(9783110852004 - Opportunities and Challenges of Bilingualism) Opportunities and Challenges of Bilingualism

Opportunities and Challenges of Bilingualism
WDE
G
Contributions to the Sociology of Language
87
Editor
Joshua A. Fishman
Mouton de Gruyter
Berlin · New York
Opportunities and Challenges
of Bilingualism
edited by
Li Wei
Jean-Marc Dewaele
Alex Housen
Mouton de Gruyter
Berlin · New York 2002
Mouton de Gruyter (formerly Mouton, The Hague)
is a Division of Walter de Gruyter & Co. KG, Berlin
© Printed on acid-free paper which falls within the guidelines of

the ANSI to ensure permanence and durability.
Library of Congress — Cataloging-in-Publication Data
Opportunities and challenges of bilingualism / edited by Li Wei,

Jean-Marc Dewaele, Alex Housen.
p. cm. — (Contributions to the sociology of language ;
87)
Includes bibliographical references and index.
ISBN 3 11 017305 0 (cloth : alk. paper)
1. Bilingualism I. Wei, Li, 1961— II. Dewaele, Jean-
Marc, 1962. III. Housen, Alex, 1964- IV. Series.
PI 15 .067 2002
404'.2—dc21
2002072426
Die Deutsche Bibliothek — Cataloging-in-Publication Data
Opportunities and challenges of bilingualism / ed. by Li Wei

— Berlin ; New York : Mouton de Gruyter, 2002
(Contributions to the sociology of language ; 87)
ISBN 3-11-017305-0
© Copyright 2002 by Walter de Gruyter GmbH & Co. KG, D-10785 Berlin
All rights reserved, including those of translation into foreign languages. No part of this
book may be reproduced or transmitted 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.
Printed in Germany.
Acknowledgements
We would like to thank Professor Joshua A. Fishman, the editor of the series to
which this volume belongs, for his enthusiasm, support, practical advice and
contribution. He has been instrumental is getting this project off the ground and
seeing it through to its publication. It is no exaggeration to say that without him
this volume would not have been possible. We are also grateful to Marilyn
Martin-Jones and Christina Bratt Paulston, who, for various reasons, could not
send in their contributions to the volume but gave us equally valuable support
throughout the project. We are particularly grateful to Dr. Anke Beck and
Rebecca Walter at Mouton de Gruyter for their support and patience. Anna-
belle David put in a great deal of time as the copy editor of the volume. Need-
less to say, there would be no book at all without the contributions from the
authors. As editors, we thank them all most sincerely for their work.
Contents
Acknowledgements ν
Contributors ix
Introduction: Opportunities and challenges of bilingualism

Li Wei, Jean-Marc Dewaele and Alex Housen 1
I. Theoretical frameworks
"Holy languages" in the context of societal bilingualism

Joshua A. Fishman 15
Forlorn hope?
John Edwards 25
When languages disappear, are bilingual education or human rights a
cure? Two scenarios
Tove Skutnabb-Kangas 45
Core values and nation-states
J. J. Smolicz 69
II. Bilingualism worldwide
French language policy: centrism, Orwellian dirigisme, or economic

determinism?
Harold F. Schiffman 89
The non-linearity of language maintenance and language shift: survey
data from European language boundaries
Peter Η. Neide and Peter J. Weber 105
Language shift among Siberian Estonians: pro and contra
Jiiri Viikberg 125
On attitudes towards Croatian dialects and on their changing status
Damir Kalogjera 145
Ethnolects- between bilingualism and urban dialect
Wolfgang Wölck 157
viii Contents
The development of Navajo-English bilingualism

Bernard Spolsky 171
Language ideology, ownership and maintenance: the discourse of the
Academia Mayor de la Lengua Quechua
TimMarr 199
Xhosa as a "home appliance"? A case study of language shift in
Grahamstown
Vivian de Klerk 221
Japan's nascent multilingualism
Florian Coulmas and Makoto Watanabe 249
III. Multilingual management and education
Managing multilingualism in Singapore

Xu Darning and Li Wei 275
Managing languages at bilingual universities: relationships between
universities and their language environment
Björn Η. Jernudd 297
Using descriptive inquiry to transform the education of linguistically
diverse US teachers and students
Ofelia Garcia and Cecelia Traugh 311
Coda
Changing paradigms in the study of bilingualism

William F. Mackey 329
Index 345
Contributors
Professor Dr. Florian Coulmas
Modernes Japan, FB 3
Gerhard Mercator University
47048 Duisburg
Germany
Professor Vivian de Klerk
Department of English Language and Linguistics
Rhodes University
Grahamstown, 6140
South Africa
Dr. Jean-Marc Dewaele
School of Languages, Linguistics and Culture
Birkbeck College
University of London
43 Gordon Square
London WC1H OPD
UK
Dr. John Edwards
Department of Psychology
St Francis Xavier University
P.O. Box 5000
Antigonish, Nova Scotia, B2G 2W5
Canada
Professor Joshua A. Fishman
Ferkauf Graduate School of Psychology
Yeshiva University
1300 Morris Park Avenue
Bronx, NY 10461
USA
Dr. Ofelia Garcia
School of Education
Long Island University
Brooklyn Campus
University Plaza
Brooklyn, New York 11201
USA
χ Contributors
Dr. Alex Housen

Department of Germanic Languages and
Centre for Linguistics
Vrije Universiteit Brüssel
Pleinlaan 2
B-1050 Brüssel
Belgium
Professor Björn Η. Jemudd
Department of English
Hong Kong Baptist University
Kowloon Tong, Kowloon
Hong Kong
Professor Dr. Damir Kalogjera
Department of English
Faculty of Philosophy
University of Zagreb
Ivana Lucica 3, 10 000 Zagreb
Croatia
Professor Li Wei
School of Education, Communication and
Language Science
King George VI Building
University of Newcastle upon Tyne
NE17RU
UK
Dr. William F. Mackey
Le Verre Bourg (812)
999, Beaurgard
Quebec G1V 4T9
Canada
Makoto Watanabe
Gerhard-Mercator-Universitat Duisburg
FB3 Modernes Japan
47048 Duisburg
Germany
Dr. Tim Man-
School of Area and Language Studies
London Metropolitan University
London N7 8DB
UK
Professor Dr. Peter Η. Neide
Centre de Recherche sur le Plurilinguisme
K.U. Brüssel
Vrijheidslaan 17, Ave. de la liberte
B-1081 Brussels
Belgium
Dr. Harold F. Schiffman
Department of South Asian Regional Studies
804 Williams Hall, Box 6305
University of Pennsylvania
Philadelphia, PA 19104-6305
USA
Dr. Tove Skutnabb-Kangas
Roskilde University
Department of Languages and Culture
P.O.Box 260, DK-4000
Roskilde
Denmark
Professor J. J. Smolicz
The Graduate School of Education
University of Adelaide
Education Building
245 North Terrace
Adelaide, SA 5005
Australia
Professor Bernard Spolsky
Department of English and
Language Policy Research Center
Bar-Ilan University
52 900 Ramat-Gan
Israel
Cecelia Traugh
School of Education
Long Island University
Brooklyn Campus
University Plaza
Brooklyn, New York 11201
USA
xii Contributors
Dr. Jiiri Viikberg

Institute of the Estonian Language
Roosikrantsi 6
10119 Tallinn
Estonia
Dr. Peter J. Weber
Centre de Recherche sur le Plurilinguisme
K.U.Brussel
Vrijheidslaan 17, Ave. de la liberte
B-1081 Brussels
Belgium
Dr. Wolfgang Wölck
Department of Linguistics
629 Baldy Hall
State University of New York
Buffalo, NY 14260
USA
Dr. Xu Darning
Department of East Asian Languages and Cultures
University of California, Los Angeles
290 Royce Hall
Box 951540
Los Angeles, CA 90095-1540
USA
Introduction: Opportunities and challenges of
bilingualism
Li Wei, Jean-Marc Dewaele and Alex Housen
When one senior academic proposed to teach a course on bilingualism, the uni-
versity authorities asked him "Why?". After having justified his reasons in de-
tail, and in writing, the senior academic was told that he could only offer the
course on the condition that he called it "Problems of Bilingualism". He was
puzzled as to what the "problems" were, or for whom, but was not given any
explanation by the university authorities. This event did not happen in a mono-
lingual country, nor centuries ago. It happened in the late 1970s, in Brussels,
the officially bilingual capital of both trilingual Belgium and the multilingual
European Union.
Things have moved a long way in the last two decades as popular miscon-
ceptions about the problematic nature of bilingualism have gradually been
given way to a growing recognition of the prevalence, normalcy and value of
bi- and multi-lingualism, both for the individual and for the larger community.
Although bilingualism is of all times, the need for compentency in more than
one language has probably never been greater than at the turn of the third mille-
nium. This is the result of a variety of factors, including, globalisation of busi-
ness, commerce and entertainment, massive population shifts of people from
different ethnolinguistic backgrounds, rapid urbanisation, and cheaper and
faster means of international travel and communication. These factors, together
with the spread of bilingual and second/foreign language education, have all
helped, in some way, to increase the number of people who can speak more
than just their mother tongue. Many of them are quite proud of the fact that they
are bi- or multi-lingual. Their intelligence has apparently not been negatively
affected, as was previously assumed, and more importantly perhaps, their pro-
fessional, social and economic status is often enhanced by their bi- or multi-
linguality. Even the university authorities, especially those in English-domi-
nant countries, having realised at last that overseas students are a huge financial
asset, publicly acknowledge the significance of having more people in their es-
tablishments who may be speaking a "foreign tongue".
In the meantime, the political climate has changed too. There is an in-
creased awareness of the language rights by/of both indigenous and immigrant
communities in developed countries. And multilingualism and multicultural-
2 Li Wei, Jean-Marc Dewaele and Alex Housen
ism have entered the public arena worldwide as an important social issue for
debate. This volume aims to join in such a debate by examining, from a truly
international perspective, the opportunities and challenges of societal bilin-
gualism in the new millennium.
1. What are the opportunities?
Perhaps one of the most noticeable social changes in the last two decades is the
increased opportunities for individuals to become bilingual. It has never been
easier for people to encounter and learn new languages, in schools, through
professional contacts, on the Internet, through music, arts and other forms of
entertainment, and in everyday social interaction. Contacts with people who
speak languages other than one's own are increasingly becoming part of the
daily routine, particularly in the urban environments. While it was formerly be-
lieved that such urban environments represented ideal contexts for the gradual
loss of linguistic heterogeneity, with linguistic minority groups being "ab-
sorbed" by the mainstream majority group, we now know that the exact oppo-
site is often the case and that resident patterns in the form of ghettoes or less pe-
joratively, "urban villages", tend to promote the maintenance of multilingual-
ism and linguistic diversity, particularly though by no means exclusively
among immigrants. A short walk in the downtown areas of London, New York,
Paris, Toronto or Sydney will reveal the extent of multilingualism presents. Im-
portantly, this linguistic diversity is not or no longer a feature typical of the
great metropolitan centres. It is often said, for instance, that in countries like
Britain and the USA, there is bound to be at least one Indian or Chinese take-
away in a town of 2,000 people.
Bilingualism in turn brings new opportunities to both the individual and so-
ciety. For the bilingual individual, the ownership of two languages has increas-
ingly become seen as an asset as the 'communication world' gets smaller. With
the dramatically increased amount of information available, and the ease of de-
livering it around the world, bilinguals may have become more important in the
employment market. There is a growing realisation, for example, that bilin-
guals and multilinguals are highly useful in the competitive international trade.
Many bilinguals are now working in what is perceived as prestigious profes-
sions, such as transnational business corporations, government, journalism,
and tourism.
For many people, being bilingual has also given them new life experiences.
While a monolingual person may experience a variety of cultures, from differ-
Introduction: Opportunities and challenges of bilingualism 3
ent colleagues and neighbours, for instance, who use the same language but
have different ways of life, bilingualism provides the opportunity to experience
two or more cultures more directly, to participate and become involved in the
core of a culture, and to appreciate the different systems of behaviour, rituals,
religious traditions, beliefs and values, histories, and literatures. Those who be-
came bilingual by moving to a new linguistic and cultural environment (e.g.
immigrants, refugees, educational and professional transients) have an oppor-
tunity to reflect on their linguistic and cultural heritage and, as many of such bi-
linguals claim, discover and develop new identities.
Bilingualism brings opportunities not only to the individual but also to the
society as a whole. In today's world, the economic strength of a nation on the
world market is not associated with how many monolingual speakers it may
have but how many bi- and multi-linguals it has. The number of bi- and multi-
lingual speakers a country produces is often seen as an indicator of the edu-
cational standard, economic competitiveness and cultural vibrancy of the
country. Bilingualism offers the society a bridge-building potential, bridges be-
tween different groups within the nation, bridges with groups beyond the artifi-
cial borders of a nation, and bridges for cross-fertilization between cultures.
Bilingualism has given geographically small countries, such as Singapore,
economic and cultural vitality and a distinctive national identity.
Bilingualism also prompts the society to rethink about the relationship be-
tween unity and diversity, to come round for the idea of peaceful co-existence
between different linguistic and cultural groups and to observe the rights and
obligations of each other. It has been shown that tolerance and co-operation be-
tween groups is possible only when linguistic diversity is respected. There is
now a wide-spread realisation that bilingualism is the norm for both the indi-
vidual and society.
2. What are the challenges?
With the new opportunities of bilingualism come new challenges. It is often the
case that being bilingual is not enough in itself. One has to be bilingual in the
"right" languages, in order to be socially upwardly mobile. There is a story that
when the admissions officer of a fee-paying high school in England brought in
a new pupil to see the headmaster and said "This is X. He speaks fluent Latin",
the headmaster said, in a loud and cheerful voice, "Oh good, you can go to Ox-
ford". When the same admissions officer brought in another pupil and said,
"This is X. He speaks fluent French". The headmaster said, in an equally loud
but less cheerful voice, "Oh yes, he can go to a university". Then the admis-
sions officer brought in yet another pupil and said, "This is X. He speaks fluent
Urdu". The headmaster said, in a not so loud and definitely not cheerful voice,
"Oh dear, what do we do about him?". This goes to illustrate that although from
a linguistic point of view all languages are equal, from a social, economic and
political point, some languages are more valuable than others.
A frequent debate amongst the people who aspire to be bi- or multi-lingual
concerns which language should be learned. The favoured languages tend to be
those that are perceived to be both international and particularly valuable in in-
ternational trade. A lower place is given in the status ranking to minority lan-
guages that are small, regional and of less perceived value in the international
marketplace.
The bilingual individual not only has to struggle with societal prejudices to-
wards the languages he or she possesses, but also with his or her self-identity. A
popularly voiced worry about bilingualism is whether the individuals con-
cerned would be caught in between two languages and cultures. Would a bilin-
gual belong to neither language group, feel neither Spanish nor American,
neither Irish nor English, neither Japanese nor Australian? In learning another
language and another culture, would a bilingual experience social alienation, a
conflict of personal and ethnic identity, personal disorientation, social isolation
and anxiety?
Facing such challenges, the individual's responses often vary: some quickly
switch to the new culture and lifestyle, while others try hard to retain their
native culture and language. Still others manage to bridge the two. A few may
feel a sense of rootlessness. Some have the choice of a new identity, others have
no choice, finding themselves unable to break out of the circle of economic, so-
cial and political oppression. It is important to realise and remember that when
one's identity becomes an issue in a language contact situation, bilingualism is
unlikely to be the cause. In other words, it is not language per se that causes the
identity crisis; rather, it is often the social, economic and political conditions
surrounding the development of bilingualism that generate such problems.
Where a language group is stigmatised, seen as socially inferior, economically
underprivileged, and where there is symbolic and physical violence towards
the people speaking a particular language, identity problems arise. It is not the
ownership of bilingualism but the condition in which the bilingual community
lives that may be the cause of the problem. Where bilinguals are encouraged, or
even forced, to become monolinguals in the majority language, identity, and
many other, problems occur. Such integration and assimilation may be de-
manded on behalf of the majority but are difficult to achieve. The majority lan-
guage community may be resistant to newcomers. Where race and skin colour
enter the equation, a bilingual may not be admitted into the host culture. There
may be a bar on access to the better jobs, to educational opportunities and to
achieving affluence.
Perhaps the really serious challenges of bilingualism come at the societal
rather than the individual level. Societies need to address the question: Do lan-
guages require individual territories, or can they co-exist depending on their
use, function and status? This seemingly straightforward question is often at
the heart of heated political and ideological debates.
There seems to be a tendency for the society to impose a majority vs. mi-
nority categorisation upon languages and language communities. The Euro-
pean Union (EU), for example, categorises the languages spoken in its member
states in terms of "major national" languages (e.g. Dutch, French, German, Ita-
lian, Spanish, etc.), "lesser" (-used or -taught) languages (e.g. Basque, Catalan,
Irish, Romansch, Welsh, etc.), and "minority" languages (e.g. Corsican, Fri-
sian, Gaelic, Romani, Sicilian, etc.). All its good intentions and efforts notwith-
standing, the political reality of contemporary Europe means that major re-
sources, especially educational resources, are likely to be devoted to the "major
national" languages. The so-called "lesser" languages may receive some insti-
tutional support, as much as economic policies permit, but the "minority" lan-
guages are unlikely to receive any real attention but lip service. If these "mi-
nority" languages continue to have little support, the likelihood of their disap-
pearance is substantially increased. This has been the case of minority lan-
guages worldwide. It has been suggested that languages are "dying" at a rate of
two a month. Regional authorities, such as the Provincial Council of Friesland,
have tried hard to protect "minority" languages in their localities, but it is clear
that the major national languages still have more social, educational, cultural,
and political prestige even in places where the "minority" language speakers
are a quantitative majority.
The majority vs. minority classification is problematic for various reasons.
Apart from its dichotomous nature, it lacks a solid scientific basis which makes
operationalisation very difficult. If one bases the categorisation solely on the
number of speakers, as the EU appears to have done, one finds a curiously
anomalous situation in South Africa, for example. In numerical terms, Afri-
kaans is a minority language. But it has been a majority language in a political
sense. It is the language of that part of the total community which held virtually
all of the political and economic power.
One thing to notice in the EU's categorisation of its languages is that there
is no mention of the non-European languages spoken by immigrant commu-
nities across the member states. In most EU countries, languages such as
Arabic (in various forms), Chinese (in various forms), Turkish, and the Indian
languages (e.g. Panjabi, Urdu, Gujarati) are usually described as "minority"

languages. This is not because these languages are spoken by insignificant
numbers of people - there are more people in the world claiming to be native
speakers of Arabic and Chinese than of all other languages combined but be-
cause their speakers are regarded as "minorities" in EU countries. One would
not hear French or German being described as "minority" languages in Eng-
land or Spain even though there are fewer native speakers of these language
than, say, Moroccan Arabic or Cantonese. A similar case is German in Bel-
gium. It is one of the three official languages along with Dutch and French, al-
though there are only 66,000 speakers, all of whom are bilingual in German
and French. This tiny autochtonous German part of the population enjoys all
possible political, cultural and educational provisions, including their own
government and parliament, television and radio, and bilingual schools, which
is in sharp contrast to the allochtonous minority communities which, although
far outnumber the German-speaking population, receive no provisions or sup-
port whatsoever.
Bilingualism also offers great opportunities as well as challenges to edu-
cation. For many people education is one, if not the main, road towards bilin-
gualism. Although bilingualism represents a growth area in education, both
specialists and the general public still have difficulty in coming to terms with it.
In many parts of the world, education has evolved along monolingual modes of
thinking, and consequently there has been a tendency to equate bilingual edu-
cation with a host of problematic connotations, including lowering of academic
standards, organisational and managerial problems, cost-ineffectiveness, and
general inefficiency and educational anarchy. However, the need for great
multilingual competency provoked by the factors mentioned above forces one
to reassess the traditional assumptions about the relation between language and
education.
Without recognising and accepting that different language communities
should have equal social status means the rich linguistic resources we have in
our society are wasted. There is a paradox that the so-called bilingual education
often aims to teach the majority language, such as English and French, to mi-
nority children. The trend is to appreciate majority European languages speak-
ers who learn a second, non-European language to ensure a continued major
role for such countries as the US and the UK in world politics and the world
economy. A good case in point is the Higher Education Funding Council of
England's (HEFCE) most recent report on Chinese studies in Britain. In this
conclusion, it claims that "few UK nationals can interpret and translate Chi-
nese" (Paragraph 56). What happened to the thousands of Chinese people in
Britain who constitute the third largest immigrant community in the country
and who hold British passports? Aren't they UK nationals? The same report
also claims that:
It is difficult to form a complete picture of numbers of academic staff in Chinese lan-

guages. The EACS (European Association of Chinese Studies) identified 46 full-
time staff with skills in Mandarin, of whom three also spoke Cantonese, across six
institutions. (Paragraph 52)
They obviously disregarded all the ethnically Chinese staff who are working in
the UK higher education institutions in their calculation. It is remarkable that
such biased, and possibly racist, claims found their place in government-spon-
sored reports.
In contrast, the Nuffield Languages Inquiry in Britain, a non-government-
sponsored inquiry, points out:
The remarkable linguistic diversity of the UK, reflecting our complex history, in-
cludes speakers of the indigenous languages and also of the languages of our main
Asian, European and Afro-Caribbean communities as well as hundreds of smaller
groups of speakers of other languages. Yet the multilingual talents of UK citizens are
under-recognised, under-used and all too often viewed with suspicion. Our aims
must be to recognise the opportunities offered by this multilingual wealth, ensure
that talent is nurtured in future generations and meet the linguistic and cultural needs
of individuals and communities. (Languages: The Next Generation. The final report
and recommendations of Nuffield Languages Inquiry 2000)
There is a tendency for societies to encourage and promote the study of foreign
languages for major European language (e.g. English, French, German Span-
ish) monolinguals, at great cost and with great inefficiency. At the same time
we devalue and destroy the linguistic gifts that people from non-European lan-
guage backgrounds bring to our societies.
To accord different languages and language communities equal rights and
status is perhaps one of the toughest challenges societies have to face in
today's world. Some suggest that bi- and multi-lingual societies are less well
co-ordinated and managed than monolingual and unicultural ones, and that
they tend to have certain problems which monolingual societies do not. On the
practical level, it is often claimed that difficulties in communication within a
country can act as an impediment to commerce and industry. More seriously,
however, bilingualism is seen as a problem for government. The process of
governing requires communication both within the governing institutions and
between the government and the people. This means that a language, or lan-
guages, must be selected as the language for use in governing. But the selec-
tion of the "official language" is not always easy, as it is not simply a pragmatic
issue. For example, on pragmatic grounds, the best immediate choice for the
language of government in a newly independent colony might be the old colo-
nial language, since the colonial governing institutions and records are already
in place in that language and those nationals with the most government experi-
ence already know it. But the old colonial language may not be a good choice
on nationalist grounds. For a people which has just acquired its own geo-
graphical territory, the language of the state which had denied it territorial con-
trol would not be a desirable candidate for a national symbol. Ireland has
adopted a strategy whereby both the national language, Irish, and the language
of the deposed power, English, are declared as official, and used the colonial
language for immediate, practical purposes while working on the promotion
and development of the national language. But in many other bilingual coun-
tries which do not have a colonial past, such as China, deciding which lan-
guage should be selected as the national language can sometimes lead to inter-
nal, ethnic conflicts.
Similarly, selecting a language for education in a bilingual country is often
problematic. In some respects, the best strategy for language in education is to
use the various ethnic languages. After all, these are the languages the children
already speak and school instruction can begin immediately without waiting
until children learn the official language. Some would argue, however, that this
strategy could be damaging for the nation-building efforts and disadvantage
children by limiting their access to the wider world.
It should be stressed once more that there is no scientific evidence to show
that multilingual countries are particularly disadvantaged, in socio-economic
terms, compared to monolingual ones. In fact, all the research that was carried
out in the 1960s and 70s on the relationship between the linguistic diversity and
economic prosperity of a nation came to the conclusion that a country can have
any degree of language uniformity or fragmentation and still be underdevel-
oped; whereas a country whose entire population speaks the same language can
be anywhere from very rich to very poor. It might be true that linguistic uni-
formity and economic development reinforce each other; in other words, econ-
omic well-being promotes the reduction of linguistic diversity. But it would be
too one-sided, to say the least, to view bilingualism as the cause of socio-econ-
omic problems of a nation.
3. Opportunities and challenges for the research community
The increased amount of bilingualism at both the individual and societal levels
offers the academic research community, a specific sector of society, new op-
portunity to evaluate their theories and models of language and communi-
cation. The last two decades have witnessed an upsurge of publications on is-
sues related to bilingualism and language contact. Academics have finally
come to the open to display and defend their interests in bilingual speakers and
bilingual communities. Yet, concerns of being trivialized, marginalized, preju-
diced and discriminated against have driven many researchers to focus on nar-
rowly defined linguistic and psycholinguistic issues in order that their work ap-
pears to be "scientific". Many published studies attempt to present bilingualism
as special cases which are used to evaluate theories that are developed on the
basis of monolingual evidence.
Thus it is often overlooked that bilingualism is also, and as some would
have it, first and foremost a social phenomenon. Bilingualism always grows
within specific historical, political, economic contexts. Failing to address or
even avoiding the social aspects of bilingualism is to the detriment of the very
subject area academic researchers are so keenly promoting. A challenge to the
academic community therefore is to make what may be viewed as scientific re-
search socially relevant as well. Academic researchers working in the field of
bilingualism should feel proud about the fact they not only have a lot to say
about the linguistic and psychological theories and models, but also make sig-
nificant contributions to socio-political debates about the world we are living in
today.
Political authorities in today's world do seem to realise that linguistic diver-
sity is a sensitive issue. Nevertheless one might wonder how sincere political
authorities are when they officially promote bilingualism and multilingualism.
Do they really want their citizens to be equally fluent in the national lan-
guage^) and minority / immigrant / foreign languages? Or are they using
"double-speak", preferring a "minimal bilingualism" where knowledge in the
minority/immigrant/foreign language is just enough to allow for basic com-
munication but not enough for the speaker to become or to feel bicultural?
Since culture and language are so closely intertwined, many politicians may
fear that introducing "foreign" languages / cultures too early (i.e. in kinder-
garten) might have dire political consequences: i.e. the dilution of the individ-
ual's sense of national identity and group membership. They may not want the
children to be too attracted to the other language communities, especially in
countries where there is tension between communities. Ideally they may prefer
the learners to develop a purely instrumental motivation in their learning of a
second language. Some researchers have wondered whether the emphasis on

communicative competence, often through boring artificial dialogues, at the
expense of attractive sociocultural information in second language course-
books is dictated by political concerns. It is telling that a second language is
generally introduced at the end of primary or at the beginning of secondary
education, despite consistent findings that native-like competence in the sec-
ond language is much harder to attain for those who start learning the language
later in life. When challenged, the defenders of such systems will say that the
introduction of a second language at too early an age might somehow confuse
the children. It matters little to them that studies on bilingual or trilingual first
language acquisition report no trace of confusion in the minds of the subjects.
This traditional argument is based partly on the fear of loss of national identity
and partly on an extreme interpretation of theories on linguistic relativism:
since (first) language and thought are so closely connected, the early introduc-
tion of a second language may disturb the development of thinking skills of the
young. It is up to the researchers in the field of bilingualism to lay these ir-
rational fears to rest through carefully designed experiments. It is equally im-
portant that bilingual researchers address socio-political issues head-on.
4. About this volume
The papers in this volume aim to contribute to the debates about the social as-
pects of bilingualism, focusing on the various opportunities and challenges bi-
lingualism presents to today's society.
Joshua A. Fishman's chapter, '"Holy languages' in the context of societal
bilingualism", is a critical analysis of the process of sanctification of one or
more languages within multilingual societies. Using Dürkheim's definition of
"holiness" as a point of departure, he examines different types of "holy" lan-
guage throughout human history, the consequences of language sanctity beliefs
and the staying power of sanctity. Like many of his previous writings, this
chapter by Fishman breaks new ground and sets the agenda for future dis-
cussions of language ideologies in bi- and multi-lingual societies.
John Edward's chapter, "Forlorn hope?", is concerned primarily with the
fate of the smaller languages in the world. He examines the "language ecol-
ogy" perspective on linguistic diversity, and argues that stable societal bilin-
gualism is both a desirable outcome and one that is very difficult to achieve and
maintain.
Tove Skutnabb-Kangas' chapter, "When languages disappear, are bilingual
education or human rights a cure?", continues the discussion of endangered
languages. Taking what she calls the Linguistic Human Rights (LHR) perspec-
tive, she examines the work in international law and in education to counteract
minorisation of languges and to promote the survival of linguistic diversity.
J.J. Smolicz's chapter, "Core values and nation-states", focuses on the para-
dox of ethnic and national upsurge in a globalising culture and economy. Using
the Australian situation as an illustration, he argues that linguistic diversity is
dependent on the maintenance and development of cultural core values of each
of the constituent national/ethnic groups. Such core values are indispensable
elements of the group's identity, but their existence is compatible with the de-
velopment within the nation-state of an overarching framework of shared val-
ues, built upon multicultural and multilingual rather than monistic premises.
The following papers address societal bilingualism, language ideology, lan-
guage planning, language maintenance and language shift in individual coun-
tries. Harold F. Schiffman's chapter, "French language policy: centrism, Or-
wellian dirigisme or economic determinism?", deals with the broader question
of how language policy is embedded in and proceeds from the so-called "lin-
guistic culture" by examining the case of language policies in France. In "The
non-linearity of language maintenance and language shift", Peter H. Neide and
Peter J. Weber investigate the European language boundaries. Jüri Viikberg's
chapter, "Language shift among Siberian Estonians", discusses the factors af-
fecting language maintenance and language loss in the Estonian linguistic en-
claves in Siberia. Damir Kalogjera's chapter, "On attitudes towards Croatian
dialects and on their changing status", examines the notions of "norm" and
"standard language" in the context of language contact in Croatia. Wolfgang
Wölck in "Ethnolects - between bilingualism and urban dialect", describes the
effects of language contact in the city of Buffalo in New York State. Bernard
Spolsky's chapter, "The development of Navajo-English language bilin-
gualism", explores the process of Navajo language loss in the southwestern
states of America. Tim Marr's chapter, "Language ideology, ownership and
maintenance", analyses the discourse of the Academia Mayor de la Lengua
Quechua in Peru. In "Xhosa as a 'home appliance'?", "Vivian de Klerk presents
a case study of language shift in Grahamstown in South Africa. Florian Coul-
mas and Makoto Watanabe's chapter, "Japan's nascent multilingualism", looks
at the causes of increasing linguistic diversity in modern Japan.
The three chapters that follow address the issue of language management
and language education in multilingual contexts. Xu Darning and Li Wei's
chapter, "Managing multilingualism in Singapore", examines the motivations
for the language policies advocated by the Singaporean government. Björn
Η. Jernudd, "Managing languages at bilingual universities", discusses the rela-
tionships between universities and their language environment. Garcia and
Traugh and their colleagues in "Using descriptive inquiry to transform the edu-
cation of linguistically diverse US teachers and students", describe the efforts
educator in two US settings are making to continue to support linguistically di-
verse schools and discusses how a particular research methodology could be
used to regenerate teacher training program and help deepen the understanding
of bilingual education.
The last chapter by William F. Mackey, "Changing paradigms in the study
of bilingualism", critically reviews key issues in bilingualism research in the
last twenty or so years, many of which have been discussed in some detail in
the preceding chapters. It also outlines a framework for future research. We
have chosen to use the chapter as Coda to the volume.
At the end of each chapter, there are five questions for discussion. These
questions are aimed to facilitate further debates about the issues raised and dis-
cussed in the chapters. Some of the key references are asterisked. Readers are
invited to consult these references for further discussions.
The senior academic whose experience with the university authorities we
mentioned at the beginning of the Introduction is Professor DrHugo Baetens
Beardsmore, President of the Royal Academy of Oversees Sciences, Belgium
and former Dean of the Faculty of Letters of the Dutch-speaking Vrije Univer-
siteit Brüssel. He has devoted much of his professional life to research on bi-
lingualism and bilingual education, and has been active as a consultant on lan-
guage planning and education to various official bodies in Europe and the Far
East. He has also inspired and trained a large number of young scholars right
across the world, including two of the editors of this volume. Professor
Dr Baetens Beardsmore is due to retire from the Free University of Brussels in
September 2002. We, together with all the contributors, would like to dedicate
this volume to him.
I. Theoretical frameworks
"Holy languages" in the context of societal
bilingualism
Joshua A. Fishman
1. Times and tides in the study of bilingualism
There is no better place to begin this discussion than to marvel at the Zeitgeist
change during the past quarter century vis-ä-vis considerations of societal bi-
lingualism. In the mid-sixties to the mid-seventies the climate of opinion was
still overwhelmingly a negative one. My earliest efforts to provide a more bal-
anced perspective on immigrant-based bilingualism in the USA (Fishman
1966) were met with substantial incredulity. The two favorite "flies" in the bi-
lingual ointment were (a) purportedly lowered intelligence and (b) penalized
foreign accentedness. When I demonstrated that there was no necessary link
between intelligence and bilingualism (Fishman 1965) and that everyone al-
ways spoke their local languages (whether LI or L2) with some kind of accent,
the demonization of some accents being a form of rank social prejudice, this
was considered to be an admission of "guilty as charged", rather than a clarifi-
cation of the issue. I think I may have been among the very first (and perhaps
actually the very first modern day researcher) to point out that the bulk of the
human population was and had always been bi- (or multi-) lingual and that
elitist bilingualism was and had always been considered a badge of refinement
and lifelong advantage, rather than a stigmata of poverty and marginality. We
have come a long distance since those days, with the pluses of bilingualism fi-
nally being given the attention that they fully deserve. The journals devoted to
this area, the conferences, the academic programs and this very book itself are
all evidence that there has been a sea-change in our understanding of the indi-
vidual, societal and cultural ramifications and correlates of bilingualism. Bilin-
gualism really is a many-faceted human phenomenon, some of its aspects (as is
also the case of all human phenomena) being inevitably selectively perceived
as more positive than others.
16 Joshua A. Fishman
2. Bilingualism as a sociocultural norm
We owe to the sociolinguistic enterprise the realization that bilingualism is not

