Trends in Romance Linguistics and Philology: Volume 5 Bilingualism and Linguistic Conflict in Romance 9783110848649, 9783110117240

Table of contents :
5.1 Overview
Representations of Romance: contact, bilingualism and diglossia
Language conflict in Romance: decline, death and survival
Standardization in Romance
5.2 Romance: Romance bilingualism in Europe
Galician and Castilian in contact: historical, social and linguistic aspects
Catalan: Castilian
Occitan: French
Bilingualism and language conflict in Rhæto-Romance
Language contact and Corsican polynomia
Sardinian: Italian
Italian and Italian dialects: An overview of recent studies
Balkan Romance: Rumanian
5.3 Romance: Romance bilingualism outside Europe
Québécois French and language issues in Quebec
Bilingualism and linguistic conflict in (French) creole-speaking societies
Creole Spanish and Afro-Hispanic
Portuguese and/or ‘Fronterizo’ in northern Uruguay
Aspects of Creole : Portuguese bilingualism and linguistic conflict
5.4 Second language acquisition by immigrants
Second language acquisition by migrants in Europe
The acquisition of Romance by speakers of a Romance variety in North America
The acquisition of a second Romance language by immigrants in Latin America
Subject Index
Index of Languages
Index of Authors

Citation preview

Trends in Romance Linguistics and Philology Volume 5

Trends in Linguistics Studies and Monographs 71

Editor

Werner Winter

Mouton de Gruyter Berlin · New York

Trends in Romance Linguistics and Philology Volume 5: Bilingualism and Linguistic Conflict in Romance Edited by

Rebecca Posner John N. Green

Mouton de Gruyter Berlin · New York

1993

Mouton de Gruyter (formerly Mouton, The Hague) is a Division of Walter de Gruyter & Co., Berlin.

© Printed on acid-free paper which falls within the guidelines of the ANSI to ensure permanence and durability

Die Deutsche Bibliothek

— Cataloging

in Publication

Data

Trends in Romance linguistics and philology / ed. by Rebecca Posner ; John N. Green. — Berlin ; New York : Mouton de Gruyter. (Trends in linguistics : Studies and monographs ; ...) NE: Posner, Rebecca [Hrsg.] Vol. 5. Bilingualism and linguistic conflict in Romance. — 1993 Bilingualism and linguistic conflict in Romance / ed. by Rebecca Posner, John N. Green. — Berlin ; New York : Mouton de Gruyter, 1993 (Trends in Romance linguistics and philology ; Vol. 5) (Trends in linguistics : Studies and monographs ; 71) ISBN 3-11-011724-X NE: Posner, Rebecca [Hrsg.]; Trends in linguistics / Studies and monographs

© Copyright 1993 by Walter de Gruyter & Co., D-10785 Berlin All rights reserved, including those of translation into foreign languages. No part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Typesetting and Printing: Arthur Collignon GmbH, Berlin Binding: Lüderitz & Bauer, Berlin Printed in Germany

Foreword

The first four volumes of Trends in Romance linguistics and philology (published between 1980 and 1982 under the imprint of Mouton Publishers, The Hague) were devoted, in turn, to: Romance comparative and historical linguistics, Synchronic Romance linguistics, Language and philology in Romance, and National and regional trends in Romance linguistics and philology. We had not originally intended to follow up the series with a supplement, despite the encouragement to that effect given in more than one of the generally favourable reviews, whose authors rightly pointed out that evaluative surveys tend to date quickly, and even more so the bibliographic information that underpins them. The idea for the present volume does, nonetheless, derive from a review by Dr Jana Vizmuller-Zocco, who was struck by the lack of discussion of language acquisition in Romance, especially in bilingual contexts. We are grateful to her for the original suggestion, for subsequent helpful comments on design, and for agreeing to contribute a chapter to the volume. In planning this sequel, we wished to link bilingualism with diglossia and language conflict, topics which have grown vastly in importance and resonance since the original project was conceived. An early policy decision, taken in the interests of confining the volume to manageable proportions, was to limit coverage almost exclusively to instances of contact between two or more varieties of Romance. While this limitation has excluded some major external influences on Romance, we hope that it has worked to produce a treatment with greater thematic unity and cohesion than would otherwise have been possible. Our thanks go to all our contributors, for striving to meet our early requests on content and scope, and for accepting patiently our later editorial suggestions and (sometimes major) stylistic interventions. We are particularly grateful to Alvina Byrne, for her thorough work on the indexes, and to Anthony Grant, for translating one of the articles, and for his unstinting and often ingenious assistance with the bibliographical searches. Thanks are also due to Professor Werner Winter for his readiness to accept a supplementary title into the Trends in Linguistics series and

vi

Foreword

for his practical advice on our first proposal; and above all to Dr MarieLouise Liebe-Harkort, Ms Margaret Saunders, and all their colleagues at Mouton de Gruyter head office in Berlin, for their patience with our editorial idiosyncrasies and the long delays that, despite good intentions and well laid plans, have disrupted the production schedule. Kaye Green has been, as ever, an unfailing source of practical help and moral sustenance. Rebecca Posner John N. Green June 1992

Contents

5.1 Overview John N. Green Representations of Romance : contact, bilingualism and diglossia Rebecca Posner Language conflict in Romance: decline, death and survival . . . Zarko Muljacic Standardization in Romance

3 41 77

5.2 Romance : Romance bilingualism in Europe Henrique Monteagudo and Anton Santamarina Galician and Castilian in contact: historical, social and linguistic aspects Miquel Strubeil i Trueta Catalan : Castilian Brigitte Schlieben-Lange Occitan : French Rebecca Posner and Kenneth H. Rogers Bilingualism and language conflict in Rhasto-Romance Ghjacumu Thiers Language contact and Corsican polynomia Rosita Rindler Schjerve Sardinian : Italian John Trumper Italian and Italian dialects: An overview of recent studies . . . Marius Sala Balkan Romance : Rumanian

117 175 209 231 253 271 295 327

5.3 Romance : Romance bilingualism outside Europe Richard Y. Bourhis and Dominique Lepicq Quebecois French and language issues in Quebec

345

viii

Contents

Guy Hazael-Massieux and Didier de Robillard Bilingualism and linguistic conflict in (French) creole-speaking societies John Μ. Lipski and Armin Schwegler Creole Spanish and Afro-Hispanic Fritz G. Hensey Portuguese and/or 'Fronterizo' in northern Uruguay Thomas Stolz Aspects of Creole : Portuguese bilingualism and linguistic conflict

383 407 433 453

5.4 Second language acquisition by immigrants Georges Lüdi Second language acquisition by migrants in Europe 495 Jana Vizmuller-Zocco The acquisition of Romance by speakers of a Romance variety in North America 535 Giovanni Meo Zilio The acquisition of a second Romance language by immigrants in Latin America 559 Subject Index

591

Index of Languages

606

Index of Authors

612

Contributors

Richard Y. Bourhis John Ν. Green Guy Hazael-Massieux Fritz G. Hensey Dominique Lepicq John M. Lipski Georges Lüdi Giovanni Meo Zilio Henrique Monteagudo Zarko Muljacic Rebecca Posner Rosita Rindler Schjerve Didier de Robillard Kenneth Η. Rogers Marius Sala Anton Santamarina Brigitte Schlieben-Lange Armin Schwegler Thomas Stolz Miquel Strubell i Trueta Ghjacumu Thiers John Trumper Jana Vizmuller-Zocco

Universite du Quebec ä Montreal University of Bradford Universite de Provence ä Aix-en-Provence University of Texas, Austin McMaster University, Hamilton University of New Mexico, Albuquerque Universität Basel Universitä degli Studi di Venezia Universidade de Santiago de Compostela [Emeritus] Freie Universität Berlin University of Oxford Universität Wien Universite de Provence ä Aix-en-Provence University of Rhode Island Institutul de Lingvisticä, Bucure§ti Universidade de Santiago de Compostela Universität Tübingen University of California, Irvine Ruhr Universität, Bochum Inst, de Sociolingüistica Catalana, Barcelona Universite de Corse, Corti Universita della Calabria, Cosenza York University, Ontario

5.1

Overview

J O H N Ν. G R E E N

Representations of Romance: contact, bilingualism and diglossia

l.

Introduction

The contributions to this volume share a common concern with the use of language in its socio-cultural context, and its representation in the minds of speakers. As such, they reflect the extraordinary blossoming of sociolinguistics (in the widest sense) witnessed over the past quarter century, and the sheer diversity of theoretical persuasions and practical applications that have now evolved. Some of the chapters below attempt answers to questions that would not, and perhaps could not, have been formulated a generation ago. Speculating on the factors responsible for both the prodigious level of activity and the shifts of emphasis, Lavandera (1988) suggests, perhaps not too fancifully, that the concentration of interest on sociolinguistic, pragmatic and discourse approaches to language, may be indirectly due to Chomsky — to a reaction, that is, against the abstract modelling of idealized data that characterized the early phase of generative linguistics. 1 As we shall see, she is not alone in her conviction. Nonetheless, within the sphere of European Romance, we must recognize an additional factor, not linguistic in origin, that has proved at least equally influential. The emergence of a new political order and the rise of European federalism have created conditions strongly favouring the resurgence of 'oppressed' minority languages and regional varieties that can now be harnessed as vehicles for autonomist aspirations. Such political movements have sought intellectual underpinning in the rich theoretical seam of conflict sociolinguistics pioneered by Catalan and Occitan linguists (see section 4.1 below, and the chapters by Posner, Schlieben-Lange, Strubell i Trueta, and Thiers). Reciprocally, political prominence has fuelled vigorous debate on the effectiveness of language planning and especially of policies aimed at raising the prestige of varieties whose low social status is now perceived as anomalous.

4

John Ν. Green

This introductory chapter (which is designed to be read in conjunction with those of Posner and Muljacic), by surveying some prominent theories of language, bilingualism, diglossia, and contact-induced change, aims to set a frame within which traditional and newly identified problems in Romance can be elucidated, and conversely, the explanatory potential of the theories can be evaluated against the known Romance facts. 2. Language and lects The concept of language as social semiotic goes back at least to Saussure, who seems to have been profoundly influenced in this aspect of his work by Dürkheim. For Saussure, language is above all a "fait social" (1965: 29, 30 — 32), knowledge of which is both a marker and an obligation of community membership. In Saussure's famous dichotomy, it is langue which is social and essential, while parole is individual and contingent: langue is the socially-warranted system learned by individuals and used by them to communicate thought through ephemeral speech acts. Whereas parole is multivariate and idiosyncratic, langue is represented in identical form in the mind of every adult member of the speech community and, once learned, is immune to conscious modification (1965: 38). Chomsky's no less famous dichotomy of competence and performance, often likened to its forerunner (Lavandera 1988: 1), nevertheless diverges crucially from it in the divorce of the underlying system from the social warrant. For Chomsky, competence is individual, essential, and psychologically validated; while performance is its often imperfect instantiation in social contexts (see, for instance, Chomsky 1980: 201 —205).2 At most, social interaction influences competence during first language acquisition, when the child is steered by its genetic blueprint to infer knowledge of grammar, parameters and specific rule schemata from limited and sometimes degenerate input — the performance of other speakers. Children, it seems, acquire language not through, but in spite of, social interaction. Chomsky, of course, does not deny the existence of ambient linguistic variation, but believes it is best studied, together with other aspects of performance, once real insights have been gained into linguistic universale and mental representations, which in turn can only be approached via idealization (1980: 24 — 26). Such a research agenda not only downgrades the social functions of language, but also, through its methodological assumption of the "homogeneous speech community", marginalizes bilingualism and seemingly denies the possibility of diglossia (see section 4 below).

Representations of Romance: contact, bilingualism and diglossia

5

Two principal linguistic traditions span the chronological gap between Saussure and Chomsky: anthropological linguistics, typified by Boas and Sapir, and Bloomfieldian structuralism. Bloomfield himself, trained in the comparative method and practised in the recording of hitherto undescribed Amerindian languages, was acutely aware of spatial variation and temporal change, but, perhaps inhibited by allegiance to a mechanistic, behaviourist theory of psychology, paid little regard to the systematic social functions of language. 'Language' was a convenient abstraction, studied by induction from the linguistic behaviour of a speech community, defined as "a group of people who interact by means of speech" (Bloomfield 1935: 42) and admitting a certain degree of internal complexity, manifested as dialects and levels of formality (1935: 52). Later Bloomfieldians, finding the concept of speech community problematic and certainly too fluid to serve as the chief component of a definition of language, came to rely instead on the idiolect, the linguistic behaviour of a single (but typical) individual. From this perspective, a language is simply "a collection of more or less similar idiolects" (Hockett 1958: 322), and a dialect is a sub-collection with greater overlap of linguistic features and correspondingly enhanced mutual intelligibility. (For a general critique of the notion of speech community, see Romaine 1982.) Similarly, and as if to guard against both reification and higher-level abstraction, Hall admonishes that language "has its place of existence only in the individual idiolect" (1964: 395). Although the structuralist and anthropological traditions within American linguistics subscribed to almost identical descriptive techniques (and, incidentally, agreed on the urgency of recording the diversity of languages threatened with extinction), they differed sharply in their view of the social and cultural role of language. Whereas Hockett (1958: 322) is content to note the practical impossibility of observing directly the speech habits of a whole community, most anthropological linguists and ethnographers believe that not to attempt this (however difficult the undertaking) denies the centrality of language and risks falsifying its nature. For instance, Hymes, one of the anthropologically-trained linguists credited by Lavandera (1988) with the successful development of an alternative to the Chomskyan paradigm, envisages: "a science that would approach language neither as abstracted form nor as an abstract correlate of a community, but as situated in the flux and pattern of communicative events" (Hymes 1974: 5). In this scheme, individuals must learn not only the structural properties of a linguistic system, but also how to drive the system to produce speech acts that will themselves be interpreted as

6

John Ν. Green

socially significant within the community to which they relate. In Hymes's memorable phrase, individuals strive to acquire and display communicative competence. Despite their manifold differences, the approaches to language sketched above share the assumption that the natural and predominant locus of first language acquisition is within a monolingual community with a well focused image of its language. To be sure, the community language may be internally complex, and there will always be individual instances of bilingualism as a result of exogamy or migration, but these special circumstances serve only to underscore the dominance of the monolingual pattern. This view has been forcefully challenged by Le Page and Tabouret-Keller (1985), on the basis of studies of children growing up in multilingual Creole communities where the competing codes are illfocused and where language learners, through their selection and combination of socially-valued features, create not only a language but, in a very real sense, their own social identity. Within this framework, individuals are seen as accommodating their linguistic behaviour to that of groups with which they wish to be identified — subject to the usual limitations of access to the target group, learning capacity, and motivation. 3 If, as Le Page contends, multilingualism is the norm outside the relatively few (but successful) western societies with highly focused standard languages and near-universal literacy, Chomsky's concept of competence is seriously undermined, and fatally so if the competing codes in multilingual settings happen to be relatively diffuse (Le Page 1973). In such circumstances, even Hymes's communicative competence requires redefinition, since individuals must negotiate not only the communicative content of their speech acts, but also that part of the system of social values that accredits them as communicatively competent. The particular blend of ethnography and social psychology favoured by Le Page and Tabouret-Keller, tempered by extensive fieldwork, leads them to postulate an elaborate, fourfold definition of 'language' (1985: 188 — 193). Sense (1), about which they are openly sceptical, corresponds to the native language or mother tongue, allegedly a fundamental property of the individual members of a particular society and often appealed to for emotional or political ends. Sense (2) refers to the observable linguistic behaviour of individuals seen in its totality or subsets. Sense (3) denotes descriptions made by linguists from observed behaviour (that is, from Sense 2 data), with the proviso that no two descriptions are generatively equivalent and that the objectivity linguists bring to the task is conditioned by their theoretical preferences and methodological train-

Representations of Romance: contact, bilingualism and diglossia

7

ing. Sense (4), probably the most familiar, is that of the language perceived from outside as a single entity (as in 'the French language'), and nowadays most probably referring to a standardized language, serving a community where it constitutes both a norm and a yardstick for prescriptivism. This schema, as will be immediately apparent, has few points of direct contact with the dichotomous definitions of language discussed above — an inevitable consequence, its authors would maintain, of their recognition of the pervasiveness of multilingualism and the need for multivariate modelling. 2.1 Languages, lects and roofs in Romance Romanists wishing to test the applicability of the hypotheses sketched above will be particularly concerned with: the number of languages (in Le Page's Sense 4) to be recognized, whether or not standardization or adoption for national or other official purposes is a necessary precondition for language status, and — more contentiously — under what 'roof (see below, and the chapter by Muljacic) the varieties denied language status should be grouped. Since issues of language status and recognition arise, directly or by implication, virtually throughout this volume, it is perhaps also worth asking whether the views of Romance specialists accord with those expressed in reference works intended for a wider audience, and whether political action designed to enhance or 'normalize' the status of any particular variety (see Strubell i Trueta, this volume) has had an impact on popular perceptions outside its claimed territory. 4 There can be no doubt about the status of the five standardized languages, of which Portuguese, (Castilian) Spanish, French and Italian form what is sometimes called Romania continua to the west and centre of the territory, while Rumanian lies to the east, insulated from the others by a wide band of Hungarian and Slavonic. The five standards clearly qualify for language status on such criteria as: long association with a territory, where they are recognized as official and symbolic of national unity; accessibility to large populations as both first and second languages; endowment with a substantial written corpus of literary and cultural merit; codification by grammarians; and external recognition of their distinctness from contiguous lects. It should be borne in mind that for Rumanian, fulfilment of the criterion of widespread external recognition dates back only a little over a century, following the Ardilean reforms leading, among other things, to the adoption of a romanized orthography. Italian, on the other hand, has enjoyed a much longer if perhaps spurious recognition based on foreign misconceptions about the pervasiveness of

8

John Ν. Green

the Tuscan written standard, whereas in fact linguistic unification through bidialectalism in a Milanese-inspired koine is a very recent phenomenon, postdating the Second World War (see Foresti et al. 1989; Muljacic 1989 c; and Posner and Trumper, this volume). An interesting commentary on the hierarchy of internal and external criteria for language status is offered by the case of Spanish and Portuguese, which are contiguous, structurally very close, and virtually intercomprehensible once a salient but regular group of phonological divergences have been surmounted; nevertheless, on political and cultural grounds, their separation has been effective since the 14th century and formally recognized outside the peninsula since at least the Treaty of Tordesillas of 1494. It seems, then, that the precedence accorded to the five Romance standards flows first and foremost from their symbolic role, closely followed by the mere fact of their standardization, and no doubt bolstered by the relative numbers of speakers (for statistics, see Fleischman 1992). The impression is confirmed by important reference works: The world's major languages (Comrie 1987) devotes individual chapters to all and only these five, as does the Oxford international encyclopedia of linguistics (Bright 1992), though in each case a brief overview of the whole Romance group is also provided. More surprisingly, the treatment is mirrored in a number of recent Romance manuals (Agard 1984, for instance), which pay at best scant attention to other synchronic varieties. The astonishing omission from this list is Catalan, which one would have expected to merit inclusion on the grounds of its cultural and literary pedigree alone, though, as Goebl (1986) reminds us, neither Diez nor Meyer-Lübke accorded it separate status in their comparative grammars. Meyer-Lübke changed his mind in the early 1920s and Entwistle's stillinfluential survey of the Hispanic languages (1936 [1962]), whether consequentially or by earlier conviction, treats Catalan as a language in its own right without even a suggestion that the status might be problematic. Matching recognition has been claimed on behalf of Occitan by its contemporary activists, though in fact the two cases are not fully parallel, since the varieties now grouped under the label 'Occitan' are not coterminous with the language of the medieval troubadours — whose separate existence was acknowledged by Raynouard even before Diez and MeyerLübke — nor with the 19th-century poetic revival of Provengal championed by Mistral and his fellow filibres (see Schlieben-Lange, this volume). Nonetheless, some measure of Occitan success can be gauged from the fact that a large-scale reference work for the general reader, the Compendium of the world's languages (Campbell 1991), has entries for seven

Representations of Romance: contact, bilingualism and diglossia

9

Romance languages, with Catalan and Occitan included exactly on a par with the five national standards. If the existence of a distinguished medieval literary tradition were of itself a guarantee of present-day language status, Galician and Sicilian would surely be included alongside Provencal, for not only do they boast substantial bodies of extant lyric poetry, but, ironically perhaps, they were quite explicitly acknowledged as foreign idioms at the height of their renown, by the non-native poets who, like Alfonso The Wise, strove to take advantage of their evocative qualities. But Galician (whose revival is debated at length by Posner and Monteagudo — Santamarina, this volume), despite acceding to the legal status of regional language in Spain, and despite the great strides its advocates have taken in both corpus and status planning, has yet to achieve significant international recognition. Leaving aside the Romance languages that have become extinct (Dalmatian and Mozarabic), 5 the remaining varieties for which language status has been, or may be, claimed all pose problems of a different order. Sardinian, Sicilian, and the lects of southern Italy sometimes grouped together as 'Calabrian', might all be considered eligible on the twin grounds of structural distance from northern Italian and difficulty of mutual intelligibility; but their internal diversity is such that both linguists and, it seems, their own speakers entertain doubts as to their unity and status (see Trumper, this volume). Sardinian was nevertheless included in the Romance canon by Meyer-Lübke, and few specialist manuals have ventured to excise it since (a separate article on Sardinian is, for instance, included in Harris —Vincent 1988). Romansh is an official cantonal language in Switzerland, where it serves a small and diminishing population, but official status has never been accorded to the wider grouping of Rhaeto-Romance which, from this standpoint, suffers the disadvantage of straddling three national frontiers. The fundamental unity ascribed by Ascoli to Rhaeto-Romance has recently been endorsed by Haiman (1988: 351), despite the well-founded objections of Francescato (1982) to the inclusion of Friulian (echoed even more forcibly by Pellegrini 1991), and the scepticism of Posner (1980) as to the justification of the entire grouping. In a few other cases within European Romance, notably Corsican (see Marcellesi 1984, and Thiers, this volume), the aspiration to independent language status is fuelled by a desire to be rid of a dominating 'roof which is judged to lack both linguistic and politico-cultural legitimacy.

10

John Ν. Green

The notion of a linguistic roof is less novel than its name. Rather as the technique of phonemic analysis creates a requirement of allophonic membership and single loyalty, so the process of language standardization may be seen to exert pressure on neighbouring lects to group themselves under the protective umbrella of the emergent standard. Lects either not attracted to this tutelage, or repelled by it, were described by Kloss as dachlos 'roofless', a metaphor refined and applied rigorously to Romance by Muljacic (see, especially, Kloss 1976, 1987; Muljacic 1989 a and this volume). While the image has protective connotations, the relationship of the lects to the standard can as easily be one of annexation and subservience, in which the 'roofed' varieties risk loss of identity through gradual assimilation, or abandonment following the erosion of what little remains of their social prestige. This all-too-predictable decline leads conflict theorists such as Marcellesi to reject the comfortable roofing metaphor in favour of satellization, defined as the tendency of a dominant ideology to "«rattacher» un systeme linguistique ä un autre auquel on le compare et dont on affirme qu'il est une «deformation» ou «une forme subordonnee»" (Marcellesi 1981 a: 9; see also section 4.1 below). Such reservations aside, valuable new insights have been gained into Romance relationships by the application of the standardological model as developed by Muljacic (1980, 1989 b, and this volume), especially on the vexed historical question of whether standard languages emerge, consensually as it were, from dialect constellations, or whether speakers of an ascendant lect annex and derive grudging support from the ambient dialects of their choosing. This, above all else, has been the cause of friction between Creole nationalists and speakers of the lexifier language, when the latter group have unthinkingly assumed that the Creole is a variety 'of their language and have consequently judged it to be debased and inferior — with all the colonialist associations that the judgement inevitably evokes. 6 3. Bilingualism and contact Bilingualism, rather like synonymy, is a concept at the mercy of its definition. Cast the net too widely ('the ability to convey a simple message in more than one language') and the result is banal; cast it too narrowly ('native command of two languages in all functional domains') and most of the interesting cases are excluded — much as 'synonymy' becomes useless if equated with total substitutability in all contexts. Notable linguists have, in fact, espoused definitions not far removed from these two extremes: Bloomfield (1935: 55 — 56) favoured the strict requirement of "native-like control of two languages", while apparently believing that

Representations of Romance: contact, bilingualism and diglossia

11

the attainment is not particularly rare and is, in principle, open to the exceptional second-language learner; whereas Haugen (1953: 7), probably in a conscious bid to shift the research paradigm, set the ability to produce "complete meaningful utterances" as the first benchmark of bilingualism. 7 (For discussion of maximalist, minimalist and intermediate stances on this issue, see Hoffmann 1991: 21—22.) In this context, it is worth noting that the popular image of the bilingual as someone who has learned two languages from infancy and is equally at home in both — an image which stresses the natural acquisition of bilingualism and its balanced outcome — is more in tune with Bloomfield than with his more liberal successors. 8 The balance of the two codes, or alternatively the degree of dominance of one of them, constitutes only one of the dimensions of bilingualism identified in the increasingly elaborate typologies now advocated. Baetens Beardsmore (1986), for instance, distinguishes over thirty types of bilingualism on the basis of different combinations of productive and receptive skills, while Hamers and Blanc (1989: 8 — 11) offer a simpler schema of variables, recognizing "relative competence, cognitive organization, age of acquisition, exogeneity, social cultural status and cultural identity", and Hoffmann (1991: 16 — 27) favours the profiling of individuals and communities using nine parameters of variation — with the twin aims of dispensing with preconceived labels and emphasizing the relativity of all aspects of bilingualism. The increasing sophistication of the field seems to have outpaced its terminology. The terms 'bilingual' and 'bilingualism' themselves remain in widespread use in the technical literature, despite their polyvalency, and researchers needing greater precision are usually content to add a prequalifying adjective (such as 'compound', 'coordinate', 'societal' bilingualism) rather than resorting to wholesale neologism. The modest innovation proposed by Hamers and Blanc (1989 [1983]), for instance, to distinguish 'bilinguality' as the psychological attribute of the bilingual individual, from 'bilingualism' in its socialpsychological and societal aspects, has won little approval and fewer imitators (Hoffmann 1991: 10). 9 Research on bilingualism has tended to split into two traditions each with its own goals and methodology, one concerned with psycholinguistic processes within the individual, and the other with the social-psychological repercussions for individuals and groups. The second tradition has laid emphasis on the factors determining language choice in particular interactions and language loyalty in the long term (Appel — Muysken 1987); on the educational implications of bilingualism and biculturalism (see Agar 1991; Hoffmann 1991: 1 1 8 - 1 3 5 ; Romaine 1989: chapter 6;

12

John Ν. Green

and the articles by Lüdi and Posner below); and on the sometimes hazy boundary between societal bilingualism and diglossia (see Laforge — Peronnet 1989; Martinet 1982; Tabouret-Keller 1982; and section 4 below). Within individual bilingualism, investigations have concentrated on the measurement of attainment in the two languages (surveyed in Baetens Beardsmore 1986: chapter 3; and more sceptically in Hamers — Blanc 1989: 1 4 - 2 9 ; and in Hoffmann 1991: 148-153); on whether bilingualism once attained remains constant or is liable to decay under unfavourable conditions (Hyltenstam —Obler 1989); and on psycholinguistic aspects of processing, with particular reference to modularity theory (Bialystok 1991 a, 1991 b; Sharwood Smith 1991). One object of fascination, precisely because it is so elusive and difficult to test, continues to be the mental representation, or set of representations, corresponding to 'coordinate' and 'compound' bilingualism (terms coined in Ervin —Osgood 1954 to designate respectively dual or single cognitive systems underlying the two linguistic codes). Here, as elsewhere in empirical investigations of psycholinguistic constructs, the findings are open to multiple and contradictory interpretations, either because experiments based on large samples fail to hold the many variables constant or because individual case studies, however meticulously observed, raise doubts as to their typicality. For instance, a recent longitudinal study of bilingual acquisition in a small child (Houwer 1990) uses the observed low level of mutual interference between morphosyntactic systems to argue in favour of the hypothesis that the two codes develop separately; but it is impossible for an observer to judge how far the child's performance is shaped by the well-focused nature of the input from the two standard languages in contact, or indeed constrained by the intensity of the monitoring. 3.1 Contact as a cause of change In his now-classic account of contact-induced change, Weinreich (1953: 1) defined the bilingual individual as the locus of contact and, by implication, the agent of transfer. Bilingualism can thus lead to change in two ways: either by percolation of linguistic features from one system to the other (usually known as interference) or by increasing the speaker's opportunities for choice of language and thereby making possible a longterm shift in preference for one language, to the detriment and perhaps ultimate abandonment of the other. Since Weinreich's goal was to elucidate the general mechanisms of change that are potentially activated whenever two distinguishable codes are in contact, he felt able to disregard the question of their sociolinguistic status: whether the varieties in contact

Representations of Romance: contact, bilingualism and diglossia

13

are usually classified as languages or dialects is immaterial to mechanisms of interference, though he did concede that the normal sense ascribed to bilingualism would not extend to speakers 'merely' adept in closely-related dialects (1953: 2). In a more recent account of the effects of contact between related systems, Trudgill (1986: 40) envisages change taking place through linguistic accommodation in face-to-face interaction (see section 2 above). This leads him to challenge Weinreich's disregard for the status of the codes in contact. Assuming that 'languages' are by definition not mutually intelligible, Trudgill contends that bilingual transfer between language systems can be prompted by communicative needs, whereas interference between mutually comprehensible dialects, though certainly not unknown, lacks communicative motivation. His model thus invites the assumption that bilingual transfer between languages will be deliberate, memorable and sustainable, whereas bidialectal transfer is likely to be adventitious and transient. Weinreich, on the other hand, considered all interference between codes to be unintentional, so that the issue of speaker motivation simply did not arise. In practice, this fundamental divergence of view does not prevent Trudgill and Weinreich from making similar predictions as to the outcome of contact, as can be seen in Weinreich's assertion that "The greater the difference between the systems, ... the greater is the learning problem and the potential area of interference" (1953: 1), with the obvious corollary that there is less scope for interference between closely related dialects. Abstract potential, however, is not the same as realization, which implies choice and opportunity. We may legitimately ask whether speakers of neighbouring dialects, who would normally have more opportunities for bidialectal exchange than they would for properly bilingual interaction, might have quite different motives for keeping their linguistic systems distinct. This brings us back to the Le Pagean concept of identity (see section 2 above): well-disposed speakers accommodate to the linguistic usage of their interlocutors to the extent that the modification does not undermine their own chosen identity or bring them too close to the linguistic behaviour of a social group from which they have intentionally distanced themselves. This, as Labov has demonstrated in a long series of illuminating studies (see especially 1963,1972 b, 1992), constitutes the social brake on change, or even the motive for reversing a change in mid diffusion.

