the development of communicative competence in two ways: (1) in the earliest phase it facilitates language learning, and, (2) it shows the youngest generation that members of the community are differentiated by age and that speech forms are to be chosen accordingly. At the other end of the lifespan is the speech of the elderly. Except as an aspect of age-induced deteriorating abilities, it has not received much attention. However, the ways in which the elderly are often spoken to once again reveal the significance of age as a determinant of linguistic choice. They are in some ways a mirror image of parentese. Giles and his associates (Kemper 1994) have observed that speech forms addressed to the elderly do not so much accommodate the communication needs of their interlocutors as the speakers' ideas of these needs. In this regard, the characteristics of "elderspeak", for example, slow production, simplified syntax, avoidance of difficult words, and exaggerated prosody, are reminiscent of other simplified speech registers, such as those directed at babies, foreigners, and pets. Sandwiched in between the languages of the very young and the very old are the languages of adolescents and adults. Characteristic features of adolescents' speech have often been described. Use of substandard, dialectal and vernacular forms, slang and innovative, often very short-lived expressions serves three main functions: (1) to appropriate the language for their own purposes; (2) to manifest group membership and construct a distinct identity; and, (3) to indicate the speakers' willingness to resist the pressure to conform to societal norms. Young and middle-aged adults form the dominant age cohort in terms of social strength in most societies, investing their behaviour including speech with norm-setting potential. Adult language is the norm, as it is the default choice for all age groups. Yet, young and middle-aged adults are set off from the other age groups in various ways, some of which are linguistically encoded. For example, in languages with complex personal pronoun systems, such as Russian, French, and German,4 the transition from adolescence to adulthood is often marked by a change of pronominal address. The general lesson to be learnt from intergenerational linguistic diversity is that linguistic choices are driven among other things by beliefs and attitudes about age divisions and notions of age-specific suitability. The observable differences in the speech behaviour of co-existing generations manifest age cohort stereotyping and the time-depth of language. The former is a motor of language change as each new generation introduces in-

The power to choose and its sociolinguistic implications

65

novations, while the latter prevents dysfunctionally rapid change because the individual needs to communicate with parents and children. 6.4.

Politeness

It has been argued that languages differ in terms of how strongly socially encoded they are (Mühlhäusler and Harre 1990) and that strongly socially encoded languages, such as Japanese, Korean and Thai, provide a rich variety of polite expressions but at the same time are more restrictive with respect to their choice than weakly socially encoded languages, such as English (Ide and Yoshida 1999). This is a contentious issue, but the fact that speakers employ various politeness strategies is incontrovertible. Almost all utterances have alternatives less or more polite, and every utterance can be judged against a standard of what is expected by the audience. Calibrating one's utterance to meet, exceed, or breach the standard is an essential part of social competence. Setting, social class, gender, age, power, and culture are the main factors speakers take into account in making their respective choices.

7.

Macro choices

7.1.

Code-switching

A major theme in code-switching research has been the question whether the mixing of, and alternating between languages and varieties is restricted by structural or social conditions. Generally speaking, as soon as a structural restriction such as, for example, the bound-morpheme condition, was proposed, it was invalidated by numerous counterexamples. Combining markedness theory with her influential Matrix Language Frame model, Myers-Scotton (1993a; 1993b) has tried to capture both the structural and social conditions of code-switching, but competing notions persist. For instance, Muysken (2000) works with the concept of 'suspension of grammar'. Once again, what is generally agreed is that speakers' choices are the issue to explain. In code-switching research the notion of choice has come to higher prominence than in other fields of sociolinguistics, perhaps because it is so striking to the outside observer that code-switchers are continuously busy making choices.

66

7.2.

Florian Coulmas

Diglossia

Another area where this is also quite obvious and which borders on codeswitching is diglossia. A relatively rigid domain separation between two varieties of the same language has been observed in a number of speech communities, for example, in India, where it is a very old tradition (Schiffman 1996), and in the Arabic-speaking world (Blau 1977). In my view, writing and literacy are crucial for the split to be brought about and to be sustained (Coulmas 2003), but the complex aetiology of diglossia cannot be discussed in the present context. What should be emphasized is that there is plenty of evidence, especially in fully literate societies such as German-speaking Switzerland, to show that diglossia is maintained not by accident but as the result of desires and norms upheld by the speech community. 7.3.

Multilingualism

In many other multilingual settings codes are held apart, but speakers make language choices on a regular basis. These choices are motivated by the social meaning they carry. In some settings language choice is more negotiable than in others. For example, in a salesperson-customer encounter in a bilingual urban milieu5 such as Brussels or Montreal negotiating the language is a regular part of many interactions, whereas in institutional settings where the language is determined by rules of procedure individual choice is more limited. However, as individuals and as groups, multilingual communities exploit the linguistic resources available which in the event comprise what are thought of as different languages. Multilingualism research has shown that the choices speakers make are not random but exhibit many regularities and hence explainable. O n e system or two?' is a question that has concerned psycholinguistic research into bilingualism for some time. From a sociolinguistic perspective the range of available options must be regarded in many cases as forming an integrated system which is perpetuated and adapted on the basis of individual cooperative choices much like a single language.

The power to choose and its sociolinguistic implications

7.4.

67

Pidginization

The cooperative nature of linguistic choice is most dramatically evidenced under very unpromising conditions where the speakers involved have no common language at their disposal. Betokening the intimate connection between choice and cooperation in language, a new language is then brought into existence, a pidgin. This only happens when interactants venture outside their territory and reference group and meet others, like them willing to communicate. Pidginization demonstrates how cooperative verbal behaviour develops in the face of extremely adverse circumstances. By choosing from their initially disparate expressive resources they must establish common ground. Although pidgins are characterized by great instability at first, that is, choices of low predictability, the cooperative nature of linguistic communication leads to incremental reductions of the range of possible choices and correspondingly increasing predictability. Rather than traditions that are carried on for tradition's sake, pidgins are a new creative intertwining of hitherto unconnected linguistic materials and as such, more obviously than other languages, bear the stamp of artefacts. Another important lesson to be learnt from pidginization is that language is the result rather than the prerequisite of cooperative action.

8.

Language decline

One more topic needs to be mentioned in this brief review of sociolinguistic issues which has received considerable attention in recent years, declining languages. A great many languages are today spoken by elderly speakers only, as the young choose to speak a language with a wider range. There is wide agreement among scholars that as a result the pool of human languages will shrink significantly in the near future (Crystal 2000). Many linguists active in this field will deny that what they consider an unfortunate development is the result of choice. I consider this a political argument. If the notion that speakers are active agents who make choices is taken seriously, it must be recognized that choice of action without full foresight and appreciation of the consequences and choice under duress is still choice. Norman Denison (1977) had a point when he asked: 'Language death or language suicide?' Whether or not linguistic diversity should be safeguarded as a valuable possession of humanity is a question unrelated to that of whether speech is a natural process beyond the speakers' control or

68

Florian Coulmas

the result of choice. The sociolinguistic phenomena I have reviewed in this paper suggest the latter.

9.

Summary

Let me sum up the main points of this paper. Sociolinguistics distinguishes itself from other language sciences, notably from biolinguistics and theoretical grammar, by focussing on those traits of language that are noticeably variable and hence subject to choice. Variability in language means that speakers are able to adjust their speech to selected aspects of their environment. This has a number of implications. -

Speakers make choices from the variety of the expressive means offered in their environment.

-

Their choices are subject to restrictions.

-

Speech communities and smaller social groups are recognizable by virtue of the restrictions they place on the linguistic choices of their members.

-

Membership is accomplished and maintained through cooperation.

-

Speech behaviour is cooperative.

-

Every language and language variety is the result of collective choice, that is, cooperative creation.

-

Where a common language exists it restricts its speakers' choices. Where no such language exists one is created by virtue of interactants' choice.

Notes 1. The idea that sound change is regular and that, accordingly, its investigation must reveal general laws (Lautgesetze) was first defended in the latter half of the 19th century by a group of linguists who called themselves 'Neogrammarians'. Their ideas gave rise to extensive debates accompanying the development of linguistic thought throughout the 20th century. And although empirical methods and techniques have changed, their principles still inform research into the mechanism of sound change. Cf. Philological Society (1978).

The power to choose and its sociolinguistic implications

69

2. If you want to get serious about the problem of freedom of the will, Wegner (2002) is an excellent account by a psychologist and Searle (1984), by a philosopher. 3. In 2002, economists Daniel Kahneman and Vernon L. Smith were awarded the Nobel Prize for this departure from the paradigm of 'rational choice theory'. See, for example, Kahneman, Slovic and Taversky (1982). 4. Cf. Head's (1978) survey of pronominal address in 100 languages. Mühlhäusler and Harre (1990) offer a critical discussion of the sociolinguistic issues involved in personal pronouns which covers a wide range of languages but is not always reliable. 5. For examples of negotiated language choice in Montreal cf. Heller (1992, 1995).

References Axelrod, R. 1984 Blau, J. 1977 Chomsky, N. 2000 Coulmas, F. 2003 Crystal, D. 2000 Denison, N. 1977

The Evolution of Cooperation. New York: Basic Books. The beginnings of Arabic diglossia: A study of the origins of Neoarabic. Afroasiatic Linguistics 4: 175-202. New Horizons in the Study of Language and Mind. Cambridge: Cambridge University Press. Writing Systems. An Introduction to Their Linguistic Analysis. Cambridge: Cambridge University Press. Language Death. Cambridge: Cambridge University Press. Language death or language suicide? International Journal of the Sociology of Language 12: 13-22.

Dhir, K.S. and T. Savage 2002 The value of a working language. International Journal of the Sociology of Language 158, 1-35. Edwards, W. (ed.) 1992 Utility Theories: Measurements and Applications. Kluwer Academic.

Dordrecht:

70

Florian Coulmas

Gumperz, J.J. 1967

Head, Β. F. 1978

Heller, M. 1992 1995

Language and communication. In The Annals of the American Academy of Political and Social Sciences 373, Β. M. Gross (ed.), 219-231. Respect degrees in pronominal reference. In Universals of Human Language, Vol. 3, Word Structure, J. H. Greenberg (ed.), Stanford: Stanford University Press, 151-211. The politics of code-switching and language choice. Journal of Multilingual and Multicultural Development 13: 123-142. Language choice, social institutions, and symbolic domination. Language in Society 24: 373^105.

Ide, S. and M. Yoshida 1999 Sociolinguistics: Honorifics and gender differences. In The Handbook of Japanese Linguistics, Natsuko Tsujimura (ed.), 444-480. Oxford: Blackwell. Kahneman, D., P. Slovic, and A. Taversky (eds.) 1982 Judgment under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press. Kemper, S. 1994

'Elderspeak': Speech accommodation to older adults. Aging and Cognition 1: 17-28.

Mühlhäusler, P. and R. Harre 1990 Pronouns and People. The Linguistic Construction of Social and Personal Identity. Oxford: Basil Blackwell. Muysken, P. 2000

Bilingual Speech: A Typology of Code-Mixing. Cambridge University Press.

Cambridge:

Myers-Scotton, C. 1993a Social Motivations for Codeswitching. Evidence from Africa. Oxford: Clarendon Press. 1993b Ohara, Y. 1997

Duelling Languages: Grammatical Structure in Codeswitching. Oxford: Clarendon Press. Shakaionseigaku no kanten kara mita nihonjin no koe no kötei [High and low pitch of the voice of Japanese from the point of

The power to choose and its sociolinguistic implications

71

view of sociophonetics]. In Josei no sekai [The world of women]. Ide Sachiko (ed.), 42-58. Tokyo: Meiji Shoin. Philological Society 1978 The Neogrammarians. Oxford: Blackwell. Popper, K. R. and J. C. Eccles 1977 The Self and Its Brain. An Argument for Interactionism. Berlin and London: Springer International. Schiffinan, H. 1996 Searle, J. 1984

Linguistic Culture and Language Policy. London and New York: Routledge. Minds, Brains and Science. Cambridge, Mass.: Harvard University Press.

Steinberg, D.D., N. Hiroshi, and D. P. Aline. 2001 Psycholinguistics. Language, Mind and World. London: Longman. Trudgill, P. 1986 Wegner, D. 2002

Dialects in Contact. Oxford: Blackwell. The Illusion of Conscious Will. Cambridge, Mass.: Harvard University Press.

How codeswitching as an available option empowers bilinguals Carol Myers-Scotton

1.

Introduction

Codeswitching between two or more languages can be both an index and a tool. As an index, codeswitching can index a speaker's self-perception, as a multidimensional person, whether as a member of a specific group, or as a member simultaneously of several groups. As a tool, codeswitching can be used in an ongoing conversation to step in - or out - of a presumed or expected identity (Myers-Scotton 1993b). This paper discusses codeswitching by bilinguals as it figures in negotiations of power differentials between individuals. Power as a theoretical construct of relations between groups and individuals is defined by its asymmetrical nature. Power exists in relationships through one participant having more control over joints outcomes, but also control over other factors that affect the other participant. Of course participants may have equal power in a relationship, but more often, they do not. While such interpersonal attributes as degree of solidarity are relatively open to negotiation in interpersonal relationship, power is harder to nudge because of its sources. That is, power asymmetries tend to rely on what participants bring to the interaction in terms of their sociolinguistic profiles. Power tends to accrue to those with the "right" attributes (e.g. a high level of formal education at the right universities, membership in the right ethnic group, practicioners in the right occupations, males rather than females, etc.). I do not suppose that any amount or pattern of codeswitching can permanently change any power imbalance; it cannot give even the cleverest speaker the right attributes. But strategic codeswitching can tip the scales in the speaker's favor - at least in the ongoing interaction. And, with luck, the aura of what counts as power in interpersonal terms may encircle the ingenious speaker to last beyond the current interaction.

74 1.1.

Carol Myers-Scotton Elite closure provides a legacy

Such micro-level codeswitching derives some of its social significance from the role it plays in establishing the preeminence of one group over others at the macro-level in any nation state. That is, codeswitching can help an educated elite in any nation establish and maintain what I have called elite closure (Myers-Scotton 1990, 1993a). Not just their repertoires, but also specific patterns of language separate this elite from other groups and especially from the common people. The phenomenon of elite closure exists everywhere, but in reference to the codeswitching cited here, it is especially relevant to such an elite in developing nations. Thanks to their high levels of education (university degrees and possibly overseas western studies for this group in the developing nations), these speakers have a facility in a prestige language that few others of their countrymen possess. By engaging in codeswitching between a strictly-local language and a language with international standing, they present themselves as different from local people who do not speak the elite language with the same facility. True, for including a local language in their speech, the elite can claim that they are simply "ordinary citizens". But when their usual speech patterns contain codeswitching to a language which also may be used locally, but is especially identified by its international connections, such patterns subvert any message of egalitarianism.

2.

Two empowering variations on the strategies of elite closure

By employing variations on the elite's patterns of codeswitching, some members of the non-elite circumvent elite closure and empower themselves. Alternatively, with some irony, the educated elite themselves can turn elite closure on its head. This paper will present examples of these opposite strategies and discuss them in regard to how they function to give individuals the power that shapes their fortunes in their interpersonal relationships.

How codeswitching as an available option empowers bilinguals

3.

75

Reliance on limited codeswitching

First, consider how the non-elite can exploit codeswitching when switches occur within the same clause. To engage in codeswitching within the same clause, speakers need to have a high level of proficiency in the language that is the source of the morphosyntactic frame of this clause. However, their ability in the other participating language(s) need not be great, if speakers do no more than insert content morphemes from this language in the grammatical frame of the clause or insert short, set phrases. MyersScotton (1993 c, 2002) refers to the language framing the clause as the Matrix Language and the language of content morphemes or set phrases integrated into the frame as the Embedded Language. Examples (1) and (2) give an indication of the structuring of such codeswitching. Content words, especially nouns, but also adjectives and verbs, may be produced in the Embedded Language. These examples come from an audio-recorded informal conversation of three young men (in their early twenties) living in the Eastleigh section of Nairobi. They come from three different ethnic groups (Kamba, Taita, Luo). They are well enough educated to be able to speak some English (they dropped out of advanced secondary school), but they probably cannot produce full clauses in what would be considered standard English in Kenya. The Matrix Language of these examples is Swahili, the expected language for such speakers for framing inter-ethnic conversations in Nairobi. In many ways, one can argue that such examples that include switches to English, constitute their usual medium of communication; in effect, codeswitching itself is their unmarked linguistic variety. (1) Comment on a payment from some casual labor: U-li-ku-wa u-na-expect one hundred and twenty? you-PAST-be you-PROGRESSIVE-expect one hundred and twenty Lakini wewe u-li-fanya But you you-past- do

job moja kidogo sana job one little very

'You were expecting one hundred and twenty? But you did a very little job'. (2) Comment on the behavior of the police: Hao wengine kama wa-me-kosana na some of them if they-PERFECT-make mistakes

bibi zao nyumbani, with wives their at home,

76

Carol Myers-Scotton

Wa-na-kuja they-NON-PAST-come

ku-release hizo to-release those

Hata saa nyingine wa-na-ku-ambia even time another they-NON-PAST-you-tell na hu-ja taste and you/NEG-NOT YET- taste

steam zao kwa watu. steam their on people u-ko drunk and disorderly, you-be drunk and disorderly

beer kwa week mbili beer for week two

'Some of them have quarreled with their wives at home, they come to release their steam/anger on [innocent] people. Even another time they tell you [that] you are drunk and disorderly, and you have not yet tasted beer for two weeks.' (Myers-Scotton, Nairobi corpus 1988) Producing clauses with such English elements as those in examples (1) and (2) requires less proficiency than framing an entire clause in English (the Embedded Language). Speakers with limited proficiency in what is locally considered to be a prestige language can claim some identity with the attributes associated with this language through such codeswitching. In the case of these young men in low cost housing in eastern Nairobi, such a pattern of switching makes them "almost" sound like the elite with their Mercedes and luxurious homes in the western suburbs of Nairobi. It empowers them, at least among their peers and at least when making firstimpressions elsewhere.

4.

Codeswitching and virtual elite membership

In a Nigerian community, the local elite, with their Western-style education, employ Ibibio/English codeswitching for their informal talk. Even though English is one of the official languages of Nigeria (along with Yoruba, Igbo, and Hausa), it is not well-known at all levels of society by any means. Such codeswitching is similar to that illustrated in examples (1) and (2), as example (3) from Ibibio/English makes clear (Essien 1995: 275). (3)

Intelligent people inamma Zkpo ntodo 'Intelligent people do not behave like that.'

How codeswitching as an available option empowers bilinguals 77 Essien (1995) reports that successful, rich local businessmen in eastern Nigeria, who have little education, but who know enough English to insert English content words into an Ibibio frame, also use Ibibio/English codeswitching. The result is that businessmen can pass as members of the educated elite. Here is Essien's comment (1995: 281): "For such people, codemixing is a wonderful opportunity to camouflage their illiteracy and join the 'we-type solidarity' of the educated class. Since even the highesteducated people code-mix, it is sometimes difficult to say exactly from the data of performance of code-mixed Ibibio who is who." Essien goes on to relate how this happens. He tells about one day when he and some university colleagues met a very rich businessman at the Governor's Office in Uyo. He knew who the businessman was, but his colleagues did not. He knew that although the businessman was very successful, he had limited education. After they were all introduced, informal conversation followed and they all "naturally mixed Ibibio and English". Later, one of the university colleagues asked which university the businessman attended. Essien reports that he told his colleague, "[His university is] His mansion in his village." (p. 281-282). In another example from Sri Lanka, a young job applicant uses codeswitching to level inequalities between himself and a professor in an interview for a university position. English has no official status in Sri Lanka, but it is known by the highly educated. (Canagarajah 1995 refers to Tamils who are bilingual in English as one of two parallel elites in Jaffna; they dominate the professions. The monolinguals are the other elite; they dominate local administration and the politico-military hierarchy, and have more significant power.) It was no surprise that the professor opened the interview in English. The job candidate replied in a codeswitching pattern with Tamil as the source of the morphosyntactic frame, but with English content word insertions. (He could assume that the professor, like him, was a native speaker of Tamil.) In analyzing this conversation, Canagarjah (1995: 205) concludes, "In gatekeeping situations ... code-switching becomes handy for those limited in English competence." Example (4) contains part of the interview, which took place in 1991 at the university (Canagarjah 1995: 204). (4) Interviewer:

So you have done a master's in sociology? What is your area of research?

78

Carol Myers-Scotton

Candidate:

naan sociology of religion-ilai taan interested enTai thesis topic vantu "the rise of local deities in the Jaffna Peninsula". "It is in sociology of religion that I am interested. My thesis topic was "The rise of local deities in the Jaffna Peninsula."

Interviewer:

Did this involve a field work?

Candidate:

Ooom, oru ethnographic study-aai tann itai ceitanaan. KiTTattaTTafouryears -aaifleld work ceitanaan. "Yes, I did this as an ethnographic study. I did field work for roughly four years."

Interviewer:

Appa kooTa qualitative research taan ceiyiraniirl "So you do most qualitative research?"

Note that after several turns in which the professor spoke only English, he switched to Tamil-English codeswitching, as well. In effect, he accommodated to the job candidate. Canagarajah (1995: 204) explains the candidate's use of codeswitching probably indicates his lack of English competence. But he also states, "We must note that speaking unmixed Tamil in this context would have displayed a poor image of the candidate, exposed his incompetence glaringly and accentuated the status difference between the interlocutors." Furthermore Canagarjah (1995: 204-5) notes, "Although [the candidate] clearly lacks the ability to form complete utterances in grammatical English, his switches are strategic. The English phrases he uses are not oftenused loans in the wider society; they are specialized vocabulary." The result is that combined with an acceptable educated Sri Lankan pronunciation, the candidate "is able to project a formidable image of himself."

5.

Irony and elite closure

Members of the educated elite may engage in codeswitching that offers the reverse of how their group members would normally speak. In some cases, such as example (5), they may be doing this to humble themselves as they ask for a favor. Example (5) comes from one half of a telephone conversa-

How codeswitching as an available option empowers bilinguals

79

tion audio-recorded in Kakamega, the major city of the Luyia-speaking area in western Kenya. The speaker is a Kakamega resident and a native speaker of Luyia. His telephone conversation is with a native speaker of Kikuyu, who comes from the central area of Kenya and perhaps the most prosperous ethnic group in Kenya. He is a businessman in Kakamega at the time of the conversation. The speaker is a local accountant and it is clear that the businessman is one of his clients. Both are well-educated. Both are trilingual; they speak two different Kenyan languages as their Lis, and they also speak both English and Swahili fluently. Because they come from different ethnic groups and are educated men talking about a business matter, English is the unmarked choice in this interaction. Note that the example contains only one side of the conversation; dots indicate the other man is talking. Swahili parts of the conversation are in bold. (5) Accountant (on the phone): Hello, Bwana Muchanga. Good Morning!... Well, I've told you good morning, but it is not a good morning. Hee... namna garti, bwana? ('What's going on?')... For so many days and so long that you have disappeared? ... No, even if you come around, you will see my mother is here... I don't know if she will die or whether she will survive. ... No, in my house in Kakamega. I took her to hospital, Bwana. We have got a poor hospital. I am telling you. I took her to the hospital at night... There was no treatment... These people claiming no medicine — nothing in the store. I think this is what I shall do - because I am (interrupted) ... Where is the time, bwana? Time is the problem with me ... Wewe, unajua mimi ttiko mwandikwa, bwana ('You know I am an accountant, mister.'). I have always to squeeze myself. Niko taabu, bwana ( Ί have worries, mister'.) ... Na sasa watu ya kusaidia taabu yangu ni watu kama wewe, tu ('And now the people who can help [me] are only people such as you.'). No, what do I do, bwana? I will come, but - huyu mzee yangu, nahitaji unisaidie, bwana na mzee yangu ('My dear old mother, I need your help, mister with my dear old mother). ... una EXPERIENCE nyingi ... Umezoea mambo hay a [na wewe] unaweza kusaidia mimi na mzee yangu. ('You have lots of experience, you're used to these matters, and you can help me with my mother'). (Myers-Scotton Kenya corpus 1977)

80

Carol Myers-Scotton

Note that in this conversation, except for the formulaic greeting, the accountant does not speak Swahili until he begins to ask the other man for help (and help implies "money" in this context). Thus, in switching to Swahili, the accountant uses the language that they have in common that indexes informality and a lower status. Swahili, as an African language (and an indigenous language to Kenya) may be used in a codeswitching pattern by the Kenyan elite for informal inter-ethnic group conversations. However, even though Swahili has official status in Kenya along with English, it clearly is more a language of the lower classes than the elite. Given its placement in this conversation and the Kenyan context, on its surface, the pattern of switches to Swahili represents a bid for shared group membership or solidarity, with the addressee. The surface interpretation is that he is making a request for "help". But given who the speaker is (educated and probably fairly prosperous, at least in the past), his use of Swahili makes it possible to interpret this request as a bid to present himself as having solidarity with the masses, as an "ordinary" person who is in need of help just like anyone else might be. That is, he is presenting himself as a non-elite person. Whether he does this ironically or not is open to interpretation. Is this honest humility? Of course it is important to recognize that this speaker (and others) can have multiple goals, and they may well be ambiguous. This analysis only suggests that the speaker's switch from monolingual English to a good deal of Swahili in a codeswitching pattern can be interpreted as a strategy to lower himself to a non-elite status in order to gain sympathy (and financial aid). Ironically, this puts the other conversational partner into a relatively higher status. Here, as in many other societies, high status people are expected to help the less fortunate. Another example from another part of Africa, South Africa, also shows codeswitching as a bid for solidarity with the common people and also is a lowering strategy. The switch here is to a language that is an unmarked choice in a very different type of interaction than where it is used here. Even more so than example (5) this is an example showing how the elite may turn elite closure on its head. The language that is used is Fanakalo (Fanagalo) and it was used in colonial South Africa in asymmetrical master-servant relationships. Fanakalo is a pidgin/creole language that has been used in South Africa at least since the arrival of the British in the mid-nineteenth century. It has elements from English, Zulu, and also some Afrikaans. (Afrikaans is the language derived from the LI of the white Dutch settlers in South Africa.). Fanakalo

How codeswitching as an available option empowers bilinguals

81

was - and still is to some extent - used in the mining industry in particular where the less powerful person in the interaction was always a black worker, usually in conversation with a white supervisor. Because of this history, Fanakalo remains an unmarked choice between unequals and may be avoided because of its racist associations. However, as the example will show, when Fanakalo is used between peers, especially outside of the work setting, it can be a marked choice and it can index shared group membership. What makes this example especially interesting is that a white speaker uses Fanakalo to other whites in a tennis match at a private club outside the South African city of Durban. Four middle-class white males are playing tennis doubles at a private club. Three of the players immigrated to South African earlier from Europe, and the fourth, a younger man, was born and brought up in South Africa. These are men whose unmarked language in this interaction would probably be English. One of the older residents (of British origin) checks on the score by asking in the following way (Adendorff 1993: 21): (6)

Ini lo telling? "What's the score?"

In saying this, the speaker combines Fanakalo (ini lo 'what's the') with Afrikaans {telling 'score'). In doing this, he is using the two "indigenous" South African languages that all the participants would be likely to know, at least in part. But note that he himself would be a native speaker of English. Adendorff (1993) comments that the speaker is indexing an assumed common South African identity among a group of status-equal white South Africans. A related interpretation is that the speaker is "showing o f f ' how much of a South African he has become (by speaking uniquely South African languages). In some ways, his use of Fanakalo and Afrikaans is an elaborate, but light-hearted, joke on his own status and that of his companions. Adendorff (1993: 21-22) makes this comment: "The speaker is confident that the other players will recognize that he is providing them with a linguistic display and that they will appreciate the display." Discussing the choice in the terms of the Markedness Model (Myers-Scotton 1993b), he goes on to say, "...this is further evidence that F [Fanakalo] is a strategic interaction resource - one connoting power and domination when it is an unmarked choice, and solidarity, when it is a marked choice".

82 6.

Carol Myers-Scotton Conclusion

In this paper I have shown how both the non-elite and the elite find ways to empower themselves by the ways in which they employ codeswitching in the micro-level of interpersonal relationships. They use a pattern of codeswitching that is originally associated with elite closure in developing nations. Elsewhere I have argued (Myers-Scotton 1990; 1993a) that one of the ways that the educated elite establish themselves is through their patterns of language use. That is, they can be identified, not just by the languages they know, but by how they use them in patterns of codeswitching. Speakers who are not part of the educated elite can project powerful images of themselves by employing some of the codeswitching strategies associated with the elite. Two examples are cited. The first is of a rich businessman in eastern Nigeria who successfully passes as a university graduate by using the language patterns of such graduates. The second comes from Sir Lanka. In this example, a job applicant for a position as an instructor at a university, where Tamil-English bilingualism is expected, makes up for his apparent lack of English proficiency by using limited, but academic, English in a codeswitching pattern. In the other pair of examples discussed in this paper, persons, whose attributes place them as members of the elite, use codeswitching to present themselves as common people. In an example from Kenya, an educated accountant uses codeswitching to project a humble identity. He switches to Swahili, the language of the wananchi ('indigenous citizens') as he becomes a supplicant in order to ask for financial aid from a businessman. There is some irony in his use of Swahili, but there is even more in the use of Fanakalo and Afrikaans in South Africa in the final example. Speaking to fellow persons of European descent, an Englishman uses the two languages that have little prestige in their circles. On the one hand, he is showing what a "real" South African he has become by showing off his ability to speak even a few words in either of these languages. On the other hand, he is ironically lowering himself from his elite status. In all of these cases, there are a number of levels of intentionality that one can argue are communicated by the particular pattern of codeswitching in the specific contexts considered. One can also interpret the choice to use codeswitching at all as an exercise of the speaker's rationality. Under a rational choice model, "rational" does not mean that the speaker is making what is objectively the best choice; rather, "rational" means that the speaker is making cognitive calculations. Based on weighing a number of

How codeswitching as an available option empowers bilinguals

83

factors, speakers make what they think will be the "best choice" for them personally. Thus, rational choices are subjective choices and they can be interpreted as having more than one social meaning (cf. Myers-Scotton and Bolonyai 2001). Certainly, one can see these examples as bids for solidarity as much as for power. One can even say that the speakers' use of codeswitching rather than of monolingual speech in a prestige language implies a certain modesty. But the ambiguity of modesty itself implies an empowering result.

References Adendorff, R. D. 1993 Ethnographic evidence of the social meaning of Fanakalo in South Africa. Journal of Pidgin and Creole Languages 8: 1-28. Canagarjah, A. S. 1995 The political economy of code choice in a "revolutionary society": Tamil-English bilingualism in Jaffna, Sri Lanka. Language in Society 24: 187-212. Essien, O. 1995

The English language and code-mixing: A case study of the phenomenon in Ibibio. In New Englishes, a West African Perspective, A. Bamgbose, A. Banjo, and A. Thomas (eds.), 269-283. Ibadan: University of Ibadan.

Myers-Scotton, C. M. 1990 Elite closure as boundary maintenance: The case of Africa. In Language Policy and Political Development, B. Weinstein (ed.), 25-42. Norwood NJ: Ablex. 1993a

Elite closure as a powerful language strategy: The African case. International Journal of the Sociology of Language 103: 149-163.

1993b

Social Motivations for Codeswitching: Evidence from Africa. Oxford: Oxford University Press.

1993c

Duelling Languages: Grammatical Structure in Codeswitching. Oxford: Oxford University Press.

2002

Contact Linguistics: Bilingual Encounters and Grammatical Outcomes. Oxford: Oxford University Press.

84

Carol Myers-Scotton

Myers-Scotton, C. M. and A. Bolonyai 2001 Calculating speakers: Codeswitching in a rational choice model. Language in Society 30: 1-28.

Section 2. Language policy and language planning: Empowering speakers of minority languages in communities and institutions

Language policy failures1 Bernard Spolsky

1.

Explaining the title

Language policy is an ambiguous and polysemous term, used in the field to refer both to the wide abstraction (the customary consensual judgments and practices of a speech community with regards to the appropriateness of a large number of significant choices among all the kinds of variants allowed in speech or writing) and to a specific policy adopted and explicitly stated for a defined circumstance and place. My concern in this paper will be with failures of the second kind of policy. A more precise way of expressing it would therefore be "failures in efforts at language management (or planning or engineering)." Failure too is ambiguous here. It can mean failure to implement a management decision (e.g. the fact that English survived the few years allowed by the Indian constitution) or failures in spite of efforts to implement (e.g. the fact that English and Afrikaans remain the main South African languages in spite of the efforts being made for the other nine languages of the constitution) or unanticipated results of language management (e.g. the fact that language revival commonly produces forms of the language that were markedly different from the traditional forms that it was intended to revive). Failure could also refer to the fact that a policy has not yet achieved its goal. There are two questions that follow. The first is why the (supposedly) managed speakers won't listen to the managers (why won't the Irish all use their language? Why do children continue to make spelling mistakes? Why do people still swear and curse and blaspheme? Why do people still speak Kurdish in Turkey or Basque in France?). The second is, why don't putative language managers (governments, language academies, education departments, style experts, priests, schoolteachers) listen to the experts (language policy and planning scholars like us) and either learn that the waves are untamable ("Leave your language alone!" some of us say (Hall 1950) or learn from us (Kaplan and Baldauf 1997) how to do good language planning.