only an individual or even a social class phenomenon, but that it may very well
be the sociocultural norm. I am using "norm" here not only in the sense of com-
monality or frequency, but also in the more interpretive sense of normality or
desirability. A great founding father of modern sociology, Emile Dürkheim,
recognized that every social norm had within itself not only an indication of
what was considered socioculturally "expected" of members in good-standing,
but also a behavioral aspect according to which members where judged to be
good, proper, decent, i.e., morally proper or improper according to sanctified
traditions and standards. Dürkheim realized that exactly what it is that is con-
sidered holy may well differ from one culture to another, but believed that in
every case it is related to normified behavior which expresses notions of the ul-
timate "good" or "desirable" in life. It should come as no surprise then, that
languages, constituting the main symbol systems of all human cultures every-
where, have often come to be considered holy too. As a result, in many bilin-
gual settings, one language or even both have had this supreme attribute as-
cribed to it (or them) by the members native to that setting and incorporated
normatively and moralistically into their communicative conventions. Dürk-
heim would say that not only is bilingualism implicitly holy wherever it corre-
sponds to the sociocultural norm, but that in many of these settings it involves
one or more languages that are explicitly holy too. The latter settings are the
ones that we will particularly explore in this paper.
3. Types of holiness
Notions of holiness differ from culture to culture. English, because of its hybrid
linguistic nature, is one of a very small sub-set of languages that has two ety-
mological roots pertaining to this notion. The words holy and sacred, from the
Germanic and the Romance components of English respectively, give us two
different glimpses of what this term can connote. Holy implies wholeness,
completeness, perfectedness, hallowedness, i.e., a specialness far above and
beyond any human approximation to such goals. Sacred, on the other hand, is
more explicitly otherworldly in its conceptualization, referring to sanctity,
saintliness, godliness. The triliterate root for the Hebrew term that is glossed as
"holy", K-D-Sh, reoccurs in meanings that imply separated off, set aside as (or
dedicated to being) special, extraordinary. Accordingly, when languages are re-
"Holy languages" in the context of societal bilingualism 17
ferred to as "holy" we must, strictly speaking, investigate the semantic (rather

than only the etymological) field of the original term that is translated into Eng-
lish as "holy", in order to find out what exactly the indigenous imagery implies.
Although we cannot pause here to do that in each case, we can at least agree
that "holy is as holy does", i.e., that in each case it may mean something some-
what different (i.e., connote or denote different attributes and be applicable to
different referents, situations or circumstances). Nevertheless, there is suffi-
cient commonality of meaning underlying the term that we can not only utilize
it cross-culturally, but even suggest reoccurring clusters of meanings or associ-
ations that the designation has in cultures around the globe. There is no human
culture without language and no human culture without the notion of the holy.
As a result, these two notions co-occur and are joined in many cultures. Nat-
urally, they also co-occur in the writings of Dürkheim, although viewed from
his more theoretical perspective that does not even require that the notion of
sanctity be perspectivally present to members of any given culture.
4. Types of holy languages1
4.1. When one of the two languages is considered inherently holy
All religions derived from Judaism, the so-called "Abrahamic religions", rec-
ognize as holy the language(s) of revelation and of their respective holy scrip-
tures. In each of these cases the language (or the "variety" of the language) that
is revelation and scripture related is by now (and has been for over a millen-
nium) a "religious classical", i.e., a variety not employed for quotidian ver-
nacular purposes and, in most instances, no longer employed for quotidian
written purposes either. Other such languages are Classical (i.e., pre-Christian)
Greek, Sanskrit, Geez, Pali, both Classical Tamils and Classical Mandarin, all
of which can also be related to purportedly revealed and hallowed texts or to in-
itially oral traditions that may only have been committed to writing centuries
after their oral initiation. These languages are holy because of having been the
vehicles of materia sancta, whether orally or in writing. Their very structure
and corpus is assumed to have been transformed or shaped by their unique
"holy vessel" function. Like all holy phenomena these languages are consi-
dered unalterable, just as are their texts. This fixed-once-and-for-all nature of
religious classicals ultimately contributed to making them unsuitable for daily
secular life, even assuming that they had been so used before their revelatory
role began. The obviously by-product of this situation is the necessary bilin-
gualization of a sociocultural aggregate, one of whose languages is strongly

sanctity linked, while the other is not. No genetic relationship between the two
is necessary, although is often present, as Ferguson (1959) has demonstrated.
4.2. When another language becomes holy by dint of association with the one
that is already holy
Precisely because holy languages either never were used for everyday speak-
ing, reading and writing, or have not been used in such functions for centuries,
other community languages have had to be put to use to teach, translate and ex-
plain the languages of direct revelation, as well as any other hallowed texts
originally and most significantly identified with them. It is in this fashion that a
progression of languages of everyday use can come to partake of holiness, be-
cause they are the means whereby subsequent generations have had materia
sancta translated, taught and explained to them. Even Latin and Ecclesiastic
Greek may be primarily in this category of "holiness one step removed", de-
pending on just how much of the Vulgate and the early Greek Bibles are be-
lieved to contain the actual words of Jesus and of his disciples, rather than
merely subsequent translations of those words. The Old Church Slavonics defi-
nitely belong in this second category, as does "Luther German" for the Hutte-
rites, Old Order Amish and Mennonites (and for many Lutherans too) and as
does Yiddish for Talmud study among almost all Khasidic Jewish groups. This
is one of the processes by which the universe of holy languages grows. The
Protestant Reformation inspired and required vernacular translations of the
Bible and of the books containing the church services. In so doing, the sancti-
fied status of English, German, and the Scandinavian languages was substan-
tially contributed to (but long maintained only in the case of German). Such
growth is probably inevitable for all successfully continuous religious cultures.
However, the "sanctified by association" languages (or the varieties most di-
rectly associated with materia sancta translation, e.g., The King James version
of the Bible) also come to resist further change and, therefore, can be replaced
for everyday use by varieties or languages that are not yet significantly relig-
iously impacted. Once again, in such circumstances the sociocultural patterns
involved are, of necessity, bilingual. In such bilingual circumstances, one
spoken language will be sanctity linked whereas the other will not (or, at most,
will be far less so).
4.3. Holiness via ethnic kinship identity, life / death metaphors or other very
special experiences and associations
Finally, more and more vernaculars are coming to be regarded as holy due to
their association with heightened ethnic contrastivity, historical ethnic griev-
ances and traditional ethnic identity. Such developments are often communally
patterned, rather than merely individual experiences, and, as such, frequently
prompt positive ethnolinguistic consciousness for the "ethnic mother tongue",
even long after an intrusive language has been adopted for everyday life. The
struggle for the life of the ethnic collectivity (or any other collectivity that
comes to be connected by a kinship myth or sentiment) begins to make use of
the metaphors of human life and death, Since the latter are linked to sanctity
(for example "life everlasting", "undying love", "immortal soul", etc.), the
former too is discussed in such sanctity suffused terms. Similarly, expressions
of endearment and intimacy, usually reserved for the closest kin, putative kin or
kin-to-be, are transferred to the ethnicity-linked language, even when (or par-
ticularly when) its secular use has become attenuated and many of its erstwhile
workaday functions are being discharged by another language still linked to the
hurly-burly of daily life. The tendency toward psycho-ostensives is particularly
noticeable among recessive ethnicity-linked languages (Matisoff 1978) and, so
also, is the tendency toward calls for acts of moral obligation and for the dis-
charge of duties due to one's own dearest and nearest of kin. Metaphors per-
taining to the beauties of nature, to the return of spring, to the inevitable waxing
of the moon, the reappearance of seasonal flowers and animals, are all part and
parcel of the tendency to romanticize and to sanctify the "ethnically own" lan-
guage in comparison with the workaday one more recently borrowed from the
outside. As a result, we are once again faced by bilingual populations in which
one language is viewed (consciously or unconsciously) as much more within
the pale or penumbra of sanctity than the other. It is in precisely this sense that
for "true believers" Irish is holy whereas English is not, Basques is holy and
Spanish is not, Breton is holy and French is not, Frisian is holy and Dutch is
not, Ainu is holy and Japanese is not.
5. When both languages are holy
When the language of materia sancta per se and the language in which sanctity
is learned, understood and appreciated are both considered holy, there may be
cultural recognition of greater holiness and lesser holiness. Hebrew and Yid-
dish are often viewed in precisely this fashion among many Orthodox Jews.
Both Rabbi Joseph Soloveitchik and the Grand Rabbi of Lubavitch have
pointed out that while Yiddish is not as holy as Hebrew is, nevertheless it too
has a very definite and vital sanctity of its own. Urdu is so enriched by Persian
and Arabic that Urdu too basks in the sanctity that emanates from the former.
This recognition may even be more spontaneous in Orthodox Christianity (e.g.,
Old Church Slavonic and Russian, Old Church Bulgarian and Bulgarian, Ec-
clesiastic Armenian and Armenian, Ecclesiastic Greek [Katherevousa] and
Greek, etc.) and in Hinduism (Sanskrit and Hindi, Sanskrit and Bengali, San-
skrit and Gujerati, etc.), where both languages are etymologically related. This
is also the case in much of Islam (Koranic Arabic and Modern Arabic) and Ju-
daism in Israel (Classical Hebrew and Modern Hebrew) and was also the case
in the Western Church before the Latinate-vernaculars became fully secular-
ized (e.g., French and Latin, Spanish and Latin, Portuguese and Latin and, of
course, Italian and Latin). The use of the same "Chinese" writing system
throughout much of China has transferred some of the holiness of Classical
Mandarin to Han and even to some non-Han languages as well.
6. When neither language is holy
Bilingualism today is spreading most rapidly and most significantly due to the
spread of Western languages of modernization and globalization. English, e.g.,
is now far from being considered holy (even though it may have been so con-
sidered briefly, at the time of the founding of the Church of England and the
publication of its Book of Common Prayer). As a result, when English forces
its way into daily technological, consumerist and youth culture on all conti-
nents, its appeal in Cuba, Indonesia, Taiwan, Madagascar and the Philippines
has as little to do with sanctity as does the staying power of the locally spoken
vernaculars. The entire notion of the sanctity of a language is still more fre-
quently Western than non-Western, when the world as a whole is considered.
Nevertheless, most of the growth in language sanctity claims has occurred in
the non-West within the past half century or so. Indeed, the very idea of the
sanctity of the local language often seems to be adopted in almost pristinely
Herderian terms. In resisting the Western-derived erosions of local life and
identity, the local language comes to be viewed as "the genuine spirit of the
people", a "mystic essence" enabling them "to remain true to their own way of
life", a "sacred trust and responsibility", a "key to preserving their own au-
thenticity", etc. However, there are still huge sections of the globe where no
such appeals are known, even though bilingualism may be quite widespread.
Africa, Latin America and the Pacific provide most examples of such totally
secular bilingualisms, whether in connection with the spread of Western or
local contact languages.
7. The consequences of language sanctity beliefs
Language sanctity convictions do not operate in a vacuum. Whether or not they

are accompanied by parallel notions of sanctity of the people and/or sanctity of
the land, language sanctity convictions have mobilizing, rallying, activating,
rousing, energizing potentials. These potentials are employed in conjunction
with collective efforts on behalf of democratization through local vernacular
status elevation, language maintenance consciousness raising, reversing lan-
guage shift and language corpus modernization. The practical utility of lan-
guage sanctity convictions in the political arena has probably also contributed
to the steady growth in such convictions, sanctity being every bit as much so-
cially constructed as other societal beliefs and convictions. However, although
language sanctity notions may help activate popular ethnic consciousness,
these very same convictions also tend to complicate or restrict language mod-
ernization efforts, constraining them to be more "authentic", purist, ausbau and
indigenous in orientation rather than following along the lines of "inter-
nationalization" (adopting paradigms from English and other Western lan-
guages) which corpus planning for modernization normally pursues. Modern-
ization movements and ethnic authenticity movements often co-occur among
bilingual populations. These populations may well become conflicted as to
which of their two languages should have the upper hand and just how holiness
and power are to be separated.
The result of ethnicity driven "introspective" or "self-distancing" corpus
planning is to make the two languages of bilingual speech communities less
similar, counteracting the common tendency for languages in contact to inter-
act with each other, not only lexically but also phonologically, grammatically
and semantically. Furthermore, within the sanctity-impacted language, the va-
rieties that are written are likely to be most stringently regulated (by schools,
churches and governments) and, therefore, most puristic of all and most distant
from "the other language" of bilingual sociocultural aggregates. Of course,
where both languages are considered holy in some way, their self-distancing
efforts will be mutual and reciprocal, at the same time that a complementary al-
location of functions exists between them to avoid societal conflict. When such
a complementary allocation of functions falls apart (e.g., when both languages

seek governmental, educational or modern written functions), then a longer or
shorter period of conflict must ensue until a monolingual victor or a new com-
plementary bilingual distribution is established.
8. The staying power of sanctity
More than fifteen hundred years after the collapse of the Roman Empire, Latin
continues to be learned and used - by bilinguals who already speak, read and
write some other language(s). Even more miraculous is the staying power of
Hebrew within a dispersed and persecuted minority culture, all of whose
members spoke - and many of whose members also read and wrote - other lan-
guages (some for within-group and others for between-group purposes). Be-
cause sanctity is upheld by and linked to major and powerful societal institu-
tions, it is not a "here today and gone tomorrow" phenomenon. It quite under-
standably has extraordinary staying power and this same power is transferred
to the languages most intimately suffused by it. Languages of direct sanctity
live on in their holy texts and, as such, are not subject to the winds of change
and of influence (or interference) from other languages in the way that spoken
languages are. While it is not the case that they do not change at all (as a result
of the fact that they come to be written by individuals who are thinking in an-
other language), it nevertheless remains true that the holy texts themselves con-
stitute a mighty "corrective" pulling all subsequent texts back into their struc-
tural orbit.
But the staying power of sanctified languages within bilingual repertoires is
also noticeable in yet another way: they do not come and go the way quotidian
vernaculars do. They wax and wane and have a seemingly phoenix-like capac-
ity to arise again out of their own ashes. This latter capacity too is a reflection of
the sanctity attributed to them. Every sociocultural system has its own frus-
trations, disappointments and defeats. Religion is a system of comfort, reset-
ting of priorities, decentering of one's own importance in the grander scheme
of life and recentering on a greater design and a Supreme Designer. As a result
of such recentering experiences, there have been cycles of "return to religion"
or of "renewed fundamentalism" throughout human history. During such
"renewal" or "rebirth" periods the original sacred texts are "rediscovered" and
the languages of these texts, seemingly peripheralized, lost and/or forgotten,
are again focused upon, studied and pondered by growing numbers of adher-
ents. Although almost never revemacularized (because the bilinguals who read
and study these texts already have perfectly functional vernaculars), their most
striking expressions, figures of speech and turns of phrase are frequently
adopted by their accompanying vernaculars, thereby adding another touch of
sanctity to the latter as well.
9. Conclusions
All in all, the realm of sanctity is doubly influential in connection with socio-
linguistic processes: (1) it both maintains and renews the vitality of the lan-
guage most impacted by religion and (2) it appreciably influences the accom-
panying language as well, if for no other reason that the speakers of the latter
are exposed during so many hours, days and years to the texts, translations or
elucidations of the former. This process of "sanctification of the vernacular by
dint of long term association with the language of materia sancta" is commonly
much stronger than the process of "secularization of the sacred". Obviously,
some of the latter also may occur over time, given that the languages of the
same community can rarely be fully compartmentalized, and particularly so if
the language of religion also comes to attain widespread secular reading and
recording functions. On the whole, however, secular vernaculars have come
and gone throughout human history, whereas religious classicals and the ver-
naculars that they have influenced most thoroughly (e.g., the "Luther Bible" of
the trilingual Old Order Amish) seem to dig-in and hang-on almost "eternally".
The full or partial sanctification of one or more languages within multilingual
societies is a factor which contributes to the stability of societal bilingualism
per se, to the longevity of language movements world-wide and to authenticity
strivings within corpus planning more generally. There is a particularly high
likelihood that the Η variety within diglossic societies will be religion related
and that more learned realizations of the L variety will also be implemented by
borrowing lexical and grammatical items from the H.
The secularized monolingual sociocultural world occupies a distinct minor-
ity status in the world at large, in which religion impacted societal bilingualism
is the quantitative norm.
Note
1. Numerous examples of languages involved in each of the three types of societal bi-
lingualism discussed below, in which holiness is ascribed to one or both languages,
may be found in Fishman (1996). The section on Sanctity in that book is organized
in a manner that closely follows the organization of the present paper. There the
reader will find scores of references to sanctity-claiming quotations pertaining to
languages from all parts of the globe.
References
Ferguson, Charles A.
1959 Diglosia. Word 15: 325-340.
Fishman, Joshua A.
1965 Bilingualism intelligence and language learning. Modern Language
Journal 49: 227-237.
1966 Language Loyalty in the United States. The Hague: Mouton.
* 1996 In Praise of the Beloved Language: The Comparative Study of Positive
Ethnolinguistic Consciousness. Berlin/New York: Mouton de Gruy-
ter.
Matisoff, James A.
1978 Blessings, Curses, Hopes and Fears: Psycho-Ostensive Expressions.
Philadelphia: ISHI Press.
Questions for discussion
1. Why would sociologists or sociolinguists who are followers of Emile Dürk-

heim tend to believe that all speech-communities possess one or more holy
languages?
2. Why do many speech communities believe that some languages are holier
than others?
3. What effect does the conviction that one or more of its languages is/are
holy have upon the rate and degree of change in that /those languages?
4. What effect does the conviction that one or more of its languages is / are
holy have upon the direction of change among other languages used by the
same community?
5. Why does the number of languages considered holy by their respective
speech communities continue to grow?
Forlorn hope?
John Edwards
1. Introduction
From Talavera to Waterloo, Richard Sharpe's tumultous military career has

been traced by Bernard Cornwell in a series of novels and television pro-
grammes. One of the most adventurous exploits of this fictional hero of the
Peninsular War was leading the "forlorn hope" in the assault on Badajoz in
1812 (Cornwell 1983). The term refers to a body of volunteers, whose often
suicidal task was to be the first to storm the breach. Showing a courage some-
times born of desperation, those who lived were rewarded in two ways. All sur-
vivors were praised, of course, and their courage marked them for life; in some
regiments, laurel-wreath sleeve insignia were awarded. More tangible was im-
mediate promotion: the commanding lieutenant became a captain, and his ser-
geants were made ensigns. A "forlorn hope" has now achieved a degree of
popular usage, roughly signifying any undertaking or expectation that is un-
likely to be fulfilled. But there are further points of interest here.
The term derives from the Dutch verloren hoop, "the lost troop". (The Ger-
man equivalent is the same, verlorener häufe, while the French enfants perdus
reminds us directly of the derivation of the word infantry.) Through misattribu-
tion or folk etymology - but a particularly apt sort of error, in this case - Dutch
hoop becomes English "hope". (The old mariners' version, with a mispronun-
ciation no doubt influenced by their environment, was "flowing hope".) The
hoop / hope confusion thus links a group of men who are usually without
hope - hoop translates literally as "heap", which itself seems tragically appro-
priate in the circumstances - to any sort of lost or unpromising cause, to peril-
ous enterprises, even to losers at the betting table. The broadening of the initial
military meaning dates to at least 1500, and it is this extended sense that re-
mains with us now - faint hope or, indeed, "hope against hope".
Most modern usage, in fact, suggests less than faint hope. Nowadays, if you
describe someone as entertaining forlorn hopes, you most probably mean that
they are fooling themselves, that their expectations are groundless, that there is
no chance of success at all. So, through inaccurate etymology and word-play,
and with the passage of a couple of centuries, a dangerous undertaking with
26 John Edwards
slim likelihood of success has become a hopeless or unwarranted one: a foray

from which only a few will return has become a kamikaze mission.
We have, then, an intrinsically interesting example of cross-language
transfer, of how an incomplete bilingualism leads to misunderstandings - even
if, as in this instance, the misapprehension itself is meaningful. The move from
verloren to "forlorn" was a reasonable enough translation - although not, of
course, a perfectly accurate one: by the fourteenth century, the English word had
already acquired the additional sense of "sad", "miserable" or "wretched".
Readers whose taste runs more to Dickens than to military fiction will remember
the depressed Mrs Gummidge (inDavid Copperfield): "I'm a lone, lorn creetur,
and everythink goes contrairy with me." But the transition from hoop to "hope"
was linguistically baseless, if an appropriate enough error in other ways. Such
etymological matters are of historical, social and linguistic interest - and since
derivations (and their fleshing-out along literary and other lines) generally in-
volve more than one language, they represent a type of bilingual study. But I
begin with this little discussion of forlorn hopes for another reason.
The argument here is that some sort of bilingual accommodation is now the
best hope, the most reasonable hope - but possibly also the forlorn hope - of
those concerned with the fate of small languages in a world made increasingly
safer for the large ones. And, if such accommodation can be seen as a forlorn
hope, then we must further ask if, in this connection, the term signifies outside
chances or none at all.
2. The ecological thrust
Any current interest in small languages and their fates is contextualised in the
three great modern themes of sociolinguistics and the sociology of language.
These are the relationship between language and identity, endangered lan-
guages and the new ecolinguistics (which often, though not inevitably, involves
a concern for language rights). All are obviously intertwined.
When, in 1985,1 published a book called Language, Society and Identity,
the final word of the title was not nearly so widely used in the literature as it is
now. For a long time, of course, there has existed a body of work dealing with
identity, with ethnicity and nationalism, with "groupness" in all of its forms.
Within the last few years, however, identity per se has been highlighted to an
unprecedented extent, and the explosion of interest has occurred across a broad
swathe of academic disciplines. From sociology, political science, anthropol-
ogy, history, language and linguistics, psychology and education has come a
Forlorn hope? 27
continuous stream of papers and books treating identities in every context, and
with all their ramifications. And it is not just from within social science that this
treatment has emerged. Even faster-flowing, perhaps, is the current whose
source is in literary and cultural studies - more a torrent than a stream here,
often in the full fig of deconstructed, post-modernist glory. Finally, it seems as
if every social or political philosopher is now writing about the accommodation
of "groupness", trying to reconcile (for example) collectivist demands for the
legal, political and social recognition of group identity with the more tradi-
tional liberal-democratic adherence to individual rights.
This eruption of interest is not, of course, a random occurrence. Our age is
one of transition and transitions are, almost by definition, painful. The break-
up of the Soviet Union, the protracted unravelling of Yugoslavia, the other re-
alignments in eastern Europe, the march towards federalism in a European
Union which is also increasingly asked to recognise a "Europe of the Regions",
the continuing agonies in Africa, and the emergence and arrangement of the
Asian Pacific power-house are all illustrative. So, too, are the politics of multi-
culturalism and accommodation in states like Canada, Australia and the United
States, countries in which the interplay between indigenous and immigrant
populations has come to be just as important as that among immigrant groups
themselves, or that between immigrants and "mainstream". When we consider
these sorts of settings, we see contexts in flux, languages and cultures in con-
tact, identities up for negotiation or renegotiation. Identity is very much on the
global agenda.
To talk of identity is to talk of culture and, perforce, language; to talk of
identities in flux is often to talk of languages at risk. The academic response to
endangered languages remains mixed, but changes are visible. Twenty years
ago, the late Dwight Bolinger (1980) commented upon the traditional reluc-
tance of linguists to engage in what could be called the "public life" of lan-
guage, a reluctance based upon, first, the feeling that linguistic evolution - even
to extinction - was somehow a "natural" process that ought not to be interfered
with and, second, the belief that intervention would, in any event, be unsuc-
cessful. Among linguists whose particular province is "small" languages, how-
ever, engagement can often be assumed. A recent example is Joshua A. Fish-
man (e.g., 1992) who has, more unambiguously than most, laid his cards on the
table, and argued for academic commitment to the conservation and mainten-
ance of cultural and linguistic diversity.
In the same year, Michael Krauss made a more pointed argument, one
which continues to set the tone for much contemporary debate. Linguists, he
noted, will be "cursed by future generations" if they do not actively intervene to
stem the "catastrophic destruction" now threatening nine out of ten of the
28 John Edwards
world's languages (1992: 7-8). More traditional linguistic documentation is

seen to be insufficient; social and political action and advocacy are required. A
response by Peter Ladefoged, arguing for a continuation of the linguist-as-dis-
interested-scientist role, ensured that the debate would continue: "one can be a
responsible linguist," he said, "and yet regard the loss of a particular lan-
guage (...) as far from a 'catastrophic destruction'" (1992: 810). A third par-
ticipant in this initial Language debate was Nancy Dorian, who pointed out
that, at the very least, this is an "issue on which linguists' advocacy positions
are worth hearing" (1993: 579; fuller details here can be found in Edwards
1994). We have, indeed, heard more and more of these positions, and important
contemporary collections include those of Dorian (1989), Robins and Uhlen-
beck (1991), Brenzinger (1992, 1998), Grenoble and Whaley (1998a) and
Nettle and Romaine (2000); as well, a useful overview has been written by
Crystal (2000).
One of the most important of the recent positions is embedded in what is
called the "ecology of language". As a term and a focus of study, ecology is a
coinage of Ernst Haeckel (perhaps as early as the 1860s) and, as its Greek root
(oiicos = 'home') implies, the emphasis is upon the holistic study of environ-
ments within which lives are lived and intertwined (for useful discussions pre-
dating the general extension to language, see Bates 1953 and Glacken 1967).
Haeckel was concerned with what Darwin had called the "conditions of the
struggle for existence" within the "web of life", stressing, therefore, both the
"beneficial" and the "inimical" relationships linking plants, animals and, in-
deed, inorganic surroundings. Ecology is about adaptations, then, whose
necessity arises from inevitable linkages. Implicit in the earliest conceptions of
this "economy of nature" is the scientific investigation of the "natural condi-
tions" constituting environments and, from this perspective, ecological aware-
ness (broadly speaking) is very old. It can be traced, for instance, at least to Ar-
istotelian concepts of "design in nature" and the idea that the world is essen-
tially ordered, it attracted further philosophical and religious elaborations
(which argued, for example, that God was the designer), and it underpins con-
temporary secular science which replaces divinity with natural laws. We are
now all well accustomed to hearing about ecology, as well as offshoots like
ecosystems and ecospheres.
A predictable extension of a concept which initially focussed upon plants
and animals involved culture. An ecological anthropology thus folded culture
into the mix, reminding us of the reciprocity between what is given and what is
constructed (pre-ecolinguistic sources here include Steward 1955 and Clifton
1968). Again, what is sometimes taken as a modern idea has longstanding
roots: Plato and Aristotle, after all, thought that climate was an important factor
Forlorn hope? 29
in human affairs and, in the eighteenth century, Montesquieu built an elaborate

philosophy on this basis, a philosophy linking climate (and topography) to all
manner of individual and collective traits - from exploratory activity to re-
ligion.
Apparently, the first specific reference to the ecology of language is found
in a chapter by Voegelin etal. (1967), but the term is particularly associated
with Einar Haugen (see his 1972 collection). His intent was to emphasise the
interconnectedness of languages with their environments, with particular re-
gard to status and function, and he produced a list of ten contextualising ques-
tions - about who uses the language, its domains, varieties, written traditions
and family linkages, the degree and type of support it enjoys, and so on (for
fuller details and typological expansions, see Edwards 1992, discussed further
by Grenoble and Whaley 1998b). These are, in themselves, neutral in tone.
However, in a book forthrightly called Blessings of Babel, Haugen - who had
earlier published a piece on the curse of Babel (1973: a shortened version ap-
pears in the book) - gives a hint as to his own preferences. Commenting on the
desirability of bilingualism in America (this is also Haugen's solution to the
"curse", incidentally: he was not, after all, simply trying to cover all fronts at
once), he argues that this is a "problem of social ecology: keeping alive the va-
riety and fascination of our country, diverting the trend toward steamrollering
everything and everyone into a single, flat uniformity" (1987: 11). A dislike of
a monotonic landscape is clear, although the quotation itself is not. Haugen
probably did not mean to imply that "social ecology" was essentially devoted
to the promotion of diversity; perhaps he meant only that any such promotion
would fall within its remit (see also his 1972 collection). In any event, a broadly
sympathetic view of linguistic diversity need not extend to blanket endorse-
ment in all circumstances.
I now wish to turn more directly to bilingualism, to consider its role within
an ecological perspective. A more thorough job here would necessarily involve
fuller analysis of that perspective - one that is in many ways unsatisfac-
tory - but this must be largely put aside for present purposes (but see Edwards,
in press). As conclusion to this section, and introduction to the next, I can make
only two points.
The first is that the breadth of the ecology-of-language view, a breadth that
would logically follow from its parent discipline, has been progressively re-
duced and the label of ecology increasingly co-opted. Ecology, to go back no
further than Haeckel's formulation, involves adaptation and struggle within re-
lationships that can range from the "beneficial" to the "inimical". In this sense,
earlier non-interventionist linguistic views - now sometimes discredited on the
grounds that it is wrong to simply stand aside and watch - were fuller than later
30 John Edwards
ones, since the former often acknowledged a Darwinian sort of linguistic