14

John Ν. Green

3.2 Bidialectalism and convergent change in Romance While of course it is not to be expected that the psycholinguistic processes involved in the bilingual storage and production of varieties of Romance will differ from putative universal patterns, or that the processes themselves will change over measurable time, Romance does offer a challenging testbed for theories of bilingual acquisition, performance, and contactbased change. In particular, the rich diversity of linguistic forms across lects which still maintain a high degree of structural overlap and intelligibility, throws into sharp relief the problem of distinguishing between bilingualism, conceived as the mastery of two quite distinct linguistic codes, and various shades of bidialectalism. If bidialectalism is to be used as a tool for the analysis of accommodation between individuals and groups of speakers, a precondition of its application is the ability to measure linguistic distance and to locate dialect boundaries. Both are problematic within European Romance, where the dialect continua and catenate intelligibility produced by gradual historical divergence are criss-crossed by countervailing (re-)convergent trends particularly evident in the present century (see Sanga 1981; and Posner, Monteagudo — Santamarina, and Trumper, this volume). Since Romance attempts to elaborate a theory of dialectometry grew out of classic dialect geography and the wish to make the best use of linguistic atlases, it was natural that the structural divergence of systems located in Euclidean space should be seen as directly correlated with geographical distance (Goebl 1991; Walter 1990). This was the premise underlying the pioneering work of Jean Seguy, and it has borne valuable fruit in the hands of his associates and successors, notably in the triangulation technique developed by Guiter under the title of methode globale (see, among many revealing applications, Guiter 1991 on dialect frontiers in southern France and northern Spain, and Lazard 1985 on the delimitation of Friulian). 10 However valuable in their own right, approaches predicated on geographical distribution do not address the issue of comprehensibility and its limits across non-contiguous lects. A promising alternative approach, based on multivariate analysis and normal-distribution testing, is now being developed by Belgian researchers seeking a statistical validation of 'supralocal' dialects, a construct which they believe is particularly revealing for patterns of morphological divergence (see Melis et al. 1984; and Verlinde - Derinck 1991)." Though dialect and language boundaries can, of course, subsist without bilingualism, the history of Romance has been one of almost continuous permeability to outside influences, and within Romance linguistics there

Representations of Romance: contact, bilingualism and diglossia

15

has been a strong tendency to view language contact as a powerful cause of change. In the context of linguistic geography and the cartographical representation of variation, 'contact' is perhaps an unfortunate term, implying friction along the interfaces between discrete zones. Contact in this limited sense might amount to the importing of culture-specific vocabulary by adventurous individuals who had crossed to the other side (in much the same way as early transatlantic explorers brought back previously unknown plants and fruits together with their names in Arawak, Nahuatl and other Amerindian langages); such contact does involve language learning and transfer within the individual, but on a far smaller scale than implied by any standard definition of bilingualism (see section 3 above). Transfer without significant bilingualism could also occur along a dialectal interface if speakers on one side became aware of a regular and salient pronunciation — such as the substitution of a uvular [R] for an apical trill — which they chose to imitate, originally perhaps in a spirit of parody. The need to distinguish different kinds of contact has led Dardel (1991) to refer to isoglosses representing the limits of structural innovations as 'inorganic', in contrast to the organic ones where there is regular interchange in the minds of bilingual individuals. In Dardel's view, the latter type now predominates, many inorganic frontiers having developed during the period of Common Romance and become fixed shortly afterwards. Contact in Weinreich's (1953) sense, and the structural interference to which it can lead, have been well studied and illustrated by Romanists (see, for instance, Sala 1985, 1988; and Meo Zilio, this volume). In many documented cases, the interference takes place not as a result of infant bilingualism but following the acquisition of a second language by adult learners. Particularly revealing are the adaptive patterns shown by migrants between Romance-speaking areas (see Lüdi and Vizmuller-Zocco, this volume). In such cases, because of the similarity of codes, migrants do not uniformly perceive their task as the learning of a foreign language, nor are they always conscious of backwash effects on their first language: Grosjean and Py (1991), for instance, document a tendency among Spanish migrant workers in the Neuchätel area to caique idioms and syntactic constructions from French into Spanish, and to give grammatical acceptability judgements implying that they have edited out of their first language some prominently Spanish characteristics. From the evidence hitherto available, it is difficult to know whether the adaptive changes observed in their language (in Le Page's Sense 2; see section 2 above) would be reversed after a period of reimmersion in the original cultural

16

John Ν. Green

matrix, or whether, as Grosjean and Py believe, a permanent retrospective change has been wrought in their (Chomskyan) competence. If the latter, the mechanisms of the change remain explicable by accommodation theory, but the extent and permanence of the change go far beyond what its proponents would normally envisage.12 Perhaps the best-known (and certainly the most notorious) explanations advanced for externally motivated change in Romance — as befits an area whose linguistic history has been significantly shaped by invasions and population movements — are the concepts of substratum and superstratum championed by Wartburg (for a critical assessment, see Kontzi 1982; Pellegrini 1980; and Schlemmer 1983). Both are concerned with the consequences of second-language learning by adults, with the difference that substratum effects are said to well up from the first language of speakers compelled to learn the language of foreign conquerors while superstratum effects trickle down from a dominant group which has chosen to learn the local language. In both cases, a period of societal bilingualism must have accompanied the linguistic changes, with transitional variants persisting long enough to outlive the corrective pull of models originally acknowledged to be 'native'. In a context characterized by unequal power relations, it is a moot point whether the resulting symbiosis would be more akin to bilingual contact as envisaged by Weinreich or to conflictual diglossia (see section 4 below); certainly Marcellesi (1981b) is happy to abandon the term 'contact' when, as here, there is complete territorial overlap of the two codes. Rather as analogy is offered as the intelligent brake on the destructive regularity of sound laws, so bilingual contact may be presented as the convergent force counteracting 'natural' divergent change. The evidence is clear that bilingual contact can act as a conduit for change in the observable behaviour of individuals. Whether the contact leads to longterm change or merely increases the range of synchronic variation, must depend on other, sociolinguistic factors. Especially when the number of individuals affected is fairly small and their bilingualism is exogenous to the community (Hamers — Blanc 1989: 10 — 11), there is no reason to assume that any resultant changes will be diffused and adopted by nonbilinguals; they are as likely to be dismissed as mistakes. Diffusion of new variants takes place when chains of individuals are sufficiently pleased with the innovations to want to imitate and communicate them to others — a process no doubt accelerated when the innovations originate in a high-status group; real change only occurs when new generations of language learners construct different underlying systems from the surface

Representations of Romance: contact, bilingualism and diglossia

17

variants to which they are exposed. In the history of Romance there are well-documented instances of prestige variants being imitated and diffused, but without a major impact on longer-term evolution. When, for instance, Catherine de Medici married Henri II in 1533 and brought with her an abundant Italian-speaking retinue, imperfect bilingualism at Court gave rise to a fashion for very italianate French which seems to have spread to upwardly aspiring circles beyond the Court; but it passed, bequeathing to modern French some technical vocabulary (most of which probably came via artisanal links) and few if any detectable consequences in morphosyntax. 1 3 Within Romance, there is little doubt that the divergence produced by early dialectalization and abetted by the breakdown of communications after the collapse of the western Empire, has been counteracted since the Renaissance by growing cultural contacts and inter-Romance borrowing, but above all by the availability, at a crucial stage in the standardization of western Romance, of Latin as a standard and roofing language (for a summary of unifying factors in Romania, see Malkiel 1977). Once there is sufficient momentum, convergent languages tend to innovate independently in similar ways and to borrow from similar sources, so accelerating their convergence (Renfrew 1989). This is not to say that convergence at a macrolinguistic level, produced according to Stehl (1989) by vertical contact between a dominating language and its dominated dialects, cannot be interrupted or cut across by more localized currents if, for instance, convergent processes become refocused onto the relationship between a national language and the related dialects spoken within the same territory. That is the phase through which most Romance languages have been passing for a century or so. In the terms we have developed above, successive generations of speakers of non-prestigious varieties have probably experienced: monolingualism in a regional lect with at best passive competence in the national language; bidialectalism; and complex monolingualism in a koine no longer identical to the former prestige variant. As the goal of national convergence nears its attainment, the stage now seems to be set for a major shift in attitudes. 4. Diglossia Few concepts in linguistics have proved so successful, so durable and so adaptable to new circumstances as that launched by Ferguson in his 1959 article "Diglossia". As Ferguson himself noted, the concept of functional differentiation of linguistic varieties was far from new, and the term he proposed to apply to it was novel only to the extent of a latinate suffix

18

John Ν. Green

and anglicized pronunciation. He derived it from diglossie, which is ascribed by Prudent (1981) to the Hellenist J. Psichari, who used it in his two-volume history of modern Greek to describe the alternation of katharevousa and dhimotiki and revived the term, some forty years later, in a popular and influential article published in Mercure de France. Appropriately, the same alternation figures as one of the four case studies in Ferguson's article, together with Cairene Arabic, Swiss German, and Haitian Creole French. For each pair of 'High' and 'Low' varieties, Ferguson was able to demonstrate striking parallels in both intrasystemic characteristics and sociolinguistic attributes. Η is the prestige variety, standardized, dignified by grammatical tradition, not acquired naturally but learned through schooling, and giving access to a valued literary and intellectual heritage. L is its workaday counterpart, used for everyday transactions, family conversation, and — if written at all — for political slogans and folktales. The functional differentiation is absolute, and misjudged usage of one or other variety risks provoking ridicule. Surprisingly, instances of diglossia are often stable and remarkably persistent — a duration of several centuries is not uncommon, and the communities themselves often see nothing problematic in the situation. Ferguson's analysis of the purely linguistic characteristics of Η and L and the extent of their mutual influence, is predicated upon one of his premises that was soon to be hotly contested: that diglossia involves two varieties of the 'same' language. This assumption necessarily colours, if not determines, the range of possible outcomes, among which is the gradual emergence of an interlect preserving most of the morphosyntactic features of L but with heavy admixture of Η vocabulary. Such an interlect, Ferguson claims, has a fair chance of evolving into the standard language, whereas the Η variety will rarely do so unless it is also the standard language of an outside, and prestigious, community. Indeed, he sees standards in general as the product of interacting linguistic forces, and doubts whether their adoption ever owes much to legislation or to the fervour of their proponents. We cannot survey in detail all of the vicissitudes of a term which, over thirty-odd years, has been subject to the usual evolutionary adjustments, periodic redefinition and what some would see as factional expropriation (for a fairly neutral account of its history see: Berger 1990, Mackey 1989, McConnell 1989, and Hudson 1992; and for a more polemical approach, Kremnitz 1981 and Prudent 1981). One of the earliest modifications has proved the most enduring. Ferguson, by directing attention firmly onto

Representations of Romancc: contact, bilingualism and diglossia

19

societal bilingualism, had rather side-stepped the relationship between diglossia and individual bilingualism; and by concentrating on case studies where the two varieties were historically related, had arguably biased the debate on interlects and standardization (if Η and L are genetically related, H, though prestigious, is by definition in decline — a decline that can only be reversed by a political decision to commit massive resources to an educational programme; see Fishman 1991). It is Fishman who is credited with reorienting discussion towards the "domains of language choice" (1965: 86) open to bilingual individuals, and with establishing the conceptual independence of diglossia and bilingualism (1967), realigning the terms so that bilingualism relates primarily to individuals and diglossia to society. Fishman's four-term system, distinguishing bilingualism with or without diglossia from diglossia with or without bilingualism, has survived some determined attacks to become canonical; but it was probably Fishman's recognition that diglossia must be freed from the straitjacket of genetic relationship that did most to launch an extensive and fruitful programme of research. Now, in the words of Mackey (1989: 28): "on a pu redefinir la diglossie comme etant simplement la repartition des fonctions langagieres dans la societe, quel que soit le nombre ou la variete des langues qui en font partie". Even so, a recent critical survey (Spolsky 1988) discusses language functions and social attitudes under the heading of "bilingualism", with only a single throwaway remark on diglossia. And diglossia, unlike bilingualism — as McConnell (1989) reminds us — for all its original attractions and more recent halo of controversy, has remained a linguist's notion, never catching the popular imagination or developing a life of its own. 4.1 Romance applications Broadly speaking, there have been two strands to the synchronic application of diglossia in Romance. The first, which may be thought of as evolutionary, has tested the original Ferguson/Fishman concept in new linguistic contexts, refining, modifying, and sometimes refuting its claims to usefulness; appropriately, since one of Ferguson's original case studies dealt with Haitian Creole, the relationship of Creoles to their lexifier languages has been a persistent theme. The second, which quickly acquired ideological overtones, has emphasized the conflictual aspects of diglossia, especially in the context of varieties of European Romance which have been downgraded, denied recognition, or actively suppressed for political motives.

20

John Ν. Green

The earliest and arguably still the best-known instance of planned linguistic discrimination in Romance is, of course, the espousal by the postrevolutionary Convention Nationale in 1794 of Gregoire's recommendation to promote French as a symbol of national unity and progress, with the concomitant aim of crushing — Gregoire used the word aneantir — patois and regional varieties (see Grillo 1989 and Joseph 1991).14 Although the policy remained formally in place for the best part of two centuries, during which French did rise to the status of major world language and the dialects did decline, it is now generally viewed as having had little direct effect on linguistic behaviour except in irredentist enclaves where it may actually have been counterproductive (for a critical evaluation of the linguistic legacy of the Revolution on the occasion of its bicentenary, see especially Bochmann 1989 a, 1989 b, and Marcellesi 1989, the latter examining the opposing discourses of national unity and individual human rights). In fact, the hegemony of French was mainly due to vigorous colonial expansion, and the decline of the dialects to improved communications, mobility, and universal primary schooling — conditions which were replicated elsewhere in 19th-century Europe and with similar linguistic consequences. At most, the Revolutionary policy shaped attitudes in favour of a national standardized norm, for which the only practical model was the educated usage of Paris, and so established an attitudinal frame that has persisted largely unaltered to the present day (see Paltridge-Giles 1984). Although Ferguson's original conception of diglossia certainly carried the implication of social stratification and elite access to literacy (and thence to the Η form and social mobility), it was only through the struggle for the linguistic emancipation of Catalan that diglossia became overtly linked with power and dominance. Catalan was ruthlessly repressed during the early years of the Franco dictatorship, as was Basque and, indeed, any regional linguistic and cultural difference that might feed autonomist aspirations and so undermine Spanish national unity. Despite the denial of any official status, public discouragement, and severe restrictions on its use in the printed media, Catalan was showing clear signs of resurgence by the mid 1960s, so that its rehabilitation was an urgent and popular priority for the Generalitat after the restoration of partially devolved government a decade later. The linguists who did much to raise the international profile of Catalan — notably Aracil, Badia i Margarit, Ninyoles and Vallverdu (for an evaluation of their work, see Posner and Strubeil i Trueta, this volume) — originally wrote, not of 'diglossia', but of 'linguistic conflict' in Catalonia and Valencia.

Representations of Romance: contact, bilingualism and diglossia

21

Indeed, before the theoretical widening of the term, to describe the Catalan situation as diglossic would have been to concede two of the principal arguments. First, the undoubted genetic relationship of Catalan and Spanish is that of co-eval branches and carries no implication of historical precedence. Second, and even more important, the division of functions under the Franco regime was in no sense organic, but imposed: each language had previously enjoyed the status and degree of elaboration necessary for use in any of Ferguson's list of roles, and each had its internal Η and L registers, as tends to happen in any 'elaborated' language (see Muljacic below). The claim made on behalf of Catalan was for the restoration of usurped rights, not for the internal development of the language to make it suitable for Ή ' functions. For this reason, the vigorous language planning policy adopted by the Generalität has paid relatively little attention to ' n o r m a l i z a t i o n ' (or corpus planning) and considerably more to 'normalization' (elsewhere called status planning; again see Strubeil i Trueta, below, and for a more partisan approach to three case studies, Lamuela 1987). While Fishman's realignment of the terms made it possible to apply diglossia in a context like that of Catalan, where throughout the repression there had been widespread bilingualism, the explicit linking of diglossia and conflict owes much to the Occitanist movement and the ideologically-committed brand of sociolinguistics associated with the journal Lengas (Montpellier: Universite Paul Valery, 1977 — ) and in particular with the names of Lafont and Kremnitz (for a sober assessment of the Occitan revival and its theoreticians, see Schlieben-Lange, this volume). The work of this prolific group has undoubtedly influenced other autonomist movements in Romance, in both theoretical and practical aspects of language planning, for instance in the choice of a single developmental norm or the deliberate recognition of 'polynomic' diversity — as advocated for Corsican by the Corte research group (see Thiers, this volume, and more generally: Boyer 1986; Calvet 1987; G a r d y - L a f o n t 1981; Jardel 1979; Kremnitz 1981, 1987, 1991; Lafont 1982, 1984; Marcellesi 1979, 1981 a, 1981 b; and Marconot 1990). A particular point of attack for this group has been Ferguson's contention that diglossia can be stable. From a conflict perspective, stability could only be the product of an exact balance of strongly countervailing forces, which is rejected as implausible in anything but the short term. Rather, Kremnitz views the relationship as a dynamic between languages which "sont toujours des poles d'attraction ou des poles de refus, souvent les deux ä la fois" (1981: 71) and, indeed, cites approvingly Lafont's claim that diglossia is always an "in-

22

John Ν. Green

dicateur de crise" (Kremnitz 1987: 211). In a recent re-evaluation of theoretical constructs, Kremnitz (1991) distinguishes four terms — bilingualism, diglossia, linguistic conflict and diglossic usage (what Lafont calls "fonctionnements diglossiques") — and, on the issue of complementarity, argues that diglossia excludes neutrality: the exchange of one variety for the other will always be interpreted as having glottopolitical significance, so that speakers are denied the freedom of consenting bilinguals to switch 'innocently' between the two varieties. The relationship of a Creole language to its lexifier is, as we implied above, potentially one of diglossia and conflict. The conflictual model has been adopted by a number of Francophone creolists, working primarily on the French-lexicon Creoles that remain in close political contact with French, notably those of Guadeloupe and Martinique, which are still territorially part of France and where French is therefore the language of official administration and education, despite the fact that almost the whole population uses Creole as a family and transactional language, so that the great majority of children join primary school as monolingual Creole speakers. The resulting educational tensions in Martinique are graphically described by Kremnitz (1983), who dwells on the personal conflict created for many teachers, caught between official policy, strong but heterogeneous personal attitudes to Creole and French, and practical educational imperatives. While such tensions may be amenable to conflictual analysis, it is less obvious that they are truly diglossic, when observers report that the linguistic behaviour of individuals is not so much polarized at the clearly defined extremes of basilectal Creole and local acrolect, as located somewhere along the continuum between the two. 15 In some ways, the most problematic case has proved to be that of Haitian, the French-lexicon Creole with the highest number of speakers and probably the greatest degree of internal dialectalization. Haiti gained independence from France in 1804 and for most of the intervening period has been ruled by a French-speaking oligarchy; but French is a foreign language for large sectors of the population, not mutually intelligible with, or shading into, Haitian Creole. Under these circumstances, some linguists argue that there is neither continuum nor diglossia. Dejean (1983), for instance, sees a bilingual dominant class in conflict with a monolingual, Creole-speaking majority (which, in turn, leads him to query the whole notion of diglossia). Valdman, on the other hand, while still rejecting Ferguson's original statement of Haitian diglossia, adopts a more measured, evolutionary approach, reanalysing some recent Creole

Representations of Romance: contact, bilingualism and diglossia

23

innovations — such as front vowel rounding and the reintroduction of post-vocalic /r/, both of which could be seen as instances of decreolization in the direction of French — as part of a crystallizing opposition between urban and rural Creole (see Valdman 1988,1991). Clearly, classic diglossia does not fit all Creole situations and undue procrusteanism is bound to lead either to untenable analyses or to conceptual collapse ("eclatement du concept", in the words of Prudent, 1981: 20), above all in those cases where three or more codes are in competition and where contact has long been severed between the Creole and its lexifier (see Christie 1990 for a microcosm case in Dominica). Nevertheless, despite strongly held views in some quarters, the balance of opinion among creolists is that diglossia remains a useful analytical tool (Winford 1985), even if the resultant analyses are somewhat idealized and depend crucially on the maintenance of traditional language attitudes (Carrington 1990). In summary, diglossia has had a major impact on studies of synchronic Romance contact, but almost always in its extended or conflictual reformulations, and rarely as Ferguson himself first defined the term. Reciprocally, Romance applications have played a major part in the refinement of the concept, questioning the stability of diglossic patterns, engaging with a dialectic of power, and adding pejorative overtones which, for conflict theorists at least, have turned diglossia into a rallying call for direct remedial action. 4.2 'Diglossia' in Late Latin / Early Romance Although our primary concerns are synchronic, a word is surely not amiss on one potential application of diglossia not addressed elsewhere in this volume, namely the relationship of Latin and vernacular during the formative period of Romance. The precise form in which 'Latin' persisted during the centuries between the fall of the Empire and the Renaissance (after which it had to be learned quite consciously as a second or foreign language) has long been a matter of speculation and controversy. A brief paragraph in Ferguson's article implies classic diglossia: "The vernacular was used in ordinary conversation but Latin for writing or certain kinds of formal speech" (1959: 337). The first part of the assertion is unexceptionable (tautologous, indeed), but the second masks some highly questionable assumptions, particularly if the comments are meant to apply to the period between the collapse of the western Empire and the Carolingian renaissance. The manifold definitions of diglossia discussed above, including Ferguson's own, require two spoken codes functioning in complementary

24

John Ν. Green

distribution and at least one written form corresponding to the Η variety, though not necessarily representing it directly. The application of diglossia to Early Romance would therefore predict that a high-register form of spoken Latin survived for centuries, stable and intact, alongside spoken Romance, and that the literate minority was fully conscious of the distinctiveness of Latin and Romance, and skilled in switching between the two codes in response to social exigencies. In the view of Wright (1982), this was indeed the conception (erroneous, and expressed here in deliberately stark terms) of an earlier generation of Romance philologists: Wright argues, for instance, that Menendez Pidal's explanation for the survival of 'semi-learned' vocabulary with conservative phonetism must presuppose the continuity of regional spoken Latin alongside true, illiterate, vernacular Romance. In fact, there is no evidence before the 9th century, and little before the 12th, for the existence either of identifiable communities speaking what would have been perceived as an archaic form of the vernacular, or of a stratum of individuals who were credited with being bilingual. There is evidence that writing was the preserve of a small elite; that it was equated not merely with a representational technique but with a complete linguistic superstructure known as ars grammatics, that it had to be learned laboriously; and that access to this highstatus knowledge was very largely controlled by the Church — to the point where scribes in later medieval monastic scriptoria may have been taught to copy symbols without thereby acquiring literacy (see, specifically, Troll 1990; and on the extent of Carolingian literacy and the value placed upon it, McKitterick 1989). The counter view to early diglossia, as advanced by Wright (1982, 1991 a), is that as Romance developed and dialectalized up to the end of the 8th century, its written form became increasingly remote and abstract, preserving grammatical features long eliminated from the spoken language, just as modern written French preserves grammatical concord where it is no longer heard, and perhaps no longer psycholinguistically real. Spoken Romance would have had dialects and high registers used by educated speakers, resulting in the complex monolingualism that we now view as sociolinguistically normal for languages serving large and diverse populations; but it would be wrong to conceive of two parallel spoken codes. What motivated the Carolingian reforms was a profound theological objection to the manifest fact that the liturgy was being enacted in diverse local forms that were not mutually comprehensible. Alcuin is believed to have devised a pronunciation, essentially based on grapheme to phoneme correspondences, that could be taught to priests,

Representations of Romance: contact, bilingualism and diglossia

25

so that the liturgy would be pronounced in the same way throughout western Christendom. There were two consequences: firstly that the rite became unintelligible to the congregation — which must have been apparent to the reformers, and accepted as the lesser of two evils; and secondly, that 'Latin' was perforce revealed as something different from Romance, an Abstandsprache in Klossian terminology. This revelation, whose far-reaching implications were almost certainly not anticipated at the time, led to the reconceptualization of Romance and Latin as separate entities, a process that has led Wright to claim that Alcuin "invented" medieval Latin (1982: 104-144; reiterated, after criticism, 1991 a: 109). Wright's new perspective on late Latin and Early Romance, though still controversial in some quarters, has been generally welcomed, even by those who remain sceptical about important points of detail. 16 The major implications for our reading of early Romance texts are now being worked out, with promising results for the interpretation of such enigmas as the interlinear glosses of San Millän and Silos (see the papers in Wright 1991 b). Nonetheless, some important questions remain unanswered, such as the motivation for early attempts to write Romance in a new orthography, the dating of general awareness of a linguistic split between Latin and Romance, and above all the point at which awareness crystallized into the widespread use of language names (see especially Janson 1991 and Lloyd 1991).17 If, as Posner argues (1992: 388-390, and in press), full awareness of linguistic distinctiveness was a Renaissance phenomenon, it is not until that period that we can properly apply the term diglossia. By then, of course, a second written code was in play and the status of the Η spoken variety had shifted, not merely to that of a second language, but to that of a revived dead language, refined by 15th-century humanists to such a forbidding extent that from then onwards it could only "vegetate as a university jargon" (Kahane — Kahane 1979: 187 — 188).18 While there can be no doubting the importance of medieval Latin as a source of lexical borrowing and an agent of linguistic convergence, it does seem that the constantly shifting relationship between Latin and Romance never perfectly fits the diglossic model, nor is helpfully explicated by it. 5. Conclusion: representations of Romance Linguists of very different persuasions use the term 'representation' in connection with the mental image that speakers form of their language. For Chomsky (1980), representation is a psycholinguistic concept, associated with such intrasystemic considerations as the storage of grammat-

26

John Ν. Green

ical and lexical information and the rule schemata underlying its retrieval. For a group of French linguists drawing inspiration from the work of Bakhtine and Moscovici, on the other hand, representation is a sociological construct, heavily conditioned by attitudes and value judgements, which occupies the intermediate ground between abstract conceptualization and perceptions of reality (see Boyer — Peytard 1990: 3 — 5; and, for a history of the notion, Peytard 1990). Representation in this sense acts as a codeterminant of behaviour and shapes what Boyer (1990: 102) calls "l'orientation des communications sociales". Significantly, Boyer exemplifies his approach from conflictual diglossia, in which the speakers' representation of their language may be eroded by subordination and acculturation to the point where they become sufficiently alienated to acquiesce in its replacement by the dominant code. The case studies of Romance bilingualism and diglossia examined in this volume, most of which adopt a mainstream social-psychological perspective, stress the intimacy of the links between language and social identity and offer ample support for the view that speaker attitudes can indeed affect linguistic systems in contact. In an article provocatively entitled, "Qui est latin?", Pohl (1988) concludes that the primary meaning of such epithets relates to language: we attribute characteristics to individuals and societies on the basis of the language we perceive them to be speaking and we use the label of the language or lect as a convenient shorthand (see, again, Paltridge — Giles 1984). Representations and beliefs do matter: speakers may borrow freely from what they perceive as another language in order to satisfy a communicative need, or simply to follow a fashion, but they may not borrow from a lect of their 'own' language without thereby affecting, and perhaps compromising, their own linguistic identity. It is not necessarily easy, however, for linguists to ascertain what language informants believe they are speaking, or to form a judgement on the appropriacy of the belief. Herman (1991), for instance, attempts to reconstruct the view of Latin held by literate speakers in the last centuries of the Empire — a view of fundamental unity, which he nevertheless argues was objectively unsustainable. 19 Schlieben-Lange, reporting on the current state of Occitan (this volume), points to a detectable — and sometimes gross — discrepancy between what speakers claim to be doing and what an impartial observer perceives to be 'objectively' true. Likewise, Meo Zilio (this volume) relates how a cocoliche-speaking informant who was asked to speak in Italian on the occasion of his first interview and in Spanish on the occasion of the second meeting a few

Representations of Romance: contact, bilingualism and diglossia

27

days later, did not demur at the requests, yet produced speech on related topics that overlapped so significantly in its structural characteristics that the analyst now cannot be sure whether or not two codes are present. (An analogous problem in the description o f f r o n t e r i z o is documented by Hensey, this volume.) If the speaker is convinced of having access to two distinct codes, on what basis, if any, is the observer to challenge this belief? The mere existence of overlapping surface realizations is no proof that they can be traced to a single underlying representation. Indeed, the very fact that the overlap falls short of identity can be argued to imply divergent representations, in the same way that for classic generative grammar the existence of ambiguous surface structures is taken as ample justification for postulating twin or even multiple deep structures, each with its own, unambiguous semantic representation. It seems likely that the existence of representations of 'a language' in Le Page's Sense 4 (see section 2 above), as opposed to the mere awareness of variable linguistic behaviour, is a fairly recent phenomenon, connected with widespread literacy, standardization, and the acceptance of prescriptive authority over language (see Milroy— Milroy 1991). Standardization implies the elaboration of something that already exists, on which the standardizing process confers the stability and prestige that result in eloquence and power (see Joseph 1987; Muljacic 1990 b, and this volume). But the precondition must be the ability to identify the variety to be singled out for standardization, and identification involves both naming and reification (see Marcellesi 1981 a: 8, on the crucial importance of naming as a symbol of linguistic autonomy). 2 0 Indeed, an important contributory factor may be metalinguistic: the expectation that speakers should perceive the distinctness of, and be able to name, the language or lect they profess to control. Informants of limited educational attainment when consulted on their use of a named lect, may give more or less coherent replies, but they do not usually round on the investigator by challenging the legitimacy of the language label embodied in the question. As the consciousness of language representations and the potential for conflictual diglossia are heightened in literate societies, especially through the adoption of explicit language planning programmes, it must be expected that politicization will lead to new demands for democratic participation in, and control over, decisions affecting language status. Whether this will lead to better decisions and the resolution of linguistic conflict, only time will tell. What is certain is that, in the case of the multiple instances of linguistic overlap in Romance, forces have been set in motion that will be difficult, if not impossible, to reverse.

28

John Ν. Green

Notes 1. For a discussion of the application of generative theory to Romance languages, see Green 1981. In the same volume, Lavandera (1981) surveys Romance applications of sociolinguistics, and Rogers (1981) attitudes towards linguistic nationalism. For a programmatic view of the priorities for research on language in its social context, see Labov 1972 a, which retains its freshness and vitality. 2. Chomsky views linguistics as a part of psychology: "The theory of particular and universal grammar, so far as I can see, can be sensibly regarded only as that aspect of theoretical psychology that is primarily concerned with the genetically determined program that specifies the range of possible grammars for human languages and the particular realizations of this schematism that arise under given conditions. One may perfectly well choose to study language and grammar with other purposes in mind and without concern for these questions, but any significant results obtained will nevertheless be a contribution to this branch of psychology." (1980: 202). 3. For enlightening applications of the model see, among others, Christie 1990 and Tabouret-Keller 1987. The notion of identity is common to a number of social psychological approaches to language (see, for instance, Eastman 1984 and Ross 1979). Le Page's work on attitudes and accommodation has much in common with that of scholars like Howard Giles and Wallace Lambert, whose ideas have been fruitfully applied to Quebec French and are more fully discussed by Bourhis and Lepicq, this volume. 4. It is beyond the scope of this section to give a systematic evaluation of the many taxonomies proposed by Romance scholars for the Romance languages and lects. For enlightening discussion, see Hall 1974, Malkiel 1978, and Posner 1980. The monumental Lexikon der romanistischen Linguistik (Holtus et al. 1988 — ), which aims to offer detailed coverage of every variety of Romance and which refers to each by the common substantivized adjective (as in "Okzitanisch"), is so vast as to be able to avoid conceding or withholding language status in its choice of titles, though, to be sure, there is an implication of relative importance in the number of pages and articles devoted to each variety. 5. Muljacic (1990 a) documents a Romance variety spoken near Scutari/Shkoder, Albania, which he calls 'Labeatic' and which may be a survival of southern Dalmatian. See also Posner, this volume, sections 2.2 and 2.5. 6. For detailed discussion of the relationship of the Romance Creoles to their respective lexifiers, see Green 1988, and the chapters in this volume by Hazael-Massieux — Robillard (on French), Lipski — Schwegler (on Spanish), and Stolz (on Portuguese). Stolz (section 5.1) is impressed by the mechanism proposed by Dulce Fanha as the most likely explanation for lusitanization: bilingualism leads to structural insecurity and consequently to interference via code-switching. 7. It is clear from a survey of bilingualism and contact written twenty years later (Haugen 1973), in which he reiterates that too narrow a definition leads to "virtual sterility", that Haugen had defined his terms quite intentionally in order to highlight the need for remedial research on aspects of bilingualism that were proving a handicap for diverse minority groups in America: "However, when bilingualism surfaced as a social or personal problem, it was nearly always because of some disability or handicap imposed on one or the other of the languages in contact. The bilingual was observed to fall short of the norms in one or both languages, and his failure was rightly or

Representations of Romance: contact, bilingualism and diglossia

8.

9.

10.

11.

12.

13.

14.