88

Bernard Spolsky

My topic, then, can be more simply restated: is language management possible?

2.

Some successes and failures

In this part of the paper, I want to list and briefly comment on a number of the better-known and more intensively studied cases of language management. The list is an ad hoc and personal one that could easily be expanded.2 One might start, following the principle of looking first at the case of bad language (Spolsky 2004: Chapter 2), by noting the general failure of religious authorities to stamp out blasphemy and of myriad educators and parents and city governments and censorship boards to ban obscenity. The most popular television programs regularly warn against the "adult language" they contain. Similarly, one notes the conclusion of one scholar (Pauwels 1998: 221) that twenty years of efforts to control sexist language have not produced promising results. This suggests that it is as hard to manage language as to control nature, as King Cnut demonstrated to his subjects with the waves. Examples like these abound, but as students of language policy generally prefer to deal with issues concerning the status and form of national languages, I will move to these. Hebrew is regularly cited as the greatest success story in language management, with its revival involving the restoration of natural intergenerational transmission, which I call revitalization (Spolsky 1991), and of daily use, which Fishman (1991) calls revernacularization. Looking at other aspects of Hebrew revival more closely, one finds that the Hebrew Language Academy and the Ministry of Education have been signally ineffective in maintaining the purity of the language from foreign influence, in preventing major changes in phonology and syntax so that the revived language is fundamentally altered (Izre'el 2003), in defending against an incursion into numerous domains of English, and in reforming a chaotic and unstandardized spelling system. In addition, scholars have pointed out the deleterious effects of the new Hebrew hegemony in replacing traditional Jewish plurilingualism by Hebrew-English bilingualism (Ben-Rafael 1994) as well as in contributing the death of numerous Jewish varieties (Spolsky and Shohamy 1999) and in weakening the position of Arabic (Amara 2002; Saban and Amara 2002).

Language policy failures

89

In contrast (and the contrast is perhaps best expressed by 0 Laoire (1999)), Irish is regularly cited as a great failure, so that a century of language revival efforts have not led to renewed use of the language though they have kept it alive ( 0 Riägain 1997; 2001). Barbour and Carmichael (2000: 37) note that it is the only national language of a sovereign European state spoken by a small minority of the population. But it remains true that efforts continue to be made to offer immersion and other programs for children, and that one of the tasks taken on for North-South cooperation has been the establishment of a North-South Language Board to improve the positions of Irish and of Ulster Scots. If prizes were to be offered for effort in language management, France with a score or more agencies and academy and committees and a full set of language laws and regulations would be close to the winner (Ager 1999; Judge 2000). French language management started with Cardinal Richelieu who established the Acadimie frangaise as an instrument to bolster central power (Cooper 1989), and was proclaimed in the Jacobin decision in the early years of the French Revolution that valued equality (which it interpreted as everyone needing to speak French) over liberty. France established the principle of "one nation, one language" which has since then dominated the language ideologies of nation-states, setting out to ensure that everyone under French political rule spoke standard French. This policy took a long time to implement, and in the 1880s, ministers of education were still trying to find enough French speaking teachers; in the 1920s, they were still concerned with religious schools sheltering other languages; in 1975 a law was passed requiring French for commerce and public life; in 1992 (in time for the Maastricht Treaty) the constitution was finally amended to proclaim French as the only language; and in 1994 the Toubon law spelled out what this should mean. From the beginning, a major role of the Academie was to protect the purity of the French language; this has more recently been interpreted as to keep out English. The policy was applied to the French colonies, which as a general rule offered education only in the metropolitan language. With the collapse of the colonial empire after France's defeat in the Second World War, this language policy was the base for a program to establish a political union around francophonie. What about the results? If you read French papers, you can believe that French is still under serious threat from English, which is the main foreign language taught not just in France but in most of Europe. The languages of the periphery stayed alive long enough to start picking up protection from recent European language rights policy, and there are reversing language

90

Bernard Spolsky

shift activities for Breton, Basque, Catalan, Corsican and Occitan (whose attrition can be blamed on economic and demographic changes rather than language policy), and threats of finding ways to protect the varieties of Arabic (Algerian with over half a million speakers, Moroccan and Tunisian with over 200,000 each) and Berber (over 600,000 speakers of different varieties). French language policy then does not seem to have had the success that all the effort seems to warrant. Norway (like other countries that achieved national independence in the 19th century) was soon faced with the challenge of developing a variety that could be considered symbolically national. The subsequent conflict over the form of language became the topic of the first major scholarly study of language management (Haugen 1966). The problem was to create a national variety of Norwegian out of the wide range of local dialects. Two conflicting approaches emerged; one preferred the language of the cities, while the other preferred traditions associated with the villages. After long and at times bitter political debate, a compromise was reached to require all schools to teach both written varieties to their pupils, while being careful to leave the local spoken dialects alone (Vikor 1989), so that now 80% of the population use one variety and 20% the other. The main cost presumably is to the school children who have to learn both. In spite of all the effort put into Norwegian language management, Norwegians are more and more finding it useful to learn and use English. An alternative method of resolving language conflict is territorial, as exemplified in Switzerland, Finland, and Belgium. The Swiss Confederation established in 1848 allowed linguistic autonomy to each of the twentysix cantons into which the country was divided: seventeen chose German, four French, three both, one Italian, and one Romansch, Italian and German. Schools teach standard German, but the Swiss continue to use their own variety in order, Ammon (1997) suggests, in order to keep their identity clear. Bilingual cantons (like Fribourg-Freiburg) are struggling to find a way to draft a language item for their constitution; Zurich is the first canton to declare English its second language. German-speaking cantons teach High German to their dialect-using pupils, but French-speaking cantons have a major problem if they feel the need to teach both High and Swiss German. In Finland, too, the solution was finally territorial: after independence from Swedish rule, Finnish became the national standard language, but the minority Swedish, spoken in southwestern coastal areas by 5% of the population, received constitutional equality in the 1919 constitution. There was rivalry for 20 years, but after the Second World War, it

Language policy failures

91

was accepted that Finnish speakers should learn Swedish just as Swedish speakers were forced to learn Finnish. In Belgium, the conflict between the speakers of Germanic dialects who support Dutch and of Romance dialects who support French has been an ongoing struggle for a century and a half. The latest compromise has been federalism, with devolution of the authority to the regions (Covell 1993). Dutch, French and to a lesser extent German have their own institutions in their defined regions, and contact is decreasing except in Brussels. The European Court has upheld the territorial solution, but there continues to be friction at the seams. In schools, dialect speakers (the majority) and minority immigrant groups are expected to acquire the appropriate standard language. Language management was an inevitable result of the conquest and resettlement associated with the spread of political and religious and military empires. Aramaic, Latin, Greek, Arabic, and Mayan were all spread in this way in the ancient world (Brosnahan 1963). Greek settlement established the language in the eastern Mediterranean; Roman settlement of retired soldiers and Church use provided the basis for the official role of Latin in Western Europe and the development of the Romance languages; Islamic military and religious conquest replaced Aramaic with Arabic. In the modern world, the Spanish conquest of Central and South America, combined with a policy of teaching only Spanish and aiming for a pure Spanish (echoing Spanish contempt for the racial mixtures that were inevitable in the homeland purged of Arabs and Jews and the colonies with rampant intermarriage with indigenous peoples) or Portuguese contributed for centuries to the extermination of the autochthonous languages. As with the French efforts to wipe out the peripheral languages, the limitations in the provision of standard language education left the languages alive outside the main areas of settlement, so that attrition depended on demographic and economic factors rather than direct policy. Many are still alive, and so it has been possible to attempt to maintain or revive them now that there has been some recognition of civil and language rights in Latin America (Hornberger and King 2001). Russian (pre- and post-Soviet) and Soviet language policies, interpretable as an example of imperialism (Grenoble 2003), have gone through a number of stages, each with its failures and partial successes. Russification of conquered areas was not uncommon in the 19th century (Weeks 2002). Soviet rule after 1917 developed a policy that granted exceptional rights to many languages (Martin 2002) in order to develop literacy and speed up political education. This policy, embedded in the Constitution and sup-

92

Bernard Spolsky

ported by Lenin, started to be reversed in the 1930s, when under Stalin a process of favoring Russian over the other languages began. This policy too was later modified during perestroika but there is recently new evidence within Russia of encouragement of Russian over alternatives. Since the break-up of the Soviet Union, each of the successor states (including Russia) has started to work to strengthen the claims of its titular language, although the policy is made more complex by the status and role of Russian outside Russia and the European insistence on minority language rights in the Baltic States. In the early 20th century, one of the most elaborate cases of language management was Turkey. The main goal of language reform was to purify Turkish of Arabic and Persian influences, as part of developing a secular and westernized state. One major step was the switch from Perso-Arabic script to Roman; the second was a campaign to replace Arabic lexical items with Turkish (or European) words. The script change was effective, leaving large quantities of historical and literary texts blocked to access for modern Turks who do not learn a second script. More seriously, Lewis (1999) argues, the very success of language reform has been a catastrophe in blocking access to texts written 50 years ago. Other early 20th century language management endeavors include the Treaty-supported efforts to impose language rights in the territories previously under control of the defeated powers. Central Europe and the Balkans were intended to set up model multilingual states, which held together for a number of years but have more recently collapsed under post-Soviet ethnic nationalism. The independent Arabic speaking states carved out of the Turkish empire all chose diglossia by accepting Classical Arabic as a single standard (Suleiman 1994); in the Maghreb, this has involved a slow and incomplete process of Arabization (Djite 1992) leaving problems of the continuing use of French for education at the post-primary level and a failure to come to grips with the issue of Berber. The major period of modern language planning endeavors followed the break-up of the European-based empires after the end of the Second World War. Generally speaking, former French and Portuguese colonies where only the metropolitan language had been used in education and government had little choice but to continue this, although many of them had very small and elite French- or Portuguese-speaking populations (Vilela 2002). In North Africa, this was modified by Arabization programs of various degrees of success. In Africa, former French colonies generally follow the

Language policy failures

93

recommendation of the 1944 Brazzaville conference to use only French for education. In former British colonies, a choice was offered. In India in the 19th century, there was a conflict between Anglicists who believed that education should be offered only in English and Orientalists who favored offering westernized education in local languages and vernaculars. As a result of compromise solutions (Khubchandani 1997), nationalist movements were able to support their languages, so that when India became independent, it adopted a constitution that recognized a dozen official languages, allowed for initial education in other vernaculars, and planned the replacement of English by Hindi as a working national official language within a few years. This led to a flurry of language management activity, including language acquisition and lexical elaboration and standardization to support a widely proclaimed three language formula. The first big failure of Indian language policy was the recognition that English would not go away (Dua 1996) and is firmly entrenched as language of post-primary education. The second was that only a third of the classified languages were in fact used in those mainly rural areas where initial education is in the vernacular. The third is that language conflict continues, as languages other than Hindi make continued claims. Former British colonies in Africa have also continued colonial models, with vernacular languages generally restricted to initial primary instruction. Only a few countries have moved at all towards developing their national vernaculars as functioning educational and governmental languages: Tanzania considers Kiswahili an official language and uses it widely, but conducts secondary and tertiary education in English (Vavrus 2002); Malawi recognizes Chichewa but uses English for advanced education and political life (Matiki 2003). Given their political and economic and health problems, one would not expect African countries to be able to devote scarce resources to the social engineering and language management needed to cultivate local vernaculars effectively. A central feature of US language policy is its decentralization, based on the constitutional decision to leave education to the individual States. Any Federal activity then may be seen as a reluctant decision to interfere with State activities. The most basic of these has been language management as enforcement of civil rights, derived from the notion of equal protection reaffirmed in the Fourteenth Amendment to the US Constitution. This was the basis for ruling that States could not prevent private bilingual education, and was further developed in Title VI of the 1964 Civil Rights Act,

94

Bernard Spolsky

currently implemented in Executive Order 13166 (August 2000) entitled "Improving access to services for people with limited English proficiency." Having withstood Congressional efforts to revoke it and in spite of firm opposition from an English Only movement, this order is starting to have major impact on the provision of interpreting and translation services in courts, hospitals, and offices. It is however certainly premature to label it a success. A second effort building on civil rights was the Bilingual Education Act, passed in 1968 and allowed to die quietly in 2002 (Crawford 2002). Intended as a method of teaching English to speakers of Spanish, it was enlarged to serve other language groups and regularly interpreted as encouraging maintenance of other languages. It is clear that this was not the original intention of the sponsors, but it was regularly attacked as aiming to substitute other languages for English. In the end, it demonstrated the weakness of the Federal government in influencing school language teaching. The third effort is constituted by a number of Federally initiated and funded programs to make up for the failure of State-mandated education systems to maintain the national language capacity (Brecht and Rivers 2000) required for national security, which is within the purview of the Federal government. One such program was the Second World War Army Specialized Training Program, intended to make up for the shortage of personnel able to speak the many languages needed for a global war. A second burst of activity, this time Cold War related, was supported by the National Defense Education Act of 1958, which included support of research and training in uncommonly taught languages. Too small to have major impact, the program showed that the Federal government could find a way to modify language education policy. The third burst of activity now follows recognition of the US role in international affairs, with the passage in 1991 of the National Security Education Program and the current post9/1 1 realization that the US has available only 30% of the language capacity required by intelligence and defense agencies. Again, the picture is of emergency patching rather than the development of a sound long-term policy, although if the new Department of Defense road map is implemented, the US Armed Forces will be plurilingually proficient. Have there been more successful language management efforts? What about the regeneration of Maori in New Zealand? Benton and Benton (2001) are pessimistic and convince Romaine (2002); Spolsky (2003) sees important gains but no signs of restored intergenerational transmission. Or

Language policy failures

95

the defense of French in Quebec? Certainly the extensive legal and political activity (Larrivee 2002) has slowed the shift to English, but (Bourhis 2001) remains cautious. In Spain, efforts for Basque have had limited effects (Azurmendi, Bachoc, & Zabaleta 2001), but Catalan is holding on against the pressure of Castilian (Strubell 2001). In Singapore, strong government action has switched education largely to English (Tickoo 1996) and has maintained roles for Malay, Tamil and Chinese (Kaplan and Baldauf 2003). In Indonesia, a new national language has been implemented (Dardjowidjojo 1998), but questions remain about the extent of its use (Goebel 2002). A final cluster of language management efforts to mention are those concerned with the rescue of endangered languages. Two kinds of preservation have been tried (often together). The first is the museum variety like salvage archeology - involving recording the linguistic facts of the language before the last speakers die so as to ensure that the enormous variety of human culture and language is not lost. The second, more socially motivated, is to attempt to preserve the language as a spoken living instrument, following the stages set out in Fishman (1991), with developments over the next decade documented in Fishman (2001). Only a small proportion of the 5-6000 languages at risk are included in these salvage management efforts, and results are generally far from promising. Romaine (2002) notes that fewer than 4% of the world's languages have any kind of official status, and regrets the "weak linkages between policy and planning which render ineffective most policies aimed at assisting endangered languages." (2002:3)

3.

The complexity of language management

Is this record of management failure in the area of language not reflected in many other human domains? Are we that much better at managing peace, or economies, or health, or natural resources? Perhaps so, but we still need to consider language management in its own right. The first question to consider, then, is whether or not the brief review in this paper has done justice to such efforts? Studies of spelling and orthography reform (Geerts, Van den Broeck, and Verdoodt 1977; Gold 1977; Gregersen 1977; Ο Murchü 1977) make clear that such activities include success and failure. France may not have homogenized its population, but

96

Bernard Spolsky

it has maintained its national consensus on the beauty and importance of French; Irish has kept its national importance even if not its functional use. We need then to clarify the notion of success. Take for instance the case of Maori. In thirty years of mobilization of resources around what was a threatened language, its status has been re-established (it is now official, and widely if symbolically supported by government) and a major effort in schools has corrected the distribution of Maoris who know their own language, so that the majority are now young. It is true that few of these young people are using the language much outside school, but they are committed to its continuation. It can fairly be said to have been regenerated (Hohepa 2000) if not revitalized. Given the pervasiveness and complexity of language in human society, it is not surprising that language practices are so hard to influence. Perhaps it will help to restate the model of language policy that I argue for (Spolsky 2004). Language management occurs when someone with (or claiming) authority over others sets out to modify some part of their language practices or of their language beliefs. It is language management when a parent tries to prevent children from swearing, or when a grandparent tries to encourage them to use a heritage language; it is language management when a government tries to outlaw a language or require all its citizens to use the same language. These are examples of language policy in the widest sense, but it is more useful to divide management into three major independent but often related tasks. The first is changing the status and domain of approved use of a language (e.g. banning Kurdish in Turkey, re-establishing the primacy of Estonian in Estonia, switching from Yiddish to Hebrew in Jewish settlements in Ottoman Palestine, making Irish an official language in the European Community, deciding on the language of instruction for a school). The second is cultivating an existing language (e.g. developing or modifying a writing system, reforming a spelling system, switching from one script to another, developing new lexicon for modernization). The third is teaching a language to people who do not speak it. While these are distinct, there are logical interconnections: when you make a language official, you need to cultivate it to carry out official functions and teach it to citizens. What drives the whole process? More precisely, why does someone or some body take on language management? While in most speech communities it is generally considered impolite to draw attention to someone's speech problems, it seems to be universal for parents, teachers and other caretakers to take responsibility for correcting the speech of their charges.

Language policy failures

97

The most reasonable goal of this is to assure the widest possible comprehension, for the standard form of the language is assumed to have the widest intelligibility. A second goal is to avoid giving offense to others. A third is to produce speech that is not just not bad (unclear or stigmatized) but that is positively good, the goodness defined by the absence of impure (foreign) features and the presence of valued identity markers. Other institutions take up the task - religions obviously, and peer groups (whose language values may well work in the opposite direction) and governments who apply the sometimes conflicting criteria of efficiency (hoping that every citizen can be easily communicated with and can easily communicate with others) and identity (using the language as the symbol for national identity). At the simplest level, the fact that there are competing managers (parents, schools, peers, churches, local and national governments) and each with divided and competing goals guarantees that there is no easy row to hoe, but rather (to mix the metaphor) a complex set of overgrown paths through the jungle. Nor is it to be assumed that the stated reason is the most important one. Calls for purity may well be the surface expression of a desire to maintain difference and identity. Similarly, an argument for efficiency (such as continuing to use a former colonial language) may mask an interest in maintaining elite power of the small group who managed to achieve proficiency in it during the colonial period (Mazrui and Mazrui 1996). Should we be studying success in terms of stated or underlying motivation? Ager (2001), who argued that French language policy was motivated by identity, insecurity and image (Ager 1999), adds more "i"s to his list: ideology, inequality, and integration and instrumentality, and while Paulston (2002) fairly criticizes him for failing to give clear cut cases illustrating each, she agrees that the approach is a useful starting point for understanding the relations between attitudes and language management. But it turns out to be much more complex to pin down motivation. Analyzing one of the cases, Maori in New Zealand (Spolsky 2003), one quickly finds how complex the situation was. It is hard to distinguish the original motivation of the group who in the 1970s started out in the Te Ataarangi movement to learn a language that was (Benton 1981; 1991) already well on the way to what Fishman (1991) classified as the lowest stage on his Graded Intergenerational Disruption Scale. No doubt they constituted part of the growing movement for ethnic revival (Fishman, Gertner, Lowy, and Milan 1985) for their language-focused activities paralleled socio-political demands for recognition of Maori rights that culmi-

98

Bernard Spolsky

nated in the establishment of the Waitangi Tribunal. The mobilization for Maori language revival then continued as one part of a more general concern for re-establishment of identity and power, with the main work of the Tribunal turning out to entail financial settlement with the various Maori tribes for the loss of land, hunting, fishing and other rights. The language movement took on a life of its own, especially as it led to Maori control over education, first in the pre-school Te Kohanga Reo movement, then in the establishment of independent Maori-immersion primary schools (Kura Kaupapa Maori), where the insistence on the use of only Maori language set up barriers within which identity and control could develop. More recently, the movement has come under strain from a number of directions. One has been the attempt by the government to reestablish control over these schools by demands for accountability; the second has been the tension within the schools between the ideological claims of Maori identity and the instrumental arguments for learning English; and the third has been the fact that while Maori language regeneration has been mainly (as Irish revival) an urban matter among essentially detribalized Maori, the weight of Tribunal activities has moved money and power back to the Tribes. Even within the language movement, then, there remain supporters of the three major positions, assimilationist, amalgamationist and separatist, that are seen as possible solutions to the challenge produced by the ethnic divide between white and Maori. But the actions of the New Zealand government are not much simpler to account for. In New Zealand as a British colony, there was the common debate between the supporters of education through the vernacular and education through English, with the second group victorious towards the end of 19th century and strongly supported by assimilationist Maoris. During the 20th century, with English firmly in place, amalgamationists were happy to start teaching Maori language as a subject in schools, teaching it to Maori and white alike. With the growth of Maori activism in the 1970s, the Department of Education too came under pressure to provide bilingual programs, which it did reluctantly (Spolsky 1989). A decision of the Waitangi Tribunal that Maori language maintenance had been promised in the 1840 Treaty provided legal backing, and the government was persuaded to pass the Maori Language Act making Maori "an official language of New Zealand". The act gave rights for interpreters in court, and established a Maori Language Commission but did little else. It was only ten years later that a combination of legal and civil or human rights arguments were used to persuade

Language policy failures

99

Cabinet to adopt a Maori language strategy which included steps to be taken by government agencies to strengthen the place of Maori language. The main areas of action and tension continue to be schools, where a large number of Maori children are now in Maori immersion classes in government-funded schools, and broadcasting, where there are steady demands for more hours of Maori language radio and television. So there is no single motivation, but rather various motivations of the various parties. Consider then the results. The fact that the government has accepted the need to publish its documents in Maori means that there are plenty of jobs for the graduates of the school programs - their literacy skills make up for their lack of oral fluency. At the same time, the existence of Maori immersion schools (like the growing if still small number of Irish immersion schools in Dublin) provides a focus for parents who wish to express their identity by sending their children to school (rather than by having to learn and use the language themselves). Mobilization around language is certainly a less harmful focus than other channels for political and national activity. The development has helped to disappoint the hopes of the pure assimilationists, those who hoped that within a century Maori and white would merge completely. Nor has it satisfied the extreme separatists, for language regeneration has not built a new closed society. Rather, it fits the complex ongoing negotiation between the two communities, maintaining difference without producing new barriers. Not failures, then, or successes, but modifications seem to be possible results. The wise language managers will acknowledge, as Cnut did, that they cannot stop the waves, but will aim to be surfers who ride them successfully.

4.

Conclusions

While language management seldom achieves the goals hoped for, there is sufficient evidence that, intelligently applied, it helps mitigate some of the troubles it is intended to cure. Just as in other spheres of human and social policy making, its complexity frequently produces unwanted or unsatisfactory results. The challenge to develop a workable theory remains. Such a theory needs to spell out not just the nature of language policy (Spolsky 2004) but survey the contributions that linguists have made to its study and sketch the nature of the conditions under which language managers can

100

Bernard Spolsky

intervene successfully in the language practices and beliefs of a speech community.

Notes 1. This paper was drafted while I was a Senior Research Scientist at the Center for Advanced Studies in Language at the University of Maryland. It is the account of a first effort to move from considering language policy as a whole to looking at one of its components, language management, and must be considered as work in progress. 2. Given the title, I may have unconsciously erred on the side of selecting cases where management has not been successful.

References Ager, D. E. 1999 2001 Amara, M. 2002 Ammon, U. 1997

Identity, Insecurity and Image: France and Language. Clevedon, Philadelphia and Adelaide: Multilingual Matters Ltd. Motivation in Language Planning and Language Policy. Clevedon, England, Buffalo, USA: Multilingual Matters Ltd. The place of Arabic in Israel. International Journal of the Sociology of Language 158:53-58. National-variety purism in the national centers of the German language. In Language Choices: Conditions, Constraints, and Consequences, M. Pütz (ed.), 161-178. Amsterdam, Philadelphia: John Benjamins.

Azurmendi, M.-J., E. Bachoc, and F. Zabaleta 2001 Reversing language shift: The case of Basque. In Can Threatened Languages be Saved? J. A. Fishman (ed.), 234-259. Clevedon, Avon: Multilingual Matters Ltd. Barbour, St. and C. Carmichael (eds.). 2000 Language and Nationalism in Europe. Oxford, New York: Oxford University Press.

Language policy failures Ben-Rafael, E. 1994 Benton, R. A. 1981

1991

101

Language, Identity and Social Division: The Case of Israel. Oxford: Clarendon Press. The Flight of the Amokura: Oceanic Languages and Formal Education in the Pacific. Wellington: New Zealand Council for Educational Research. The Maori Language: Dying or Reviving (Alumni-in-Residence Working Papers Series). Honolulu, HA: East-West Center.

Benton, R. A. and N. Benton 2001 RLS in Aotearoa/New Zealand 1989-1999. In Can Threatened Languages be Saved? J. A. Fishman (ed.), 422—449. Clevedon/Avon: Multilingual Matters Ltd. Bourhis, R. Y. 2001

Reversing language shift in Quebec. In Can Threatened Languages be Saved? J. A. Fishman (ed.), 101-141. Clevedon/Avon: Multilingual Matters Ltd.

Brecht, R. D. and W. P. Rivers 2000 Language and National Security in the 21st Century: The Role of the Title VI/Fulbright-Hays in Supporting National Language Capacity. Dubuque Iowa: Kendall-Hunt Publishing Company. Brosnahan, L. H. 1963 Some historical cases of language imposition. In Language in Africa, J. Spencer (ed.), 7-24. Cambridge: Cambridge University Press. Cooper, R. L. 1989 Covell, M. 1993 Crawford, J. 2002

Language Planning and Social Change. Cambridge: Cambridge University Press. Political conflict and constitutional engineering in Belgium. International Journal of the Sociology of Language 104: 65-86. Obituary: The Bilingual Education Act: 1968-2002. URL: http:// ourworld.compuserve.com/homepages/JWCrawford/T7obit.htm

Dardjowidjojo, S. 1998 Strategies for a successful national language policy: The Indonesian case. International Journal of the Sociology of Language 130: 35-47.

102

Bernard Spolsky

Djite, P. G. 1992

Dua, H.R. 1996

Fishman, J. A. 1991

The Arabization of Algeria: Linguistic and sociopolitical motivations. International Journal of the Sociology of Language 98: 15-28. The spread of English in India: Politics of language conflict and language power. In Post-Imperial English, J. A. Fishman, A. Rubal-Lopez, and A. W. Conrad (eds.), 557-588. Berlin and New York: Mouton de Gruyter. Reversing Language Shift: Theoretical and Empirical Foundations of Assistance to Threatened Languages. Clevedon/England: Multilingual Matters Ltd.

Fishman, J. A. (ed.) 2001 Can Threatened Languages be Saved? Reversing Language Shift, Revisited: A 21st Century Perspective. Clevedon/Avon: Multilingual Matters Ltd. Fishman, J. Α., Μ. Η. Gertner, Ε. G. Lowy, and W. G. Milan 1985 The Rise and Fall of the Ethnic Revival: Perspectives on Language and Ethnicity. Berlin: Mouton de Gruyter. Geerts, G., J. Van den Broeck, and A. F. Verdoodt 1977 Successes and failures in Dutch spelling reform. In Advances in the Creation and Revision of Writing Systems, J. A. Fishman (ed.), 179-245. The Hague and Paris: Mouton. Goebel, Z. 2002

Gold, D. 1977

Code choice in interethnic interactions in two urban neighbourhoods of Central Java, Indonesia. International Journal of the Sociology of Language 158: 69-87. Successes and failures in the standardization and implementation of Yiddish spelling and romanization. In Advances in the Creation and Revision of Writing Systems, J. A. Fishman (ed.), 307369. The Hague and Paris: Mouton.

Gregersen, Ε. Α. 1977 Successes and failures in the modernization of Hausa spelling. In Advances in the Creation and Revision of Writing Systems, J. A. Fishman (ed.), 421^140. The Hague and Paris: Mouton.

Language policy failures Genoble, L. A. 2003 Hall, R. A. 1950 Haugen, E. 1966 Hohepa, P. 2000

103

Soviet Language Policy. Dordrecht: Kluwer Academic Publishers. Leave Your Language Alone! Ithaca: Cornell University Press. Language Conflict and Language Planning: The Case of Modern Norwegian. Cambridge, MA.: Harvard University Press. Towards 2030 AD: Maori Language Regeneration: Examining Maori Language Health. Paper presented at the Applied Linguistics Conference, Auckland, New Zealand.

Hornberger, Ν. Η. and Κ. A. King 2001 Reversing language shift in South America. In Can Threatened Languages be Saved? J. A. Fishman (ed.), 166-194. Clevedon/Avon: Multilingual Matters Ltd. Izre'el, S. 2003

Judge, Α. 2000

The emergence of spoken Israeli Hebrew. In Corpus Linguistics and Modern Hebrew, B.H. Hary (ed.), 85-104. Tel Aviv: Tel Aviv University. France: One state, one nation, one language? In Language and Nationalism in Europe, S. Barbour, and C. Carmichael (eds.), 44-84. Oxford and New York: Oxford University Press.

Kaplan, R. B. and R. B. Baldauf 1997 Language Planning from Practice to Theory. Clevedon and Avon: Multilingual Matters Ltd. 2003

Language and Language-in-Education Planning in the Pacific Basin. Dordrecht: Kluwer Academic Publishers.

Khubchandani, L. M. 1997 Language policy and education in the Indian subcontinent. In Encyclopedia of Language and Education. Vol. 1: Language Policy and Political Issues in Education, R. Wodak, and D. Corson (eds.), 179-187. Dordrecht: Kluwer Academic Publishers. Larrivee, P. (ed.) 2002 Linguistic Conflict and Language Laws: Understanding the Quebec Question. Basingstoke UK: Palgrave.

104

Bernard Spolsky

Lewis, G. 1999 Martin, T. 2002 Matiki, A. J. 2003

The Turkish Language Reform: A Catastrophic Success. Oxford: Oxford University Press. The Affirmative Action Empire: Nations and Nationalism in the Soviet Union 1923-1939. Ithaca NY: Cornell University Press. Linguistic exclusion and the opinions of Malawian legislators. Language Policy 2(2): 153-177.

Mazrui, A. M. and A. A. Mazrui A tale of two Englishes: The imperial language in post-colonial 1996 Kenya and Uganda. In Post-Imperial English, J.A. Fishman, A. Rubal-Lopez, and A. W. Conrad (eds.), 271-302. Berlin and New York: Mouton de Gruyter. O Laoire, M. 1999

O Murchü, Μ. 1977

0 Riägain, P. 1997 2001

Paulston, C.B. 2002

Pauwels, A. 1998

Athbheochan na heabhraise: Ceacht don Ghaeilgel (Revival of Hebrew: Example for Irish). Baile Atha Cliath: An Clochohar Tta. Successes and failures in the modernization of Irish spelling. In Advances in the Creation and Revision of Writing Systems, J.A. Fishman (ed.), 267-289. The Hague and Paris: Mouton. Language Policy and Social Reproduction: Ireland 1893—1993. Oxford: Clarendon Press. Irish language production and reproduction 1981-1996. In Can Threatened Languages be Saved? J.A. Fishman (ed.), 195-214. Clevedon and Avon: Multilingual Matters Ltd. Review of Dennis Ager. Motivation in Language Planning and Language Policy, and Kas Deprez and Theo Du Plessis, Multilingualism and Government. Language in Society 31(5): 790796. Women Changing Languages. Longman.

New York: Addison Wesley

Language policy failures Romaine, S. 2002

105

The impact of language policy on endangered languages. MOST Journal of Multicultural Studies 4(2).

Saban, I. and M. Amara 2002 The status of Arabic in Israel: Reflections on the power of law to produce social change. Israel Law Review 36(2): 5-39. Spolsky, B. 1989

Maori bilingual education and language revitalization. Journal of Multilingual and Multicultural Development 9(6): 1-18.

1991

Hebrew language revitalization within a general theory of second language learning. In The Influence of Language on Culture and Thought: Essays in Honor of Joshua A. Fishman 's sixty - fifth Birthday, R.L. Cooper and B. Spolsky (eds.), 137-155. Berlin: Mouton de Gruyter.

2003

Reassessing Maori regeneration. Language in Society 32(4): 553-578.

2004

Language Policy. Cambridge: Cambridge University Press.

Spolsky, B. and E. Shohamy The Languages of Israel: Policy, Ideology and Practice. Cleve1999 don: Multilingual Matters. Strubeil, M. 2001

Suleiman, Y. 1994

Tickoo, M.L. 1996

Vavrus, F. 2002

Catalan a decade later. In Can Threatened Languages be Saved? J.A. Fishman (ed.), 260-283. Clevedon and Avon: Multilingual Matters Ltd. Nationalism and the Arabic language: An historical overview. In Arabic Sociolinguistics: Issues and Perspectives, Y. Suleiman (ed.), 3-24. Richmond: Curzon Press. Fifty years of English in Singapore: All gains, (a) few losses? In Post-Imperial English, J. A. Fishman, A. Rubal-Lopez, and A.W. Conrad (eds.), 431-456. Berlin and New York: Mouton de Gruyter. Postcoloniality and English: Exploring language policy and the politics of development in Tanzania. TESOL Quarterly 36(3): 373-397.