struggle. (This is not, incidentally, to be taken as a broad endorsement of non-
intervention per se. I make the observation only in passing, as it were.) While
there are some contemporary researchers who would claim an ecological per-
spective that reflects a range of possibilities (from linguistic health all the way
to extinction, perhaps) - see the recent chapter by Mufwene (2000), for in-
stance - the field now generally argues for more pacific interaction. As
Mühlhäusler (2000: 308) has noted in a recent review article, "functioning
ecologies are nowadays characterized by predominantly mutually beneficial
links and only to a small degree by competitive relationships (...) metaphors of
struggle of life and survival of the fittest should be replaced by the appreciation
of natural kinds and their ability to coexist and cooperate." As in ecology writ
large, so in the ecology of language. We have a view of a world in which there
is room for all languages, where the goodness of diversity is a given, where
"the wolf also shall dwell with the lamb." This is certainly a kinder and gentler
picture, but surely the key word here is "should", surely the key question is
whether the desire is also the reality. We could remember Woody Allen's re-
working of that passage from Isaiah: "the lion and the calf shall lie down to-
gether, but the calf won't get much sleep." In brief, the "new ecology" is predi-
cated on a number of assumptions, argues for an association between biologi-
cal and linguistic diversity, and is motivated chiefly (as Mühlhäusler and others
have observed) by the accelerated loss of languages. It might be argued that the
ecological idea, as applied to languages, is largely of metaphoric relevance, but
it is certainly a more tangible concept in many eyes. It is now a watchword with
increasingly specific connotations, a concept co-opted, an agenda particula-
rised.
That loss of linguistic diversity noted above underlies, for example, many
organisations that share the watchword, that proclaim an ecological outlook
and that advocate some form or another of active intervention. And, to return to
a point implied at the beginning, science is typically allied with a sense of
morality, insofar as linguistic conservation, preservation and maintenance are
considered to be basic human rights. We can briefly note here that an ecological
awareness translated into action means leaving the academic cloisters and
implies intervention in the lives of people who, in many instances, will be at
some social, political or economic disadvantage. It is this potential application,
of course, that makes close examination of the area so important. Some of the
better-known organisations here are the Endangered Language Fund, the Com-
mittee on Endangered Languages and Their Preservation, and Terralingua:
Partnerships for Linguistic and Biological Diversity (based in the United
States), The Foundation for Endangered Languages (in England), Germany's
Forlorn hope? 31
Gesellschaft für bedrohte Sprachen, and the International Clearing House for
Endangered Languages (Japan). Similar concerns motivate the European pro-
grammes of Linguasphere and the Observatoire Linguistique (Dalby 2000) as
well as those whose more pointed purpose is language-rights legislation. Fuller
details may be found in Crystal (2000) and Maffi (2000).
The second point - and the one that returns us to the central theme here - is
that the desire for language diversity that unites many under the ecological
banner is not an argument for monolingualism. Those who wish to see an ex-
tended future for the small languages of the world no longer propose - if, in-
deed, they ever seriously did - that Welsh or Frisian, Haida or Mi'kmaq, will be
sufficient for their speakers' purposes and aspirations across all domains. Ar-
guments now are not for existence per se but, more precisely, for the coexist-
ence of the lions and the calves.
3. Bilingual accommodations
Within an ecological thrust that aims to support faltering varieties, the empha-
sis upon bilingualism as a longer-term solution for small and threatened lan-
guages is clearly a key component. Benton (1981), for instance, accepts that a
bilingual accommodation with English is the best that Maori can now hope for.
This is just one specific example of a more general acknowledgement. Thus,
Mühlhäusler (1996) argues for what he calls an "equitable" bilingualism. Maffi
(2000) notes the desirability of a stable and non-subtractive bilingualism that
links mother tongues with languages of wider communication. Wurm says the
same, endorsing the "possibility of speakers of (...) endangered languages
being bilingual in their own language and a given large metropolitan language"
(1998: 194). And Crystal observes that the duality is "perfectly possible,
and (...) highly desirable. Because the two languages have different pur-
poses - one for identity; the other for intelligibility [sic] - they do not have to
be in conflict" (2000: 29); later in his book, Crystal expands upon the modus vi-
vendi offered by "healthy" bilingualism - i.e., one in which languages are com-
plementary and not competitive. These ideas are exactly in tune with the newer,
gentler ecology.
Bilingualism is, in fact, a reasonable solution - reasonable on theoretical
grounds, and reasonable in the de facto sense that more people in the world are
bilingual (or better) than are monolingual. It is also worth restating the truisms
that bilingualism per se involves no cognitive penalties, that it is a capacity that
can be almost effortlessly acquired by the youngest of children, and that the
lack of education (and, indeed, illiteracy) of most of the world's multilinguals
32 John Edwards
further attests to its unexceptional character: bilingualism is not rocket science.

It is a reasonable and practical response to diversity; it is, indeed, an obvious
one. The question here is not whether bilingualism can allow people to enjoy
that social and psychological duality noted above - the answer is clear. The
question is whether bilingualism will have this function, particularly where
small languages are involved.
Bilingualism as solution is embedded in an ecological perspective which
makes a number of assumptions. Two of these are general enough that, inter-
twined, they form a backdrop for everything else. Linguistic diversity is, of
course, taken as an unalloyed good, to be defended wherever it seems to falter.
It is an outcome implicit in recent language-rights manifestos, covenants and
declarations. These have a chequered provenance, make several sorts of lin-
guistic claims, and have received various degrees of official response (see
Brumfit 2001; Maffi 2000 and Crystal 2000 on an important topic which, again,
I shall discuss more fully elsewhere). Apart from the legalistic approach of for-
mal proclamations, there are several bases upon which a defense of diversity
can rest, and Mühlhäusler (2000) has summarised these in moral, scientific,
economic and aesthetic terms.
The morality of diversity - beyond, that is, assumptions of inherent rights
which are, themselves, predicated upon a moral argument - suggests that lan-
guage attrition means loss of accumulated experience and knowledge. Sec-
ondly, it is argued that multilingual societies produce higher achievement, and
that linguistic "encounters" aid scientific advance. This in turn suggests that
language diversity is economically beneficial; it is noted, more specifically,
that emphasis placed upon lesser-used varieties will prove more worthwhile
than simply broadening the base of those who learn "big" varieties. The econ-
omic argument is itself broadened when diversity is seen to support identity
and "social healing" - these are among the "positive externalities [that] need to
be internalised and made part of economic argumentation about the cost of lan-
guage planning" (Mühlhäusler 2000: 334). Finally, an aesthetic appreciation
values all diversity, and regrets all loss.
These, as I say, represent Mühlhäusler's four-fold rationale. Although other
arguments altogether can be made, and although the substantiation of these
four themselves could be done along different lines, the assumptions made here
are quite typical. All are debatable. Have we lost the knowledge of the Greeks
and the Romans, for example? Is their "world view" gone forever? Perhaps
fairer examples would cite small languages with only oral traditions. If, for the
sake of argument, we accept that, when they go, they take all their cultural in-
sights with them, wouldn't this be a rationale for the promotion of liter-
acy - and, if so, how does this square with a frequently-expressed ecological
Forlorn hope? 33
view (see below) that a move to print is a move to danger for small commu-
nities? Scientific arguments that repertoire expansion involves enhanced intel-
lectual capacity are, to say the least, not proven, and it seems disingenuous (or
psychologically naive) to include under such headings the observation that "the
top achievers in the British university system (...) are increasingly students
with an Asian or continental European language background"(333). As for di-
versity per se aiding discovery - the western awareness of Chinese script "rev-
olutionised the enterprise of systematic knowledge"(333), we are told - well, it
could just as easily be said that language differences typically constitute a bar-
rier (though not, of course, a very solid one) to international exchange. Econ-
omic rationales for diversity are even harder to sustain; this is so at "macro" le-
vels, but social costs and social responses are, after all, built upon individual
coral. Mühlhäusler observes that "the value of a hundred expatriate economists
speaking Polish, Mongolian or Fijian may be much greater than the training of
an additional hundred speakers of Japanese or French"(334). No doubt true,
but - in the absence of special incentives - only as a reflection of perceived
need (which is to say, economic considerations). This seems to take us directly
back to those odious market forces and rational choices which are seen to fa-
vour Japanese, French and so on. Otherwise, who would opt for Mongolian
over French, who would voluntarily narrow their scope? Bill Mackey pointed
out, many years ago, that "only before God and the linguist are all languages
equal" (1978: 7), and the point is sharpest in crass economic terms. Finally, we
come to the aesthetic argument, one which holds diversity valuable beyond any
such crass instrumentality. It may seem churlish to advance any alternatives
here; indeed, on logical grounds, it may be pointless. De gustibus, and so on.
Still, educated arguments have in fact been made for the propriety and elegance
of uniformity. One recalls, for instance, that Matthew Arnold - whose most fa-
mous work was devoted to culture - had a deep interest in Celtic history and lit-
erature, an interest which coincided with a desire for the rapid disappearance of
spoken Welsh and the assimilation of all the Celtic peoples (see, e.g., Arnold
1883). And John Stuart Mill, whose arguments about representative govern-
ment have endeared him to those believing that national and state boundaries
should coincide, could also argue that "inferior and more backward" groups
(he mentioned the Basques and the Bretons) would be better assimilated within
the "current of the ideas and feelings of a higher civilised and cultivated
people" than to remain some "half-savage relic of past times" (1964 [1861]:
363).
For the moment at least, I wish only to point out that these sorts of argu-
ments are eminently debatable, and that the logical defense of linguistic diver-
sity is not as straightforward as some might imagine. "Clear your mind of
34 John Edwards
cant," said Johnson to Boswell in 1783 - the word is italicised in the Life be-
cause Johnson said that, while it was often socially de rigueur to utter mindless
pleasantries, to trot out platitudes or, indeed, to be downright hypocritical, one
ought never to think insincerely. Good advice still, particularly in highly-
charged areas where fact and opinion overlap. More recently, and more specifi-
cally, we could recall Ladefoged's point that "statements such as 'just as the ex-
tinction of any animal species diminishes our world, so does the extinction of
any language' are appeals to our emotions, not to our reason" (1992: 10). Does
this suggest that the case for diversity has no basis at all? Quite the contrary.
But it is important, I think, to see on what base it really rests and this, I think, is
clear - even if it is not always presented clearly or directly. It is a base con-
structed of perceptions of morality and aesthetic preference. These are the es-
sential animating articles of faith that underpin all ecological expression. The
first, which is itself predicated upon arguments about human rights or, some-
times, ius naturale, is of course more debatable than the second, where the non-
rational nature of de gustibus is its strongest element. (My own view, inciden-
tally, is to agree [as I almost always do] with Johnson who, on his famous tour
of the Hebrides [in 1773], said to Boswell, "I am always sorry when any lan-
guage is lost".) But we must surely aim to separate preference from evidence,
or - and this is the last spasm of my Johnsonian paroxysm - hope from experi-
ence.
And a final note here: rationales for diversity need not, in any event, have
the slightest relevance to actual policy or planning in a world where, as Kedou-
rie (1961: 125) observed, "it is absurd to think that professors of linguistics and
collectors of folklore can do the work of statesmen and soldiers (...) academic
research does not add a jot or a tittle to the capacity for ruling." This is an overly
blunt assessment, and one which ignores conduits of knowledge (which, ad-
mittedly, are often very sluggish and unpredictable in their flow) and possibil-
ities of persuasion. Still, apart from its useful nudge to modesty and caution, it
is perhaps a reminder that many arguments are academic in more than one
sense.
In the Irish revival effort of the late nineteenth and early twentieth centuries,
the close association of the language with Catholicism - Irish was, for
example, seen to be "the casket which encloses the highest and purest re-
ligion", and "the instrument and expression of a purely Catholic culture" (see
Edwards 1995:113-114) - involved two central elements. The first is the belief
that a particular language and a particular religion (or culture) are indeed
linked. The second - which logically flows from the first but which can cer-
tainly be exploited even in the absence of that strong initial belief - is that it
makes good sense to try and bolster a weak or disputed quantity (here, Irish) by
Forlorn hope? 35
linking it with a stronger cultural pillar (Catholicism). This is an analogy that

occurs to me when turning to the second broad ecological assumption of a con-
textualising or backgrounding nature: the linkage made between linguistic di-
versity and its biological counterpart. It is clearly valuable to proclaim this
linkage in a world increasingly aware of environmental issues; the advantages
of adding anxieties about language decline to concerns with pollution, loss of
plant and animal habitats and industrial depredation are obvious. But how rea-
sonable, in fact, is this addition?
At a metaphoric level, the association works well enough. Doesn't diversity
make the world richer and more interesting, and isn't linguistic diversity one
part of a larger attractive mosaic? And, if we intervene to save the whales, or to
clean up oil spills - or, indeed, to keep historic buildings from the wrecker's
ball, or to repair and preserve rare books and manuscripts - then why should
we not also stem language decline, ensure a future for all varieties, prevent
larger languages from swallowing smaller ones, and so on? A popular case
could theoretically be made here, even if based only upon (alleged) morality
and aesthetics. (Of course, by using the word "only" here, I don't mean to be-
little the force of either morality or aesthetics. I simply mean that, while it is ea-
sier to sway broad public opinion when more concrete arguments can be made
for some course of action, it can also be moved on these less tangible grounds.)
It is interesting, however, that recent ecological arguments have attempted
to make the link between linguistic and other types of diversity much more than
metaphoric. In one sense, this just returns us to those other rationales - scien-
tific and economic - already noted: after all, the case for saving whales and re-
ducing carbon emissions is not only made on moral or ethical grounds, but also
rests upon practical, forward-looking and economic arguments. But a rather
more pointed assertion has been made. Harmon (1996) has suggested that there
may be more than analogy between linguistic and biological diversity, that
areas in the world rich in one are also extensive in the other, that the two diver-
sities are "mutually supportive, perhaps even coevolved" (Maffi 2000: 175; see
also Mühlhäusler 2000). Apart from difficulties of measurement and interpre-
tation, however, we should be reminded here of the classic principle that cor-
relation need not imply causation. But, putting aside the dubious argument im-
plied in the quotation, there is in any event a practical problem that purported
linkages between animals and languages cannot overcome: it is much more dif-
ficult to maintain the latter than it is to preserve the former. I don't mean to say,
of course, that saving rare species is easy. But, when we have been able to
muster sufficient resources and to garner enough support, we have intervened
with some success in the lives of snails and whales. We have passed regulations
forbidding some sorts of hunting and fishing, and allowing stocks to recover.
36 John Edwards
We have forbidden the importation of materials whose removal damages the

environment elsewhere. We have outlawed clear-cutting and prescribed refor-
estation. It is clear enough that we haven't always done very well, that the en-
vironment continues to be harmed in important ways, and so on. But we have a
potential level of control here that is impossible with human societies and then-
languages - unless, of course, we were willing to act in the dictatorial ways
(benevolent though they may be) that are open to us with plants and animals.
It is on the stage set by these two assumptions - the essential goodness of
linguistic diversity, and its links with the organic world - that bilingualism-as-
solution is presented. Two basic elements are involved here: first, weaker va-
rieties may have to be stabilised (or rejuvenated, or revived); second, the dua-
lity of the smaller and the larger languages has to have some hope, at least, of
an enduring diglossic relationship. The first of these is always theoretically
possible, if not always practicable, and the second is problematic. Somali and
Guarani are examples of indigenous languages that have, in the face of power-
ful linguistic competition, undergone significant "development" and "modern-
isation" - success stories, indeed, compared to many others. Bilingual accom-
modations here have, consequently, also proved more longstanding than in
other settings. But it would be naive to think that such cases can be replicated at
will and it should be remembered that the standardisation of small varieties
does not preclude further depredations by large ones: Guarani and Somali re-
main much less useful, in broad perspective, than Spanish or English, and their
hold on the rural, informal and intimate contexts where they continue to thrive
ought not to be taken for granted. Bilingualism, as we know very well from all
sorts of situations, is often a way-station on the road to new linguistic arrange-
ments.
It is a linguistic truism that all languages can be adjusted as their speakers'
needs dictate. It is a social and political truism that, because of the fortunes of
their speakers, some languages are stronger than others. The language decline
that is at the heart of the issue here is often a symptom of contact between
groups of unequal power, an effect of a larger cause. It follows that attempts to
arrest it are difficult: you don't cure measles by covering up the spots. The logi-
cal - indeed, the ecological - approach is to unpick the social fabric that has
evolved, and which has brought about language decline, and reweave it into a
new pattern. But while this is possible, it is also difficult, inasmuch as those
concerned to stem decline typically want only some selected reworking of so-
cial evolution, not wholesale revolution. Overall, the implication is that a stable
diglossia between smaller and larger varieties is difficult to ensure, especially
in an era of change and mobility. Of course, if people are willing to remain in
an isolated condition - the isolation can be physical or psychological - the
Forlorn hope? 37
possibilities for long-term bilingualism are more propitious. Indeed, isolation

has not only stabilised bilingualism, it has very often accounted for the endur-
ance of monolingualism in the original variety. Twenty years ago, Fishman
noted:
Stable bilingualism and biculturism cannot be maintained on the basis of open and
unlimited interaction between minorities and majorities. Open economic access and
unrestricted intergroup interaction may be fine for various practical and philosophi-
cal purposes. Indeed, they strike most of us as highly desirable legal and social prin-
ciples; but they are destructive of minority ethnolinguistic continuity. (Fishman
1980: 171)
I have cited this quotation before, because it seems to summarise the situation
quite well. Here is an avowed supporter of minority languages, of stable bilin-
gualism, of "ethnolinguistic continuity", suggesting that the price of stability is
higher than most have evidently been willing to pay. Curious, to say the least,
that open access and interaction are seen as "fine" - these are the sorts of things
that people struggle and migrate for, not simply desirable principles. Would we
advocate voluntary minority-group self-segregation - as has appealed to a
handful of groups, generally along religious lines - or would we take steps to
ensure that Gaelic speakers never get off Cape Breton, remaining in their island
fastness for the sake of their language and culture?
Well, these lines of argument are well known, and fuller details can easily
be found. I want to conclude this section by referring to three arguments that re-
cent ecological writings have resuscitated. These are arguments intended to
bolster linguistic stability and the enduring bilingualism to which it is seen to
lead, but - in line with the block quotation above - we see that each involves a
price that may be too high to pay.
First, there is the view (touched upon above) that illiteracy favours stability.
Mühlhäusler (2000) has criticised arguments that link literacy to "empower-
ment", and has noted that its development - whether formally-sponsored or
not - need not involve preservation of linguistic diversity. He goes a bit further
when he approvingly cites a source arguing that literacy promotion actually
works against "linguistic vitality" (355; see also the extended discussion in
Mühlhäusler 1996). In Maffi (2000), we find a citation to work (by Abram
1997) which emphasises the importance of orality for the persistence of in-
digenous varieties. (Is it irrelevant, I wonder, to note that Abram's book is en-
titled The Spell of the Sensuous!). As well, literacy is often seen as a sort of
bully, in the same way that large languages are the villains, and small ones the
victims - written varieties can push oral ones aside, writing is seen as sophis-
ticated and, indeed, more likely to bear the truth, and so on. It is also sometimes
38 John Edwards
seen as a sort of Trojan horse: Rhydwen (1998), for instance, feels that speak-
ers of at-risk varieties can be falsely lulled into security once writing arrives. It
is certainly reasonable to point out the cruel fallacy that literacy inevitably
leads to social or political improvement, or to refer to the single-mindedness of
literacy campaigns. It is also true that writing does not automatically augment
veracity (do you believe everything you read in the papers?). It would surely be
yet another instance of "isolationism", however, to keep literacy at bay in the
service of language maintenance.
A second, and related, point suggests that formal education is not always
the ally of enduring diversity and bilingualism. Benton (1981: 171) - i n a view
reproduced by Mühlhäusler (1996) - expresses doubt that "schools can have
any positive role in ensuring the continued viability of Oceanic languages" (the
"can" and "any" are interesting here; in his endorsement of the statement,
Mühlhäusler [1996: 267] is more circumspect, saying that "it is probably not
difficult to agree with Benton that education and school language policies have
done little [...]"). These authors are hardly against education per se. Rather,
they wish to draw our attention to its often intrusive qualities, to its champion-
ing of literacy over orality, to the way in which it imposes foreign (i.e., western)
values and methods upon small cultures. Again there is the idea of cultural bul-
lying. Maffi (2000: 177) notes, for example, that formal education need not
come "at the expense of learning traditional knowledge" and, indeed, that
school is not the only venue for the acquisition of "valuable knowledge" - these
notes are, of course, unexceptionable, and presumably would not have been
made in the absence of the juggernaut power of education. It is not difficult to
sympathise with laments about supposedly intrusive "foreign" education para-
digms but - given that all education worthy of the name is multicultural in na-
ture - the argument may be self-defeating. Formal education necessarily in-
volves broadening the horizons, going beyond what is purely local and "tradi-
tional". In an unequal world - one whose disparities create risks for languages,
in fact - education will perforce become yet another evidence of those dispar-
ities. Those concerned with gaining a place in the media for minority languages
have learned that they are double-edged swords: while it is clear that access to
them is important, they also facilitate the transmission of those larger in-
fluences upon decline. There are similar "risks" associated with the medium of
education.
Finally, Maffi (2000) is undoubtedly correct - though hardly original - in
noting that language endangerment (and, by implication, all that is associated
with it) is a political matter. Proponents of the new ecology, unsurprisingly,
tend to see modern social and political developments as harmful to diversity;
globalisation is the latest culprit. But they sometimes overstate their case, im-
Forlorn hope? 39
agining conscious policies and ideologies whose prime purpose is to flatten the
linguistic landscape. Thus, unwelcome "streamlining, diversity-reducing ideo-
logy" is mentioned by Mühlhäusler (2000), who goes on to note that the "free-
market" (312) concept, as applied to languages, is not only complex but, in-
deed, "suspect". It is, of course, easy to find specific examples of direct and
conscious intervention to support Mühlhäusler's claim that, for language plan-
ning, the "main question has been that of constraining or reducing diversity"
(314). But more general conspiracy models are both unlikely and unparsimoni-
ous: officialdom has not generally been directly concerned with language per
se, although linguistic fall-out is of course predictable on the basis of broader
social policies, and the linguistic shrinking in evidence today is virtually en-
tirely explicable without recourse to such models. In a footnote, Mühlhäusler
himself acknowledges that promoting modernisation is seen by "many policy
makers as being sufficient to get rid of linguistic diversity" (361). (This pre-
sumes, of course, close bureaucratic interest in languages in and of themselves;
as well, while modernisation is certainly promoted for profit, its basic attrac-
tion does not rest with "push" factors alone.)
Current ecological models, in their "greenness", naturally tend to identify
some types of political villains more readily than others: unrestrained free-
market capitalism, unfettered industrialisation, galloping globalisation. It is
not uncommon to find disparagement of the scientific culture and concern for
the "privileging" of its knowledge over "folk wisdom" (see Rhydwen 1998).
The small-is-beautiful argument is occasionally trotted out, although rarely as
blatantly as we find in Salminen (1998: 62): "without romanticizing or idealiz-
ing the indigenous cultures, it is clear that they are superior to the mass culture
because their members retain the capability of living in at least relative har-
mony with the natural environment."
When considering arguments, however, that certain societies, groups or in-
stitutions can be singled out for criticism here, one or two generalities should
be borne in mind. The desire for mobility and modernisation, for example, is
(with some few notable exceptions) a global phenomenon. Whether one looks
at the capitalist world or the former communist one, at contemporary times or
historical ones, at empires or small societies, at immigrant minorities or in-
digenous groups, one sees a similarity of pressures which take their toll, force
change and throw populations into transitional states that have certain conse-
quences. Situations alter, people move, needs and demands evolve; and decline
in the existence or the attractions of traditional lifestyles often entails decline in
languages associated with them.
40 John Edwards
4. Conclusions
Despite recent arguments, the new ecology is not particularly new. Haugen
(1987) himself noted that the term language ecology is essentially a reworking
of an "old model", intended to unite insights found in areas like dialectology
and, more recently, sociolinguistics and the sociology of language. The essen-
tial idea, that language issues are political and social, and must be considered in
their contexts, has been long accepted. To put it another way, the view that it is
linguistic attributes that account for language dominance or subordination is
now a very dated one. The modern ecological thrust - as exemplified in the
writings I have cited here - has, however, a rather more narrowed focus. It is an
environmentalism that makes a specific case for diversity. I don't suggest that
this is an illegitimate stance per se, although - as my arguments have tried to
show - it is not a sturdy one in all respects. But it is surely reasonable to have
some general misgivings about an area which styles itself broadly while mar-
shalling its activities along quite specific lines.
The important issues here do not centre upon preferences or assumptions
about diversity. I take it for granted, for instance, that settings in which lin-
guistic diversity is maintained (without either draconian or ill-considered inter-
vention, on the one hand, or at the cost of desired social flexibility, on the other)
are more attractive, and can sometimes be more beneficial, than the alternative.
Like Haugen, I prefer varied landscapes. But I would think it unwise to pre-
sume that preferences are always commensurate with broadly desirable possi-
bility.
In a discussion of endangered languages, Dauenhauer and Dauenhauer
(1998: 78) note the paradoxical situation that languages will die "unless we do
something", but may also die "even if we do something". This is an echo of what
I referred to (in 1985) as the "paradox of the Gaeltacht" - if nothing is done, the
erosion of these Irish-speaking areas continues, but conscious intervention does
not appear to have stopped the decline, either. The same forces that bear upon
original-language decline also contribute to the instability of bilingualism and,
in my view, the ecology of language offers no new insights into them, nor any
novel means of influencing them. It is disingenuous and may, in fact, be dan-
gerous, to present old wine in new bottles (to paraphrase St Matthew), to give
the appearance that some improved articulation is at work. With or without the
"new" ecological perspective, stable bilingualism remains both a desirable out-
come and one often difficult to achieve and maintain. It is, then, a forlorn hope.
Whether the opportunities are slim or virtually non-existent will depend very
much on circumstance. But going into the breach is always perilous.
Forlorn hope? 41
References
Abram, D.
1997 The Spell of the Sensuous: Perception and Language in a More-Than-
Human World. New York: Vintage.
Arnold, M.
1883 On the Study of Celtic Literature and on Translating Homer. New
York: Macmillan.
Bates, M.
1953 Human ecology. In: A. Kroeber (ed.), Anthropology Today, 82-93.
Chicago: University of Chicago Press.
Benton, R.
1981 The Flight of the Amokura. Wellington: New Zealand Council for
Educational Research.
Bolinger, D.
1980 Language: The Loaded Weapon. London: Longman.
Brenzinger, Μ. (ed.)
1992 Language Death: Factual and Theoretical Explorations with Special
Reference to East Africa. Berlin/New York: Mouton de Gruyter.
1998 Endangered Languages in Africa. Köln: Rüdiger Kopp.
Β rumfit, C.
*2001 Individual Freedom in Language Teaching. Oxford: Oxford Univer-
sity Press.
Clifton, J. (ed.)
1968 Introduction to Cultural Anthropology. Boston: Houghton Mifflin.
Cornwell, Β.
1983 Sharpe's Company. London: Harper Collins.
Crystal, D.
2000 Language Death. Cambridge: Cambridge University Press.
Dalby, D.
2000 Linguasphere unites a thousand world voices. Times Higher Edu-
cation Supplement, 26 May.
Dauenhauer, N. and R. Dauenhauer
1998 Technical, emotional and ideological issues in reversing language
shift: Examples from southeast Alaska. In: L. Grenoble and L. Whaley
(eds.), Endangered Languages: Current Issues and Future Prospects,
57-98. Cambridge: Cambridge University Press.
42 John Edwards
Dorian, N. (ed.)
1989 Investigating Obsolescence: Studies in Language Contraction and
Death. Cambridge: Cambridge University Press.
Dorian, N.
1993 A response to Ladefoged's other view of endangered languages. Lan-
guage 69: 575-579.
Edwards, J.
1985 Language, Society and Identity. Oxford: Blackwell.
1992 Sociopolitical aspects of language maintenance and loss: Towards a
typology of minority language situations. In: W. Fase, Κ. Jaspaert and
S. Kroon (eds.), Maintenance and Loss of Minority Languages,
37-54. Amsterdam: Benjamins.
1994 What can (or should) linguists do in the face of language decline? In:
M. Harry (ed.), Papersfrom the Seventeenth Annual Meeting of the At-
lantic Provinces Linguistic Association, 25-32. Halifax: Saint Mary's
University.
1995 Multilingualism. London: Penguin.
* in press The ecology of language revival. Current Issues in Language Plan-
ning.
Fishman, J. A.
1980 Minority language maintenance and the ethnic mother-tongue school.
Modern Language Journal 64: 167-172.
1992 Conference summary. In: W. Fase, Κ. Jaspaert and S. Kroon (eds.),
Maintenance and Loss of Minority Languages, 395-403. Amsterdam:
Benjamins.
Glacken, C.
1967 Traces on the Rhodian Shore. Berkeley: University of California
Press.
Grenoble, L. and L. Whaley (eds.)
* 1998a Endangered Languages: Current Issues and Future Prospects. Cam-
bridge: Cambridge University Press.
Grenoble, L. and L. Whaley
1998b Toward a typology of language endangerment. In: L. Grenoble and
L. Whaley (eds.), Endangered Languages: Current Issues and Future
Prospects, 22-54. Cambridge: Cambridge University Press.
Harmon, D.
1996 Losing species, losing languages: Connections between biological
and linguistic diversity. Southwest Journal of Linguistics 15: 89-108.
Forlorn hope? 43
Haugen, Ε.
* 1972 The ecology of language. In: A. Dil (ed.), The Ecology of Language:
Essays by Einar Haugen, 324-339. Stanford: Stanford University
Press.
1973 The curse of Babel. In: E. Haugen and M. Bloomfield (eds.), Lan-
guage as a Human Problem, 33-43. New York: Norton.
* 1987 Blessings of Babel: Bilingualism and Language Planning. Berlin/
New York: Mouton de Gruyter.
Kedourie, Ε.
1961 Nationalism. London: Hutchinson.
Krauss, Μ.
1992 The world's languages in crisis. Language 68: 4-10.
Ladefoged, P.
1992 Another view of endangered languages. Language 68: 809-811.
Mackey, W.
1978 The importation of bilingual education models. In: J. Alatis (ed.),
Georgetown University Round Table on Languages and Linguistics,
1-18. Washington: Georgetown University Press.
Maffi, L.
2000 Language preservation vs. language maintenance and revitalization:
Assessing concepts, approaches and implications for the language
sciences. International Journal of the Sociology of Language 142:
175-190.
Mill, J.
1964 [1861] Considerations on Representative Government. London: Dent.
Mufwene, S.
2000 Language contact, evolution and death: How ecology rolls the dice.
In: G. Kindell and M. Lewis (eds.), Assessing Ethnolinguistic Vitality,
39-64. Dallas: Summer Institute of Linguistics.
Mühlhäusler, P.
* 1996 Linguistic Ecology: Language Change and Linguistic Imperialism in
the Pacific Region. London: Routledge.
*2000 Language planning and language ecology. Current Issues in Lan-
guage Planning 1: 306-367.
Nettle, D. and S. Romaine
2000 Vanishing Voices: The Extinction of the World's Languages. Oxford:
Oxford University Press.
44 John Edwards
Rhydwen, M.
1998 Strategies for doing the impossible. In: N. Ostler (ed.), Endangered
Languages, 101-106. Bath: Foundation for Endangered Languages.
Robins, R. and E. Uhlenbeck (eds.)
1991 Endangered Languages. Oxford: Berg.
Salminen, T.
1998 Minority languages in a society in turmoil: The case of the northern
languages of the Russian Federation. In: N. Ostler (ed.), Endangered
Languages, 58-63. Bath: Foundation for Endangered Languages.
Steward, J.
1955 Theory of Culture Change. Urbana: University of Illinois Press.
Voegelin, C., F. Voegelin and N. Schutz
1967 The language situation in Arizona as part of the southwest cultural
area. In: D. Hymes and W. Bittle (eds), Studies in Southwestern Eth-
nolinguistics, 403-451. The Hague: Mouton de Gruyter.
Wurm, S.
1998 Methods of language maintenance and revival, with selected cases of
language endangerment in the world. In: K. Matsumura (ed.), Studies
in Endangered Languages, 191-211. Tokyo: Syobo Hituzi.
1. What are the constraints on the feasibility of long-term bilingual accommo-