29

wrongly attributed to his knowledge of the other." (1973: 507). Mackey (1956, 1962) also used definitions as a means to reorientate research. Among the many reliable manuals on bilingualism now available (with inevitable differences of emphasis and quality of coverage), we should cite: Appel — Muysken 1987; Baetens Beardsmore 1986; Grosjean 1982; H a m e r s - B l a n c 1989; Hoffmann 1991; and Romaine 1989. Psycholinguistic aspects of processing and the maintenance of bilingual competence are addressed respectively in Bialystock 1991 b, and in Hyltenstam — Obler 1989. Spolsky 1988 is a brief survey of (chiefly Anglo-American) research preoccupations, concentrating on social aspects of bilingualism. For a more European perspective on bilingualism and language contact, see: Denison 1988; Lüdi —Py 1986; and the conference papers collected in Lüdi 1987, and in Ureland 1980, 1981, 1982 and 1985. Zuanelli Sonino 1989 is an interdisciplinary reader on literacy, as experienced in both monolingual and multilingual communities. The word 'bilinguality' is used by Weinreich (1953: 81) in a context referring to a property of the individual, but he does not offer a definition or appear to invest it with any theoretical significance. The field has unfortunately been marred by disputes between prominent scholars on the validity of statistical techniques and the extent of their originality: see Goebl's three-volume account of dialectometrical methods (1984) and its hostile review by Guiter (1985), and their exchanges at the Trier round table on dialectology (Goebl 1991; Guiter 1991). It is not within our scope to give a detailed survey of recent developments in dialectology, which were more fully covered in earlier volumes of this series (see, for example, my Introduction to Posner—Green 1981). For appraisals of current trends in general and Romance dialectology, see respectively Alinei 1991 —92 and Pfister 1987. For discussion of continua and the Schuchardtian concept of catenate intelligibility, see Hockett 1958: 3 2 2 - 3 2 6 ; I o r d a n - O r r - P o s n e r 1970: 5 1 - 5 3 ; and Posner 1980: 1 2 - 1 8 , and below. The effects of dialectal and other types of linguistic variation on first and second language acquisition are investigated in Adamson 1988 and Preston 1989 b, with Preston 1989 a concentrating on attitudinal factors in dialect shift. General principles of research on multilingualism are addressed by the papers in Gendron —Neide 1986, and Raith et al. 1986. Patterns of adaptive change resulting from the contact of English and varieties of French in Canada have been well studied: see Mougeon — Beniak 1991, and Bourhis —Lepicq, this volume. Berruto et al. 1988 is a detailed study of the effects of multilingual contact on formal registers of Italian: the authors examined the written work of Italian students at the University of Zurich, finding that many morphosyntactic features are very unstable, and that non-native speakers and bilinguals make similar numbers and types of mistakes when compared with native speakers. In a more light-hearted vein, Wandruszka (1985) reports on more tolerant attitudes towards southern pronunciation variants, and even the adoption into Parisian French of some 'francitan' expressions — as a result, he avers, of the success of the French rugby team, almost wholly composed of southerners. It is not clear whether the innovations can survive a series of defeats or replacement team membership. The much earlier Edicts of Villers-Cotterets, of 1539, had certainly had discriminatory effects, but the ostensible legislative goal had been to resolve potential ambiguities in the interpretation of statutes. Schlieben-Lange (this volume) debates whether the crucial phrase of the Edicts, "en langage maternel frangois et non aultrement", was meant to

30

15.

16.

17.

18.

19.

20.

John Ν. Green tolerate or exclude southern vernaculars. In section 4.2 below, we discuss whether the Carolingian reforms amounted to a conscious act of language planning. We cannot here debate the theoretical validity of the continuum postulate, which is widely accepted and used in anglophone creolistics (for example, Bickerton 1975), but regarded with some suspicion elsewhere, either because empirical studies fail to locate all speakers within a single dimension of variation (see Le Page — Tabouret-Keller 1985: 180 — 200), or because a continuum may be held to imply inevitable decreolization as the acrolect exerts an irresistible 'pull' on Creole usage. It is beyond our scope to examine all the points of detail. The reviews, some of them substantial, include: Steven D. Kirby, La coronica 12 ( 1 9 8 3 - 8 4 ) : 2 9 5 - 2 9 7 ; Saul Levin, General linguistics 24 (1984): 1 9 4 - 1 9 9 ; Paul M. Lloyd, Hispanic review 52 (1984): 367 — 377; Francisco Marcos Marin, Revista de filologia espanola 64 (1984): 129 — 145; Ian Michael, Modem language review 83 (1988): 925 — 926; Frank H. Nuessel, Language 60 (1984): 1 8 4 - 1 8 5 ; Arnulf Stefenelli, Revue de linguistique romane 47 (1983); 4 4 6 - 4 4 7 ; and Maximo Torreblanca, Journal of Hispanic philology 7 (1982 — 83): 141 — 143. Of these, Kirby and Lloyd are completely convinced by the basic hypothesis, while Torreblanca thinks it is wholly erroneous. Wright originally claimed that the reconceptualization dated from the early 9th century in Carolingian France and sometime after 1080 in northern Spain; more recently, he has accepted that these are 'first possible' dates for awareness in the minds of some exceptional individuals, with the idea being diffused very much more slowly through the general population. Posner (in press) takes issue both with Wright's chronology and with his explanation for the emergence of written Romance; the invention of a Romance orthography in northern France, she believes, took place later than generally supposed and owed much to pre-existing Germanic models. The precise nature of medieval competence in Latin remains obscure. It is noticeable that scholars do not usually speak of 'bilingualism'. Can one, indeed, be bilingual in one living and one dead language? Balibar's (1985) history of French from Carolingian times to the Republic, faces the problem in its title, by speaking of 'colingualism'. Varvaro 1991 offers support from a rather different angle: the 'collapse' of Latin after the fall of the western Empire can be seen as the loss of the centripetal force that had previously held together variants that were obvious without being destructive. Perceptually, what is important is the moment when speakers "lose, or more precisely feel they have lost, the consciousness of belonging to a whole and acquire instead a sense of local identity" (Varvaro 1991: 48). It is well known that language descriptors in Latin were adverbs derived from adjectives denoting geographical or ethnic origin, hence romanice meant '[speaking] in the Roman way / like the Romans do'. The change from adverbial reference to nominal designation (via a metaphor lingua latina 'the Latin tongue'), marks the beginning of reification of language, as well as the sociolinguistic fact of its autonomy. For discussion of the importance of naming as a trigger for the consciousness of linguistic dichotomy in Late Latin / Early Romance, see Janson 1991; Lloyd 1991; Uytfanghe 1991; and section 4.2 above.

Representations of Romance: contact, bilingualism and diglossia

31

References Adamson, Hugh D. 1988 Variation theory and second language acquisition. Washington DC: Georgetown University Press. Agar, Michael 1991 "The biculture in bilingual", Language in society 20: 167—181. Agard, Frederick B. 1984 A course in Romance linguistics. Volume 1: A synchronic view. Washington DC: Georgetown University Press. Alinei, Mario 1 9 9 1 - 9 2 "Whither dialectology?", Quaderni di semantica 12: 2 0 7 - 3 3 3 and 13: 9 3 139. Ammon, Ulrich — Norbert Dittmar — Klaus J. Mattheier (eds.) 1987 — 88 Sociolinguistics. An international handbook of the sciences of language and society. 2 vols. Berlin: Walter de Gruyter. Appel, Rene — Pieter Muysken 1987 Language contact and bilingualism. London: Arnold. Baetens Beardsmore, Hugo 1986 Bilingualism: basic principles. (2nd edition.) Clevedon, Avon: Multilingual Matters. Balibar, Renee 1985 L'institution du fran^ais. Essai sur le colinguisme des Carolingiens ä la Republique. Paris: Presses Universitaires de France. Berger, Marianne R. 1990 "Diglossia within a general theoretical perspective: Charles Ferguson's concept 30 years later", Multilingua 9: 2 8 5 - 2 9 5 . Berruto, Gaetano — Bruno Moretti — Stephan Schmid 1988 "L'italiano di parlanti colti in una situazione plurilingue", Rivista italiana di dialettologia 12: 7 - 1 0 0 . Bialystock, Ellen 1991 a "Metalinguistic dimensions of bilingual language proficiency", in: Ellen Bialystock (ed.), 1 1 3 - 1 4 0 . Bialystock, Ellen (ed.) 1991 b Language processing in bilingual children. Cambridge: Cambridge University Press. Bickerton, Derek 1975 Dynamics of a Creole system. Cambridge: Cambridge University Press. Bloomfield, Leonard 1935 Language. (3rd, British, edition.) London: Allen and Unwin. Bochmann, Klaus 1989 a "La glottopolitique dans la Revolution Fran?aise et ses reflets Italiens", Beiträge zur romanischen Philologie 28: 237 — 246. 1989 b "Die Französische Revolution und ihre Sprachpolitik — nationale and internationale Kurz- und Langzeitwirkungen", Zeitschrift für Phonetik, Sprachwissenschaft und Kommunikationsforschung 42: 561 — 563. Boyer, Henri 1986 "«Diglossie»: un concept ä l'epreuve du terrain. L'elaboration d'une sociolinguistique du conflit en domaines Catalan et occitan", Lengas 20: 31 —54.

32

John Ν. Green 1990

"Materiaux pour une approche des representations sociologiques. Elements de definition et parcours documentaire en diglossie", Langue franfaise 85: 102-124. Boyer, Henri — Jean Peytard (eds.) 1990 Les representations de la langue: approches sociolinguistiques. (= Langue franςaise 85.) Paris: Larousse. Bright, William (ed.) 1992 Oxford international encyclopedia of linguistics. 4 vols. New York: Oxford University Press. Calvet, Louis-Jean 1987 La guerre des langues et les politiques linguistiques. Paris: Payot. Campbell, George L. 1991 Compendium of the world's languages. 2 vols. London: Routledge. Carrington, Lawrence D. 1990 "The instrumentalization of St. Lucian", International journal of the sociology of language 85: 71 - 8 0 . Chomsky, Noam 1980 Rules and representations. Oxford: Blackwell. Christie, Pauline 1990 "Language as expression of identity in Dominica", International journal of the sociology of language 85: 61—69. Comrie, Bernard (ed.) 1987 The world's major languages. London: Croom Helm. Dardel, Robert de 1991 "Types d'isoglosse en roman commun", in: Dieter Kremer (ed.), 478 — 484. Dejean, Yves 1983 "Diglossia revisited: French and creole in Haiti", Word 34: 189-213. Denison, Norman 1988 "Language contact and language norm", Folia linguistica 22: 11—35. Eastman, Carol 1984 "Language, ethnic identity and change", in: John Edwards (ed.), Linguistic minorities, policies and pluralism (London: Academic Press), 259 — 276. Entwistle, William J. 1936 The Spanish language, together with Portuguese, Catalan and Basque. London: Faber and Faber. [Second edition, revised by William D. Elcock, 1962, with subsequent reprints.] Ervin, Susan — Charles E. Osgood 1954 "Second language learning and bilingualism", Journal of abnormal and social psychology 49 (supplement): 139 — 146. Ferguson, Charles A. 1959 "Diglossia", Word 15: 325-340. Fishman, Joshua A. 1965 "Who speaks what language to whom and when?", La linguistique 2: 67 — 88. 1967 "Bilingualism with and without diglossia, diglossia with and without bilingualism", Journal of social issues 23: 29 — 38. 1991 Reversing language shift. Theoretical and empirical foundations of assistance to threatened languages. Clevedon, Avon: Multilingual Matters.

Representations of Romance: contact, bilingualism and diglossia

33

Fleischman, Suzanne 1992 "Romance languages", in: William Bright (ed.), 3: 337-343. Foresti, Fabio — Elena Rizzi — Paola Benedini (eds.) 1989 L'italiano tra le lingue romanze. Atti del XX congresso internazionale di studi. Bologna, 25 — 27 settembre 1986. (Pubblicazioni della Societä di Linguistica Italiana 27.) Roma: Bulzoni. Francescato, Giuseppe 1982 "Rhaeto-Friulian", in: Rebecca Posner - John N. Green (eds.), 131-169. Gardy, Philippe — Robert Lafont 1981 "La diglossie comme conflit: l'exemple occitan", Langages 61: 75 — 91. Gendron, Jean-Denis — Peter H. Neide 1986 Plurilinguisme en Europe et au Canada. Perspectives de recherche. Bonn: Dümmler. Goebl, Hans 1984 Dialektometrische Studien. 3 vols. Tübingen: Niemeyer. 1986 "Problems and perspectives of contact linguistics from a Romance scholar's point of view", in: Peter H. Neide — P. Sture Ureland — Iain Clarkson (eds.), Language contact in Europe (Tübingen: Niemeyer), 125 — 150. 1991 "La dialectometrie — pourquoi faire?", in: Dieter Kremer (ed.), 332 — 341. Green, John N. 1981 "Generative grammar", in: Rebecca Posner — John N. Green (eds.), 369-407. 1988 "Romance Creoles", in: Martin Harris — Nigel Vincent (eds.), 420—473. Grillo, Ralph D. 1989 Dominant languages. Language and hierarchy in Britain and France. Cambridge: Cambridge University Press. Grosjean, Francois 1982 Life with two languages: an introduction to bilingualism. Cambridge, Mass.: Harvard University Press. Grosjean, Frangois — Bernard Py 1991 "La restructuration d'une premiere langue: l'integration de variantes de contact dans la competence de migrants bilingues", La linguistique 27.2: 35 — 60. Guiter, Henri 1985 Review of Hans Goebl 1984, in: Revue de linguistique romane 49: 201 —207. 1991 "Applications d'une methode geolinguistique en galloroman et iberoroman", in: Dieter Kremer (ed.), 352 — 361. Haiman, John 1988 "Rhaeto-Romance", in: Martin Harris - Nigel Vincent (eds.), 351-390. Hall, Robert A. Jr. 1964 Introductory linguistics. New York: Chilton Books. 1974 Comparative Romance grammar. Volume 1: External history of the Romance languages. New York: Elsevier. Hamers, Josiane F. — Michel Η. A. Blanc 1989 Bilinguality and bilingualism. Cambridge: Cambridge University Press. Harris, Martin — Nigel Vincent (eds.) 1988 The Romance languages. London: Croom Helm. Haugen, Einar 1953 The Norwegian language in America. A study in bilingual behavior. 2 vols. Philadelphia: University of Pennsylvania Press.

34

John Ν. Green 1973

"Bilingualism, language contact, and immigrant languages in the United States: a research report 1956 — 1970", in: Thomas A. Sebeok (ed.), Current trends in linguistics. Volume 10: Linguistics in North America (The Hague: Mouton), 5 0 5 - 5 9 1 . Herman, Jozsef 1991 "Spoken and written Latin in the last centuries of the Roman Empire", in: Roger Wright (ed.), 2 9 - 4 3 . Hockett, Charles F. 1958 A course in modern linguistics. New York: Macmillan. Hoffmann, Charlotte 1991 An introduction to bilingualism. London: Longman. Holtus, Günter — Michael Metzeltin — Christian Schmitt (eds.) 1988— Lexikon der romanistischen Linguistik. Tübingen: Niemeyer. Houwer, Annick de 1990 The acquisition of two languages from birth: a case study. Cambridge: Cambridge University Press. Hudson, Alan 1992 "Diglossia: a bibliographic review", Language in society 21: 611 —674. Hyltenstam, Kenneth — Loraine K. Obler (eds.) 1989 Bilingualism across the lifespan. Aspects of acquisition, maturity, and loss. Cambridge: Cambridge University Press. Hymes, Dell 1974 Foundations in sociolinguistics. An ethnographic approach. Philadelphia: University of Pennsylvania Press. Iordan, Iorgu — John Orr — Rebecca Posner 1970 An introduction to Romance linguistics: its schools and scholars. (Revised, with a supplement "Thirty years on", by Rebecca Posner.) Oxford: Blackwell. Janson, Tore 1991 "Language change and metalinguistic change: Latin to Romance and other cases", in: Roger Wright (ed.), 1 9 - 2 8 . Jardel, Jean-Pierre 1979 "De quelques usages des concepts de bilinguisme et de diglossie", in: Gabriel Manessy — Paul Wald (eds.), Plurilinguisme: normes, situations, strategies (Paris: L'Harmattan), 2 5 - 3 8 . Joseph, John E. 1987 Eloquence and power: the rise of language standards and standard languages. London: Pinter. 1991 Review of Ralph D. Grillo 1989, in: Language 67: 1 5 9 - 1 6 3 . Kahane, Henry — Renee Kahane 1979 "Decline and survival of western prestige languages", Language 55: 183 —198. Kloss, Heinz 1976 "Über Diglossie", Deutsche Sprache 4: 3 1 3 - 3 2 3 . 1987 "Abstandsprache und Ausbausprache", in: Ulrich Ammon et al. (eds.), 302-308. Kontzi, Reinhold (ed.) 1982 Substrate und Superstrate in den romanischen Sprachen. Darmstadt: Wissenschaftliche Buchgesellschaft.

Representations of Romance: contact, bilingualism and diglossia

35

Kremer, Dieter (ed.) 1991 Actes du XVIIF congres international de linguistique et de philologie romanes, Universite de Treves (Trier) 1986. III. Grammaire diachronique et histoire de la langue — Diabetologie et geographie linguistique — Textes non-litteraires. Tübingen: Niemeyer. Kremnitz, Georg 1981 "Du «bilinguisme» au «conflit linguistique». Cheminement de termes et de concepts", Langages 61: 63 — 74. 1983 Frangais et Creole: ce qu'en pensent les enseignants. Le conflit linguistique ä la Martinique. Hamburg: Buske. 1987 "Diglossie, possibilites et limites d'un terme", Lengas 22: 199 — 213. 1991 "Y a-t-il des «diglossies neutres»?", Lengas 30: 2 9 - 3 6 . Labov, William 1963 "The social motivation of a sound change", Word 19: 273 — 309. 1972a "The study of language in its social context", in: William Labov 1972b: 183-259. 1972 b Sociolinguistic patterns. Philadelphia: University of Pennsylvania Press. 1992 Principles of linguistic change. Volume 1: Internal factors. Oxford: Black well. Lafont, Robert 1982 "La privation d'avenir ou le crime contre les cultures", in: Robert Lafont et al. (eds.), Langue dominante, langues dominees (Paris: Edilig), 15 — 36. 1984 "Pour retrousser la diglossie", Lengas 15: 5 — 36. Laforge, Lome — Louise Peronnet (eds.) 1989 Bilinguisme et diglossie. (= Revue quebecoise de linguistique theorique et appliquee 8.2). Trois Rivieres, Quebec: Presses Universitaires. Lamuela, Xavier 1987 Catalä, occitä,friülä: llengües subordinades i planificaciö lingüistica. Barcelona: Edicions dels Quaderns Crema. Lavandera, Beatriz R. 1981 "Sociolinguistics", in: Rebecca Posner - John N. Green (eds.), 1 2 9 - 2 2 8 . 1988 "The study of language in its socio-cultural context", in: Frederick J. Newmeyer (ed.), 1 — 13. Lazard, Sylviane 1985 "Les frontieres du dialecte frioulan: etude dialectometrique ä la lumiere de la methode globale d'Henri Guiter", Revue de linguistique romane 49: 27 — 70. Le Page, Robert B. 1973 "The concept of competence in a creole-contact situation", York papers in linguistics 3: 31 —50. Reprinted in: David Feldman (ed.), Studio gratularia in honor of Robert A. Hall Jr (Madrid: Playor), 1 7 3 - 1 8 9 . Le Page, Robert B. — Andree Tabouret-Keller 1985 Acts of identity. Creole-based approaches to language and ethnicity. Cambridge: Cambridge University Press. Lloyd, Paul M. 1991 "On the names of languages (and other things)", in: Roger Wright (ed.), 9-18. Lüdi, Georges (ed.) 1987 Devenir bilingue — parier bilingue. Actes du Τ colloque sur le bilinguisme, JJniversite de Neuchätel, 20 — 22septembre 1984. Tübingen: Niemeyer.

36

John Ν. Green

Lüdi, Georges — Bernard Py 1986 Eire bilingue. Berne: Peter Lang. Mackey, William F. 1956 "Towards a redefinition of bilingualism", Journal of the Canadian linguistic association 2: 4 — 11. 1962 "The description of bilingualism", Canadian journal of linguistics 7: 51 —85. 1989 "La genese d'une typologie de la diglossie", Revue quebecoise de linguistique theorique et appliquee 8.2: 11—28. Malkiel, Yakov 1977 "Factors in the unity of Romania", Romance notes 18: 263 — 271. 1978 "The classification of Romance languages", Romance philology 31: 467 — 500. Marcellesi, Jean-Baptiste 1979 "Quelques problemes de l'hegemonie culturelle en France: langage national et langues regionales", International journal of the sociology of language 21: 63-80. 1981a "Bilinguisme, diglossie, hegemonie: problemes et täches", Langages 61: 5 — 11.

1984

"La definition des langues en domaine roman: les enseignements a tirer de la situation corse", in: Jean-Claude Bouvier (ed.), Actes du XVIF congres international de linguistique et philologie romanes. V: Sociolinguistique des langues romanes (Aix-en-Provence: Presses Universitaires), 309 — 314. 1989 "Les deux discours glottopolitiques issus de la Revolution Franijaise: la periode 1790-1794 et le Rapport Giordan (1982)", Zeitschrift für Phonetik, Sprachwissenschaft und Kommunikationsforschung 42: 564—571. Marcellesi, Jean-Baptiste (ed.) 1981 b Bilinguisme et diglossie. (= Langages 61.) Paris: Larousse. Marconot, Jean-Marie (ed.) 1990 La parole et le pouvoir. Diglossies dans l'histoire. (= Lengas 27.) Montpellier: Universite Paul Valery. Martinet, Andre 1982 "Bilinguisme et diglossie. Appel ä une vision dynamique des faits", La linguistique 18.1: 5-16. McConnell, Grant D. 1989 "Les concepts de bilinguisme et de diglossie historique, developpement et application", Revue quebecoise de linguistique theorique et appliquee 8.2: 43-55. McKitterick, Rosamond 1989 The Carolingians and the written word. Cambridge: Cambridge University Press. Melis, Ludo — Serge Verlinde — Patricia Derynck 1984 "La notion de dialecte supralocal: essai de definition dialectometrique", Orbis 33 [1989]: 7 0 - 1 3 2 . Milroy, James — Lesley Milroy 1991 Authority in language: investigating language prescription and standardisation. (2nd edition.) London: Routledge. Mougeon, Raymond — Edouard Beniak 1991 Linguistic consequences of language contact and restriction. The case of French in Ontario, Canada. Oxford: Clarendon Press.

Representations of Romance: contact, bilingualism and diglossia

37

Muljacic, Zarko 1980 "Per una classificazione 'standardologica' delle lingue romanze", in: Herbert J. Izzo (ed.), Italic and Romance. Linguistic studies in honor of Ernst Pulgram (Amsterdam: Benjamins), 85 — 96. 1989 a "Über den Begriff Dachsprache", in: Ulrich Ammon (ed.), Status and function of languages and language varieties (Berlin: Mouton de Gruyter), 256 — 277. 1989 b " H a n n o i singolari diasistemi romanzi «emanato» le «loro» lingue standard (come di solito si legge) ο hanno invece le lingue standard romanze determinate in larga misura a posteriori i «loro» dialetti?", in: Fabio Foresti et al. (eds.), 9 - 2 5 . 1989 c "The emergence of the Florentine > Italian language", in: Thomas J. Walsh (ed.), Synchronic and diachronic approaches to linguistic variation and change (Washington DC: Georgetown University Press), 221 —226. 1990 a "Sul dalmatico meridionale (o labeatico)", Abruzzo, rivista dell'Istituto di Studi Abruzzesi 2 3 - 2 8 (1985-1990): 3 6 9 - 3 8 6 . 1990 b "Ausbau-Universalien und quasi-universalien", Zeitschrift fur Dialektologie und Linguistik 57: 167 — 173. Newmeyer, Frederick J. (ed.) 1988 Linguistics: the Cambridge survey. Volume IV: the socio-cultural context. Cambridge: Cambridge University Press. Paltridge, John — Howard Giles 1984 "Attitudes towards speakers of regional accents of French", Linguistische Berichte 90: 7 1 - 8 5 . Pellegrini, Giovan Battista 1980 "Substrata", in: Rebecca Posner - John N. Green (eds.), 4 3 - 7 3 . 1991 La genesi del reto-romanzo (o ladino). (Beiheft zur Zeitschrift für romanische Philologie 238.) Tübingen: Niemeyer. Peytard, Jean 1990 "Evaluation sociale dans les theses de Mikhail Bakhtine et representations de la langue", Langue franqaise 85: 6 — 21. Pfister, Max 1987 "Dialettologia italiana e dialettologia romanza (1976 — 1986)", Rivista italiana di dialettologia 11 [1989]: 3 9 - 9 3 . Pohl, Jacques 1988 "Qui est latin?", La linguistique 24.2: 3 9 - 7 3 . Posner, Rebecca 1980 On the Romance languages. Oxford: Clarendon Press. [Inaugural lecture.] 1992 "Language", in: Richard Jenkyns (ed.), The legacy of Rome. A new appraisal (Oxford: Oxford University Press), 3 6 7 - 3 9 8 . in press "Latin to Romance (again!): change or genesis?", in: Jaap van Marie (ed.), Papers from the tenth international conference on historical linguistics (Amsterdam: Benjamins). Posner, Rebecca — John N. Green (eds.) 1980 Trends in Romance linguistics and philology. 1: Romance comparative and historical linguistics. (Trends in linguistics, studies and monographs 12.) The Hague: Mouton Publishers. 1981 Trends in Romance linguistics and philology. 2: Synchronic Romance linguistics. (Trends in linguistics, studies and monographs 13.) The Hague: Mouton Publishers.

38

John Ν. Green 1982

Trends in Romance linguistics and philology. 3: Language and philology in Romance. (Trends in linguistics, studies and monographs 14.) The Hague: Mouton Publishers. Preston, Dennis R. 1989 a Perceptual dialectology. Non-linguists' views of areal linguistics. Dordrecht: Foris. 1989 b Sociolinguistics and second language acquisition. Oxford: Blackwell. Prudent, Lambert-Felix 1981 "Diglossie et interlecte", Langages 61: 1 3 - 3 8 . Raith, Joachim — Rainer Schulze — Karl-Heinz Wandt 1986 Grundlagen der Mehrsprachigkeitsforschung. Forschungsrahmen, Konzepte, Beschreibungsprobleme, Fallstudien. (Zeitschrift fur Dialektologie und Linguistik Beiheft 52.) Stuttgart: Franz Steiner. Renfrew, Colin 1989 "Models of change in language and archaeology", Transactions of the philological society 87: 103-155. Rogers, Kenneth H. 1981 "Studies on linguistic nationalism in the Romance languages", in: Rebecca Posner — John N. Green (eds.), 229 — 256. Romaine, Suzanne 1982 "What is a speech community?", in: Suzanne Romaine (ed.), Sociolinguistic variation in speech communities (London: Arnold), 13 — 24. 1989 Bilingualism. Oxford: Blackwell. Ross, John A. 1979 "Language and the mobilization of ethnic identity", in: Howard Giles — Bernard Saint-Jacques (eds.), Language and ethnic relations (Oxford: Pergamon), 15 — 26. Sala, Marius 1985 "Langues en contact: evaluation des resultats", in: Ursula Pieper — Gerhard Stickel (eds.), Studio linguistica diachronica et synchronica Werner Winter sexagenario MCMLXXXIII gratis annis ab eius collegis amicis discipulisque oblata (Berlin: Mouton de Gruyter), 751—756. 1988 El problema de las lenguas en contacto. Mexico: Universidad Nacional Autonoma de Mexico. Sanga, Glauco 1981 "Les dynamiques linguistiques de la societe italienne (1861 — 1980): de la naissance de l'italien populaire a la diffusion des ethnicismes linguistiques", Langages 61: 93 — 115. Saussure, Ferdinand de 1965 Cours de linguistique generale, publie par Charles Bally et Albert Sechehaye avec la collaboration d'Albert Riedlinger. (3rd edition.) Paris: Payot. Schlemmer, Gerd 1983 Die Rolle des germanischen Superstrats in der Geschichte der romanischen Sprachwissenschaft. Hamburg: Buske. Sharwood Smith, Michael 1991 "Language modules and bilingual processing", in: Ellen Bialystock (ed.), 10-24.

Representations of Romance: contact, bilingualism and diglossia

39

Spolsky, Bernard 1988 "Bilingualism", in: Frederick J. Newmeyer (ed.), 100 - 1 1 8 . Stehl, T h o m a s 1989 "Typologie des contacts linguistiques: langues romanes, Creoles frangais et dialectes Italiens", in: Fabio Foresti et al. (eds.), 115 — 124. Tabouret-Keller, Andree 1982 "Entre bilinguisme et diglossie: du malaise des cloisonnements universitaires au malaise social", La linguistique 18.1: 17 — 43. 1987 "Parier Creole, devenir creole: le cas complexe du district de Cayo, ä Belize", in: Georges Lüdi (ed.), 227 — 241. Troll, Denise A. 1990 "The illiterate m o n k and written communication. The work of the medieval scribe", in: Richard L. Enos (ed.), Oral and written communication. Historical approaches (Newbury Park: Sage), 4: 96 — 125. Trudgill, Peter 1986 Dialects in contact. Oxford: Blackwell. Ureland, Per Sture (ed.) 1980 Sprachvariation und Sprachwandel. Probleme der Inter- und Intralinguistik. Akten des 3. Symposions über Sprachkontakt in Europa. Tübingen: Niemeyer. 1981 Kulturelle und sprachliche Minderheiten in Europa. Aspekte der europäischen Ethnolinguistik und Ethnopolitik. Akten des 4. Symposions über Sprachkontakt in Europa, Mannheim 1980. Tübingen: Niemeyer. 1982 Die Leistung der Strataforschung und der Kreolistik: typologische Aspekte der Sprachkontakte: Akten des 5. Symposions über Sprachkontakt in Europa, Mannheim 1982. Tübingen: Niemeyer. 1985

Glotto- und ethnogenetische Aspekte europäischer Sprachen: Akten des 6. Symposions über Sprachkontakt in Europa, Mannheim 1984. Tübingen: Niemeyer. Uytfanghe, Marc van 1991 "The consciousness of a linguistic dichotomy (Latin-Romance) in Carolingian Gaul: the contradictions of the sources and of their interpretation", in: Roger Wright (ed.), 1 1 4 - 1 2 9 . Valdman, Albert 1988 "Diglossia and language conflict in Haiti", International journal of the sociology of language 71: 67 — 80. 1991 "Decreolization or dialect contact in Haiti?", in: Francis Byrne — T h o m Huebner (eds.), Development and structures of creole languages (Amsterdam: Benjamins), 75 — 88. Varvaro, Alberto 1991 "Latin and Romance: fragmentation or restructuring?", in: Roger Wright (ed.), 4 4 - 5 1 . Verlinde, Serge — Patricia Derynck 1991 " L a dialectometrie de la delimitation des zones dialectales", in: Dieter Kremer (ed.), 3 4 2 - 3 5 1 . Walter, Henriette 1990 " L a differentiation geographique en frangais et en italien", La linguistique 26.2: 3 5 - 4 5 .

40

John Ν. Green

Wandruszka, Mario 1985

"Plurilinguisme au sein de la Romania", in: Jean-Claude Bouvier (ed.), Actes du XVIFcongres de linguistique et philologie romanes. 2: Linguistique comparee et typologie des langues romanes (Aix-en-Provence: Presses Universitäres), 2: 65-95. Weinreich, Uriel 1953 Languages in contact. Findings and problems. (Publications of the Linguistic Circle of New York 1.) New York: Linguistics Circle. [Sixth printing, The Hague: Mouton, 1968.] Winford, Don 1985 "The concept of 'diglossia' in Caribbean Creole situations", Language in society 14: 345 — 356. Wright, Roger 1982 Late Latin and Early Romance in Spain and Carolingian France. Liverpool: Francis Cairns. 1991 a "The conceptual distinction between Latin and Romance: invention or evolution?", in: Roger Wright (ed.), 103-113. Wright, Roger (ed.) 1991 b Latin and the Romance languages in the Early Middle Ages. London: Routledge. Zuanelli Sonino, Elisabetta (ed.) 1989 Literacy in school and society. New York: Plenum.