106

Bernard Spolsky

Vikor, L. S. 1989 Vilela, M. 2002

Weeks, T. R. 2002

The position of standardized vs. dialectal speech in Norway. International Journal of the Sociology of Language 80: 41-60. Reflections on language policy in African countries with Portuguese as an official language. Current Issues in Language Planning3(3): 306-316. Religion and russification: Russian language in the catholic churches of the 'Northwest Provinces' after 1863. Kritika: Explorations in Russian and Eurasian History 2(1): 87-110.

Empowerment through the community language A challenge1 Michael Clyne

This paper explores to what extent it is possible to empower immigrants and their children by raising the status of their 'minority' languages through status planning, especially in the education system. This is discussed in the context of deep-seated monolingual/monocultural attitudes in the community and an environment in which disabling changes have outweighed enabling changes, and especially where there has been a move away from comprehensive, cohesive language policy development. I will begin with a few words about Australia's changing ethnolinguistic demography. For most of Australia's non-indigenous history, the British Isles had been the predominant and preferred source of immigrants. The exclusion of immigrants of non-European background, strong resistance to largescale immigration from continental Europe, and a rigid assimilation policy marked the period 1901-1945. The postwar mass immigration scheme, initiated to build up Australia's secondary industry and to increase its population as protection from an anticipated Asian invasion, brought large numbers of non-English-speaking immigrants to the country because the supply of immigrants from the British Isles had become too limited. Over the years, new immigrants started coming from countries culturally more and more distant from the British. With the abolition of racially based immigrant restrictions in the early 1970s, Asian countries became an important source, and increasingly immigrants were selected on the basis of educational skills and wealth and entered the higher rungs on the socioeconomic ladder. According to the 2001 Australian Census, about 16% of the population then used a language other than English at home. In Sydney and Melbourne, the proportion was 29% and 27% respectively. Table 1 indicates the top 20 languages other than English in 2001 and changes from the 1991 Census. It will be noted that there has been a decline in the home use of the European languages brought to Australia mainly in the post-World War Π mass immigration program and a steady increase in the use of languages of Asian origin, some of which, such as Chinese and Indonesian, have also

108

Michael Clyne

been taught in schools and universities because of their economic and strategic importance to Australia. Table 1. Top 20 languages other than English spoken at home, 2001 Language

2001

% change from 1991

Italian

353606

-15.6

Greek

263718

-7.7

Cantonese

225307

+38.9

Arabic

209371

+28.6

Vietnamese

174236

+58.1

Mandarin

139288

+155.9

Spanish

93595

+3.4

Tagalog (Filipino)

78879

+33.4

German

76444

-32.6

Macedonian

71994

+11.7

Croatian

69850

+10.7

Polish

59056

-11.8

Turkish

50692

+20.8

Serbian

49202

+102.2

Hindi

47817

+110.4

Maltese

41392

-21.9

Dutch

40187

-14.7

French

39643

-12.9

39528

+100.1

38724

+42.4

Korean Indonesian

2

(Based on Clyne and Kipp 2002)

Empowerment through the community language 1.

109

Lack of empowerment through community languages in assimilationist Australia

Non-English-speaking immigrants are disempowered by not knowing enough English, for Australian society and most of its institutions operate in English. This means that those without adequate English cannot participate fully in Australian society; in fact in many cases, they cannot earn their livelihood except through unskilled repetitive work. In the immediate postwar period, English was the only language of the administrative and educational domains, and with a few exceptions of the electronic media. To receive access to most information or to improve your status, you needed adequate English. For most of the 20th century, empowerment of immigrants through language could occur only through English. This meant developing English as a Second Language programs for classes on ships, in migrant camps 3 , in cities, on radio, in schools, and later in workplaces. For the first two decades, care was taken that English classes were conducted monolingually in the target language and the class composition was generally controlled to avoid any use of another language in class. Maintenance of the original languages was discouraged and had to take place privately. At the time, (a) hardly any immigrant languages were taught in schools; (b) there were severe restrictions on radio programs in 'foreign languages'; (c) 'foreign languages' in schools were gatekeeper subjects unrelated to communication. The main language taught in schools was French. If a language such as German, Russian or Italian was taught, the motivation was not related to its use in Australia, and discriminatory measures were developed to ensure that students 'with an unfair advantage' did not do too well, and schools teaching German were generally not in areas where German was spoken much. Discrimination included the overuse of translation into the English, in which the English was assessed, in 'foreign language' subjects and severely penalizing particular 'errors' characteristic of second generation bilinguals as well as variable assessment according to background, which was identified illicitly; (d) very few languages were accredited as subjects in the matriculation examination - they were subjects that were taught and examined at university;

110

Michael Clyne

(e) interpreters and translators were employed by banks and department stores to help their customers; in areas such as hospitals, courts, and police stations, much of the interpreting was done by friends and family members, especially young bilingual children; (f) the only holdings of local libraries in languages other than English were a few bilingual dictionaries and school editions of a few French texts. There was a substantial shift to English as the language in which many children spoke and responded to their parents and this is reflected in data from languages of the postwar era, such as Dutch, German, Maltese, Hungarian, Latvian, Lithuanian, and in the second generation Italian. Table 2. Language shift of selected groups to English as home language, 1996 and 2001 Birth place

% shift (Gl 4 , 2001)

% shift (Gl 1996)

% shift (G2 1996) Endogamous

Mother Father ExogaX X mous aggregated

G2 Aggregated

54.4

48.3

80.0

89.4

92.2

91.1

89.7

Germany 54.0

48.2

77.6

9.0

93.6

92.0

89.7

Hungary

35.0

31.8

64.2

85.9

90.7

89.4

82.1

Italy

15.9

14.7

42.6

73.1

80.9

79.1

57.9

Latvia

38.2

na

na

na

na

na

na

Lithuania 41.7

na

na

na

na

na

na

Malta

38.2

36.5

70.0

92.0

94.0

92.9

82.1

Netherlands

62.6

61.9

91.1

95.5

97.2

96.5

95.0

Austria

The German- and Austrian-born 45-54 year olds, who came to Australia with their parents in the 1950s and generally answered their parents in English (Clyne 1967: 27) record about 70% shift to English as the home

Empowerment through the community language

111

language (70.4% shift in the Austrian-born and 69.2% in the German-born members of this age group.). The shift in the parental home was accompanied by two tendencies. One was a low self-estimation of the parents as the children did not appreciate the linguistic capital the parents possessed in their LI, together with a change in family dynamics, as it was often the children who had to act as intermediaries between the family and the wider community. The other tendency in some families was minimal communication between parents and children, for the parents had too little English and the children's competence in LI was insufficient for meaningful interaction to take place. In Sections 2 to 5,1 will outline how conditions were developed to empower bilinguals by raising the status and functions of their 'other' language. In the final sections, I will discuss some of the recent impediments to empowering bilinguals in this way.

2.

New developments in the interests of social equity

The term 'community languages' has been in use since 1975 when it was devised by a migrant education action group in Melbourne to legitimate those languages other than English within the Australian community. It reflected changes that had been taking place since the election of the Whitlam government in 1972. This Labor government was elected after 23 years of conservative rule on a platform of radical reform in social and foreign policy towards a more inclusive and independent-minded Australia. The push for ethnic minority rights reflected similar developments in the US and Canada. In Australia, the policy of multiculturalism both acknowledged demographic reality and also was intended to provide Australians of non-English-speaking backgrounds, also termed 'ethnic Australians', with social equity. The measures introduced responded to the demands of a wide coalition of interests lobbying for the recognition of cultural diversity in policy, including ethnic groups, teacher associations, trade unions, and academics, and expressed through Migrant Workers' Conferences and Migrant Education Action conferences and committees. Many of the measures implemented in Australia relating to language date from the early to mid 1970s. They include:

112

Michael Clyne

(a) the establishment of a telephone interpreter service to enable people to communicate adequately with doctors, lawyers and other service professionals, with the bureaucracy and sales personnel; (b) public notices in community languages to give all people access to vital information; (c) the lifting of restrictions on broadcasting in languages other than English; (d) the establishment of a state-run multilingual radio network in Sydney and Melbourne and a public (community access) station in Melbourne within the Australian Broadcasting Commission; (e) Government subsidies to community-run after hours ethnic schools, so that language maintenance could become a national and not a private activity. They empowered many people of non-English-speaking background to participate more fully in Australian society and gave them the conviction that their bilingualism is valued and worth transmitting and maintaining. Soon after, the Federal Government gave grants to universities for the expansion of programs in community languages such as Modern Greek and various Slavic languages. Some measures were introduced at the level of states or local government areas. For instance: (a) the expansion of languages available in schools and the number of languages accredited for public examination - in Victoria including such languages as Lithuanian and Latvian; (b) the inclusion of holdings in the major community languages of the district in local libraries. Some made it possible for non-English-speaking migrants to function more easily in Australian society regardless of their proficiency level in English. In the late 1960s, adult ESL programs ceased to be based on the monolithic assumption that learners started off with 'no language' and all had the same limited needs in English. Programs for more advanced and professionally qualified migrants were started, English for Special Purposes courses developed, and eventually some classes were taught bilingually.

Empowerment through the community language

3.

113

Mainstreaming

Subsequently, a government multicultural TV network was introduced in 1980. This was within a policy of 'mainstreaming multiculturalism'. Most of the programs were films in languages other than English with English sub-titles. The intention was to demonstrate to all sections of the Australian community that good cinema is available in all languages and from all over the world. An English-language international news drawn from all over the world also introduced a more global and multicultural perspective than the news on other TV stations, where it could be concluded that anything that happened in non-English-speaking countries other than war or bombings was of little significance. Unlike the situation on radio, community languages were not allocated a specific amount of time on TV and the availability and quality of films were the main criteria for inclusion. In the 1990s, language-specific satellite news broadcasts in 16 languages were introduced on TV. They occupy the morning slots. Mainstreaming of community languages also occurred in education, with the teaching of many community languages to groups of students from a range of backgrounds. During the 1970s and early 1980s, in response to demands from multicultural lobby groups, the presence of a significant community became a reason for the teaching of their language as an 'Australian language'. This sent the signal that bilingualism and bilinguals were worth maintaining, regardless of the economic, demographic and political strength of the core nation(s) of the language. At the same time, there was a push for the introduction of languages other than English into primary schools, something that occurred first ad hoc and experimentally, and more generally in some states than in others. Because there was no recent tradition of language teaching in primary schools, and primary education was more sensitive to community needs than were secondary schools, this was an opportunity to introduce community languages which had not traditionally been taught. Thus the gulf between the non-English-speaking home and the school could be bridged in some cases, and the status of the children and the parents could be boosted. It also meant that children from both English- and non-English-speaking backgrounds got the message from the outset that school was not a monolingual domain. The position of the ethnic groups on this development varied. The Italian community groups in most states pushed strongly for the teaching of languages in primary schools and had their government subsidies for parttime ethnic schools reallocated to Italian 'insertion classes' in mainstream

114

Michael Clyne

schools, i.e. language classes run in school time by teachers appointed by the (Italian) ethnic community. In this way, Italian was taught to all students in a school, regardless of their background. Italian is now the second most widely learned language in Australian schools, after Japanese. Greek culture gives people of Greek descent strong propriety rights over their language, and the Greek Orthodox groups saw Greek programs as being mainly for their children. They were concerned about maintaining control over the teaching of Greek culture and about standards - they wanted it to be taught the way it had been taught in Greece. The Greek left saw education, including the teaching of community languages as a civic right to be available through the public education system in normal school hours through mainstream schools. They had instigated the migrant rights movement, which pushed for community languages to be available to all students. We will return to this issue in due course. At this stage, it should be pointed out that school education is a state prerogative and that each of the six states and two territories ('quasistates'), each of which has its own language policy and priorities; within these, there are differences between government, Catholic and other (independent) schools. By the 1980s, about 20 languages were taught in schools in the state of Victoria, for instance. A number of schools offered bilingual programs - some in the interests of continuity of education for recently arrived secondary students, some to provide initial literacy and numeracy skills in the stronger language (transitional in some cases), and some, building on community resources in the district, a content-based program for those with and (predominantly) without some background in the language. However, only a small number of languages can be offered in any one school. In Victoria, there is an institution within the state Education Department called the Victorian School of Languages (VSL), which runs classes on Saturdays in normal state schools for students wishing to take a language not available at their school 5 . This institution, which now prepares students for the final examination in over forty languages, has been replicated in New South Wales, South Australia, and the Northern Territory. So Farsi, Tamil, Turkish, and Ukrainian are treated in the same way as French, German or Japanese, giving recognition to any bi- or multilingualism as academically worthy. Each of the 33 centres of the VSL offers a range of languages (up to a dozen) which are relevant to the local community, and teachers are accredited by the Education department. Many of them teach in mainstream schools during the week. Like the part-time ethnic schools, they play a complementary role to the day schools but the ac-

Empowerment through the community language

115

knowledgement of language learning and maintenance as an important Australian activity is enhanced, at least marginally, by the fact that the institution is not ethnospeciflc.

4.

The national policy on languages

Multiculturalism became a bipartisan policy. Many policies in this area were initiated or expanded by the Fraser government (1975-83), conservative successor to the Whitlam government. In response to pressure from professional community groups, the Fraser Government set up a parliamentary committee to investigate the need for a National Languages Policy in the 1980s. Following hearings and submissions from the public6. Five of the 45 languages examined throughout Australia, Bengali, Czech, Latvian, Lithuanian and Slovenian, are currently under suspension because of veiy low enrolments; a report was tabled and four guiding principles were established, which became the basis of subsequent language policy statements at the federal and state levels: 1. 2. 3. 4.

Competence in English; Maintenance and development of languages other than English, both community and indigenous languages; Provision in services in languages other than English; Opportunities for learning second languages.

The National Policy on Languages (Lo Bianco 1987) established the complementarity of English and other languages used in Australia. It therefore not only gave status to the other languages but also empowered immigrants and their children both through raising their proficiency level in English and by giving them more opportunities to use, and improve their proficiency in the community language. Social, economic and cultural arguments were presented for multilingualism being a national asset in Australia. Although various domains of language use were covered, the emphasis was on education, and it was here that implementational strategies and budgetary spending were recommended and accepted by Parliament.

116

Michael Clyne

5.

Economic approaches

Soon after, economic neo-liberalism became the basis of all government policies for both major political parties, and short-term economic strategies, especially labour market training, became the rationale for language policy. This meant two things: an emphasis on improving English literacy standards, which was to undermine the emphasis on multilingualism, and the increasing differentiation between more and less important languages on the basis of short-term economic forecasts. Both of these new emphases were pushed in 1991 in a new policy document, written by the Minister of the amalgamated portfolio, Employment, Education and Training, John Dawkins and his senior advisors (Dawkins 1991). It was entitled Austrcilia 's Language: Australian Language and Literacy Policy (ALLP), somewhat inappropriately for a nation with so many languages. It emphasized short-term economic needs, especially those of the labour market, and therefore pushed English literacy and languages considered important for Australia's trade and tourism interests. Initiatives in indigenous and community language maintenance were moved more to the margins, particularly in relation to funding. The ALLP introduced the dichotomy 'language' and 'literacy' which has haunted Australian language in education policy since (See Section 8, below). The report required states to prioritize eight languages from a list of 14 for extra funding. Major community languages were included especially if they had other grounds for prioritization (e.g. Italian, Chinese and German in all states but Spanish only in two and Arabic in one) but some community languages were included in the eight, regardless of wider status (Greek in six out of eight states and Vietnamese in four). However, economic neo-liberalism also emphasized the value of multilingualism through the commodification of languages (Heller 1999). They played an important part in Prime Minister Keating's policy of Productive Diversity, which was based on the idea that Australia should utilize the linguistic and cultural diversity residing in the workforce as resources in trade, tourism, and workplace management. To the extent that this is happening it has benefited bilingual and bicultural individuals, companies, and the nation as a whole, both in status and financially. Among the success stories have been the Australian telecommunications and transport industries in Hungary, Poland, Lithuania, and the Ukraine after 1990, legal firms in Vietnam, and many areas of trade with Hong Kong, Taiwan, and India established by, or with the aid of bilingual and bicultural Australians from

Empowerment through the community language

117

the appropriate ethnic background. Included in these success stories are some languages that are not among those usually prioritized in education for economic reasons. At the same time, European forms have established their Asia-Pacific headquarters or call centres in Australia because of its resources in Asian and European languages. The status of bi- and multilinguals has been enhanced a little through the allowance which public servants receive for demonstrating proficiency in a language other than English. However, it has been found (Kipp, Clyne and Pauwels 1995) that, in the first generation, non-recognition of some overseas qualifications and levels of post-survival English proficiency and in the second, limited registers in the community language have prevented many bilinguals from benefiting from Productive Diversity. It should be noted that, though 18% of Australian residents speak a language other than English in the home, Australian chief executive officers have a lower rate of multilingualism than their counterparts in 27 other countries (Rosen et al. 2000).

6.

Australia's changing language demography

Up to the 1990s, it was generally assumed that community languages were of European origin and that Asian languages were taught and learned for external reasons. In fact, there are rapidly increasing communities of Cantonese, Vietnamese, Mandarin, and Filipino speakers, and the enrolment of students from south-east Asia at universities and the senior classes of secondary schools has become one of Australia's most lucrative industries. Nevertheless, the presence of communities using the languages in Australia did not seem to play a role in the discussion surrounding the report Asian Languages and Australia's Economic Future (Rudd 1994) which recommended that 60% of all Australian schoolchildren and 15% of those in Years 11 and 12 should take one of four 'super-prioritized' languages, Chinese (Mandarin), Indonesian, Japanese, or Korean. This formed the basis of the National Asian Languages and Studies in Australian Schools strategy which was funded from 1995 to 2002.

118

Michael Clyne

7.

Problems with 'empowerment' through community languages in education

The presence of students with a background in these Asian languages has been perceived as a problem by some of those making decisions on the evaluation of students. This appeared to rekindle the same urge as in the 1950s and 1960s to ensure that those who knew the language best would not get better scores in the language than those who were not 'advantaged'. At that time, the languages were German, Italian and Russian. Now they are especially Chinese, Japanese, Korean, and Indonesian. The first three are particularly difficult languages for English speakers to learn, so the discourse of the 'unfair advantage' was reemerging on very fertile ground. There were a number of other factors: 1.

While bilingualism was still perceived by some educationists in deficit terms, it was not something that needed to be guarded - but now that it could be seen as an advantage, particularly an economic one, it seemed important that this advantage should be available to the 'mainstream' group at least as much as to other groups! 2. A nationwide global scaling system means that students all over Australia end up with a single score for the examination at the end of secondary school which counts for entry score into tertiary institutions. The actual competence in a particular subject is obscured by statistical formulae ('scaling') which relate such a grade to that of the other candidates and of all of them in other subjects, as well as a general aptitude test which all students take. All results in some languages, such as French, German, Hebrew, Japanese as a Second Language, and to a lesser extent Chinese and Modern Greek were scaled up in 2002 because of the scores of the students in other subjects, and those in some other languages, such as Arabic, Macedonian, Turkish, and Japanese as a First Language were scaled down7. This should be seen in the context of very stiff competition for places in universities, especially some universities and some faculties. 3. One of the reasons why some students have done badly in their nonlanguage subjects is that they are recent arrivals and their English level is not so high. This is partly counterbalanced by their eligibility to sit for English as a Second Language instead of English.

Empowerment through the community language

119

Let us digress to enumerate the background categories of students in language classes (Clyne, Isaakidis, Liem and Rossi Hunt 2004; Clyne, Fernandez, Chen and Summo-O'Connell 1997): (a) Students with a home background and (some) overseas experience of formal education through the language as a medium of instruction. (b) Students with an active home background in the language and some formal instruction (primary and/or ethnic school) in Australia. (c) Students with an active home background in the language and no formal instruction prior to secondary school. (d) Students with an active home background in a variety of the language but not in the standard language, in which classes are conducted with or without formal instruction in the language here or elsewhere. (Examples are Cantonese, or the various national varieties of spoken Arabic). (e) Students with a passive home background in such a language. (f) Students with no home background in the language but formal instruction in the language at primary school. (g) Third language learners, whom we treat as a separate subgroup in our study. (h) Students with a passive family background (usually one parent or grandparent/s) and no formal instruction in the language prior to secondary school. (i) Students with a passive family background in a variety of the language and no formal instruction in the standard language prior to secondary school. (j) Students with no home background and no prior knowledge of the language. Garcia (this volume) refers to an even greater complexity of student background in her Spanish/English and Chinese/English bilingual schools in New York. Such diversity is clearly more complicated than a dichotomy and, in terms of assessment, wherever you try to draw a line between the 'advantaged' and the 'disadvantaged', you are creating discrimination. Moreover, there are many opportunities for the sharing of community languages, which ought to be utilized. We have been conducting a research project on community languages in secondary schools, supported by the Australian Research Council, in conjunction with the Victorian Department of Education and Training and the

120

Michael Clyne

Catholic Education Office. In this project, the Research Unit for Multilingualism and Cross Cultural Communication at the University of Melbourne has been working with the schools, among other things, to develop models to cater for programs with such a range of student backgrounds and to utilize community resources in language learning and teaching. The intention is to demonstrate that any second language learner is advantaged in Melbourne by living in a city with such linguistic diversity, and it is important to develop strategies so that everyone, regardless of their background, will be motivated and provided with input and output opportunities, to develop their linguistic potential to the fullest. We have been working with one school program each in Chinese (Mandarin), Greek, Spanish, and Arabic. Research staff have been developing activities in which students with different types and degrees of background are working collaboratively and utilizing the rich community resources in the local shopping centre and the wider community. These include visits to and conversations in a day care centre for elderly Spanish speakers, the creation and editing of a Spanishlanguage electronic newsletter with school and community news, visits of community members interacting in Standard Arabic about their work, and transactions and interactions in local shopping centres in Greek and Mandarin. Another part of the project involves the production of resources to help speakers of Cantonese or of a Lebanese variety to learn (and to differentiate) Mandarin and Standard Arabic respectively. Students with and without a background in the target language are affected by two types of publicity when it comes to studying languages at senior secondary levels: (a) News of discrimination against students of particular backgrounds. This can have the effect of discouraging them from taking the subject, demotivating them from maintaining their community language, and in turn diminishing the opportunities for the empowerment of young bilinguals. (b) Assumptions that students without a background cannot succeed in the language. This has had the effect of discouraging them from selecting this language for study or demotivating them from proceeding beyond the period when a language other than English is compulsory. It would appear that the removal of certain students would create a more egalitarian situation with a more 'level playing field' and scope for hardworking monolingual Australians to not be discriminated against. This is

Empowerment through the community language

121

typical of the myth of egalitarianism in Australia. In fact, a number of Establishment (elite, high fee charging) independent schools have been lobbying the Victorian Curriculum and Assessment Authority and through the media8 ensure that students such as theirs from Anglo-Australian backgrounds can do well in Chinese. Such schools are well-known for their outstanding university entrance results. They do not see empowering bilingual children particularly in their interests, although they are attended by many children of Chinese background. The problem, however, lies in the 'global scaling' notion and the competition for university places. The position of the independent schools has also been supported both individually and at times collectively through the Chinese Teachers Association by Chinese teachers from government schools, who believe that unless students from Chinese backgrounds are discriminated against, other students will fear their competition and not opt for the subject in the higher secondary years. It is interesting to note that independent schools have the highest percentage of students taking Chinese in Victoria, and most of the students of non-Chinese background with outstanding results in Chinese attend these schools. It is also paradoxical that many Chinese teachers themselves have problems teaching students without a home background in Mandarin. They tend to speak disparagingly about 'inferior' students and cherish the change of Chinese from compulsory subject to elective in Year 9 to avoid teaching such students. Many of them experience discipline problems with students who do not share the Chinese ethos of learning and respect for the teacher. This problem is likely to be overcome by the employment of second generation Chinese-Australians as Chinese teachers so that their motivation to maintain the language and to stay with the subject is very important. It is true that there is a perception among non-home background students that Chinese is too difficult for a person from another background to do well and there is a similar perception for Greek, Arabic and, to a lesser extent, Spanish because of competition from 'background students'. On the other hand, other studies (Clyne, Fernandez, Chen and SummoO'Connell (1997: 122); Clyne, Jenkins, Chen, Tsokalidou and Wallner 1995) demonstrate that friends from Chinese backgrounds have actually motivated a large percentage of other children to learn the language and have provided them with input and output opportunities in it. In a parental focus group meeting in our most recent project, parents from Latin American backgrounds (Clyne, Isaakidis, Liem and Rossi Hunt 2004) indicated that their children's language maintenance has been motivated by the en-

122

Michael Clyne

thusiasm with their friends from non-Spanish-speaking families for learning their language. The 1996 Census showed a very high intergenerational loss between the first and second generations originating from the PRC, Hong Kong or Taiwan, far more so than between the two generations in families originating in south-eastern Europe or the Middle East (Clyne and Kipp 1997). Chinese families from Hong Kong and Taiwan studied for an in-depth study of language use and maintenance (Clyne and Kipp 2002) indicated a strongly pragmatic attitude towards language, favouring English, and Mandarin over Cantonese or Taiwanese, for their children on instrumental grounds. Examination authorities see the solution in two or three different examinations in language subjects, with entiy requirements determined on the basis of 'background'. Some earlier stages of this saga have been discussed in Clyne, Fernandez, Chen and Summo-O'Connell (1997) and Clyne (1997). At present, in Victoria, a separate examination for students with a particular type of background takes place only in four Asian languages, those especially prioritized by the NALSAS Strategy (see above, 6). These are the languages with which international students are most likely to come into the senior years of secondary education. However, identification of this group is based on a points system, in which the number of years of education through the medium of the target language becomes the basis on which students can enrol for one of the other level. The only language with three examinations is Chinese. Like Indonesian, Japanese and Korean, Chinese has the differentiation between LI and L2 examinations based on separate syllabuses. But unlike these other languages, Chinese has both an L2 and an L2 (Advanced) examination. The latter is not based on a separate syllabus but simply taken by students categorized according to their background, not according to their self-perceived proficiency level as one might expect from the name 'Advanced'. The criterion for enrolment in 'Advanced' is periods of education through the medium of target language from 6 months to 7 years. New South Wales has 'Background Speakers' examinations in Chinese, Indonesian, Japanese, Korean, Malay, Persian, and Russian, while in South Australia, there are three levels for learning languages - Beginners, for those starting the language at senior secondary level, Continuers, for those who began the study in Year 8, and Background Speakers, for students with "substantial linguistics and cultural background in the language". 9

Empowerment through the community language 8.

123

More general impediments

In the course of the project, we have become aware of a number of impediments to community language programs contributing to the empowering of bilinguals in Australia. They are mainly connected with monolithic and consequently monolingual thinking. One is the dichotomy 'language' and 'literacy' which has been part of the discourse on education in Australia since Dawkins's Language and Literacy Policy. The implicit and underlying assumption is that for a person to be literate (in English), they must have acquired literacy in English, and in English only. Part of the argument runs that people have to be more literate today than ever before but that literacy standards are falling. This means that more time needs to be spent on English literacy rather than other languages, which do not contribute to (English) literacy. A child having problems with English literacy is often withdrawn from a second language program, even where the language is their home language. Insights on the interdependence of languages and on literacy transfer are not understood or believed by the average school principal or by other school administrators (Cummins 1998). Another figure of speech is 'overcrowded curriculum', which implies that demands on time on the curriculum such as 'literacy' in the lower primary years, computer studies, consumer studies, and popular electives such as self-defence, photography and cookery delegitimates the time required by 'languages other than English' as a key learning area' and the consequent status as a compulsory subject from the preparatory year to Year 10, leading to some schools making it an elective beyond Year 8 or not introducing it until middle primary school. The shortage of appropriately trained teachers for many languages, unnecessarily traditional teaching methods in some languages (such as Arabic and Modern Greek), the inavailability of materials for some languages, especially for those from bilingual backgrounds, and the inability of some programs to address the needs of a diverse student population are further impediments. The return of the monolingual mindset as the dominant way of thinking in Australia leaves much less scope for empowerment of immigrants and their families through status change of their language. We need to wait for a new change of attitude similar to what occurred in Australia in the 1970s and 1980s. However, other countries with less of an anti-intellectual tradition may be better able to implement this due to belief in globalization, and

124

Michael Clyne

the exigencies of international trade and collective security while we keep experimenting with the sharing of languages.

9.

Some concluding remarks

There are clearly ways in which bilinguals can be, and have been empowered through the raising of the status of their language, as either a symbol or a commodity. Through the commodification of languages, bilinguals can be seen to contribute an asset to the nation. However, empowerment means changing power structures even though what may be intended is language sharing and facilitating widespread multilingualism. When it comes to the crunch, at most times in history, those in power will not share it. Thus, better ways of spreading multingualism and a better understanding of language throughout the community are needed. In 1988, Joshua Fishman, a keen observer of the Australian language situation, wrote (1988: 137): We (the Americans) are a long, long way from a positive language policy such as the Australians have just adopted. By 1991, in his study of reversing language shift, he did not see Australian policy implementation delivering successful results (Fishman 1991: 277): Australian policies and practices constitute a positive but ineffective approach to reversing languages shift on behalf of recent immigrant languages... Ten years later, he was quite negative (Fishman 2001: 479): The comfort that minority languages could take a decade ago from Australian language policy...has largely vanished in the interim. Instead of Australia being the forerunner of a sea change in the valuation of aboriginal or immigrant languages on the part of English-dominant host-countries, the direction of influence has run in the opposite direction. The traditional American negativeness towards such languages has crossed the Pacific. While Fishman has correctly identified a retrograde development, I am inclined to think that Australia is perfectly capable of retreating from a strong multilingual policy without American influence, for monolingualism has been the dominant policy in Australia for most of its colonized history. However, no state of Australia has declared English its official language; nor is there any political movement in favour of such an enactment.

Empowerment through the community language

125

As I have indicated earlier in this paper, there are many innovations in education, the media, and other areas of public life where multilingualism remains well established and is empowering Australians of non-Englishspeaking background. They include the over forty languages examined as subjects in the end-of-secondary-school examination, radio programs in over eighty languages on multilingual radio stations, a government-run multicultural TV station with English sub-titled films in other languages, a telephone interpreter service functioning in 190 languages, and public libraries with holdings in the main community languages of the district. The issue of what can happen when multilingualism is mainstreamed, both positively and negatively, may be of interest to people in other countries with linguistically and culturally diverse populations so that collaborative solutions may be found to the benefit of bilinguals and biculturals and the entire nation.

Notes 1. I am grateful to Sue Fernandez for helpful suggestions, to Maree Dellora for useful information, and to the Australian Research Council for supporting the project, Community Languages in Secondary Schools, conducted by the University of Melbourne in partnership with the Victorian Department of Education and Training and the Catholic Education Office. 2. The increase shown for Indonesian is for 1996-2001 only, due to the fact that the 1991 Census coded the closely related languages Indonesian/Malay as one entry. If we add the Malay figures to the Indonesian ones for 2001 for purposes of comparison, then the percentage increase from 1991-2001 is 61.6%. 3. In Australian English, 'migrant' is synonymous with 'immigrant'. 4. Gl = 1st generation. 5. Five of the 45 languages examined throughout Australia, Bengali, Czech, Latvian, Lithuanian and Slovenian, are currently under suspension because of very low enrolments. 6. During the life of the committee, the government changed and the next government under Hawke (Labor) continued the inquiry. 7. Personal communication, Maree Dellora. 8. Yvette Slaughter is studying the lobbying as part of her thesis on policies towards Asian languages in Australia. 9. Personal communication, Maree Dellora.

126

Michael Clyne

References Clyne, Μ. 1967

Transference and Triggering. Mouton: Martinus Nijhoff.

Clyne, M., S. Fernandez, I. Chen, and R. Summo-O'Connell 1997 Background Speakers. Melbourne: Language Australia. Clyne, Μ., T. Isaakidis, I. Liem, and C. Rossi Hunt 2004 Developing and sharing community language resources through secondary school programs. International Journal of Bilingual Education andBilingualism 7: 255—278. Clyne, M., C. Jenkins, I. Chen, R. Tsokalidou, and T. Wallner 1995 Developing Second Language from Primary School. Canberra: Language Australia. Clyne, M. and S. Kipp 1997 Trends and changes in home language use and shift in Australia. Journal of Multilingual and Multicultural Development 18: 154— 73. 2002 Cummins, J. 1998 Dawkins, J. 1991 Fishman, J.A. 1988

Australia's changing community language demography. People and Place 10: 29-35. Language, Power and Pedagogy. Clevedon: Multilingual Matters. Australia's Language: The Australian Language and Literacy Policy. Canberra: Australian Government Publishing Service. 'English only' - its ghosts, myths and dangers. International Journal of the Sociology of Language 74: 125—40.

1991

Reversing Language Shift. Clevedon: Multilingual Matters.

2001

From theory to practice (and vice versa). In Can a Threatened Language be Saved? J.A. Fishman (ed.), 451-483. Clevedon: Multilingual Matters.