dations?
2. Why is identity at the heart of discussions of the "social life of language"?
3. What is the particular contribution that the study of endangered languages
makes to the broader sociology-of-language enterprise?
4. What can (or should) scholars do in the face of language decline?
5. Can any meaningful analogies be drawn between linguistic diversity and its
biological counterpart?
When languages disappear, are bilingual education
or human rights a cure? Two scenarios
Tove Skutnabb-Kangas
1. Introduction
There are probably between 6.500 and 10.000 spoken (oral) languages in the
world, and possibly an equal number of sign languages. The most comprehen-
sive listing of the world's (mostly oral) languages is in the 14th edition of Ethno-
logue, edited by Barbara F. Grimes of the Summer Institute of Linguistics.1 This
lists more than 6,700 languages spoken in 228 countries. Europe is linguistically
poor. With some 225 languages, it accounts for only 3% of the world's oral lan-
guages and of these 225, 94 are "endangered", according to Krauss (1992: 5).
81% of the world's oral languages are in Africa, Asia and the Pacific.
Nine countries in the world have more than 200 and another thirteen more
than 100 languages each (Krauss 1992: 6). These top 22 countries, just over
10% of the world's countries, probably account for some 75% (over 5,000) of
the world's oral languages. The top 10 oral languages in the world, in terms of
number of mother tongue speakers (Mandarin Chinese, Hindi, Spanish, Eng-
lish, Bengali, Portuguese, Russian, Japanese, German and Wu Chinese) repre-
sent a mere 0.10-0.15% of the world's oral languages, but account for close to
50% of the world's oral population. The next 10 are Javanese, Korean, French,
Vietnamese, Telugu, Yue Chinese, Marathi, Tamil, Turkish and Urdu. There
are around 80 languages with more than 10 million speakers, together account-
ing for far over 4 billion people. Fewer than 300 languages are spoken by com-
munities of 1 million speakers and above. Over half of the world's oral and
most of the sign languages are used by communities of less than 10,000 speak-
ers, and half of these by 1,000 speakers or less. Around 10% of the world's lan-
guages are spoken by less than 100 speakers each. The median number of
speakers of oral languages is probably some 5-6,000 people. Somewhat over
80% of the languages are endemic; they exist in one country only and are there-
fore obviously extremely vulnerable if they are small in numbers.
Languages are today being killed2 at a much faster pace than ever before in
human history. As a consequence, linguistic diversity is in decline. Krauss (1992)
divides the (oral) languages into three groups, the moribund, the endangered and
the safe languages, on the basis of three criteria: intergenerational transfer from
46 Tove Skutnabb-Kangas
parents to children, numbers, and official status. The basic criterion is whether or
not children are learning the language (and this is partly why educational lan-
guage rights are so important, see below). "Moribund languages" are thus lan-
guages which are no longer being learned by children; "endangered" are lan-
guages which, though now still being learned by children, will - if the present
conditions continue - cease to be learned by children during the coming century;
and "safe languages" are languages which are neither moribund nor endangered
(Krauss 1992: 5-7). The more detailed UNESCO Red Books for Endangered
Languages3 operate with the same criteria and stress that the first two (children's
learning, numbers) are the most important ones.
Linguists agree that many languages face extinction as spoken (or sign) lan-
guages, if present trends continue. Krauss estimates the number of oral lan-
guages that are assured of still being around in 2100 as only around 600, which is
less than 10% of present-day oral languages. Not only will most of the languages
with less than 10,000 speakers, over half of today's languages, disappear, but so
will most of those which have between 10,000 and 1 million speakers.
Some of the direct main agents for this linguistic (and cultural) genocide are
formal education and the mass media, and behind them macro-level economic,
military and political agents. Even if schools alone cannot save languages, as
many sociolinguists, like Fishman (e.g. 1998: 414-415) have pointed out,
schools can kill them more or less on their own.
There is an expanding body of scholarship addressing issues of linguistic in-
equality. The concept linguicism was created by analogy with racism and sexism
to refer to "ideologies, structures and practices which are used to legitimate, ef-
fectuate, regulate and reproduce an unequal division of power and resources
(both material and immaterial) between groups which are defined on the basis of
language" (Skutnabb-Kangas 1988: 13). Most education systems worldwide re-
flect linguicism (Skutnabb-Kangas 2000a). Often state education systems not
only violate the linguistic human rights (LHRs) of minorities but they contribute
to linguistic genocide.
2. From linguistic genocide to linguistic human rights
2.1. A continuum
What is linguistic genocide? I see it as an end point on a continuum where the

other end point is full enjoyment of all linguistic human rights. Languages
which are official languages obviously enjoy many rights (see de Varennes
When languages disappear, are bilingual education or human rights a cure? 47
1996 for an overview) and in most cases one might suggest that native speakers
of these languages enjoy all LHRs. This includes state support for the intergen-
erational transmission of their languages in the state school system, through the
use of these languages as the main media of education.
Linguistic genocide, a possibly provocative and emotionally charged ex-
pression, also has to be scientifically described and legally defined. From a re-
search point of view, using Cobarrubias' taxonomy, linguistic genocide in-
volves actively "killing a language" (without killing its speakers, as in physical
genocide) or, through passivity, "letting a language die". "Unsupported coexist-
ence" mostly also leads to minority languages dying (see Skutnabb-Kangas
1999a, 2000a for discussions about the difference between 'language death' and
'language murder' / linguistic genocide). From an international law point of
view, we can use definitions of genocide and linguistic genocide from the UN
Genocide Convention. Two types of UN definitions are relevant. The first type
are the following two definitions, which still are part of the 1948 UN Inter-
national Convention on the Prevention and Punishment of the Crime of Geno-
cide (E 793 1948):
Article 11(e), "forcibly transferring children of the group to another group"; and
Article 11(b), "causing serious bodily or mental harm to members of the group" [em-
phasis added]
The transmission of languages from the parent generation to children is the most
vital factor for the maintenance of languages. When more and more children get
access to formal education, much of the more formal language learning that ear-
lier happened in the community takes place in schools. If an alien language is
used in schools, i.e. if children do not have the right to learn and use their lan-
guage in schools as the main medium of education, the language is not going to
survive because children educated through the medium of an alien language are
not likely to pass their own language on to their children and grandchildren. In
this case the educational system has, through forced assimilation, participated in
linguistic genocide.
Juan Cobarrubias (1983) has developed a taxonomy of policies which a
state can adopt towards a (minority) language, with the following stages: "1. at-
tempting to kill a language; 2. letting a language die; 3. unsupported coexistence;
4. partial support of specific language functions; 5. adoption as an official lan-
guage". A directly LHRs-related similar taxonomy or continuum of LHRs might
start at (a) linguistic genocide, and continue through (b) discrimination on the
basis of language, (c) non-discrimination prescription and (d) conditional accept-
ance of some LHRs, to (e) full unconditional LHRs.
2.2. Linguistic genocide - first scenario
The first educational scenario to be sketched is the one we have today for most
indigenous peoples, linguistic minorities, and those majority populations,
often in ex-colonial countries, where one of the old colonial languages still is
the main medium of education. This scenario fits UN definitions of linguistic
genocide. It does not support intergenerational transfer of languages: it leads to
disappearance of linguistic diversity.
First language attrition and loss have been described extensively in research
literature and fiction. Sandra Kouritzin (1999) describes many cases in Canada
where immigrant minority children have lost a language within one generation
so that they as adults, for instance, are no longer able to speak to their parents.
Lily Wong Fillmore (1992) has described the consequences for families in the
USA. Peter Mühlhäusler (1996) discusses results of linguistic imperialism in
the Pacific. Pirjo Janulf (1998) shows in her longitudinal study that of those
Finnish immigrant minority members in Sweden who had had Swedish-medi-
um education, not one spoke any Finnish to their own children. Even if they
themselves might not have forgotten their Finnish completely, their children
were certainly forcibly transferred to the majority group, at least linguistically.
This happens to millions of speakers of threatened languages all over the
world. For oral minority students education through the medium of a dominant
majority language often leads to the students using the dominant language with
their own children later on. Over a generation or two the children are linguisti-
cally and often also culturally assimilated, forcibly transferred to a dominant
group. Since there are no alternatives in formal education (i.e. schools or
classes which teach mainly through the medium of the threatened indigenous
or minority languages), the transfer happens by force. For it to be voluntary, al-
ternatives should exist, and parents would need to have enough reliable in-
formation about the long-term consequences of the various choices. None of
these conditions are usually fulfilled for indigenous or minority parents and
children. Therefore, the situations where children lose their first language
through forced assimilation can often be characterised as genocide according
to Article 11(e), "forcibly transferring children of the group to another group".
Since most Deaf children are born to hearing parents, parents and children
do not have the same mother tongue by origin, and many of the Deaf children
will in their turn have hearing children. Deaf children of hearing parents are in
many countries still taught through oral methods, i.e. taught lip-reading and
speaking in a dominant majority language, to the exclusion of a sign language.
They are not learning their 'own' language, a sign language, which is for all
Deaf children the only type of language through which they can express them-
selves fully, i.e. it is their mother tongue by competence. Thus Deaf children,
taught predominantly through the medium of a dominant oral majority lan-
guage, are undergoing linguistic genocide: an attempt is made to forcibly
transfer them from their own language group to dominant majority language
group. This subtractive attempt is doomed to fail, however, since it is imposs-
ible for Deaf children to hear.
There is also a wealth of research and statistics about the mental harm that
forced assimilation causes in education and otherwise. This also entails threats
to democracy and equality. A few examples follow, from Africa, Australia and
North America, and from Deaf education - there are hundreds of other
examples in Skutnabb-Kangas (2000a).
Williams (1998) did two large-scale empirical studies, testing almost
1,500 students, in Zambia and Malawi in grades 1-7 and interviewing and ob-
serving many. In Zambia, children were (supposed to be) taught through the
medium of English from grade 1 onwards, and to study a local language as a
subject. This is known as submersion education (see Skutnabb-Kangas 1996c,
2000a for definitions of various models). In Malawi, children were taught
through local languages, in most cases their mother tongues, during the first
4 years of primary school, while studying English as a subject. From grade 5
onwards, children in Malawi also studied through the medium of English. Even
when the Zambian children had had all their schooling in English, their test re-
sults in the English language were no better than those of the Malawi children
who had only studied English as a subject. In fact the children in Malawi were
doing slightly better than the children in Zambia.
In both countries there were huge differences in the results in English be-
tween urban and rural children, meaning English language results are socially
not enhancing democracy. Similarly, there were considerable gender differ-
ences, meaning English language results do not support gender equality. Many
of the Zambian pupils could not even be tested in the local language because
they could not read it. On the other hand, when the Malawi children were tested
in the local language, there were almost no differences between urban and rural
pupils, or between the genders. Large numbers of Zambian pupils are claimed
to "have very weak or zero reading competence in two languages" (Williams
1998: 62). The "Malawian success in teaching reading in the local language",
on the other hand, is "achieved despite the almost complete absence of books
and classes with an average of around 100 pupils, many of which are taught in
the open" (Williams 1998: 62). We often hear that there is no money for teach-
ing in the many languages, in Africa or Asia. Echoing Indian evidence (e.g.,
Pattanayak 1988), Williams concludes that "[the] moral of the Malawian
achievement would appear to be that if resources are scarce, there is a greater
likelihood of success in attempting to teach pupils a known local language,

rather than an unknown one" (Williams 1998: 62). Since 74-89% of the
children in grades 3-6 are judged as not adequately comprehending a text in
English that is judged to be at their level (Williams 1998: 63), "it is likely that
they cannot understand their content subject course books" (Williams 1998:
63), and therefore it is "difficult to see how the majority of pupils in Zambia
and Malawi could learn other subjects successfully through reading in English"
(Williams 1998: 63).
Teaching through an African language thus produces more democracy and
equality, whereas using a foreign language as the measure of status and as a medi-
um of education harms the children and also society as a whole. Williams con-
cludes that "[f]or the majority of children in both countries the test results, and
classroom observations, suggest there is a clear risk that the policy of using Eng-
lish as a vehicular language may contribute to stunting, rather than promoting,
academic and cognitive growth" (Williams 1998:63-64; emphasis added). This
fits the UN genocide Article 11(b) definition of "causing serious (...) mental harm
to members of the group". The study confirms a pattern in many postcolonial
contexts. World Bank policy employs a rhetoric of endorsing local languages, but
funding exclusively strengthens European languages (Mazrui 1997; Brock-Utne
1999).
A similar conclusion is reached in Australia by Anne Lowell and Brian
Devlin in an article (1999) describing the "Miscommunication between Abo-
riginal Students and their Non-Aboriginal Teachers in a Bilingual School". It is
clearly demonstrated that "even by late primary school, children often did not
comprehend classroom instructions in English" (Lowell and Devlin 1999:
137). "Communication breakdowns occurred frequently between children and
their non-Aboriginal teachers" (Lowell and Devlin 1999: 138), with the result
that "the extent of miscommunication severely inhibited the children's edu-
cation when English was the language of instruction and interaction" (Lowell
and Devlin 1999: 137; emphasis added). In the conclusions and recommen-
dations the authors say that "the use of a language of instruction in which the
children do not have sufficient competence is the greatest barrier to successful
classroom learning for Aboriginal Children" (Lowell and Devlin 1999: 156;
emphasis added). This is also "causing serious (...) mental harm to members of
the group".
John Baugh from Stanford University, in an article called "Educational Mal-
practice and the Miseducation of Language Minority Students" (2000) draws a
parallel between how physicians may maltreat patients and how minority stu-
dents (including students who do not have mainstream US English as their first
language, for instance Ebonics/Black English), are often treated in education.
The harm caused to them by this maltreatment and miseducation also fits the UN
definition of "causing serious (...) mental harm to members of the group".
When Deaf students are taught orally only and sign languages have no place
in the curriculum, this also often causes mental harm, including serious preven-
tion or delay of cognitive growth potential (e.g. Branson and Miller 1998; Jo-
kinen 2000).
All these are examples of genocide according to Article 11(b). In all the
cases above it is also a question of the school "prohibiting the use of the lan-
guage of the group", as in the second type of UN definition. This is the specific
definition of linguistic genocide, which was included in the final Draft of the
Convention. In preparatory work for the UN Genocide Convention, linguistic and
cultural genocide were discussed alongside physical genocide, and were seen as
serious crimes against humanity (Capotorti 1979). When the Convention was ac-
cepted in the UN General Assembly, Article 3 covering linguistic and cultural
genocide was voted down by 16 states (Capotorti 1979), and it is thus not in-
cluded in the final Convention of 1948. But even when the states members of the
UN in 1948 voted down the Article on linguistic and cultural genocide, there
was wide agreement about how to define the phenomena. Thus what remains is
a definition of linguistic genocide which most states then in the UN were pre-
pared to accept:
Article III( 1) "Prohibiting the use of the language of the group in daily intercourse or
in schools, or the printing and circulation of publications in the language of the
group".
'Prohibition' can be direct or indirect. If there are no minority teachers in the pre-
school / school and if the minority language is not used as the main medium of
education, the use of the language is indirectly prohibited in daily intercourse and
in schools, i.e. it is an instance of linguistic genocide. Most minority education in
the world is thus tantamount to linguistic genocide, as defined by the UN. So is the
education that most indigenous first nations have had and that many of them still
have (see e.g. Hamel 1994; Jordan 1988; Fettes 1998). Skutnabb-Kangas (2000a)
gives hundreds of examples of this prohibition and the harm it causes. Subtract-
ive formal education which teaches children something of a dominant language
at the cost of their first language (i.e. mother tongue by origin, as for oral minor-
ity children and for Deaf children with Deaf parents, or mother tongue by com-
petence, as for Deaf children of hearing parents), is genocidal. All subtractive
models of 'bilingual' education can be and often are genocidal. This includes not
only submersion and segregation models but also most transitional models, es-
pecially early-exit transition.4
This stands in sharp contrast to the enriching outcomes of additive forms of

bilingualism education, where new languages are added to the child's linguistic
repertoire at no cost to her first language(s). Scholars such as Baetens Beards-
more (e.g. 1986, 1992, 1993, 1995) and others have demonstrated, both with
empirical and theoretical arguments, how additive language learning, with high
levels of multilingualism as a goal and a result, can be achieved in varying con-
texts, all over the world. Additive bilingual education models can take many
forms, including language maintenance models, immersion, two-way models,
and the European Schools model (for detailed overviews, see Baker and Prys-
Jones 1998; the contributions to Skutnabb-Kangas 1995; and Skutnabb-Kan-
gas 2000: chapter 8).
2.3. Can present linguistic human rights counteract linguistic

genocide? - Prerequisites for the second scenario
For the maintenance and development of languages (and thereby linguistic di-
versity on earth), educational language rights, including the right to mother
tongue medium education, are absolutely vital. Binding LHRs, education
rights in particular, may be one of the necessary (but not sufficient) ways of
counteracting linguicide and linguicism. Our second scenario is one where all
education is based on full educational (and other) LHRs for everybody. Do we
currently have a proper basis in international law for these rights to be imple-
mented?
There is a growing scholarly literature on LHRs, with a convergence be-
tween the concerns of lawyers (e.g. de Varennes 1996), sociolinguists (Skut-
nabb-Kangas and Phillipson 1994; Hamel 1997), media researchers (Hamelink
1998), economists (Grin and Vaillancourt 2000), political scientists (McRae
1997; MacMillan 1998), educational sociologists (May 1999, 2001), anthro-
pologists (Maffi etal. 1999) and others. Kontra etal. 1999 is an example of
multidisciplinary efforts. The entire field is explored in depth, and related to
overall language policy and to language ecology in Skutnabb-Kangas (2000a)
and Phillipson (2000). Minorities have some support for other aspects of using
their languages in areas such as public administration, courts, the media, etc.5
So far, however, the picture about necessary legal educational prerequisites for
the second scenario is pretty dim.
In many of the post-1945 human rights documents, language, along with
race, sex, and religion, has been seen as one of the basic characteristics on the
basis of which individuals are not to be discriminated against in their enjoy-
ment of human rights and fundamental freedoms (e.g. the United Nations
Charter, Art. 13). But when we move from the non-duty-inducing phrases in
the preambles of the human rights instruments to the binding clauses, es-
pecially to the educational clauses, there is a change (see Skutnabb-Kangas and
Phillipson 1994 for an overview). The clauses or articles about other character-
istics create obligations: the states are firm dutyholders and are obliged to
('shall') act in order to ensure the specified rights (i.e. positive rather than
negative rights). They contain demanding formulations where modifications,
opt-out clauses and sliding-scale alternatives are rare. In binding educational
clauses, however, language often disappears completely (see Skutnabb-Kangas
1998). An example is the Universal Declaration of Human Rights (1948) where
the paragraph on education (26) does not refer to language at all. Another
example is the International Covenant on Economic, Social and Cultural
Rights (1966) where the educational Article (13) omits reference to language
or linguistic groups (which have been mentioned in its general Article 2.2):
(...) education shall enable all persons to participate effectively in a free society, pro-
mote understanding, tolerance and friendship among all nations and all racial, ethnic
or religious groups (...)
The European Convention on Human Rights and Fundamental Freedoms of

1950 is equally silent on not only language rights in education but on even
more general minority rights. Several new Declarations and Conventions
aimed at protecting minorities and/or minority languages, passed in the 1990s,
also omit language in the education clauses.
If language-related rights are included and specified, the Article dealing with
these rights, in contrast to the demanding formulations and the few opt-outs and
alternatives in the articles dealing with other characteristics, is typically so weak
and unsatisfactory that it is virtually meaningless, as many researchers show (see
e.g., Guillorel and Koubi 1999; Kontra etal. 1999; Phillipson and Skutnabb-Kan-
gas 1994,1995,1996; Skutnabb-Kangas 1996a, b, 1999b; Skutnabb-Kangas and
Phillipson 1994, 1997, 1998; Thornberry 1997; de Varennes 1996). Compare the
unconditional formulation in Article 1 with the education Article 4.3 in the UN
Declaration on the Rights of Persons Belonging to National or Ethnic, Religious
and Linguistic Minorities (1992, not in force yet because of too few ratifications;
see Skutnabb-Kangas 2000b):
1.1. States shall protect the existence and the national or ethnic, cultural, religious
and linguistic identity of minorities within their respective territories, and shall en-
courage conditions for the promotion of that identity.
1.2. States shall adopt appropriate legislative and other measures to achieve those
ends.
4.3. States should take appropriate measures so that, wherever possible, persons
belonging to minorities have adequate opportunities to learn their mother tongue or
to have instruction in their mother tongue.
One can ask what constitutes "appropriate measures", or "adequate opportun-

ities", and who is to decide what is "possible"? We see the same phenomenon in
the European Charter for Regional or Minority Languages (1992, in force since
March 1998). The formulations in the education Article 8 include a range of
modifications, including "as far as possible", "relevant", "appropriate", "where
necessary", "pupils who so wish in a number considered sufficient", "if the
number of users of a regional or minority language justifies it", as well as a
number of alternatives, as in "to allow, encourage or provide teaching in or of the
regional or minority language at all the appropriate stages of education" (empha-
sis added). Just as in the UN Declaration above, the opt-outs and alternatives per-
mit a reluctant state to meet the requirements in a minimalist way, which it can
legitimate by claiming that a provision was not "possible" or "appropriate", or
that numbers were not "sufficient" or did not "justify" a provision, or that it
"allowed" the minority to organise teaching of their language as a subject, at their
own cost. Likewise, in the Council of Europe Framework Convention for the Pro-
tection of National Minorities (1994, in force since February 1998; see Thornber-
ry 1997, for a thorough critique), the Article covering medium of education is so
heavily qualified that the minority is completely at the mercy of the state (em-
phases added):
In areas inhabited by persons belonging to national minorities traditionally or in sub-

stantial numbers, if there is sufficient demand, the parties shall endeavour to en-
sure, as far as possible and within the framework of their education systems, that
persons belonging to those minorities have adequate opportunities for being taught
in the minority language or for receiving instruction in this language.
The rights of indigenous peoples might improve somewhat with the UN Working
Group on Indigenous Populations Draft Declaration on the Rights of Indigenous
Peoples.6 A still more recent attempt to promote language rights, a draft Universal
Declaration of Linguistic Rights, handed over to UNESCO in Barcelona in June
1996, also suffers from similar shortcomings, even if for several beneficiaries
(language communities and, to some extent, language groups) it represents great
progress in relation to the other instruments described. Still, indirectly its edu-
cation section forces all others except those defined as members of language com-
munities (which roughly correspond to national territorially based minorities) to
assimilate. The Declaration is under revision at UNESCO7. In conclusion, most
educational clauses do not oblige states to more than non-discrimination on the

basis of language.
3. Work to counteract minorisation and promote the survival of

linguistic diversity, in international law and in education
There are few if any universal promotion-oriented rights in the binding clauses on
language rights in education in international law. There are some small recent im-
provements in 'soft law', though, specifically in a General Comment of 6 April
1994 from the UN Human Rights Committee (UN Doc. CCPR/ C / 21 / Rev.l /
Add.5,1994) on Article 27 of the UN International Covenant on Civil and Politi-
cal Rights (which still grants the best binding protection to languages), and in the
Hague Recommendations Regarding Minority Education Rights, from the OSCE
(Organisation for Security and Cooperation in Europe, with 55 member states)
High Commissioner on National Minorities8. The rights are also seen as applying
to immigrant minorities when the two documents are read together. In the Hague
Recommendations section "The spirit of international instruments", bilingualism
is seen as a right and responsibility for persons belonging to national minorities
(Art. 1), and states are reminded not to interpret their obligations in a restrictive
manner (Art. 3). In the section on "Minority education at primary and secondary
levels", mother tongue medium education is recommended at all levels, including
bilingual teachers in the dominant language as a second language (Ar-
ticles 11-13). Teacher training is made a duty on the state (Art. 14). Finally, the
Explanatory Note states that "submersion-type approaches whereby the curricu-
lum is taught exclusively through the medium of the State language and minority
children are entirely integrated into classes with children of the majority are not
in line with international standards" (p. 5). If the Hague Recommendations were
to be implemented, linguistic genocide in education could be stopped and
children would have some of the most vital LHRs. The 1997 Harare Declaration
from an OAU (Organisation for African Unity) Conference of Ministers on
Language Policies in Africa and The Asmara Declaration on African Lan-
guages and Literatures (17 Jan. 2000)9 are examples of similar positive devel-
opments. Several edited books (e.g. Skutnabb-Kangas 1995; May 1999) about
successful programmes show how education leading towards high levels of
multilingualism can be organised.
Several international organisations work for the promotion of linguistic diver-
sity. Some are collecting the basic data needed. Ethnologue, mentioned above, is
one. The World languages report, a UNESCO project whose aim is to describe
the linguistic diversity of the world, to study the evolution and current state of the
languages, and explain the problems which affect them in the different regions of
the world, wants to draw attention to the need for proper management of the
world's linguistic heritage10. Another UNESCO-supported project is the Inter-
national Clearing House for Endangered Languages, in Tokyo. It works with
UNESCO on the Red Books on Endangered Languages, functions as a data bank
of materials on endangered languages, and publishes a newsletter11. The Euro-
pean Bureau for Lesser Used Languages, EBLUL, is an example of proactive
work to counter the minorisation of languages in European Union countries12
Terralingua is an international organisation devoted to preserving the world's lin-
guistic diversity and to investigating links between biological and cultural diver-
sity. Terralingua's web-site13 has lists of and links to organisations working with
both endangered languages, including various types of salvage operations, and
with language rights.
Language scholars who have discussed the ecology of languages, starting
with Einar Haugen (e.g. 1972) have been aware of the threat to linguistic diversity
and many are trying to counteract it, through analysis and action (see, e.g. Brenz-
inger 1992; Fill 1993, 1998; Fishman 1991, 2000; Grenoble and Whaley 1996,
1998; Hinton 1994; Huss 1999; Mühlhäusler 1996). In the struggle against lin-
guicism and linguistic imperialism and domination, speakers of threatened lan-
guages are using many strategies. In Europe and neo-Europes (settler countries
populated from Europe to the virtual extinction of earlier populations), most of
those numerically non-dominant groups who have succeeded in gaining legal
protection for their languages nationally have initially been politically and /or
economically 'powerful' groups: Swedish-speakers in Finland, Afrikaans- and
English-speakers in South Africa, Russian-speakers in the Baltic states. Subse-
quently, a few small national minorities and indigenous peoples have also
gained some linguistic protection (e.g. in Canada, Finland, New Zealand); a
backlash also occurs (cf. Australia, California). Sign languages are mentioned
in the constitutions of 8 states, Uganda and Finland being the first ones. Rom-
any is starting to get some protection in a few countries. Kurdish, forbidden in
Turkey, is extensively used in diaspora, and Kurdish satellite television has,
despite serious threats, succeeded in broadcasting Kurdish-medium pro-
grammes to Kurdistan from Europe (Hassanpour 1999, 2000).
Revitalisation and even the reclaiming of earlier minorised languages are also
taking place. An encouraging example is given in Amery (2000). He describes
work on reclaiming Kaurna, an Australian Aboriginal language where the last
speaker died some 60 years ago. The reclamation is mainly based on missionary
documents from around 1850. The Maori, Hawaiians and Sämi use "language
nests", in which pre-schoolers are taught in the indigenous languages by lin-
guistically and culturally proficient elders. Their pre-school teachers and par-
ents also often develop more proficiency in the ancestral language. In immer-
sion programmes for these indigenous children, they are taught through the
medium of indigenous languages which they initially do not know. The train-
ing of teachers and journalists in, for and through the medium of several small
indigenous languages is expanding. In Arctic areas, for example, indigenous
peoples are also establishing their own universities. Master-Apprentice-pro-
grammes in California (see Hinton 1994) pair off proficient indigenous elders
with younger people for 6-12 months, for instance 20 hours a week, for lan-
guage revitalisation purposes, where the only requirement is that they use an
indigenous language These are just a few examples.
Despite such work, strategies to counteract the linguistic dominance and hier-
archisation that may ultimately lead to the disappearance of the majority of
today's languages are urgently needed. Today's efforts are insufficient. Without
basic major changes of the kind Baetens Beardsmore (2000) and others have been
advocating for decades, the linguicidal scenario continues. But as we know, lack
of LHRs is not only an information problem. The political will of states to grant
LHRs is the main problem. Human rights, especially economic and social
rights, are, according to human rights lawyer Katarina Tomasevski (1996:
104), to act as correctives to the free market. She claims that the "purpose of in-
ternational human rights law is (...) to overrule the law of supply and demand
and remove price-tags from people and from necessities for their survival" (To-
masevski 1996: 104). These necessities for survival include not only basic food
and housing (which would come under economic and social rights), but also
basics for the sustenance of a dignified fife, including basic civil, political and
cultural rights - and LHRs are a part of cultural rights. The message from sociol-
ogists like Zygmunt Bauman, human rights lawyers like Katarina TomaSevski
and many others is that unless there is a redistribution of resources for imple-
menting human rights, progress will be limited. It is probably not even of any use
to spread knowledge of human rights as a basis for self-directed human develop-
ment unless the resources for implementation follow, and that can only happen
through a radical redistribution of the world's material resources. But the impli-
cations of lack of educational LHRs and the resulting linguistic genocide reach
much further than education or linguistic diversity.
4. Implications for language ecology
A comparison of the world's linguistic and biological megadiversity countries

shows a very high overlap (Harmon 1995, in press); both languages and bio-
logical species become thicker on the ground the closer to the equator one
moves, and arctic areas have fewer species and languages.
Making a very simple calculation for the year 2100, estimating the rate of
extinction with today's situation as the starting point, yields the following re-
sults (see Skutnabb-Kangas 2000a for details): according to a pessimistic real-
istic estimate, 20% of all today's biological species will be extinct in
100 years' time, while according to an optimistic realistic estimate the figure
would be 2%. The corresponding figures for plants are between 13 and 17%.
In contrast, "in the last five hundred years about half the known languages of
the world have disappeared", according to Hans-Jürgen Sasse (1992: 7), and
90% of today's languages would be extinct in 100 years' time, according to
Krauss, and over half even according to the most optimistic estimates. Thus
the threat to biodiversity is relatively mild, as compared to the threat to lin-
guistic diversity.
But the relationship between linguistic and cultural diversity on the one hand
and biodiversity on the other is possibly not only correlational. There seems to be
mounting evidence that it might be causal (see Posey 1999, edited for the United
Nations Environmental Program, for some of the evidence, especially Maffi etal.
1999; see also Skutnabb-Kangas 2000a). The strong correlation need not indicate
a direct causal relationship, in the sense that neither type of diversity can prob-
ably be seen directly as an independent variable in relation to the other. But lin-
guistic and cultural diversity seem to be decisive mediating variables in sustain-
ing biodiversity itself, and vice versa, for as long as humans inhabit the earth.
Today it is safe to say that there is no pristine nature left - all landscapes are cul-
tural landscapes; they have been and are influenced by human action, even those
where untrained observers might not notice this immediately. Ethnobiologists,
human-ecologists and others have proposed theories of "human-environment
coevolution", including the assumption that "cultural diversity might enhance
biodiversity or vice versa" (Maffi 1996). In this perspective, the first conference
investigating this relationship between humans and their environment, "Endan-
gered Languages, Endangered Knowledge, Endangered Environments" (Maffi
2000) stressed "the need to address the foreseeable consequences of massive dis-
ruption of such long-standing interactions" [i.e. the human-environment coevol-
ution] (Maffi 1996). The processes of language loss also "affect the maintenance
of traditional environmental knowledge - from loss of biosystematic lexicon to
loss of traditional stories" (Maffi 1996). Thus loss of languages on a massive
scale may also entail loss of some of the basic prerequisites for maintaining life
on the planet. Both bilingual education and human rights are a small but necess-
ary prerequisite for a cure - but much more is needed.
Notes
1. http://www.sil.org/ethnologue/.
2. Using "killed" rather than "dying" or "disappearing" highlights the fact that it is
neither natural (in the same sense as for biological organisms) nor agentless for lan-
guages to disappear. And if there are agents responsible for and /or contributing to
the killing, the scope for action may also be broader than if one thinks one is fighting
against a natural development. Processes leading to linguistic assimilation and
therefore often languages disappearing include linguistic genocide. Besides, all the
verbs, kill, disappear, die, are equally metaphorical.
3. The Red Books for Europe and Northeast Asia can be found at <http://www.helsinki.
fi/~tasalmin/>; Asia, Africa and the Pacific at <http://www.tooyoo.Lu-tokyo.ac.jp/>
and Russia at <http://www.eki.ee/books/redbook/>.
4. For these models, see Baker and Prys Jones 1998, articles in Skutnabb-Kangas 1995
and Skutnabb-Kangas 2000, chapter 8.
5. Frowein, Hofmann and Oeter's edited books (1993, 1994) about minority rights in
European States give excellent overviews of the European situation.
6. <http ://www.unhchr.ch/html/menu4/subres/9445 .htm>.
7. News and/or details about some human rights instruments can be checked at the fol-
lowing web-sites: The European Charter for Regional or Minority Languages
<http://www.coe.fr./eng/legaltxt/148e.htm>; Framework Convention for the Pro-
tection of National Minorities <http://www.coe.fr./eng/legaltxt/157e.htm>; Draft
Universal Declaration of Linguistic Rights <http://www.linguistic-declaration.
org>. Many documents on language and law can also be downloaded from Mer-
cator Linguistic Law and Legislation's web-site <http://www.troc.es/ciemen/
mercator/index-gb.htm>.
8. <http://www.osce.org/>.
9. <http://www.outreach.psu.edu/C&I/A110dds/declaration.html>
10. <http://www.unescoeh.org>.
11. <http://www.tooyoo.Lu-tokyo.ac.jp>.
12. EBLUL Head Office: <eblul@indigo.ie>; EBLUL Information Centre: <pub00341
@innet.be>.
13. <http://www.terralingua.org>.
References
Amery, R.
2000 Warrabarna Kaurnal Reclaiming an Australian Language. (Multilin-
gualism and Linguistic Diversity.) Lisse: Swets and Zeitlinger.
Baetens Beardsmore, H.
1986 Bilingualism: Basic Principles. (Revised second edition). Clevedon:
Multilingual Matters.
1992 Bilingual education. In: J. Lynch, C. Modgil, and S. Modgil (eds.),