REBECCA POSNER

Language conflict in Romance: decline, death and survival

1. Introduction My aim in this chapter (which takes as read Green's preceding discussion of general problems of contact, bilingualism and diglossia) is to survey the ways in which linguistic conflict has arisen between Romance languages and to ask whether the conflict can be resolved or regulated. I have tried to cover the whole Romance area, while concentrating on problems that are not fully treated elsewhere in the volume. Contact between different Romance varieties, and between them and Latin, has been intense throughout our era and may be held responsible for some of the changes the Romance languages have undergone in the course of their history. Distanciation and competition between the languages, consequent on moves towards standardization, is, I maintain, a post-medieval phenomenon, but conflict arose only in comparatively modern times, as an aspect of assertion of ethnic identity. 2. Language conflict Conflict is a term associated with a Marxist view of social change, engendered by the competition between classes to gain power (or, more exactly, authority) in a Herrschaftsverband — Max Weber's term translated by Dahrendorf (1959) as an "imperatively coordinated association". A class for Marx was identical with a social stratum; Dahrendorf modifies this to indicate a group of people with the common political aim of achieving authority, in competition, or conflict, with another group. In this view, social conflict can be associated with language in much the same way as it might be linked with religion: in simple terms a subjected class may have in common a language different from that of the dominant class. The conflict between the classes, if regulated, could lead to a resolution of the language problem, but this would be based on a shift in the personnel in positions of domination within the society.

42

Rebecca Posner

Language conflict can refer, however, to two different, though sometimes overlapping, types of situation. In the first, bilinguals may experience conflict of allegiance between two languages, or more exactly between the two communities to which, for them, the languages belong. They may then desert one language or, unable to resolve the conflict, lapse into a state of anomia, psychologically alienated from their surroundings. The other 'language conflict' situation is not fundamentally different from that of social conflict, involving competition or hostility between communities each using a different language. This sort of conflict, which may even erupt into violence, is often associated with feelings of resentment and persecution by one of the linguistic communities and by the exercise of dominance and intolerance by the other. In neither case can it properly be said that language as such is at the root of the conflict, rather than other socio-psychological, socio-economic or political factors. However, since the late 18th century, language has been regarded as the badge of ethnic or national identity, and even may today, with increasing uniformity of life-styles, be the only differentiating feature of many communities. Conflict of loyalties for an individual bilingual speaker can be painlessly resolved if the environment appropriate to each language is clearly delimited, and if each language has similar prestige value. The conflict can be acute in circumstances in which loyalty to family and friends clashes with the desire for social integration and advancement in a wider society. A bilingual community where one language is treated as inferior to the other can be split asunder by such tensions. Socially mobile members of a monolingual dominated community will wish to acquire, or even shift to, the dominant language, arousing resentment among more loyal and less ambitious fellows, while possibly incurring scorn from the 'superiors' they ape. Where the dominant linguistic community uses political repression and social disapprobation to impose its language on the subservient community, dire consequences can ensue. 2.1 Language and nationhood Although it is probably true that bi- or even multi-lingualism is more widespread in the world than monolingualism, modern Western thought has taken for granted the idea of the monolingual nation state (Laponce 1984). Until recently bilingualism, like racial admixture, was seen as undesirable, or even dangerous. Nazi German theorists, in particular, claimed that bilinguals were disloyal, unstable and intellectually backward (Weinreich 1953: 117 — 120). The Fascist regimes of both Italy and Spain

Language conflict in Romance: decline, death and survival

43

espoused the ideology of the 'national language', advanced by the French Revolution (Emsley 1988), and took vigorous steps to suppress minority languages and dialects. The notion of 'language conflict' within Romance developed particularly in relation to the conflict between Catalan (or more precisely Valencian) and Castilian (el habla cristiana) during the Franco era (Aracil 1965; Ninyoles 1969), though the term had been used earlier by Terracini (1957) in discussion of language 'death', of translation, and of the relation of language to culture, and probably even earlier by social historians, especially about Austro-Hungarian tensions (Inglehart — Woodward 1972). The term was taken up by Haugen (1966) in discussion of Norwegian. It is during the last fifteen years, however, that it has become a major theme of language contact studies (Aracil 1982; Calvet 1987; G a r d y - L a f o n t 1981; Kremnitz 1979, 1981; McRae 1983; Neide 1980, 1986a, 1986b, 1987, 1988; Ninyoles 1977; Oksaar 1984; Trudgill 1986; and Wardhaugh 1987). We may well ask why this theme should now have come to the fore, though it was hardly mentioned by the founder of contact linguistics, Uriel Weinreich, except obliquely in discussing 'language loyalty' (1953: 100—101) and more overtly in discussing psychological problems of bilingualism (1953: 86, 120). It may be that the attrition of cultural differences between communities in the modern world has provoked a backlash, an 'ethnic revival' (Gellner 1983; Smith 1971, 1979, 1981), and that, in the European context at least, language is now the peg on which militant political irredentist action hangs (Williams 1984). Modern life, moreover, requires more developed linguistic skills — a high standard of literacy, ability to present and assess arguments, and so on — among a wider range of the population than hitherto, as well as presenting the possibility, or even necessity, of greater geographical, professional and social mobility. Language has therefore become more important to more people. Whereas in earlier eras a relatively small elite had access to education, much of which was (Classical) language-based, now the majority must acquire, through the vehicle of language, the skills necessary for advancement or even survival. When a speaker's mother-tongue is not a vehicle for education, precious schooling time has to be devoted to acquiring a school language, rather than other skills. In the ancient world, evidence about which language was used in any interchange is hard to find; indeed the Greeks and Romans showed little curiosity about foreign languages. One assumes that important people were well served by interpreters. Within medieval Europe, communication

44

Rebecca Posner

between the 'literate' (see Troll 1990) would be effected in Latin. However, knowledge of foreign vernaculars seems to have been fairly widespread (Bischoff 1961), and comments on variety in language are frequent (Lusignan 1987). In Spain intercommunication between Christians and Moors often had to be conducted through bilingual Jews. Problems arose, however, when vernaculars took over the functions of Latin and the masses sought to play a greater part in public life. It has been suggested (Gellner 1983: 78) that if Western Europe had moved towards industrialization in the High Middle Ages, there might have been a Latin or Romance nationalism; this might mean that competition between vernaculars, and the communities that have adopted them as standard languages, would not have developed as it did. It is probably not an accident, however, that technological advances — like printing and the production of gunpowder — were accompanied by a growing desire of the submerged majority for self-esteem and self-expression, manifested by the questioning of authority, in religion and in politics, and the promotion of vernacular languages as an appropriate vehicle for serious debate. Was it accidental that simultaneously the idea of the nation-state was beginning to take seed? (see Fishman 1972:15). We recall that Voltaire believed that the European identity was stronger than national differences (Godechot 1988). It was however Rousseau (following Herder) who probably got it right, with his insistence on national identity symbolized by language (Siccardo 1984). In the event, the Jacobin doctrine, that all citizens should learn and use the 'national language', to the exclusion of their regional varieties, in order to play a full part in the political life of a free nation, seemed the simplest and most efficient solution to linguistic problems (Balibar — Laporte 1974; Gordon 1978). Even in France, however, it was a long haul and eventually provoked reaction from regionalist groups (Grillo 1989). The ideology adopted by most nation-states in Europe was that a common language promotes cohesion and facilitates communication between its citizens. In this way, too, each state would be maximally differentiated from its neighbour, potentially an enemy, and communication within the frontiers would be enhanced, while intercourse across frontiers would be discouraged or even forbidden. The education system, a state religion and universal military service are important means of cementing the nation as a linguistic unit. A linguistic, religious, or cultural minority would be a potential threat, a 'fifth column': foot-loose, lawless, and preferentially bilingual, gipsy communities are the prototype of such minorities, often the target of irrational fear and hatred. Minority lan-

Language conflict in Romance: decline, death and survival

45

guages associated with a finite territory may be tolerated, but any spread from the allotted ghetto arouses apprehension. The Swiss solution to the potential problems of multilingualism, we recall, is based on the 'territorial principle' (McRae 1983) — the language(s) traditionally used in the territory alone receive official sanction (though, in true Swiss fashion, it is usually possible to buy, by private finance, education or translation services in another Swiss language). As fear of small wars between neighbouring states has receded, and as collaboration between states has increased, and with, also, a continual rise in population, the arguments for imposition of national languages become weaker. Where minority language speakers are reasonably numerous, it is feasible nowadays to provide school and university education for them. Even the problems of book provision are lessened by technological developments in printing techniques. Within the European Community, it is just as likely that Galicians or Sardinians will seek work in Germany, as outside their own region of Spain or Italy: there is no compelling reason that they should be more proficient in Spanish or Italian than in German, as their second language. If eventually they settle permanently in Germany, they may willingly shift their language loyalties to German, or, at any rate, accept that use of the mother-tongue is necessarily limited to family-circle interchange. At home, however, they may insist on the right to use the mother-tongue for all normal language functions. It is in these circumstances that minority language speakers may resent the involuntary bilingualism imposed on them by resolutely monolingual national language speakers (Aracil 1982). Significantly enough, in Switzerland, where language conflict is muted, the majority German speakers are, or at least used to be, more ready to become bilingual than French speakers. Symmetrical bilingualism has been one of the strategies at the basis of Canadian attempts to resolve English-French language conflicts. 2.2 Language shift and language death The overwhelming fear of minority language speakers is that bilingualism and consequently diglossia are merely steps towards language shift, attrition and even 'death'. Psychologists seek to link these phenomena to language 'loss' in individuals, as a result of brain damage or ageing. The processes of loss in bilingual individuals may indeed be akin to attrition and shift in a bilingual community (Sharwood-Smith 1989). However, it can be claimed (Woolard 1989 a) that 'shift' is more often associated with failure to acquire a language by a community's children than with 'loss'

46

Rebecca Posner

by older speakers. Yet there is evidence that in some aphasic conditions, the language most easily lost is that acquired latest, and that there is recall of the earliest mother-tongue, even if it has not recently been used. If this is so, it would help to explain a puzzling finding of dialectologists in some Romance areas, like Northern France, where dialects were, eighty years ago, said to be known only by the very old, yet where it is now found that present-day old people still show knowledge of the dialects. However, it is to be noted that in minority language communities it is often claimed that children no longer have knowledge of the language, but nevertheless, unknown to their elders, the children may covertly use the 'underground' language among themselves. Perhaps the very old, no longer concerned with social prestige, are more ready to bring their knowledge of the language into the open. Language 'death' has attracted much attention in recent times (see especially Dorian 1989 and Dressier 1988), in tune with current interest in 'ecological' matters (Haugen 1972). Threatened languages arouse the same concern in caring circles as disappearing animal species: the richness of our environment is seen as diminished by the loss, usually seen as consequential on the socio-economic pressures of modern life. Language shift on the part of an individual who chooses a new life in a new land may be the source of little anguish, though it can have the dire result of insecure command of any language by the individual ('semilingualism'). In an immigrant minority language community, new modes of discourse will probably develop, involving 'language-mixing' or 'code-switching' or 'long-term accommodation' and the creation of 'interlects' (see section 3.4 below). Where, however, a linguistic community, in its own patrimonial territory, is threatened, by political pressure or by demographic decline, with extinction as a discrete entity, it may seek to defend itself. Influx of immigrants is often regarded as a particular threat — and indeed in the past resettlement has been used as a means of consolidating the political grip of a dominant group on a recalcitrant territory: examples within the Romance area are the repopulation strategies of Reconquista Spain or Mussolini's encouragement of immigration to the South Tyrol. The metaphorical 'death' of a language can occur by the extinction of all its speakers (possible if the community is small) or by their shift to another language (usually over the space of two or three generations). During the Roman Imperial period, languages were undoubtedly lost by the latter process and possibly by the former. To call Latin a 'dead' language is to invoke another process, for, in the relevant sense, the Romance languages

Language conflict in Romance: decline, death and survival

47

today are living Latin varieties. Their 'dead' ancestor consists of a finite corpus of texts and linguistic traditions. Change brought about by isolation, by 'accommodation' in contact situations, and by deliberate distanciation of one variety from another, left the traditional inert variety with fewer and fewer functions, culminating in its virtual abandonment in the Roman liturgy in 1965. Today it is a foreign language for everybody, and the second language of only a few. For an account of the 'death' of medieval Latin, see Kahane —Kahane 1979. Romance languages that have 'died' in the last two millennia have probably all succumbed by way of the second process. Dalmatian, although surviving precariously till the 16th century in Ragusa (Dubrovnik), where its speakers are said to have lived in symbiotic relationship alongside Slavic neighbours, disappeared, possibly after converging with colonial Venetian, finally in the 19th century. German is known to have rolled back the Northern frontiers of Romance, most recently in Switzerland and Austria: so-called Rhaeto-Romance varieties survive in mountain fastnesses at the frontier. In Britain, too, Romance (or 'Vulgar Latin') varieties must have given way to Germanic, probably as a result of language shift rather than of extinction of the romanized population. In North Africa, the spread of Islam must have displaced Romance varieties, of which we have no direct evidence, as it might have eventually done in the Iberian peninsula, were it not for the Reconquista (Burns 1984). That some isolated varieties have survived and even flourished — Rumanian comes to mind — is a matter of surprise. Much seems to depend on the timing of revival: Rumanian profited from the popular tide of 19th century nationalism. In the second half of the 20th century pressures towards uniformity, better communications and educational requirements have threatened the varieties of formerly isolated and backward communities, so that many of them seem condemned to extinction. The only Romance language that was twice nearly, though not quite, lost by the first process — extinction of its speakers — is Judeo-Spanish, sefardi. Suppressed in Spain after 1492, by the persecution and expulsion of the Jews, but nevertheless flourishing under the Ottoman Empire, it barely survived Nazi attempts at systematic genocide. In Israel, the national language ideology has encouraged a shift to Hebrew, but there remain pockets of speakers in virtually every continent (see Sala, this volume).

48

Rebecca Posner

2.3 Linguistic purism Within Romance, fear of language death focuses mainly on language shift, as the real threat. However, transformation of a formerly glorious language into a new variety, by the 'accommodation' process of change is sometime lamented as ignominious, in emotive and value-loaded terms like 'pure' and 'unadulterated' versus 'bastard' and 'degenerate'. Revival of a past stage of the language, accompanied by tastefully cosmetic modernization of the lexicon, after the model of Modern Hebrew, is an ideal sought, particularly today by minority language activists in Spain, and earlier in France and Italy. Where one Romance variety is 'threatened' by another, linguistic distance being small, the fear of 'contamination', with creation of interlects, is heightened by purist attitudes. Any interlect is condemned to a subordinate, and despised, status; the submerged community continues to suffer humiliation, without even the dignity of its own distinctive language system. All the Romance national languages were inspired originally by the purist attitudes inherited from the Latin tradition (see Jernudd — Shapiro 1989). A strictly regulated written code is regarded as necessary for cultural expression, envisaged as an elite domain. Spoken vernacular was allowed more latitude, as long as it served only the functions of interpersonal oral communication and entertainment. But when the vernacular was elevated to higher roles — first for administration, and, especially from the 16th century on, replacing Latin as the vehicle of serious discourse — a more 'illustrious' version, modelled on Latin, was fashioned, and eventually imposed, through the education system, on the whole political unit. Thus the ideology of the monolingual nation-state has been closely associated in Romance speaking countries with linguistic purism (as, indeed, nationalist ideology usually stresses ethnocultural 'authenticity' and 'purity': Fishman 1972: 8). Attitudes to new varieties, interlects or Creoles, have been contemptuous, and sometimes linked, covertly or overtly, to racist prejudices. For subjugated linguistic communities, the choice has been between acceptance of a subservient role (and consequent withdrawal from many aspects of modern society), assimilation, or revolt. Self-confident, prosperous communities, with militant intellectual leadership, a historical tradition and an urbanized population are more likely to assert their independence. The fashioning of a standardized Catalan, or of dialect koines in industrial regions of Northern Italy, are examples: in both cases, unskilled immigrants from less advanced areas are effectively linguistically excluded from the privileged magic circle. Conflict here arises where central authority, as in the Fascist

Language conflict in Romance: decline, death and survival

49

era, attempts to suppress the linguistic aspirations of the privileged groups, and may persist even after the cessation of oppression, as the language issue has become associated with wider political programmes. 2.4 Isolates Where the submerged linguistic community is composed of small rural units, with difficult intercommunication, economic advancement is often sought by way of assimilation or emigration. Emigrant groups, implanted in an alien environment, may resist assimilation more strongly than immobile ones. Thus, to give some Romance examples: support for the establishment of a Galician academy in 1905 came from emigrants to Cuba; Val Verzascans in California have preserved their dialect often with more pride than those who remained in Tessin (Sanga—Tuttle 1989); tiny isolated Daco-Rumanian communities, probably fleeing Ottoman domination, preserved their language in the Istrian peninsula for centuries, although assimilated culturally to their Serbo-Croat environment and dominated successively by Italians, Austrians or Jugoslavs (see Sala, this volume); Catalan-speakers in Alghero in Sardinia, isolated for centuries from contact with other Catalan varieties, and, indeed, unaware until the mid-19th century that their language was not a kind of Spanish, have continued to defend it, in spite of pressures exerted by Standard Italian and the Sassarese variety of Sardinian (Posner 1989); sefardi, we repeat, survived for centuries in Asia Minor, the Balkans and North Africa, after the expulsion of its speakers from Spain. Why low-prestige languages should be maintained in face of all the odds is a much-discussed question: 'ethnic identity' is the buzz-phrase in much of the discussion (Ryan 1979; Woolard 1989 b), but it is still not wholly understood and different conditions hold in each case. Such isolates must, however, be bilingual, and diglossic, communities, in that their vernacular is used only for close personal interaction, in the family, in the fields or fishing-boats and in the bar, or perhaps as an ingroup secret code, while another language is required for communication with the world outside. Commentators predict the attrition and death of the vernacular, as its privileged domains shrink, and as intercourse with the exterior increases. Alternatively, the isolated community, pauperized by economic changes and weakened by persistent endogamy, may so decline that eventually the language dies with its speakers. Rural communities today, however, can breed their own share of able and educated sons and daughters, who reject the polarization between local loyalty and social mobility, and who seek to modernize and promote

50

Rebecca Posner

their own vernacular, often to the derision of those who acknowledge the quaint charm and folkloric authenticity of local varieties, but who refuse them 'language' status. Conflict thus arises between older-style dialectologists and younger more sociolinguistically oriented investigators. 2.5 The case of Spain — past and present At present, conflict is most virulent in Spain, following the relaxation of the oppressive linguistic policies of the Franco period, where the claims of Aragonese and Asturian activists are laughed out of court by some prominent academics (Posner 1992). Whereas Catalan and Galician (see Strubeil i Trueta and Monteagudo — Santamarina, this volume) are recognized, sometimes reluctantly, as having a literary tradition, and with Valencian (Byrne 1989; Casanova 1988; Cremades Marco 1982; Ferrando Frances 1980, 1988), are recognized by Statute as co-official with Castilian, in their respective territories, the vernacular varieties that survive in Asturias and Aragon are treated as 'local modalities' (Alvar 1986 a), and efforts to establish a normalized koine are vigorously opposed in some quarters. In these Northern territories, probably latinized late and then staying free of Moorish domination, the Romance varieties developed independently from Latin (Cano Gonzales 1987). However as the Reconquista advanced they were replaced, for many functions, by other Romance varieties. In the East, Catalan became the official language of the Kingdom of Aragon, which after the 15th century unification with Castile, gave way to Castilian. In the West, Galician expanded into Portugal (see section 3.2 below), and Asturian into the reconquered Leon region: Castilian is often seen as an innovating variety of Leonese (Lloyd 1987: 178 — 180), in its turn a diluted Asturian, which did replace Latin as the official language in the Kingdom of Leon until the 13th century. Soon after, the union of Castile and Leon (1230) led to the replacement of Leonese by Castilian: today it survives only as a number of regional features in Castilian (Llorente Maldonado 1986). Asturian local varieties, known since the 18th century as babies, are however still extant in rural areas (Neira 1982; Penny 1969, 1978), and among the working-class in industrial centres (D'Andres 1987). There is little consciousness of linguistic separateness in the underpopulated, rural area of Leon (Holmquist 1988; Lang 1982), even though in Estremadura the isolated 'Hurdanos', who featured in a 1923 Bunuel film, have until recently been popularly regarded as genetically and linguistically quite un-Spanish (Catani 1983). Nevertheless, in Aragon (Conte et al. 1977; Monge 1989) and, even more so, in Asturias (Muljacic 1988; Tolivar Alas 1989), there is some popular

Language conflict in Romance: decline, death and survival

51

feeling in favour of promotion of the local languages, even though Castilian has made great inroads into their territories, especially since the 16th century. Romance varieties used in Southern Spain, like Murcian and Andalusian, are treated historically as variants of Castilian, following repopulation, after the conquest of the Moors. Seeing it as a regional and popular variety of Castilian, speakers have low consciousness of the separate identity of Murcian (Munoz Garrigos 1986; Perez-Alonso 1979) except, as panocho, in a satirical comic setting. Probably originally the outcome of joint resettlement of Castilian and Catalan speakers, with the admixture of pre-conquest varieties, it has been influenced by Castilian from the 14th century on, adopting Southern features, like the seseo and the loss of plural -5. The resettlement of the extreme South of Spain was not completed until the 16th century, when the codification of standard Castilian was well underway. In Andalucia, the language replaced by Castilian, Old Andalusian (Malkiel 1988), is sometimes called mozarabic, from the name (must a'ribun) given to Christians under Arabic rule, though there may have been very few of these, after about the 12th century. It was also spoken by the Jews and perhaps by the muwallads, or Muslims of Iberian stock (though there is evidence to suggest that these had by now adopted a variety of Arabic). Our knowledge of the 'mozarabic' language is scant and uncertain: it seems to have been a conservative dialect that did not share specifically Castilian developments. The contact of confident, and by now conflictive and intolerant, Castilian conquerors in their finest hour, with what must have been a demoralized and backward local population probably produced a language shift among Romance speakers towards the prestige variety, though forced conversion and repression was clearly unsuccessful, and culminated in the expulsion of the 'converted' moriscos in the early 17th century (see Boase 1990). How far features of the submerged languages were transferred to the adopted Castilian speech is disputed. Mena Calvo (1986) rejects vehemently the idea that Andalusian varieties are "mestizajes impuros de raiz morisca y judia", to which one might add the 'contamination' of gipsy settlement in the South. There is however early evidence for some Andalusian phonological features (Mondejar 1988). We cannot be sure whether Andalusian is historically a dialect of Castilian or a castilianized form of a language that developed separately from Latin (Mondejar 1986). Today many Andalusians will claim that theirs is a more authentic Castilian than that of Madrid (Carbonero 1982): in Seville, regional features are

52

Rebecca Posner

spread through all classes of the population (Lamiquiz — Carbonero 1987) and carry no opprobrium. The relative closeness of Southern Spanish speech to American Spanish no doubt adds to its prestige (Carbonero 1991). Consciousness among speakers of the identity of their language is fairly high: a survey in Seville in 1978 (Ropero Nüfiez 1982) shows that nearly 40% (most of them younger informants) believe they speak 'andaluz', some 22% 'espanol' and 27% 'castellano', mainly among older informants (with 5% opting for 'espanol andaluz' and nearly 4% for 'castellano con acento andaluz'). In the country, more informants named a local variety as their mothertongue, but 20% still opted for 'andaluz'. Partly through the influence of tone-setting individuals, Southern pronunciation has become quite modish, and Andalusian speakers have none of the linguistic insecurity noted for Murcian (Lamiquiz et al. 1982 — 87). On the contrary, their variety is regarded as witty and racy, and is adopted by entertainers in preference to slower, grave and raspy Castilian styles. In discussing the Spanish linguistic situation, we should not omit mention of Aranes, which is the only Occitan variety with official status (see Schlieben-Lange, this volume). In the far Northwestern corner of Catalunya on the Northern slopes of the Pyrenees, the Spanish Val d'Aran juts out into French Bearnese territory. Aranes is a Western Occitan variety, linguistically fairly close to the Aragonese used to the Southwest, but more clearly differentiated from Catalan. Nearly 60% of the permanent population of this tourist area regard themselves as mothertongue Aranes speakers, compared with 30% Castilian — and a mere 9% Catalan-speakers, though 80% of the population is competent in Catalan. About 30% claim to speak French. Nearly all are bilingual, many are trilingual, or even triglossic, but very few are quadrilingual (Climent 1986; Viault 1987). Attitudes towards Aranes are positive among the great majority of the population, especially since the Catalunya Language Statute, which speaks of 'el habla aranesa' rather than 'modalidades', decreed that it should be taught in schools as a written, as well as a spoken language: at present all primary schools should teach a normalized variety for one or two hours a week. The orthographic norm is adapted from the Occitan model. Variation within Aranes has been studied by Winkelmann (1989), with particular reference to phonology and morphology. 70% of the variants appear to be endogenous, with contact variants mainly geographically distributed (for example, the definite article forms etjel). Younger speakers are more influenced by Catalan and Castilian, but also, surprisingly, by Languedoc; older speakers retain

Language conflict in Romance: decline, death and survival

53

more Gascon features. One distinguishing Gascon feature — the ke 'enonciatif which prefixes all assertive main clauses — is rare except in the extreme North, and is used more by older speakers. By all accounts, variation does not hinder the acceptance of the need for a norm, and conflict between the languages spoken in the Val is minimal. 2.6 Other Romance areas In the more tolerant Italian atmosphere, where regional autonomy dates further back, linguistic conflict is less evident than in Spain (see Trumper, Rindler-Schjerve and Posner — Rogers, this volume). The hegemony of standard French, on the other hand, has barely been weakened by mild decentralization policies. Occitan and Corsican language advocates often lament their lot in conflictual terms (see Schlieben-Lange and Thiers, this volume). In Switzerland and in the South Tyrol, Rhaeto-Romance varieties, though paid lip-service in official circles, are under threat, and their ultimate demise is predicted by some (see Posner — Rogers, this volume). It is German, rather than another Romance language, that is encroaching on their territory. In Friuli, in contrast, the Rhaeto-Romance language is in contact with neighbouring Venetan varieties and with standard Italian, but appears to be in no imminent danger (for the Friulian varieties spoken in Rumania see Sala, this volume). In Switzerland, Italian, although an official language, is somewhat under threat, in closer contact with German rather than with the other Swiss Romance languages. Because of a certain distrust of standard Italian, consonant with their loyalty to their local Tessinese dialects, and to the Federation, Swiss Italian speakers feel that they must defend their linguistic heritage, especially in face of the influx of immigrant workers from Italy, and they turn to parochialism and folkloric revival (Bianconi 1980; Lurati 1976), while tending to use German for many language functions. The French-speaking Romands traditionally believe that the Tessinese are their natural allies against the Germanophones, because they are perceived as sharing a Romance heritage, but the Romands rarely learn Italian, expecting, without justification, French to be the common language (Knecht — Rubattel 1985: 152). Italian is in fact hardly used in Switzerland outside the Tessin, except by immigrants. An attempt to preserve it as a federal language gave rise to the so-called 'Modell Schweiz', proposing that passive knowledge of Italian should be promoted in order to encourage polyglot interchange (Watts 1991: 99). Rumanian varieties spoken in Greece, Bulgaria and Albania have little hope of survival in the future; speakers who settle in Rumania tend to

54

Rebecca Posner

assimilate to the Daco-Rumanian standard language, while those who stay in place are often condemned to an archaic and obsolescent life-style (see Sala, this volume). 3. Language survival and revival Frequently elderly loyal speakers in the small communities accept, with regret, that language death will occur, often lamenting, as old people do, the callousness and disrespect of younger generations. More politically conscious speakers may, we repeat, be steeled to active resistance by the example of apparent success in reviving dead or dying languages — as in Israel or Wales. Often they will not only accuse their persecutors — imposers of a dominant language — but also blame their fellows for lack of courage or 'self-hate' (Catalan auto-odi: Ninyoles 1969), for accepting others' low evaluation of their language and for selfishly seeking personal advancement at the cost of disloyalty to their community. As bilinguals, they may take pride in their greater aptitude for language acquisition than their 'slow-witted' monolingual neighbours (Neide 1986 a), but still resent imposed asymmetrical bilingualism. Experience suggests that immortality, or at any rate longevity, for a language depends on its extension in space, and its status, measured by the number and importance of its functions (Mackey 1985). To this must be added a positive will on the part of speakers to preserve their language, and their support for leaders in a political campaign for language retention. Sometimes such leaders are accused of manipulation of the community's imperfectly formulated wishes, and exploitation for ulterior motives of a deeply-felt desire for preservation of linguistic heritage as a sign of community identity. It is true that there have been examples of campaigners who abandoned the linguistic struggle once other more narrowly beneficial aims had been attained (the Lliga Regionalista founded in 1901 in Catalunya is one example: Woolard 1989 b: 25 — 26). Today even moderate advocates of language maintenance are sometimes lumped together, by opponents, with irredentists and militant underground organizations (see Thiers, this volume). Studies of language maintenance and language death point to the conclusion that to secure survival of one's mother tongue concerted and vigorous action is necessary to devise a superordinate koine, which will cover local dialects, in order to extend its spatial validity, and to widen the functional relevance of the koine by rendering it apt for use in a greater range of domains. The former requirement can be met by the recognition of a more or less abstract diasystem — a sort of common

Language conflict in Romance: decline, death and survival

55

denominator of features shared by related varieties, which forms the basis for a shared written code. The latter requires a conscious attempt — status planning or 'normalization' (see Muljacic, this volume) — to extend the lexical coverage of the resultant koine, so that discourse on, say, philosophy or nuclear physics can be as adequately conducted in the language as on, say, harvesting, fishing or the upbringing of children. The consequences of these two processes — koineization and normalization — must mean that the modernized language is not identical with any 'authentic' local usage, the principal function of which is to signal parochial solidarity. Critics may sneer at the artificial character of the modernized language, which lacks the homeliness and variety of village dialects, a criticism most often levelled in Spain (Posner 1992). Another feature regarded by some as essential to ensure the survival of a threatened language is codification (corpus planning or normativization: see Muljacic, this volume). The European belief that mother-tongue speakers need to learn the correct grammatical use of their own language is one that dies hard and is probably based on traditional methods of teaching Latin (Lusignan 1987). True, school text-books are usually associated with simplification of complexities. Where, however, simplification and levelling of variants goes counter to the intuitions of native speakers, it is likely to be counterproductive. Puristic, proscriptive normativisation, whether harking back to obsolete prestigious literary usage, or appealing to some assumed logical base, may defeat its own purpose — to provide mother-tongue speakers with a viable and accessible means of expression for all the domains of modern life. More appropriately, language planners must permit variants that do not seriously hinder comprehension, and language users must be prepared to accept, as part of their own language, regional or social usages that they themselves may habitually shun. 'Polynomic' standardization of this sort (see Thiers, this volume) is in tune with the more open society of the modern world, whereas the purist, elitist standardization of earlier eras was concerned more with exclusion than with inclusion. 3.1 Linguistic distance and accommodation The greater the linguistic distance between languages the less likely is language shift to occur (Mackey 1980). Romance languages are all so close to one another that shifts are easily, and sometimes unconsciously, effected. Variants of one language often coincide with those of another: speakers may be unsure which language they are currently using (see, for

56

Rebecca Posner

instance, Meo Zilio and Sala, this volume). In their anxiety to hinder language shift, speakers will frequently favour variants in their mothertongue that are maximally differentiated from the language by which they feel threatened (examples abound in contributions to this volume: see in particular Monteagudo — Santamarina and Thiers). A 'language' may be identified, in Kloss's terminology (Muljacic, this volume) as Abstand, 'standing off from its neighbours. Here the covert assumption is that 'purity' in language is desirable, and that to be legitimate bilingualism must be 'co-ordinate' — with access to each language effected by different routes, rather than 'compound' — with access to one language via the other (Baetens Beardsmore 1986). Where Romance varieties are concerned, strict compartmentalization is artificial and impractical. Apart from their shared Latin heritage, the languages have throughout their history interacted with each other, borrowing extensively from each other and from Latin. Some of the features that distinguish one national language from another are of fairly recent origin — and not necessarily present in all regional or social varieties. 'Accommodation' between speakers of different varieties, even across administrative boundaries, must be accepted if intercommunication is to be maintained. 'Accommodation' (Giles 1984; Giles —Smith 1979) is the generally accepted term to designate the way speakers mutually adjust their usage in a verbal interaction, seeking to appear co-operative and considerate. Some theorists (Giles et al. 1973) believe that there are universal features of this process; in this perspective, similarity of language and culture would favour accommodation between Romance speakers. However, others insist that accommodation practice varies according to different social habits and traditions (Stieblich 1986). 'Long-term accommodation', or convergence (Orioles 1989), applies to more stable practices shared by members of a community in frequent contact with another, and can lead to the establishment of a more or less stable interlect: Trudgill (1986) sees this process as a major factor in language change. There is a clear relationship between these notions and Weinreich's (1953) 'interference', for which Clyne (1987) substitutes 'transference'. Interdialects reflecting convergence between different Romance varieties are, we repeat, regarded with contempt by supporters of linguistic 'purity' or 'integrity'. Until quite recently 'transitional dialects' formed part of a chain of mutual intelligibility from one end of the Western European Romance area ('Romania continua') to the other. With the polarization, or 'focussing' (Le Page —Tabouret-Keller 1985), entailed by standardization and the 'national language' syndrome, or by koineization