Garcia, O. this volume

Lost in Transculturation: The Case of Bilingual Education in New York City.

Empowerment through the community language

Heller, M. 1999

Linguistic Minorities and Modernity: A Sociolinguistic graphy. London: Longman.

127

Ethno-

Kipp, S., M. Clyne, and A. Pauwels 1995 Immigration and Australia's Language Resources. Canberra: Australian Government Publishing Service. Lo Bianco, J. 1987

National Policy on Languages. Canberra: Australian Government Publishing Service.

Rosen, R., P. Digh, P. Singer, and C. Phillips 2000 Global Literacy: Lessons on Business Leadership and National Cultures. London: Simon and Schuster. Rudd, K. 1994

Asian Languages and Australia's Economic Future. Canberra: Australian Government Publishing Service.

Pidgins and Creoles between endangerment and empowerment: A dynamic view of empowerment in the growth and the decline of contact languages, especially in the Pacific Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

1.

Introduction

During the 20th century, there has been a long tradition of neglecting the study of Pidgins and Creoles in linguistics and it has only been during the last few decades that publications in the theoretical domain have begun to fill the gap. Typically, Pidgins develop in the context of trade, temporary employment and similar non-permanent joint ventures. Where contacts become permanent, as when African slaves were permanently settled on the plantations of the New World, the Pidgins creolize, i.e. they extend their functional range and acquire structural complexity in the process of becoming the first language of the new community of speakers. The social context in which Pidgins and Creoles developed was typically one of asymmetric power relations, racism and colonialism and the memory of these circumstances has influenced speakers, planners and professional linguists alike. Errington (2001: 19ff) comments on an important aspect of colonial linguistics: "Colonial regimes created conditions that engendered Creole languages, but which linguists largely ignored"; this reflects the view held by both colonizers and the majority of linguists (with the exception of figures such as Hugo Schuchardt) that Creole languages were degenerate. A small number of Pidgins and Creoles were described when they were perceived as useful for containing or controlling the 'natives'. Thus a West African Pidgin English grammar and phrasebook was compiled for use as a language in the German colonial army (Hägen 1908) and similar materials also were produced for Swahili in German East Africa and Tok Pisin in German New Guinea. For Pidgins and Creoles that were used as missionary languages (e.g. Negerhollands, Sranan and again Tok Pisin) similar structural and lexical accounts were compiled. Some of these colonial Cre-

130

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

oles became means of empowerment in the struggle for political independence (eg. Bislama, Swahili, Seychellois, and Hiri Motu) but the majority of these languages have remained very poorly documented. Speakers of most surviving Pidgins and Creoles continue to restrict the use of these languages to private and powerless domains. Language planners typically do not seriously raise the status and/or promote the lexical ausbau of these languages and the discourses of linguists about the endangerment of the world's languages unsurprisingly make little mention of Pidgins and Creoles. Historically, language policy making and planning for these languages was concerned with meeting the aims of external colonizing powers to: expedite social control minimize administrative costs replace indigenous languages and cultures in order to establish a more efficient communication system. As with environmental biology and other areas of ecological studies, it is dangerous for linguists to focus on charismatic 'pure', 'authentic' species only, because they tend to overlook an important part of the linguistic systems in use worldwide. The reason for many contact-induced languages remaining under-documented is that they are not considered as 'pure' languages, (i.e. they are often designated "non-standard") as well as their association with undesired social processes such as slavery and disintegration of traditional societies. Generally, the speakers of Pidgins and Creole languages have neither an attentive nor a positive attitude towards their languages and, having internalised what Keesing (1990: 149ff) refers to as the 'colonial ideology', they often do not even consider them as autonomous linguistic systems. Australian Aboriginal people speaking mixed Koines or Creoles often refer to them as 'shit languages'; the first contact varieties between Melanesian and French in Southern New Caledonia were called 'kapone' with the same signification. Few users of Pidgins and Creoles argue for their linguistic rights. Furthermore, contact-induced varieties have in several instances been identified as the principal cause of the disappearance of traditional languages. For instance researchers working in Papua New Guinea (Mosel 2002; Wurm 2001: 286-287) have at times attributed the disappearance of smaller local languages to the spread of Tok Pisin. The negative impact of

Pidgins and Creoles between endangerment and empowerment

131

these languages is only one part of the picture, however. As observed by Drechsel (1997), Pidgins and Creoles can also fulfil an important role in maintaining diversity in complex language ecology. A long tradition in the linguistic mainstream and the linguistic discourse has been inclined to underrate the importance of contact-induced varieties of language in this function. Sociolinguists and language planners have tended to regard Pidgins and Creoles as undeserving or incapable of becoming instruments of political and social empowerment. The Hawai'ian language, for instance, was reinstated as an official language some years ago and this recognition was accompanied by significant funds being provided for its use in education and the public domain. At the same time, Hawai'ian Pidgin English (Da Kine) spoken by a far greater number of residents, continues to be marginalized and denigrated. The policy of the Republic of Vanuatu again illustrates the powerlessness of Pidgins and Creoles. Crowley (1990:1) comments on the fact that whilst Bislama was declared the national language of Vanuatu, only the official languages French and English were permitted to be used in the education system. Officially, the use of Bislama is still forbidden in the school but, as Tryon and Charpentier (2004: 453-454) have commented: The twenty years which followed independence have been marked by inaction. This inability to change and plan has been very beneficial to Bislama, which looks very much as if it will become the default language, after the double failure of attempts to set up a French/English bilingual system and a system of vernacular education. In this paper we will present data from two endangered English-related Creoloids from the South Pacific: Norfolk and Palmerston. Both languages'came into being through acts of linguistic power, and indeed replaced the Polynesian languages of the female members of these two communities, but over time they developed into languages of identity and, in the case of Norfolk, social empowerment. 2 They illustrate that in addition to the risks threatening all languages with small communities of speakers it is, paradoxically, their relation to English that poses an additional threat to their survival. Owing to their many superficial resemblances to their socially privileged lexifiers, such languages must not only be described as abstract, decontextualised structural systems but also from the perspective of their use and functions. They can be seen as typical strategies of using the community's linguistic resources for specific aims in specific lan-

132

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

guages: this observation is obviously transferable to the study of many other endangered languages in similar situations. From the earliest days of contact between Europeans and Pacific islanders, small mixed 'beach communities' were set up, often Utopian and of very limited duration. The best-known of these communities is the community of British sailors and their Tahitian consorts who, subsequent to the mutiny on the Bounty, settled on unoccupied Pitcaira Island, where a new contact language (called Pitcairaese, Pitcairn English or Pitkern) developed. This community and their language were subsequently relocated to Norfolk Island, where the language is referred to as Norfolkese, Norfolk English, Norfolk or Norf k. Much less well known is the story of a second English lexifier mixed language that developed on Palmerston Island in the mid 19th century under slightly more peaceful conditions. The initial development of both languages was a traumatic intercultural encounter which led to the loss of much of the language and culture of the Polynesian women.3 This development can be characterized as 'creative destruction' as it promoted a range of dynamic linguistic processes and the development of new support systems for new identities. The fact that mixed languages rather than acrolectal English initially became the medium of communication on both islands would seem to illustrate the role of Pidgins and Creoles4 as buffers that attenuate the shock resulting from cultural clashes. Because we are dealing with a shallow time depth and small communities, these processes are more visible than in other language contact and conflict situations.

2.

The Pitkern NorPk languages

2.1.

General description of Pitcairn and Norfolk Island

Pitcairn Island is situated in an isolated part of the Central South Pacific Ocean (24° 01S χ 130° 06W), its distance from New Zealand from where it is administered being greater than that between Sweden and India. Its landmass is less than five square kilometres and its present population around 50, with a possibility that it will be abandoned. There is an ongoing conflict between Britain and the Pitcairn Islanders following the decision of the British government to prosecute a significant proportion of the adult male islanders for sexual misconduct. The ongoing court case has been interpreted by many Pitcairners as an attempt to close down the island and

Pidgins and Creoles between endangerment and empowerment

133

for the first time in their history the islanders have argued that they are not British subjects. Norfolk Island is located 1,575 kilometres east of Australia in the South Pacific Ocean (24° 05S χ 167° 59E). It occupies an area of about 40 square kilometres and has a permanent population of about 2,600. It is visited by about 35,000 tourists per annum, with projected numbers reaching 50,000 in the near future. The political status of the island has been a matter of dispute between the Federal Government of Australia, the State of New South Wales and the descendants of the Pitcairn Islanders (O'Collins 2002). Pitcairn descendants today comprise fewer than half of the permanent residents of the island. There are moves by some islanders to take Australia to the International Court of Justice for committing 'cultural genocide', their arguments including the denigration and destruction of the Norfolk language. Norfolk Island was discovered by Captain Cook in 1779, and because of its ample natural resources and isolated position, was made a British penal colony in 1788. The first penal settlement was abandoned in 1814, but a second penal settlement was built in 1825 at a location for the 'extremist punishment short of death' (Hoare 1982: 35) and 'a cesspool of sodomy, massacre and exploitation' (Christian 1982: 12). Following much criticism, the settlement was closed down in 1854. This is where the story of the Norf k language begins. Rather than abandon the island the British government decided on what was referred to as 'the experiment' - to settle a small community of simple god-fearing English-speaking people on an isolated island and watch their moral progress. To this purpose, in 1856 the entire population of Pitcairn Island was relocated to Norfolk Island. The 'guinea pigs' in this experiment were the Pitcairn Islanders, whose origins we shall briefly comment on. The story of the mutiny on the Bounty has been popularised by numerous novels, plays and films, and Pitcairn Island, where the Bounty mutineers settled in 1790, has come to stand as a metaphor for a South Sea Utopia. When nine British sailors, twelve Tahitian and Tubuaian women and six Tahitian men arrived on Pitcairn, the island was uninhabited. By 1800, following a period of violence, the Englishman John Adams was the sole male survivor with 10 Tahitian women and 23 children. When he died in 1829 the island had become a model Christian community of about 80. As John Adams approached the end of his life he realized that the maintenance of Christian values and the English language required outside help and in the following years three male British subjects settled on the island

134

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

and married local women. From the mid 1820s all children were taught English literacy by native English speakers and Tahitian was not encouraged and within a generation died out on Pitcairn. Earliest references to an English-Tahitian contact language date to 1789 when the British sailors, to taunt their captain, deliberately mixed Tahitian words into their language. On Pitcairn, the Polynesians communicated with the British mutineers in a Pidgin exhibiting a mixture between Tahitian, West Indian Creole and English. Ross and Moverley (1964) characterise what they called Pitcairnese as the outcome of language mixing, and provide numerous details about Tahitian lexicon and grammar, as well as details on dialect features. They provide details on the provenance and likely dialect affiliation of the mutineers (1964: 49/137). The imperfect knowledge of Tahitian among the first generation children born on Pitcairn is suggestive of the low esteem in which Tahitian culture and language were held by the mutineers. Tahitians were excluded from land ownership. In spite of very unfavourable demographic conditions (by 1800 there were 10 Tahitian women, 23 mixed-race children and one Englishman), English remained the dominant language and both English and Christianity were taught after 1793 (Maude 1964: 50). The dominant linguistic socializers were British males, in particular: Edward Young, the story teller, who contributed a number of St. Kitts pronunciations and lexemes, [1] for [r] in words such as stole, 'story' or klai 'cry'; and morga 'thin'. John Adams, the patriarch, who created the social conditions in which standard acrolectal English, against all demographic odds, could prevail as the dominant language of the community. As outsiders began to visit the island after 1808, Europeanization accelerated and Tahitian ways of clothing, food preparation and food collection declined. Maude (1964: 51-57) comments on the community's identification with the European side of their heritage and notes several examples of racial friction. Though Tahitian disappeared in the 1830s, the mixed contact vernacular continued to be used side by side with English in most families. It is not clear to what extent the original white settlers used it and for the first generation women it was probably a convenient means of intercommunication. For the first children born on Pitcairn it became the informal way of speaking. However, ever since the first outsiders settled on Pitcairn in the 1820s

Pidgins and Creoles between endangerment and empowerment

135

and took control of the domains of education and religion, the language was restricted to non-official domains and functions and occupied the low position in a stable diglossic situation, where English was the superordinate language and where an increasing number of outside settlers did not even have to learn Pitkera5. According to Laycock (1989: 622), the revaluation of the Pitkern language occurred in 1831 when, because of food and water shortages, the entire community attempted to resettle in Tahiti. This disastrous experiment ended after five months. Infectious diseases ravaged the Pitcairners and they were disgusted with the low moral state of their hosts as well as some of the British sailors they met. The Pitkern language thus became a symbol of non-identity with outsiders as well as a positive marker of a separate community, though it certainly continued to be subordinate to English. In 1839 the population had grown to 100, and by 1850 it had reached 156. In 1853, as fish became scarce and the island degraded, the inhabitants solicited the aid of the British Government to transfer them to another island. In 1856 all 194 Pitcairn Islanders were relocated to Norfolk, but a number of families returned to Pitcairn shortly afterwards to be joined there by various newcomers over the years. 2.2.

Notes on the language

The ambivalent attitudes towards Tahitian language and culture are reflected in several areas of language mixing. First, it is remarkable that words of Tahitian origin tend to be predominant in marked domains of language: taboo words, negative characterisations, undesirable and unnatural phenomena and properties. Examples (in unstandardised spelling) include: eeyulla

'adolescent, immature, wet behind the ears'

gari

'accumulation of dirt, dust, grime, grease, etc.'

hoopaye

'mucous secreted in the nose'

howa-howa

'to soil one's pants from a bowel movement, have diarrhoea'hullo 1) 'a person of no consequence', 2) 'having nothing of

136

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

any value; dirty, poor' iti

'any of the wasting diseases, but mainly referring to tuberculosis'

iwi

'stunted, undersized'

loosah

'menses, menstruation'

maioe

'given to whimpering or crying a lot, like a child, but not necessarily a child'

nanu

'jealous'

pontoo

'unkempt, scruffy'

po-o

'barren or unfertile soil'

tarpou

'stainsfruits on the caused from peeling some andhands vegetables'

tinai

(1) 'to gaze at with envy', (2) 'an avaricious person'

toohi

'to curse, blaspheme, or swear'

uuaa

'sitting ungraciously'

uma-oola

'awkward, ungainly, clumsy'

The dominance of the British males can be seen in the way the new environment was named: Thus, no Tahitian placenames are found on Pitcairn. However, a surprisingly large proportion of Pitcairn localities and its fauna and flora are named after English speaking individuals: Adam's Rock

'John Adams' fishing site'

Down-under-Johnny-fall

'the cliff, from which John Mills fell in 1814'

Ed's fence Austin grass

'an enclosure belonging to Edward Young'

Big Jack

'Hilo grass' (introduced by Austin Young) 'Indian mallow' (probably named after Jack Evans who introduced it as goat feed)

Hilda flower

'tiger lily' (introduced by Hilda Young)

Pidgins and Creoles between endangerment and empowerment

137

Hattie tree

'pink orchid tree' (after the missionary Hattie Andre)

Izzie Fish

named after Isobel Coffin

Frederick

a kind of fish

George

a small pointed sea shell

Whereas the lexicon is suggestive of the wish of the Pitcairners to distance them from their Tahitian heritage (something also manifested in their switch to European dress, food, pastimes and religion), there are other linguistic features suggesting close integration. For instance, words of English, Tahitian and other provenance do not differ, as they do in most contact languages, in their susceptibility to morphosyntactic rules, suggesting a full integration of the two languages. This corresponds to the definition of a mixed language and similar integration of two languages is in evidence in Pidgins that arose in egalitarian circumstances such as the numerous village-to-village trade Pidgins of pre-colonial Papua New Guinea. (a) progressive marker -en Yu tuhien

'You are swearing'

Mais aanti kuken fkresmes

'My aunt Christmas'

is

cooking

for

(b) stages of comparison agli - aglia - aglies

'ugly - uglier - ugliest'

pill - pilia - pilies

'sticky - stickier - stickiest'

meyameya - meyameyara meyameyares

'withered - more withered most withered'

morga - morgara - morgares

'thin - thinner - thinnest'

The Norf k language, as spoken today, differs in many ways from the records of the early 19th century contact language and indeed from contemporary Pitkern. It reflects contacts with other languages such as the Melanesian Pidgin English spoken in the Melanesian Mission community on

138

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

Norfolk Island between 1867 and 1920, differences in the colonial power, a different sense of identity and a great deal of internal lexical and grammatical development. The attitudes of Norfolk Islanders towards the variety spoken on Pitcairn remain variable and ambivalent. Some contemporary Norfolk Islanders believe that it is the same language, others that the two varieties are "as different as chalk and cheese". Some hold the romantic view of Pitkern as their true ancestral language; others regard it as crude and degenerate. Mention should also be made of the fact that, in spite of attempts to define a standard variety of Norfolk (Buffett 1999), it remains a somewhat unfocussed language, with speakers from different families varying in pronunciation, grammar and views about its separateness from English, a phenomenon reminiscent of the situation in colonial Belize (former British Honduras) described by Le Page and Tabouret-Keller (1985). 2.3.

Social history of the N o r f k language

Life on Norfolk for Pitcairners in the first couple of decades after 1856 underwent little changes. If anything, the isolation from the outside world was even greater and the spiritual and educational well-being of the Pitcairners remained in the hands of the reverend George Hunn Nobbs, helped mostly by members of his family. In 1859 an outsider from Hertfordshire, England, Thomas Rossiter, was appointed by the Australian government as a schoolmaster with the aim of improving "the tone of the children" (Mercer 1987: 4). The intention of the government was reflected by Governor Young in 1862 (Mercer 1987: 5): ... it is indispensable to continue Mr. Rossiter's service for years to come ... Upon the school must be placed the main dependence from preventing these interesting colonists from relapsing into the listlessness which the climate and abundance with which they are surrounded are so apt to superinduce; without it there might ensue a complete forgetfiilness of the habits and pursuits of civilised life. One notes, that 'the experiment' required constant intervention by the experimenters. Strict control of Norfolk education was somewhat relaxed when two consecutive headmasters of islander background were employed between 1884 and 1906 but as early as 1897 an expert from the Australian mainland had recommended the appointment of a headmaster from New South Wales, which eventuated in 1906 and which has become the common procedure followed to date. As the teachers came from mainland Aus-

Pidgins and Creoles between endangerment and empowerment

139

tralia and did not speak N o r f k, they showed far less sympathy for the language than earlier members of staff and over the following years the education system became the principal means of assimilating the community to mainstream Australian English ways of speaking. The annual reports by headmasters and New South Wales school inspectors illustrate clearly the persecution of the N o r f k language. 6 In 1912 Mr. Ray inspected the Norfolk Island Public School and in his report to the New South Wales Education Department made extensive comments on the problems of teaching English: English: Much difficulty is experienced in teaching English. The chief obstacle is the prevalence of the Norfolk Island "language" outside school. This jargon, which is the everyday medium of conversation, between the islanders and adults as well as children is in no respect a language, it is not even a 'patois'. It is said to be a mixture of English and Tahitian. As a matter of fact bad English, spoken by the Bounty men and imperfectly imitated by the Tahitians. The usual greeting is 'whatawayou?' derived from 'In what way are you', or as we would say, 'How are you' - The reply is frequently 'Wallthankyer' 'Well thank you'. The word for "people" is "sallen" which comes from 'sillen' and the Tahitian women's way of saying 'children'. In like manner 'all of you' is corrupted into 'yorlya'. Again, the islanders rarely use the plural of nouns or the past tense of verbs when trying to speak English. Ί send him two case orange yesterday' is typical. This jargon is so habitual in the homes (most of them) and elsewhere that children at school are painfully slow when they try to speak good English. When a question is put or a suggestion is made to them it is plain their first impulse is to answer in their jargon. They then have to translate their ideas into English which they deliver in a jerky stilted manner. In written composition, too, they lapse into the common faults found in their speech. A lad who was a candidate for the Pacific Cable Service failed in English composition. He passed in all other subjects. During an address to parents on the 13th instant, I pointed out that the continuance of the use of the Norfolk Island language was likely to hinder the progress of their children and seriously hamper them when in competition with others in afterlife. I advocated the suppression of the language in their homes. The people appeared sympathetic, but it remains to be seen whether the reform will be carried out.

140

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

At any rate the use of 'Norfolk Island' in or about the school grounds should be prohibited.

From 1914 until about 1930 headmaster's report after headmaster's report makes similar critical comments about the Norfolk language. A. A. Matthews who was headmaster in 1914 was no exception: The different classes were examined at the end of the quarter and fairly satisfactory results were obtained. I must admit the results are far below the standard reached by New South Wales youngsters, but still there has been all round improvement. The teachers here work at a great disadvantage. The little ones come to school scarcely able to talk or understand any English. The Norfolk Island 'jargon' is almost exclusively spoken in their homes, and I have known cases, where children have been ridiculed by the Islanders for speaking proper English. The Islanders are proud of their 'language' as they call it, and in some cases openly tell you it is useless for the teachers to try to get rid of it. I feel sure, however, that with steady insistence against its use at school and careful teaching it will ultimately disappear. The good work done by the upper classes in English warrants my belief.

Passmore, who was appointed headmaster in 1915, entertained in his unpublished diary (probably written in 1916) a rather negative view of the language matter: The dialect is of comparative recent growth. Very little of it came from Tahiti. Most of the words are corruptions of English. 'Sullen' for 'children' and 'larn' 'to tell'. 'Larn a little sullen no do da', 'Tell the little children not to do that'. If you pretend not to understand the lingo as most English people do in self-defence you will hear one say sneeringly 'He's agamonin he car was it!!' 'Car' is the negative of 'to do' and 'to know'. It means Ί cannot' or Ί do not know' and is the same for all persons. The parent who hears his child speak in correct English in his own home will cry angrily, "Oh! You wawaha" and the child will shrink as under a curse. 'Wawaha' is one of the Tahitian words and means 'proud' or 'traitor' ... The early people came under the influence of an English teacher sent out by the Home Government but when he died his place was taken by one of his pupils. I knew him well, and an excellent man he was but he allowed the dialect to be used in the school to the almost total exclusion of English, hence, I think the present situation. As a concession to Island prejudice one of the members of the staff of the school is a Norfolker and this naturally helps to foster the lingo

and he continues:

Pidgins and Creoles between endangerment and empowerment

141

I love these people but I hate their language because of its limitations and for another reason in which my opinion has been lately corroborated by Professor Adams. He says that peoples' language determines their moral character. If the Norfolk dialect could be wiped out I am convinced that there would be a moral uplift.

This is not the place to present the full set of derogatory remarks on the Norf k language, nor the strategies employed by different headmasters and teachers to eradicate the language. Corporal punishment was very common to discourage the use of Norf k as indeed it was, and continues to be, in other instances of standard language imposition. By 1930, these efforts began to bear fruit and the first source mentioning that Norfolk was becoming an endangered language appeared in the Pacific Islands Monthly of 26 August 1932 (p. 11). Its writer mentions several of the reasons for this endangerment which are presented below. One of the reasons given is the increasing number of outsiders (referred to as mainlanders on Norfolk) who come to settle. The latest available statistics for Norfolk Island (census for 2001) indicate that Islanders of Pitcairn descent are not only a minority of the population of the island, comprising 1,574 permanent residents plus about 500 tourists and 500 TEPs (temporary entry persons) at any one time, but have also become a minority among the permanent residents (about 47%). No reliable statistics are available for residents of mixed Pitcairner-mainlander descent but informed observers confirm that marriages within the Pitcairner community are becoming rare. The language in mixed households is almost always English. A second reason given is 'the passing of the eldest generation', who were social and linguistic role models. The absence of Norf k-speaking role models continues to be a major problem. The majority of economically successful islanders are monolingual English speakers; the overwhelming majority of teachers are recruited from the Australian mainland, as are the clergy. The global youth culture is increasingly embraced by the younger generation, though there are a small number of locally produced songs and an annual song competition. Some of these are in the Norf k language. The author also comments that education policies have had the intended effect of making the islanders ashamed of their language (Anon 1932: 11): As it is, for some mistaken reason, they seem ashamed to live as their fathers and mothers did and to speak the tongue that is a thousand times superior to the ugly English they learn in the State school.

142

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

The eradication of the N o r f k language was no longer official education policy after 1930 but as Harrison (1972: 22) remarked: ... one gathers that teachers at the local school who are usually mainlanders appointed by the New South Wales government, think that their pupils' regular use of Norfolk outside the classroom is a drawback to their progress. Many parents, also, being anxious for their children to receive a sound education, feel that if they naturally talk and think in Norfolk instead of English their schoolwork is likely to suffer. The result of all of these factors was a further decline in use and status of N o r f k and by the end of the 1960s, English became the dominant language for most children outside the classroom and this has continued today. The decline of N o r f k illustrates how deliberate linguicide combined with neglect and linguistic and social assimilation has lead to the decrease of power of an already weak language. The revival of the N o r f k language in recent years is due to a range of factors, some of them bearing close similarity to the situation on Martha's Vineyard (Labov 1972). As more and more Norfolk islanders went to study or work on the Australian mainland, and as the number of tourists and temporary residents from Australia increased, a new pride and sense of identity developed, combined often with a feeling of antagonism against Australian colonial policies. Language occupies a central place in this attitude. Two publications that appeared in the late 1980s (Nobbs Palmer 1986; Buffet and Laycock 1988) not only documented the language but suggested ways of writing it down in order to preserve it and to increase its use. Community members began to give lessons in N o r f k at the Central School and the status of the language has been enhanced by a number of means: -

its use in the Assembly and for official government functions its mainstream position in the education system its use in song writing, poetry reading and other cultural events meetings of community members to discuss questions of language revival.

On December 15th the Norfolk Assembly passed the Norfolk Island Language Bill which declares N o r f k to be an official language on Norfolk Island. It is far too early to ascertain whether this legislation and the above mentioned activities will lead to a further revival of the language. Supportive legislation, positive feelings and the sense of identity in themselves are

Pidgins and Creoles between endangerment and empowerment

143

unlikely to reverse the decline of Norf k and the availability of materials again may help preserve but not revive the language. Moreover, the recent positive attitudes towards Norf k are fragile. In the wake of two much publicised murder cases on the island, the language has been labelled a "'dialect for murder' (The Australian, 2 June 2004, p. 3) and the recent investigation into sexual misconduct on Pitcairn Island has prompted the reporter for the Adelaide Advertiser (23 September 2004, p. 21) to refer to the 'blunt pidgin language of Pitcairn Island' and to illustrate its bluntness with the observation that "7ow 'sa dirty dawg is a common insult." Such stereotyping highlights the continuing powerlessness of Norf k vis ä vis English, a major reason why the shift to English continues. There would seem to be two ways to address this issue: (a) seeking strategies for empowering Norf k and (b) seeking ways of disempowering English. In the circumstances, (b) would seem to be the more promising strategy. For instance, instead of granting Norfolk Islanders the right to use Norf k in drawing up land title documents one can take away the legal status of documents drafted in English. As long as monolingual English speakers can settle, be employed and be elected into the legislature, there is no instrumental motivation to learn Norf k.

3.

Palmerston English (PE)

3.1.

Social history of Palmerston Island

Palmerston Island is part of the Cook Islands, in the Polynesian part of the South Pacific. The island is situated 434 km from the main island Rarotonga. The whole atoll has only 2,6 km2 of land area, which are shared by six islets, whereas the coral reef covers 1,457 hectares of a lagoon up to 11 km large. For hundreds of years, the island was used as a relay station for Polynesian travellers and fishermen; archaeological findings (adzes, bones) prove this fact. There was no continuous settlement, however, possibly because of the weak situation of the flat island - the highest point is a sandy hill of 6m - during violent hurricanes.

144

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

Captain Cook was the first European to sight the island in 1774, on his second voyage, and he landed only on his third and last voyage in 1777. He named it after the Lord of the British Admiralty. In 1811,•η there was a first unsuccessful attempt by Europeans to settle on the island. In 1863, the Englishman William Ma(r)sters settled on Palmerston with several Polynesian women of the same family from Penrhyn Island in the Northern Group of the Cooks. The oral tradition mentions his three wives; there might have been one or two more and also some men who worked with him to establish coconut plantations on the atoll. An initial survey and contacts with the archives of the Hamilton Library and the Bishop Museum in Hawai'i, les Archives Territoriales de la Polynesie Fran9aise in Tahiti, the Auckland University Library in New Zealand, the National Archives in London and personal communications through the researchers of the Natural Heritage Project of the Cook Islands indicate that there remains a lot of sociolinguistic research in this field to be carried out (exact origin of the first settlers, land ownership, contacts to other islands). The forefather ruled on the island in an autocratic style. Only English was allowed to be spoken and it is recorded that his children spoke English with a strong regional British accent. The social life of his numerous offspring living on the island had also to be organized and therefore, William divided the island into three parts according to the three clans stemming from his wives Akainangaro, Tepou and Matavia. Before his death in 1899, he established a taboo for intermarriage within the same clan which is still respected. After a period of fighting for dominance, William Π of the Akainangaro branch took authority over the island and he is said to have perpetuated his father's will to keep the original English as the only language on Palmerston. In the 1920s and 1940s, the island was extremely isolated due to severe hurricanes in 1923, 1926, 1931, 1935, 1942 and also due to WW Π. Whenever it was possible, Palmerston people had sought partners in the Pacific neighbourhood but they could hardly do so during this period and there were several marriages between two partners from the island. Creolization is likely to have occurred with this locally-oriented third or forth generation. From the end of WW II until the 1980s, Palmerston English underwent stabilization due to the continuous isolation. During the last two decades, migration to Rarotonga and to New Zealand has increased and it is threat-

Pidgins and Creoles between endangerment and empowerment

145

ening to deplete the island's population, to weaken its lifestyle and to threaten its language. 3.2.

General description of the language

The existence of a distinct Palmerston English Creole has not been widely acknowledged and no mention of such a language is made in the Ethnologue List (www.ethnologue.com). Like Pitkern and Norf k, Palmerston English (PE) is a Polynesian-English contact language combining elements from a variety of mid-19th-century British English (with Gloucestershire and Yorkshire being mentioned in the oral tradition, information which will be confirmed by future research, a research project on the dialectal origin of Palmerston English is in preparation) and Polynesian, especially in its variants of the islands of Manihiki and Rakahanga in the Northern Cook Islands. In the bilingual Marsters family, Polynesian speakers were compelled by the English-speaking family head to give up their mother tongue and to use English, a situation similar to the one on Pitcairn. The linguistic features of PE show evidence of the modification of English used as an L2 by LI Polynesian speakers. Polynesian influence can be noticed in the phonology: the glottal stop (indicated by ') appears as a variant of fhJ or /T/; there is a partial loss of the voiced/voiceless contrast in the consonant system: dis 'ausis 'these houses'; the vowels tend to present a basic 5-vowel-system in stressed position and a centralized schwa-like vowel (indicated by l@l) in unstressed position). There are expressions originating in archaic or dialectal English like tu ev @ ya@n 'to have a yarn' [talk to each other] and yond@ for 'under, over there'. Verb inflections too may be of dialect origin. Navigational terms from English are over-generalized and also used for movements on land. The official language of the Cook Islands, English, and the socially powerful Cook Island Maori remain the main sources of the lexicon. The fact that Palmerston is part of the Cook Islands means that Northern Cook Maori words are used for fishing and agriculture and Rarotongan Maori words in the fields of administration and Polynesian culture with no equivalent in the Palmerston tradition (administration, Polynesian dance). The syntactic structures of PE deserve far more thorough study than has previously been possible. In fact the only provisional account is Ehrhart-

146

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

Kneher (1996). The pronouns are of particular interest, with dual constructions and inclusive/exclusive differentiation following the Polynesian system: yu tu en matavia a@ i@ 'you and Matavia are here'. In the verbal group, aspect is a central subject for further observation, in particular constructions of the type bin + verb to express perfectivity or anteriority. Repetition expressed by reduplication is frequent: de boi noknok 'is 'ed 'the boy is/was knocking his head repeatedly' and might be a trace of the Polynesian substrate. A comparison with reduplication in NF (Mühlhäusler 2003) would suggest that reduplication is much rarer there - a contrast for which it would be interesting to uncover the reason. The verb ending for the present tense is -s throughout: I goes, you goes, we goes, dej goes ..., except for the third person singular: he go. According to several of our informants, sentences like ai tu:k a paka soup en ai uen tu de fare kaukau Ί took a piece of soap and I went to the bathroom' go back to the first generation of language contact (the Polynesian women trying to speak their husband's language). Nowadays, the speakers dissociate themselves from this kind of speech, which they consider to be corrupt, unstable and of little prestige. The lack of stability of PE in the early stages of contact is explained by the fact that the words of the two languages were just placed side by side or mixed arbitrarily, rather than being properly integrated, as would be the case in the present. 3.3.