Cultural Diversity and the Schools: Education for Cultural Diversity:
Convergence and Divergence, volume 1, 273-283. London: The
Falmer Press.
1993 European models of bilingual education: practice, theory and devel-
opment. Journal of Multilingual and Multicultural Development 14
(1 / 2): 103-120.
* 1995 The European school experience in multilingual education. In:
T. Skutnabb-Kangas (ed.), Multilingualism for All, 21-68. Lisse:
Swets and Zeitlinger.
2000 Taalverscheidenheid: inzet en strategieen [Language diversity: issues
and strategies]. Bulletin des Seances Academie Royale des Sciences
d'Outre-Mer 46 (4): 427-437.
Baker, C. and S. Prys Jones
1998 Encyclopedia of Bilingualism and Bilingual Education. Clevedon/
Philadelphia: Multilingual Matters.
Baugh, J.
2000 Educational malpractice and the miseducation of language minority stu-
dents. In: J. Kelly Hall and W. G. Eggington (eds.), The Sociopolitics of
English Language Teaching, 104-116. Clevedon: Multilingual Matters.
Branson, J. and D. Miller
1998 Nationalism and the linguistic rights of Deaf communities: Linguistic
imperialism and the recognition and development of sign languages.
Journal of Sociolinguistics 2 (1): 3-34.
Brenzinger, Μ. (ed.)
1992 Language Death: Factual and Theoretical Explorations with Special
Reference to East Africa. Berlin/NewYork: Mouton de Gruyter.
Brock-Utne, B.
1999 Whose Education for All? Recolonizing the African Mind? New York:
Garland.
Capotorti, F.
1979 Study of the Rights of Persons Belonging to Ethnic, Religious and Lin-
guistic Minorities. New York: United Nations.
Cobarrubias, J.
1983 Ethical issues in status planning. In: J. Cobarrubias and J. A. Fishman
(eds.), Progress in Language Planning: International Perspectives,
41-85. Berlin/New York: Mouton de Gruyter.
Fettes, Μ.
1998 Life of the Edge: Canada's Aboriginal languages under official bilin-
gualism. In: T. Ricento and B. Burnaby (eds.), Language and Politics in
the United States and Canada: Myths and realities, 117-149. Mahwah,
NJ: Lawrence Erlbaum.
Fill, A.
1993 Ökolinguistik. Eine Einführung [Ecolinguistics. An introduction], Tii-
bingen: Gunter Narr Verlag.
1998 Ecolinguistics - State of the art 1998. In: A. Fill, AAA - Arbeiten aus
Anglistik und Amerikanistik 23 (1): 1-16. Tübingen: Gunter Narr Ver-
lag.
Fishman, J. A.
* 1991 Reversing Language Shift. Theoretical and Empirical Assistance to
Threatened Languages. Clevedon, UK: Multilingual Matters.
1998 Review of Tove Skutnabb-Kangas' Multilingualism for All. Language
in Society 27 (3): 413-415.
Fishman, J. A. (ed.)
*2000 Can Threatened Languages be Saved? Reversing Language Shift,
Revisited: A 21" century Perspective. Clevedon, UK: Multilingual
Matters.
Frowein, J. Α., R. Hofmann and S. Oeter
1994 Das Minderheitenrecht europäischer Staaten, Teil 1,2 (Beiträge zum
ausländischen öffentlichen Recht und Völkerrecht, Band 108 / 109).
[Minority rights in European States, volumes 1, 2 (Contributions to
foreign public law and international law)]. Berlin: Springer-Verlag.
Grenoble, L. A. and L. J. Whaley
1996 Endangered languages: current issues and future prospects. Inter-
national Journal of the Sociology of Language 118: 209—223.
Grenoble, L. A. and L. J. Whaley (eds.)
1998 Endangered Languages: Current Issues and Future Prospects. Cam-
bridge: Cambridge University Press.
Grin, F. and F. Vaillancourt
2000 On the financing of language policies and distributive justice. In:
R. Phillipson (ed.), Rights to Language. Equity, Power and Education,
102-110. Mahwah, NJ: Lawrence Erlbaum Associates.
Guillorel, H. and G. Koubi (eds.)
1999 Langues et droits. Langues du droit, droit des langues [Languages and
rights. Language of rights, the right of languages]. Bruxelles: Bruylant.
Hamel, R. E.
1994 Indigenous education in Latin America: policies and legal frameworks.
In: T. Skutnabb-Kangas and R. Phillipson (eds.), in collaboration with
M. Rannut. Linguistic Human Rights. Overcoming Linguistic Dis-

crimination, 271-287. (Contributions to the Sociology of Language 67).
Berlin/New York: Mouton de Gruyter.
1997 Linguistic human rights from a sociolinguistic perspective. Special
Issue. International Journal of the Sociology of Language 127: 1-24.
Hamelink, C. J.
1994 Trends in World Communication: on Disempowerment and Self-em-
powerment. Penang: Southbound and Third World Network.
1998 Human rights. Gazette. The International Journal for Communication
Studies 60 (1).
Harmon, D.
1995 The status of the world's languages as reported in the Ethnologue.
Southwest Journal of Linguistics 14,1-33.
Harmon, D.
in press In Light of Our Differences: How Diversity in Nature and Culture
Makes us Human. Washington, D.C.: The Smithsonian Institute.
Hassanpour, A.
1999 Language rights in the emerging world linguistic order: the state, the
market and communication technologies. In: M. Kontra, R. Phillipson,
T. Skutnabb-Kangas and T. Värady (eds.), Language: a Right and a
Resource. Approaching Linguistic Human Rights, 223-241. Buda-
pest: Central European University Press.
Hassanpour, A.
2000 Theory and practice in language rights for the Kurds. In: R. Phillipson
(ed.), Rights to Language. Equity, Power and Education, 33-39.
Mahwah, NJ: Lawrence Erlbaum Associates.
Haugen, E.
1972 The Ecology of Language. Essays by Einar Haugen. Stanford: Stan-
ford University Press.
Hinton, L.
1994 Flutes of Fire. Berkeley: Heyday.
Huss, L.
1999 Reversing Language Shift in the Far North. Linguistic Revitalization
in Scandinavia and Finland. Acta Universitatis Upsaliensis. Studia
Uralica Upsaliensia 31. Uppsala: Uppsala University.
Janulf, P.
1998 Kommer finskan i Sverige att fortleva? En Studie αν sprakkunskaper
och spräkartvändning hos andragenerationens sverifinnar i Botkyrka
och hos finlandssvenskar i Abo. [Will Finnish survive in Sweden? A

study of language skills and language use among second generation
Sweden Finns in Botkyrka, Sweden, and Finland Swedes in Turku, Fin-
land]. ActaUniversitatis Stockholmiensis, StudiaFennica Stockholmien-
sia 7. Stockholm: Almqvist and Wiksell International.
Jordan, D.
1988 Rights and claims of indigenous people. Education and the reclaiming of
identity: the case of the Canadian natives, the Sami and Australian Abo-
rigines. In: T. Skutnabb-Kangas and J. Cummins (eds.), Minority Edu-
cation: from Shame to Struggle, 189-222. Clevedon: Multilingual
Matters.
Jokinen, M.
2000 The linguistic human rights of Sign language users. In: R. Phillipson
Mahwah, NJ: Lawrence Eribaum Associates.
Kontra, Μ., R. Phillipson, T. Skutnabb-Kangas and T. Värady (eds.)

1999 Language: a Right and a Resource. Approaching Linguistic Human
Rights. Budapest: Central European University Press.
Kouritzin, S.
1999 Face[t]s of First Language Loss. Mahwah, NJ: Lawrence Erlbaum
Associates.
Krauss, M.
1992 The world's languages in crisis. Language 68 (1): 4-10.
L6ger, S. (ed.)
1996 Vers un agenda linguistique: regardfuturiste sur les nations unies, To-
wards a Language Agenda: Futurist Outlook on the United Nations.
Ottawa: Canadian Centre for Linguistic Rights, University of Ottawa.
Lowell, A. and B. Devlin

1999 Miscommunication between Aboriginal Students and their Non-Abo-
riginal Teachers in a Bilingual School. In: S. May (ed.), Indigenous
Community-based Education, 137-159. Clevedon: Multilingual
Matters.
MacMillan, C. M.
1998 The Practice of Language Rights in Canada. Toronto: Toronto Univer-
sity Press.
Maffi, L.
1996 Language, knowledge and the environment: Threats to the world's bio-
cultural diversity. Terralingua Newsletter 2.
2000 Linguistic and biological diversity: the inextricable link. In: R. Phillipson
Mahwah, NJ: Lawrence Erlbaum Associates.
Maffi, L., T. Skutnabb-Kangas and J. Andrianarivo

1999 Linguistic diversity. In: D. A. Posey (ed.), Cultural and Spiritual Values
of Biodiversity. A Complementary Contribution to the Global Biodi-
versity Assessment, 19-57. London: Intermediate Technology Publi-
cations, for and on behalf of the United Nations Environmental Pro-
gramme.
Maffi, L. (ed.)
2001 On Biocultural Diversity. Linking Language, Knowledge and the En-
vironment, Washington, D.C.: The Smithsonian Institute.
May, S.
1999 Language and education rights for indigenous peoples. In: S. May (ed.),
Indigenous Community-based Education, 42-66. Clevedon: Multilin-
gual Matters.
* 2001 Language and Minority Rights. London: Longman.
Mazrui, Alamin
1997 The World Bank, the language question and the future of African edu-
cation. Race and Class 38 (3): 35-48.
McRae, K. D.
1997 Conflict and Compromise in Multilingual Societies. Finland. Water-
loo, Ontario: Wilfried Laurier University Press.
Mühlhäusler, P.
1996 Linguistic Ecology. Language Change and Linguistic Imperialism in
the Pacific Region. London: Routledge.
Phillipson, R.
* 1992 Linguistic Imperialism. Oxford: Oxford University Press.
Phillipson, R. and T. Skutnabb-Kangas

* 1995 Linguistic rights and wrongs. Applied Linguistics 16 (4): 483-504.
*1996 English only worldwide, or language ecology. TESOL Quarterly:
429-452. (T. Ricento and N. Hornberger (eds.), Special Topic Issue
Language Planning and Policy).
Phillipson, R. (ed.)
* 2000 Rights to Language. Equity, Power and Education. Mahwah, NJ: Law-
rence Erlbaum Associates.
Posey, D. A. (ed.)
1999 Cultural and Spiritual Values of Biodiversity. A Complementary Con-
tribution to the Global Biodiversity Assessment. London: Intermediate
Technology Publications, for and on behalf of the United Nations En-
vironmental Programme.
Robinson, C. D. W.
1993 Where minorities are in the majority: language dynamics amidst high
linguistic diversity. In: K. de Bot (ed.), Case Studies in Minority Lan-
guages. AILA Review 10, 52-70.
Sasse, H.-J.
1992 Theory of language death. In: M. Brenzinger (ed.), Language Death:
Factual and Theoretical Explorations with Special Reference to East
Africa, 7-30. Berlin: Mouton de Gruyter.
Skutnabb-Kangas, T.
1988 Multilingualism and the education of minority children. In: T. Skut-
nabb-Kangas and J. Cummins (eds.), Minority Education: from Shame
to Struggle, 9-44. Clevedon, UK: Multilingual Matters.
1996a The colonial legacy in educational language planning in Scandina-
via - from migrant labour to a national ethnic minority? International
Journal of the Sociology of Language 118: 81-106.
1996b Educational language choice - multilingual diversity or monolingual re-
ductionism? In: M. Hellinger and U. Ammon (eds.), Contrastive Socio-
linguistics, 175-204. Berlin/New York: Mouton de Gruyter.
1998 Human rights and language wrongs - a future for diversity. Language
Rights 20 (1): 5-27. (Special issue: Language Sciences.)
1999a Language attrition, language death, language murder - different facts or
different ideologies? In: A. F. Christidis (ed.), 'Strong' and 'Weak' Lan-
guages in the European Union. Aspects of Linguistic Hegemonism,
59-73. Thessaloniki: Centre for the Greek Language, Aristotle Univer-
sity of Thessaloniki.
1999b Linguistic diversity, human rights and the 'free' market. In: M. Kontra,
R. Phillipson, T. Skutnabb-Kangas and T. Varady (eds.), Language: a
Right and a Resource. Approaching Linguistic Human Rights,
187-222. Budapest: Central European University Press.
* 2000a Linguistic Genocide in Education - or Worldwide Diversity and
Human Rights? Mahwah, NJ: Lawrence Erlbaum Associates.
2000b The relevance of educational language rights in the EU enlargement
debate. Available at http://www.ecmi.de/activities/minority_congress
_2000_speeches.htm
Skutnabb-Kangas, T. (ed.)
1995 Multilingualism for All. Lisse: Swets and Zeitlinger.
Skutnabb-Kangas, T. and Robert Phillipson
* 1994 Linguistic human rights, past and present. In: T. Skutnabb-Kangas and
R. Phillipson (eds.), in collaboration with M. Rannut. Linguistic
Human Rights. Overcoming Linguistic Discrimination, 71-110. (Con-
tributions to the Sociology of Language 67.) Berlin/New York: Mouton
de Gruyter.
Skutnabb-Kangas, T. and Robert Phillipson

1997 Linguistic human rights and development. In: C. J. Hamelink (ed.),
Ethics and Development. On Making Moral Choices in Development
Co-operation, 56-69. Kampen, The Netherlands: Kok.
1998 Linguistic human rights. Gazette. The International Journal for Com-
munication Studies 60(1): 27-46. (Special volume: Human Rights.)
Skutnabb-Kangas, T. and R. Phillipson (eds.), in collaboration with M. Rannut

1994 Linguistic Human Rights. Overcoming Linguistic Discrimination.
(Contributions to the Sociology of Language 67.) Berlin/New York:
Mouton de Gruyter.
Stoel, Max van der

1996 The Hague Recommendations Regarding the Education Rights of
National Minorities and Explanatory Note. The Hague.
Thornberry, P.
1997 Minority rights. In: Academy of European Law (ed.), Collected
Courses of the Academy of European Law, 307-390, Volume VI,
Book 2. The Netherlands: Kluwer Law International.
Tomaäevski, Κ.
1996 International prospects for the future of the welfare state. In: Reconcep-
tualizing the Welfare State, 100-117. Copenhagen: The Danish Centre
for Human Rights.
de Varennes, F.
1996 Language, Minorities and Human Rights. The Hague/Boston/London:
Martinus Nijhoff.
Williams, E.
1998 Investigating Bilingual Literacy: Evidence from Malawi and Zambia.
(Education Research No. 24.) London: Department For International De-
velopment.
Wong Fillmore, L.
1991 When learning a second language means losing the first. Early Child-
hood Research Quarterly: 323-346.
1. Try to assess the education of one indigenous people (if there are any), one
national minority and a few immigrant minorities in your country. Where
on the continuum from linguistic genocide to full linguistic human rights
would you place their education. Argue for your choice.
2. Check what international and regional human rights conventions your
country has signed and ratified (use the relevant websites mentioned in the
Notes). Look at the articles in them which refer to language, especially lan-
guage in education. Do you think your country and a school of your choice,
one you know fairly well, live up to the requirements in the human rights
documents? If not, why not?
3. Read the whole text (with the Explanations) of the Hague Recommen-
dations Regarding Minority Education Rights. Does minority education in
your country live up to the requirements in relation to national and immi-
grant minorities? If not, what should be done?
4. Find all the arguments in the article that you agree with. Why do you agree?
Then find those that you disagree with. Why do you disagree? If you work
in a group, choose a few arguments where some in the group agree and
some disagree. Have a debate. After the debate, assess where you feel you
did not have enough knowledge to be able to defend your arguments. Where
can you get more knowledge?
5. What are your feelings when you read about 1. linguistic genocide, and
2. the threat to the planet from the killing of both biodiversity and linguistic
diversity, and the consequences for our future. How can you use these feel-
ings constructively?
Core values and nation-states
J. J. Smolicz
1. Introduction
Tenacity of national identity and attachment to the language and culture of

one's own group are evident in most parts of the world, as much in "estab-
lished" nations, such as the English and the French, as in the newly indepen-
dent nations, eager to consolidate their own states, which have emerged from
the ruins of the former Russian / Soviet empire. The first act of independent
Moldova was to discard Cyrillic from the writing of its Romance language,
much as Lithuania had jumped at the chance to do the same, even before the
collapse of Tsarist Russia during World War I (Royal Institute of International
Affairs 1970). Nor should newly emerged national identities be invariably per-
ceived as destructive as opposed to those older ones which have succeeded in
being grounded into a nation-state (Davies 1997). The dedication to national
identity, its acceptance and activation, is so ubiquitous, so deep and capable of
producing such powerful movements that it is futile to pretend that it can be
eliminated as anachronistic or dismissed as a mere invention of the elite (Con-
nor 1993, 1994).
The paradox of ethnic or national upsurge in a globalising culture and econ-
omy, has puzzled many academics and opinion leaders. In the post World War
II climate, ethnic separatist movements within nation-states dominated by the
established nations were labelled as a carry-over from the past, to be found
mainly in the under-developed, backward, non-democratic parts of the world.
The Western ideological blindness to its own problems was succintly illus-
trated by McGarry (1995) with special reference to Northern Ireland, while
similar examples could be provided from Corsica, the Basque country or even
from the uncertain fate of some of the more recent immigrant groups (Skut-
nabb-Kangas and Phillipson 1996). Ethnic tensions are indeed world-wide,
many of them being of long standing as, for example, the situation of Muslims
in the Southern Philippines which has confronted first the ruling Spaniards,
later the Americans, and finally the Republic of the Philippines (Zialcita 1995).
William Safran (1995: 2) concisely summarizes the present understanding
of the current ethnic pluralist phenomenon in his statement that a "consensus
has emerged that the state cannot cope 'neatly' with ethnic reality short of dis-
70 J. J. Smolicz
posing of it by expulsion, extermination, ghettoization, forcible assimilation

and other methods now widely considered to be oppressive, undemocratic or at
least 'inelegant'. But there is consensus about little else". It is a contention of
this paper that it is the traditional model of monoethnic and monolingual
nation-state that constitutes the main problem, due to its inability to respond to
a pluralist national reality based upon the diverse cultural and linguistic heri-
tage of the country's constituent groups.
2. European ideal of a monolingual nation-state
The attempt to build mono-lingual nation-states in Europe has often been jus-
tified by the identification which many peoples have developed between the
consciousness of being a distinct national group and their native tongue. It is
widely recognised that a language often becomes the symbol of a group's iden-
tity and the core value of its culture. This link between cultural identity and lan-
guage persists to this day for a great number of national/ethnic groups and is
likely to continue in the future. It is difficult to deny the cohesive power of a
language as a building block of national solidarity, around which people can
gather, especially in times of stress or danger to their survival.
What is at question therefore is not whether a language may be regarded as
the most important element of a given culture, its carrier and core value (Smo-
licz 1981,1991,1995), but whether this identification between language and a
cultural group requires in each instance the creation of a separate political en-
tity for its protection and development. Is it possible for a number of linguistic
groups to co-exist within the same political organism, without fear that their
cultural heritage is thereby so threatened that the only recourse to ensure sur-
vival is to separate into another state, however precarious its economic base or
geographic position. According to the argument presented in this paper there is
no intrinsic link or nexus between a language and a state. This view is built
upon a clear differentiation between a nation, for which a particular language
can be a core value, and a state which may harbour more than one language and
more than one cultural group (Znaniecki 1952; Smolicz 1990, 1997).
3. State, nation and linguistic pluralism
The state, as a political and territorial entity, is concerned with managing politi-
cal, economic and legal structures, including issues pertaining to citizenship.
Core values and nation-states 71
The nation, on the other hand, is identified by its members with a range of cul-
tural values which are perceived as reflective of their past and as influencing
both their present existence and future development. Only very few states now
can with justification claim the title of nation-state, in which there is a near per-
fect fit between the single nation and a state. Not more than a dozen out of some
186 states in the modern world would qualify. The problem is that even among
multi-ethnic countries, which clearly are the main type of state in the world
today, the politically and culturally dominant group often still perceives the
state as the extension of its own nation (Smolicz 1990; Connor 1993).
Rabinowitz (2001: 64-64) succinctly exposes this false assumption by
showing how it hinges on a non-problematized division of the globe into a
series of idealized "ultimate territories" each ostensibly forming a coherent
homogenous and representative entity, implying a perfect fit between "terri-
tory", "the people living in it" and "their culture", all contained withn the em-
bracing "superstructure of the state". He labels this nation-state construct as
etatism, an "umbrella of ideological and political etatist cliches (which is) par-
ticularly predatory regarding ethnic and national minorities".
Recent history has confirmed that such mono-centrism can lead to disas-
trous results, or constitute a time bomb which multi-ethnic countries neglect at
their peril. Attempts to suppress minority languages are not only degrading of
human dignity, but also an invitation to separatism, and an incitement to frag-
mentation into mini-states. The result may be suffering, not only for subordi-
nate groups, but also for the dominant group which is attempting to achieve a
monolingual state by imposing its own language on all others.
The danger point is reached when minority groups are placed under such
heavy pressure by the dominant one that they lose faith in its intentions and feel
so insecure in the common state as to perceive that separation is the only alter-
native to the ultimate demise of their linguistic heritage. It is up to the majority
then to convince the minorities of the security of their languages and cultures
within the state. If the mutual confidence is ever utterly broken (as it has been in
Bosnia, Chechnya, Rwanda or the Middle East), the culturally creative power
of groups is redirected to destructive pursuits that consume energies which
could have been used to advance the languages and cultures of all concerned.
It is to avoid such a situation that the pluralist solution postulates the exist-
ence of a state that upholds the principles of both political and cultural democ-
racy. Political democracy argues that all permanent residents should be given
an opportunity to acquire citizenship with full political rights. Cultural democ-
racy requires that the cultural and linguistic rights of all citizens be guaranteed
within a framework of overarching or shared values, which include such items
as democratic political institutions and the rule of law (Smolicz 1997). While
72 / . J. Smolicz
such shared values normally involve one or more lingua francas for communi-
cation between all citizens, there is no implication of a monopoly for such lan-
guages, rather safeguards are provided for the development of other regional or
community languages in a variety of domains, including in particular all levels
of education. Following the restoration of democracy, Spain has provided an
example of the way cultural and political democracy can be achieved in prac-
tice, through a variety of constitutional guarantees to its autonomous commu-
nities which are intended to protect their ethnospecific languages as core values
of their groups, while recognizing the role of Castillian/Spanish as the lan-
guage of the whole country. The unsolved problems which remain in the
Basque parts of the country illustrate the danger of long periods of suppresion
and the minority's consequent frustration which cannot be easily forgotten,
even if eventually conditions are created to correct the former abuses (Conversi
1997).
4. Nation-state and globalisation
The accumulation of "pluralist dilemmas" over the recent years in both Europe
and the post-colonial world demonstrates with increasing clarity that the
nation-state, in the form in which it was conceived in Western Europe - mono-
national and equipped with all the culturally centralising powers of intact sov-
ereignty - is a thing of the past. The combined impact of economic global-
isation and ethnic revival is eroding much of its cultural and political omni-
potence. Over time much of its former charisma has faded. The paradox is that
the process of the decharismatisation of the nation-state is proceeding under
the simultaneous impact of the denationalising of the economy and the re-
nationalising of politics (Sassen 2000). According to Eisenstadt (2000: 108):
The growing diversification of the understanding of modernity (...) (has moved) far
beyond the homogenic and hegemonic vision of modernity that were prevalent in the
fifties (...) At the same time the various components of modern life and culture were
refracted and reconstructed in a way which went beyond the confines of any institu-
tional boundaries, especially those of a nation-state - giving rise to a multiple pattern
of globalization.
The effects of such global multiple trends upon national/ethnic configurations

have been very complex. On the one hand, they have moderated the national-
ism among traditional "post-industrial" nations (Dogan 2000), while on the
other, they have highlighted conflicts arising from the long term subjugation of
national/ethnic groups in Central and Eastern Europe and from European col-
onisation in Asia and Africa. These seemingly contradictory phenomena have
been singled out by Dogan (1993, 2000) as a sign of an "asynchronic" situ-
ation, whereby a decrease of nationalist sentiments in the West is contrasted
with its rise in the non-Western world (Asia, Africa and Eastern Europe). The
rise of nation-states and the associated doctrine of nationalism (both deemed to
have been generated in Western Europe), are assumed by this author to have
reached a state of maturity in the West, enabling it to achieve growing detach-
ment from the former nationalist ideologies.
In contrast, the non-Western world, without nation-state doctrines until a
much later date, is perceived to be still in an immature state and hence much
more liable to nationalist sentiments. Such a vision of the West, apparently al-
ready liberated from the extremes of nationalism, can be regarded as distinctly
Euro-centric, since it focuses upon the more positive mutual relations between
dominant European nations (such as France, Germany and possibly the UK)
and ignores ethnic tensions within both historical / territorial and migrant mi-
norities. This positive image of the West omits the distinct lack of eagerness on
the part of the traditional nation-states to respond positively to the aspirations
of "submerged" nations, or ethnic groups within their midst.
Dogan's vision on the maturation of the Western model of nation-state in
Europe, together with the implied need to have the process replicated else-
where in the world as the only remedy against the bacillus of nationalism is
contradicted by those Asian scholars that claim that there is no need to acquire
the infection in the first place (Ooman 1997a; Tombiah 1986). The model of
nation-state developed in the West, for all its claimed egalitarian and unifying
virtues, has left a trail of subjugation and cultural destruction in its wake, with a
consequent reaction against it, even in those parts of Western Europe that es-
caped enforced homogenisation by the dominant nation-state. These "sub-
merged nations" are at long last claiming their cultural rights and searching in a
unified Europe for an escape route from cultural suffocation within the con-
fines of the mono-national nation-state (Grant 1997). Indeed, it is through the
incipient de-nationalisation of the nation-state's territory, that the forces of
unification and globalisation have helped to create conditions for the renais-
sance of local cultures and ethnic identity.
There is surely no need for Asian countries to retrace this pathway at a time
when European nations themselves are discovering the limitations and draw-
backs of the mono-national formula and searching for an all-European inter-
culturalism that holds the prospect of forming bonds based upon cultural ex-
periences that cross existing borders, embracing all groups, both dominant and
non-dominant.
74 J. J. Smolicz
The solution to pluralist dilemmas is not to be found in the fragmentation of

multi-ethnic states into ever smaller and less viable nation-statelets. Safran
(1995:2) argues that "ethnic pluralism need not be state destroying"; that "ethnic
minorities, far from being impediments to modernisation, may in fact be as mod-
ern - if not more modern - than 'national' majorities and that they may be im-
portant modernising agents". Furthermore, he contends that "members of com-
munities who identify themselves ethnically are not ipso facto striving for sep-
aratism or even weaker forms of political autonomy". This is in agreement with
Ooman's (1997b: 6) view that "the assumptions that each nation strives to have
its own state and that a national culture would be safe only within its own state
are unsustainable when viewed in a historical and comparative perspective".
To avoid separatism, minorities must be assured that cultural survival does
not depend on escaping from a common state. The positive approach is to con-
struct a culturally pluralist state, as exemplified in their diverse ways by coun-
tries as different from one another as Spain, Australia and India, or through the
gradual incorporation of nations within a common supra-national entity that re-
spects cultural and linguistic diversity within the overall community, as well as
within each of its component parts. The preservation of such diversity is de-
pendent on the maintenance and development of cultural core values of each of
the constituent national/ethnic groups. Such core values are seen as indispens-
able elements of the group's identity, but their existence is compatible with the
development within the nation-state of an overarching framework of shared
values, built upon multicultural rather than monistic premises.
5. Australia as a multicultural nation-state
Australia can be taken as an example of a country which claims to be building a

multicultural nation and state on the basis of the cultural contributions of both
majority and minority groups. After the post war attempt to build a monoethnic
nation-state along Anglo-Saxon lines proved impossible to achieve, there was
initially a tendency to avoid the term "nation" because of its negatively per-
ceived "nationalist" connotations, and to replace it with "society", especially in
the use of the phrase "multicultural society". Where the term nation has been
used, it has been understood as "civic society", with reference to national iden-
tity often replaced by "civic identity" instead (Home 1994). Kloskowska
(1997: 70) points out, however, that the concepts of nation and society are not
synonymous, with the term society "displaying qualities which are instrumen-
tal, practical, concrete that do not require the existence of an inner bond or of
cultural unity in the other spheres, nor call for the emotional attitudes" en-
countered in the case of a nation. For this author 'nation' demands the presence
of common cultural allegiance, creating a sentimental and emotional bond ex-
pressed as "love for the nation". From this perspective, it has proved futile to
escape from the concept of a nation altogether, since there is always a danger
that its strong emotional connotations can be usurped in an aggressive way,
giving rise to exclusive nationalism that is hostile to neighbours and minorities
(Huntington 1996).
As the Australian experience shows, for a multi-ethnic state to survive and
develop along multicultural lines, it is necessary for more than the common
political machinery of the state to evolve. It requires also the cultivation of a set
of shared, over-arching cultural values that extend beyond political structures
and include other aspects of culture which not only reflect the majority group's
core values, but also take account of the minority groups' aspirations and their
own particular core values.
Ever since the prime ministership of Malcolm Fraser (1975-1981), multi-
culturalism as a national ideal has sought to uphold and develop such an over-
arching framework in which the right of individuals from minority back-
grounds to maintain their ethnic identity was assured. This approach presup-
poses support for different cultures to contribute to the overarching framework,
while maintaining the central elements or core values that are essential for their
integrity and creative force.
The balance between sharing and diversity rests on the degree of consensus
which has developed on a number of cultural and political issues in areas such
as democratic forms of government, the economy and the law, as well as exclu-
sion from the framework of any requirement for uniformity in relation either to
race and ancestry or to religion. The idealized model to which Australia aspires
is therefore free from the divisions that are most difficult to bridge, as when one
particular religion is made mandatory or when racial or ancestral character-
istics are regarded as exclusion markers that set the limits of nationhood.
However, the Australian model has been interpreted in different ways ac-
cording to the degree of multiculturalism that people have been prepared to ac-
cept. Some have perceived the shared cultural framework to be essentially dy-
namic in its capacity to adjust to the existing, as well as the future, complexity
of the population. They point to the fact that the framework has already proved
its flexibility through the belated political incorporation of Aboriginal-Austra-
lians within the arch as voters since 1967, and the abolition of the White Aus-
tralian Policy, sealed by the acceptance of the Vietnamese refugees during the
1970s (Jupp 1988), even if much remains to be achieved in the sphere of "rec-
onciliation" between the descendants of immigrants and the indigenous popu-
lation. Others, afraid of fragmentation, have much more limited notions of plu-
76 J. J. Smolicz
rality and prefer the framework to be grounded mainly in Anglo-Saxon or