Language conflict in Romance: decline, death and survival

57

of regional dialects, more clear-cut frontiers are now discernible. For Rumania, the 20th-century political boundaries were designed to coincide broadly with linguistic frontiers, though as the events of 1989 and 1990 have made known to newspaper readers, they embraced a sizable Magyarspeaking region and omitted Rumanian-speaking former Soviet Moldavia (Moldova). Within Daco-Rumanian, however, there is remarkably little regional variation — a fact which may be related to the circumstances of the Ottoman administration, or may, some claim, suggest that the language was comparatively recently introduced North of the Danube, by settlement of immigrants from further South. In Western Romania, none of the national boundaries coincide precisely with the extension of one discrete grouping of local Romance varieties (Auer —di Luzio 1988). In France, however, the national language has spread out over all the territory of the state, and even beyond, though not wholly eliminating all other languages (Vermes 1988; Vermes — Boutet 1987), which include the Romance varieties collectively labelled Occitan, the Catalan spoken in Roussillon, the Italianate Ni^ois, Corsican and some remnants of Franco-Provenpal and Northern French local varieties. In Spain a similar process, begun by the 16th century, is not complete, as we have seen. In Italy, unified only since the mid-19th century, diffusion has been even slower, though it has increased apace since the second world war. Belgium, officially bilingual, shelters several Romance language varieties — besides standard French, Walloon, Picard and the Lorraine variety gaumais still have some currency, while quadrilingual Switzerland recognizes three Romance languages — French, Italian and Romansh — and is the home, still, of some local Franco-Provengal and Lombard varieties. 3.2 Dialect continua and interlects: the case of Spanish and Portuguese In Southern Italy, Southern France, Northern Spain, Sardinia and Corsica there still subsists to some degree a dialect continuum. Particularly interesting are the interlects in the West of the Iberian peninsula, especially where the national boundary between Spain and Portugal runs through them. Although Portugal was a separate state from the 12th century, it is probable that the Portuguese language was not perceived as autonomous before the 16th century (Monteagudo 1988). Before then the linguistic position of Gallego-Portuguese in relation to Castilian was not dissimilar to that of Galician today (see Monteagudo —Santamarina, this volume). The Western varieties of the peninsula were differentiated from

58

Rebecca Posner

Castilian mainly by phonological features, perhaps reflecting a greater orientation towards the south, as compared with the northern slant of Castilian (Carvalho 1991); but all Spanish Romance varieties were probably more mutually intelligible than they are today. The literate made use of Galician and Castilian for different literary genres — the lyric being the principal domain of Galician. Until the reign of Alfonso el Sabio, Galician was the language of official documents in the region, and it went on being used, alongside Latin, in notarial documents until the 16th century. Latin was the preferred language of Church documents until the end of the 13th century, and was used alongside Galician till the 15th century. Latin remained in use in the University until early in the 19th century (Filgueira Valverde 1982). In the North, from the 13th century on, Castilian features appeared in Galician writings, but the perception of a gulf between Galician and Portuguese did not emerge before certain radical phonological changes had occurred in Southern Portuguese varieties and in Castilian in the 16th century. Even before the period of Castilian rule in Portugal (1580—1640) literary bilingualism was normal among Portuguese writers. Gil Vicente, for instance, wrote between 1502 and 1536 twelve autos in Castilian (using also the Leonese variety sayagues for rustic characters), fifteen in Portuguese and nineteen in both languages, with characters interacting in a way that implies mutual comprehension (Teyssier 1959). That there was equality of prestige between the two languages is disputed. Vazquez Cuesta (1981) argues that even when it was modish for Castilian courtiers to write lyrics in Galician, the language used was a sort of castrapo — Castilian with stereotyped Lusisms — and that by the 16th century in Portugal there was diglossia, with Castilian treated, alongside Latin, as the High variety, whereas Gallego-Portuguese was homely, archaic and subordinate. The King of Portugal from 1521 to 1557, Dom Joäo III, for instance, appointed Castilians to teach at Coimbra University, and in 1567 the Portuguese mathematician, Pedro Nunes, translated his work into Castilian to make it more accessible to the cultured public. The social devaluation of Portuguese gathered momentum in the 17th century, so much so that the King immediately after the period of Castilian rule, Dom Joäo IV (1640—1656), preferred to write a treatise on music in Castilian, rather than Portuguese. The later 17th century reaction against Castilian domination, both political and cultural, led to a desire to differentiate Portuguese maximally from Castilian, in spite of the salient similarities in their history, culture and language. Thus the 'inflected infinitive' and older features such as the intercalation of object pronouns

Language conflict in Romance: decline, death and survival

59

in the synthetic future, pronoun enclisis or the use of Latin pluperfect reflexes with tense, rather than modal, function, are held to be specifically Portuguese, and are today, especially in Brazil, markers of a high prestige social dialect (Koike 1983). Some such 'Lusisms' are found in other Western Spanish varieties — in Galician, of course, but also in the remnants of Leonese. But it is the phonological features of the Lisbon norm — salient nasalization, weakening and closing of atonic vowels, shushing of implosive /s/ (chiante), velar realization of /rr/ — that differentiate the language more obviously from Castilian, which in its turn has followed certain other innovating phonological paths (for instance, devoicing all sibilants and merging jbj with /v/). Some of the more distinctive phonological features of standard European Portuguese are absent or attenuated in Brazilian (which has nevertheless innovated elsewhere, as in the palatalization of dentals before front vowels, fala aberta). We deduce (with Revah 1958) that at the time of colonization of Brazil, Portuguese had not undergone these changes. Some Brazilian regions — especially Rio Grande do Sul — use Portuguese varieties that are phonologically closer to Spanish, leading to the suggestion that they are strongly influenced by Spanish, via early settlers or modern contact, but it may be that they are merely more conservative. Fuertes Alvarez (1964: 364) claims "el Rio Grande del Sur es donde mejor se habla el portugues" (which is reminiscent of the 'Portuguese is badly pronounced Spanish' joke), whereas other observers talk about a portunol used in Southern Brazil. The hostility between the two national languages of the peninsula is replicated in those South American countries where the two languages are in contact. Camara (1964) speaks of "dois mundos culturais que ... näo se interpenetram". Certainly Portuguese speakers show an unwillingness to spend study time acquiring the neighbouring language, as they rarely have difficulty in understanding, or even speaking Spanish. Thus one year's compulsory study of Spanish was dropped in Brazilian higher secondary schools in 1962, as more students opted for English or French. Spanish mother-tongue speakers have much more problems in acquiring competence in Portuguese (Garcia Morejon 1964) — though whether this is a matter of psychological resistance or genuine linguistic difficulty is hard to tell. Conflict between Portuguese and Castilian (or 'el idioma nacional') is most felt in Uruguay (see Hensey, this volume). On the other Brazilian frontiers less work has been done, though frontier varieties in Colombia are currently being investigated at the Instituto Caro y Cuevo in Bogota. Uruguay, we recall, became independent in 1828, set up with

60

Rebecca Posner

British help, as a buffer state between Argentina and Brazil, but large tracts of the interior were not permanently settled until the late 19th century. That Uruguayans feel that the Portuguese-Spanish interlects extensively used on their Northern frontier, and in pockets elsewhere, are a threat to national unity, can be seen from the report of an Academy Commission (Academia Nacional 1982). The mainly gaucho regions, virtually unpenetrated before the 18th century, because of hostile natives, fell prey to cattle rustlers and military incursions from Brazil well into the 19th century. Attempts to develop Spanish-language schooling in the region were hampered by the predominantly Portuguese-speaking settlers, whose loyalty was to Brazil, sending their children there for education, and even baptism. Today many of the inhabitants are bilingual, though mother-tongue Spanish speakers more readily acquire Portuguese than vice-versa, possibly because the fronterizo variety is more easily accessible to them, or because work opportunities across the border are better. 'Fronterizo' is a term used loosely in Uruguay for any popular variety, but in the North it seems to be a form of Portuguese that has converged with Spanish, or perhaps been protected, by its contact with Spanish, from the forces that have differentiated the two languages in modern times. In either case it suggests that bilingualism has been widespread enough to cause extensive transference and convergence. Even though Rona (1963) suggests that the interlect is comprehensible to neither Spanish nor Portuguese speakers, other observers maintain that local varieties of each language are mutually intelligible, and that therefore there is no community functional reason to acquire both. Contact between Spanish and Portuguese varieties in the Iberian peninsula has a different historical background and carries different sociolinguistic connotations. In the North, there is a continuum of dialect variation from West to East, disrupted in modern times by the incursion of the innovating standard Castilian variety, which has been imposed, most effectively in the Franco era, as the unique language of culture in Spain. Repopulation strategies created more of a mosaic further South and in some parts of the frontier between Spain and Portugal natural features, like the Duero/Douro, or man-made buffer areas lined with frontier fortresses represent barriers between the peoples, that have kept the languages apart. But elsewhere the frontier demarcation was disputed until well into the 19th century and there has been interaction between villagers over the border (often in smuggling activities). The linguistic consequences of these contacts have been studied more from the Portuguese (since Vasconcellos 1886, 1902), than from the Spanish side (but

Language conflict in Romance: decline, death and survival

61

see Krüger 1925, for a description of Sanabrian-Portuguese varieties in Zamora province). Boleo's 1964 [1974] survey gives a bibliography: his students in Coimbra worked in Northern and Central frontier areas, mainly during the 1960s and 1970s. Further work is proceeding, among Lisbon researchers, on the sociolinguistic situation in Barrancos, half a mile from the Andalusian border (Alvar 1986 b: 21—22; Vasconcellos 1955). Although the Coimbra studies so far published (Maia 1977; Matias 1986 and Santos 1967) take up themes initiated by Weinreich (1953) on language contact and interference, they use the more traditional methodology of dialectology, concentrating on the usage of a single, elderly informant for each locality, and relying on questionnaires. Actual bilingual practices are little recorded, for contact across the frontier is usually effected in mutually intelligible local dialect, though sometimes Portuguese citizens do their shopping or consult doctors in neighbouring Spanish towns, necessitating some approximation to standard Castilian. Where Portuguese speaking pockets persist in Spain (as in Olivenza, Campo Major, Almedilla, the Sierra de Jälama villages and Hermisende) Portuguese dialectal usage is overlaid by school-taught Castilian. Within Portugal, besides Barrancos, the only recorded 'Spanish' localities are, it is assumed, former Galician or Leonese settlements (such as some villages in the Vinhais region, Spanish until the late 19th century, and the Riba Coa incorporated into Portugal in 1297) and the contrast between local dialect and standard Portuguese is not much starker than is the case for dialects that are considered 'Portuguese'. Where there is interruption of the chain of mutually comprehensible dialects — as in Zamora province where Leonese is being swamped by Castilian — there seems to be rather less contact at the personal level. Interaction was interrupted by the closing of the frontier at the time of the Spanish Civil War, but today, with both Spain and Portugal members of the European Community, and more widespread schooling, we anticipate changes in the situation. The dialects examined are transitional varieties, rather than contact varieties, not converging from originally distinct languages, but sharing common historical changes, or, more often, retaining common archaic features. The cleavage between the two standard languages is not reflected in these dialects, except insofar as they are now converging towards their respective standards, as transport and educational facilities improve. These factors also encourage geographical mobility — to urban centres within the state, or to industrial regions in other countries. Cross-frontier migration — as distinct from day-to-day exchange — is comparatively rare, though intermarriage does occur and may, linguistically, be worth

62

Rebecca Posner

studying. The people of the frontier region seem not to be in conflict, sharing similar life-styles and often, in the past, cooperating in agricultural or smuggling enterprises. The most notorious cross-frontier locality is the village of Riodonor/Rionor, cut in two by the national boundary, but sharing a single interlect and daily activities. Krüger (1925: 122) speaks of "un verdadero campo de batalla entre dialectos", but his account concentrates on phonological and morphological features and makes little reference to speakers' attitudes. Santos (1967: 49, 132) paints a picture, for 1960, of constant cooperation and interaction, rather than conflict, but comments little on the impact of schooling, which must have increased in the last thirty years. Transitional varieties in Spain — between Galician and Asturian, between Aragonese and Catalan (Alvar 1976), as well as the fronterizo varieties just discussed, or interlects between any variety and standard Castilian — tend to be stigmatized, as even the minority language supporters uphold a purist attitude to language questions. In Italy, once the heartland of linguistic purism (with the foundation of the Accademia della Crusca in 1582) nowadays a more relaxed attitude to variation is taken by the powers that be, so that regional varieties of Italian, representing convergence between dialect and standard are acceptable. By and large, dialect speakers — still the majority of Italians — are reconciled to a diglossic situation, with regional Italian alternating with a dialect koine and/or a local dialect. European interlects, often part of a dialect continuum, have presumably shared over time similar influences and transformations. Those of overseas territories are more likely to have developed from contact between two distinct languages. We have already seen that fronterizo varieties in South America and in Europe, although superficially alike, are fundamentally different. The cocoliche of Buenos Aires (Meo Zilio, this volume) has no direct parallels in modern Europe, though the crusaders' lingua franca with elements from French, Italian and Catalan was probably similar (Kahane — Kahane 1976). 3.3 Creole continua Whether 'decreolized' forms of Creoles result from a similar contact between two distinct varieties, or whether there was always overlap, or a continuum from basilect to acrolect, is a matter for dispute. Romance Creoles in contact with related metropolitan standards — as in the French Departements d'Outre-Mer, Guadeloupe, Martinique and Reunion, or in the former Portuguese Cape Verde islands — will have among their

Language conflict in Romance: decline, death and survival

63

variants 'decreolized' forms that converge towards the standard (see Hazael-Massieux — Robillard and Stolz, this volume). Where, however, there is no such political contact, as in Dominica, there is less likelihood of the formation of an interlect. In Haiti and Curapao, attempts at standardization, with the devising of writing systems that will enable basic education, at least, to be conducted in the Creole, may entail some convergence (or decreolization) towards prestigious international languages, French and Spanish respectively. One apparent example of convergence between related non-standard Romance languages is to be found in Louisiana, where two historically distinct French varieties are recorded, neither of them in regular contact with standard French for some two hundred years. 'Cajun' continues the Canadian variety brought to the bayou country by Acadians, expelled from Canada in the mid-18th century, after wanderings elsewhere. This patois, and the life-style of its speakers, was reported as clearly distinguished from the Creole of ex-slaves, in the late 19th century (Fortier 1884 — 85, 1891). Today the varieties seem to have merged and no longer distinguish black and white speakers. In some localities, usage leans more towards the Creole, in others toward the patois. In any case both varieties may be used in family conversational interchanges without any apparent breakdown of communication (Marshall 1982; Morgan 1959, 1970; Neumann 1985; Tentchoff 1975; and Valdman 1992). 3.4 Language mixing and code switching Relatively stable interlects, like cocoliche or fronterizo, are nowadays infrequent, but short-term accommodation and language-mixing — where speakers change frequently, apparently unpredictably and often unconsciously from one language to another, to form 'instant pidgins' — occur frequently in interactions between speakers of different Romance mothertongues. Language-mixing and code-switching are topics that have aroused much interest recently among psychologists and linguists alike (Auer 1984; Clyne 1987; Di Sciullo et al. 1986; European Science Foundation 1990; Gardner-Chloros 1987,1990; Heller 1988; L i n d h o l m - P a d i l l a 1978; Lipski 1985; Poplack 1980; Poplack - Sankoff 1984, 1988; Poplack - S a n k o f f - M i l l e r 1988; R e d l i n g e r - P a r k 1980; S a n k o f f - P o p l a c k 1981; Singh 1985; Stieblich 1986; Woolford 1983). Terminology is fluctuating: here we use 'code-switching' for circumstances in which bilingual speakers, with clear differentiation between their two languages, make a choice of language for each utterance, usually depending on the identity

64

Rebecca Posner

of the interlocutor or other external circumstances, though also for stylistic effects. Switching may be regarded as a "safety-valve that conserves the integrity" of each language (Woolard 1989 b), and so is compatible with purist linguistic attitudes. 'Language-mixing', on the other hand, refers to a way a speaker changes from one language to another in the course of an interaction, an utterance, or even a sentence, and is much more problematic. How far mixing is different from nonce-borrowing, or from other types of variation, and what syntactic or social constraints operate in mixing, are as yet unresolved questions. The loci within a sentence at which a change of language is made, can give clues to the way bilinguals store their languages and gain access to them. Empirical studies are therefore of importance to determine whether there are any universal characteristics of this process: so far results are inconclusive and sometimes contradictory. In purist circles mixing is stigmatized, as showing an insecure command of the distinction between the two languages. In some social environments, however, mixing can be regarded as a mark of politeness, an extreme form of accommodation, or as a witty and emphatic conversational ploy. Where two Romance varieties are involved it is sometimes impossible to discern whether or where a change has been made. At the extreme, the mixing strategies acquire some sort of stability, and an interlect, like cocoliche, results (Meo Zilio, this volume). 3.5 Bilingual exchange and passive bilingualism The closeness of different Romance varieties favours another interactional strategy — that of bilingual exchange, in which respective speakers use their own mother tongue, with some degree of accommodation. This strategy is sometimes agreed, by implicit or explicit contract between monolingual speakers eager to communicate. I was once the amused, and bemused, witness to a protracted, and not always completely successful, interchange, on a long and boring train journey, between Italophone and Francophone passengers, who merely wanted to while away the time in jolly conversation. Another personal experience of a more long-term tacit contract, concerned a Galician and a Catalan, students working in the summer as chambermaids in a Swiss hotel, who habitually chatted in bilingual exchange, in preference to the use of their mutually shared language, Castilian, which was introduced only when there was danger of incomprehension on practical matters. Code-switching among Italian and Spanish migrant workers in French-speaking Switzerland is described in Del Coso-Calame et al. 1985, where it is claimed that family conver-

Language conflict in Romance: decline, death and survival

65

sations are habitually bilingual and unconscious use is made of interlectal forms (see also Quilis 1986). Similarly, Grosjean — Py 1991 show how Spanish immigrants to Neuchätel adjust their mother tongue grammar to their adopted language; and Vale Ferreira 1983 shows how the written Portuguese of children of Portuguese immigrant workers in France has adapted to French in its orthography, morphosyntax and vocabulary. Moretti 1992 describes how versions of Italian serve as a lingua franca among Romanophone immigrants in German-speaking Switzerland (and even among Greek immigrants), with frequent accommodation and codeswitching. Further studies on the way Romanophone immigrants adjust to adopted Romance languages — for instance an examination of the adopted Catalan of immigrants from the rest of Spain (see Sole 1983) — will doubtless yield interesting results. Multilinguals, in most circumstances, prefer to adjust to their interlocutors, switching in response to the language that initiates the exchange. Sometimes this is an effect of politeness, at others of some degree of ostentation. They may also switch, sometimes violating a tacit contract, because of impatience at their interlocutor's halting use of a non-native tongue, fearing breakdown of communication, or anxious to display their own linguistic prowess. Courtesy, on the other hand, may induce a bilingual to persist in the use of one language, in order to encourage an interlocutor's efforts to gain experience in that language. Comparatively little study has been made of bilingual interchange in Romance: indeed, as we have noted, where the distance — social, geographical or linguistic — between the two varieties is small, it may be difficult to distinguish accommodation, mixing, switching and bilingual interchange. Some Catalan theorists have advocated the encouragement of passive bilingualism (bilingiiismo passiu) as a solution to the problem of asymmetrical bilingualism, fostering the peaceful coexistence of Castilian with Catalan. Bastardas (1989) suggests that psychological factors hinder adoption of this strategy, but a study of interaction between young Castilian and Catalan mother-tongue speakers in a mainly Catalanspeaking district of Barcelona (Calsamiglia — Tuson 1980) shows that it can be favoured, though even here Catalan speakers are ready to give way when exchange is threatened, except insofar as they are ideologically motivated to exert their linguistic rights. Polyglot exchange has been suggested as a solution to the linguistic problems of the European Community, but when mooted at the College of Europe, Bruges in 1978, the proposal received short shrift (Coulmas 1991 a: 5).

66

Rebecca Posner

4. Is there a solution? Whether any strategy can prevent the eventual death of a language is doubtful — in Lord Keynes's words, "In the long term we are all dead". But conservationism, in the modern age of tolerance, can slow down the rate of language attrition, as long as it has the support of those most affected — the speakers of the language. References Academia Nacional 1982 Estudio sobre el problema idiomätico fronterizo. Montevideo: Academia Nacional de Letras. Alvar, Manuel 1976 La fronter a catalano-aragonesa. Zaragoza: Institution Fernando el Catolico. 1986a "Modalidades lingiiisticas aragonesas", in: Manuel Alvar 1986c: 133 — 142. 1986 b "Cuestiones de bilingüismo y diglossia en el espanol", in: Victor Garcia de la Concha (ed.), El castellano actual en las comunidades bilingües de Espana (Salamanca: Junta de Castilla y Leon), 11—48. Alvar, Manuel (ed.) 1986 c

Lenguas peninsulares y proyeccion hispänica. Madrid: Fundaciön Friedrich Ebert / Instituto de Cooperation Iberoamericana. Ammon, Ulrich — Norbert Dittmar — Klaus J. Mattheier (eds.) 1987 — 88 Sociolinguistics. An international handbook of the science of language and society. 2 vols. Berlin: Walter de Gruyter. Aracil, Lluis V. 1965 Conflit linguistique et normalisation linguistique dans l'Europe nouvelle. Nancy. [Mimeo. Reprinted as "Conflicte lingüistic i normalitzaciö lingüistica a l'Europa nova", in: Lluis V. Aracil 1982: 23 — 38.] 1982 Papers de sociolingiiistica. Barcelona: Edicions de la Magrana. Auer, Peter 1984 Bilingual conversation. Amsterdam: Benjamins. Auer, Peter — Aldo di Luzio (eds.) 1988 Variation and convergence: studies in social dialectology. Berlin: de Gruyter. Baetens Beardsmore, Hugo 1986 Bilingualism: basic principles. (2nd edition.) Clevedon, Avon: Multilingual Matters. Balibar, Renee — Dominique Laporte 1974 Le franqais national — politique de la langue nationale sous la Revolution. Paris: Hachette. Bastardas, Albert 1989 "Sobre el bilingüisme passiu: canvi i persistencia dels compartaments lingüistics", Revista de Catalunya 31: 43 — 54. Bianconi, Sandro 1980 Lingua matrigna. Italiano e dialetto nella Svizzera italiana. Bologna: II Mulino. Bischoff, Bernard 1961 "The study of foreign languages in the Middle Ages", Speculum 36: 209 — 224.

Language conflict in Romance: decline, death and survival

67

Boase, Roger 1990 "The morisco expulsion and diaspora: an example of racial and religious intolerance", in: David Hook — Barry Taylor (eds.), Cultures in contact in medieval Spain. Historical and literary essays presented to L. P. Harvey (London: King's College), 9 - 2 8 . Boleo, Manuel Paiva 1964 "O estudo das relasöes mütuas do portugues e do espanhol no Europa e na America, e influencia destas linguas em territörios da Africa e da Asia". [Reprinted with additional bibliography in: Manuel Paiva Boleo, Estudos de linguistica portuguesa e romänica (Coimbra: Universidade de Coimbra, 1974), 1: 3 5 5 - 3 9 8 . ] Burns, Robert I. 1984 Muslims, Christians, and Jews in the crusader kingdom of Valencia. Cambridge: Cambridge University Press. Byrne, Alvina 1989 A sociolinguistic study of Valencian: its usage and status. [Unpublished Masters thesis, University of Dublin.] Calsamiglia, Helena — Empar Tuson 1980 " l i s i alternanga de llengües en grups de joves d'un barri de Barcelona: Sant Andreu de Palomar", Treballs de sociolingüistica catalana 3: 11—82. Calvet, Louis-Jean 1987 La guerre des langues et les politiques linguistiques. Paris: Payot. Camara, J. Mattoso Jr. 1964 "O espanhol no Brasil e a conexäo entre ο portugues e ο espanhol no America do Norte", in: Presente y futuro..., 337 — 343. Cano Gonzales, Ana Maria 1987 "Averamientu a la hestoria de la llingua asturiana", in: Informe so la llingua asturiana (Oviedo: Academia de la Llingua Asturiana), 11—23. [French version, 65 — 77.] Carbonero, Pedro 1982 " N o r m a estandar y actitud sociolingüistica", in: Vidal Lamiquiz et al. (eds.), 1: 1 3 7 - 1 4 6 . Carbonero, Pedro (ed.) 1991 Habla de Sevilla y hablas americanas. (Sociolingüistica andaluza 5.) Sevilla: Publicaciones de la Universidad de Sevilla. Carvalho, Joaquim Brandäo de 1991 "«Cantabrie» et «Mozarabie»: de quelques divergences entre espagnol et portugais", La linguistique 27.2: 61 —73. Casanova, Emili 1988 "El valenciano dentro del diasistema lingüistico catalän", in: Aurora Juärez Blanquer (ed.), 1 3 - 2 4 . Catani, Maurizio 1983 "El estatus del habla local en relacion con la lengua nacional: el caso de Las Hurdes (Cäceres, Espana)", in: Louise Dabene et al. (eds.), 55 — 65. Climent, Teresa 1986 Realitat linguistica a la Val d'Aran. Barcelona: Gencralitat de Catalunya. Clyne, Michael 1987 "Constraints on code-switching: how universal are they?", Linguistics 25: 139-764.

68

Rebecca Posner

Conte, Anchel — Chorche Cortes — Antonio Martinez — Franco Nagore — Chesus Vasquez 1977 El aragones: identidad y problemätica de una lengua. Zaragoza: Libreria General. Coulmas, Florian 1991 a "European integration and the idea of the national language. Ideological roots and economic consequences", in: Florian Coulmas (ed.), 1 —37. Coulmas, Florian (ed.) 1991 b A language policy for the European Community: prospects and quandaries. (Contributions to the sociology of language 61.) Berlin: de Gruyter. Cremades Marco, F. de Borga 1982 La llengua valenciana, en perill. Valencia: Grup d'accio valencianista. Dabene, Louise — Monique Flasaquier — John Lyons (eds.) 1983

Status

of migrants'

mother

tongues

/ Le Statut des langues

d'origine

des

migrants. Strasbourg: European Science Foundation. D'Andres, Ramon 1987 "La situation social de la llingua asturiana", in: Informe so la Uingua asturiana (Oviedo: Academia de la Llingua Asturiana), 25—44. [French version, 79-94.] Dahrendorf, Ralf 1959 Class and class conflict in industrial society. London: Routledge and Kegan Paul. Dann, Otto — John R. Dinwiddy (eds.) 1988 Nationalism in the age of the French Revolution. London: Hambledon. Del Coso-Calame, Francine — Francois de Pietro — Cecilia Oesch-Serra 1985 "La competence de communication bilingue. Etude fonctionnelle des codeswitchings dans le discours de migrants espagnols et Italiens ä Neuchätel (Suisse)", in: Elisabeth Gülich — Thomas Kotschi (eds.), Grammatik, Konversation, Interaktion. Beiträge zum Romanistentag 1983 (Tübingen: Niemeyer), 377-398. Di Sciullo, Anne-Marie — Pieter Muysken — Rajendra Singh 1986 "Government and code-mixing", Journal of linguistics 22: 1 —24. Dorian, Nancy C. (ed.) 1989 Investigating obsolescence. Studies in contraction and death. Cambridge: Cambridge University Press. Dressier, Wolfgang U. 1988 "Language death", in: Frederick J. Newmeyer (ed.), Linguistics: the Cambridge survey. IV Language: the socio-cultural context (Cambridge: Cambridge University Press), 184-192. Emsley, Clive 1988 "Nationalist rhetoric and national sentiment in revolutionary France", in: Otto Dann - John R. Dinwiddy (eds.), 3 9 - 5 2 . European Science Foundation 1990 Papers for the workshop on concepts, methodology and data. Basel 12—13 January, 1990. Strasbourg: ESF scientific network on code-switching and language contact. Ferrando Frances, Antoni 1980 Consciencia idiomatica i nacional dels valencians. Valencia: Universität.

Language conflict in Romance: decline, death and survival 1988

69

"Presente y futuro de la normalization lingüistica en el Pais Valencian", in: Aurora Juarez Blanquer (ed.), 25 — 34. Filgueira Valverde, Xose 1982 "O galego escrito, entre ο latin e ο castelan, no medievo", in: Dieter Kremer - Ramon Lorenzo (eds.), Tradition, actualidad e futuro do galego. Adas do Coloquio de Treveris 13 a 15 de novembre de 1980 (Santiago de Compostela: Xunta de Galicia), 127-130. Fishman, Joshua A. 1972 Language and nationalism. Two integrative essays. Rowley, Mass.: Newbury House. Fortier, Alcee 1884 — 85 "The French language in Louisiana and the Negro dialect", Transactions of the modern languages association of America 1: 96 — 111. 1891 "The Acadians in Louisiana and their dialect", Publications of the modern language association 6: 64 — 94. Fuertes Alvarez, Dionisio 1964 "La lengua espanola en el Rio Grande del Sur (Brasil)", in: Presente y futuro..., 361—366. Garcia Morejon, Julio 1964 "Creadon y desarrollo del hispanismo en Brasil", in: Presente y futuro..., 345-359. Gardner-Chloros, Penelope 1987 "Code-switching in relation to language contact and convergence", in: Georges Lüdi (ed.), Devenir bilingue — parier bilingue: actes du Τ colloque sur le bilinguisme, Universite de Neuchätel, 20 — 22 septembre 1984 (Tübingen: Niemeyer), 9 9 - 1 1 5 . 1990 Language selection and switching in Strasbourg. Oxford: Clarendon Press. Gardy, Philippe — Robert Lafont 1981 "La diglossie comme conflit: l'exemple occitan", Langages 61: 75 — 91. Gellner, Ernest 1983 Nations and nationalism. Oxford: Blackwell. Giles, Howard (ed.) 1984 The dynamics of speech accommodation. Berlin: Mouton. [ = International journal of the sociology of language 46.] Giles, Howard — Robert N. St Clair (eds.) 1979 Language and social psychology. Oxford: Blackwell. Giles, Howard — Philip M. Smith 1979 "Accommodation theory: optimal levels of convergence", in: Howard Giles - Robert N. St Clair (eds.), 4 5 - 6 5 . Giles, Howard — Donald M. Taylor — Richard Y. Bourhis 1973 "Towards a theory of interpersonal accommodation through language: some Canadian data", Language in society 2: 177—192. Godechot, Jacques 1988 "The new concept of the nation and its diffusion in Europe", in: Otto Dann - John R. Dinwiddy (eds.), 1 3 - 2 6 . Gordon, David C. 1978 The French language and national identity. The Hague: Mouton.