Patterns of communication for PE

Due to the high mobility of Palmerston Islanders in the Pacific region and the consequent dilution of their traditions, the number of fluent speakers of PE is rapidly decreasing, which makes the language a highly endangered one. The core area of PE is situated on Palmerston Island with about 50 fluent speakers, the entire population of the island. The diaspora is much larger, though still small in absolute terms, with about 400-500 more speakers living on other islands of the Cook group and a further c. 400-500 persons in the Pacific region having a passive or an active knowledge of PE. There is a very strong migration movement from and to Palmerston (often with Rarotonga as an intermediate stage) and it is considered as natural by the families to be mobile, despite the very poor maritime transport conditions. There is no airstrip either because the three family branches were not able to come to an agreement: space is very scarce as they share all the

Pidgins and Creoles between endangerment and empowerment

147

islets and such a project needs the approval of the whole community. The island and its language and cultural traditions have an important symbolic unifying role for the whole family and are held in high esteem by the family members not living on the island. For those with Palmerston as place of birth and with a long period of stay on the Island, this link is particularly important. On the other hand, in the Island tradition, every person who is accepted by the Island community and sets foot on Palmerston soil is considered as being part of the family. For most Palmerston Islanders living abroad, however, the knowledge of the language can be considered as dormant and it is not automatically transmitted to the next generation while living overseas. On Palmerston, PE is the language of internal communication. The patterns of language use can be described as diglossic, since other languages are used for the contact with the outside world: English in its Rarotongan or in its New Zealand variety, Rarotongan Maori and the Polynesian languages of the Northern Group of the Cooks. English is mastered at different levels of proficiency depending on the individuals; the knowledge of Maori is passive or very poor in most of the cases. The local radio operator gets some special training in Maori in order to be able to communicate with the other outer islands of the Cooks. The perspectives for the language depend heavily on the island's demography. How many of the young people want to stay on the island? At which stage of their lives do Palmerston Islanders leave their island? Under which circumstances do they wish to come back? If they have precise plans, are they then intending to stay definitively or do they have another project of migration for later on? What kind of English variety do they bring back from abroad? If they get married abroad, where is their partner's place on Palmerston? For the moment, everybody on Palmerston carries the family name of Marsters. Is it possible to imagine a future for the Island with other family names being introduced? Palmerston English is even more strongly threatened outside the home island, in the diaspora, which is where the bulk of the speaker community lives - a phenomenon not uncommon for Pacific Island cultures. It will be interesting to study the links between language use and identity and we will try to measure whether the pull of the homeland is strong enough to counterbalance the loss of tradition and the de-focussing/diffusion (Le Page and Tabouret-Keller 1985) of its linguistic system and the associated discourse traditions.

148

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

Palmerston English, like many other powerless forms of English (e.g. the Bass Straits English spoken by the descendants of American and European whalers and their Tasmanian consorts or Tristan da Cunha English) are not well documented varieties. The only widely available published account so far is a brief sketch based on Ehrhart's fieldwork of 1991 (Ehrhart-Kneher 1996). There are incidental observations and comments in books written by South Seas travellers and locally-based researchers, but there is no video documentation of either the island or the language. Recent material, which has been published by travellers/sailors through the WorldWide Web offers interesting glimpses but is very often too limited in quantity and quality to provide a basis for a scholarly analysis. Information on language use is not easy to obtain because of speakers' attitudes, which reflects PE's status as a pyjama language, a language, which is used only in a very intimate circle.8 There is no written tradition in PE. Some elements of local documents composed traditionally in Standard English (land contracts, official documents or epitaphs on tombstones) show the influence of a PE substrate. Until the very recent past, writing was not an important aspect of island life, which among other things is reflected in a high rate of illiteracy, especially amongst the old people. The Island community is aware of the need for modernisation and is planning to commit to writing important aspects of their oral tradition, in particular narratives on early settlement, important family events and the foundation myth. PE is a contact language of high interest. Compared to the better described cases of Pitcairn and Norfolk, it remains an underdocumented language. Hence, data from PE have not been considered in linguistic discussion in which they would clearly have been important (cf., e.g. Baker and Huber's 2001) comparison of world-wide features in English-derived contact languages and the systematic comparison of PE and Pitkern, NF, Ngatik and other English-derived contact varieties suggested in Ehrhart-Kneher (1996) remains to be undertaken. The aim of our comparison of Norf k and Palmerston English is to bring together two communities with historical connections and with obvious similarities in their sociolinguistic development and in the ecolinguistic environment in which they communicate. For example, it is striking that in both instances the numerically dominant female speakers of Polynesian were not permitted to pass on their language to the mixed offspring. On Palmerston this was due to the influence of a single English-speaking male,

Pidgins and Creoles between endangerment and empowerment

149

and on Pitcairn Island after 1800, again a single Englishman succeeded in imposing the English language and much of English culture. In the linguistic history of both islands the children were an important factor because it was they who developed a contact vernacular on the basis of the pidginized English used in intercommunication in the initial years of settlement. Finally, on both islands written and formal English (initially established mainly through the Bible and religious literature, latterly also through education and literacy in general) has remained unchallenged as the High variety. Understanding the factors that sustain and weaken small island contact vernaculars could be of potential importance to an understanding of wider issues of language decline. The areas in which a structural comparison between Norf k and Palmerston English would seem particularly promising are dual and plural pronoun systems, prosody (yet to be described) and the extent of Polynesian influence in the lexicon.

4.

Conclusions

Creole languages that developed on remote isolated islands are often portrayed as test cases for linguistic and sociolinguistic theories (Calvet and Karylemou 1998, Reinecke et al. 1975, Chaudenson 1992). Their role is both to strengthen existing generalizations about languages and to highlight singularities in language development. The two languages focussed on in this paper developed at the remote edge of the British colonial empire but nonetheless did not escape the linguistic ideologies of 18th and 19th century Britain. They illustrate what Silverman (1967: 203) has referred to as the "triumph of English". On Pitcairn (and subsequently Norfolk) and Palmerston acrolectal English has remained the high language against all geographic and demographic odds and Standard English is likely eventually to replace the mixed Creoles of these islands, just as the Polynesian languages of the Polynesian females in these beach communities were replaced. One of our central questions has been concerned with the extent to which the languages of the Polynesian consorts of powerful white men nevertheless could continue to remain influential in the face of outright suppression. To answer this question it is necessary to document both the external (institutional) and internal (private) instruments of power. In the case of Pitcairn Norfolk and Palmerston, external measures became operative

150

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

mainly in the 20th century, particularly in the form of educational policies and practice. For the earlier phases the power differential within the communities was of crucial importance. In these communities, white AngloSaxon patriarchs were powerful role models and their agenda (civilizing through assimilation to English language and customs and Christian religion) prevailed. However, their influence was attenuated by a range of pragmatic factors and by the ability of their Polynesian women to undermine their power in a subtle way. The linguistic distance between the Europeans and Polynesians was one of the factors that required a pragmatic solution. The Polynesians initially simply were not able to acquire full competence in the patriarchs' language. The resulting mixed language was sufficiently transparent for the white males to enable communication with the women and to satisfy their wish to be in control of communication. This impression was reinforced by the predominance of lexical items of English origin and a longitudinal study is likely to confirm that an increasing number of Polynesian words were replaced by English ones, except in female domains such as cooking, childrearing and gardening. In the case of Pitkern, words expressing undesirable, marked or taboo concepts typically are of Polynesian origin (see Section 2.2.). The development of a Creole- acrolectal English diglossia was again the outcome of pragmatic factors. The Creole served to index women, children and other powerless members of the community and it should be noted that some members of the better families (usually descended from later arrivals such as teachers and missionaries) never spoke the basilect. However, the Creole also fulfilled other functions. In the case of Pitcairn, it was desirable to have a language that outsiders could not understand, as for many years the population lived in fear of being brought to justice by the British navy. As contacts with outsiders increased it also became a means of defining a separate identity. Whereas Tahitian dress, dancing, building styles and lifestyle were eroded by the successive efforts of outsider missionaries and educators, the distinct language remained and, in times when outsider influence was resented, became a symbol of resistance. The current renaissance of Pitkern and Norf k reflects the growing tensions between their speakers and the British and Australian government respectively. One of the outstanding tasks for analysts is to document how Polynesian ways of thinking are perpetuated in what many of the users of Pitkern, Norfolk and Palmerston regard as a dialect of English.

Pidgins and Creoles between endangerment and empowerment

151

A preliminary analysis suggests a relatively small number of grammatical features that can be directly related to a Tahitian substratum, including relatively free conversion of word classes and dual pronouns. Additional insight could be gained from exploring systematically the conventions of speech use such as different norms for asking and answering questions. Eades' account of communicative rules in South East Queensland Aboriginal English (Eades 1991) provides an appropriate model. Not being of English origin should not be equated with Polynesian origin, however. A large number of constructions and pragmatic rules in Pitkern-Norf k and Palmerston English appear to reflect learners' grammar (e.g. absence of inflections and the adaptation of the Creoles to the specific island conditions [spatial orientation systems]). It is noted that in Norf k the principal reference point for an absolute spatial system is the seat of power (the administrative capital Kingston). Other locations are defined in terms of horizontal and vertical distance. Both the inhabitants of Norfolk Island and those of Palmerston operated in terms of a stable diglossic system which they took for granted until fairly recently. Demographic changes, increasing outsider control of island affairs and information technology have triggered dramatic changes and there is now a growing awareness among younger speakers that the survival of their island Creoles can no longer be taken for granted. This realization appears to have resulted in different reactions. Palmerston Islanders are eager to see their language documented and preserved but are not very strongly motivated to empowering it. Given that Palmerston Islanders have tended to have a relatively powerful status within the Cook Islands their main aim is to preserve a distinct identity. The case of Norf k is different. Many Norfolk Islanders feel that they have been let down by the Australian government and that their rights and privileges on the island have been eroded. Their discourse about the revival of Norf k includes the strengthening of a separate identity and, at the same time, linguistic rights and empowerment through language. Norf k has just been given co-official status with English and is likely to be used in a growing number of public domains. It can also be expected that Norf k will be mainstreamed in the education system. It remains to be seen whether such positive external changes will be matched by changes in the internal vision on language matters among the population of the island.

152

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler

Notes 1. We shall employ the glottonyms Palmerston English for the language of Palmerston Island, Pitkern - Norf k for the varieties of Pitcairn and Norfolk Island and we shall distinguish between Pitkern and Norf k where indicated. The spelling Norf k was made official by the Norfolk Island Language Act of December 15th 2004. 2. Because of the diminutive population size of Pitcairn and Palmerston empowerment is less an issue. 3. The Polynesian women of both societies left their home islands as young girls. Because of their age and their belonging, in most instances, to the lower classes, their cultural knowledge may have been limited. 4. The linguistic nature of the various English-related Pacific contact vernaculars, like their Atlantic counterparts spoken on Tristan da Cunha and St. Helena, remains an issue, and labels such as 'Pidgin', 'Creole', 'Creoloid' that have been applied to them, remain problematic (see Mühlhäusler 1998b). For the purpose of our argument, a resolution of these terminological issues is not required and our choice of the labels Palmerston English and Norf k is not meant to prejudge the issue of typology of contact languages. 5. The extent to which later arrivals such as George Hunn Nobbs (1828-1884), the religious dictator Joshua Hill (1832-1834) or the various Seventh Day Adventist missionaries had passive knowledge of the language is not known. However, the fact that Pitkern had become an esoteric insider language combined with the social rules for its use, suggest that even passive knowledge is minimal. This is reminiscent of other Pidgin and Creole societies where the elite does not bother to acquire competence in the lower variety. During Australian colonial control of Papua New Guinea, for instance, even long-time expatriate residents had little or no knowledge of Tok Pisin: "A rough and hopeful guess is that one in fifty can understand Pidgin as spoken by the indigenes to each other" (Bell 1971: 38). 6. Originals of these internal reports are located in the Mitchell Library Sydney and copies in the Flint collection of the Fryer Library, University of Queensland and the School Library Norfolk Island Central School. 7. "Six European were landed by Captain Fodger of Daphne, murder of two of the men by the native." Colonial Secretary Index, State Records of New South Wales, Australia. 8. Ehrhart's standing in and contacts with the island society for over a decade (and recordings with members of the family that have now passed away) have been very helpful for the recordings.

Pidgins and Creoles between endangerment and empowerment

153

References Anonymous 1932

Now all yolye: The quaint speech of Norfolk is. The Pacific Islands Monthly, p. 11.

Baker, P. and M. Huber 2001 Atlantic, Pacific, and world-wide features in English-lexicon contact languages. English World-Wide, Vol. 22: 2. Bell, H. L. Language and the army of Papua New Guinea. Australian Army Journal 264: 31-42. Buffet, A. 1999 Speak Norfolk Today. An Encyclopedia of the Norfolk Language. Norfolk Island: Himii Publishing. Buffett, A. and D. C. Laycock 1988 Speak Norfolk Today. Norfolk Island: Himii Publishing. Calvet, L.-J. and M. Karylemou (eds.) 1998 Des lies et des Langues, special issue n° 15 Plurilinguismes, Paris. Chaudenson, R. 1992 Christian, G. 1982

Des lies, des Hommes, des Langues. Essai sur la Creolisation Linguistique et Culturelle. Paris: L'Harmattan. Fragile Paradise - the Discovery of Fletcher Christian BountyMutineer. London: Hamish Hamilton.

Crowley, T. 1990

Beach-la-Mar to Bislama. Oxford: Clarendon Press.

Dening, G. 1992

Mr. Bligh's Bad Language. Cambridge: CUP.

Drechsel, A. 1997 Eades, D. 1991

Mobilian Jargon. Oxford: Clarendon Press.

Communicative strategies in Aboriginal English. In Language in Australia, Romaine, S. (ed.), 84-93. Cambridge: CUP. Ehrhart-Kneher, S. 1996 Palmerston English. In Atlas of Languages of Intercultural Communication in the Pacific, Asia, and the Americas, S.A.

154

Sabine Ehrhart, Christian Mair and Peter Mühlhäusler Wurm, P. Mühlhäusler and D.T. Tryon (eds.), 523-532. Berlin: Walter de Gruyter.

Errington, J. J. 2001 Hagen, G. 1908

Harrison, S. 1972 Hoare, M. 1982 Keesing, R. 1990

Labov, W. 1972 Laycock, D. 1989

1990

Colonial linguistics. Annual Review of Anthropology 30 (1): 1939. Kurzes Handbuch für Neger-Englisch an der Westküste Afrikas unter besonderer Berücksichtigung von Kamerun. Berlin: Dingeldey and Werres. The Language of Norfolk Island. Honours thesis. Sydney: Macquarie University. Norfolk Island: An Outline of its History 1774-1981. St. Lucia: University of Queensland Press. Solomons Pijin: Colonial ideologies. In Language Planning and Education in Australasia and the South Pacific, R. Baldauf and A. Luke (eds.), 149-165. Clevedon: Multilingual Matters. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press. The status of Pitcairn-Norfolk: Creole, dialect or cant? In Status and Function of Languages, U. Ammon (ed.), 608-629. Berlin: Mouton de Gruyter. The interpretation of variation in Pitcairn-Norfolk. In Development and Diversity: Language Variation across Time and Space. A Festschrift for Charles-James N. Bailey, J. A. Edmonson, C. Feagin, and P. Mühlhäusler (eds.), 621-627. Arlington: Summer Institute of Linguistics and the University of Texas.

Le Page, R. B. and A. Tabouret-Keller 1985 Acts of Identity. Cambridge: CUP. Maude, Η. Ε. 1964

The history of Pitcairn Island. In The Pitcairnese Language, A. S. C. Ross and A. W. Moverley (eds.), 45-101.

Pidgins and Creoles between endangerment and empowerment Mercer, B. 1987 Mosel, U. 1998

155

An Island Education - A History of the Norfolk Island Public School. Norfolk Island: P. & C. Association. Some Pacific island Utopias and their languages. 15:27-47.

Plurilinguismes

2002

Description of the Project "The Teop language of Bougainville, Papua New Guinea", financed by the VW foundation for endangered languages, Hannover, Germany: www.volkswagenstiftung.de/foerderimg/foerderinitiativen/bewilligungen/proldoku _d.ht

2003

A note on reduplication in Pitkern-Norfolk. In Twice as Meaningful, S. Kouwenberg (ed.), 239-244. London: Battlebridge.

O'Collins, M. 2002

An Uneasy Relationship: Norfolk Island and the of Australia. Canberra: Pandanus Books.

Commonwealth

Reinecke, J. E., S. M. Tsuzaki, I. F. Hancock, and R. E. Wood (eds.) 1975 A Bibliography of Pidgin and Creole Languages. Honolulu: University Press of Hawaii. Ross, A.S.C. and A.W. Moverley The Pitcairnese Language. London: Andre Deutsch. 1964 Silverman, D. 1967 Wurm, S. A. 2001

Pitcairn Island. Cleveland and New York: World Publishing Company. Language endangerment in the insular Greater Pacific area, and the New Guinea area in particular. In The Boy from Bundaberg Studies in Melanesian Linguistics in Honour of Tom Dutton, A. Pawley; M. Ross & D.T. Tryon (eds.), 283-298. Canberra: Pacific Linguistics.

Wurm S. Α., P. Mühlhäusler and D. T. Tryon (eds.) 1996 Atlas of Languages of Inter cultural Communication in the Pacific, Asia, and the Americas. Berlin: Mouton de Gruyter.

Lost in transculturation: The case of bilingual education in New York City* Ofelia Garcia

1.

Another West Side Story by way of introduction

I started teaching in an alternative public school in New York City in 1970 in what was then called Hell's Kitchen. The neighborhood was then a seething mix of tenements and factories not too unlike the description given to us by Jacob Riis in his How the Other Half Lives (1890). The community was that of Leonard Bernstein's West Side Story, with violence, tragedy, and despair coexisting with love and hope. Prostitution and drugs were visible, but so were struggling families, anxious to build a better life for themselves. At the time, Puerto Ricans, US citizens by virtue of their colonial status, made up eighty percent of the New York Latino population, and bilingual education was just coming into being. Responding to the dismal failures of Puerto Ricans in New York schools, as well as to the political pressure of a Puerto Rican Civil Rights Organization, Aspira, the New York City Board of Education supported the creation of transitional bilingual education programs where newly arrived PUerto Rican children could be educated in Spanish while getting intensive English as a Second Language instruction. I became a bilingual teacher overnight, teaching English as a Second Language as well as subject matter in Spanish. Thirty years after my initial teaching appointment, the transformation of Hell's Kitchen into what is now called "Midtown West" mirrors changes in the City and its schools. Cruise ships are now docked in the once abandoned piers. The sex shops have been transformed into fusion restaurants that blend different ethnic cuisines. The tenements have been replaced by luxury condos, such as the newly opened AOL Time Warner Center where apartments sell for two million to thirty-two million dollars. The few tenements that remain are now crowded by mostly Mexican, Central and South American immigrants, as many Puerto Ricans have moved out, leaving symbolically behind the Puerto Rican Traveling Theatre. The moving images of cruise ships and a traveling theatre, the hybrid smells and tastes of fusion restaurants, and the technological triumph

158

Ofelia Garcia

evoked by the AOL Time Warner building are suggestive of the flux and the transcultural context in which we find ourselves in the 21st century. Our postmodern era has been described as a post-industrial period in which new technologies, media images, and service economies characterize a global, multinational, capitalist marketplace (Appadurai 1996; Graddol 1997; Sefranek 2003). And yet, for a short time in the days following September 11, 2001, we were reminded of how life in the flux could be an illusion. The image of modern travel, the airplane, was turned into a weapon that made time stop for all of us. As the airplane merged with the weapon, we were brought face to face with how our global interdependence and our local realities coexist, often inextricably tied in tension. The image of the airplane-turned-weapon is instructive today to analyze the turn that bilingual education has taken in the United States as immigration has increased and our linguistic profile has become more heterogeneous and complex. In the face of the new linguistic flux and hybridity brought about not only by the proliferation of languages in the United States, but also by the speed of communication and travel made possible by technology, the transitional bilingual education models of the past are being substituted by two-way dual language bilingual education models (TWDL from now on) where children of different linguistic profiles are integrated for instruction in two languages. These dual language bilingual education programs clearly respond and are able to better accommodate the greater language and cultural hybridity of the 21st century. And yet, they contain within them the risk of turning their own hybridity into a weapon capable of destroying the minority language. In this paper we point to the possibilities and the dangers of dual language bilingual education programs1, arguing that unless these programs reclaim diglossia (see below) and extend it in ways that support the new hybridities, as well as the old differences, they run the risk of losing themselves in English.

2.

Diglossia and U.S. schools

The study of societal multilingualism and the scholarly recognition of the role of minority languages in the 20th century owe much to the work of Joshua A. Fishman. Building on the model of diglossia posited by Ferguson in 1959, Fishman (1972) has argued that stable societal multilingualism requires the functional allocation of languages and their compartmentalization in social life. That is, in order for two languages to be maintained

Lost in transculturation

159

intergenerationally, a diglossic arrangement is necessary, with each language being used for distinct purposes or in different domains. Transitional bilingual education programs in the developing world has been a form of educational development and a way to empower the minority group (Benson 2004). The programs implemented in US elementary schools at the end of the 20th century usually have one Latino bilingual teacher who teaches recently arrived students in Spanish and English, often going back and forth between the two languages. By violating diglossic principles, transitional bilingual education programs are agents of language shift to English (Fishman 1976; Garcia 1993). And yet, transitional bilingual education programs are useful "safe houses" for these newly arrived students. In these programs, Latino students are able to learn about the ways of schooling in the United States, as well as acquire English, with a teacher who is a cultural and linguistic broker (Garcia 1991; Garcia and Baker 1995). But the United States' sociolinguistic and socioeducational situation was soon to get more complicated as the world became globalized. At the end of the 20th century, a change in U.S. immigration law made it possible for many more Latin Americans, Asians and Africans to immigrate to the United States. At the same time that there was freer movement of speakers, new technologies made it possible for languages to move over a multiplicity of communication channels and at a speed unheard of in the 20th century (Maurais and Morris 2004). On the other hand, globalization made it possible for English to increase its supremacy in language domains with high prestige, especially science, mass media, technology, popular culture, international business, and education. The paradox created by the greater heterogeneity and hybridity of discourse alongside the greater homogeneity and hegemony of English at the beginning of the 21 st century has impacted how immigrants are being educated in the United States, often leaving them with much less than what they had in the late 20th century. Today, transitional bilingual education has been outlawed in Arizona, California, and Massachusetts (Crawford 2000). And it is the model known as Two-way Dual Language or Dual Immersion (TWDL) that often has become the only viable alternative to transitional bilingual education for immigrant children (Cloud, Genesee and Hamayan 2000; Lindholm-Leary 2001, Torres-Guzman 2002). Unlike transitional bilingual education programs, dual language bilingual education programs follow both diglossic and additive principles of bilingualism. In dual language programs the two languages are strictly compartmentalized, and the children stay in the pro-

160

Ofelia Garcia

gram for the duration of their elementary schooling. Two-way dual language programs are also better models of second language acquisition and social integration, since recently arrived immigrants are schooled with those who are fluent in English and born in the United States. Given that mass migration and electronic mediation mark today's world, the Two-Way Dual Language models are better able to accommodate the hybridities and flux of today. But two-way dual language bilingual education programs are not always able to sustain linguistic differences and bilingualism. Although the programs compartmentalize the two languages strictly, they keep the children, all of whom show different degrees of bilingualism, integrated for instruction throughout the entire day. The TWDL model then sustains diglossia by compartmentalizing the languages of instruction, but violates it and works against it because children who have no language in common creatively mix the languages in an effort to communicate. Although the linguistic arrangement of dual language programs can support the complex multilingualism and cultural and linguistic hybridity of the 21st century, the TWDL model can also work well to promote English language monolingualism, in much the same way as transitional bilingual education programs have done in the past.

3.

New York City's sociolinguistic profile

New York City is one of the most multilingual cities in the world (Garcia and Fishman 2001). As we enter the 21st century, only 52 percent of the school-aged population (5 to 17 years old) is English monolingual, and 29 percent of the school-aged population speaks Spanish at home. According to the 2000 Census, there were 405,522 school-aged children who spoke Spanish at home in New York. In addition, there were 50,665 students who spoke Chinese; 27,549 who spoke Russian; 25,285 who used Yiddish at home; 18,844 who spoke French Creole; and 12,515 who spoke French. There were also 11,599 students who were speakers of Korean; 10,340 who spoke Hebrew; 9,892 who used Arabic at home; 9,702 who spoke Urdu; 9,072 who used Italian; and 7,557 who spoke Polish (U.S. population Census 2000). New York City has a very different sociolinguistic profile today than when I started teaching, when speakers of languages other than English were mostly Puerto Ricans. Today, Spanish is still the number one lan-

Lost in transculturation

161

guage, although the speakers of Spanish are no longer Puerto Ricans and the varieties of Spanish heard around the city are more diverse than ever. Although Puerto Ricans are still the majority Latino group in the city, they have consistently lost dominance, dropping from 90 percent in 1960 to 37 percent in 2000. And although the Dominican population has risen to make up 19 percent of Latino New Yorkers, there has been an enormous growth of Mexicans, Central Americans and South Americans. This increase has challenged the supremacy of Caribbean Spanish in New York City. In 1960, 90 percent of Latinos in New York City were speakers of Caribbean Spanish, whereas in 2000 only 57 percent of New York Spanish speakers could be considered speakers of Caribbean Spanish (Zentella 2001; Garcia 2001, 2002, 2003a, 2003b). Spanish has a very large presence in New York City. Of the two Spanish-language daily newspapers, El Diario La Prensa sells 60,000 copies a day and Noticias del Mundo sells 28,000 copies. The second most heard radio show in New York City is a Spanish language show, El vacilon de la manana, heard in La Mega, one of four radio stations that transmits Spanish language programs in New York City. As the city has grown to be more Spanish-speaking, and more diversified in the varieties of Spanish used, the Spanish language itself has gained more global status. For example, the Lnstituto Cervantes, a non-profit organization created by the Spanish government to teach Spanish and to contribute to the advancement of the Spanish and Hispanic American cultures throughout non-Spanish speaking countries has become firmly established. In 2004, the Lnstituto Cervantes opened the doors of an unequaled Spanish language and cultural center in New York, offering academic and cultural programs in Spanish. Likewise, Spanish has become an important commodity to do business with the growing Spanish speaking population in the United States, as well as with Latin America. The Latino market's current purchasing power in the United States has been assessed at $206 billion (Zentella 2001). In June 2002, Coca-Cola, Wendy's, Sears and other major corporations spent fifty million dollars in Spanish language commercials for the World Cup that was being transmitted by Univision. It is significant that Univision, the major Spanish language network in the United States, advertises itself by saying: "Univision, as American as flan," claiming a Spanish-language Americanness akin to the custard-like desert of all Spanish-speaking nations, and increasingly of the U.S. (Garcia 2003c).

162

Ofelia Garcia

Despite the more global aspects of Spanish today, Spanish in the United States continues to be seen as the language of those who are poor and nonwhite. Recently, for example, the author of an advice column of a mainstream journal, Vanity Fair, told a reader: "Forget Spanish As for everyone's speaking it, what twaddle! Who speaks it that you are really desperate to talk to? The help? Your leaf blower?" (Ask Dame Edna 2003:116).

4.

Dual language bilingual education in New York City: Supporting diglossia and hybridity

The increased presence of Spanish in New York City, coupled with the improved image of Spanish in the world's stage, has been a determining factor in the creation and development of two-way dual language programs in Spanish and English. In the late 1990s TWDL programs, especially at the elementary level, were encouraged in New York State as a way to meet the exigencies of regulations concerning English Language Learners and new immigrants, and as a way to integrate a growing Latino population that had been segregated for instruction in transitional bilingual education programs. With the exception of one Chinese (Mandarin)/English two-way dual language bilingual education program, the rest of the TWDL programs in New York City are Spanish/English. According to a document produced by New Visions for Public Schools, two-way dual language bilingual education programs have two defining characteristics: (1) "There is separation of the two languages for instruction. Sustained periods of monolingual instruction promotes linguistic development better than mixing languages within the same lesson. The non-English language is used at least 50 percent of the time for instruction." (2) "The program serves a balanced ratio of students who speak each language, ideally 50/50 but not to go below 70/30." The first criteria of language separation sets the stage for diglossia, while the second criteria of a balanced number of speakers of each language sets the stage for hybridity.

5.

Diglossia

The language separation or diglossic instructional arrangements in the many Spanish/English programs follow different patterns. In one of the

Lost in transculturation 163 Spanish/English TWDL schools, Spanish instruction takes place for an entire week with a Spanish language teacher, and then students switch over to the English language teacher who follows up the instruction in English (ialternate week compartmentalization). In another TWDL school, students are taught in English one day and Spanish the next {alternate day compartmentalization). Other schools alternate time of day, with one language being used in the morning for instruction, and the other language used in the afternoon. Some schools alternate the order of languages, following what is called a "roller coaster" model in which if Spanish is taught in the morning in Day 1, it is taught in the afternoon in Day 2. Most TWDL schools maximize diglossia by having different teachers teach in the different languages (alternate teacher compartmentalization). In the Mandarin/English DL school, the two languages are clearly delineated. English is taught in the morning and Mandarin is taught in the afternoon by the same bilingual teacher. In addition, from 3:00 PM until 5:30 PM there is free and compulsory Mandarin-only instruction taught by teachers who are not certified and are not employees of the New York City Department of Education. Despite the external diglossic structure of dual language bilingual education models, the diversity of children, teachers and ways of using the two languages reflects a hybridity that has little to do with the language and cultural differences experienced in the classrooms of Hell's Kitchen in the early 1970s.

6.

Hybridity: The children

Most of the children who attend the Spanish/English dual language schools seem to be Latino. But, these children, for the most part, have little in common with the Spanish-speaking teacher. In a country like the United States where language shift advances quickly, most of the Latino children are no longer speakers of Spanish. Many second generation children hear Spanish in family situations, if not from the parents, from the grandparents. But for many, the language exists only in their memory or in the cultural attachment of the growing Latino popular culture of the United States - in the songs of Shakira, Jennifer Lopez, and the others. Many other children are products of mixed marriages, and Spanish is not the language of the home. Others come from Latin American countries that are officially Spanish speaking, but from regions or families where Spanish has never been

164

Ofelia Garcia

spoken. This is the case of the large Garifuna-speaking population in New York City, Hondurans who might speak Spanish as a result of schooling in Latin America, but for whom Spanish is not a home language and who use Garifuna, and not Spanish, in raising their children. This is also the case of the growing population of indigenous-language-speaking Latinos in New York City, many speakers of Quechua/Quichua, of Guarani, and of Mayan languages. As we have seen, even when the children are Spanish speakers themselves, there is tremendous heterogeneity in the Spanish language varieties spoken in New York City classrooms. The entire sociolinguistic complexity of Latin America can be observed in one dual language classroom. And this linguistic complexity is often juxtaposed in the texts produced by the children, written many times in groups and by groups of speakers that represent the entire linguistic continuum not only of Latin America but across the Americas, with U.S. Spanish also becoming one more variety. In the Chinese/English TWDL school, an outsider is able to identify the few non-Chinese students in the school, mostly students of African descent from the Anglophone Caribbean whose parents support the school's traditional discipline and structure. At a glance, Mandarin seems to be the heritage language of most of the children. Little by little, however, the linguistic complexity of these seemingly Chinese children emerges. On the one hand, there are children who are no longer speakers of Mandarin, and who struggle, alongside their English monolingual classmates, to acquire Mandarin. On the other hand, there are many children who are ethnically Chinese but who live today as children of white Anglophone parents, having been adopted at birth. And then, there are many who live in homes where another Chinese language is spoken - especially Cantonese, Fuzhounese, Shanghainese, Wenzhounese and Taiwanese. The linguistic complexity of all of China and of all of Taiwan is concentrated and evident in this small school, alongside the linguistic complexity of Anglophones not only in the United States but in the world. And many times, in one child's discourse one can hear evidence of it all. This is what makes dual language programs in the United States such a rich context for the study of multilingualism, for in a small space multilingual interactional negotiations are constantly taking place.

Lost in transculturation 7.

165

Hybridity: The teachers

In the Spanish/English TWDL programs, most bilingual teachers are of Latino background, although increasingly the teachers are not of the same national origin as the children. As we have seen, 80 percent of children of Latino background in New York City in 1960 were Puerto Ricans. Today, students of Puerto Rican background represent approximately a third of Latino students in the city. However, the majority of Latino students of Puerto Rican background, now most often second or third generation, are in monolingual classrooms and not in bilingual education classrooms, even of the dual language kind. But teaching has been a veiy attractive profession for New York Puerto Ricans who make up the majority of the city's Latino teachers. In Dual Language classrooms, Latino teachers, mostly of Puerto Rican background, are teaching Latino students who are not of Puerto Rican background, and increasingly not of Caribbean background. And thus, in these classrooms, all varieties of Spanish and ways of using Spanish are manifested, with the power of the teacher giving preference to Caribbean Spanish with its radical pronunciation and its exaggerated histrionics, and the power of the number of students giving preference to the more conservative pronunciation and the more understated Central and South American discourse. Increasingly these competing Spanish language discourses are transformed and hybridized as they're used in collaborative activities and in learning, and as teachers attempt to accommodate linguistically to the many Spanishes and the many Englishes of the classroom. More and more, teachers in Spanish/English TWDL classrooms are also Anglos. As a result of the higher standards for teachers that resulted in the establishment of the New York State Teacher Certification exams, Latino bilingual teachers without native-like English writing proficiency have been shut out of classrooms (Garcia and Trubek 1999). The shortage of bilingual teachers has encouraged the recruitment of former Peace Corps volunteers, generally Anglos who have acquired Spanish as a second language in the rural communities of Guatemala, Honduras, Mexico, and the like. The Spanish language these teachers bring is very different from that which their students speak. As second language speakers of Spanish, these teachers are confronted with varieties of Spanish that are sometimes not only alien to them, but that are juxtaposed with English, many times African American English or varieties spoken in the Anglophone Caribbean. The distance between the Spanish these teachers struggle to speak and the languages of their students is great.