Anglo-Celtic core values (Bullivant 1981). Minority cultures are then expected
to contribute only peripherally, chiefly in relation to food and the celebration of
colorful customs and festivals. In the extreme residualized form, where minor-
ity ethnic core values are largely dissipated, this model ceases to be multicul-
tural - an aim openly propagated by the One Nation Party which currently has
the support of approximately 10% of the Australian electorate.
At present, the multicultural model officially prevails and has been sup-
ported both by formal resolutions passed in the Houses of Parliament and by
statements of the Governor-General of Australia (Deane 1997). The ultimate
success of the Australian multicultural nation-state, however, will depend on
the degree to which the core values of the minority groups, including the in-
digenous population, are given the opportunity to develop and flourish, not
only within the confines of the home and local community but also in the public
arena, and in particular to be included in the education of the next generation of
Australians. What are the core values which need to be maintained and devel-
oped in order to sustain the multicultural nature of a pluralist state? The answer
to this question, which may vary from group to group, requires a more compre-
hensive analysis of the concept of core values itself.
6. A core values perspective
The concept of core values was originally proposed by the author in his earlier
studies of minority cultures in Australia in collaboration with M. J. Secombe
(Smolicz 1981, 1992; Smolicz and Secombe 1985, 1986, 1989). The concept
has been developed with special reference to those values that are regarded as
forming the most fundamental components of a group's culture, symbolic of
the group and its membership. From both the insiders' and outsiders' perspec-
tive, it is core values that enable social groups to be identified as distinctive cul-
tural communities. The idea of core values has proved of interest to other re-
searchers in a number of contexts, for example, Baker and Prys-Jones (1998:
114-115); Clyne (1991:91-105); Conversi (1990,1997:165-166); Fedorowicz
(1997: 83-86), Stern (1988), Moore (1984) and Skutnabb-Kangas (1986: 165).
The particular core values which act as 'markers' of a group's identity have
been identified as, for example, ethno-specific language, family structure, re-
ligious beliefs or love of the homeland. The nature of core values can be most
clearly discerned when the group concerned is under threat and needs to defend
its culture against external pressures. The Greek and Latvian languages, for
example, can be seen to have emerged particularly clearly as core values for
their respective groups, when they were under threat of suppression by neigh-
bouring powers. These groups were, therefore, already familiar with commu-
nity strategies which they could put in place to ensure the survival of what they
regarded as the critical element of their culture in the pluralist immigration
countries such as Australia, United States or Canada (Smolicz 1999).
Our studies in Australia would appear to confirm that ethnic groups show
important variations in the aspects of culture which could be regarded as their
core values. Clyne (1991: 91-92), has commented at length on the core value
theory, which he found "tempting because of its ready applicability to the ex-
planation and prediction of language maintenance trends among ethnic
groups". He has, however, identified problem areas, such as the "possible dis-
continuity between language use and language attitudes", but noted that subse-
quent research by Smolicz and Secombe (1989) had "very successfully ad-
dressed the (possible) discrepancy (...) (by) examining a wide range of ethnic
groups" and differentiating between negative indifferent, general positive and
personal positive evaluations of the native tongue, in contradistinction to its ac-
tual usage.
7. Core values of the Australian ethnic minorities
Studies on ethnic cultural diversity in Australia have shown the way in which
some groups focus predominantly upon one particular aspect of their culture,
which they wish to defend at all costs, while in the case of others, it is possible
to identify two or even three foci, with the chances of group survival enhanced
through mutually supportive core values, such as family and religion, or family
and language (Smolicz, Hudson and Secombe 1998; Smolicz, Secombe and
Hudson 2001). Greek-Australians have proved particularly successful in com-
bining the maintenance of their language, with their collectivist family patterns
and their reliance upon Greek Orthodox Church, to produce a triad of core val-
ues. A number of our Greek-background respondents pointed to the advantages
of this triad of values in resisting the assimilation pressures of the mainstream
society, while integrating into its civil, economic and political structures.
In the case of the Polish-background respondents, there appeared to be two
core values which went hand in hand - collectivist family patterns and the
Polish language. The Catholic church, to which most Polish-Australians be-
longed, did not seem to function as a core value in the same way, as the Greek
Orthodox church for the Greek-Australians. In spite of the wide-spread net-
work of Polish-Catholic chaplains who held religious services in Polish, Ca-
tholicism, with its hierarchical territorially-based parishes and dioceses, was
78 J. J. Smolicz
not specific to the Polish people alone, but was shared with others, particularly
the dominant mainly Irish-origin majority within the (O'Farrell 1986; Smolicz
1994; Smolicz and Secombe 1981).
For the Italian-background respondents, the family appeared to be the pre-
eminent core value with varying significance ascribed to other values. Al-
though the more highly educated respondents, as well as those from specific re-
gions of Italy, accorded high value status to the standard forms of the literary
language, most of the respondents from Southern Italy were content to modify
and adjust their language usage in order to ensure effective communication in
the family. Such subordination of the language to family life could be taken as
indicative of the relative importance of family bonds within this immigrant ori-
gin group (Chiro and Smolicz 1997, 1998).
In the case of the Chinese-background respondents, Mandarin was recog-
nised as a vital symbol of Chinese ethnic identity for Chinese people from dif-
ferent parts of China and Southeast Asia, as well as from a variety of socio-
economic backgrounds. The Chinese family in its collectivist form, however,
remained the bedrock of Chinese cultural survival in Australia. It was accepted
by many of our respondents as a core value which was so manifest that it hardly
needed any prolonged discussion (Smolicz, Secombe and Hudson 2001).
The responses of the Latvian-Australians pointed to a different set of core
values, which did not include collectivist family structures. The very high
status accorded to Latvian language as the core value of the group's culture was
evident in the insistence that the second and third generations master it at the
literary level, an expectation which even surpassed the importance accorded to
it as a language of everyday usage in the family. This evaluation of Lativian
language by the respondents and the desire to preserve it appeared to reside in
their consciousness of being entrusted with the cultural heritage of the Latvian
people. In the face of russification in their Soviet-ruled homeland, these Lat-
vian-Australians assumed the role of 'keepers' of the language to save it from
extinction. For this group, the native tongue acquired an autotelic significance
of a value per se (Kloskowska 1985: 22) which functioned as a symbol of
identification for all members, not only among the Latvian community in Aus-
tralia, but also for Latvian people world-wide, both in Latvia itself and in the
diaspora.
The educational and literary activities of the Latvian group could be re-
garded as an illustration of a stage on Fishman's (1991: 95) Graded Intergen-
erational Disruption Scale (GIDS), where a socially integrated and ethnolin-
guistically active community is attempting to ensure not only the intergener-
ational transmission of its informal oralcy but also its reinforcement through
literacy. The Latvian community also showed its ability to organize itself on a
world-wide basis, going well beyond the neighbourhood intimacy of the local
context through the social and educational structures of the World Federation
of Free Latvians which have provided support for language maintenance ef-
forts in countries such as Germany and USA and Australia, with Latvia itself
becoming increasingly the focus of such activities following its liberation from
Soviet rule (Darzins 2000).
Australian research on linguistic minority communities confirms Fishman's
(1991: 413) contention of the core importance of the home-family-community
as the basic institutions in minority language maintenance, although the "com-
munity" did not invariably correspond to the "little culture" of the local neigh-
bourhood, but could stand for a national entity, which preserved its cultural
cohesiveness in spite of being scattered over many countries and continents.
These studies demonstrated how the linguistic maintenance efforts of minority
ethnic groups were intertwined with their familial and other ethnic social struc-
tures in ways that reflected their core values, acting as symbols of their ethnic
identity within the context of a multicultural Australia.
The extent to which the minorities' linguistic core values are maintained in
Australia can be seen to depend on several inter-related factors. These include
not only the tenacity of a particular group's members in upholding their histori-
cally evolved and continually revived core values, but also the group's current
geographical situation, which has been demonstrated, for example, in the
special efforts to maintain Latvian and other Baltic languages. There has also
been a growing realisation of the Australian indigenous communities' almost
total isolation from the rest of the world and the way past linguistic suppression
has undermined their way of life. This is now resulting in a determined effort to
reclaim their languages, including those such as Kaurna on the Adelaide Plains
of South Australia, which had already ceased to be spoken in the course of the
nineteenth century (Amery 2001). Other factors reviewed above include the
presence of other language - supporting core values, such as family structure
and ethno-specific religion. The majority group's cultural, social and edu-
cational attitudes and policies also have a bearing on the ability of minorities to
arrest or even reverse language shift to English. Related factors include the ex-
tent of overlap between the minority and majority cultural values, taken as a
whole, and the extent to which members of the minority are deemed to be so-
cially acceptable by the majority in terms of primary social relationships. The
final factor which will influence the vitality of minority linguistic cores will be
the educational policy pursued by the country in question.
Although the existence of the plurality of languages in Australia is now
probably more accepted than at any time since the period of laissez faire plu-
ralism in the mid-nineteenth century, toleration of languages is less firmly en-
80 J. J. Smolicz
trenched in the Australian ethos than religious pluralism. Minority languages

are taught in many schools and ethnic schools run by the minority communities
receive government subsidies. Nevertheless, it is difficult to assess the extent to
which schools have contributed to stabilising the country's linguistic pluralism.
For practical purposes English dominates the scene, even when minority com-
munity languages are taught as school subjects in mainstream or independent
schools. Those minority groups which depend upon their language as their
main core value are more vulnerable than those which are anchored in other
cultural core values, such as religion or family structure.
Despite therefore the significant multicultural reforms in Australian
schools, the multilingualism of the country, from the perspective of community
languages appears to be fragile, although new migration waves and changing
policiess may alter this sutuation. The absence of ethnic strife and the existence
of policies that favour cultural as well as political democracy create a milieu
which is favourable to minority cultural mantenance, with an ethnically plural-
ist state placing upon individuals and their groups the onus of the extent to
which they wish to maintain their cultural distinctiveness.
The dramatic changes in the concept of "Australianess" and what consti-
tutes a "real Australian" demonstrate a radical departure from the former
monolingual and mono-ethnic assumptions that prevailed in that country from
the time of establishment of the Australian Federation in 1901 until the mid
1970's. They suggest the possibility of a successful rejection of the European
ideal of the monolingual nation-state in favour of acknowledging the right of
minorities to their cultural heritage, exemplified by their core values, when
taken within the framework of overarching values which are shared by the Aus-
tralian nation as a whole.
References
Amery, R.
2001 Reclaiming the landscape: Kaurna naming activity, signage and public
art in Adelaide and Adelaide Plains. Paper delivered at the conference
of Australasian Universities' Language and Literature Association
(AULLA), Adelaide University, February.
Baker, Colin and Sylvia Prys-Jones (eds.)

1998 Encyclopedia of Bilingualism and Bilingual Education. Clevedon:
Bullivant, Brian M.
1981 The Pluralist Dilemma in Education. Sydney: Allen and Unwin.
Chiro, G. and J. J. Smolicz

1997 The activation and evaluation of Italian language and culture among
tertiary students of Italian ancestry in Australia. In: R. Watts and J. J.
Smolicz (eds.), Cultural Democracy and Ethnic Pluralism, 213-242.
Frankfurt am Main: Peter Lang.
1998 Evaluation of language and social systems by a group of tertiary stu-
dents of Italian ancestry in Australia. Altreitalie 18: 24-65.
Clyne, Michael
1991 Community Languages: The Australian Experience. Cambridge:
Cambridge University Press.
Connor, Walker
1993 Beyond reason: the nature of the ethnonational bond. Ethnic and
Racial Studies 16 (3): 373-389.
* 1994 Ethnonationalism: The Questfor Understanding. New Jersey: Prince-
ton University Press.
Conversi, Daniele
1990 Language or race? The choice of core values in the development of
Catalan and Basque nationalisms. Ethnic and Racial Studies 13: 5-70.
:
1997 The Basques, the Catalans and Spain: Alternative Routes to National-
ist Mobilisation. London: Hurst.
Darzins, D.
2000 Personal Communication from President of the Associations of Lat-
vian Organisations in South Australia.
Davies, N.
1997 West best, east beast. Oxford Today 9 (2): 28-31.
Deane, W. (Sir)
1997 Multiculturalism our Australian way. Multicultural Life 2: 3.
Dogan, Mattei
1993 Comparing the decline of nationalism in Western Europe: the gener-
ational dynamic. International Social Science Journal 36: 177-198.
2000 Nationalism in Europe: Decline in the West, Revival in the East. In:
Ε. K. Scheuch and D. Sciulli (eds.), Societies, Corporations and the
Nation State, 181-200. (The Annals of the International Institute of
Sociology, New Series, 7.) Leiden/Boston/Köln: Brill.
Eisenstadt, Shmuel N.
2000 The Contemporary Scene - Multiple Modernities. In: Ε. K. Scheuch
and D. Sciulli (eds.), Societies, Corporations and the Nation State,
97-108. (The Annals of the International Institute of Sociology, New
Series, 7.) Leiden/Boston/Köln: Brill.
82 J. J. Smolicz
Fedorowicz, Η. M.
1997 The multicultural society in Central Europe: Towards a communi-
cation strategy. In: R. J. Watts and J. J. Smolicz (eds.), Cultural
Democracy and Ethnic Pluralism, 77-94. Frankfurt am Main: Peter
Lang.
Fishman, Joshua A.
* 1991 Reversing Language Shift: Theoretical and Empirical Foundations of
Assistance to Threatened Languages. Clevedon: Multilingual Matters.
Grant, N.
1997 Democracy and cultural pluralism: towards the 21st century. In:
R. Watts and J. J. Smolicz (eds.), Cultural Democracy and Ethnic Plu-
ralism, 25-50. Frankfurt am Main: Peter Lang.
Hörne, D.
1994 Public Culture: an Argument with the Future. Pluto Press.
Huntington, Samuel
1996 The Clash of Civilizations and the Remaking of World Order. New
York: Simon and Schuster.
Kloskowska, Antonina
1985 Symbolic values in social conflicts. The Polish Sociological Bul-
letin 4: 19-29.
1997 Skad i po co narod? [The nation - from where and for what?] Znak 49
(502): 69-78.
Jupp, J.
1988 The Australian People: an Encyclopedia of the Nation, its People and
their Origins. Sydney/London: Angus and Robertson.
McGarry, John
1995 Explaining Ethnonationalism: The Flaws in Western Thinking.
Nationalism and Ethnic Politics 1 (4): 121-142.
Moore, J.
1984 Exporting European core values: British and French influences on
education in Mauritius. European Journal of Education 19: 39-52.
O'Farrell, Patrick J.
1986 The Irish in Australia. Sydney: New South Wales University Press.
Ooman, Τ. K.
1997a Citizenship, Nationality and Ethnicity: Reconciling Competing Iden-
tities. Cambridge: Polity Press.
1997b Citizenship, Nationality and Ethnicity: The Context of Globalisation,

Paper presented at the World Congress of the International Institute of
Sociology, Köln, Germany.
Rabinowitz, D.
2001 The Palestinian citizens of Israel: the concept of trapped minority and
the discourse of transnationalism in anthropology. Ethnic and Racial
Studies 24 (1): 64-85.
Royal Institute of International Affairs

1970 The Baltic States. Westport, CT: Greenwood Press.
Safran, W.
1995 Nations, ethnic groups, states and politics: A preface and an agenda.
Nationalism and Ethnic Politics 1 (1): 1—10.
Sassen, S.
2000 Beyond Sovereignity: De-Facto Transnationalism in Immigration Pol-
icy. In: Ε. K. Scheuch and D. Sciulli (eds.), Societies, Corporations
and the Nation State, 128-148. (The Annals of the International Insti-
tute of Sociology, New Series, 7.) Leiden/Boston/Köln: Brill.
Skutnabb-Kangas, T.
1986 Conflicting paradigms in minority education research. In: B. Spolsky
(ed.), Language and Education in Multilingual Settings, 153—181.
Clevedon: Multilingual Matters.
Skutnabb-Kangas, T. and R. Phillipson.
1996 Minority Workers or Minority Human Beings? A European Dilemma.
International Review of Education 42 (4): 291-307.
Smolicz, J. J.
* 1981 Core values and cultural identity. Ethnic and Racial Studies 4 (1):
75-90.
1988 Tradition, core values and intercultural development in plural so-
cieties. Ethnic and Racial Studies 11 (4): 387-410.
1990 The mono-ethnic tradition and the education of minority youth in Ger-
many. Comparative Education 26 (1): 27-44.
* 1991 Language core values in a multicultural setting. International Review
of Education 37 (1): 35-52.
1994 Multiculturalism, religion and education. Education and Society 12:
22-47.
1995 Language - a bridge or a barrier? Languages and education in Aus-
tralia from an intercultural perspective. Multilingua 14 (2): 151-182.
84 J. J. Smolicz
1997 Australia: from migrant country to multicultural nation. International

Migration Review 31 (1): 164-186.
* 1999 (M. J. Secombe and J. Zajda, eds.) J. J. Smolicz on Education and Cul-
ture. Melbourne: James Nicholas Publishers.
Smolicz, J. J., D. M. Hudson, and M. J. Secombe
1998 Border crossing in 'Multicultural Australia': A study of cultural val-
ence. Journal of Multilingual and Multicultural Development 19:
318-36.
2001 Family Collectivism and Minority Languages as Core Values of Cul-
ture Among Ethnic Groups in Australia. Journal of Multilingual and
Multicultural Development, London (in press).
Smolicz, J. J. and M. J. Secombe

1981 The Australian School through Children's Eyes: A Polish-Australian
View. Melbourne: Melbourne University Press.
1985 Community languages, core values and cultural maintenance: The
Australian experience with special reference to Greek, Latvian and
Polish groups. In: M. Clyne (ed.), Australia - Meeting Place of Lan-
guages, 11-38. Canberra: Pacific Linguistics, Australian National
University.
1986 Italian language and culture in South Australia: A memoir approach.
In: C. Bettoni (ed.), Altro Polo: Italians Abroad: Studies on Language
Contacts in English Speaking Countries. Sydney: Frederick May
Foundation for Italian Studies.
1989 Types of language activation in an ethnically plural society. In:
U. Ammon (ed.), Status and Function of Languages and Language
Varieties, 478-514. Berlin/New York: Mouton de Gruyter.
2000 Language resilience and educational empowerment: Philippines and
Australia. In: R. Phillipson (ed.), Rights to Language: Equity, Power
and Education, 164-169. Mehaw, NJ: Lawrence Erlbaum.
Stern, A.
1988 Education policy towards Circassian minority in Israel. Paper deliv-
ered at a conference On the Maintenance and Loss of Minority Lan-
guages. Nijmegen: Institute of Applied Linguistics, University of
Nijmegen.
Tombiah, S. J.
1986 The nation-state in crisis and the rise of ethno-nationalism. In: Ε. N.
McAllister (ed.), The Politics of Difference, 124-143. Chicago: Uni-
versity of Chicago Press.
Zialcita, F. N.
1995 State formation, colonialism and national identity in Vietnam and the
Philippines. Philippine Quarterly of Culture and Society 23: 72-117.
Znaniecki, F.
1952 Modern Nationalities: A Sociological Study of How Nationalities
Evolve. Urbana: University of Illinois Press.
1. Distinguish between the concepts of Nation and State and give examples of
each.
2. Is monolingual, mono-ethnic nation-state out of date?
3. What is the fate of linguistic minorities in a state as (2)?
4. Discuss languages as a core value of culture and provide examples.
5. Discuss cultural groups which may display more than one core value.
II. Bilingualism worldwide
French language policy: centrism, Orwellian
dirigisme, or economic determinism?
Harold F. Schiffman
1. Introduction
To students of language policy, France has what can be considered to be the

most centrist of centrist language policies in the world.1 The idea that centrism
can be effective as a language policy rests upon some notions, however, that
need to be examined. Centrism means of course that decisions are made at the
center, i.e. the center of power in the polity, and as far as France is concerned,
the center/periphery dichotomy is perfectly represented by the language situ-
ation: at the geographic center of the hexagon is the French language, while at
the periphery - Bretagne, le Midi, Corsica, Alsace - there is almost no region
where French is native. The history of France is the history of the spread of
French (le francien) out from the lie de France into these marginal territories.
Dirigisme is defined in my dictionary as the attempt to direct and control
things from one central place, in particular to centrally direct and control a
nation's economy. That is, dirigisme is economic centrism. Though France was
never part of the Soviet bloc, the idea of central planning of the economy in-
volved strong state intervention in economic and financial affairs, going so far
as to construct and attempt to carry out five-year plans, to move and relocate in-
dustry around the country, and do other kinds of things that are more typical of
the Soviet economic model than, e.g., the American one. The idea that lan-
guage could be controlled in this same way, by central decision-making, has
been around in France for a long time, dating to the promulgation of the Or-
donnance de Viller-Cotterets in 1539. The French Revolution gave us other
kinds of intervention on behalf of language, such as measures to annihilate the
idiomes, patois, jargons and other kinds of non-standard French, deemed
counterrevolutionary or even worse. The decrees and ordonnances of the Rev-
olution appropriated and perpetuated the monarchic view that language could
be controlled from the center, and the view that central control of language was
just possibly undemocratic (as it would be seen, e.g., in Anglo-Saxon coun-
tries) has never been part of French language policy.
90 Harold F. Schiffman
2. Goal of this paper
It is one thing to think of decisions being made at the center that affect French
banking, or agriculture, or the railroad system but when the matter is a lin-
guistic one, one must ask oneself whether any government can realistically ex-
pect to control the linguistic habits of its citizenry in any meaningful way.
Though the term dirigisme is used by no-one in France to refer to its linguistic
policies, the parallels between economic dirigisme and what I will call lin-
guistic dirigisme (or dirigisme linguistique) are striking. Using his own system
of analysis that looks at most social interaction as a set of economic exchanges,
Bourdieu (1982) has placed the economic language model at center stage, so
my plan here is to examine both the historical attempts to control language
overtly from the center in France, and the more subtle system of controls that
Bourdieu delineates.
My overall goal is to show how my own notion (Schiffman 1996), namely,
that language policy is embedded in and proceeds from what I call linguistic
culture2, is a larger framework through which to view the role of dirigisme lin-
guistique, which I see as an unquestioned assumption3 of French language pol-
icy and the culture in which that policy is embedded.
3. Review of the literature
French language policy, broadly speaking, has been studied by Ager 1990;
Bedard and Maurais 1983; Catach 1991; de Certeau, Julia and Revel 1975;
Haitz 1988; Gordon 1978; Guilhaumou 1989; Hell 1986; Levy 1929; Philipps
1975, 1978, 1982; Schiffman 1996; Tabouret-Keller and Luckel, 1981; and
Vermes and Boutet 1987. It seems clear to me that many writers about French
language policy impute results and outcomes of the policy to the strong overt
language policy direction that the French state supposedly exerts; as I and
others have shown, however, until recently, France actually lacked the explicit
directives on language that many thought it possessed, but because they and the
French public at large (Catach 1991) believe a myth about these directives and
their explicitness, the French tend to submit to policies they believe to be in ef-
fect, when they were not in fact part of French law.
As for Bourdieu's thinking about language and language policy, we can dis-
cern also a dirigiste view, but with different effects. His ideas on this topic, ex-
pressed most cogently in his 1982 work (Ce que parier veut dire) throw a dif-
ferent light on this issue; but though he takes a "marketplace" approach and
French language policy 91
sees language as a commodity with symbolic value that is exchanged for value
in a larger marketplace of symbolic values, I will try to show that there are as-
pects of his analysis that could only be true for a polity like France, where the
centralized control of "cultural products" (as he puts it) and the creation of a
"unified market" in "linguistic products" have, in his view, made the explicit
language policy work. That is, because of economic determinism and the de-
sire to participate and attain upward mobility in this marketplace, French
citizens accept standard language as part of the cultural capital that increases
their personal worth, resulting in behavior that appears to be controlled by the
central authority. But he specifically abjures the power of the state to make
people speak a certain way, imputing the power of this policy instead to much
more subtle pressures of social structure.
We will examine his ideas in more detail below; first, I would like to look at
the historical background of dirigisme linguistique.
4. Dirigisme and jacobinisme
Though the French do not use the term dirigisme for all examples of the central
controlling of much of French material life, they do have a term for this ten-
dency, namely, Jacobinisme. This term has its origin in one of the revolutionary
factions that had various names ("Friends of the Constitution" or "Friends of
Liberty and Equality"), that met in a former Dominican convent in Paris, the
Dominicans being known locally as Jacobins. The popular term Club des jac-
obins was then applied to this faction, which, though not originally pro-repub-
lican, gradually became more and more radically revolutionary, losing its more
conservative or moderate members after 1792 and the elimination of the mon-
archy, and in fact becoming, through its dominance of the Committee of Public
Safety, and its association with Maximilien de Robespierre, chief architect of
the 'Reign of Terror', emblematic of revolutionary excess and of arrogation of
central control through state-sponsored violence and terrorism. After the fall of
Robespierre (9 Thermidor, an II, or July 27, 1794), the Jacobins were tempor-
arily, and then eventually permanently, banned. Jacobinism, however, whatever
its name, still seems to remain a feature of French political life, and some fea-
tures of it, such as attempts to control language (an enduring project of the
French Revolution), persist to this day.
As I have tried to show (Schiffman 1996), the French Revolution adopted a
policy on language that was very different from the kind of policy that other
democratic nations see as appropriate. I am not the first to notice that the out-
come of the American and the Soviet revolutions were to either disengage gov-
ernment from control of language (i.e. establish in the US a kind of laissez-
faire language policy), or to empower previously unempowered linguistic
groups (such as in the Soviet Union), whereas in the French revolution, as is
well known, languages other than French were disenfranchised, and were
treated as counterrevolutionary activities.4 In fact, Brunot, in his monumental
history of the French language, declares the chief linguistic accomplishment of
the French Revolution to be the firm establishment of the Monarchic language
policy, rather than the elimination of it, as everything else associated with the
monarchy was eliminated.5 That monarchic policy was dirigiste, centrist, auth-
oritarian, controlling, and after the fall of the monarchy, more inflexible than
before. As in other things, it involved the notion of divine right6, and as we can
see from almost any discussion of French language policy, a similar notion per-
vades the topic to this day.7
4.1. Before the Revolution
We must not conclude from the foregoing that French language policy before
the French Revolution was inept and somehow laissez-faire. As is well known,
France had exhibited dirigiste tendencies as early as 1539, when Francois I is-
sued the Ordonnance of Villers-Cotterets, according to which all official docu-
ments were to be henceforward written in "French". Though the issuance of
edicts and ordonnances may have zero effect on actual linguistic behavior, it
legitimizes the notion that such edicts can control something, or will control
something, and that the monarch has the right and authority to control lan-
guage.8
When most people think of language control in France, it is the Academie
frangaise that comes to mind.9 The Academy was founded in 1635, as a private
organization, and then chartered by Richelieu in 1637; its tasks were to estab-
lish a poetic, rhetoric, a grammar, and a dictionary and it was this last task, the
dictionary, that it is most famous for. It arose out of the perception that French
was somehow threatened by Italian and to some extent Spanish; loan words
from these languages were being borrowed freely, and fears that French would
be swamped and even wiped out by these languages prompted the desire to es-
tablish an academy. A language academy, of course, is an example of dirigisme
since it rests on the notion that rules and decisions about language can be made
by a body given the authority to make them, and language users will follow
those rules. It never had a mandate to standardize French, however, and es-
pecially not its orthography; its members disavowed attempts to impose this
task on them.10 Rather, they felt the priority was to greffer 1'usage (lit.: "graft
usage") and in fact "use the best usage". It was hampered by the fact that its
own members failed to follow their own rules or spell French in any uniform
way (Voltaire was famous for this) and the committee of 40 was forced to sub-
contract the task to Vaugelas, who struggled on his own.
4.2. The threat offranglais
What is clear, however, is that even if the French Academy does not see its task
as being one of standardizing the French language, the public at large thinks it
does, and cannot understand why it does not act to save the language from the
incursions of the more modern menace, le franglais.11 As Catach points out
(Catach 1991: 55) with the abdication by the Academy of some of its duties,
the tasks of controlling and standardizing things has in effect devolved upon a
number of bodies - the state itself (the Premier, the Minister of Education) ap-
points commissions, who make reports to the Academy, which discusses them,
etc. etc. Thus the body that has the highest public expectations of it, imbued as
French linguistic culture is of the idea that language can be controlled from a
central body, itself abjures any notion of control, even of something as mun-
dane as spelling. The French state is forced to take action, creating councils,
high commissions, which consult each other, make reports and recommen-
dations, and seem to be following acceptable bureaucratized procedures. But
the responsibility for various things is diffused, and for the public at large, the
perception is that things are mired in red tape.
4.3. Who controls what?
One solution to the language control problem is to control that which can be
controlled, i.e., the written language, and this is of course what many polities,
including France, have fallen back on. The French sociologist Bourdieu has
written extensively about how language, especially language in France, has
come to occupy a place where not only is language controlled, language con-
trols. His view of language and culture is that they constitute a system of ex-
changes, operating in a sort of linguistic market, and by constituting a unified
market-place where only one kind of language is permissible, no French citizen
can escape the system.
Bourdieu's point of view, as far as official language in France is concerned,
converges, in certain ways, with mine, so my thesis, that linguistic dirigisme is
related to economic dirigisme is not so far-fetched. Bourdieu is at great pains to
point out that though centralism is indeed a characteristic of the habitus of the
country, it is not through decrees and ordinances that the state controls language.
The official language is linked with the State. And that is true both in its genesis and
in its social usage. It is in the process of state constitution that the conditions that con-
stitute the unified linguistic marketplace are created and dominated by the official
language: obligatory at official occasions and in official spaces (the school, public
administration, political institutions, etc.), this State Language becomes the theor-
etical norm against which all linguistic practices are measured objectively. No-one is
able to ignore the linguistic law, which has its body of judges, the grammarians, and
its agents of imposition and control, the teachers, who are invested with the power to
make their speaking subjects submit their linguistic performance without exception
to examinations and juridical sanctions of the scholastic kind.12 (Bourdieu 1982: 27)
But, he goes on to note:
For one mode of expression among others (...) to impose itself as the only legitimate
one, the linguistic market must be unified and different dialects (class, regional, or
ethnic) must be measured against the standards of the language or by legitimate
usage. Integration into one single 'linguistic community' (...) is the [necessary] con-
dition for the establishment of the state of linguistic domination. (Bourdieu 1982:28)
Until the French Revolution, the process of linguistic unification was indistinguish-
able from the process of the construction of the monarchical state. (...) The dia-
lects (...) [especially in the north of France, or pays de langue d'oil ] begin to give
way, progressively, from the 14th century onwards, (...) to the common language
which was being elaborated in Paris in cultivated domains, and promoted by a statute
of official language (...) i.e. utilized in the form that educated use had given it, i.e. the
written form. (Bourdieu 1982: 29)
During the revolution, he points out, the dialects were unusable as the 'langue
pratique' for decrees etc. because the available vocabulary didn't have com-
mon meanings; revolutionaries were thus forced to "forge" a common lan-
guage [langue moyenne] with particular attention given to spelling. Thus the
fixation on meanings, of trying to control the official language so that it could
be inculcated into the heads of the new citizens, making them think more
clearly, etc. This concern with these fussy issues still goes on today, resulting in
the movements concerned with the "defense de la langue", which he calls a
kind of "mind control". (Bourdieu 1982)
The imposition of the legitimate language against the idiomes and patois is part of
the political strategies destined to assure the perpetuation of the gains of the Revol-
ution by the production and reproduction of the new man [I'homme nouveau].
Language becomes a "method" which allows people to identify the revol-