70

Rebecca Posner

Grillo, Ralph D. 1989 Dominant languages: language and hierarchy in Britain and France. Cambridge University Press. Grosjean, Frangois — Bernard Py 1991 "La restructuration d'une premiere langue: l'integration de variantes de contact dans la competence de migrants bilingues', La linguistique 27.2: 35 — 60. Haugen, Einar 1966 Language conflict and language planning. The case of modern Norwegian. Cambridge, Mass.: Harvard University Press. 1972 Ecology of language. Stanford: Stanford University Press. Heller, Monica (ed.) 1988 Code-switching, anthropological and linguistic aspects. Berlin: Mouton de Gruyter. Holmquist, Jonathan Carl 1988 Language loyalty and linguistic variation. A study in Spanish Cantabria. Dordrecht: Foris. Holtus, Günter — Georges Lüdi — Michael Metzeltin (eds.) 1989 La corona de Aragon y las lenguas romänicas. Miscelänea de homenaje para Germän Colon. Tübingen: Narr. Inglehart, Ronald F. — Margaret Woodward 1972 "Language conflicts and political community", in: Pier Paolo Giglioli (ed.), Language and social context (London: Penguin), 358 — 377. [Excerpted from a longer article published in 1967.] Jernudd, Bjorn — Michael Shapiro 1989 The politics of language purism. Berlin: Mouton de Gruyter. Juarez Blanquer, Aurora (ed.) 1988 Las lenguas romänicas espanolas tras la constitucion de 1978. Granada: Tat. Kahane, Henry — Renee Kahane 1976 "Lingua franca: the story of a term", Romance Philology 30: 25 — 51. 1979 "Decline and survival of Western prestige languages", Language 55: 183 — 198. Knecht, Pierre — Christian Rubattel 1985 "Problemes sociolinguistiques en Suisse romande", in: Iwar Werlen (ed.), Probleme der schweizerischen Dialektologie. Problemes de la dialectologie suisse (Fribourg: Editions universitäres), 141 — 158. Koike, Dale Α. 1983 "The inflected infinitive as a marker of social dialects", in: John J. Bergen — Garland D. Bills (eds.), Spanish and Portuguese in social context (Washington DC: Georgetown University Press), 92 — 104. Kremnitz, Georg 1979 Sprachen im Konflikt: Theorie und Praxis der katalanischen Soziolinguistik. Tübingen: Narr. 1981 " D u bilinguisme au conflit linguistique. Cheminement de termes et de concepts", Langages 61: 63 — 74. Krüger, Fritz 1925 "Mezcia de dialectos", in: Homenaje ofrecido a Menendez Pidal (Madrid: Hernando), 2: 1 2 3 - 1 6 6 . Lamiquiz, Vidal — Pedro Carbonero 1987 Perfil sociolingüistico del sevillano culto. Sevilla: Instituto de Desarrollo Regional.

Language conflict in Romance: decline, death and survival

71

Lamiquiz, Vidal et al. (eds.) 1982 — 87 Sociolingüistica andaluza. 4 vols. Sevilla: Publicaciones de la Universidad de Sevilla. Lang,Jürgen 1982 Sprache im Raum. Zu den theoretischen Grundlagen der Mundartforschung. Unter Berücksichtigung des Rätoromanischen und Leonesischen. Tübingen: Niemeyer. Laponce, Jean A. 1984 Langue et territoire. Quebec: Les Presses de l'Universite Laval. Le Page, Robert — Andree Tabouret-Keller 1985 Acts of identity. Cambridge: Cambridge University Press. Lindholm, Kathryn J. — Amado M. Padilla 1978 "Language mixing in bilingual children", Journal of child language 5: 327-335. Lipski, John Μ. 1985 Linguistic aspects of Spanish-English language switching. Tempe: Arizona State University, Center for Latin American Studies. Llorente Maldonado de Guevara, Antonio 1986 "Las hablas vivas de Zamora y Salamanca en la actualidad", in: Manuel Alvar (ed.), 1 0 7 - 1 3 2 . Lloyd, Paul M. 1987 From Latin to Spanish. Philadelphia: American Philosophical Society. Lurati, Ottavio 1976 Dialetto e italiano regionale nella Svizzera italiana. Lugano: Banca Solari & Blum. Lusignan, Serge 1987 Parier vulgairement. Les intellectuels et la langue fran Leonese 2 > Leonese 3 to be read: Leonese abstand + ausbau language (1); Leonese abstand language as a 'heterogeneous dialect' of the Spanish ausbau language (2); Leonese as a dialect of the Spanish abstand language (3) (see Muljacic 1990, 1991). Better still are nonsynonymous pairs like: Florentine > Italian; Castilian > Spanish. Despite the existence of a flourishing literature and a consolidated ausbau language, a 'co-official' language functions in diglossic conditions of a new type (true, it can be the vehicle for all types of oral and written texts but its speakers cannot use their language in all situations in their own land); and, given time, it can be ousted from its status. That the struggle against diglossia is the driving force behind linguistic development (Bier-

96

Zarko Muljacic

bach — Neu-Altenheimer 1982: 113) is only a half-truth. The same can be said of the expansion of diglossia. Between 'total independence' and 'total dependence' there is a mid position, as pointed out by Montes Giraldo (1986). It is more frequent than commonly supposed and will very likely increase and generate triglossic conditions with the growth of European economic unity, at least in some text types. Not all eleven official languages of the twelve member states will be 'business languages' (and there are new candidates for the second place in possible homogenized repertories: 1. English, 2. Catalan ausbau language, 3. Catalan dialect). The concept 'dialect through subordination' will evolve in the years to come. Many idioms have been, are, and will be both languages and 'dialects' so Hegelian discourse {sowohl ... als auch 'both ... and') offers the basis of a relativistic model of linguistics. 5.5 Summary We have tried to rid our discussion of chauvinistic or sentimental overtones. The concepts national state and its corollary national language — at least in the Western community and its imitators — are not everlasting, and indeed are being eroded under our very eyes from two directions: supernational integrations and (sub)national emancipations. The language question not only of 'new' but also of 'old' languages will recur in new contexts in the decades to come. Nobody knows whither Europe is bound (a Europe of states or a Europe of regions). Whatever happens, it will grow from a synthesis and not from a summation. And languagemaking is an endless process which "involves much more than merely the construction of systems of signs. It is also the essential process by which men construct a cultural identity for themselves, and for the communities to which they see themselves as belonging" (Harris 1980: i). Discussion of the whole spectrum of linguistic means available to speech communities (especially if the main idioms of their repertories are 'bi-dependent') will, if seen in these terms, represent only one and not the final resting place of standardization. Notes 1. See, for this distinction, Bartsch 1987, Joseph 1987, and Joseph 1989. Every standard language is a corpus of language standards. Even dialect speakers whose dialect has no written grammars or dictionaries know its norms; they can decide if a word, and so on, is correct or not correct. 2. The best term would be standardology. 3. In German, of course, these terms are written with capitals: Abstandsprache, Ausbausprache. See Kloss 1967 and 1976.

Standardization in Romance

97

4. This single term covers nascent, half developed and fully fledged 'languages by elaboration'. 5. Allen 1969: passim emphasizes the resemblance between standard language learning by dialect speakers and foreign language learning. Recent research has shown that the learning of the standard language differs from the learning of a second/foreign language in at least four points. 6. See Corbeil 1987: 561 ff.; D a o u s t - M a u r a i s 1987: 12ff. 7. See Domasnev 1978; Svejcer-Nikol'skij 1986. 8. In West German praxis a variety (Germ. Varietät) is a sum of differential variants (Germ. Variante). 9. See the discussion of this neologism in Garvin —Mathiot 1968: 371 —372. 10. The term language planning has been literally translated (It. pianificazione linguistica, Germ. Sprachplanung) or rendered with coinages (Fr. glottopolitique, amenagement Unguistique). These terms suggest the idea of conflict and allude to a third type of planning prestige planning. See Corbeil 1987; D a o u s t - M a u r a i s 1987; and Marcellesi 1986. 11. Central are endings of substantives and adjectives, and tenses of the indicative and subjunctive. Peripheral are, for example, the different substitutes of the negative imperative; compare niente correre! 'do not run!' (in all persons) with Standard Italian non correre!, non corra!, and so on. This form is not considered dialectal but a component of the Regional Italian of Sicilia. 12. Because of the neologism standardology it would be possible to use a less shocking term. Normative linguistics must however be distinguished from normative grammar. 13. A colloquium held at the University of Corti (Corsica) in September 1990 was dedicated to the theme 'Les langues polynomiques'. See also Thiers, this volume. 14. Svejcer-Nikol'skij studied situations where the same language serves different nations, and emphasize (1986: 20): "The term standard language variant was applied to the variety of a standard language, limited to a certain national area (for example Standard English in the USA, Standard German in Austria, Standard French in Canada, Standard Spanish in Argentina and other Latin American countries). The totality of a standard language variant and the substandard (dialect) varieties (territorial and social) within the same area forms a national language variant. The relationship between the standard language variant and the substandard idioms within the same area is the same as between the standard language and the dialects within a language serving one nation". See also Muljacic 1988 d. 15. Francorsu (The French of indigenous Corsicans) must be distinguished from frangais regional de Corse of speakers whose mother tongue is French. 16. If the dominating language is very similar to the dominated one, an ecart maximum is desirable not only in lexicon but in phonology and syntax too. See Proces-verbaux 1931: 3 0 3 - 3 0 4 ; Muljacic 1989e. 17. Brozovic (1970: 29 — 30) gives two lists of Romance languages: the first does not include Moldavian, the second does not include Occitan and Sardinian. Compare Muljacic 1980. 18. Munske (1983: 1009), speaking of "dismantling" phenomena (Germ. Abbauerscheinungen) in some German colloquial languages (in transitional forms of adjustment between the German Standard Language and certain dialects), states that the genuine development of dialects was interrupted: "Durch solchen Sprachwandel..., sie verlieren ihre ... Selbständigkeit und treten nunmehr in ein abhängiges, deszendentes Verhältnis zur

98

Zarko Muljacic Hochsprache ... Die ursprünglich rein funktionale Überdachung des Dialektes durch die Hochsprache (Diglossie) wird aufgrund des Dialektabbaus, der Homogenisierung im Sprachenkontakt, zu einer strukturellen Überdachung".

Appendix I: Chronological survey of some relevant

meetings

[Note: these titles are given in their original form, except that multilingual titles have been abbreviated to the English version only. The Appendix is not intended as a list of published proceedings, though some bibliographical information is given indirectly, through crossreferences.] Premier congres des philologues slaves, Prague, 1929. See Theses 1929; Vachek 1983 a, 1983 b. Reunion phonologique internationale, Prague, 1930. Vc congres international des linguistes, Bruxelles, 28 aoüt —2 septembre 1939. Summaries and answers to questions of this congress (which did not take place) have been published: Pintelon 1973; S j o e s t e d t - J o n v a l 1973. VIII congresso internazionale di studi romanzi, Firenze, 3 — 8 aprile 1956. Eighth international congress of linguists, Oslo, 1957. Section: Languages in contact. Ninth international congress of linguists, Cambridge, Massachusetts, August 27 — 31, 1962. Section: Language and society. UCLA sociolinguistics conference, Los Angeles Campus, Center for Research in Languages and Linguistics, May 11 — 13, 1964. See Ferguson 1966; Haugen 1971. Language problems of developing nations, Airlie House, Warrenton (Virginia), November 1 - 3 , 1966. e X congres international des linguistes, Bucarest, 28 aoüt—2 septembre 1967. Section: Sociolinguistique. Can language be planned? Sociolinguistic theory and practice for developing nations, University of Hawaii, Honolulu, April, 7 — 10, 1969. Giornate internazionali di sociolinguistica: congresso internazionale di scienze dell'Istituto Luigi Sturzo, Roma, 1969. XIII e congres international de linguistique et philologie romanes, Universite Laval (Quebec, Canada), du 29 aoüt au 5 septembre 1971. Themes: VI, VII, VIII, X. Eleventh international congress of linguists, Bologna-Firenze, August 28 —September 2, 1972. Section: Sociolinguistics. XIV congresso internazionale di linguistica e filologia romanza, Napoli, 15 — 20 aprile 1974. Round table: I problemi politico-culturali delle lingue romanze nel mondo attuale. VIII congresso internazionale di studi, SLI, "Aspetti sociolinguistici dell'Italia contemporanea, Bressanone, 31 maggio — 2 giugno 1974. VIII world congress of sociology, Toronto, August, 1974. Centre Internacional Escarre sobre les minories etniques i nacionals [ = CIEMEN]. Primeres jornades del C I E M E N , Abadia de Cuixa, 22 — 29 d'agost de 1976, "Dret i minories nacionals. Relacions lingüistiques occitano-catalanes". Colloque sur les minorites linguistiques. Universite Laval, 15 — 18 avril 1977, Quebec. Conseil International de la Langue Frangaise, "Le fransais en contact avec: la langue arabe, les langues negro-africaines, la science et la technique, les cultures regionales", Sassenage, 1 6 - 2 0 mai 1977. XI congresso internazionale di studi, SLI, "I dialetti e le lingue delle minoranze di fronte aH'italiano", Cagliari, 2 7 - 3 0 maggio 1977.

Standardization in Romance

99

Segonesjornades del C I E M E N , Abadia de Cuixä, 1 6 - 2 2 d'agostde 1977. "Balears-Pitiüses, Cörsega, Sardenya. Per les reivindicaciones nationals". Twelfth international congress of linguists, Vienna, August 28 —September 2, 1977. Section: Language and society. Romanistentag, Gießen, 1977. Thema IV, 2: Sprachliche Minderheiten. 5 έπκ colloque international de linguistique fonctionnelle (Ioannina, 10 —15 juillet 1978). Theme II: Planification linguistique et choix de langue. Terceres jornades del C I E M E N , Abadia de Cuixä, 16—23 d'agost de 1978, "Les autonomies en diferents estats. Experiencies i perspectives". 2. Symposion über Sprachkontakt in Europa, "Standardsprache und Dialekte in mehrsprachigen Gebieten Europas", Mannheim 1978. Colloque "Sociolinguistique. Approches, theories, pratiques", Universite de Rouen, Faculte des Lettrcs de Mont-Saint-Aignan, 27 novembre —2 decembre 1978, G R E C O . Colloque international "Langues et cooperation europeenne", Strasbourg, Palais de l'Europe, 17 — 20 avril 1979, Centre d'Information et Recherche pour l'Enseignement et l'Emploi des Langues (CIREEL). Quartes jornades del C I E M E N . Abadia de Cuixä, 16 — 23 d'agost de 1979, "Fet nacional; llengua, territori, migrations. Paisos Catalans, Euskadi, Galicia, Bretanya, Occitänia, Flandes". Colloque sur la sociolinguistique dans les pays de langue romane, Francfort-sur-le-Main, 2 4 - 2 5 septembre 1979. See Dittmar-Schlieben-Lange 1982. Progress in language planning: international perspectives, Paterson State College, Paterson (NJ), 1979. See Haugen 1983. 3. Symposion über Sprachkontakt in Europa "Sprachvariation und Sprachwandel. Probleme der Inter- und Intralinguistik", Mannheim 1979. Symposium "Contact + confli(c)t" [I], Brussels, 1979. Research Centre on Multilingualism, Brussels. XVI congres internacional de Iingüistica i filologia romäniques, Ciutat de Mallorca, 7 — 12d'abril de 1980. Taula rodona: Sociolingüistica i Iingüistica romänica. Sectio III: Histöria externa de les llengües romäniques i sociolingüistica. Cinquenes jornades del C I E M E N , Abadia de Cuixä, 18 — 24 d'agost de 1980, "Ensenyament de la llengua i mitjans de comunicaciö social". First international conference on minority languages, Glasgow University, September 8 — 13, 1980. Journees Pluriel, Rennes, 2—4octobre 1980, Universite de Haute-Bretagne, Rennes II. 4. Symposion über Sprachkontakt in Europa, "Kulturelle und sprachliche Minderheiten in Europa. Aspekte der europäischen Ethnolinguistik und Ethnopolitik", Mannheim 1980. Problemi linguistici ed unitä europea, Merano/Meran, 25 —28 ottobre 1980, Libera Universitä degli Studi di Trento. Coloquio cientifico internacional "Tradition, actualidade e futuro do galego, Trier/Treveris (13 a 15 de novembro de 1980). Simposio sobre sociolingüistica. Madrid, 17 —20 de diciembre de 1980. Sociedad Espafiola de Lingüistica. eme 2 colloque international "Langues et cooperation europeenne. Status et gestion des langues", Urbino, Centro Alti Studi Europei, 1 6 - 2 0 settembre 1981. CIREEL. Seminario sull'educazione plurilingue in Italia, Frascati, 12 —14 ottobre 1981. Centro europeo dell'educazione (CEDE), Villa Falconieri (Frascati).

100

Zarko Muljacic

Colloque international de sociolinguistique "Les situations de diglossie", Montpellier, 3 — 5 decernbre 1981, Universite Paul-Valery, Montpellier III. See Bierbach — NeuAltenheimer 1982. Fifth Burdick-Vary symposium, April 1 —3, 1982, Institute for Research in the Humanities of the University of Wisconsin (Madison). See Scaglione 1984 a, 1984 b; Malkiel 1984. XVI congresso internazionale di studi, SLI, "Linguistica storica e cambiamento linguistico", Firenze, 7 — 9 maggio 1982. See Muljacic 1985 a. Symposium "Contact + confli(c)t II", Brussels, June 2 — 5, 1982, Research Centre on Multilingualism, Brussels. 5. Symposion über Sprachkontakt in Europa "Die Leistung der Strataforschung und der Kreolistik. Typologische Aspekte der Sprachkontakte", Mannheim 1982. XHIth international congress of linguists, August 29 —September 4, 1982, Tokyo. Working groups 12 and 13: European contact linguistics. "Das Romanische in den Ostalpen", Institut für Romanistik der Universität Salzburg, 6 - 1 0 Oktober 1982, Salzburg. 17. linguistisches Kolloquium. Sektion II: Mehrsprachigkeit und Gesellschaft, Brüssel 1982. "Le minoranze linguistiche. Stato attuale e proposte di tutela". Convegno della Societä Italiana di Glottologia, Pisa, 16 e 17 dicembre 1982. Second international conference on minority languages, Turku/Äbo, June 6 — 12, 1983. XVIpmc c o t l g r e S international de linguistique et philologie romanes (Aix-en-Provence, 29 aoüt — 3 septembre 1983). Section: Sociolinguistique des langues romanes. See Bochmann 1985 a. Colloque international "Langues et conflits", Paris, decembre 1983, Institut de politique internationale et europeenne, Universite de Paris X — Nanterre. "Nation et nationalstes en Espagne". Col loqui internacional sobre nacio i nacionalitats a Espanya (segles XIX i XX), Fondation Singer-Polignac, Paris, 28 — 31 mars 1984. Congressu internaziunale nantu ä u bislingvisimu — Congres international sur le bilinguisme, Bastia, 12 — 15 avril 1984. "La formazione delle lingue letterarie", Convegno della Societa Italiana di Glottologia, Siena, 1 6 - 1 8 aprile 1984. See Muljacic 1985b. XL colloque international de linguistique fonctionnelle, Bologne, 2 —7 juillet 1984. Theme III: Langue nationale et langues regionales. Symposium "Contact + confli(c)t III", Brüssels, August 11 — 12, 1984, Research Centre on Multilingualism, Brussels. VIII congresso internazionale di studi, SLI, "L'italiano regionale", Padova — Vicenza, 1 3 - 1 6 settembre 1984. Symposium international "Problemes de glottopolitique", Universite de Rouen, MontSaint-Aignan, 2 0 - 2 3 septembre 1984. I congresso internacional da lingua galego-portuguesa na Galiza, Associa^om Galega da Lingua (AGAL), Ourense, 20 — 24 setembro 1984. 6. Symposion über Sprachkontakt in Europa, "Entstehung von Sprachen und Völkern. Glotto- und ethnogenetische Aspekte europäischer Sprachen", Mannheim 1984. See Schmitt-Brandt 1985. Internationales Kolloquium "Die soziolinguistische Variation — theoretische Probleme und romanische Applikationen", Leipzig, 1 — 3 November 1984. See Bochman 1985 b; Marcellesi 1985. Colloque sur le statalisme, Institut des Hautes Etudes de Belgique, Bruxelles, 13 —14 mars 1985.

Standardization in Romance

101

Symposium "La production d'identite", Sommieres, les 30, 31 mai et l " j u i n 1985, Universite Paul-Valery, Montpellier III. Curso de verano "Lenguas peninsulares y proyeccion hispänica", Malaga, julio —agosto 1985. Fundaciön Friedrich Ebert y Instituto de Cooperation Ibero-americana, Madrid. Theodor-Gartner-Tagung "Rätoromanisch und Rumänisch", Vill/Innsbruck, 23 — 25 September 1985. Primo congresso internazionale "Le minoranze etniche e nazionali", Comune di Piana degli Albanesi — Facoltä di lettere dell'Universitä di Palermo — Facoltä di lettere e filosofia dell'Universitä della Calabria, Palermo — Piana degli Albanesi, 4 — 7 dicembre 1985. Ciclo " M a p a lingüistico de la Espana actual", Madrid, 22 de abril —13 de mayo de 1986, Fundaciön Juan March. Segon congres internacional de la llengua catalana, 30d'abril —11 de may 1986. See Garvin 1986. XVIIT congres international de linguistique et de philologie romanes, Universite de Treves (Trier), du 19 au 24 mai 1986. Section IV: Linguistique pragmatique et linguistique sociolinguistique. See Goebl 1988. International colloquium on language planning, Ottawa, May 25 — 29, 1986. See C I R B — ICRB 1986. Third international conference on minority languages, University College, Galway (Ireland), June 2 1 - 2 6 , 1986. Xlth world congress of sociology of the International Sociological Association (ISA), New Delhi, 1 8 - 2 3 August 1986. See Ammon 1989. 13c colloque international de linguistique fonctionnelle, Corfou, 24 — 29 aoüt 1986. Theme III: La normalisation des langues. XX congresso internazionale di studi SLI "L'italiano tra le lingue romanze", Bologna, 25 — 27 settembre 1986. See Muljacic 1989 c. 24° convegno dell'AIMAV, "Isole linguistiche e cultural! all'interno di culture minoritarie: problemi psico-linguistici, sociolinguistici, educativi", Udine, 13 — 16maggio 1987. XIII conceyu internacional de llingiies y cultures amenazaes, Uvieu/Oviedo, 24 — 26 de xunetu de 1987. XIVth international congress of linguists, Berlin, August 10—15, 1987. Sections: VIII and IX. Convegno regionale "Scuola, lingue e culture locali", Villa Manin di Passariano, 4 — 5 settembre 1987. II congresso internacional da lingua galego-portuguesa na Galiza, AGAL, Santiago-Ourense, 2 3 - 2 7 setembro 1987. Symposium on languages of minor diffusion, Santiago de Compostela, September 24—26, 1987. Convegno internazionale "Koine in Italia dalle Origini al Cinquecento", Milano —Pavia, 2 5 - 2 6 settembre 1987. Colloque "Contact de langues. Quels modeles?", Nice, 28 — 30 septembre 1987, Centre d'Etudes des Plurilinguismes. "Tipologie della convergenza linguistica", Convegno della Societä Italiana di Glottologia, Bergamo, 17 - 1 9 dicembre 1987. Colloquy "Lesser used languages in primary education", Ljouwert/Leeuwarden (Netherlands), April 1988, Fryske Akademy. Lega per le lingue delle nazionalita minoritarie (LeLiNaMi) dello stato italiano, "Convegno di studio sulla normalizzazione e standardizzazione delle lingue friulana, ladina e sarda",

102

Zarko Muljacic

Darte/Arta (UD), 2 8 - 2 9 maggio 1988, Vich/Vigo di Fassa (TN), 17 settembre 1988, Santu Lussurgiu (OR), 25 settembre 1988. International symposium "Contact + confli(c)t [IV]", Brussels, June 2—4, 1988. Research Centre on Multilingualism, Brussels. XIII jornades internacionals del C I E M E N , Abadia de Cuixä, 16 — 23 d'agost 1988, "Les nacions petites amb estat: les seves problemätiques i els seus exemples per a les nacions sense estat". 2° congresso internazionale "Le minoranze etniche e linguistiche", Piana degli Albanesi, Comune di Piana degli Albanesi. V centenario della fondazione, 7 — 11 settembre, 1988. XVIII convegno di studi dialettali italiani "Fra dialetto e lingua nazionale: realtä e prospettive", Lugano, 1 1 - 1 5 ottobre 1988. See Muljacic 1991. VII xornaes d'estudiu "La normalization de la llingua", Uvieu, 26 — 28 ochobre 1988. Academia de la Llingua Asturiana. See Muljacic 1989 a. 5. romanistisches Kolloquium "Zum Stand der Kodifizierung romanischer Kleinsprachen", Eichstätt, 6 —9 Dezember 1988. Katholische Universität Eichstätt. See Dahmen et al. 1990. XXIII congresso internazionale di studi SLI, "Storia dell'italiano e forme deH'italianizzazione", Trento — Rovereto, 18 —20 maggio 1989. See Muljacic 1990. Fourth international conference on minority languages "Comparative research and development of theories", Ljouwert/Leeuwarden (Netherlands), June 20 — 24, 1989, Fryske Akademy. XIX congreso internacional de lingüistica e filoloxia romänicas. Universidade de Santiago de Compostela do 4 ό 9 de setembro de 1989. Secciön III: Lingüistica pragmätica e sociolingüistica. Section VI: Galego. Convegno europeo "Innovazione nella tradizione: problemi e proposte delle comunitä di lingua minoritaria", Comitato Nazionale Federativo Minoranze Linguistiche d'Italia ( C O N F E M I L I ) äderente a The European Bureau for Lesser Used Languages, UdineCodroipo 14 - 1 6 settembre 1989. X V I P colloque international de linguistique fonctionnelle, Leon, 5 —lOjuillet 1990. Theme 3: Les langues dans l'Europe des douze. Colloque international "Les langues polynomiques", Universite Pasquale Paoli, Corti, 1 7 - 2 2 septembre 1990. Languages in contact, languages in conflict, Societas Lingüistica Europaea Annual Meeting, Berne, September 1 8 - 2 1 , 1990. I. gemeinsames Kolloquium der deutschsprachigen Lusitanistik und Katalanistik, Berlin, 2 0 - 2 3 September 1990. III congresso internacional de lingua galego-portuguesa na Galiza, AGAL, Vigo —Santiago/ Ourense, 28 — 30 de setembro e 1 — 2 de outubro de 1990. Colloque international "Standardisation linguistique. Elements pour une theorie de la standardisation linguistique", Parpan —Chur (Suisse), 15 — 20 avril 1991, organise par l'Academie Suisse des Sciences Humaines en collaboration avec la Commission de l ' U N E S C O pour la Suisse. II Γ colloque du groupe d'etude sur le plurilinguisme europeen, Uni versite des Sciences Humaines de Strasbourg, 30 mai — 1 juin 1991. Section: Politiques linguistiques en Europe. XVIIF colloque international de linguistique fonctionnelle, Prague, 12 —17juillet 1991. Theme 2: Les langues dans l'Europe, de l'Atlantique a l'Oural (situation et Statut des diverses langues).

Standardization in Romance

103

First international seminar on language planning (in territories with their own language under pressure from other hegemonic languages), Galician Culture Council, Santiago de Compostela, September 2 5 - 2 8 , 1991. Symposium "State and plurilingual population", Barcelona, 16 —17 December, 1991.

Appendix II: Chronological II (a) Journals

list of journals and series

Europa ethnica. Vierteljahresschrift für Nationalitätenfragen... Problems of nationalities, Wien, 18 (1958) — . Its 'predecessor' Nation und Staat, volumes 1 — 17, Wien — Leipzig, 1927 — 1944, is irrelevant for the present theme, as it studied almost exclusively German ethnic groups abroad. Langages, Paris: Didier-Larousse, 1966 — , Rassegna italiana di sociologia, 9.2 (1968). La monda lingvo-problemo. Rotterdam: Center for Research on International Language Problems, 1969 — . It ends with volume 6 (1974). For the continuation, see Language problems and language planning (below). Plural society. Quarterly journal, The Hague: Foundation for the Study of Plural Societies, 1970 — . Language in society, ed. Dell Hymes, Cambridge: Cambridge University Press, 1972 — . Sociolinguistics. Journal of the research committee on sociolinguistics of the International Sociological Association, Dordrecht — Providence (RI): Foris Publications, 1972 — , Canadian review of studies in nationalism — Revue canadienne des etudes sur le nationalisme, Charlottetown: University of Prince Edward Island, 1974 — . International journal of the sociology of language, ed. Joshua A. Fishman, Berlin: Mouton de Gruyter, 1 9 7 4 - . Lingua e contesto. Nuovi studi di dialettologia linguistica, geografica, sociologica, pragmatica. Serie di studi e monografie diretta da Armistizio Matteo Melillo et al., Manfredonia: Atlantica Editrice, 1 9 7 5 - , It ended with volume 6 (1983). La grotta della vipera. Rivista trimestrale di cultura, dir. Antonio Cossu, Cagliari, 1975 — . Pluriel. Recherches en sciences sociales sur les relations inter-ethniques, les relations raciales, les contacts culturels, les problemes de minorites et la question nationale, resp. PierreJean Simon, Paris, 1975 — . Cahiers de linguistique sociale, Rouen: GRECSO, Universite de Haute-Normandie, 1976 — . Minoranze, Milano: Associazione Internazionale per la Difesa delle Lingue e delle Culture Minacciate (AIDLCM), 1 9 7 6 - . It ended with volume 12. Ladinia. Sföj cultural dai Ladins dies Dolomites, San Martin de Tor: Istitut Ladin "Micura de Rü", 1 9 7 7 - . Langage et societe, dir. Pierre Achard, Paris, 1977 — . Language planning newsletter, ed. Joan Rubin, University of Hawaii, East-West Center, 1977-1986; Central Institute for Indian Languages, 1 9 8 7 - . Language problems and language planning (Lingvaj problemoj kaj lingvo-planado). Volumes 1—3 (1977 — 1979), Rotterdam: Center for Research on International Language Problems. Volume 4 (1980) — , Austin: University of Texas Press, with the cooperation of CRILP. Lengas. Revue de sociolinguistique, Montpellier: Universite Paul-Valery, Montpellier III, 1977 — . Rivista italiana di dialettologia. Scuola, societä, territorio, eds. L. Cöveri — F. Foresti, Bologna: CLUEB, 1 9 7 7 - 1 9 8 7 . Volume 12 ( 1 9 8 8 ) - , ed. F. Foresti.

104

Zarko Muljacic

Treballs de sociolingiiistica catalana, Valencia: Climent, 1977 — , Sprache & Herrschaft. Zeitschrift für eine Sprachwissenschaft als Gesellschaftswissenschaft, ed. Peter Bettelheim, Wien, 1 9 7 8 - . Interrupted (temporarily?) 1986. Etudes Creoles. Culture, langue, societe, Ottawa: Didier, 1979 —. Journal of multilingual and multicultural development, Clevedon (Avon), UK: Multilingual Matters, 1 9 8 0 - . AmiraslReperes occitans, dir. Robert Lafont, Montpellier, 1982 — . Etnie, dir. Tavo Buratti, Milano, 1982 — . Lletres asturianes. Boletin oficial de l'Academia de la Llingua Asturiana Principäu d'Asturies, Uvieu/Oviedo, 1 9 8 2 - . Multilingua. Journal of cross-cultural and inter-language communication, ed. Richard J. Watts, Berlin-New York: Walter de Gruyter, 1 9 8 2 - . Revista de llengua i dret, Barcelona: Escola d'Administracio Publica — Generalitat de Catalunya, 1 9 8 3 - . Linguaggi. Bollettino quadrimestrale, dir. Stefano Gensini, Roma: Coop Spazio Linguistico, 1 - 3 (1984-1986). Contact. The European Bureau for Lesser Used Languages, Luxembourg, Commission des communautes europeennes, 1984—. Neue Romania, Berlin: Institut für Romanische Philologie der Freien Universität Berlin, 1984-. Limbas. Rivista de istudios linguisticos, Nugoro/Nuoro (Sardinna), 1986 — . Diverse lingue. Rivista semestrale delle letterature dialettali e delle lingue minori, Udine: Campanotto Editore, 1986 — , Sociolinguistica. International yearbook of European sociolinguistics, ed. U. Ammon — K. J. Mattheier — P. H. Neide, Tübingen: Niemeyer, 1987 — . Europa de les nacions. Revista sobre el moviment nacionalitari europeu, Barcelona: CIEMEN, 1988 —.