166

Ofelia Garcia

In the Chinese/English TWDL School, the teachers are of three kinds. Firstly, there are Chinese American teachers who are bilingual and certified in New York State to teach childhood bilingual education. Secondly, there are Taiwanese nationals, who are most often Mandarin monolinguals and who are paid through private sources, often by Chinese American associations and by the Taiwanese government. Thirdly, there are Chinese elders in every classroom, supported by Chinese American associations. Each of the three groups has a different sociolinguistic profile. The Chinese American teachers are often English dominant, since the exigencies in academic English of the three tests of the New York State Teacher Certification Examination are certainly much higher than those required in Mandarin in the Bilingual exam. All teachers are of Chineseethnic background, with one teacher being half-Chinese, half-Anglo. And although all are bilingual by New York State standards, some speak Mandarin hesitantly by Chinese standards. In fact, there's a special program that takes some of these Chinese American teachers to Taiwan in the summer in the hopes of strengthening their Chinese language proficiency. In contrast, the Taiwanese Chinese teachers are not proficient in English, and although many are speakers of more than one Chinese language, they use solely Mandarin in instruction. On the other hand, the Chinese elders display the linguistic heterogeneity of the Chinese American community. Not all of them are speakers of Mandarin. Most are speakers of Cantonese, Fuzhounese, and Shanghainese, the languages of many children for whom instructional support in their mother tongue is not available. In the preschool and kindergarten classrooms, these Chinese elders abound, making it possible for Chinese children from non-Mandarin speaking homes to feel at home in the school.

8.

Hybridity: The language other than English

The different and conflicting discourses of the different children and the different teachers in dual language classrooms deeply affect the ways in which the language other than English is taught and used in instruction. As Spanish/English bilingual education programs came into being in the 1970s, the issue of which Spanish standard was to be used in instruction and in texts was hotly contested. Puerto Rican teachers often complained about the Spanish language material, mostly written in a Mexican American Spanish variety that was generally absent in the Northeast. Now, with

Lost in transculturation

167

the heterogeneity of Spanish varieties in the city, and especially the diversity of Spanishes spoken by students for whom Spanish is a first language, a heritage language or a second language, the question of which Spanish standard variety should be used cannot even be posed. TWDL programs today exhibit all Spanish language varieties of Latin America, Spain and the United States. In the Chinese/English DL school, while the Chinese American teachers use a Reading and Writing Workshop approach to literacy that involves Read Alouds and Shared Readings, the Chinese after-school teachers use traditional approaches that emphasize mechanical reading. The Chinese American teachers have child-centered classrooms where children work individually in Independent Reading and in pairs and small groups on Paired Reading or Guided Reading, but the Chinese after-school teachers only use whole class instruction. Chinese American teachers have students respond to authentic literature in journals or have them write book reports, poetry and creative pieces; the Chinese after-school teachers have students copy from the blackboard or from the book. Only structured writing activities are provided. Although at a glance it seems that there is compartmentalization not only of language, but also of the discourse that surrounds teaching and learning, a closer look enables us to understand the hybridity of teaching Mandarin itself. The students at this school are mostly from non-Mandarin speaking areas of Mainland China - from Guangdong and Fujian. Even though they're mostly from the People's Republic of China, it is the Taiwanese government that provides the books and the teachers of Mandarin. The People's Republic of China simplified Chinese characters in 1964. This simplified script is different from the traditional characters that immigrants who settled long ago in Chinatown have continued to use all over its streets. But Taiwan also continued to use the traditional Chinese characters. So it turns out that students are being taught to read the traditional script of Taiwanese texts, a script no longer used in their place of origin. Furthermore, to teach Chinese reading, the school uses Zhu Yin Fu Ha, the phonetic logographic symbols used to learn to read Mandarin in Taiwan. But this way of teaching to read in Mandarin is very different from the phonetic Romanized alphabet of Hanyu Piny in used in the People's Republic of China. So again, students are being taught to read through symbols that are never used in their place of origin. It is clear that when one observes two-way dual language classrooms, strongly child-centered as a result of having children of different linguistic

168

Ofelia Garcia

proficiencies in classrooms where only one language is used at a time, one is confronted by a heteroglossia, a hybridity of language use, usually unheard in more traditional classrooms and even in transitional bilingual programs. Dual language classrooms certainly advance transculturai understandings and act as a "third space," a discursive dialogic space in which, as Bakhtin has made clear, competing discourses coexist and languages are hybridized (Bakhtin 1981). But by not providing linguistically homogeneous groups with differentiated instruction, especially in literacy, the TWDL model can, in practice, work against diglossic arrangements that would sustain bilingualism over the course of a students' life. At the same time, by not providing opportunities to contrast the two languages and juxtapose them in instruction to develop the bilingual awareness of children, the hybridity they reflect can work against the minority language. The continued interactional negotiation of the languages and discourses of children in an instructional space where only one language is used but where the children have different degrees of proficiency, eventually cedes the space to English, the language of power, of assessment and promotion, even when a Language Other than English is officially used in instruction.

9.

Lost in transculturation in the USA. From bilingual/bicultural to multilingual/transcultural

Transitional bilingual/bicultural education programs were started as ways to promote the English language acquisition and the bilingualism of the US Spanish speaking population. The emphasis was on the bi- for a language minority group, even if for a short period of time. As such, these programs acted as "safe houses" in the sense given to the term by Mary Louis Pratt, a social and intellectual space where the group could constitute itself as "horizontal, homogeneous, sovereign communities with high degrees of trust, shared understandings, temporary protection from legacies of oppression" (1991: 39). On the other hand, two-way dual language programs are brave attempts to extend the bilingualism of the few in the United States to the many who are English monolinguals. The programs have the potential to shape the new transculturai identities that emerge from the interculturality of multiple knowledge bases that come together in the classrooms. The emphasis is on the multi- and the shaping of the trans-, a reality in a globalized world and in a post-colonial order.

Lost in transculturation 169 But increasingly, the standards-based instruction that now dominates schooling in the United States, coupled with the English-only assessments that have become the only measure of academic progress, is restricting the multiple languages, literacies and knowledge-bases that are present in U.S. dual language classrooms (Garcia 2003 a; Garcia and Menken forthcoming). Faced with No Child Left Behind legislation that mandates strict English-only assessment guidelines for all children, the present structure of dual language models cannot be successful in sustaining the multiplicity of voices and heteroglossia that they were meant to develop. In this transcultural and globalized space, especially as experienced from within the United States' hegemony, the power of English homogenizes and with ease shrinks the multi- and the trans- into a monolingual U.S. reality. Paradoxically, the transcultural heterogeneous space of dual language classrooms erodes the local minority language reality, as difference is lost. Michael Clyne, referring to the more dynamic language contact of today's world and the understanding that bilingualism cannot be studied from monolingual perspectives talks aboutplurilingualism (2003). We have seen in Dual Language classrooms the more dynamic, more encompassing and more inclusive use of the multiple languages and literacies of the 21st century, with each of them touching, "shrinking with intimacy" as the Mexican-American Gloria Anzaldüa would say (1987). But the tension is between the plurilingualism of the classroom, exercised by children and teachers with little power, and the monolingualism of assessment of the all-powerful state educational authorities. The power dimension of languages is a most important variable in considering the success of language policy efforts, and especially language policy in education (Bratt Paulston 2004; Clyne, this volume; Edwards, this volume; Spolsky, this volume). It is clear that in the United States the power of English gives little room to other languages. The U.S. educational system does not support the kind of plurilingualism that Clyne speaks about, where bilingual children would be taught and evaluated without being continuously compared to monolinguals. In fact, more than ever, United States schools assess bilingual children from monolingual Englishonly perspectives. The way in which United States educational authorities have chosen to interpret the "transcultural" context of the 21 st century has the potential of robbing the powerless of the few "safe houses" which they had been able to build for themselves at the end of the twentieth century. Little does this "transcultural" context have anything to do with the concept of "transcul-

170

Ofelia Garcia

turation" as coined by the Cuban ethnologist Fernando Ortiz. In his Contrapunteo Cubano del Tabaco y del Aziicar Ortiz (1978/1940) had said: We understand that the word transculturation better expresses the different phases of the transitive process of one culture to another, because this consists of not only acquiring a different culture, which is really what the Anglo-American word acculturation means, but the process also necessarily implies the loss or lack of hold of a first culture, that which can be called a partial deculturation, and it also points to the consequent creation of new cultural phenomena that could be called neoculturation. In effect, as the Malinowski school claims in all embraces of cultures there is something of what happens in the genetic copulation of individuals: the child always has something of both progenitors, but it is always different from each of them (p. 96, my translation). But this embrace where both are permanently changed is not part of the United States ethos. In two-way dual language classrooms, children of all linguistic profiles embrace for a short time, but the power of English and the insistence that a full academic, professional and social life can only be led as monolingual English-only individuals relentlessly affects only the Spanish- or Chinese-speaking children, leaving the English speakers mostly untouched. So what has been gained from going "transcultural"? What has been lost in leaving the bicultural? What have we gained from our linguistic hybridity? What have we lost in abandoning more traditional ways of defining bilingualism? I would submit that the impact of globalization on bilingual education for powerless groups in the United States has only created a cosmetic change, one which gives the appearance of integration, of transculturation, of hybridity, of flux. But seen from the perspective of the minority group itself, this flux has worked against their empowerment, as the homogeneous spaces in which they worked with each other have shrunk. Transitional bilingual education programs that provided a "safe house", even for a short time, are under attack and have been increasingly disbanded. Even in transitional bilingual education programs, teachers are increasingly told not to speak Spanish to the children. Bilingual education teachers have lost their jobs. Spanish in US public schools has often receded to after-school programs or to instances when the teacher can close the door. Spanish continues to lose itself in the apparent transculturation of the USA.

Lost in transculturation 10.

171

Conclusion

It is clear from the foregoing analysis that the dual language models of the 21st century respond to a changed transcultural context where hybridity is a defining characteristic. But hybridity increases risks of disorder, and unless it is used purposely by the educators in charge, it could provide a way in which the language of power exerts itself. In the United States we seem to have gone from a clearly non-diglossic model of transitional bilingual education to a mostly transglossic model of dual language education. We have created the "illusion" of an honest linguistic and cultural embrace, but in so doing, we have destroyed the educational spaces in which speakers of languages other than English could use those languages to acquire knowledge, status and power, and could do so without sharing space with those who already speak the language of power and who will, no doubt, do better in the high-stakes English-only assessments. The challenge of sustained multilingualism in our globalized world, and of the survival of local languages and cultures, will increasingly depend on transdiglossic models of schooling and other ways of behaving socially that both build on our increased hybridity, as well as protect our unique differences. Louis-Jean Calvet (1999) talks of a "gravitational model" of diglossia in which global powerful languages like English can coexist with and not threaten local languages. This is the view also expressed for Nigeria by Igboanusi and Wolf (this volume) and for South Africa by Alexander (this volume) and Finlayson and Slabbert (2004). Two-way Dual language models have the potential for a transdiglossic education that would truly and over time support bilingualism in the United States. But they need to carefully balance their structure, now fully diglossic, with a more hybrid structure, just as they would need to balance their participants' practices, now fully hybrid, with more diglossic practice. The two criteria of the dual language program mentioned above would have to be balanced in the following ways: 1.

2.

The strict compartmentalization of languages needs to cede some instructional space for putting the two languages alongside each other for purposes of study and comparison in order to develop children's metalinguistic awareness of their own bilingualism and biliteracy; The instructional integration of the linguistically different children needs to make some space for instruction of linguistically homogene-

172

Ofelia Garcia ous students at times, especially for literacy and most certainly during the years of emerging bilingualism.

Two-way dual language bilingual education programs today often do only what the left side of Figure 1 displays. To sustain our differences and develop the capacity for a multilingual citizenry, we need to balance the left side of Figure 1 with the right side. That is, besides language separation, TWDL bilingual education programs must ensure that teachers have opportunities to compare and contrast the students' languages, and on instances, use code-switching for instructional purposes (Myers Scotton, this volume; Van der Walt, Mabule, de Beer 2004). It is only by putting languages alongside of each other that students become conscious of their bilingualism and biliteracy. And besides integrating children with different sociolinguistic profiles, enabling the joint co-construction that is so important to scaffold learning of language minority students (Walqui 2002), TWDL programs must also separate them for specific targeted instruction in the child's first language. Diglossia

+

Hybridity

Language separation

+

Integration for language comparison and study

Hybridity

+

Diglossia

Children with different linguistic profiles always mixed for instruction

+

Children of same linguistic profile separated for some instruction, especially literacy and during the early years of bilingual instruction

Figure 1. Transdiglossic-balanced two-way dual language bilingual education The question for those of us who live our differences in the United States, of course, is whether the development and spread of transdiglossic twoway dual language bilingual education models is at all possible, given the Anglophone world's hegemonic position in the global order. But at least it is important for those of us who are involved in the education of language minorities to understand and reveal the Oreo nature of many US educational programs - dual, bi, and even multi and trans on the outside, but just as mono as monolingual instruction on the inside. It is also essential that

Lost in transculturation

173

we guard against the potential of these programs to turn the flux of a linguistically shared space into a weapon that may destroy the language other than English that they were meant to expand. We cannot avoid the hybridity of the 21st century, and educational programs certainly must respond to it and reflect it. Two-Way Dual language bilingual education programs are a step in the right direction. But for the languages of the less powerful to survive within our increased hybridity, diglossia is still a useful concept, one that has to be reclaimed, as well as reshaped to reflect the greater dynamism and complexity of language use in the 21st century. The alternative would be to lose ourselves in the powerful languages of a new world order.

Notes * I wish to thank Leah Mason for her careful reading of this article. 1. My 1985 paper with Ricardo Otheguy, "Tending Our Own Garden" (unpublished) already pointed to the danger of trends to include language majority children in bilingual education programs. More recently, Guadalupe Valdes (1997) has raised serious issues regarding dual language programs and language minority children.

References Alexander, N. this volume Anzaldüa, G. 1987 Appadurai, A. 1996

Socio-political factors in the evolution of language policy in postApartheid South Africa. Borderlands/La Frontera: The New Mestiza. San Francisco: Aunt Lute Books. Modernity at Large. Cultural Dimensions of Globalization. Minneapolis: University of Minnesota Press.

Ask, Dame Edna 2003 Vanity Fair, p. 116. Bakhtin, Μ. M. 1981 The Dialogical Imagination. Austin: University of Texas Press.

174

Ofelia Garcia

Benson, C. 2004

Bratt P. C. 2004

Calvet, L.-J. 1999

Bilingual Programs as Educational Development: Access, Quality, Empowerment and Equity. (LAUD Series A: General and Theoretical Paper No. 605). Essen: University of DuisburgEssen. Preliminary Thoughts on Extrinsic Linguistic Minorities. (LAUD Series A: General and Theoretical Paper No. 592). Essen: University of Duisburg-Essen. Pour une ecologie des langues du monde. Paris: Plön.

Cloud, N., Genesee F. and E. Hamayan 2000 Dual Language Instruction: A Handbook for Enriched Education. Boston, MA: Heinle & Heinle. Clyne, M. 2003 this volume Crawford, J. 2000 Edwards, J. this volume

Dynamics of Language Contact. Cambridge: Cambridge University Press. Empowerment through the community language - A challenge. At War with Diversity. US Language Policy In an Age of Anxiety. Clevedon: Multilingual Matters. The power of language, the language of power.

Ferguson, C. A. 1959 Diglossia. Word 15: 325-40. Finlayson, R. and S. Slabbert 2004 Is Instruction in the Mother Tongue Always the Optimal Choice? South African Case Studies with Reference to the Inclusion/Exclusion Debate. (LAUD Series A: General and Theoretical Paper No. 602). Essen: University of Duisburg-Essen. Fishman, J. A. 1972 1976

The Sociology of Language. Rowley, Ma: Newbury House. Bilingual Education. An International Sociological

Perspective.

Rowley, Ma: Newbury House. Freeman, R. 1998

Bilingual Education and Social Change. Clevedon, Avon: Multilingual Matters.

Lost in transculturation Garcia, O. (ed.) 1991 Bilingual Education: Focusschrift in Honor Fishman. Amsterdam: John Benjamins. Garcia, O. 1993

of Joshua

175

A.

Understanding the societal role of the teacher in transitional bilingual classrooms: Lessons from sociology of language. In Bilingual Education in Friesland. Κ. Zondag (ed.), 23-38. Leeuwarden: Gemeenschappelijk Centrum voor Onderwijsbegeleiding in Friesland.

2001

New York's multilingualism: World languages and their role in a U.S. City. In The Multilingual Apple: Languages in New York City, O. Garcia and J. A. Fishman (eds.), 3-50. Berlin: Mouton.

2002

Students of Latin American descent in New York City schools. Keynote address. New York University, November 1, 2002.

2003a

Nouvelles esperances et barrieres dans le domaine de l'education aux Etats-Unis. Hommes et Migrations (France) no. 1246. (Novembre-Decembre): 17-27.

2003b

La ensenanza del espanol a los latinos de los EEUU. Contra el viento del olvido y la marea del ingles. Insula julio-agosto, num. 679-680 (julio-agosto, Spain).

2003c

The politics and policies of Spanish in NY. Keynote lecture given at the Symposium on The Politics of Spanish. Graduate Center, CUNY; April 4, 2003.

Garcia, O. and C. Baker (eds.) 1995 Policy and Practice in Bilingual Education. Extending the Foundations. Clevedon, Avon: Multilingual Matters. Garcia, O. and J. Trubek 1999 Where have all the minority educators gone and when will they ever learn. Educators for Urban Minorities. 1: 1-8. Garcia, O. and J. A. Fishman (eds.) 2001 The Multilingual Apple: Languages in New York City. Berlin and New York: Mouton de Gruyter. Garcia, O. and K. Menken. forthcoming The English of Latinos from a plurilingual transcultural angle: Implications for assessment and schools. In Dialects, Other Englishes, and Education, S. Nero (ed.). Mahwah, New Jersey: Lawrence Erlbaum.

176

Ofelia Garcia

Graddol, D. 1997 Internet Websites

The Future of English? London: The British Council. Instituto Cervantes, www.cervantes.org.

Lindholm-Leary, K. 2001 Dual Language Education. Clevedon: Multilingual Matters. Maurais, J. and M. Morris (eds.) 2004 Languages in a Globalising World. Cambridge: Cambridge University Press. Myers-Scotton, C. this volume Ortiz, F. 1978 Pratt, M. L. 1991 Sefranek, M. 2003 Spolsky, B. this volume

How codeswitching as an available option empowers bilinguals. Contrapunteo cubano del tabaco y el azücar. Caracas, Venezuela: Ayacucho. (Original worked published in 1940). Arts of the Contact Zone. Association of Departments of English. Bulletin, 33^0. Multiliteracies and hybridity. Unpublished paper. Teachers College, Columbia University. Language policy failures.

Torres-Guzman, Μ. E. 2002 Dual language programs: Key features and results. Directions in Language and Education. 14: 1-16. United States Population Census 2000 File 3. Washington, D. C. Valdes, G. 1997

Dual-language immersion programs: A cautionary note concerning the education of language-minority students. Harvard Educational Review 67: 391—429.

Van der Walt, C., D. R. Mabule, and J. J. de Beer 2004 Letting the LI in by the back door: Code-switching and translation in Science, Mathematics and Biology classes. Tydskrif vir Taalonderrig 35, no. 2-3.

Lost in transculturation Walqui, A. 2002

177

Scaffolding instruction for English learners. Conceptual framework for quality teaching for English learners. West Ed unpublished manuscript.

Wolf, H.-G. and H. Igboanusi this volume Empowerment through English: A realistic view of the educational promotion of English in post-colonial contexts: The case of Nigeria. Zentella, A. C. 2001

Spanish in New York. In The Multilingual Apple: Languages in New York City, O. Garcia and J. Fishman (eds.), 167-202. Berlin and New York: Mouton.

Language policies in Spain: Accommodation or alteration? JoAnne Neff-van Aertselaer

1.

Introduction

This paper deals with changes in language policies in Spain from the early 1990s up to the 2004 national elections. It focuses on reactions from the so-called "historic" autonomous communities - the Basque Country, Catalonia (and, recently, Valencia), and Galicia - which differed according to the degree to which their own languages define their unique identity and according to which political parties were in power. Early in the new century, the conservative Partido Popular (PP), then enjoying a large parliamentary majority, passed two education laws, one affecting state control over primary and secondary education, the LOCE (Ley Organicα de Calidad de la Education, the Law for Quality in Education) passed on December 23rd, 2002, and the law affecting state control over tertiary education, the LOU (Ley Orgänica de Universidad, University Organic Law) passed on December 20th, 2001. The control of educational policies, and especially language education policy, has been associated with extended levels of autonomy for minority-language groups demanding a greater measure of self-determination (cf. measures taken in Wales and Catalonia, in May 2001, and by the Sämi parliament in Norway (Marten, this volume). Although education policy alone cannot achieve a language shift in a society (Fishman 1991), support for, or at least tolerance of minority-language policies (Alladina and Edwards 1991: 20; May 2001: 13) has come to be considered central to the preservation and extension of cultural and linguistic rights. Thus the enactment of these laws under the PP administration, considered a strong affront by the historic communities, especially in the Basque Country and in Catalonia, was seen as one more step in a series of obstacles instituted by the PP in order to impose their conservative views on education and to limit the institutional expansion of the autonomous languages, and in the case of the LOU, as an attempt to limit university autonomy, including language requirements for teachers.

180

JoAnne Nejf-van Aertselaer

However, the 2004 national elections have altered the political climate in Spain and, as a result, the central government's relationship with the autonomous communities has changed. The newly elected President, Rodriguez Zapatero of the Spanish Socialist Party, has already suspended some of the provisions of the LOCE1 by modifying the calendar for application of its provisions. In addition, the Zapatero government has stated that it will also reform the application of certain provisions of the LOU. In spite of these very changing circumstances in present-day Spain, there are two major points included in the first versions of this paper that still hold true, and one of them contributed in large part to the major turnover in the victorious election results obtained by the Spanish Socialist Party in the election of March 14th, 2004. The first major point has to do with the way the three historic communities carried out their cultural renovation at the end of the 1800s. These diverse processes have led to differences in the role that language has played in the construction of their identity and the kinds of language policies they were able to establish. As well, the three historic communities were treated differently from the other autonomous communities under the 1978 constitutional provisions. The second point has to do with the Aznar government's (Partido Popular) power struggle with two of the three historic autonomous governments (the Basque Country and Catalonia), from 2000 to 2004. Some political commentators (Aizpeolea 2004: 22) have stated that during this second legislative period, the Aznar government, after blocking further devolutions to the autonomous governments, then set out on a course based on three types of measures: (1) the recuperation of national symbols, with the constant exhibition of the national flag; (2) the accentuation of Spanish nationalism in sport events; and, (3) the discourse on the "unity" of Spain as an answer to the discourse of "uniqueness" of the autonomous communities. While this perspective on the Aznar government's preoccupation with the cultural hegemony of the nation-state is essentially valid, it overlooks the interrelationship between cultural and political features of language (Fishman this volume). In the struggle over the official status of the languages of the historic autonomous communities, the Aznar government took various rather unsubtle measures to obstruct new language policies enacted by the different autonomous communities. Rather than just blocking further devolutions, what Aznar's second-term government did was to begin a very concerted effort to take back from the autonomous governments certain previous devolutions which these governments had used to

Language policies in Spain: Accommodation or alteration?

181

strengthen their distinctiveness, and, in many cases, the conservative government used Spain's European Union (EU) participation to hinder these communities in their demands for direct participation in the EU. During the first Aznar administration (from 1996-2000), these issues were not allowed to come to the fore because the Aznar government needed both the elected governments of the Basque Country (Partido National Vasco) and of Catalonia (Convergencia i Unio) to maintain a parliamentary majority (Fernandez and Laborda 2002). But, during the second Aznar administration (2000-2004), the PP, then with a clear majority, began to be less cooperative with these two historic communities whose support it no longer needed. Thus, while the historic autonomous communities had begun to reformulate their minority status vis-ä-vis the nation-state ideology, the central government began to retreat into an evermore hegemonic stance, especially regarding the institution of new language policies. One of the most active agents in this process, hence, was the Ministry of Education and Culture (MEC), which put forward laws (the LOCE and the LOU) involving the following issues: 1. 2.

Number of hours of teaching of Castilian in comparison to those dedicated to the autonomous languages (the LOCE). The establishment of national evaluation agencies which were to set the agenda for both the evaluation ("Acreditacion") and the contracting ("Habilitacion") of university professors (the LOU).

Each of these measures was meant to undermine the "uniqueness" discourses of the historic autonomous communities and to return control of education to the central government. Clearly, these issues, along with others,2 played a major part in antagonizing the 2004 voters both within and outside of the historic autonomous communities.

2.

National formations and cultural renovation

National minorities display different degrees of self-consciousness which influence the measures of autonomy they demand from central governments, including linguistic claims. The Basque Country, Catalonia, and Galicia all had ratified Statutes of Autonomy during the Second Republic (1931-36). When the new constitution of Spain was approved in 1978, these three autonomies were in a key position to initiate the route toward

182

JoAnne Neff-van A ertselaer

full autonomy. By a transitional disposition, these three communities automatically received the status of autonomous and initiated devolution procedures, while other regions had to comply with a five-year restricted autonomy.4 Through the study of the dimensions of language claims, it is possible to focus on significant differences between Catalonian and Galician nationalism, on the one hand, and Basque nationalism, on the other, and to show how their language claims arise from different socio-historic circumstances during the revitalization periods. Revitalization movements, defined by Wallace (1956: 274) as "movements which represent a deliberate, organized and conscious effort on the part of members of a society to construct a more satisfactory society", include: 1.

2.

3. 4. 5.

6.

Mazeway reformulation: Reformulation of identity markers of a cultural system or field organized by the individual's own experience in order to take on new values and meanings. Communication: New pathways (cultural/linguistic organizations) of communication must be established in order to permit the group to organize and carry out activities. Organization: Charismatic leadership, coupled with, among other resources, the use of language for social mobilization (Fishman 1997). Adaptation: Resistance from within and without must be dealt with. Cultural transformation: Formulation of a group action program, particularly regarding language policies which might link a language variety to stronger cultural and political functions. Routinization: If the group action program in non-ritual spheres is effective in reducing stress-generating situations, it becomes established as normal, i.e. a new steady state is created.

Although all three of the historic communities shared a past history within the same nation-state structure (especially during the Francoist period, in which their languages were heavily repressed, but at the same time, fortified by the conflation of nationalism-democratic struggles against the dictatorship), Catalonia and Galicia have the most similar histories as far as the reformulation of cultural systems and cultural transformation due to group action. Both went through periods of regeneration. In the case of Galicia, one of the most notable initial signs was the 1863 publication of Cantares Gallegos by Rosalia de Castro in a Rexurdimento (Regeneration) period, which began in 1853 and lasted well into the 20th century. The

Language policies in Spain: Accommodation or alteration?

183

preoccupation with the Galician language had come to the fore in the Floral Celebrations of 1861, and with the establishment of the first meeting of the Brotherhood of Galician Speakers (Irmandades da Fala), held in Coruna in 1916. This body of scholars drew up plans for the elaboration of dictionaries and other linguistic studies and also called for the use of Galician in administrative spheres and in educational systems. Catalonia, after a long history as a political entity with its own language (first recorded en 1176, according to Hoffmann 1999), went through a cultural movement, known as the Renaixenga (Renaissance) from about 1840 to the 1870s. Both regeneration movements began as literary movements and gradually spread to other types of regionalist expression. However, Catalonia had the advantage of having greater numbers of speakers and a more evenly distributed population (between rural and urban areas, and especially a greater number of middle-class speakers) who were still fluent in spoken Catalan at the end of the 19th centuiy. Most probably for this reason, as well as the notable support from community leaders, linguistic normalization in Catalonia (Pompeu Fabra, 1913-1932)5 was possible long before the same process was able to take place in either Galicia or the Basque Country.6 In the cases of both Catalonia and Galicia, however, the revitalization movements incorporated language as a principle constituent of identity. On the other hand, the Basque Country, which had retained its local statutes and charters (Fueros)7 until 1876 (Conversi 1997: 46), could not base its revitalization on language, towards which some leaders took an ambivalent position. As Conversi (1997: 53) points out, few modern nationalisms have a history that has been so thoroughly shaped by a single person, Sabino Arana y Goiri (1865-1903). He single-handedly drew up the first political program, founded its first political institution (later to become the Partido Nacionalista Vasco, according to Corcuera 1979), wrote its anthem and designed its flag. However, he was ambivalent when it came to placing the Basque language at the core of Basque identity. Arana carried out a series of studies on the Basque language, but he was well aware that, because so few Basque people actually spoke or wrote the language, insistence on the regeneration of Basque would have to be coupled with other requirements. He believed that these should be ethnic ones and insisted on Basque lineage as a way of ensuring the Basque "race". He thought that because the Basque language is notoriously difficult, this factor would prevent Castilian speakers from learning it, and through this learning, thus further weakening the Basques as an ethnic group.8

184

JoAnne Neff-van Aertselaer

The weak status of the Basque language,9 coupled with the great influx of workers from other areas of Spain, a trend also present in Catalonia during a period of enormous industrial expansion,10 meant that Basque identity suffered more than the Catalan identity. Unlike the Catalans and the Galicians, the Basques had very little resurgence literature to fall back on and much of the native Basque population could not be counted on for language maintenance. These great differences in revitalization movements have meant that both Catalans and Galicians tend toward inclusive policies and, on the other hand, Basque governments have tended toward exclusive policies (Conversi 1997). These various tendencies can be seen in the types of measures put forth by each of the autonomous governments. May (2001: 16) insists that there is an apparent "inexorable association of majority (national) languages with modernity and progress - and, conversely of minority languages with tradition and obsolescence", but this certainly has not been the case of Catalonia, far more advanced in technology and education than many Castilian-speaking areas of Spain, and this is increasingly the case for the Basque Country, although lesser so for Galicia. Nor does this seem to be the case in other European countries, notably Belgium. In any case, the coupling of minority languages with the "un-modern" is not necessarily an "inexorable association".

3.

The devolution process and Spanish education policies

During the devolution process, set in motion by the Spanish Constitution of 1978, the Spanish national government, with the Spanish Socialist Party holding office until 1996, strove to maintain equilibrium, both between the different autonomous communities and between the central government and the individual autonomous communities by retaining the central government's position as broker. For example, the central government established the Inter-Territorial Compensation Fund to maintain financial equity between different autonomies. To assure cohesion in academic requirements, the Ministry of Education (MEC) retained the right to set from 55% to 65% of the national core curriculum, while also granting a certain amount of curricular freedom to reflect local and regional priorities. During this period, the central government was not always in agreement with the measures passed by the autonomous communities, nor was the reverse true. From 1981 to 1991, the central government appealed to the Constitution Court against 120 of the more than 1500 laws (not all dealing with

Language policies in Spain: Accommodation or alteration?

185

language issues) approved by the regional governments, while the latter appealed to the same high court against 127 of the 528 laws approved by the central government (Alvarez Conde 1990, quoted in Haywood 1995: 144). Even though the decentralization process was brought under stricter control by the central government, as a result of the attempted coup of February of 1981 (Nunez Astrain 1995), the Spanish state continued ahead with some important education laws which would affect all of the regions. In 1985, the Socialist government enacted the LODE (Ley Orgmica del Derecho a la Education, Organic Law on the Right to Education) giving Spain's autonomous communities the right to administer their own schools. The LODE also established free, compulsoiy education, and required schools to respect the different languages and cultures of Spain. In a subsequent law, the Ley Orgmica de Ordenaciön General del Sistema Educativo (the Organic Law on the General Organization of the Educational System, known as the LOGSE, taking effect in 1991), the MEC recommended that all educational communities within Spain assign the same amount of time to the teaching of Spanish (Castilian) as to the language of the autonomous communities, in the case that these languages did not coincide, as is true of the Basque Country, Catalonia (along with Valencia and Balearic Islands), and Galicia. Since the passage of the national LODE and the LOGSE, each autonomous community, within the increasing devolution of competencies to the different Communities throughout the eighties and early nineties,11 established its own framework for balancing the number of hours devoted to the teaching of the autonomous language and Castilian at primary and secondary levels. For example, in the Basque Country, pupils in first (6 yrs. old) and second grades (7 yrs. old) spend four hours per week on the learning of language and literature in Castilian and the same amount of time on language and literature in Basque. However, it must be noted that, in the Basque Country, parents may choose the type of school according to their linguistic preference (e.g. Β schools - teaching is carried out in two languages, Castilian and Basque, and D schools - teaching is carried out primarily in Basque, with Castilian as an additional subject). 12 In Galicia, there are fewer choices as far as the language of instruction, although there are provisions for areas, for instance, where more Castilian is needed. In first and second years of primary school, students receive four hours of Galician and four of Castilian and another half-an-hour per week is dedicated to the study of grammatical structures common to the two languages.