utionary language with revolutionary thought: to "reform language, purge it of
usages linked to the former society and [re] impose it thus purified, is to impose
a thought process itself purged and purified." We must not think of political and
linguistic unification simply as a technical process, of communication, or
simply to see it as a kind of statist centralism intent on crushing local particu-
larisms, however. "The conflict between the French of the revolutionary intel-
ligentsia, and the idiomes or patois is a struggle for symbolic power over the
formation and re-formation of mental structures." [emphasis mine] (Bour-
dieu 1982: 31)
Thus,
it is not just a question of communicating, but of recognizing a new discourse of

authority, with its new political vocabulary, its terms of address and of reference, its
metaphors, its euphemisms, and the representation of the social world which it auth-
orizes, and which, because it is linked to the new interests of new groups, is inex-
pressible in the local speech forms fashioned by usages linked to specific interests of
peasant groups. (Bourdieu 1982: 31)
While one must not forget the contribution which political unification movements
(which are visible in other domains, such as that of the law) bring to the fabrication
of the language that linguists accept as a natural given, one must also not impute to it
the entire responsibility for the generalized use of the dominant language, (a) dimen-
sion of the market unification of symbolic goods that accompanies the unification of
the economy, and also cultural production and cultural circulation. It is easy to see in
the case of the matrimonial exchange market (Bourdieu 1982: 35)
What he is referring to here is the marriage market, where linguistic 'products'

until then destined to circulate in the protected enclosure of local markets are
perceived to be suddenly devalued and revalorized as "peasant values", result-
ing in the collapse of the peasantry (especially its male component) who are
then condemned to celibacy / bachelorhood; this is because women won't
marry men who speak like peasants if they can marry men who speak the 'legit-
imate' language.
Confirming Labov's observation (Labov 1972) that lower-middle class
women are the quickest to adopt linguistic features of higher prestige, varying
more widely between their informal and formal speech than any other class or
gender, Bourdieu goes on to describe why French women are the first to adopt
standard speech in rural areas:
and since women are quicker to adopt the legitimate language (or the proper pronun-
ciation) (...) especially because they specialize in the domain of consumption (of
various products, especially the language) and by the logic of marriage, which is for
them the principal route to social mobility, where they move upward, they are pre-
disposed to accept, beginning at school, the new demands of the symbolic capital
market. (Bourdieu 1982: 35)
Here Bourdieu points out what is the real causative factor in promoting official
language, namely, that the imposition of another language does not happen be-
cause the state decrees it, but because of other social factors correlated with the
officialization.
Thus, the effects of domination which are correlated with the unification of the mar-
ket are not exercised except by means of a whole complex of specific institutions and
mechanisms, of which language policy proper, along with the intentional interven-
tions of pressure groups, are only very superficial factors. (Bourdieu 1982: 35)
In other words, you don't have to have the "overt" kinds of pressure associated
with officialization, because they are only superficially effective; the market
and other institutions allied with it do the job for you.
And the fact that [juridical or quasi-juridical constraints] presuppose political and
economic unification, which they contribute, retroactively, to the strengthening of
does not at all imply that one should impute the progress of the official language to
the direct efficacity of [these] constraints, (which can only impose the acquisition
[my emphasis, hs] of the official language but not the general use of it, and at the
same time, the autonomous reproduction of it.) Every symbolic domination presup-
poses on the part of those who submit to it a form of complicity, which is neither
passive submission to an exterior constraint, nor free acceptance of its values. The
recognition of the legitimacy of the official language has nothing of a belief delib-
erately professed and therefore revocable, nor of an intentional act of acceptance of a
"norme" ·, it is rather inscribed in the practical state of dispositions which are subtly
inculcated, as part of a long and slow process of acquisition, by the sanctions of the
linguistic marketplace and which are often adjusted, without any cynical calculation
or of any consciously resented constraint, to the chances of material and symbolic
profit which the laws of the price-formation characteristic of a certain market objec-
tively promise to the possessors of a certain linguistic capital. (Bourdieu 1982:
35-36)
This means, he adds in a footnote (Bourdieu 1982: 37), that "linguistic customs
[moeurs] cannot be modified by decrees the way partisans of a volunteerist pol-
icy involving " d e f e n s e de la langue" seem to believe.
4.4. Bourdieu and Whorf
It is also interesting to note that Bourdieu specifically discerns a kind of folk-

Whorfian (Mertz 1982) world-view at work in the imposition and functioning
of this model. Teachers in French schools are on the front lines, as it were,
working constantly to "inculcate a clear faculty of expression and of each emo-
tion," i.e. through language. They work to replace the patois, which is nothing
but a jumble of confusion, with standard French, itself the only "clear and
fixed" thing that deserves to be in their heads, and trying to get them to perceive
and feel things in the same way. The work of the teacher is "to erect the com-
mon conscience of the nation." Bourdieu calls this a Whorfian or Humboldtian
theory of language, which sees scholarly action as "intellectual and moral in-
tegration" (Bourdieu 1982: 32). Teaching language, therefore, is a kind of
'mind control;' instilling the standard language in the heads of children will re-
program them to think clearly.
4.4.1. Written language and spoken language

Bourdieu's work does not always distinguish between written and spoken lan-
guage, but it is clear from his writing that the kinds of controls he talks about
(above and elsewhere) are primarily of the written kind, though control of
spoken language, i.e. correct pronunciation of official French is also involved.
That the French state is much better at controlling a linguistic marketplace of
written language, rather than spoken language, however, is clear even from the
above presentation. Given what else he says about the evaluation of 'linguistic
products' it is also clear that written tokens of language will be more harshly
evaluated than will spoken ones, and because they are also scarcer, they have
higher symbolic value. But attention to subtle differences of pronunciation will
also result in evaluations that can only benefit those speaking the 'legitimate'
language, in its standard (Parisian, or at least northern) pronunciation.
Though a large part of language is invariable, there exist, whether in pronunciation,

lexicon, or in grammar, a whole set of differences associated meaningfully with so-
cial differences which, though negligible to the linguist13, are pertinent from the so-
ciological point of view, because they enter into a system of linguistic oppositions
which retranslates a system of social differences. (Bourdieu 1982: 41)
5. Problems with Bourdieu's analysis: the linguistic black

market
Though Bourdieu takes great pains to establish that there is a unified market-
place for language, and that the state creates and controls this market (thus ex-
cluding everyone who does not participate in it) he fails to explain why it is that
in the end, some people continue to use non-standard languages, the ones
known in France as patois, idiomes etc. Despite the relentless march of cen-
trism, and the domination of this market, a sort of linguistic black-market con-
tinues to persist, with values (Labov (1972) calls this "covert prestige") at-
tached to these 'products' that seem to be determined by other factors, perhaps
beyond the pale of the centrist control. Once we begin to think of ways that of-
ficial markets are undermined or resisted, we can imagine, not only a black
market, but markets purveying linguistic contraband, stolen or illegal goods, or
other kinds of under-the-counter activities. In economic terms, we know that
black markets exist because they can provide things that are scarce; illegal sub-
stances are sold because people want to buy them, whether or not they publicly
admit it. Perhaps we need to think offranglais as one token of a black-market
linguistic commodity, illegally imported and 'consumed' because it has covert
prestige, and consumers in the linguistic market place want it, irrespective of
its legality. Thus the economic model seems to call for a way to deal with more
than just the official products of the linguistic marketplace, but contraband
commodities that have their own symbolic value must be taken into consider-
ation, too.
5.1. An Orwellian language policy?
Though French people pay lip-service to the notion that le franglais must be
banned, if a dirigiste strategy is to work, French speakers must change their
stripes, since the French readily admit that they are unwilling to do what they
are told to do. In other words, resistance to franglais will come about if each
French citizen not only pays lip-service to the official policy about it, but also
exercises self-control in their consumption and exchange of such illegitimate
linguistic commodities. But if speakers do not do this, the only recourse gov-
ernments then have is to set up an Orwellian police-state such as existed during
in the Nazi-occupied Alsace during World War II (when people speaking
French in public were "deported" to the interior of France, or worse.) It is not
clear that any French citizen wants a linguistic police-state, which in any event
would require a gendarme linguistique to shadow every French citizen eight-
een hours a day, to make sure all utterances were grammatically acceptable;
only E. Germany, with a Stasiu agent shadowing large numbers of its citizenry,
ever achieved this level of control, and never in the linguistic realm. It would
entail secret denunciations, electronic surveillance, perhaps deportations, re-
education camps, and state terrorism, in short, all the trappings of a totalitarian
state. No polity has ever managed to erect such a draconian language policy,
and French citizens certainly do not wish to pay this price. Is there no way out
of this conundrum?
Perhaps by now it is clear that it is not state dirigisme that works to control
language, but other social forces; and that attempts by the center to totally con-
trol language are not going to succeed in eliminating regional languages, non-
standard forms of French, or contraband linguistic products (French corrupted
by franglais) any time soon. Other social forces may act to do this, but they will
be social forces outside the control of the government, such as urbanization,
computerization, the channel tunnel, the globalization of the economy (and the
concomitant spread of the English language), perceptions about the cultural
capital that standard language provides and so on. The creation of a more open
European Union, now that a unified currency has come into effect, may in fact
have even more profound effects on the regional languages, and indeed on the
French language.
5.2. The fallacy of linguistic dirigisme
The fallacy of linguistic dirigisme, therefore, is based on supposed parallels

with economic dirigisme, which controls the economy by controlling the
money supply, the means of production, wages, prices, exchange rates, interest
rates, imports, or any other commodities, currencies, or things and substances
of value, the supply of which is finite, or can be made finite by state control.
Language, however, is not a finite substance or commodity that can be con-
trolled in this way, for a number of reasons.
1. Utterance supply: One is that each speaker (of any language whatsoever)
in the polity generates his/her own supply of utterances. This is a basic
fact about language and how it works that is often misunderstood, and not
just in French culture.
2. Divine origin of language: In many cultures of the world the idea that lan-
guage has divine origins, or stems from some Platonic higher conscious-
ness, leads culture-bearers to assume that language can therefore be con-
trolled (or must be controlled) in various ways, and that the utterances that
speakers make can be discounted or devalued if they do not meet some pre-
established standard.15
3. Covert prestige: It seems to me that French linguistic culture would have it
that self-generated utterances of a non-standard sort are like debased cur-
rencies or contraband, and must be driven out of existence. The problem is,
as Gresham's law has it, bad money drives out good, so the existence of cor-
rupt language has paradoxical effects; people pay lip service to 'good' lan-
guage, but non-standard language also has symbolic value to its speakers16
because it authentically represents par excellence their personal, social, re-
gional, or even sexual identity in ways that the standard language never can.
4. Belief systems: Though there is no proof that dirigisme is ever effective
when applied to language and linguistic habits, the belief that it works is
firmly grounded in French ideas about language (Catach 1991; Schiffman
" 1996). It is now being called upon to save French from the corruption and
perturbations brought on by wholesale borrowing offranglais (English and
American words and phrases), which bring with them an unsavory ideology
and life-style, which, if not resisted at all costs, will undermine and debase
French culture beyond recognition.
Notes
1. "Notre centralisme linguistique, le plus ancien et le plus consomme d'Europe, est

bien connu: Philippe le Bel, Frangois Ier, Richelieu, Colbert, la Convention, Napo-
leon, Jules Ferry et d'autres encore l'ont illustre". (Catach 1991: 7)
2. I have defined this as "the sum total of behaviors, assumptions, cultural forms,
prejudices, folk belief sytems, attitudes, stereotypes, ways of thinking about lan-
guage, and religion-historical circumstances associated with a particular language".
(Schiffman 1996: 5)
3. Some would call it ideology, e.g. Haitz 1988.
4. See the words of Barere or Gregroire, e.g. in Schiffman 1996: 99-111.
5. Gordon (1978) echoes this view.
6. Miller (1982) shows how in Japan, as certain state-sponsored myths have crumbled,
language myths have become stronger.
7. The revolution in Turkey in the 1920's however seems to have emulated the French
Revolution in linguistic matters, using a "purified" Turkish as a symbol of national-
ism, and banning the use of other languages such as Armenian, etc.
8. As I have claimed, however, it is the implementation of language policy that is
usually its achilles' heel; if a decree has no teeth, it will not be implemented, es-
pecially if there is a cost to its effective implementation, and no cost (or penalty) for
failure to implement.
9. [The] Academie frangaise, or French literary academy, [was] established by the

French first minister Cardinal de Richelieu in 1634 and incorporated in 1635, and
existing, except for an interruption during the era of the French Revolution, to the
present day. Its original purpose was to maintain standards of literary taste and to
establish the literary language. (Encyclopedia Brittanica, website.)
10. "Depuis le debut de ce siecle, 1'Academie a fait savoir qu'elle ne se sentait plus ni
le droit ni les moyens d'assurer seule ses fonctions. Elle a signifie ά plusieurs re-
prises qu'elle entendait seulement rester de 'greffer de 1'usage' [...]. Elle 'ne se re-
connaitpas le droit', declarait-elle en 1935, et encore tout recemment, 'de reformer
I' orthographe'." (Catach 1991: 55)
11. This is the term given to English loan words such as le weekend, le fast-food, le self-
service that have been borrowed on a large scale in recent decades, and which are
found (Flaitz 1982) to menace the very foundations of French culture.
12. La langue officielle a partie liee avec Γ Etat. Et cela tant dans sa genese que dans
ses usages sociaux. C'est dans le processus de constitution de Γ etat que se creent
les conditions de la constitution d'un marche linguistique unifie et domine par la
langue officielle: obligatoire dans les occasions officielles et dans les espaces offi-
ciels (ecole, administrations publique, institutions politiques, etc.), cette langue
d'Etat devient la norme theorique ά laquelle toutes les pratiques linguistiques sont
objectivement mesurees. Nul n'est cense ignorer la loi linguistique qui a son corps
de juristes, les grammairiens, et ses agents d' imposition et de contröle, les maitres
de 1' enseignement, investis du pouvoir de soumettre universellement ά l'examen et
ά la sanction juridique du titre scolaire la performance linguistique des sujets par-
lants. (Bourdieu 1982: 27)
13. Bourdieu faults most linguists, from Saussure on down (though he probably
exempts sociolinguists) for accepting standard language as the domain of their en-
quiry, and ignoring the fact that they legitimate it by focussing exclusively on la
langue and ignoring la parole.
14. Stasi is the abbreviation of the German term Staatssicherheitsdienst or State Secur-
ity Service.
15. See Schiffman 1996 for examples of mythological ideas about language.
16. Labov refers to this as covert prestige because though speakers overtly deny the
value or validity of non-standard forms, they retain and use them for certain pur-
poses, because at some level, they have meaning for them (Labov 1972). We can
perhaps see the contradiction here as parallel to the paradox of drug interdiction in
western countries - people want the drugs banned, and their importation inter-
dicted, but covertly, some consumers want them, and are willing to pay for them, so
the supply continues to flow.
References
Ager, Dennis Ε.
1990 Sociolinguistics and Contemporary French. Cambridge: Cambridge
University Press.
Bedard, Edith and Jacques Maurais
1983 La Norme Linguistique [The linguistic norm]. Paris: Le Robert.
Bourdieu, Philippe
* 1982 Ce que parier veut dire [What speaking means]. Paris: Fayard.
Catach, Nina
1991 L'Orthographe en debat [Spelling in debate]. Paris: Editions Nathan.
de Certeau, Michel, D. Julia and J. Revel
* 1975 Une politique de la langue. La Revolution frangaise et lespatois: I'en-
quete de Gregoire [A policy about language. The French Revolution
and dialects: an inquiry by Gregoire], Paris: Gallimard.
Falch, J.
1973 Contribution ά 1'etude du Statut des langues en Europe [Contribution
to the study of the status of languages in Europe], No. 3. Centre Inter-
national de Recherches sur le Bilinguisme. Quebec: Presses de l'Uni-
versite Laval.
Flaitz, Jeffra
* 1988 The ideology of English: French perceptions of English as a world
language. Contributions to the Sociology ofLanguage 49. Berlin/New
York: Mouton de Gruyter.
Gordon, David
* 1978 The French language and national identity 1930-1975. Contributions
to the Sociology ofLanguage 22. The Hague/Paris/New York: Mouton
de Gruyter.
Guilhaumou, Jacques
1989 La langue politique et la revolution frangaise [The political language
and the French Revolution], Paris: Meridiens Klincksieck.
Haut Comite de la langue fransaise and Premier Ministre
1975 La loi relative ά la langue frangaise [The law regarding the French
language], Paris: La documentation fransaise.
Hell, Victor
1986 Pour une culture sans frontieres: L'Alsace, une autre histoire franco-
allemande [Towards a culture without borders: Alsace, another
French / German story]. Strasbourg: BF editions.
Labov, William
1972 Sociolingusitic Patterns. Philadelphia: University of Pennsylvania
Press.
Levy, Paul
1929 Histoire linguistique d'Alsace et de Lorraine [Linguistic history of
Alsace and lorraine]. Paris: Societe d'Edition Les Belles Lettres.
Mertz, Elizabeth
1982 Language and mind: a 'Whorfian' folk theory in United States lan-
guage law. (Working Papers in Sociolinguistics 93.) Austin: South-
west Educational Development Laboratory.
Philipps, Eugene
1975 Les luttes linguistiques en Alsace jusqu'en 1945 [Linguistic struggles
in Alsace until 1945]. Strasbourg: Culture Alsacienne.
1978 L Alsace face ά son destin: la crise d'identite [Alsace facing its des-
tiny: an identity crisis], Strasbourg: Soci6te d'Edition de la Basse-Al-
sace.
1982 Le defi Alsacien [The alsatian challenge], Strasbourg: Societe d'Edi-
tion de la Basse-Alsace.
Pool, Jonathan (ed.)
* 1987 Linguistic inequality. Language Problems and Language Planning 11
(1).
Schiffman, Harold F.
1992 'Resisting arrest' in status planning: structural and covert impedi-
ments to status change implementation. Language and Communi-
cation 12 (1): 1-15.
1993 The balance of power in multiglossic languages: implications for lan-
guage shift. International Journal of the Sociology of Language 103:
115-148.
*1996 Linguistic Culture and Language Policy. London/New York: Rout-
ledge.
Smith, David M.
1973 Language, speech and ideology: a conceptual framework. In: Shuy
and Fasold (eds.), Language attitudes: current trends and prospects,
97-112. Washington, D.C.: Georgetown University Press.
Smolicz, J. J.
1979 Culture and Education in a Plural Society. Canberra: Curriculum De-
velopment Center.
Tabouret-Keller, Andree and Frederic Luckel

1981 La dynamique sociale du changement linguistique: quelques aspects
de la situation rurale en Alsace [The social dynamics of linguistic
change: a few aspects about the rural situation in Alsace]. Inter-
national Journal of the Sociology of Language 29: 51-70.
Vermes, Genevieve and Josiane Boutet (eds.)

1987 France, pays multilingue [France, multilingual country], 2 volumes.
Paris: L'Harmattan.
Whorf, Benjamin Lee

1964 A linguistic consideration of thinking in primitive communities. In:
Dell Hymes (ed.), Language in Culture and Society, 129-141. New
York/Evanston/London: Harper and Row.
1. What do you think of the author's contention that ideas about controlling
language policy in France could only arise in a linguistic culture like that of
France, and that such a policy could not or would not exist in, e.g., an
Anglo-Saxon country? What other polities in world history have tried to
control language the way France does, and what kind of government did
they have?
2. Since France isn't the only French-speaking country in the world, what do
you think the reaction of people in Canada (Quebec), Belgium, Switzerland
and francophone Africa is to the tendency on the part of France to dictate
what the French language will or should be like?
3. Now that the internet can penetrate any home in the world without govern-
ment intervention, what is the prognosis for the ability of France to control
the French language in the new millennium?
4. Given the fact that in France the media, the educational system, and the
publishing industry are controlled centrally, i.e. from Paris (whether owned
by the government, or privately), is there a possibility that Bourdieu's
analysis is more or less correct?
5. What do you think would happen if a more laissez-faire language policy
were allowed to develop in France? Would currently disenfranchised
citizens have more access to communication and education than they now
possess?
The non-linearity of language maintenance and
language shift: survey data from European language
boundaries
Peter H. Neide and Peter J. Weber
1. Introduction: multilingualism in Europe
The post World War II era has seen an increase in different forms of multilin-
gualism, which have acquired different values. While autochthonous minority
("ethnic groups", "nationalities") residents in most European states were orig-
inally the chief focus of interest, new, often socially-defined minorities such as
migrants, guest workers, emigrants returning from former colonies, refugees,
evacuees and re-settlers have, since the 1960's, been entering the European pic-
ture, a development reinforced by the opening of the East in the 1990's. This
proliferation of minority groups served to increase the awareness of the major-
ity groups, which amazingly, did not push these established minorities into the
background. Instead, new currents like the so-called "renaissance of dialects
and smaller languages" and the new regional consciousness oriented towards
smaller groups, focused the attention of research, politics, culture and the gen-
eral public more intensely on minorities, whose significance for a culturally vi-
able Europe was emphasized in the East as much as in the West.
The pressure exercised by the majority peoples to have standardized lan-
guages, and the cultural and socio-economic influence of the super-
powers - which themselves, in the process of globalization, threathened the
cultural autonomy of the majority peoples - forced the smaller ethnic groups
with no legal protection to decide whether it was desirable and/or possible to
take measures to ensure their survival. For most smaller ethnic groups in Eu-
rope this meant choosing between two rather unpalatable alternatives: either to
conform to the often economically stronger majority group and be assimilated
further, or to face conflict, the resolution and consequences of which were very
uncertain, since modest efforts to promote minorities through a concept of re-
gionalization by the European Union could nevertheless not guarantee to pro-
vide a 'counterweight' to the effects of globalization.
In present day Europe - i.e. Europe after the Maastricht II treaty, with its
strong tendency toward unification and international connections - every in-
106 Peter Η. Neide and Peter J. Weber
stance of language and culture contact between different ethnic / cultural

groups involving the identity of the group seems to imply conflict. Numerous
cultural and contact linguistic researchers are making efforts to avoid or at least
play down these conflicts, firstly by stressing the individuality and the "other-
ness" of the minority, and secondly, by putting forward different types of multi-
lingualism as potential solutions.
2. Fields of discussion: European language boundaries
Throughout the last decade, Europe has seen a revival of ethnic minorities
within the context of "European multilingualism" and it is European Union
policy to promote the conservation of cultural identity in all groups of society.
In fact, as far back as the early 80's, EU institutions have been dealing with the
situation of autochthonous language minorities. As a result both of this interest
and of the growing number of resolutions and reports - the most popular
among them being the European Charter for Regional and Minority Languages
of November 5,1992 - the European Commission has begun to grant financial
support to the language minorities in question. The formulation of respective
political initiatives, however, has been impeded by the fast-changing legal, in-
stitutional and social environment of some language communities.
This article will discuss typical aspects of language maintenance and lan-
guage shift in Europe as they can be found in Germanic-Romance and Ger-
manic-Slavic minority contact situations. The first case presented here is that of
the German-speaking minority in 'Old Belgium'. Old Belgium, with an esti-
mated German-speaking population of less than 40,000, refers to a number of
geographically disconnected areas along Belgium's eastern border which have
been part of Belgium since its foundation in 1830 (cf. map 1).1 We will focus in
particular on the German minority in southern Old Belgium and compare it to
another minority that is considered weaker: the Upper Sorbian minority in Ger-
many. This group includes all Sorbian speaking communities of the Upper
Lausitz area, most of whom are Catholics (cf. map 2; Elle 2000).
In the Euromosaic study2, language minorities are first defined as social
groups (Neide etal. 1996). By analyzing and comparing some of the linguistic
and socio-economic features which characterize these minorities, it may be
possible to define the backdrop of language maintenance and language shift
that might in turn serve as a model for other minorities.
The main theoretical goal here is to give some examples of the irregularity
of language maintenance, language shift and language loss and to show that
The non-linearity of language maintenance and language shift 107
these processes cannot be explained by intra-linguistic factors or by a linear ap-

proach alone. The discussion is underpinned by three major claims:
- Language maintenance and language shift from different minorities are
very specific processes. At a very low level the circumstances surround-
ing these processes can be only explained by analyzing each minority.
- Language maintenance and language shift are non-linear processes.
- The non-linearity of language maintenance and language shift displayed
by these minority groups illustrates and determines contact universale.
These contact universale could provide data for (minority) language plan-
ning such as that intended by Euromosaic.
The term "non-linearity" refers to the scientific notion that neither form nor
function of natural objects - in contrast to technical objects - can be explained
by causal chains, but rather by complex causal fabrics or correlations of effects
(Eilenberger 1990). According to a section of chaos theory called "non-linear
dynamics", there is no such thing as a direct correlation between cause and ef-
fect. It may be assumed that the (strong) decline in speakers of a language com-
munity does not lead to the disappearance of an ethnic group in a linear
fashion. Due to the complex correlations of effects, positive counter-effects
may occur. What ultimately matters is the correction of historical and current
statements on the results of empirical research on language boundaries. A typi-
cal example of this type of statement is given by Walter Hoffmeister (1977),
who, in his case study on language shift in Eastern Lorraine, predicted the de-
cline and death of the German language by the next generation: "so wird es im
Ergebnis zu einer Kongruenz von Staats- und Sprachgrenze kommen" [lan-
guage shift will result in a gradual disappearance of the indigenous language]
(Hoffmeister 1977: 82).
For the ensuing discussion of non-linearity in the context of European lan-
guage boundaries, it will first be necessary to draw up a profile of the two mi-
nority groups and language boundaries concerned, using empirical data from
the Euromosaic study. This profile will form the basis of two micro case studies
of language maintenance and language loss along linguistic boundaries where
in one case, German serves as a dominated minority language (Old Belgium),
and in the other German serves as the dominating majority language (Upper
Sorbian area). These case studies will demonstrate the non-linearity of the pro-
cesses of language maintenance and loss at work in these two minority com-
munities, and will pave the way for proposing a way of identifying contact uni-
versale in both empirical minority research and European language policy and
planning.
3. The minorities: two profiles
3.1. The German speaking area in Old Belgium
Following the declaration of Belgian independence in 1830, freedom of lan-

guage was proclaimed, giving citizens the right to freely choose between
Dutch, French and German when dealing with the authorities (Article 5 of the
decree of November 16, 1830). Shortly after 1839, this law ceased to be ap-
plied. In administration and jurisdiction, French was used almost exclusively.
With regard to primary education, a law passed in 1842 stated that - according
to the actual needs - basic terms had to be taught in all three languages. In 1914,
another law was passed to the effect that children had to be taught in their native
language at all levels. In reality however, this policy was often not enforced and
secondary education continued to be predominantly conducted in French.
The improvements in the status of the German language which came about
with the favorable language policy introduced during German occupation in
World War I, were abolished immediately at the end of the war. After the Ger-
man invasion of May 1940, the region of Eupen-Malmedy, nine Old Belgian
communities and Bocholz (Central Old Belgium) were annexed by Germany.
As a consequence, German was the only language in education, administration
and justice up to 1945. After these regions were reclaimed, these policies were
of course abandoned, and many people stopped using German for personal rea-
sons. German vanished from administration, education, justice and religion
and has been non-existent as a cultural and literary language in Old Belgium
ever since. Today, the use of German in Old Belgium is restricted to the home
and village level in the form of Lower Rhine-Franconian, Limburg- and Mo-
selle-Franconian dialects (see e.g. Kern 1997).
The German-speaking region of Belgium consists of parts of New and Old
Belgium (see also note 1). In Old Belgium there are three important German-
speaking areas: Montzener Land, also called "Welkenrather Gegend" (North
Old Belgium), Bocholz with the villages of Deifeld, Urt, Watermal (Central Old
Belgium), and the Arelerland on the border with the Grand Duchy of Luxem-
bourg (South Old Belgium). In total, there are some 100,000 German-speaking
people in Belgium - approximately 1 % of the population. In Old Belgium, there
are some 40,000 speakers of German dialects whose official language is French.
There are four reasons for assuming that the German-speaking minority in
Old Belgium might be jeopardized:
- The language has no legal status.3 There appears to be little intention of
providing legal protection, or indeed support for such a move.
- People do not make use of current regulations facilitating the use of the
language (e.g. in the region "Montzener Land").
- There is still a degree of animosity towards Germans, the result of recent
history.
- There is no standard German serving as umbrella language for the dia-
lects, thus contributing to a tendency towards linguistic autonomy. The
absence of standard German serving as an umbrella language for the dia-
lects contributes a special format to linguistic autonomy, which can be
characterized as diglossic - standard French replaces standard German
and German dialects serve as a vernacular. This diglossic attitude with
separated domains (French in formal, German in informal situations) can
be described as a complementary distribution of language varieties.
3.2. The Sorbian speaking area in Germany
The decline of the usage of the Sorbian language came about in the 10th cen-
tury, when the old Sorbian tribes lost their political independence. In the fol-
lowing two centuries farmers from Franken, Thüringen and Sachsen founded
settlements in their territory and German cities; commerce and traffic devel-
oped. As early as the 13th century, the Sorbian language was forbidden for the
first time. With the reformation in the 16th century, the first texts were written in
Sorbian such as the fragment of an agenda (1543) from Zossen and the trans-
lation of the New Testament (1548) by M. Jakubica from Laubnitz. The further
splitting-up of the Sorbian-speaking region into several political territories in
the course of the following centuries required the introduction of a standard
written language on the basis of the dialects. In the middle of the 19th century,
the Upper Sorbian written language based on the dialect spoken in the region of
Bautzen was introduced as a binding standard in the Sorbian-speaking region
in the Upper Lausitz. The Lower Sorbian written language based on the dialect
of Cottbus became the lingua franca of the Lower Lausitz.
Of particular note in the current discussion is the fact that the Sorbian lan-
guage before 1945 had experienced a differentiated, but permanent pressure to-
ward assimilation and Germanization, which can be attributed to the region's
social and economic structures and to the prevailing - fascist - ideology in
Germany. After 1945, formal rights to conserve and promote the language were
secured, but in effect, were limited by extensive industrialization and the domi-
nant - communist - ideology.
Today, the Sorbian language is spoken in the regions of Upper and Lower
Lausitz in the federal Provinces ("Länder") of Saxonia/Sachsen and Branden-
burg, concentrated mainly in the regions east of Kamenz around Bautzen,