II (b) Series Nationalia (Publicacions de l'Abadia de Montserrat), Barcelona: Centre Internacional Escarre sobre les minories etniques i nacionals, 1977 — . See Primer es jornades, 1976, Appendix I. Akten des ... Symposions über Sprachkontakt in Europa, ed. P. Sture Ureland, Mannheim, 1977 — . They contain the papers read at the 1977 Symposion (see Kloss 1978 b) and so on. Only the relevant symposia are mentioned in Appendix I. Trends in Romance linguistics and philology, eds. Rebecca Posner — John N. Green, The Hague: Mouton Publishers, 1 9 8 0 - . See Lavandera 1981; Rogers 1981 b. Fodor-Hagege (eds.), I — III (1983-1984); I V - V (1989-1990). Holtus et al. (eds.), 1 9 8 8 - . See Krefeld 1988: Muljacic 1988 b.

Appendix III: Chronological list of relevant

terms

abstand languages 1952/1967: Kloss 1967. ausbau languages 1952/1967: Kloss 1967. languages in contact: Weinreich 1953. language loyalty. Weinreich 1953. diasystem: Weinreich 1954. diglossia: Ferguson 1959. It was brought up to date in: Marcellesi 1981.

Standardization in Romance

105

language planning: Haugen 1959. language conflict: Aracil 1965; Ninyoles 1969. national sociolinguistic profile formulas: Ferguson 1966. apparently dialectized abstand languages: Kloss 1967. status planning: Kloss 1969. corpus planning: Kloss 1969. roofing language. Its basis Dachsprache was coined, in all probability, by H. Goebl in 1975. See Muljacic 1989 d: 263. language disloyalty: Salvador 1983. language ecology: Haarmann 1983. heterogeneous dialects (dialects through subordination): Montes Giraldo 1984. focus(s)ing: Muljacic 1985 a. apparently lingualized ausbau languages: Muljacic 1985 a. participating function: Garvin 1986. prestige planning: Haarmann 1986. national language variant (variety): Svejcer—Nikol'skij 1986 (in Russian: 1978!). satellization: Muljacic 1988 b: 291. 'paradox of elaboration': Muljacic 1989 a. 'paradox of distance': Muljacic 1989 a: 49. transsatellization: see section 5.1 above. relativistic model of linguistics: see section 5.4 above. universals of elaboration: see section 5.0 above. homogenized repertories: see section 5.4 above.

References Allen, V. 1969

"A second dialect is not a foreign language", in: James E. Alatis (ed.), Linguistics and the teaching of standard English to speakers of other languages and dialects (Washington DC: Georgetown University Press), 189—195. Ammon, Ulrich — Norbert Dittmar — Klaus J. Mattheier (eds.) 1987 — 88 Sociolinguistics: an international handbook of the science of language and society. Soziolinguistik: ein internationales Handbuch zur Wissenschaft von Sprache und Gesellschaft. 2 vols. Berlin —New York: Walter de Gruyter. Ammon, Ulrich (ed.) 1989 Status and function of languages and language varieties. (Foundations of Communication 1.) Berlin —New York: Walter de Gruyter. Aracil, Lluis V. 1965 Conflit linguistique et normalisation linguistique dans ΓEurope nouvelle. Nancy. Balibar, Etienne — Immanuel Wallerstein (eds.) 1988 Race, nation, classe: les identites ambigues. Paris: Editions La Decouverte. Bartsch, Renate 1987 Norms of language. Theoretical and practical aspects. London —New York: Longman. Baum, Richard 1984 "Defense et illustration de la grammaire traditionnelle", Travaux de linguistique et de litterature 13.1: 7—17. 1987 Hochsprache. Literatursprache, Schriftsprache. Materialien zur Charakteristik von Kultursprachen. (Impulse der Forschung 49.) Darmstadt: Wissenschaftliche Buchgesellschaft.

106

Zarko Muljacic

Berruto, Gaetano 1974 La sociolinguistica. Bologna: Zanichelli. Besch, Werner 1988 "Standardisierungsprozesse im deutschen Sprachraum", Sociolinguistica

2:

186-208.

Bierbach, Christine — Irmela Neu-Altenheimer 1982 "Table ronde: les methodes d'enquete", Cahiers de linguistique sociale (Situations de diglossie. Actes du Colloque de Montpellier) 4/5: 107 — 155. Bochmann, Klaus 1985 a "Quel modele de developpement lexical pour les langues romanes en voie d'emancipation?", Linguistique comparee et typologie des langues romanes. Actes du XVHeme congres international de linguistique et philologie romanes (Aix-en-Provence, 29 aoät—3 septembre 1983) (Aix-en-Provence — Marseille: Universite de Provence — Jeanne Laffitte), 2: 191—201. 1985 b "Zum theoretischen Status und ideologischen Wert des VariationsbegrifTes in der Romanistik", Linguistische Arbeitsberichte (Sektion Theoretische und angewandte Sprachwissenschaft. Karl-Marx Universität Leipzig) 51: 2 — 17. 1989 Regional- und Nationalitätensprachen in Frankreich, Italien und Spanien. Leipzig: VEB Verlag Enzyklopädie. Boyer, Henri 1986 "Diglossie: un concept ä l'epreuve du terrain; l'elaboration d'une sociolinguistique du conflit en domaines Catalan et occitan", Lengas 20: 21—54. Brozovic, Dalibor 1970 Standardni jezik. Teorija. Usporedbe. Geneza. Povijest. Suvremena zbilja. [Standard language. Theory. Comparisons. Genesis. History. Contemporary Reality.] Zagreb: Matica Hrvatska. Burdeau, Georges et al. 1985 "Nation" (A. L'idee de nation, B. La construction nationale, C. Nation et Ideologie, D. Le nationalisme), Encyclopaedia universalis (Paris), 14: 932-947. CIRB - ICRB 1986 Actes du colloque international sur I'amenagement linguistique. Proceedings of the international colloquium on language planning. Ottawa 25—29 mai 1986/ May 25 — 29, 1986, presentation de Lome Laforge. Quebec: Les Presses de l'Universite Laval. Clyne, Michael 1984 Language and society in the German-speaking countries. Cambridge: Cambridge University Press. Corbeil, Jean-Claude 1987 "Vers un amenagement linguistique compare", in: Jacques Maurais (ed.), 553-566. Coulmas, Florian 1985 Sprache und Staat. Studien zur Sprachplanung. (Sammlung Göschen 2501.) Berlin —New York: Walter de Gruyter. Dahmen, Wolfgang et al. (eds.) 1990 Zum Stand der Kodifizierung romanischer Kleinsprachen. Romanistisches Kolloquium V. (Tübinger Beiträge zur Linguistik 348.) Tübingen: Gunter Narr Verlag.

Standardization in Romance

107

Danes, Frantisek 1988 a "Herausbildung und Reform von Standardsprachen", in: Ulrich Ammon et al. (eds.), 1: 1506-1516. 1988 b "Sprachkultur", in: Ulrich Ammon et al. (eds.), 2: 1687-1703. Daoust, Denise — Jacques Maurais 1987 "L'amenagement linguistique", in: Jacques Maurais (ed.), 5 — 46. Dittmar, Norbert — Brigitte Schlieben-Lange 1982 Die Soziolinguistik in romanischsprachigen Ländern. La sociolinguistique dans les pays de langue romane. (Tübinger Beiträge zur Linguistik 150.) Tübingen: Gunter Narr. Domasnev, Aleksandr I. 1978 "O granicax literaturnogo i nacional'nogo jazyka" ["On the boundaries of the literary and national language"], Voprosy Jazykoznanija 2: 3 — 16. Ferguson, Charles A. 1959 "Diglossia", Word 15: 325-340. 1962 "The language factor in national development", Anthropological linguistics 4.1: 2 3 - 2 7 . 1966 "National sociolinguistic profile formulas", in: William Bright (ed.), Sociolinguistics. Proceedings of UCLA sociolinguistics conference, 1964 (The Hague: Mouton), 531-545. Fishman, Joshua A. 1972 The sociology of language. An interdisciplinary social science approach to language in society. Rowley, Mass.: Newbury House. 1987 "Research on national languages", in: Ulrich Ammon et al. (eds.), 1: 638-646. Fodor, Istvän — Claude Hagege (eds.) 1983 —1984 Language reform. History and future — La reforme des langues. Histoire et avenir — Sprachreform. Geschichte und Zukunft. With a preface by Joshua A. Fishman. 3 vols. Hamburg: Helmut Buske. 1989-1990 Idem, volumes I V - V . Hamburg: Helmut Buske. Fontan, Francois 1961 Ethnisme, vers un nationalisme humaniste. Cuneo: A. Frassino. Gardy, Philippe 1981 "La diglossie comme conflit: l'exemple occitan", Langages 61: 75 — 91. Garvin, Paul L. 1983 "Le role des linguistes de l'Ecole de Prague dans le developpement de la norme linguistique tcheque", in: Edith Bedard — Jacques Maurais (eds.), La norme linguistique (Quebec—Paris: Gouvernement du Quebec / Conseil de la langue fran^aise — Le Robert), 141 — 152. 1986 "La langue standard — concepts et processus". Segon Congres internacional de la llengua catalana, Barcelona 1986. Unpublished typescript, 26 pp. Garvin, Paul — Madeleine Mathiot 1968 "The urbanization of the Guarani language: a problem in language and culture", in: Joshua A. Fishman (ed.), Readings in the sociology of language (2nd ed.) (The Hague —Paris: Mouton), 365 — 374. Geerts, Guido 1979 "History as language planner: the Netherlandic language area", Word 30: 58-75.

108

Zarko Muljacic

Goebl, Hans 1988

"Heur et malheur d'une langue romane nouveau-nee: le corse", in: Dieter Kremer (ed.), Actes du XVIIF congres international de linguistique et de Philologie romanes. Universite de Treves (Trier) (Tübingen: Max Niemeyer), 5: 287-295.

Görlach, Manfred 1988

"Sprachliche Standardisierungsprozesse im englischsprachigen Raum", Sociolinguistica 2: 131 — 185. Guespin, Louis 1988

"Langue et 'dialecte'. La position de Gramsci et les besoins glottopolitiques de la France de 1987", Beiträge zur romanischen Philologie 27: 219-228.

Haarmann, Harald 1975 Soziologie und Politik buch Verlag. 1980 1983 1986

1988

der Sprachen Europas. München: Deutscher Taschen-

Multilingualismus (1): Probleme der Systematik und Typologie. (Tübinger Beiträge zur Linguistik 116.1.) Tübingen: Gunter Narr. Elemente einer Soziologie der kleinen Sprachen Europas. Band 1: Materialien zur Sprachökologie. Hamburg: Helmut Buske. Language in ethnicity. A view of basic ecological relations. (Contributions to the sociology of language 44.) Berlin — N e w York — Amsterdam: Mouton de Gruyter. "Allgemeine Strukturen europäischer Standardsprachenentwicklung", Sociolinguistica, 2: 10 — 51.

Harris, Roy 1980

The language-makers. London: Duckworth / Ithaca ( N Y ) : Cornell University Press. Haugen, Einar 1959 1966 a 1966 b 1971

1983

"Language planning in modern N o r w a y " , Anthropological linguistics 1.3: 8-21. Language conflict and language planning: the case of modern Norwegian. Cambridge, Mass.: Harvard University Press. "Dialect, language, nation", Anthropological linguistics 6.8: 922 — 935. "Linguistics and language planning", in: William Bright (ed.), Sociolinguistics. Proceedings of the UCLA sociolinguistics conference 1964 (2nd ed.) (The Hague —Paris: Mouton), 50 — 67. "The implementation of corpus planning: theory and practice", in: Joan Cobarrubias and Joshua A . Fishman (eds.), Progress in language planning. International perspectives. (Contributions to the sociology of language 31.) (Berlin —New York —Amsterdam: Mouton), 269 — 289.

1987 "Language planning", in: Ulrich Ammon et al. (eds.), 1: 626 — 637. Heraud, Guy 1966 Popoli e lingue d'Europa. Milano: Ferro Edizioni. 1987 "Renaissances ethniques", $ociolinguistica 1: 30—45. Holtus, Günter — Michael Metzeltin — Christian Schmitt (eds.) 1988 Lexikon der romanistischen Linguistik: IV. Italienisch, Korsisch, Sardisch, Italiano, Corso, Sardo. Tübingen: Max Niemeyer. Jakobson, Roman 1931

"Prinzipien der historischen Phonologie", Travaux du cercle linguistique de Prague 4: 247-267.

Standardization in Romance 1962

109

"Retrospect", Phonological studies [ = R. Jakobson, Selected writings /] (The Hague: Mouton), 629-658. Jedlicka, Alois—Jan Chloupek 1988 "Sprachnormierung und Kodifizierung", in: Ulrich Ammon et al. (eds.), 1: 1650-1660. Joseph, John Earl 1980 "Linguistic classification in Italy: problems and predictions", Language problems and language planning 4: 131 — 140. 1982 "Dialect, language, and 'synecdoche'", Linguistics 20: 473 — 491. 1983 "Interferences of standardization on the transition from Middle to Modern French", Journal of historical linguistics and philology 1: 35 — 63. 1984 a "The engineering of a standard language", Multilingua 3: 87 — 92. 1984 b "The elaboration of an emerging standard", Language sciences 6: 39 — 52. 1985 "'Superposed' languages and standardization", Studi italiani di linguistica teorica ed applicata 14: 35 — 52. 1987 Eloquence and power. The rise of language standards and standard languages. London: Frances Pinter. 1989 "Standard languages and language standards". MS. To appear in: V. Prakasam (ed.), An encyclopaedic dictionary of linguistics terminology (Delhi: University Grant Commission). Kloss, Heinz 1929 "Sprachtabellen als Grundlage für Sprachstatistik, Sprachenkarten und für eine allgemeine Soziologie der Sprachgemeinschaften", Vierteljahrsschrift für Politik und Geschichte 1: 103-117. 1952 Die Entwicklung neuer germanischer Kultursprachen von 1800—1950. München: Pohl. 1967 "'Abstand languages' and 'ausbau languages'", Anthropological linguistics 9.7: 2 9 - 4 1 . 1969 Research possibilities on group bilingualism. Quebec: Centre international de recherche sur le bilinguisme. 1976 "Abstandsprachen und Ausbausprachen", in: Joachim Göschel — Norbert Nail — Gaston Van der Elst (eds.), Zur Theorie des Dialekts (Zeitschrift fur Dialektologie und Linguistik. Beihefte. Neue Folge 16) (Wiesbaden: Franz Steiner Verlag), 301 - 3 2 2 . 1978 a Die Entwicklung neuer germanischer Kultursprachen seit 1800. (Sprache der Gegenwart. Schriften des Instituts für deutsche Sprache 37.) Düsseldorf: Pädagogischer Verlag Schwann. 1978 b "Sprachkontakte in Europa", in: Per Sture Ureland (ed.), Sprachkontakte im Nordseegebiet. Akten des 1. Symposions über Sprachkontakt in Europa, Mannheim 1977 (Linguistische Arbeiten 66) (Tübingen: Max Niemeyer), 1 — 17. 1984 "Umriß eines Forschungsprogrammes zum Thema 'Sprachentod'", International journal of the sociology of language 45: 65 — 76. 1987 a "Nation", in: Ulrich Ammon et al. (eds.), 1: 102-108. 1987 b "Abstandsprache und Ausbausprache", in: Ulrich Ammon et al. (eds.), 1: 302-308. Kloss, Heinz — Harald Haarmann 1984 "Introduction — Introduction", in: Heinz Kloss — Grant D. McConnell (eds.), Linguistic composition of the nations of the world — Composition

110

Zarko Muljacic

linguistique des nations du monde, vol. 5: Europe and the USSR — L'Europe et l'URSS (Quebec: Les Presses de l'Universite Laval), 7 — 75. [H. Kloss was responsible for pp. 11—43 and 70 — 72.] Krefeld, Thomas 1988 "Periodisierung", in: Günter Holtus et al. (eds.), 748-762. Kremnitz, Georg 1979 Sprachen im Konflikt. Theorie und Praxis der katalanischen Soziolinguisten. Eine Textauswahl. (Tübinger Beiträge zur Linguistik 117.) Tübingen: Gunter Narr. 1981 "Du 'bilinguisme' au 'conflit' linguistique. Cheminement de termes et de concepts", Langages 61: 63 — 74. Lafont, Robert 1984 "Pour retrousser la diglossie", Lengas 15: 5 — 36. Lamuela, Xavier 1987 Catalä, occitä, friülä: llengües subordinades i planificacio lingüistica [Catalan, Occitan, Friulian: subordinated languages and language planning]. Barcelona: Jaume Vallcorba. Lavandera, Beatriz R. 1981 "Sociolinguistics", in: Rebecca Posner — John N. Green (eds.), 129 — 228. Loman, Bengt 1988 "Sprachliche Standardisierungsprozesse in Skandinavien", Sociolinguistica 2: 209-231. Lüdi, Georges 1989 "Situations diglossiques en Catalogne", in: Günter Holtus et al. (eds.), La Corona de Aragon y las lenguas romänicas. Miscelänea de homenaje para Germän Colon. La Corona d'Aragö i les llengües romäniques. Miscel.lania d'homenatge per α Germä Colon (Tübingen: Günter Narr), 237 — 265. Malkiel, Yakov 1984 "A linguist's view of the standardization of a dialect", in: Aldo Scaglione (ed.), 5 1 - 7 3 . Marcellesi, Jean-Baptiste 1985 "Der aktuelle Prozeß der Entstehung von Sprachen in der Romania", Linguistische Arbeitsberichte 51: 17 — 26. Marcellesi, Jean-Baptiste (ed.) 1981 "Bilinguisme et diglossie", Langages 61. 1986 "Glottopolitique", Langages 83. Marcellesi, Jean-Baptiste — Evelyne Bulot — Thierry Bulot 1988 "En quoi l'ecole sociolinguistique de Rouen est-elle gramscienne?", Beiträge zur romanischen Philologie 27: 213 — 217. Mattheier, Klaus J. 1988 a "Das Verhältnis von sozialem und sprachlichem Wandel", in: Ulrich Ammon et al. (eds.), 2: 1430-1452. 1988 b "Nationalsprachenentwicklung, Sprachenstandardisierung und Historische Soziolinguistik", Sociolinguistica 2: 1—9. Maurais, Jacques (ed.) 1987 Politique et amenagement linguistique. Quebec — Paris: Gouvernement du Quebec/Conseil de la langue franchise — Le Robert.

Standardization in Romance

111

Mioni, Alberto M. 1975 "Per una sociolinguistica italiana. Note di un non sociologo", in: Joshua A. Fishman, La sociologia del linguaggio (Roma: Officina Edizioni), 7 — 56. 1984 "Distribuzione geografica delle lingue e loro funzioni sociolinguistiche", Linguaggi 1.3: 29—47. 1988 "Standardization processes and linguistic repertoires in Africa and Europa: some comparative remarks", in: Peter Auer — Aldo di Luzio (eds.), Variation and convergence: studies in social dialectology (Sociolinguistics and language contact 4) ( B e r l i n - N e w York: Walter de Gruyter), 2 9 4 - 3 2 0 . Montes Giraldo, Jose Joaquin 1984 "Para una teoria dialectal del espanol", in: Homenaje a Luis Florez. Estudios de historia cultural, dialectologia, geograßa lingüistica, sociolingiiistica, fonetica, gramätica y lexicografia (Biblioteca de Publicaciones del Instituto Caro y Cuervo 69) (Bogota: Instituto Caro y Cuervo), 72 — 89. 1986 "Lengua — dialccto una vez mas. La persistencia de un viejo problema", Thesaurus 41: 1 - 1 9 . Muljacic, Zarko 1980 "Per una classificazione 'standardologica' delle lingue romanze", in: Herbert J. Izzo (ed.), Italic and Romance. Linguistic studies in honor of Ernst Pulgram (Current issues in linguistic theory 18) (Amsterdam: John Benjamins), 85 — 96. 1985 a "Come applicare il modello standardologico comparativo in diacronia? Sulla 'focalizzazione' nella storia linguistica d'ltalia", in: Luciano Agostiniani — Patrizia Beilud Maffei — Matilde Paoli (eds.), Linguistica storica e cambiamento Unguistico. Atti del XVI congresso internazionale di studi, Firenze 7 — 9 maggio 1982 (Pubblicazioni della societä di linguistica italiana 23) (Roma: Bulzoni), 3 9 7 - 4 0 9 . 1985 b "Romania, Germania e Slavia: parallelismi e differenze nella formazione delle lingue standard", in: Adriana Quattordio Moreschini (ed.), La formazione delle lingue letterarie. Atti del convegno della Societä Italiana di Glottologia (Biblioteca della SIG 8) (Pisa: Giardini), 3 9 - 5 5 . 1985 c "II fenomeno Überdachung 'tetto', 'copertura' nella sociolinguistica (con esempi romanzi)", Linguistica 24: 77 — 96. 1986 "L'enseignement de Heinz Kloss (modifications, implications, perspectives)". Langages 81: 53 — 63. 1987 "'Statalismes' semantiques dans la Romania", in: R . C r e s p o et al. (eds.). Studies in honour of Mario Alinei. Volume II. Theoretical and applied semantics (Amsterdam: Rodopi), 339 — 356. 1988a " ' L i n g u a ' e 'dialetto' rivisitati", in: Harald Thun (ed.), Energeia und Ergon. Sprachliche Variation — Sprachgeschichte — Sprachtypologie: Studio in honorem Eugenio Coseriu. Band II. Das sprachtheoretische Denken Eugenio Coserius in der Diskussion (1) (Tübinger Beiträge zur Linguistik 300) (Tübingen: Gunter Narr), 401 —409. 1988 b " N o r m a e standard", in: Günter Holtus et al. (eds.), 2 8 6 - 3 0 5 . 1988 c "Heinz Kloss und die Kreolsprachen", Neue Romania. Veröffentlichungsreihe des Studienbereiches Neue Romania des Instituts fur Romanische Philologie der Freien Universität Berlin 7: 43 — 56. 1988 d Review of Haarmann 1986, Archivio glottologico italiano 73.1: 58 — 61.

112

Zarko Muljacic

1989 a 1989 b

1989 c

1989 d 1989e

1990

1991

Munske, 1983

"Le 'paradoxe elaborationnel' et les deux especes de dialectes dans l'etude de la constitution des langues romanes", Lletres asturianes 31: 43 — 56. "The emergence of the Florentine > Italian language", in: Thomas J. Walsh (ed.), Synchronic and diachronic approaches to linguistic variation and change (Georgetown University round table on languages and linguistics 1988) (Washington DC: Georgetown University Press), 221 —226. "Hanno i singoli diasistemi romanzi 'emanato' la 'loro' lingua standard (come di solito si legge) ο hanno invece le lingue standard romanze determinato (in larga misura) a posteriori i 'loro' dialetti?", in: Fabio Foresti et al. (eds.), L'italiano tra le lingue romanze. Atti del XX congresso internazionale di studi della SLI, Bologna, 25 - 27 settembre 1986 (Roma: Bulzoni), 9 - 2 5 . "Über den Begriff Dachsprache", in: Ulrich Ammon (ed.), 256 — 277. "Ν. S. Troubetzkoy, precurseur de la linguistique de contacts", in: Carlos Martin Vide (ed.), Adas del IV congreso de lenguajes naturales y lenguajes formales (Barcelona: Universität de Barcelona), 963 — 980. "II caso italiano vs altri casi europei. Prolegomeni alia standardologia storica comparata", in: Emanuele Banfi — Patrizia Cordin (eds.), SLI, Storia dell'italiano e forme dell'italianizzazione. Atti del XXIII congresso internazionale di studi, Trento — Rovereto, 18—20 maggio 1989 (Roma: Bulzoni), 9 — 20. "Per un approccio relativistico al rapporto: lingua nazionale — dialetti", in: Vocabolario dei dialetti della Svizzera italiana. [Atti del] XVIII convegno di studi dialettali italiani "Fra dialetto e lingua nazionale: realtä e prospettive" [Lugano, 11 — 15 ottobre 1988, a cura di Giovan Battista Pellegrini] (Padova: Unipress), 247-259. Horst Haider

"Umgangssprache als Sprachenkontakterscheinung", in: Werner Besch et al. (eds.), Dialektologie. Ein Handbuch zur deutschen und allgemeinen Dialektforschung (Berlin-New York: Walter de Gruyter), 2: 1002-1018. Ninyoles, Rafael Lluis 1969 Conßicte linguistic Valencia. Valencia: Tres i Quatre. Picchio, Riccardo — Harvey Goldblatt (eds.) 1984 Aspects of the Slavic language question. Vol. I. Church Slavonic — South Slavic — West Slavic; Vol. II. East Slavic (Yale Russian and East European publications 4 a—4 b). New Haven — Yale Concilium on International and Area Studies. Pintelon, P. 1973 "Les caracteres generaux d'une langue commune", V°me congres international des linguistes, Bruxelles, 28 aoüt—2 septembre 1939. Rapports, resumes, Dokumente, reponses (Nendeln — Liechtenstein: Kraus Reprint), Risumes: 40-56. Pisani, Vittore 1970 "Si puo parlare di unitä ladina?", in: Atti del congresso internazionale di linguistica e tradizioni popolari, Gorizia — Udine — Tolmezzo 1969 (Udine: Societä Filologica), 53 — 64. Porru, Giulia 1939 "Anmerkungen über die Phonologie des Italienischen", Travaux du cercle linguistique de Prague 8: 187 — 208.

Standardization in Romance

113

Posner, Rebecca — John N. Green (eds.) 1981 Trends in Romance linguistics and philology. Volume 2: Synchronic Romance linguistics. The Hague — Paris — New York: Mouton. Price, Glanville 1976 "Language standardization in the Romance field: a survey of recent work", Semasia 3: 7 — 32. Proces-verbaux 1931 "Proces-verbaux des seances du 18 au 21 decembre 1930", Travaux du cercle linguistique de Prague 4: 289 — 305. Ray, Punya Sloka 1963 Language standardization. The Hague: Mouton. Rogers, Kenneth H. 1981 a "Selected recent studies in linguistic nationalism in the Romance languages", Canadian review of studies in nationalism — Revue canadienne des etudes sur le nationalisme 8.2: 267 — 283. 1981 b "Studies on linguistic nationalism in the Romance languages", in: Rebecca Posner — John N. Green (eds.), 229 — 256. Rubin, Joan 1979 Directory of language planning organizations. Honolulu: East-West Center. Salvador, Gregorio 1983 "Sobre la deslealtad lingüistica", Lingiiistica espafiola actual 5: 173 — 178. Saussure, Ferdinand de 1983 Course in general linguistics, translated and annotated by Roy Harris. London: Duckworth. Scaglione, Aldo 1984 a "The rise of national languages: East and West", in: Aldo Scaglione (ed.), 9-49. Scaglione, Aldo (ed.) 1984 b The emergence of national languages. Essays by M. L. Baeumer — T. Cable - H. Goldblatt - Κ. Η. Karpat - Y. Malkiel - Η. Paolucci - Α. Scaglione. Ravenna: Longo Editore. Schmid, Heinrich 1982 Richtlinien für die Gestaltung einer gesamtbündnerromanischen Schriftsprache Rumänisch Grischun. Cuira: Lia Rumantscha. Schmitt, Christian 1988 "Typen der Ausbildung und Durchsetzung von Nationalsprachen in der Romania", Sociolinguistica 2: 73 — 116. Schmitt-Brandt, Robert 1985 "Entstehung der lateinischen Schriftsprache", in: Per Sture Ureland (ed.), Entstehung von Sprachen und Völkern. Glotto- und ethnogenetische Aspekte europäischer Sprachen. Akten des 6. Symposions über Sprachkontakt in Europa, Mannheim 1984 (Linguistiche Arbeiten 162) (Tübingen: Max Niemeyer), 121-132. Siegel, Jeff 1985 "Koines and koineization", Language in society 14: 357 — 378.

114

Zarko Muljacic

Sjoestedt-Jonval, M. L. 1973 "Langue commune, langue dominante, langue de culture", Vime congres international des linguistes, Bruxelles 28 aoüt—2 septembre 1939. Rapports, resumes, Dokumente, reponses (Nendeln — Liechtenstein: Kraus Reprint), Reponses 15-16. Srpovä, Milena 1988 "La linguistique tcheque dans les annees 80", La linguistique 24.1: 113 — 126. Stewart, William A. 1968 "A sociolinguistic typology for describing national multilingualism", in: Joshua A. Fishman (ed.), Readings in the sociology of language (2nd ed.) (The Hague: Mouton), 531—545. Svejcer, Aleksandr Davidovic—Leonid Borisovic Nikol'skij 1986 Introduction to sociolinguistics. (Linguistic and Literary Studies in Eastern Europe 14.) Amsterdam —Philadelphia: John Benjamins. Theses 1929 "Theses presentees au premier congres des philologues slaves", Travaux du cercle linguistique de Prague 1: 5 — 29. Trubetzkoy, Nikolaj Sergeevic 1939 "Gedanken über das Indogermanenproblem", Acta linguistica (hafniensia) 1: 81-88. Vachek, Josef 1972 "The linguistic theory of the Prague school", in: Vilem Fried (ed.), The Prague school of linguistics and language teaching (London: Oxford University Press), 11-27. 1976 "Sociolinguistics seen from the Prague point-of-view", in: Wolfgang Viereck (ed.), Sprachliches Handeln — soziales Verhalten. Ein Reader zur Pragmalinguistik und Soziolinguistik (München: Wilhelm Fink), 13 — 27, 551. 1983 a "Prague linguistic circle: theses presented to the first congress of Slavists held in Prague in 1929", in: Josef Vachek — Libuse Duskova (eds.), Praguiana. Some basic and less known aspects of the Prague linguistic school (Amsterdam: John Benjamins), 77 — 120. 1983 b "The Czech editor's postscript", ib., 2 7 5 - 3 0 2 . Vaz de Soto, Jose Maria 1981 Defensa del habla andaluza. Sevilla: Edisur. Weinreich, Uriel 1953 Languages in contact: findings and problems. (Publications of the Linguistic Circle of New York 1.) New York: Linguistic Circle of New York. 1954 "Is a structural dialectology possible?", Word 10: 3 8 8 - 4 0 0 . Willemyns, Roland — Piet Van de Craen 1988 "Growth and development of standard Dutch in Belgium", Sociolinguistica 2: 1 1 7 - 1 3 0 .

Romance : Romance bilingualism in Europe

HENRIQUE MONTEAGUDO and ANTON SANTAMARINA

Galician and Castilian in contact: historical, social, and linguistic aspects

1. Introduction The last 10 — 15 years have been so important for the future of Galician that, we can be sure, when they are examined 50 years from now, they will be regarded as the decisive period in the struggle between its survival and its disappearance. The 1970s started with the worst forecasts for the future of Galician. Those in power, bent on annihilating minority languages within Spain, for the first time in history had the means to succeed: an educational system so ingrained in society that its power of penetration was not recognized; new mass-media (radio and television) with wider circulation and enormous influence on the spoken language; and stringent control over both education and the media. From the 19th century, the middle classes had opted for Castilian as the language for personal advancement and social prestige. The Galician cities, which were centres of linguistic assimilation, grew steadily, and Galician-speaking immigrants settling there stopped speaking Galician to their children, who were brought up in a strongly castilianized environment. The final refuge for the language, the world of peasants and fishermen (which Galicianism had mythicized as their Arcadia), was threatened with destruction: emigration bled the active population (in the Galician hamlets all that remained were a large number of old people and a few children), and the school, radio, and television became the means of acculturation of this bastion of traditional culture. As the linguistic loyalty of this section of the population began to dwindle, so desertions towards Castilian started (Alonso 1973). The only resistance in favour of Galician was from a small group of intellectuals, armed with sporadic contributions to the daily press, a prestigious quarterly magazine (Grial), a few other publications and a publishing house (Galaxia). In the University of Santiago, the centre of

118

Henrique Monteagudo and Anton Santamarina

education for the professional classes, a tiny minority of teachers and students struggled against the current. Through the personal initiative of a philanthropic linguist (Constantino Garcia) and thanks to the work of some enthusiastic students, the first Chair of Galician Language and Literature was founded, together with an Instituto da Lingua Galega (ILG). The Chair was devoted to university teaching, whilst the Institute worked in the hitherto untilled field of basic research on Galician. 1 Meanwhile, clandestine nationalism was gaining popularity, with the first associations of Galicianist workers; and the regime tolerated cultural organizations with Galicianist tendencies and activities. They succeeded in popularizing slogans in favour of Galician, in which the 'Nova Cancion Galega' and subsequent movements were very important. Galician lost a substantial number of speakers to Castilian, but at least it began to gain strength against assimilation through new publicity, ranging from publication of Galician materials to car stickers. From 1975 on, progress was extremely rapid, with the end of the dictatorship in 1976, the promulgation of the democratic Constitution in 1978, and the proclamation of the autonomous Statute of Galicia in 1981. These political changes were preceded, accompanied and followed by important social and cultural developments, of which the beginning of the process of social spread of Galician is merely a symptom. However, in this article we concern ourselves only with the linguistic situation and with language planning in Galicia. During the last 15 years, more bibliographies on the different linguistic and sociolinguistic aspects of Galician have been produced than ever before. In Galicia itself, since the creation of University courses specializing in Galician-Portuguese philology, hundreds of graduates now teach Galician professionally, and dozens are engaged on research into the language in the ILG. Knowledge of Galician within the country, but above all abroad, was partial and inaccurate, owing to the marginal position of the language plus the lack of tradition in its cultivation and investigation. Scholars of Romance languages had access to very little information and virtually no first-hand knowledge, which explains why so little attention was paid to Galician and the unreliability of their judgements on the subject. Indeed, it is only recently that publication of such research has begun (Fernandez Rey 1988). For this reason we shall begin with a brief historical survey, before dealing with the most salient aspects of the present situation in the sociolinguistic and linguistic fields.