186 JoAnne Neff-van A ertselaer The other autonomous communities have similar frameworks. The law in effect up to 2004, the LOGSE, required only that, by the end of the sixth grade (12 yrs. old), the total number of hours of instruction given in Castilian be equal to those given in the autonomous language. It also included a provision under which autonomous communities could dedicate 10% of the schedule to teaching their co-official language. Since the LOGSE did not stipulate in which language the other content areas were to be taught, some autonomous communities were able to increase instruction in their own language,13 as well as to estipulate that their native language be used as the language of administration.14 Additionally, some autonomous communities have passed laws to attend to groups of their citizens who live in border areas and who speak a language other than the autonomous one: Galician, in Asturias, Castilla-Leon and Extremadura; Basque is an official language in the areas of Navarre where it is prevalent; in Aragon, for speakers of Catalan; and in Catalonia, for speakers of Aranese. In the late 1990s, this linguistic panorama has been greatly enriched, and complicated, by an influx of new immigrant minorities (Turell 2001), including minorities from Western and Eastern European countries, those from Black Africa, those from the Maghreb, those from North and South America and those from Asia. As Turell (2001: 18-20) rightly complains, institutional involvement in Spain's increasing linguistic diversity has been extremely lacking in spite of the educational literature which attested to the benefits of plurilingual educational contexts (Nieto 1992; Spolsky 1986). It appears that both the central government and the governments of the autonomous communities have been more concerned with the power struggle in the establishment of a rather broad (and, monolithic) linguistic policy than in attending to the linguistic realities of present-day Spain. If the new law, the LOCE (Ley Organic α de Calidad de la Education, Organic Law on the Quality of Education), had taken effect,15 all Autonomous Communities within Spain would have been required to give ten hours of instruction per week in Castilian and in the autonomous language in first and second grades, thereby reducing the number of hours remaining for the study of other subjects, including a required foreign language (only half-an-hour per week for each of these years). Some communities, such as Catalonia, protested that this new linguistic requirement would mean that either they reduce the number of hours of instruction in their own language, which they were unwilling to do, or they raise the number of hours of instruction in their own language to make it equal to those given in Cas-

Language policies in Spain: Accommodation or alteration?

187

tilian. This latter adaptation would mean, however, that the students in these two years would be spending 47% of their class time on these two languages, which was equally unacceptable. Three Autonomous Communities - Catalonia, Galicia, and the Basque Country - asked the Constitutional Court of Spain to take up the matter. Up until 2001, the universities were regulated by the LRU {Ley de Reforma Universitaria, Law of University Reform), passed under the Socialist government. It gave universities greater freedom to develop their own curriculum and financial plans with their own autonomous governments. It was this law that opened the door for the governments of the historic communities to develop language criteria for university teachers. By 1996, Catalonia, for example, had established a working group of Catalonian universities to draw up accreditation and evaluation criteria. It clearly was working towards the establishment of language requirements for teachers, which did become part of the selection process for university teachers, as set out in the 1998 Law on Language Policy and in the more specific 2003 Catalan University Law (LUC). At the end of 2001, the conservative central government (PP) presented the new Organic Law of Universities (LOU), which established a central evaluation agency. The Catalonian government had already established their own evaluation agency by this time and the Basque government countered the central government by setting up their own agency as well. In Catalonia, the autonomous government established a cultural-linguistic institute (Ramon Llull) to deal with questions outside of Spanish borders, paralleling the Cervantes Institute. If the new LOU had been allowed to continue unchanged, it would have meant that all future university teachers and all professors applying for a contracted or tenured position would have had to attend to the central government's requirements, including having to take the evaluation examination in Castilian. They would then, supposedly, be required to take the evaluation examination of their own communities, some of which would include language requirements, as in Catalonia, for example. These two educational policies, the LOCE and the LOU, were part of the many measures involving linguistic policies16 gradually taken by the conservative government in order to curtail the autonomy of the communities, and especially of the three so-called historic communities. Furthermore, the central government under the PP consistently blocked laws passed by autonomous governments, attempting thereby to prevent these languages from being used for administrative purposes. In September of

188 JoAnne Neff-van Aertselaer 2003, the Balearic Islands government, controlled by the PP, decreed that only a minimum level of Catalan would be required of administration workers; this ruling overturned a previous decree which had set the required level at "C", the highest level. In December of 2003, in Valencia, where the conservatives also controlled the parliament, the autonomous government's Directorate General for Linguistic Policy of the Generalitat Valenciana called for Valencian to be recognized as an independent language from Catalan and the MEC accepted the request.17 The Catalan Language Academy protested that the decision by the Ministry of Education to have Valencian taught separately in the Official Language Schools constituted an attack on the unity of the Catalan language. The central government appears to have been following a strategy of using language policy to "play the claims of one minority group against another" (May 2001: 16).

4.

State policies and regional differences

Throughout the early years of 2000, more conflicts arose due to the divergence between the national "unifying" objectives, as set out by the conservative Partido Popular, and those established by three different autonomous governments to attend to their uniqueness. In the months leading up to the 2004 national elections, state repression played an important role in radicalizing the ideologies of nationalist formations, particularly those of the Basque Countiy and of Catalonia. The central government, controlled by the conservative party, believed that the devolution process had come to a close, especially as far as education policies were concerned. Referring to the autonomous governments, the Minister of Education, Pilar del Castillo, stated in a newspaper interview in December of 2003 that "now that the devolution process is closed, some people think that they should keep asking for more" (Gallego-Diaz 2003: 22). Neither the Catalonian and the Basque parliaments, controlled by nationalist (regional) parties, believed, and presently believe, that the process has yet been completed. They have, therefore, supported measures or passed laws which some, both inside and outside these regions, find unsettling. For example, in 1998 the Catalonian Parliament passed the Llei de Politico Lingiiistica (Law of Linguistic Policy) which stipulated that Catalan is "the sole" language of the Catalonian administration and the language that should be used in all other public institutions (Article 2.2). The law was meant to reinforce Catalan as the predominant language

Language policies in Spain: Accommodation or alteration?

189

within Catalonia. However, the central government believed that such a ruling required a change in the Spanish constitution, which states that both Catalan and Castilian are the co-official languages of this autonomous region. The Galician government, controlled by the PP until 2005, had taken a more ambivalent stance towards closure of the devolution process. It has called for participation of the autonomous governments in European regional plans, but has not carried out coercive programs for the linguistic normalization. The PP Councilor of Education for Galicia (Campuzano 2004) stated that his department has carried out programs to stimulate and motivate speakers to use Galician, trying thereby to avoid more aggressive programs which might create a backlash. Late in 2003, the central government (PP) rectified its initial linguistic policies concerning the LOCE, allowing for the same number of hours given to Castilian and the co-official languages. But it did so only after the Valencian community, controlled by the PP, asked for a revision. This, and other decisions taken in regard to petitions from the Galician community, then also governed by the PP, suggests that if the demand came from a member of their own party, the central government considered rectifying; if not, the autonomous policies were considered "radical". There have been additional differences between communities regarding who is considered a member of the communities. All three of the historic communities have large numbers of Castilian only speakers, and in Catalonia and the Valencian Country, this is further complicated by a sizable influx of immigrants, mostly from the Maghreb, South America and Eastern Europe. Catalonia and Galicia have traditionally had inclusive definitions for autonomous citizens. Maragall, the newly-elected President of the Catalonian Generalitat has frequently declared that Catalan cultural activities are not exclusively those which take place in Catalan; rather, cultural events taking place in Castilian should be considered part of Catalonian cultural activities (Mercator News 2002b). He has also repeatedly asked for Catalan to be co-official in the rest of Spain. In none of these communities, nor in the rest of Spain, is there a concerted effort by the education councils of the autonomous communities to provide early years of education in the first language of the immigrants (Turell 2001: 12-20; also see Clyne this volume). The Basque government, controlled by the National Basque Party, appears to have been adhering to a more exclusive definition of citizenhood, seemingly based on ethnicity. Recently, the Basque Councilor of Education

190

JoAnne Neff-van Aertselaer

provoked criticism from both the Partido Popular and the Socialist Party when she sent instructions to Basque schools in which she used the term "immigrants" to refer to all those students coming into the Basque school system from non-Basque areas, thus equating students from other regions of Spain with immigrants from outside of Spain (Azumendi 2003). The High Tribunal of the Basque Country (Iriarte 2005) has recently ruled that the term "immigration" cannot be used for Spanish citizens coming into the Basque Country from other autonomous communities and that the Department of Education of the Basque Country cannot force newly-arriving students to attend only type D schools (all subjects given in Basque). However, interesting recent changes have come about in the Basque Autonomy with the election of Josu Jon Imaz as substitute for the previous leader of the Partido Nacional Vasco, 18 presently controlling the Basque Parliament. In recent statements made in Basque, Castilian and French, Imaz has opted for "civic identity" for the Basque Country rather than pertinence to a supposed Basque ethnic group. It appears that he wants to take a different stance regarding non-Basque people entering into his community, a policy which the Basque government would have to adopt if it were to gain European acceptance. All of the historic communities have called for more direct participation in the Assembly of European Regions. Each community would team up with provinces of neighboring countries, with linguistic affinities: Galicia with Portugal, Catalonia with some Pyrennese areas and some areas of southern France, and the Basque Country with the Northern Basque Country (France). With the newly-elected Socialist Party now in power, there have been some dramatic changes in the central government's linguistic policies. In April of 2004, President Rodriguez Zapatero showed his support of all of Spain's official languages by stating that he will ensure that the Constitution be translated into Basque, Catalan, and Galician. In June of 2004, he asked the EU to consider Basque, Catalan and Galician as Treaty languages. The Spanish Senate has also taken steps towards changing its ineffective statutes in order to turn itself into the representative legislative body for regional affairs, and has said that the three historic communities' languages can be used in its sessions. It remains to be seen if Spain's legislative bodies will grant Catalan, or the other two historic communities' languages, co-official status in all of the Spanish territory, or if these communities will be able to exercise political representation internationally, as the President of the Catalan community has requested.

Language policies in Spain: Accommodation or alteration? 5.

191

Conclusion

In the preceding sections, I have pointed out the importance of language and identity in political discourse between the central government and three historic autonomous communities of Spain, especially in regard to the educational policies adopted by each of the governments. The political climate in Spain has changed, but this does not mean that all of the problems of sovereignty have disappeared. Rodriquez Zapatero's government has already stated that it will not consent to important constitutional amendments, as requested by the Basque government, in order to allow them to be a "free-associated" state. It is also patent that the accelerating migratory trends will affect linguistic and other educational policies (Turell 2001). The Catalan government's linguistic survey for 2003 (constrastant.net) found that 17% of the 7257 people interviewed did not speak Catalan at home. In previous years, this government has commented that of the many immigrants received in this community, most were from Morocco, making linguistic integration more difficult, while the majority of Latin American immigrants chose the Madrid area. Nevertheless, given the scale of the immersion programs in Catalonia, it does appear that what Fishman (1991) has suggested - the reversal of language shift, with Catalan as Η and Castilian as L - will be possible. The use of Basque continues at a much slower pace and it is not primarily transmitted from parents to children, although it is now being used in some pre-primary and primary schools. In July of 2004, Behatokia, the Basque Language Observatory for Linguistic Rights, presented its 2003 annual report on the situation of language rights in Euskalherria, i.e. Basque-speaking areas of Spain and France (Mercator News 2004a). The report, based on an analysis of 737 cases, states that language rights are systematically infringed upon at all levels of public administration. In order to counter this state of affairs, the Basque government has just passed its Third Normalization Language Plan (Mercator News 2004b) aimed at making Basque the working language of more local administrations. But, this will affect only 50 of the 527 administrative units. Adult education in Basque is increasing and there are some pre- and primary schools using Basque as the language of instruction; although 62.8% of the students/parents have chosen to attend Model D school (with Basque as the principle language and Castilian as one subject), there is still limited learning of the language at home and limited use of Basque for everyday social/commercial interaction (Laza 2005).

192

JoAnne Neff-van Aertselaer

Galician seems to be making progress, even though in response to the Galician government's policies, there have been many complaints concerning the ineffectiveness of normalization practices. 19 A 1991 survey showed that 91% of the inhabitants understand Galician and 84% speak it, but only 48% use it all the time (eurolang.net 2004). One of the remaining problems concerns the establishment of official orthographic conventions for Galician. Another important issue which may affect linguistic policies is the rising importance of the Assembly of Regions within the European Community. All three of the autonomies discussed in this paper are eager to participate in a "Europe of regions" and have already participated in the Committee of the Regions. This is one way to reduce the power of the central government and this route may provide a non-violent course to more self-government and international participation, which is much needed in present-day Spain.

Notes 1. Various provisions of the LOCE were suspended on May 29th, 2004. The subject of Religion (in reality, only the Catholic confession) will continue to be voluntary and not be counted in students' grade point average; the LOCE would have made Religion compulsory (or, optionally, a Study of Religions course) and would have been counted as part of the students' school record. Students will not be required to choose just one field of study at age 14, as they would have under the LOCE. 2. The central government used the Department of Justice to close down some Basque newspapers and publishing houses, and also to outlaw the Batasuna Party (with 142,784 votes over a total Basque census of 3,613,205 voters in the most recent elections). There has also been heavy state propaganda against different phases of cooperation among the three historic communities, especially if the Basque Nationalist Party was involved. 3. In the case of Galicia, the Statutes (1931) never took effect because of the Civil War. 4. Navarre was also given autonomous regional status. 5. Pompeu Fabra Poch, trained as an engineer, enjoyed a long history of publications, for which he is credited with the normalization of Catalan: 1891, Essay on Modern Catalan Grammar, 1898, Contribution to the Grammar of the Catalan Language·, 1904, Treatise of Catalan Orthography, 1912, Grammar of the Catalan Language·, 1913, Norms of the Institute for Catalan Studies', 1917, Orthographic Dictionary·, 1918, Catalan Grammar and Intermediate Course of

Language policies in Spain : Accommodation or alteration?

193

Catalan Grammar, 1925, Orthography of Catalan·, 1930, Compendium of Catalan Grammar, 1932, General Dictionary General of the Catalan Language; and, 1956, New Grammar of Catalan. 6. In Galicia, there still seem to exist two different tendencies (Martino Paz 1999): those who see Galician as a dialect of Portuguese and therefore believe that there should be linguistic intervention in order to aid speakers and writers (especially as far as vocabulary) to come to a possible convergence with Portuguese in the distant future, and, those who believe that Galician and Portuguese are related languages (from Galaico-Portuguese), but have developed along different lines since the separation of Portugal from Spain in the 12th century. 7. Mostfueros were codified in the 17th and 18th centuries and established direct agreements between the Spanish crown and Basque regional powers (Conversi 1997: 45). 8. Arana (1965: 627-628) is often criticized for having tried to mobilize ethnic concepts with statements such as the following: "The Vizcayans are obliged both to learn their own language as well as not to teach it to the "maketos" [Castilian descendents].... the difference in languages provides a great means for saving us from contact with the Spanish and thereby avoiding the mixing of the races." 9. As Conversi (1997) has noted, unlike the Catalans, the Basque elite did not speak the Basque language, which was looked on as the language of backward rural areas. There was, thus, a certain class element in the acceptance or rejection of Basque, which no longer seems to be the case. 10. Industrial expansion took place from the 1880s onwards for the Basques, and for the Catalans, from the first third of the 19th century (mNACTEC ). 11. Seven of Spain's 17 autonomous communities (Andalusia, the Basque Country, the Canary Islands, Catalonia, Galicia, Navarre and Valencia) have almost full control over the administration of the education systems within their territories. 12. For a similar experiment in the Aran Valley, with the parents choosing between Catalan, Castilian Spanish and Aranese as a primary language of instruction for pre-primary and primary children, see Suils and Huguet 2001. 13. In Galicia, by law, each student at the age of 8 to 16 must have from two to four subjects with Galician as the language of instruction. In Catalonia, by 1990 about 30% of all pupils were receiving their education entirely in Catalan, and this percentage is now even higher, due to increased use of the language by those born in Catalonia and by recent immigrants. In 1994, more than 130,000 children participated in the Catalan Immersion Program, particularly in schools located in predominantly Spanish-speaking areas, mainly in suburban Barcelona (Mercator News 2002a). 14. In 1986, the Balearic Parliament passed the Linguistic Normalization Act (3/1986) whose principal objective was to make Catalan the normal working language of administration and the normal medium of instruction in schools.

194

JoAnne Neff-van Aertselaer

15. The LOCE would have entered into effect for first yr. of primary education during 2004-2005 and for second yr., third yr. and fifth yr. during 2005-2006, and for the remaining two years of primary (fourth and sixth yrs.) in 20062007. 16. Other strategies used by the PP to foment the disappearance of the minority languages involved having certain communities lessen the linguistic requirements for the use of these languages in public administration (the case of Navarre's failed attempt to outlaw Basque in the northern border areas of this community) or trying to encourage internal dissension with a language group (the case of supporting Valencian as a language separate from Catalan). 17. It must be noted, however, that the question of Valencian as a separate language has been an issue of contention since at least 1997, when the Constitutional Court ruled that the Valencian and Catalan are the same language, thus overturning a previous Supreme Court ruling which ordered the University of Valencia to refer to the language solely as "Valencian" (Mercator News 1997). 18. It was thought that the previous party leader, Javier Arzalluz, adhered more closely the Sabino Aranas' definition of basquehood based on ethnic grounds. 19. A study of non-normalized languages in Europe found that there is generally a non-observance of the laws in the field of education, a situation which "is not producing new Galician-speakers". The study states that in both primary and secondary schools, the legislation in force is not being observed. In addition, it noted that there were great differences between rural and urban areas, with less use in the latter areas (Mercator News 2001).

References Aizpeolea, L. R. 2004 El Gobierno derogarä la reforma legal contra el referendum del 'lehendakari'. El Pais (2nd of April: 22). Alladina, S. and V. Edwards (eds.) 1991 Multilingualism in the British Isles 1. London and New York: Longman. Arana, S. 1965 Azumendi, E. 2003

iQue somos? In S. Arana, Obras completas. Buenos Aires: Editorial Sabindiar-Batza, 627-628. PP y PSE presentan mociones de censura contra la consejera vasca de educacion. El Pais (1st of November: 26).

Language policies in Spain: Accommodation or alteration?

195

Campuzano, A. En Galicia la educacion es muy cara, [interview with Celso 2004 Curräs, Consejero de Educacion de Galicia]. In Boletin delIlustre Colegio Oficial de Doctor es y Licenciados en Filosofia y Letras y Ciencias 151: 2—5. Clyne, M. this volume

Empowerment through the community language - A challenge.

Constrastant.net 2004 Sobre les dades de l'Estadistica (www.contrastant.net, 16th of May). Conversi, D. 1997 Corcuera, J. 1979 Eurolang.net 2004

d'usos linguistics

2003.

The Basques, the Catalans, and Spain: Alternative Routes to Nationalist Mobilisation. London: Hurst & Company. Origenes, ideologia y organizacion del nacionalismo vasco, 1876-1904. Madrid: Siglo XXI. Linguistic survey: Spain, (http://www.eurolang.net).

Fernandez, T. and J. J. Laborda 2002 Espaha. ιCabemos todos? Madrid: Editorial Alianza. Fishman, J. A. this volume

Sociolinguistics: More powers to you! (On the explicit study of power in sociolinguistic research)

1991

Reversing Language Shift. Clevedon: Multilingual Matters.

1997

Language and ethnicity: A view from within. In The Handbook of Sociolinguistics, F. Coulmas (ed.), 327-343. London: Blackwell.

Haywood, P. 1995 Hoffmann, C. 1999

The Government and Politics of Spain. London: Palgrave. Language, autonomy and national identity in Catalonia. In Whose Europe? The Turn towards Democracy, D. Smith and S. Wright (eds.), 48-78. Oxford: Blackwell.

Gallego-Diaz, S. 2003 Ocho anos no es mucho para equipos de gobierno [interview with Pilar del Castillo, Minister of Education], El Pais (1st of December: 22).

196

JoAnne Neff-van Aertselaer

Iriarte, M. 2005 Laza, R. 2005

El TSJPV niega que sea inmigrante el desplazado de una region a otra. El Mundo (18th of March: 12). La vertiginosa implantation de euskara. El Mundo (2nd of April: 12).

May, S. 2001

Language and Minority Rights. London: Longman.

Martiflo Paz, R. 1999 Historia da lingua galega. Santiago de Compostela: Sotelo Blanco. Mercator News 1997

The Constitutional Court recognises for the first time the unity of the Catalan language, (http://www.ciemen.org/mercator/indexgb.htm, April).

2001

The Galician normalization law is not being implemented. http://www.ciemen.org/mercator/index-gb.htm, November).

2002a

The Catalan language in education in Spain. (http://www.ciemen.org/mercator/index-gb.htm, November).

2002b

The main Catalan political party in the opposition criticizes the Catalan model of linguistic policy, (http://www.ciemen.org/mercator/index-gb.htm, June).

2003

Spain's Ministry of Education challenges the unity of the Catalan language, (http://www.ciemen.org/mercator/index-gb.htm, December).

2004a

Report on language rights in Euskal Herria presented. (http://www.ciemen.org/mercator/index-gb.htm, July).

2004b

Basque government presents the third normalization language plan, (http://www.ciemen.org/mercator/index-gb.htm, August).

Nieto, S. 1992 Nunez, A.L. 1995

Affirming Diversity: The Sociopolitical Context of Multicultural Education. New York/London: Longman. La razon vasca. Navarra: Txalaparta.

Language policies in Spain: Accommodation or alteration? Spolsky, B. 1986

197

Language and Education in Multilingual Settings. Clevedon: Multilingual Matters.

Suils, J. and A. Huguet 2001 The Occitan speech community of the Aran Valley. In Multilingualism in Spain, Μ. T. Turell (ed.), 141-164. Clevedon, UK: Multilingual Matters. Turell, M. T. (ed.) 2001 Wallace, A. 1956

Multilingualism in Spain. Clevedon, UK: Multilingual Matters. Revitalization movements. American Anthropologist 281.

58: 264-

The potential of parliaments for the empowerment of linguistic minorities: Experiences from Scotland and Norway Heiko F. Marten

1.

Introduction

Numerous academics and politicians have in recent years contributed to the description and analysis of language policy for the benefit of smaller languages. The present paper tries to add to these by taking up the notion of yet another aspect of politics and language, exemplified by two case studies. The political aspect is the decentralization of parliamentary power for the benefit of minority languages. The two case studies deal with the relationship between the Scottish Parliament and the Gaelic language on the one hand, and between the Norwegian Sämi Parliament, the Sameting, and the Sämi language on the other hand. The underlying idea is to consider whether parliamentary bodies may contribute to the empowerment of speakers of minority languages regarding the language of individual choice in as many instances as possible. This applies to any domain of language use, but in particular public bodies, education, and the media, at local, regional and national levels. In line with traditional power structures of minority-majority relations, Gaelic and Sämi speakers are mostly tied to a general pattern in which speakers are obliged to choose the dominant language (English or Norwegian) if they wish to participate in higher levels of power. The choice of Gaelic or Sämi is tantamount to remaining at the margins. Perceptions by speakers such as "you get nowhere with Gaelic" or "I suffered from speaking Sämi, I do not want my children to go down the same road" continue to be detrimental to attitudes towards the languages and their intergenerational transmission. These attitudes are frequent both among speakers of Gaelic and Sämi, and within the majority population. Empowerment of linguistic minorities thus intensely deals with overcoming the notion that English or Norwegian provide potential access to power and resources,

200

Heiko F. Marten

whereas sticking to Gaelic or Sämi implies restrictions and limited access to participation in decision-making. Fishman (2001) suggests a graphic representation of the use of a dominant and a threatened language in a society. Domains of language use are broadly classified into those associated with power (P), to which parliaments undoubtedly belong, and less powerful domains (n-P). The traditional situation for Gaelic and Sämi speakers would see English and Norwegian as the languages used in legislative bodies. In the Fishman notation, this would be symbolized as: Ρ n-P

English, Norwegian Gaelic, Sämi

Empowerment of Gaelic and Sämi speakers would imply an elevation of Gaelic and Sämi from n-P to Ρ domains, i.e. from the lower to the upper part of the figure. Yet, as it is hard to imagine that Gaelic and Sämi speakers are easily empowered to use their less dominant languages in any powerful context, the aim should be, initially, to achieve a position of shared functions of the dominant and non-dominant languages in the powerful domains. To aim for more, i.e. the unlimited use of the minority languages in any Ρ situation, should be, more realistically, a long-term goal. The graphic situation to aim for at first should therefore be: Ρ n-P

English/Gaelic, Norwegian/Sämi Gaelic, Sämi

Throughout this paper, we will thus keep in mind questions such as: How can parliaments contribute to the empowerment of minority language speakers? How have parliaments enabled speakers to choose Gaelic or Sämi and yet be able to participate at high political levels? And how have power structures and access to power changed through language policy initiated by new decentralized parliamentary bodies? A question resulting from this is how the development of minority language policy depends on the type of parliamentary body. As I will show, both the Gaelic and the Sämi models of decentralization may contribute to minority language speakers' empowerment in relation to the dominant language group.

The potential ofparliaments for the empowerment of linguistic minorities

201

The following discussion is divided into four parts. In the next section, I will comment in more detail on the functions of parliaments and their implications for linguistic minorities. I will then address in turn the Scottish Parliament and its impact on Gaelic policy and the Sameting in Norway and Sämi language policy. The final section will try to illuminate the question of how parliamentary bodies may contribute to the empowerment of linguistic minorities.

2.

The importance of (decentralized) parliaments for linguistic minorities

Let us now reflect in more detail on the way in which parliaments are of importance for linguistic minorities. Spolsky (2004) explicitly links the discussion of language and power to political entities, mostly states or their sub-divisions. Parliaments as central political institutions in a state provide a chance for minorities to participate in power. According to Hague and Harrop (2001), functions of legislatures are, broadly speaking, representation, deliberation/debating, legislation, authorizing expenditure, and the making and scrutinizing of governments. In the following we will see that these functions are all of potential relevance for minority language speakers. Concerning representation and deliberation, there is a high symbolic value of multilingualism in a parliament to bridge the abstract concept of a political unit and the individual. Parliaments are in this way one piece of a mosaic of items contributing to personal linguistic identity. Only when the population feels genuinely represented in such an institution does the political entity which the parliament refers to have a positive value. The presence of minority issues is thus highly symbolic in the political system through representation in a parliament, which can guarantee a certain degree of awareness of minority issues in a state. Broderstad (1995), in the context of the Norwegian Sameting, discusses three core functions of parliaments directly related to the population. First, parliaments speak on behalf of the various groups within the political entity. Second, they are representative, i.e., they display the structure of the population according to certain criteria such as age, class, gender, or language. These two functions are generally in line with the Hague/Harrop function of representation. In opposition to these, there is a third function, the responsiveness to wishes and demands and the promotion of these at different

202

Heiko F. Marten

layers of decision-making. Turan (1994) looks closer at this responsiveness of a legislature, and identifies four dimensions in which this may occur: policy, service, allocation, and symbolism. Symbolic responsiveness implies a focus on attitudes towards the legislature, rather than the behaviour of the legislators per se. The other three dimensions closely relate to the Hague/Harrop functions of legislation, expenditure and government. Language policy making is linked to legislation as the basis of policy making. Authorizing expenditure is related to the provision of services and the allocation of means for the funding of language planning projects. The making and scrutinizing of governments is, finally, of potentially least practical relevance for minority language speakers - there must be a considerable degree of strength to directly influence the formation of a government. In a similar line of argumentation, May (2001) also discusses parliaments as central institutions which canalise the will of the people and their relation to linguistic minorities. In the persisting political structure with nation-states as the fundamental unit of political action, parliaments have kept their dominant position as the centres of pluralist policy-making. In order to enable minorities to participate adequately in power structures, May calls for a rethinking of nation-states as multicultural. Linguistic minority groups should have distinct rights in the form of self-government or special representation (i.e., at least adequate representation, one might add). Both legitimization of a language through official legislation and institutionalization through its regular use in official bodies "highlight the power relations in situations of minority language shift" (May 2001: 146). A multicultural solution would thus imply empowerment of minority language speakers through participation and an increase in choice opportunities. The ultimate aim is that language varieties different to the dominating one would become "normalized", i.e., taken for granted in any context, including parliaments. Summarizing, we can thus broadly assign two relevant components for linguistic minorities to parliaments. On the one hand, they are a highly symbolic part of the domain of public bodies, where the empowerment of minority language speakers lies in the representation and the creation of awareness through the fact that their language is present in one of the most observed state institutions. On the other hand, through its language policy and planning potential, it is a major component in the exertion of power, and as such a channel of empowering minority language speakers if they adequately participate in decision-making, in particular concerning legislation and the distribution of means.

The potential ofparliaments for the empowerment of linguistic minorities

203

The question to follow from this is how a maximum of awareness and of influence on decision-making can be achieved. One evident option is the decentralization of parliamentary power. The Encyclopedia of Democracy (Lippert 1995: 335) defines decentralization as "the devolution of power in a state to elected local authorities", which in practice is closely linked to democratic principles. This delegation of power to subordinate bodies implies that decision-making is distributed more widely and brought closer to the point of service or action. Applied to linguistic minorities, it is obvious how decentralization is for their benefit: the likelihood that minority members are part of a parliament increases as the political entity and its population decreases in size. In other words, it is much more likely that, for instance, Gaelic speakers would be represented in the Scottish Parliament which stands for roughly 5 million people, than in the Westminster Parliament in which over 50 million people are represented. This change in presence of minority issues affects both the representation of minority language speakers and the potential for policy-making. In recent years, we have become used to such decentralization processes in democratic societies. For instance, the changes in political structure in Spain since the 1970s have had undeniable impacts on the support for languages such as Catalan and Basque (see also Neff van Aertselaer, this volume, on current changes in language policy after the government change in Spain in April 2004). Another example is the Nunavut Territory in the North of Canada. Established as a new political unit in 1999, it serves a largely homogenous Inuit population as a means of decentralized power. Their main language, Inuktitut, is used as a language of government, and administration, including the regional parliament. Inuktitut speakers have thus received a considerably higher degree of power in regional government affairs (cf. http://www.gov.nu.ca/cley/englisManguage.htm). These are two very clear examples of how political decentralization may result in a strong increase in language policy efforts and support of minority languages. In both Scotland and Norway, there are new parliamentary bodies as players in language policy in the political structures. Representation of minority language speakers and responsiveness to their wishes have thus received a chance of gaining more attention than previously. In the following two sections, we will see how these chances have been used.

204

Heiko F. Marten

3.

The Scottish parliament: Its impact on Gaelic policy

Let me now come to the Scottish Parliament and its impact on Gaelic policy. The Scottish Gaelic language has, after centuries of decline and a public policy ranging from ignorance to active suppression, experienced a strong increase in support since the 1980s (cf. MacKinnon 2000). However, speaker numbers are still declining, to a level of just below 60,000 in the most recent census of 2001, and the demographic position of Gaelic remains extremely weak. After the establishment of the decentralized Scottish Parliament in 1999, hopes were held that this would also have an impact on the chances of Gaelic to prosper (cf. McLeod 1998). Westminster

National

Regional Regional Councils Cooperation on Gaelic Issues Presence of Gaelic issues

Figure 1. The position of Gaelic issues in the Scottish political structure Figure 1 shows the presence of Gaelic issues in institutions within the political system of Scotland. They are dealt with to varying degrees by both regional councils and the Scottish Parliament, as well as the Scottish Executive. Circles in the figures vary from bodies where Gaelic is not dealt with at all to bodies which devote a substantial proportion of their work to Gaelic. In Scotland as a whole, Gaelic is an issue in the Scottish Parliament and the

The potential of parliaments for the empowerment of linguistic minorities

205

Scottish Executive, even though, not surprisingly, the role of Gaelic is fairly small. As opposed to the Scottish Parliament, the Scottish Executive as a successor of the Scottish Office, which was responsible for Scottish affairs on behalf of the government in Westminster, is no truly new player in the system. Yet, originating from the Scottish Parliament, it is much more likely to respond to the Scottish population's wishes. At a position above the Scottish level, a number of issues, most prominently broadcasting, remain under the control of the Westminster Parliament. Here, Gaelic plays an even smaller role than in Scottish national bodies, coming close to near nonexistence. The centre of attention in this paper, however, shall be the Scottish Parliament as the new institution. In its proceedings, Gaelic issues have played a role in numerous contexts. In fact, support for Gaelic is quite common among the Members of the Scottish Parliament (MSPs). It is important to note that this support is independent of political affiliation: Members of all political parties have spoken in favour of Gaelic at various occasions. Symbolic instances of Gaelic presence are the bilingual English-Gaelic signage in the Scottish Parliament building, the possibility of using Gaelic in communication with the parliament, and that Gaelic is indeed used - albeit at a Table 1.