Galician and Castilian in contact

119

2. Historical roots of the contact between Galician and Castilian 2.1 Galician as the original language of Galicia We assume basic knowledge of the process which began with the latinization of the northwest of the Iberian Peninsula, and which led, through the development of spoken Latin in this area (which Romans called Gallaecia), to the emergence of a new linguistic variety, 'Galician'. The most characteristic and decisive linguistic changes in this variety were completed and consolidated when, for well-known historical reasons, the speakers spread this new-born language south of its original territory between the 11th and 13th centuries. Until the 13th century, there was no written system sufficiently distinguishable from Latin to enable this variety to be consistently recorded. From the 13th to the 16th century, we find the first attempts to establish a written variety of the Galician language. This written language was to become instrumental in the regulation of social life and in satisfying the cultural demands of the dominant classes of two emerging nationalities: Galicia and Portugal. The birthplace of the Galician language was politically fragmented in the years before these attempts at early 'linguistic promotion': the beginnings of this political division seem to be rooted at the very end of the 11th century, when the county of Galicia split with the county of Portugal. The 12th century saw the transformation of the counties of Galicia and Portugal into two separate kingdoms. 2 The 13th century witnessed the union of the kingdoms of Galicia and Castile and the consolidation of the autonomous kingdom of Portugal; the 14th century, the frustration of attempts to separate Galicia from the kingdom of Castile and the failure of attempts to unify Galicia with Portugal; the 15th century, the final submission of the Galicians and the beginnings of their political dependence, together with the complete confinement of Portugal along its peninsula border, which led Portugal to embark on its celebrated voyages of discovery. The lasting political fragmentation of the Galician-Portuguese linguistic domain, with the resulting rupture of social interchange and of cultural contact between the communities (which had characterized the relationship between the two countries during the period of the formation of the language), had inevitable linguistic consequences. It is certain that until the 16th century the independent evolution of varieties of Galician and Portuguese did not produce a great fissure in their unity (Maia 1986), but the continuation of the isolation for another five centuries widened the gap between them to such an extent that it now seems appropriate

120

Henrique Monteagudo and Anton Santamarina

to regard Galician and Portuguese as two separate languages (see Teyssier 1980: 4 8 - 4 9 , and sections 6.5.1 and 6.5.2 below). Essentially, there were six areas of cultivation of written Galician (or Galician-Portuguese) language in the medieval period in Galicia: (1) literature and history, the most outstanding aspect being the brilliant school of lyric poetry which was active certainly during the 13th century and the first quarter of the 14th; note also the growth of prose, both narrative and historical (as far as we know, it lasted from the 13th to the middle of the 15th century), generally translated from Castilian or Latin; (2) administration: regulations for guilds and brotherhoods, acts and agreements of the councils; (3) the economy: private contracts, wills, account books for the different institutions; (4) the administration of justice: requests and pleas, lawsuits; legal treaties (translated from Castilian and Latin); (5) science: all that remains is a veterinary treatise from the 15th century; and (6) religious life: lessons and sermons. In fields (2), (3), (4), and (6), we know that the vitality of Galician lasts from the beginnings of the 13th century until the middle of the 16th. From our point of view, they have a special relevance, because cultured literature reached only a small section of the public, and had a less important social function than the regulation of socio-economic activities. We should point out that in the period in which Galician was cultivated for these purposes, owing to the lack of a political or linguistic centre, there was no development of a stable written variant, although the parameters of variation of Galician were not much wider than those of any other Romance variety in that era. Nor did a common spoken variety which could be recognized as standard become established. Examination of the copious documentation of medieval Galicia which has been preserved shows us that in writings up to the end of the 14th century the only languages current were Latin and Galician. On the other hand, personal records and the documentation produced by the urban organizations (councils, brotherhoods, chapters, and so on) are practically unanimous in their use of Galician between the 13th and 15th centuries, especially so during the 14th. However, the Royal Court of Castile almost from the reign of Alfonso X (1252 — 1284) onwards used Castilian with its Galician subjects, which forced Galicians to use that language in their relations with the Court (Monteagudo 1985). In any case, we can say without exaggeration that in this first period, as far as documents for internal circulation were concerned, the most common practice was the use of Galician, with Castilian serving only for communication with the

Galician and Castilian in contact

121

Royal Court, and limiting Latin to ever diminishing roles. As a result, Galician was on the road to becoming a national language before the intervention of Castilian. 2.2 The introduction of Castilian into Galicia: the Dark Ages The political developments that took place between the middle of the 14th century and the end of the 15th were the most decisive factors in the introduction of Castilian and the loss of ground of Galician as a literary and cultural language. Firstly, between about 1335 and 1385, came the complete destruction of the Hispanic Burgundian dynasty previously supported by the Galician nobility and its replacement by the triumphant dynasty of Trastamara. The new rulers rewarded Leonese and Castilian supporters with endowments of land and sinecures in Galicia. Moves towards separation or incorporation into independent Portugal failed. Secondly, throughout the 15th century social revolts brought into conflict different sectors of the popular classes with diverse noble factions, or even clashes of various noble factions fighting among themselves. The highpoint came between about 1465 and 1475, when a general revolution erupted in Galicia (the 'Guerra Irmandina'), in which the bourgeoisie led the peasants against the nobility, which eventually triumphed. Dynastic struggles also weakened Galician society, which was at the mercy of the central powers of the crown of Castile. The events of these two historical periods led to increasing centralization of the state. In this, the absolute power of the crown and the submission of the nobility to it played a crucial role. These trends intensified between the 16th and 18th centuries, the era of the Antiguo Regimen. The submission of the leading classes of Galicia to the central power and their linguistic and cultural assimilation came about in two ways, firstly through the massive introduction of nobles from outside Galicia who occupied the highest posts in the hierarchy and formed the first castilianizing nucleus in Galician society; and secondly through the assimilation of the Galician nobility, who educated their children at the Castilian court and who became more concerned with gaining court sinecures, and less with the management of their landed properties, which they had to abandon to fight abroad or serve at court. This highest sector of nobility of Galician origin became the first castilianized social nucleus of Galicia (see Chacon 1979). It can be shown that the use of Castilian in documents increased in the second half of the 14th century, became even stronger in the 15th, and eventually replaced Galician in the 16th century. The process of

122

Henrique Monteagudo and Anton Santamarina

castilianization began in the higher echelons of the nobility (the archbishop of Santiago, other bishops, high ranking dignitaries selected by the king) and filtered downwards through the social pyramid. Documents clearly show that the process of linguistic substitution followed the pattern of political and military occupation in Galicia and the decapitation of its society (see Vazquez 1980: 707; Villares 1985: 93). The marginalization of Galician is attested in 1550 by a foreign visitor to Compostela (Alonso 1974: 305) and in 1597, when the chapter of Santiago cathedral complained that "in this archbishopric, as in the bishopric of Ourense, ... they gave the best sinecures not only to those from foreign kingdoms but those who penetibus ignored Galician" (Lopez 1989: 9). Galician stopped being used as a written form soon after the introduction of printing in Galicia and the foundation of the first and so far the only Galician university. The consequences are clear. Galician lost its written uses and stopped being cultivated, which meant: that the language was deprived of its formal registers; that the process of formation of a common variety did not continue, rather, the dialects grew freely and became more and more diverse; that Galician failed to achieve the linguistic enrichment brought by later advances of civilization; and thus that Galician remained defenceless to the penetration of Castilian. In 1607, the Portuguese linguist Duarte Nunes de Liäo expressed this magnificently, commenting: "when compared with Galician, the Portuguese language progressed as much in its profusion as in its elegance, as we can see. This was caused because Portugal had kings and courts, which are the office where the vocabulary is forged and refined, and the source for the rest of the citizens, something that never took place in Galicia" (Buescu 1983: 220; Monteagudo 1988). Between the 16th and the 18th centuries the castilianizing of Galician society was superficial, as only small sections (still very significant) of the population had access to the written language, and the petty nobility, still dependent on the soil, kept Galician as an everyday language. Therefore, we postulate that in this period the social direction of castilianization must have been selective and horizontal. Castilian was adopted only in the most privileged sections, those closest to the apparatus of church and state. A linguistic split was produced along class lines, so that Castilian remained for the dominant classes the symbol of their social status (compare Ninyoles 1978: 41—66 on Valencian). Galician was excluded from the most prestigious social areas, the highest themes, and the most formal functions, and began to be associated with the lower classes and with informal and uncultured areas and communicative sit-

Galician and Castilian in contact

123

uations. Its prestige was weakened at the same pace as the increase of self-hate among Galicians. During the 18th century a new factor appears which reinforces these negative connotations associated with Galician in the social consciousness: extension of schooling, even though it still only reached small sections of society. A royal decree of 1768 made the use of Castilian obligatory as the vehicular language. We have records of the complaints of Father Martin Sarmiento (1695 — 1772) about the marginalization of Galician in schools and his wish "... that the Galicians will get rid of the hatred they feel for their own language" (Alonso 1973: 64). 2.3 From the 19th century to the present day 2.3.1 The social evolution of Galicia and emigration Attempts to modernize Galicia are even today not wholly successful: political democratization has had to face up to powerful traditional forces and is only very recently consolidating its position in Spain and Galicia; the substitution of a precapitalist economy based on agriculture for an economy founded on industry and trade is still far from complete in Galicia; Galicia remains underdeveloped with respect to proletarianization of the peasants, concentration of the population in urban areas, and the spread and improvement of communications; primary education only became widely available in the 60s and 70s, whilst access to higher and further education was restricted until quite recently. We should note however the advance of the mass media: newspapers, which had a considerable circulation during the 19th century; radio, from the 1930s, and television, from the 1960s. During this era the centralized state grew steadily stronger despite resistance. Galicia remained in a situation of dependence, exaggerated by the absence of a bourgoisie with specific Galician interests, since a major sector of the industrial and commercial bourgeoisie during the 18th and 19th centuries came from outside Galicia, especially from Catalonia and the centre of the Peninsula. Probably Galician survived only because of the failure of the central state to modernize the country. Around the weak industrial, commercial, and service centres created in Galicia, small cities developed, where sectors of the Galician-speaking population came into contact with Castilian. Even more relevant to our argument was emigration abroad. Already during the 17th century, many Galicians occupied lowly positions in Castilian cities. For the 18th century, we have evidence of the sheer volume of Galician emigration to Portugal and other areas of the Peninsula: some 500,000 people had to find work

124

Henrique Monteagudo and Anton Santamarina

outside Galicia, out of a population of hardly more than a million towards the end of that century. But, from the middle of the century to the present day, emigration reached exodus proportions: between 1860 and 1910 some 500,000 Galicians emigrated to South America, most of them as permanent settlers; 3 a similar number between 1915 and 1930; and roughly 350,000 between 1946 and 1965. More recently, the main destination for emigration has been industrialized Europe, where between 1960 and 1977 more than 200,000 people went to work (Sixirei 1988). The population of Galicia thus declined in absolute terms between 1950 and 1970 (see Santamarina — Fernandez Rey 1992). The present demographic structure of Galicia can be seen from the following figures. The total number of inhabitants is almost 3,000,000 people, 7.4% of the population of Spain. Population growth has been extremely slow during this century (for example, in 1919 Galicia had almost 2,000,000 inhabitants, 10.6% of the Spanish total). The Galician population is very widely dispersed in small rural nuclei with no fewer than 31,894 population units; 42% of Galicians live in units of less than 10,000, and only 33% in units of more than 50,000 inhabitants. The economic activity which employs the largest section of the population is agriculture (41 % of the active population, compared with 16% in industry and 30% in the tertiary sector). These figures give a general idea of the socio-economic structure in Galicia: dispersed population, little urbanization, underdevelopment, and, in consequence, reliance on the primary sector and strong external emigration. 2.3.2

Sociolinguistic

evolution

The public use of Galician gained momentum only at the beginning of the 19th century. The first opportunity was offered by the Napoleonic invasion, which produced a rupture in the central state. The main resistance came from a popular army of volunteers (the 'guerrilla'), which was organised independently in Galicia. The first important and widely circulated printed matter in Galician was related to the formation of this popular army: appeals for volunteers and reports of their prowess. Until the middle of the century, Galician was used in written form in a discontinuous and sporadic way, basically in pamphlets and polemic writings, usually published in the burgeoning press. For the most part, Galician was used by defenders of liberal and progressive stances, although it was occasionally adopted by conservative publicists (see Marino 1990).

Galician and Castilian in contact

125

Nevertheless, it was not long before Galician was cultivated in the privileged field of poetry. The second half of the 19th century saw a renaissance of the Galician language (the 'Rexurdimento'), like that of Provengal and Catalan, supported by acculturated elites who identified themselves with their country. This renaissance had its origins in literary works, especially after 1863, with the publication of Cantares gallegos (1863) by Rosalia de Castro (1837-1885). M. Curros Enriquez (1851 1908), E. Pondal (1835-1917), and Rosalia together represented the 'three crowns' of Galician literature. Soon afterwards scientific philological works appeared, like those of J. A. Saco y Arce (1868) — the only reference book for Romanists before the 1970s — and of M. Valladares Nünez (1884). Historians like M. Murguia or A. Lopez Ferreiro searched for the historic roots of Galicia. The foundation of the Academy in 1905 had been long awaited by the many who wanted to put an end to linguistic anarchy (Alonso 1988). The marginalization of Galician for centuries fostered a false impression amongst its speakers of speaking a corrupt language. Because Galician is very close to the dominant language, it suffers from "very strong cultural pressure: heavy structural borrowing" (the fifth degree on the borrowing scale of Thomason — Kaufman 1988: 74 and 100; see section 5.2.2 below). The renaissance writers used a language spoken mainly by the illiterate, with the mission of sweeping away the prejudices of a population which grossly undervalued its own language. This, amongst other reasons, justifies periodic publications promoting the defence and illustration of the Galician language. Eventually, this led to the birth of the Academia Galega, whose emblem bears the magnificent inscription 'Colligit, expurgat, innovat', but the wishes of its founders were only partially fulfilled, as the institution did not even produce a dictionary or a normative grammar. Beyond these elite groups, the presence of Galician in non-rural areas was becoming rare. The castilianizing action of the church and state was felt with more and more impact, so that the 'degaleguization' of the new urban classes, especially the emerging middle class, and the educated sector, continued to increase. Castilian added to its exclusive presence in the traditional fields, an all-powerful hegemony in the new social and cultural areas. 4 With the introduction of new methods of education and communication, and with the relative social mobility induced by the capitalist economy, Castilian now expands vertically, and spontaneously. As Castilian is essential for social advancement, as soon as the population gains the means to learn it, individuals and aspiring social groups will

126

Henrique Monteagudo and Anton Santamarina

adopt it for themselves and for their families. Thus, there is massive language shift towards Castilian alongside cultivation and vindication of Galician by the enlightened minority. Only after 1916, with the formation of the 'Irmandades da Fala' ('Brotherhoods of Speech'), did Galician begin to transcend the limits of literature. By this time, the first steps towards political proselytism were taken. The opening essays and works on local scientific research were written in Galician, and it was also in this period that there appeared regular publications with more ambitious scope than in the past. The 15 years before the Spanish Civil War (1936 — 1939) saw the appearance of an important new generation of writers: artists, scientists and politicians like A. D. R. Castelao (1886-1950), V. Risco (1884-1963), R.Otero Pedrayo (1887 — 1976), and others. For the first time, a united group of highly accomplished intellectuals (known collectively as the 'Xeracion Nos') 5 put Galicia to the forefront of their tasks. They were responsible for putting Galicia on the road to recovery. From the linguistic point of view, it is an extremely important stage in the history of Galician, as they made great progress in the elaboration (Ausbau) of Galician, with regard to modernization of terminology and stylistic development (see Haugen 1987: 61 —62). For the first time, Galicia achieved legal autonomy, which recognized co-officiality of the Galician language (1936). However, none of this group had professional linguistic training, so language was not standardized during this period. The work of this generation and of its successor (the parauniversity group Seminario de Estudos Galegos), was frustrated by Franco's triumph, which annihilated Galicianism, forcing its protagonists into exile or silence. Through censorship, Galician became a forbidden language outside the family domain (de facto, as dictatorships do not normally outlaw languages de jure). The coercive castilianizing nature of the dictatorship, in an era in which emigration of Galician speakers increased apace, schooling of the infant population became more widespread and a new powerful mass-medium — television — appeared, placed Galician in a anguished situation, which democracy has inherited. We shall now examine the main features of the resultant picture. 3. Societal

bilingualism

States which deny the rights of a minority begin by not acknowledging its existence. Thus, the official population censuses carried out in Spain never collected data about Galician. This explains why all the quantitative data we shall cite throughout this study are based on random samples,

Galician and Castilian in contact

127

and not on an exhaustive linguistic census, which is absolutely necessary for appropriate language planning. The polls we used in collecting the following statistics were carried out at different times from 1970 onwards, using a variety of methods, and polling people from a wide cross-section of society, from housewives to clergymen, but with particular emphasis on teachers and schoolchildren, owing to the ease with which they can be surveyed (Siguän 1992). Table 1 gives a brief description of the polls we collected. 6 Table I

Linguistic polls taken

Poll

Ν

F-70 A-74 S-75 R-77 M-81 W-84 M-84 G-85 C-87 C-87 C-88 C-88 C-88 C-89 C-89

306 1200 300 874 837 104 1925 363 1683 912 446 4335 236 639 129

=

Occupation

Area of poll

Housewives Housewives Population over 18 Primary school teachers School children 5 — 14 Adults over 18 School children 1 0 - 1 4 School children 1 0 - 1 4 School children 1 0 - 1 4 Primary school teachers University lecturers University students Non-academic university staff Clergymen Seminarists

Galicia Arousa, Pontevedra Galicia Galicia Santiago Towns and cities Galicia Ourense, Limia Galicia Galicia Galicia Galicia Galicia Galicia Galicia

3.1 First language (L 1) and linguistic ability in Galician and Castilian In this section we shall refer to the basic data used as a starting point by every study on the social dimensions of pluri-individual bilingualism: the quantification of individuals according to their L 1 and the analysis of their proficiency in basic skills in the co-present languages. We will offer mainly data for Galician, because the surveys were focused on the figures for this language. We should bear in mind that most of the data collected refers to abilities as declared by those interviewed (unfortunately, most of the polls conducted did not test them). However, when gauging the extent of bilingualism, we should be aware of the linguistic proximity of Galician and Castilian. From table 2, we draw the obvious conclusion that Galician is still the L 1 of the vast majority of the Galician population. This represents a

128

Henrique Monteagudo and Anton Santamarina

Table 2

F-70 A-74 W-84

L 1 of adults Galician

Castilian

Bilingual

76% 84% 67%

24% 15% 21%

* *

12%

(Sources: F-70: 1265; A-74: T12; W-84: Tl.) Note that the percentage differences between F-70 and A-74 can probably be explained by the difficulties of collecting representative samples in Galicia, owing to the extreme dispersion of the population. For W-84, note also that this survey was carried out exclusively in urban and semi-urban areas.

solid starting point for correctly devised and implemented language planning to achieve significant results in the medium term. It would not appear difficult to achieve bilingualism in Galicia, with Galician the normal language of the whole community, and Castilian learnt and used fundamentally as an instrumental language for special purposes and external communication. However such an aim meets complex problems, not the least of which is the present legal design of co-officiality of Galician and Castilian which, as we will show, privileges the latter. As regards competence in oral comprehension, in all the polls the vast majority of the Galician population (over 90%) say they understand spoken Galician. In some samples, especially those collected from schoolchildren, around 25% may have some difficulty with this skill. In oral expression, the vast majority of Galicians feel themselves capable of speaking Galician, even though this percentage is lower than for oral comprehension. The extent of oral bilingualism is explained by the close linguistic similarity between Galician and Castilian, together with the fact that Galician is the L 1 of the majority, with Castilian being used by the school and mass-media. As regards reading ability, it is important to take into account the general illiteracy rate in Galicia. According to successive censuses, of Galicians over 10 years of age 9.5% of were illiterate in 1970, and 6.5% in 1981. As Galician was absent from the written media until recently, many adults feel incapable of reading Galician. This becomes even more evident when compared with the percentages of schoolchildren who received some instruction on Galician (in M-81: 72% read 'easily', 24% 'fairly easily'). That adults understand written Galician, despite the absence of the language from the written media and the schools, may be due to its linguistic proximity to Castilian, but, more significantly, a number may have received some type of education of Galician in the

Galician and Castilian in contact Table 3

129

Competent speakers of Galician (self-reported)

Poll

Total

Fluent

Fairly fluent

Poor

F-70 A-74 S-85 W-84 G-85 C-87 C-88

92% 95% 96% 87% 86% 93% 83%

*

*

*

*

*

*

81% 72% 48% 38%

15% 15% 38% 55%

4% 13% 13% 7%

*

*

*

(Sources: F-70: T18.45; A-74: T9; S-85; W-84: T2; G-85: Tl; C-87: T5/A; C-88: Till.12.) Note that from Poll C-87 we took only data for 13 — 14 year old school children; and from poll C-88, only for university students.

period of the Republican schools, and had the opportunity to read press articles and perhaps a book in Galician. During the dictatorship, the presence of Galician in the press was marginal, but effective, and some books in Galician achieved very impressive sales figures. However, the difficulties experienced by adults in reading Galician are persuasive in preventing Galician newspapers from publishing in the language. In this way, illiteracy in Galician functions as a feed-back mechanism: adults find reading Galician difficult, so the press does not publish in Galician, thus reading skills do not improve. In written expression, it is easy to understand why the percentages of adults who consider themselves capable of writing in Galician are low: 20% in W-84 to 37% in S-85 (7% 'well' and a further 30% 'with some difficulty'). As to that, compare the figures in table 4 for university lecturers (C-88 a) with figures for university students (C-88 b). Although the figures are not included in the tables, we should highlight the data for R-77, taken from teachers (only 25% regarded themselves as capable of writing Galician), and those for M-81 from schoolchildren in Santiago (shortly after the introduction of Galician in the curriculum, 26% of them stated that they could write well and 63% with some difficulty). The results of the marginalization of Galician explain the figures presented in table 5, in which the women interviewed (mostly Galicianspeaking but also capable of speaking Castilian) declared themselves more skilled in reading and writing Castilian than Galician. The analysis of the above tables gives a very general picture of the situation, which must be refined by considering at least two important discriminatory factors: location and level of income.

130

Henrique Monteagudo and Anton Santamarina

Table 4

Competence of adults in reading and writing Galician

Poll

Reading

Writing

F-70 A-74 S-75 W-84 C-88 a C-88 b C-88c

42% 50%

24% 22% 37% 20% 44% 62% 48%

*

37% 75% 79% 71%

(Sources: F-70: 1264; A-74: T9; S-75: 227; W-84: T2; C-88: TIII.4.12,20.) Note that W-84 only includes figures for those who responded positively to the questions " D o you read easily?" and " D o you write well?". Poll C-88 is broken down as: C-88 a university lecturers, C-88 b university students, C-88 c non-academic university staff. Table 5

Linguistic skills in Galician and Castilian

Speaking Reading Writing

Galician

Castilian

95% 50% 22%

82% 83% 80%

(Source: A-74: T9,10.)

3.2 Linguistic skills according to location To gauge the importance of the first discriminatory factor, we offer two tables (6 a and 6 b) which need virtually no explanation. Briefly, one must bear in mind the predominantly rural character of Galician society, and also that peasant society and its culture are in rapid decline. Obviously, higher language maintenance in the non-urban environment can be explained by the marginalization of the rural population and its closer adherence to traditional values, whilst the pressure of Castilian in the cities has played havoc with Galician usage. Table 6 a

L 1 of housewives according to place of origin

Galician Castilian (Source: A-74: T28.)

Vigo city

Province -Vigo

Whole province

64% 35%

91% 8%

84% 15%

Galician and Castilian in contact Table 6 b

131

L 1 of school children aged 10 —14 by location

Galician Castilian

City centre

Borough Suburb

Village

Hamlet

Total

7% 93%

16% 84%

36% 64%

72% 28%

42% 56%

12% 88%

(Source: M-84: T2.) Note that 'Borough' and 'Suburb' refer to urban population centres with more than 25,000 inhabitants; 'Village' refers to centres with some 5 — 25,000; 'Hamlet' refers to centres with less than 5,000. Under 'Total', 2% of answers are missing.

Table 7a

Linguistic skills in Galician: adults by location

Speaking Reading Writing

Rural

Urban

Metropolitan

96% 37% 23%

74% 48% 26%

72% 78% 29%

(Source: F-70: T18.45.) Note that it is not clear how this survey defined the terms 'Urban' and 'Metropolitan' in relation to the population of Galicia.

Table 7b

Observed speaking ability of pre-school children in Galician and Castilian, by location Good

Hamlet Village City

Fairly good

None

Monolingual

Bilingual

G

C

G

C

G

C

G

C

G/C

85% 45% 14%

35% 70% 95%

13% 23% 29%

23% 25% 1%

2% 31% 56%

42% 5% 2%

42% 5% 2%

2% 31% 56%

66% 64% 42%

(Source: C-87: T9/A, 15/A.) The children were approximately 5 years old and had received no previous education. The row 'City' is missing 1 % of replies on competence in Galician and 2% on Castilian.

Location is also a clearly discriminatory factor in linguistic skills. Owing to the higher standard of living in the cities, with greater ease of access to education and increased diffusion, density, and influence of the mass-media, and also because the cities were the nuclei for the Galician cultural movement and democratic resistance, paradoxically, despite the fact that less Galician is spoken, the percentage of people able to read and write is far higher than elsewhere. In table 7 b we should highlight two conclusions: first, it seems that village children are more commonly bilingual before starting school than

132

Henrique Monteagudo and Anton Santamarina

town children; secondly, in the same circumstances, similar percentages of village children from hamlets have knowledge of Castilian as city children have of Galician. The semi-urban area is the transitional zone between the predominance of Galician or Castilian, and is also the area where today the processes of castilianization and regaleguization are most intense. 3.3 Linguistic skills and L 1 according to level of income For the correlation between the level of income and linguistic competence in Galician and Castilian, we present the following two tables: Table 8 a

LI Reading Writing

Level of income as a discriminatory factor in linguistic skills in Galician and Castilian Very low G C

Low G

97% 30% 17%

83% 38% 15%

3% * *

C

Medium G C

High G

17% * *

68% 51% 36%

48% 66% 30%

32% * *

C 50% * *

Very high G C 20% 54% 21%

80% * *

(Source: F-70: T18.47.) Table 8 b

Profession as a discriminatory factor in L 1

Professionals with university degrees High ranking officials, directors and managers White-collar workers Skilled professionals (electricians, plumbers) Employers in small/medium businesses and shopkeepers Non-white-collar workers in businesses and shops Non-active population (unemployed, retired) Farmers and fishermen

Galician

Castilian

22% 30% 41% 64% 68% 73% 76% 93%

78% 70% 59% 36% 32% 27% 24% 7%

(Source: M-84: T18.)

Tables 8 a and 8 b are particularly informative about the social distribution of Galician and Castilian, and its congruence with the locational distribution: the social sectors with the lowest incomes are essentially those living in the countryside, employed in agriculture-related professions, farming, and fishing. This subordinate situation of Galician is another clear indication of the social retardation of Galician society. Language planning aimed only at spreading basic linguistic skills to the

Galician and Castilian in contact

133

whole population (that is, simply seeking to create bilingualism throughout the population), will probably result in slowing down, but not in halting, the language shift away from Galician. Policy aimed at the social spread of Galician should stress the socio-cultural enhancement of the Galician-speaking community. Where communities are in a state of inequality, a fair democratic policy does not mean treating them as equal, but practising positive discrimination. The results of M-81 indicate a significant change of trend in the factors which influence skills in Galician, principally due to the introduction of Galician in schools and the general change in attitudes towards the language. Among children, social class is not yet a clearly discriminating factor in linguistic competence in Galician (although one must remember that these are self-evaluating interviewees), given that the percentage of individuals who regard themselves as capable of speaking fluently and fairly fluently in Galician is almost invariably 85%, regardless of social class. Nor does location seem to be a discriminatory factor: the 85% does not vary significantly between children from urban and rural backgrounds. On the other hand, these factors do discriminate between adults: thus, in W-84: 2, we find that adults over 36 from lower class and semirural backgrounds (remembering that this survey does not offer statistics for rural areas) consider themselves more able in speaking Galician than younger people from the middle class and urban areas, but less able in reading and writing. 3.4 The dynamics of bilingualism: code switching and language shift The data offered by M-84 on the language habitually used in oral communication according to generation indicate the extent of language shift: Table 9

Galician Castilian

Customary language by generation Grand parents

Parents

Children

83% 17%

69% 31%

42% 56%

(Source: M-84: T32.2.) These are figures given by children on the linguistic behaviour of their parents and grandparents. Under 'Children', 2% are missing.

Linguistic behaviour inside the family is a good indicator of the level of bilingualism of Galicians and of language shift. We can analyse language

134

Henrique Monteagudo and Anton Santamarina

use in relations between different generations of families by drawing on the extremely revealing statistics in table 10. The figures for tables 9 and 10 clearly indicate the trend of language shift: as Castilian steadily increases, Galician monolinguals continue to decrease. Comparing the percentages of bilinguals in the family domain through the three generations, in the passage from the present parents' generation to the children's, the practice of bilingualism inside the family unit has diminished. This results in a polarization between monolingual Castilian-speaking children and monolingual Galician-speaking children. It is the calculus of social advancement that explains the phenomenon of Galician-speaking parents who use Castilian when speaking to their children. Factors related to the cultural prestige of Castilian and the prejudices which this produces are, in our view, secondary and overemphasized by some scholars. 7 This social factor must be taken into account in language planning if we wish to avoid a linguistic bias which would probably lead to failure. In the light of this, the policies of successive Spanish and Galician governments with regard to Galician may be interpreted as fake promotion which in practice justifies and legitimizes the language shift in favour of Castilian. Poll M-81, which paid most attention to language maintenance and language shift over the last two generations, gives the following results, which must be highlighted: — Galician receded more as a first language ( L I ) for children than it actually did in use. — Children retain the linguistic behaviour of their parents far more than the previous generation. — The percentage of degaleguization diminishes whilst that of regaleguization doubles in the transition from parents to children, although regaleguization is still less than degaleguization. — In recent years, the rate of abandonment of Galician has declined, owing mainly to changes in the linguistic behaviour of the urban middle class. Poll M-84 studies the relationship between the customary language of schoolchildren and the language of their families. The conclusion is that the social and linguistic environments exert a powerful influence on the language of the children: so, for children from galeguized family environments (those in which their grandparents and parents speak to them entirely in Galician), we find that in urban backgrounds 56% of them

Galician and Castilian in contact

ο C ο ε

S 3 υ 3 ÖÜ ο α cο 13 ε

£τ—