Major events relating to Gaelic in the first year of the Scottish Parliament and the Scottish Executive's work

Date

Event

1999 October

Deputy Mnister for Gaelic at the Royal National Mod Festival: the first Gaelic speech of a government minister

November

Agreement to appoint Gaelic Officer for Parliament

December

Announcement of Gaelic Task Force

2000 February

Announcement of Gaelic parliamentary and legal dictionary

March

First debate in Gaelic (on Gaelic issues) in a parliament in modern times

March

Education, Culture and Sport Committee meeting on Gaelic parents' organization petition

March

Increased funding for Royal National Mod announced

June

First meeting of cross-party Gaelic group

206

Heiko F. Marten

very limited number of occasions - in plenary debates and committee meetings. Given its symbolic presence in the Scottish Parliament right from the start of operation, the beginning of the decentralized era was quite encouraging for Gaelic. This view is supported by data seen in Table 1, which lists a selection of occasions in which Gaelic played a role in the first year of the Scottish Parliament's work, between October 1999 and June 2000: As the table indicates, within a few months after the Scottish Parliament had been established, Gaelic featured regularly at various levels. These included both symbolic steps to represent Gaelic in the Parliament, as well as concrete responses to the Gaelic communities' demands through policy preparation, such as the inauguration of the Gaelic Task Force to prepare a report of the present state of Gaelic, and the announcements of increased funding. The Deputy Minister for Gaelic's speech at the Royal National Mod Festival in October 1999, the major annual Gaelic Cultural Festival, was the first modern Gaelic speech of a government minister. A month later, the Scottish Parliament agreed to appoint a Gaelic Officer. On March 2nd, 2000, the Scottish Parliament saw the first Gaelic debate in a parliamentary institution in modern times. After the first year of operation, it could thus legitimately be described as pointing towards a coherent Gaelic policy (cf. Marten 2004). However, when looking at what has happened since then, we find that initial hopes did not hold in the years that followed. In fact, there were a number of major disappointments for Gaelic activists. One first major setback in 2000 was the extremely unsatisfactory debate on the role of Gaelic in the Education Act launched by the Scottish Executive. A right to Gaelicmedium education was denied, despite strong lobbying by Gaelic activists and support by a number of MSPs. In the years 2000 and 2001, several reports on Gaelic that had been commissioned by the Scottish Executive were published. The Gaelic Task Force Report took account of the sad state of affairs of Gaelic. The Gaelic Broadcasting Report was bound to show little influence, given that broadcasting belongs to the policy fields still under the responsibility of the UK government. The report "A Fresh Start for Gaelic", followed by the "National Plan for Gaelic" in 2002, identified concrete policy issues as suggestions for a holistic reversal of previous Gaelic policy. However, this possible kick-off failed to generate much follow-up. After publishing these reports, the necessary political devotion to bring about substantial change was lacking. As a result, in November 2002,

The potential ofparliaments for the empowerment of linguistic minorities

207

an opposition MSP, who had argued constantly in favour of Gaelic measures, dealt with his impatience with the Scottish government by introducing a Gaelic Bill to the Scottish Parliament. This Bill was, however, rejected by the Executive. The most concrete achievement of the period was the establishment of Bord na Gaidhlig, the Gaelic Language Board. As an institution for the development, coordination, advice and monitoring of Gaelic policy initiatives, its major aims are to increase speaker numbers, to strengthen Gaelic in families and communities, and to extend the use of and access to the Gaelic language in any aspect of life across Scotland. Its strategic priorities lie in the fields of education and learning, culture and arts, and economic and social development (cf. Bord na Gaidhlig Web Site, at: http://www.bord-na-gaidhlig.org.uk). The establishment of this central planning agency in itself was certainly an important step forward. However, problems arose concerning the restriction of powers to advisory functions instead of enforcement competence, and the lack of a statutory basis. Furthermore, the appointment of the Board's members caused disappointment, as many candidates whom activists would have liked to see as members were rejected. At the end of the first term of the Scottish Parliament in May 2003, there had been some promising efforts for Gaelic, but there was still no coherent policy. In the subsequent election campaign, the ruling Labour Party promised that, if re-elected, it would introduce a Gaelic Bill within the first six months of the new term, in other words, by autumn 2003. Since the government indeed managed to be re-elected to another term in office, we may ask if some type of Gaelic legislation has been achieved by now, after more than two further years. A draft bill, published by the Executive in October 2003, was discussed by the Scottish Parliament with the participation of interest groups and scholars throughout 2004. This process led to the introduction of a Bill to the Scottish Parliament in September 2004, which was debated in the parliament's institutions until it finally passed unanimously as Gaelic Language Act in April 2005 (see http://www.scotland.gov.uk/Topics/ArtsCulture/gaelic/17912/12316 for submissions by the public to the debate on the Bill, and http:// www.scottish.parliament.uk/business/bills/billsnotlnProgress2/gaelicLanguage.htm for the Act and the relevant proceedings in the Scottish Parliament). However, activists and opposition MSPs share the view that the Act is much too weak. It establishes Board na Gaidhlig by statute and requires it to develop a National Gaelic Policy. One of the major aspects is that the Board may demand of public bodies to introduce a Gaelic

208

Heiko F. Marten

policy. There is, however, no true obligation - if a public body rejects such a step, there is little the Board can do. Bodies which did not wish to support Gaelic in the past are thus very unlikely to change their attitude in the future. This lack of a rights-based approach also applies to Gaelic education. Further, several important fields of language planning remain untouched, e.g. the use of Gaelic broadcasting. As a whole, the Language Act is a step forward, but as it stands, its impact on helping the language to survive will be limited. Table 2 summarizes major events in Gaelic policy from 2000 to 2005: Table 2, Major events in Gaelic policy 2000-2005 Year

Gaelic Policy Event

Result/Evaluation

2000

Debate on the role of Gaelic in the Education Act

Unsatisfactory: No right to Gaelic-medium education

2000

Gaelic Task Force Report

Account of the sad state of affairs

2001

Gaelic Broadcasting Report

Some, but not sufficient commitment to Gaelic

2002

Report "A Fresh Start for Gaelic" and the "National Plan for Gaelic"

Suggestions for a possible kickoff with little follow-up

2002

Establishment of the Gaelic Language Board

Only restricted powers granted

2002

Private Member's Bill in 2002

Rejected by the Executive

2003 2004

The 2003 Bill by the Scottish Execu- A step into the right direction, but lacking ultimate dedication tive

2005

Gaelic Language Act passed in April An important step forward, but too weak

All in all, we find that Gaelic has indeed become a much more important part of the agenda of public bodies through the impact of the Scottish Parliament, albeit to varying degrees. The Scottish Executive has established a policy that supports Gaelic to a degree formerly unknown in the UK, and attention towards Gaelic and the steps taken have improved its situation. The existence of the Scottish Parliament has led to more pressure being exerted and has generally increased the amount of attention being focused on Gaelic. Most activists agree that it is very doubtful that the responsible political authorities would have given so much attention to Gaelic issues

The potential ofparliaments for the empowerment of linguistic minorities

209

without the Scottish Parliament. The symbolic presence of the language in the Parliament supports the awareness of the Gaelic heritage in Scotland, and Gaelic speakers may now feel represented at the highest level of politics in democracy. The recent establishment of the Gaelic Language Board points in a direction which might result in a coherent policy. However, this careful and positive approach is undermined by a number of disappointments. Bord na Gäidhlig's role could still be strengthened. The Gaelic Bill introduced by the opposition was rejected by the government. Despite the general consensus that the Executive Bill will give Gaelic a certain degree of official recognition, the proposal has met considerable disappointment. Many Gaelic activists feel that much of the policy of the new administration only constitutes lip service. Many parts of Gaelic policy would need to be framed in terms of obligation instead of mere recommendation. Policies have to be much more binding for authorities. The position of Gaelic is still far from safe and its decline is unlikely to be stopped by the measures introduced. As a whole, despite some positive changes, views among activists on language maintenance still only range from pessimistic to very sceptically optimistic.

4.

The Sameting in Norway: Its impact on Sämi language policy

Let us now move to the second case study, the Sämi language in Norway. The position of the Sami language is similar to the position of Gaelic, in particular concerning its century-long suppression and a revival of interest in the language in the past decades. Also the number of speakers is similar; estimations suggest roughly 30 to 40,000 speakers. Säpmi, the traditional settlement area of the Sämi, is spread over four countries, namely Norway, Sweden, Finland and Russia. The largest Sämi group by far lives in Norway, and we will therefore look at the situation in Norway as a case study (cf. Magga 1995). It will soon become clear that the Sämi situation in many respects is a model to follow, especially when looking at Arctic neighbours outside Scandinavia. Norway, in the past years, has also seen a political decentralization, at least with regard to the Sami. If we look at the position of Sämi issues in the Norwegian political system, we again see that a number of different political levels are involved. Figure 2 displays Sami issues in the system, similar to the corresponding figure relating to Scotland. Also similarly to the situation in Scotland, Sämi plays a varying role in several of the numerous local

210

Heiko F. Marten

and regional bodies, in the Norwegian Parliament, the Storting, and the Norwegian government's departments. In addition to the familiar system of political stratification, however, there is the Sameting, the Sämi Parliament, as a body elected by the Sämi population and dealing with Sämi issues only. The Sameting is outside the traditional system, and it cooperates with political institutions at all levels. Ending a decade-long process of minority rights debates in Norway, the Sameting was opened as a body of political representation for the Sämi population in 1989. Here, Sämi demands are democratically legitimized and centrally administered. It thus empowers Sämi speakers by giving them an institutionalized voice (Figure 2). The chances of choosing to use Sämi in public institutions have thus increased - and the Sameting itself coined the phrase "to manage in Sämi" in any context as an aim of a normalizing language policy. Storting

Government

Sameting

National

Regional Councils

\

Regional

Local Municipality Councils Cooperation on Sämi Issues Presence of Sämi Issues

Figure 2. The position of Sami issues in the Norwegian political structure

Yet, there is also a fundamental difference with regard to the impact of constitutional changes in Norway to those in Scotland. The composition of the government in Norway remains unaffected by the Sameting. The Sameting is similar to the Scottish Parliament in its democratic representations and

The potential ofparliaments for the empowerment of linguistic minorities

211

closer responsiveness to people's demands. Concerning the legislative function, those who eventually take the fundamental decisions (i.e. the Scottish and Norwegian governments) are influenced in very different ways: directly by the Scottish Parliament which elects the Scottish Executive, but only very indirectly by the Sameting, which has only gained influence as an institutionalized and democratically legitimized channel of lobbying. As one of its first major political projects, the Sameting helped the Storting to develop a Sämi Language Act which came into force in 1992. Through coordinated decision-making in the Sameting, Sämi speakers have thus been empowered by gaining influence on the shaping of fundamental linguistic rights, most notably through the establishment of six municipalities in the core Sämi area as the Sami Administrative Area, where Sami is co-official with Norwegian (cf. Sara 2002). Given that the Sameting and a detailed Sämi language policy have existed around ten years longer than such a Gaelic policy in Scotland, it is not surprising that the fundamental parameters of this policy were fixed several years earlier, most remarkably by the Sämi Language Act. Current language policy displays both political functions which contribute, among others, to legislation, and administrative functions which are in line with the expenditure function in the Hague/Harrop list. Two recent examples of how the Sameting deals with language will demonstrate these differing functions. The first example is a meeting of the Sameting's plenum in February 2003. The topic under discussion was the extension of the Sämi administrative area to cover one additional municipality. The proceedings were conducted as in any other parliamentary body: A motion was introduced to the plenum, and then subsequently debated, amended and finally approved. The next step, however, was to "ask the (Norwegian) government to introduce to the Storting an amendment to the Sämi language Act to make the changes as agreed in Sameting". Here, we clearly see the limits of the Sameting within the system: The Sämi have been given a coordinated body which fulfils the responsive and representative functions of parliaments. Once a decision has been taken, however, it depends on negotiations with the Norwegian government to decide if this decision is implemented as a binding policy. A second example of how the Sameting deals with language planning is the Sameting's Language Department's leadership meeting in March 2002. It discussed mainly core language planning issues: The creation of a language prize, the demand that Sämi be used on the products of the state dairy company, the plea for language cooperation with the Sämi outside Norway, and several terminology issues ranging from ΓΓ to administration, and those

212

Heiko F. Marten

concerning reindeer farming. Here again, Sämi language policy was directly in Sämi hands. Through a budgetary competence assigned to it by the Norwegian government, it is able to authorize expenditure on various aspects of language policy. Thus, the main importance of the Sameting concerning language is that it is able to carry out administrative tasks fairly independently of other bodies. In areas where the Sameting administers a budget for funding language planning projects, the empowerment of Sämi through the Sameting as a body responding directly to the Sämi population's needs becomes obvious. In those cases where the Sameting can only suggest policy and legislation to the government, empowerment in the strict sense is not given - although the Sameting, as an institutionalized body, is much more likely to succeed than a less coordinated approach of lobbying would be. Sämi efforts regarding language planning have thus been centralized and are in Sämi hands to a much higher degree than they used to be before the Sameting's establishment. Additionally, the Sameting's symbolic function contributes to selfawareness, political organization and determination in putting forward the cause of the Sämi language to a degree which is entirely out of reach for the function of the Scottish Parliament for Gaelic. As a consequence, initiatives to support the Sämi language are regular, and they are targeted at numerous aspects of language policy, from highly political to mainly administrative, from core status questions to down-to-earth corpus planning. The Sameting is also consistently striving for more influence. Negotiations with the government, e.g., on financing more Sämi language projects, show that pressure may be exerted and often proves to be successful. Despite some rights and budgetary guarantees, however, Sämi language policy still depends on the financial and legal goodwill of the Norwegian authorities, and on the cooperation of the state, county and municipality authorities involved.

5.

Conclusion: Can parliaments contribute to the empowerment of minority languages speakers?

To conclude, let us now go back to the question raised at the beginning: How can parliaments contribute to the empowerment of minority language speakers? By presenting two quite different examples of decentralized parliamentary institutions, it has been my intention to point out the different roles

The potential ofparliaments for the empowerment of linguistic minorities

213

such a body can play. In both cases, the decentralization of parliamentary power has resulted in an increase in language policy initiatives. It can be concluded that the existence of these two parliaments makes a difference. The establishment of both parliaments has directly resulted in official legislation for the languages. However, there are distinctions between the two parliaments. Considerably more lobbying is needed in Scotland for Gaelic to become part of the agenda there. Political measures usually do not go as far as desired by many activists. If a policy receives parliamentary support, however, a great deal of the political process has been shortened; implementation is controlled by the highest political bodies and is thus much more likely to succeed. In the Sameting, on the other hand, there is much more devotion to language issues than in the Scottish Parliament. But after a decision is taken by the Sameting, implementation, at least for far-reaching issues, still depends on negotiation with the Norwegian government. Concerning the relation of parliaments and linguistic minorities, we may deduce that the answer, from the evidence presented, to the question whether parliaments may contribute to the empowerment of minority language speakers is a very careful Yes - within clear limits. Of the two broad functions of parliaments for minority language speakers, the function of representation and creating awareness has substantially increased. In relation to the second function, the chance to cany out language planning, decentralization of parliaments has brought minority language speakers closer to decision-making. Choosing between Gaelic or Sami does not automatically signify the restriction of finding oneself within less powerful domains any longer. Language policy has been enhanced for the benefit of minority languages under a certain participation of their speakers. Parliamentary bodies may thus indeed serve the empowerment of minority language speakers and ensure participation of these groups in mainstream decision-making, with increased chances of a real language choice. Both symbolism and representation on the one hand, and decisionmaking, legislating and the impact on expenditure, on the other, have thus been seized to improve participation of Gaelic and Sami speakers. Similarly, regarding demands, legitimization and institutionalization are also fulfilled to a certain degree. The Scottish Parliament and the Sameting have both contributed to multicultural (in the form of multilingual) awareness on the difficult path to normalization - which is indeed a term regularly referred to by politicians in both countries. The Scottish Parliament provides the chance of representing multilingual rights and showing awareness for

214

Heiko F. Marten

these within the existing framework. The Sameting, on the other hand, allows for a limited degree of self-government. In the Fishman model, we have today come closer to the following situation. Ρ

Gaelic/English, Sami/Norwegian

n-P

Gaelic, Sami

This seems to be the present relationship of Gaelic and Sämi to powerful and less powerful domains. Gaelic and Sami can be used in at least some political bodies. Policy initiatives are now taken on behalf of Gaelic/Sami speakers in a representative parliamentary way. And yet, the impact of the parliaments on empowerment is clearly limited. Decisions ultimately still depend on the majority population, and positive changes so far have been restricted, either to certain domains of language use (Gaelic), or geographically (Sami). As a whole, both cases show that the status of minority languages and the rights of their speakers continue to be subject to the political goodwill of the majority. Implementation and interpretation - despite some structural improvements and legal guarantees - still depend on negotiation with political bodies dominated by the majority population. Ultimate power is still with mainstream society. The two parliaments have created more control mechanisms and structural chances to empower minorities. But if the majority wishes to ignore these demands, it is still able to do so. The example of the limited strength of the Gaelic Act shows that much more could be achieved.

References Broderstad, E. G. 1995 Samepolitikk - interessemaksimering eller identitetsskaping? (LOS i Nord-Norge Notat 39). Troms0: Universitetet i Troms0. Fishman, J. A. 2001

Why is it so hard to save a threatened language?, In Can Threatened Languages be Saved? Reversing Language Shift, Revisited: A 21st Century Perspective, J. A. Fishman (ed.), 1-22. Clevedon: Multilingual Matters.

The potential of parliaments for the empowerment of linguistic minorities

215

Hague, R. and M. Harrop 2001 Comparative Government and Politics. An Introduction. Houndmills: Palgrave. Lippert, S. M. (ed.) 1995 The Encyclopedia of Democracy. Vol. II. London: Routledge. MacKinnon, K. 2000 Magga, Ο. H, 1995

Marten, H. F. 2004

May, S. 2001 McLeod, W. 1998 2001

Scottish Gaelic. In Languages in Britain and Ireland, G. Price (ed.), 44-57. Oxford: Blackwell. The Sami Language Act. In Linguistic Human Rights, T. Skutnabb-Kangas and R. Phillipson (eds.), 219-234. Berlin and New York: Mouton de Gruyter. The establishment of Scottish parliament: What difference does it make for the Gaelic language? In The Future has Already Begun. Recent Approaches in Conflict Linguistics (Plurilingua XXV), J. Darquennes, P. H. Neide, and P. J. Weber (eds.), 71-80. St. Augustin: Asgard. Language and Minority Rights: Ethnicity, Nationalism and the Politics of Language. Harlow: Pearson Education. Scotland's languages in Scotland's parliament. Scottish Affairs 24: 68-82. Gaelic in the New Scotland: Politics, rhetoric and public discourse. Journal on Ethnopolitics and Minority Issues in Europe, at: http://www.ecmi.de/jemie/download/JEMIE02MacLeod2811-01.pdf.

Neff - van Aertselaer, J. this volume Language policies in Spain: Accommodation or alteration? Sara, A. 2002

Regional Characteristics of Säpmi and the Sami people, Presentation at the Conference: The Regional Identity of Ethnic Groups in Europe, in collaboration with The Land of Schleswig-Holstein, the Federal Union of European Nationalities and the European Academy Sankelmark. 13-14 June 2002. available at: http ://www.fuen. org/pdfs/20020614Saami_Sara.pdf

216

Heiko F. Marten

Spolsky, B. 2004 Turan, I. 1994

Language Policy. Cambridge: University Press. The Turkish legislature: From symbolic to substantive representation. In Parliaments in the Modern World, G. W. Copeland and S. C. Patterson (eds.), 105-128. University of Michigan Press.

Web Sites BÖRD NA GÄIDHLIG: http://www.bord-na-gaidhlig.org.uk SAMI POLICY OF THE NORWEGIAN GOVERNMENT: http://odin.dep.no/krd/norsk/tema/same/bn.html THE GOVERNMENT OF NUNAVUT LANGUAGE POLICY: http: IIwww. gov. nu. ca/ cley/English/language. htm THE SAMETING: http://www.samediggi.no THE SCOTTISH EXECUTIVE: http://www.scotland.gov.uk THE SCOTTISH PARLIAMENT: http://www.scottish.parhament.uk

The dominance of languages and language communities in the European Union (EU) and the consequences 1 Ulrich Ammon

1.

The public discourse of multilingualism and the growing predominance of English

The public discourse of EU politicians on language policy stresses the need to maintain all the languages, at least the indigenous languages of the EU as part of the community's cultural wealth and as an essential component of European identity. As a corollary, general individual multilingualism is stipulated to guarantee communication between EU citizens, ideally the knowledge of three languages: mother tongue + two foreign languages. Such views were presented over and again, for example during the European Year of Languages in 2001. The EU's linguistic reality appears to be different, in that its perhaps most striking feature is the ever greater predominance of just one language, English. This fact is often felt to be in conspicuous contrast with the EU's language-policy objectives. However, at close view the predominance of English and multilingualism are at least partially compatible. They could for instance coincide, in the long run, in the following overall EU language situation: 1. on the societal level, (a) a monoglossia for the English-speaking countries, (b) a diglossia for other linguistic majorities or for linguistic minorities in the English-speaking countries: national language/ minority language + English, (c) a triglossia for linguistic minorities outside the English-speaking countries: minority language + national language + English, (with English always the "highest" language1,"highest" in diglossia terminology) plus, in all three cases, the additional use of a number

218

Ulrich Ammon

ber of other foreign languages for special purposes or in special domains. 2. on the individual level, the corresponding (a) prevalent monolingualism of the native speakers of English (b) prevalent bilingualism for the diglossia regions or, respectively, (c) prevalent trilingualism for the triglossia regions, plus, in all three cases, additional foreign language skills for special sections of the population (quadrolingualism, etc.) Societal and individual multilingualism would be largely maintained, except for the English-speaking countries, and no language would have to disappear. At the same time, communication throughout the entire EU would be guaranteed. Nevertheless, such a situation would hardly be wellreceived generally. The major reason is perhaps the obvious inequality of languages, especially the predominance of English, which - among other things - would allow the English-speaking countries to be excepted from the general European multilingualism. Typical criticism often is about the imperialism of the English-speaking world led by the U.S.A., imposing their own language on the entire globe, including Europe. To avoid misunderstanding I want to point out right at the beginning that I myself do not believe that the recent predominance of English in the EU and world-wide is mainly caused through continued language enforcement by the English-speaking countries. It is more likely a consequence, in my opinion, of rational behavior by those who acquire skills in English, i.e. from individuals', institutions' or governments' own choice. We may glean the underlying rationality through examination of some simple economic considerations such as the following: (1) On the individual level: A (native) speaker of French, German or any other language reaps more benefit, on average, from learning English than a speaker of English does, on average, from learning French or any other language. (2) On the societal level: A whole country mostly benefits more from placing English at the top of its school curricula than it would from any other language. (3) On an admittedly very abstract level: It would be less costly for any entire language community other than English to acquire skills in English than it would be for the entire English language community (the native speakers) to acquire skills in any other language. This can be

The dominance of languages and language communities in the EU

219

deduced from the fact that the Total Income of the English language community exceeds the Total Income of any other language community. May these broad proposals suffice here to indicate that there are more weighty reasons for the spread of English today than political pressure or ruse. This is not to deny that the rationality of individuals' choices, and social entities' choices, in favor of English is dependent on the overall power, especially the economic power, of the English-speaking countries. Nor do I deny that the English-speaking countries have had, and still have, policies of spreading their language that most likely enhance the spread of English (Phillipson 1993). Numerous other countries also try to spread their own language as best as they can but with less success. One could, however, still argue that the English-speaking countries' endeavors are unfair considering their superior might - if one holds certain standards of fairness in international politics. What I want to point out is that criticism of English language predominance finally boils down to the criticism first, of language communities' unequal power and second, of their making use of that power or letting it work to their advantage. This view - if correct reveals how difficult it is to actually change language predominance. It does, however, not imply that nothing could be done about it. The reasons suggested for the predominance of English hold similarly, in principle, for the secondary predominance of other languages, further down the language hierarchy, like French, German, Spanish, Italian, etc.

2.

Potential biases in the approaches to £ U language policy

There have been numerous analyses of the EU language situation. Recent studies have been, among others, books by Robert Phillipson (2003), Andreas Ross (2003) and Peter A. Kraus (2004), and articles, or book chapters, by Theo van Els (2002), Abram de Swaan (1999; 2001: Chapter 8), or Ammon (2003a, 2003b, and 2003/4). Phillipson (2003) has sounded the alarm against English becoming the predominant language of the EU by exaggeration, expressed even in his book title English-only for Europe? He does not say, of course, that English is about to completely substitute all the other languages in Europe, only that it is about to become its sole lingua franca and sole working language, or even sole official language, of European institutions, of the EU or perhaps also the Council of Europe.

220

Ulrich Ammon

The possibility of English substituting all the other languages even in everyday life in the long run seems implied, however, in Phillipson's discourse. One could ask, provocatively, whether it would really be a catastrophe. Have linguists shown convincingly, or could they, that the over-all losses of an English-only situation (in the radical sense) would really be greater than its benefits? Linguists' unsatisfactory answers to this question may be among the reasons why their advice is often not taken seriously by law makers or politicians. An example is the recent draft of an EU Constitution. All the linguists participating in the discussion were in favor of a language paragraph in the Constitution highlighting the value of multilingualism, but they could not move the Convent to comply. Often, linguists' sole objective seems to be the preservation of maximal language diversity without seriously considering potential social or economic downsides or how communication functions. Maximal language diversity is, of course, in line with linguists' particular interests. Whether, however, it coincides with other peoples' interests remains an open question, and one which linguists avoid tackling seriously, or which is even taboo among them. Some views expressed on the EU language situation do not, or at least not openly, oppose English language predominance. Examples are David Laitin (1997), Abram de Swaan (1999; 2002: Chapter 8), or Theo van Els (2002). To Laitin's proposal, I will return shortly. As for de Swaan or van Els, both their views seem to be pro-English predominance, at least latently. Their inclination seems compatible with the interests of their own language community, Dutch, a language which has no real chance of either becoming a lingua franca or an institutional working language of the EU. It would, therefore, be easier for the Dutch language community to have just one clearly predominant language on which to focus foreign-language studies, rather than several languages that have to be studied and managed. In contrast, language communities like the French, the German, the Spanish or the Italian have different interests in that they still hope that their languages will play a role, if only a limited one, as a lingua franca and an institutional working language of the EU in the future. They therefore would like to restrict the predominance of English. It seems, in my view, reasonable to assume that all individuals dealing with language policy are affected by their own linguistic (language community's), national and professional interests. I am too, of course, being a member of the German language community and, in addition, a professor of German. I also feel that Ross' (2003) or Kraus' (2004) books may be affected in their support of multilingualism by their own language commu-

The dominance of languages and language communities in the EU

221

nity's particular interests, especially in their support of several working languages for the EU institutions and their rejection of English as the only working language. As scholars, we try, of course, to keep our personal interests in check and to argue "objectively". However, the only guarantee for an overall somewhat balanced view is, in my opinion, that participants from as many language communities, countries, and scholarly or scientific fields as possible contribute to the discussion.

3.

A rough framework for EU language policy and the focus of this article

Being aware of the possibility of my own bias, in the following, I will focus on various aspects of the EU's language policy: (1) institutional working languages, (2) lingua francas and, as a corollary, (3) foreign language studies. These questions are of primary interest for the German language community and for language communities with languages of a similar status or function. Any comprehensive EU language policy, of course, has to deal with numerous more or less contested and conflictive topics. One can catch glimpses of this diversity from the viewpoint of a typology of EU languages according to status and function. One possibility of such a typology, which I have presented elsewhere (Ammon 2003/2004), is partially hierarchical and implicational, though not with respect to all the types, and thus forms a partial taxonomy. The features of the different types can roughly be deduced from their names. (1) (2) (3) (4) (5)

European lingua franca(s) EU institutional working language(s) EU institutional official languages EU member countries' national official language(s) EU member countries' regional official language(s) EU member countries' acknowledged (indigenous) minority language^) (6) EU member countries' unacknowledged indigenous minority language^)

222

Ulrich Ammon

(7) EU member countries' unacknowledged exogenous minority language^) (8) Languages with debated language or dialect status (9) Dialects The types are not disjunctive with respect to single languages. Some languages belong only to one of the types (e.g. Letzeburgish to 3, or Galician to 4), while others, especially the numerically stronger, belong to several (e.g. German to all of 1 - 6, with different sections of its language community). Principally, all language communities are interested in raising the status or function of their own language on the typological hierarchy, or to maintain their present levels. Accordingly, language communities' interests diverge. Another notable difference between the EU language communities is, of course, their different clout regarding the pursuit of their own interests.

4.

Differences in acceptability of English language predominance for the EU as compared to some multilingual countries

In spite of the predominance of English several other EU languages have continued to play a role as foreign languages taught in school, or studied privately, within and outside the EU. They also serve, to some extent, for transnational, or rather translingual, contacts within and beyond the EU. These languages also mark a crucial difference, among others, between the EU and countries like India, which David Laitin (1997) has proposed as a linguistic model for the future EU, especially if the EU once turned into a federal state. Languages like French, German, Italian or Spanish resemble Hindi in that they function as institutional working languages and as a lingua franca within the EU (cf. Sociolinguistica 15/2001) together with, though to a lesser extent than, English. They, however, differ from Hindi, in that they are also studied as foreign languages, and used for translingual contacts, beyond the EU, while Hindi has these functions practically only within the Indian Union. Is the role of English for the EU the same as it is for India, or should it be in the future? There too is an important difference. For the EU, English is an indigenous language, for India it is imported and traditionally foreign. In the EU, it therefore does not have the neutrality it has in India. This limits its acceptability within the EU in a manner similar to Hindi, which is

The dominance of languages and language communities in the EU

223

not accepted as the Indian Union's sole lingua franca, especially not in Southern India. Europeans, the linguistically sensitive at least, feel that granting English the predominance it has in India would carry with it enormous linguistic privileges for two EU member states, or then citizens: Britain and Ireland. Robert Phillipson (2003: 101, 127, 142 f.) presents Australia, South Africa and Sweden as potential models of functioning multilingualism for the EU. However, he neither highlights that English is much more predominant in these countries than it is in the EU as a whole, nor that English has a different status in these countries than it does in the EU, namely as an entirely foreign, i.e. a "neutral" language with respect to the indigenous language communities in Sweden, or as the dominant language imposed by colonialism in Australia and South Africa. Is English on its way to a similarly overwhelming predominance in the EU? Let us have a brief look at its actual predominance in the EU and at some EU programs intended to counteract such a predominance.

5.

The predominance of English in the EU

5.1.

Number of speakers of a native and foreign language

The EU comprises 25 member states, ten of which joined on 1 May, 2004; before that date there were 15 member states. There are data collected in intervals {Eurobarometer) on the number of speakers, actually speaker claimants, of languages as a native and as a foreign language which show the change of figures for English in comparison to other larger EU languages. A representative sample of EU-citizens of each country was asked, first, regarding their native tongue, and, second, about the languages (except native tongue) in which they could carry on a conversation. In the year 2000 survey, the question was changed into "Which language (except native tongue) do you know?" (for informants in the German-speaking countries: "Welche Sprache ... können Sie?"), which is hardly a question clear enough to produce valid answers and whose results are therefore usually not taken seriously (figures in Table 1). There are no exactly comparable figures for the ten new member states which joined the EU in May 2004, but various data clearly show that the overall rank order for the central-eastern European countries is English > German > French > Italian or Spanish.

224

Ulrich Ammon

Table 1. Number of speakers of major EU language in the EU of 15 countries (percent of citizens) (Eurobarometer 28 (1987): 78; 41 (1994): 36; 50 (1998): 110; for (2000) cf.: http://europa.eu.int/comm/education/languages.html2 Foreign Tongue ? ? ? ? ?

Sum Total 36 25 27 19 14

Mother Tongue 17 24 16 17 9

1994 Foreign Tongue 25 7 13 1 5

Sum Total 42 31 29 18 14

Foreign Tongue 31 8 12 2 4

Sum Total 47 32 28 18 5

Mother Tongue 16 24 16 16 ?

2000 Foreign Tongue 40 10 19 3 7

Sum Total 56 34 35 19 18

1987

English German French Italian Spanish

Mother Tongue ? ? ? ? ?

English German French Italian Spanish

Mother Tongue 16 24 16 16 11

1998

5.2.

Number of learners of a foreign language

Comprehensive figures from regular schools are available for four languages (cf. Figure 1). Additional data show that Italian is the language following Spanish in the rank order of languages studied in the EU countries (cf. eg. Schlüsselzahlen zum Bildungswesen in Europa 2002\ 172). The national FL curricula of practically all EU countries (present and future member states) require English as an obligatory subject for all pupils. Other languages, mainly French, German, Spanish, or Italian - are either an obligatory choice or merely optional subjects with mostly less learning time allotted to them than to English.

The dominance of languages and language communities in the EU

225

English

French

ι 1. ,Η

11.Η I

ω!, (0.J.I.C). J (:) !