Hoyerswerda, Weißwasser, Spremberg and Cottbus - townships and villages
with strong groups of Sorbian native speakers. In 1998, the estimated number
of Sorbs was approximately 40,000. Some 32,000 of these are Saxonian Upper
Sorbs (Upper Lausitz), and approximately 7,000 to 8,000 are Brandenburg
Lower Sorbs (Lower Lausitz). (For more information, see e.g. Faßke 1997).
Just like the German-speaking minority in Belgium, the future of the Sor-
bian population is likely to come under threat, for three key reasons:
- The Sorbs are an isolated minority in one of the affluent countries of the
European Union.
- Before the joining of Austria, the Sorbs, together with the Macedonian
and Pomak communities in Greece, were the only Slavic minorities in the
European Union. But whereas the Macedonians in Greece can readily
refer to their Northern neighbours and the Pomaks to the Northern Bul-
garians, the Sorbs are the only Slavic minority without any relationship to
another minority group of reference outside Germany.
- The case of Sorbian shows that even guaranteed protection by two differ-
ent political authorities (Saxonia and Brandenburg) is insufficient to stop
the process of decline in the life of a minority language or culture.
4. The specificity of language maintenance and shift: two micro

case studies
The above observations highlight the differences between the two minorities.
Their actual situations, however, are not vastly different: both are endangered
minorities in Europe. In order to operationalize these differences, the analysis
of the two case studies will primarily focus on the sectors of education and
economy by means of a portfolio (see Gabele 1993). The idea of a portfolio,
adjusted to the subject of minority research, is aimed at guiding expected re-
sources to those social sectors of the minority which have the most favorable
prospects and which provide the minority with the best chances of using the
language and keeping it alive (cf. Weber 1997). These sectors can be detected
by finding correlations between the minorities, based on comparing categories
(cf. Table 1).
Table 1. Comparing minorities
Comparing category German in Old Belgium Sorbian in Germany

Education French as official language Two different schooling sy-
is dominant, increasing iso- stems, increasing isolation
lation of German. of Sorbian.
Economy Germany as a natural pole No natural pole of attrac-

of attraction, but no linkage tion, no linkage to the Ger-
to the German-speaking man-speaking market.
markets.
The definition of strengths and weaknesses based on the comparing categories

makes it possible to identify those configurations which form the greatest dis-
crepancies between the minorities and which thus might constitute potentially
successful spheres of activity for the weaker minority. It will then be possible
to use this comparative methodology to single out areas and strategies which
may be of special importance for the future of these minorities.
Both minorities are, to differing degrees, in an unfavorable situation. This
situation could be described in both cases as follows:
- Education: In Old Belgium, no German education is provided, but Sor-
bian speakers have their own Sorbian domains in schools, where students
use Sorbian as medium of education.
- Economy: The German minority is stronger due to the background of the
German nation state than the peripheral Sorbian minority without any
neighboring "homeland".
According to the comparing categories, our observation leads us to conclude
that although both minorities are facing similar dangers, these dangers can be
attributed to different developments. The explanations for this divergent devel-
opment are linked to eco- and socio-linguistic factors, which provide some of
the reasons why the two minorities have not disappeared in spite of their in-
creasingly unfavorable chances of survival. The following provides a statistical
view of the situation of minority speakers of German and Sorbian, which sup-
ports the principle of non-linearity. In addition, specific reasons are given as to
why the death of the language does not come about in a linear fashion.
4.1. The German speaking area in Old Belgium
In 1930 87% of the residents in the village of Martelingen claimed to speak

German (cf. Table 2). Exactly 17 years later (1947) only one and a half percent
of the people of that same language community still claimed to use German as
their first and main language. There must have been a strong reason for their
change in attitude since census data can be interpreted as being relevant for the
identity, solidarity and loyality feelings of a bilingual population. Almost
everyone seems to be bilingual in this border area. Our minority population,
with a high degree of flexibility in cultural and linguistic matters must, after the
end of the Second World War which marked the end of the "fashionable period
of facism" of the 1930's, have decided, to vote in favor of the dominant French
groups and to give up German as their symbol of identity.
Table 2. Census data (Old Belgium)
The German communities of the districts of Arel and Bastnach
1962 1846 1930 1947
inhabitants German French German French German French
Arel 13.520 78,02 16,48 13,45 82,53 0,72 95,07
Athem 7.085 - - 23,04 70,72 1,46 94,03
Attert 718 99,0 0,99 82,73 14,05 28,28 67,56
Bonnert 2.060 98,07 1,91 80,00 15,29 4,72 90,74
Diedenberg 517 100,0 0,00 90,82 4,13 21,88 71,36
Elcheroth 745 98,76 1,23 92,22 3,89 4,85 91,91
Girsch 210 91,11 8,82 82,29 12,46 6,55 89,52
Heischlingen 2.777 97,61 2,38 10,66 84,81 11,67 82,12
Herzig 1.560 98,40 1,58 86,20 9,12 1,03 94,83
Hewerdingen 473 99,32 0,67 92,15 3,51 50,99 40,48
(Holdingen) 3.391 51,29 48,70 25,78 69,79 2,41 92,60
Hondelingen 621 99,67 0,32 91,84 4,86 21,02 69,70
Ibingen 3.411 95,08 4,91 11,75 82,03 0,34 96,02
Martelingen 1.616 96,23 3,76 87,16 10,02 1,56 84,35
Feiteler 791 42,65 57,34 23,57 73,06 4,01 92,44
Metzig 2.856 97,12 2,87 82,22 12,41 8,89 86,30
Niederelter 1.638 95,25 4,74 83,05 13,53 10,02 83,09
Nothum 342 - - 93,67 1,09 32,93 58,90

The German communities of the districts of Arel and Bastnach
1962 1846 1930 1947
inhabitants German French German French German French
Selingen 613 - 88,26 7,56 14,17 72,46
Tintingen 322 90,00 10,00 77,33 20,55 14,03 80,36
Tontelingen 456 - 89,87 3,63 3,39 93,62
Tömich 651 97,70 1,29 86,75 8,28 6,41 89,96
Wolkringen 645 - 88,47 6,80 40,67 47,77
Source: Neide 1979: 70
The need to adopt a non-linear perspective is reinforced by patterns of use

within the private sphere (families). It is not easy to find an exact description of
language use and language shift in multilingual linguistic areas in the "prelin-
guistic" literature of the 19th century. Yet when one examines the written
sources dealing with Old Belgium of circa 1850, one obtains a relatively clear
picture of the language use within the family. Whereas the older generation still
spoke the local variety of German (monolinguals), the middle generation had
already changed over to the official French while retaining the local dialect
(diglossic speakers). The young generation, however, had lost the mother
tongue and had shifted completely to French (monolinguals again), able to
understand, yet not speak the language of their ancestors.
Fifty years later, the situation remained unchanged, with the grandparents
still using German, the parents bilingual and the school-age generation prefer-
ring French. Reviewing literature of the 1930s, the participant observer is con-
fronted with identical linguistic behaviour: the older generation spoke only
German, the younger adults spoke two languages, and the youngsters had again
replaced their low variety by a higher French one. This unchanging state of af-
fairs seems to defy logic: if the observations are all true, the low variety should
not have appeared in the 20th century. But this is only if one takes a linear ap-
proach. Many of the young people that moved away from the Germanic low
variety at school, due to official instruction in French, have obviously returned
to their linguistic roots. Returning from the coal and steel area of Wallonia,
where income prospects were better, and inheriting parental farms, houses or
businesses in their local region, they must have felt the need to conform and re-
identify with the local vernacular.
This example illustrates the limitations of a linear approach and highlights
the importance of taking non-linear approaches into consideration.
4.2. The Sorbian speaking area in Germany

The Sorbian minority shows the same statistical development. During the Ger-
man national phase until 1871, the number of Sorbian speakers decreased
steadily (1880/1884: 166,000 according to Macica Serbska 1991). The Nazis
disapproved of the Slavic character and of the Slavic Sorbian minority in par-
ticular. In 1936/38, only 111,000 Sorbian speakers were counted in the area
(Macica Serbska 1991). Under the communist regime, the minority apparently
prospered. On one hand, this was a good example of a seemingly liberal social-
ist society, but on the other hand, it was not popular for a socialist regime to tol-
erate growing diversity with a political aim of equality. Probably for this rea-
son, no more than 81,000 Sorbian speakers could be found in 1955/56 (Macica
Serbska 1991).
Today's era of neo-liberalism, with its focus on material values, also has a
negative impact on the situation of peripheral minorities. But at the same time,
the move towards globalization brings with it a move towards regionalization
and the growing importance of traditional values. Nowadays, according to the
Euromosaic report, there are still some 40,000 Sorbian speakers. Estimates
based on recent qualitative empirical surveys assume that today there are only
30,000 speakers of Upper Sorbian left. Statistical estimates at the turn of the
century, however, predicted the death of the Sorbian language along the Slavic-
German language boundary before the end of the 20th century.
Parallel to the development of the Old Belgium German-speaking minor-
ity, in particular in places such as Nothum (North of the province capital
Arel), re-immigration played an important part for the (reversed) non-linear
shift of language from Sorbian to German. There are four reasons for re-immi-
gration:
- Marriages: In 1996 the first marriage conducted exclusively in Lower
Sorbian took place. At the same time there were an increasing number of
mixed marriages in the Upper and in the Lower Sorbian area.
- Inheritance: As we know from official sources, the ten years to come will
be the years of material inheritance. A lot of young people will inherit the
patrimony of parents and grand-parents which could be an opportunity for
the Sorbian minority to expand with re-immigrants (EU-terminology:
"remigrants").
- Official Positions: The minorities are becoming increasingly "academic"
and service-oriented. Furthermore, there are also a large number of offi-
cial jobs specifically for local unemployed minority speakers.
- Roots: The older generation living in a boom society until the 60's, has re-
discovered other values than materialistic ones. Many of them are frus-
trated with capitalist society and want to live in a different society in old
age - another chance for re-immigrants to preserve their first language.
Re-immigration is accompanied by a strong process of re-identification
with traditional folkloristic values. A typical example is the "Hochzeitsmarkt"
(wedding market), to which only speakers of Sorbian are admitted. This phe-
nomenon of increasing differentiation can also be observed in the economic
sector. There are rural cooperatives, for instance, reserved for speakers of Sor-
bian. According to the Euromosaic study, this high degree of homogeneity of
the community with regard to attitudes and identity implies a positive attitude
with respect to the minority language. This assumption is confirmed in Table 3,
as 73% (276 of 300) of the interviewees declared they felt Sorbian.
Table 3. Sense of belonging
yes no
Sorbian 73% 27% 100%

German 32% 68% 100%
European 29% 71% 100%
Other 5% 95% 100%
According to these results, neither national (German) nor supranational (Euro-

pean) concepts achieve significant results. This feeling of local identity ex-
pressed in the opinion of the interviewees is also influenced by a certain inter-
est on the part of the local authorities (Neide, P. H. etal. 1996).
5. Language maintenance and shift: a non-linear process

These specific processes of language maintenance and loss of two minorities
show how ideology, religion, politics, psychology and economy serve as "eco-
linguistic" (ecology of language) categories of non-linearity. With regard to the
Sorbs, the different political systems and the ideologies associated with them are
an essential factor in the non-linear development of the language. Finally,
neither the national socialist, the communist nor the market-economy system
has managed "to wipe out" the Sorbs completely. Within each of these three
ideological systems the Sorbs have been able to create political niches where
they could co-exist with the government of the day. Religion, in particular
among the Upper Sorbs, was a key factor in creating a counterbalancing sense of
identity in the face of unfavorable political conditions. The Protestant church
could not fulfil this supporting task for the Lower Sorbian minority. As a con-
sequence, this group has vanished almost completely.
Table 4. Ecolinguistic factors of the non-linearity of language shift
The German minority The Sorbs in Germany

in Old Belgium
Ideology - Change of political systems
Religion - Catholic religion as a factor to

strengthen the own identity.
Politics rural area

traditionally conservative
voting for the Christian-Socialist
Party.
industrial area
traditionally progressive
voting for the Socialist Party
Psychology The minority wants to speak the The minority tends to speak the
more prestigious French more prestigious German
language, but is unable to give up language, but the Sorbs are nev-
the German cultural heritage. ertheless attached to the Sorbian
cultural heritage.
Economy Minority in the periphery Minority in the periphery
Depending on the economic characteristics of the particular region, the Ger-

man-speaking minority in Old Belgium is more politically than ideologically
influenced. Traditionally, rural areas in Belgium are rather conservative and
support the Christian-Socialist Party, whereas industrialized areas of Old Bel-
gium (belonging to Wallonia) tend to be more progressive and prefer to vote for
the Socialist party. This has led to the disappearance of traditional regional cul-
ture, which has been replaced - in the case of the Sorbs - by an "international
workforce" like in the ideologized system of the Sozialistische Einheitspartei
Deutschlands [socialist union party of Germany] in the former GDR.
There are certain similarities in terms of political behaviour, and both mi-
norities show an "emotional" bond to their cultural heritage. Both try not to
speak the more prestigious language - except with non-minority peers, as their
minority language is associated with stereotypes such as backwardness or so-
cial inferiority. At the same time, they are unable or unwilling to give up their
original German or Sorbian culture. With regard to the Sorbs, this has led to the
coinage of the term "Wends", applied to those people who do not speak Sor-
bian any more but still cultivate Sorbian cultural heritage (see also Geske and
Schulze 1997).
The backwardness associated with minorities can be attributed to the geo-
graphic border situation in which they live. The German-speaking and Sorbian-
speaking minorities both live in economically outlying areas with strong rural
features due to re-agriculturization in recent history. Now however, these areas
have developed niches within a variety of economic sectors and as a result, the
economic border situation should not lead to a further decline of speakers. This
development is due to the counter-effects of regionalization which promote
small and medium-sized businesses of all types (see the EU-Agenda 2000, in
particular as to future agricultural policy) and also to the new media, which
makes contrasts between central and fringe regions less relevant.
These new media will certainly play an important part as a factor of non-
linearity in language maintenance and loss. Contact-linguistic research has
shown that every multilingual situation is inherently asymmetrical. There is al-
ways a hierarchy among the speakers or a difference in power. If this asym-
metry is institutionalized, languages become national languages and they and
their speakers dominate other ("smaller") languages. This process is strongly
influenced by the traditional electronic and printed media as the mouthpiece of
the national state. That is how the leading role of the more powerful group and
their language and culture is finally established (see Barret-Boyd etal. 1996).
Today, the new media allow cultures to meet and create more favourable
conditions for minority languages, regardless of a central nation state and in-
creasingly, new international media, subject to entirely different rules, are
supplementing national electronic and printed media (see Schröder and
Zimmer 1995).
National States International Organizations

(e.g. Germany) (e.g. forum "small" languages)
National/regional media Global media

(e.g. ARD) (e.g. digital radio)
Monolingual media Multilingual media
(e.g. German) (e.g. Minority languages in the EU)
One-way media: Separate media Interactive media: Integrated media
Figure 1. Media and society
Whereas media in national states were designed for one particular language
region only, global media of international organizations is orientated towards
their multilingual audiences. At the same time, advanced technology allows for
the integration of different media and interactivity.
The new media could form an essential factor for ethnic groups in Europe.
In a world of international and multicultural communities, it could facilitate the
forming of international unions of speakers of "smaller" languages, consisting
of many sub-groups. In this way, the new media could ultimately create a sense
of identity and thus represent an aspect of non-linearity. All these communities
are bound by common interest which need not necessarily rank below the in-
terest of majorities.
6. Conclusion: contact universale for defining a European

language policy
Descriptions of the features of the two minorities discussed in this article have
pointed, directly or indirectly, at their connection to a majority - French-speak-
ing Walloons for the German-speaking minority in Old Belgium and German
speakers for the Sorbian-speaking minority in Germany. Using ecolinguistic
categories of description, it is possible to examine universal features of contact
between minorities and majorities. For instance, in all cases there is a deter-
mined connection between the "backward" character of minorities and the
maintenance of their language in contact with speakers of majority languages.
A non-linear approach focuses on aspects of complex intra-group behaviour on
the part of the minorities which cannot be easily attributed to rational par-
ameters. This can lead to a further splitting-up of the group, creating "extreme"
and "balancing" characters. A further essential insight is that extra-linguistic
factors are more important for language maintenance or loss than intra-lin-
guistic factors and that language can be seen as a secondary symbol of under-
lying socio-economic differences in language contact. These differences deter-
mine the asymmetry of the language contact situation and in the end its
"winners" and "losers".
For an adequate differentiation of a European language policy, the follow-
ing conclusions have to be drawn:
- The needs of the language groups in Europe are so heterogeneous that a
centralized or uniform language policy cannot meet the requirements of
all ethno-linguistic groups and minorities.
- The recently frequently emphasized European Union's principle of sub-
sidiarity can be considered as an important step towards the consideration
of the interests of small language communities.
- Balanced socio-economic support of underprivileged groups contributes

to the survival of minorities.
- Results and findings with respect to strong groups should be transferred to
smaller and weaker groups.
- Europe has developed two different concepts for a future national lan-
guage policy: Whereas the (Mediterranean / Romance) South prefers
complex legislation for the recognition, protection and maintenance of
language communities, the North rather wants to help the language com-
munities to help themselves.
Given the heterogeneity of European minorities and the complexity of lan-
guage policy linear statements such as those made by Hoffmeister and cited at
the beginning of this article would seem unjustifiable. Accordingly, language
policy in the context of research on language maintenance and shift at language
boundaries should instead focus on areas where global and local concepts over-
lap. The consequences of on-going globalization and - at the same time - the
consequences of regionalization ("glocalization") showing specific features of
European minorities do not allow for generalization. Future research will
therefore have to develop new methods and models of multilingual research as
they are currently emerging from majority-minority contexts and from the new
economic status of minorities. It is one of the chief tasks of contact linguistics
to take on this challenge and to concern itself more intensively than in the past
with this field. This endeavour can thus serve as an outstanding example of ap-
plied linguistics in the 21th century, the significance of which for the viability
and survival of minorities in Europe, with more than 200 different languages
(nearly one hundred of which endangered according to Kraus 1992), cannot be
overstated.
Appendix
Map 1. German-speaking Belgium

Sorbische
Sprachgebiete
Lübben
Lubin
Cottbus
Calau Cboiebuz
Kaiawa
Sprembcrg
Senften-
bcrg ·
Hoyel
Wojei
Kamenz (
Kamjenc
Bischofswerda
niedersorbisch
Obergangs-
dialckte
obersorbisch
Map 2. Sorbian-speaking area in Germany (Source: Domowina [ed.] 1992: Die Sorben
in der Lausitz. Bautzen: Domowina.)
Notes
1. The German-speaking parts of Belgium that constitute Old Belgium are not admin-
istratively related to the eastern German-speaking region known as 'New Belgium'
(population: appr. 68,000) which was ceded by Germany to Belgium after the First
World War (Neide 1979).
2. The Task Force Human Resources, Education, Training and Youth of the European
Commission (GD XXII) decided in the Autumn of 1993 to commission a large-scale
study on the present situation of so called "lesser-used" languages of Europe. It in-

vited four of the tendering centers to form a team and work together in a multidis-
ciplinary fashion concerning the production and reproduction of minority lan-
guages. The centers investigated 45 languages through desk and field research for a
period of two and a half years. Almost 6,000 minority speakers have been inter-
viewed or were involved through Language Use Surveys. Euromosaic and Euromo-
saic II were conducted by 80 researchers and Language Group Correspondents.
3. This in contrast to New Belgium where German is an official language.
References
Barret-Boyd, Oliver, John Nootens and Anthony Pugh

1996 Multilingualism and the mass media. In: Hans Goebl, Peter H. Neide,
Zdenfek Stary and Wolfgang Wölck (eds.), Contact linguistics. An In-
ternational Handbook of Contemporary Research, 426-431, Vol-
ume 1. Berlin/New York: Walter de Grayter.
Eilenberger, Gert
1990 Komplexität. Ein neues Paradigma der Naturwissenschaften [Com-
plexity: a new paradigm in science]. In: H. Ditfurth and E. P. Fischer
(eds.), Mannheimer Forum 89 / 90 (Studienreihe Boehringer Mann-
heim), 71-136. München: Piper.
Elle, Ludwig
2000 Today's situation of the Sorbian language and concepts of its revital-
ization. In: L. Satara and S. Hose (eds.), Maintenance, Revitalization
and Development of Minority Languages, 17-21. Bautzen: Sorbian
Institute.
Faßke, Helmut
1997 Deutsch-Sorbisch. In: Hans Goebl, Peter H. Neide, ZdenSk Stary and
Wolfgang Wölck (eds.), Contact linguistics. An International Hand-
book of Contemporary Research, 1790-1797, Volume 2. Berlin/New
York: Walter de Gruyter.
Gabele, Eduard
1993 Portfolio-Planung [Portfolio-planning]. In: Vahlens großes Wirt-
schaftslexikon, Volume 2. München: Beck. [First published 1967.]
Geske, Anja and Jana Schulze
1997 Das Sorbische als Minderheitensprache. Darstellung einiger Prob-
leme des Spracherwerbs. [Sorbian as a minority language. Problems
in language acquisition]. In: Jürgen Erfurt and Angelika Redder
(eds.), Spracherwerb in Minderheitensituationen [Language acquisi-
tion in language minority situations], 128-164. Oldenburg: Druckhaus

Dresden.
Hoffmeister, Walter
1977 Sprachwechsel in Ost-Lothringen [Language shift in Eastern Lor-
raine]. Wiesbaden: F. Steiner.
Kern, Rudolf
1997 Französisch-Deutsch. In: Hans Goebl, Peter H. Neide, ZdenSk Stary
and Wolfgang Wölck (eds.), Contact linguistics. An International
Handbook of Contemporary Research, 1130-1136, Volume 2. Berlin/
New York: Walter de Gruyter.
Krauss, M.
* 1992 The world's languages in crisis. Language 68 (1): 4—10.
Macica Serbska
1991 Die Sorben in Deutschland [The Sorbs in Germany]. Bautzen.
Neide, Peter H. and W. Amian
* 1979 Plädoyer für eine soziolinguistische Betrachtungsweise einer westeuro-
päischen Minderheit [Pleading for the sociolinguistic view on a West-
European minority]. In: Peter H. Neide (Ed.), Deutsch als Mutter-
sprache in Belgien [German as a mothertongue in Belgium], 1-6. Wies-
baden: F. Steiner.
Neide, Peter H.
1979 Volkssprache und Kultursprache [Vernacular and standard]. Wies-
baden: F. Steiner.
Neide, Peter H., Miquel Strubeil and Glyn Williams
* 1996 The production and reproduction of the minority language groups in
the European Union. Luxemburg: Office for official publications of
the European Communities.
Schröder, Hartmut and Dagmar Zimmer
1995 Begegnung der Kulturen im Netzwerk? Neue Medien und Transkul-
turalität [Encounters of cultures in the network? New media and trans-
culturality]. Zeitschrift für Kulturaustausch 1995 / 3: 357-364.
Weber, Peter J.
1997 Sind Mehrsprachige erfolgreicher im neuen Europa? [Are multilin-
guals more successful in contemporary Europe?] In: Peter Η. Neide
(ed.), Sociolinguistica 11, 184-191. Tübingen: Niemeyer.
1. Which linguistic factors can be observed at language boundaries?

2. Why are non-linear processes important for the explanation of language
maintenance and language shift?
3. How far does the language profile of the Slavic Sorbs differ from that of the
Old Belgian Germans?
4. Which factors play a decisive role for "remigrants" (re-immigrants) to shift
back to the mother-tongue at a certain age?
5. Which ecolinguistic factors are significant for language shift?
Language shift among Siberian Estonians: pro and
contra
Jiiri Viikberg
1. Introduction
Some people emigrate from their native country to live among foreigners. An
individual would find it hard, no doubt, to maintain his or her identity against
the pressure of the new environs. Ethnic groups, however, have a chance to keep
up their own language and culture for a longer period. By investigating various
groups of immigrants, scholars have discovered a number of circumstances on
which the maintenance or loss of the mother tongue seems to depend. On the
basis of the studies by Gaarder (1977), Kloss (1966), Fishman (1972, 1980),
Haugen (1973) and Heye (1979), Grosjean lists several factors that are vital for
the survival of a migrant group's language (Grosjean 1982:107-112). The most
successful in this respect are those migrants who regard their stay abroad as
temporary, thus seeing no need to integrate into the foreign community. If the
sojourn continues, however, the ability to survive linguistically and culturally
starts to depend increasingly on the size and geographic concentration of the
group in question. If the group is numerous enough and lives in a relatively
compact area, its members can go on using their mother tongue outside their
homes, making it the basis of their communication and cultural life. A dis-
persed settlement and intermarriages, however, forecast a quick shift to another
language. It is also very important how positive (or negative) the immigrants'
own attitude to their language is and how liberal (or assimilative) the language
policy of the host country is. Those parameters have also been considered in the
present analysis of the linguistic situation of Siberian Estonians. It should be
noted, however, that many factors are ambivalent, that is, they may affect the
situation in opposite directions. In the following discussion we will try to pay
due consideration to both possibilities, starting from the positive effect that cer-
tain circumstances may have on immigrant language maintenance and proceed-
ing to the negative influences of the same circumstances.
The bulk of the following data comes from local informants during the ex-
peditions of author to Siberian Estonian villages in the period of 1984-1991.
Material from literary sources (Granö 1905; Maamägi 1980; Nigol 1918; Päll
1980) is illustrated by facts from oral narratives (about 300 hours of tapes).
126 Jüri Viikberg
2. Historical background
The Estonian settlements in Siberia can be divided into (1) older settlements
(colonies), founded by deported criminals in the nineteenth century, and (2)
younger settlements (villages), founded by volunteer migrants in the late nine-
teenth and early twentieth century.
Siberia has been considered a part of Russia since 1582, and it became a re-
gion where criminals and subversives from the European part of Russia were
sent for penal servitude. Estonia was incorporated into Russia in 1721, but
there are no reports of any Estonians having arrived in Siberia during the eight-
eenth century. The picture clears up in the nineteenth century when the
numbers of Lutheran settlers such as Estonians, Latvians, Finns, Germans, etc.,
had become sufficient to allow their own villages to be founded.
In 1845 the czar Nicholas I issued an ukase according to which Lutherans
were to be concentrated in the Ryzhkovo village, West Siberia (minor of-
fenders), and in the Minusinsk region, East Siberia (graver offenders), so that
they could maintain their religion and language. In 1861 some ethnic villages
had already been built up there; Estonians had their Viruküla (officially Revel)
in western Siberia and Pulan (officially Verhnjaja Bulanka) in eastern Siberia.
Ryzhkovo (founded in 1804) and Verhne Suetuk (founded in 1850) remained
with a mixed population.
Around the 1840s the idea of emigrating to Russia for free land and free life
became increasingly popular among Estonian peasants. Starting in 1855 Esto-
nians began to emigrate, first southward to Samara and the Crimea, then further
into the Russian inland until they reached Siberia at the end of the nineteenth
century. In 1897, 12% of all Estonians lived outside their native country; 4,202
of them lived in Siberia. The history of Estonian settlements in Siberia culmi-
nated in 1918, when 40,000 Estonians lived in more than 100 Siberian villages
(Nigol 1918: 39). According to census data the number of Estonian settlers in
Siberia stood at 32,000 in 1926, 20,000 in 1979 and 17,000 in 1989 (see
Table 1).
Most of the family heads who applied for permission to emigrate from Es-
tonia were over 35 years old. The majority were estate workers, farm hands, or
cottagers. In rare cases a tenant, a schoolteacher, a merchant, a cabman, or a
caretaker might apply. A remarkable number of Estonian villages have sur-
vived in Siberia up to the present. The earlier settlements are dominated by the
fourth and fifth generations of settlers, while in the younger ones the second
and the third generations set the fone. Differences in origin and native dialects
persist in Siberia: at least, North Estonians and South Estonians used to live in
separate villages.
Language shift among Siberian Estonians: pro and contra 127
Table 1. The number of Estonians in Siberia, 1939-1989
District 1939 1959 1970 1979 1989

Omsk 7,005 6,053 5,160 4,544 4,069
Tomsk 2,287 1,850 1,374 1,054 916
Kemerovo 2,986 2,562 2,008 1,688 1,454
District 1939 1959 1970 1979 1989
Novosibirsk 3,948 4,219 2,745 2,353 1,974
Krasnojarsk 10,979 11,075 8,529 6,993 6,061
Total in Siberia 30,764 29,033 22,548 19,465 17,273
Source: Village soviets, the Karatuzskoe CPSU District Committee, the Central Board
of Statistics of the Estonian SSR, and official census accounts.
3. Factors affecting language maintenance
First we will discuss the social factors that depend, first and foremost, on the
ethnic group in question: its size, status, activism and attitudes.
3.1. Size of group
The Siberian Estonian villages have traditionally been rather populous. In ear-
lier times a normal village used to have 300-500 inhabitants, while on better
soil the numbers could be even bigger: for example, 700 in Zolotaja Niva and
Vambola, western Siberia, 750 in Haidak, and over 900 in Verhnjaja Bulanka,
eastern Siberia (Nigol 1918: 42-60). There were also some rather small vil-
lages (e.g. of 70, 80,170, or 190 people), but those were either quite new (often
situated close to each other) or detached from bigger settlements that had be-
come overcrowded. Major changes were caused by the optation in 1920-1926,
and especially by the collectivization of the 1930s. Many villages disappeared
altogether and the remaining people moved to other villages (Viikberg 1988:
284).
128 Jüri Viikberg
3.2. Continued immigration
From the point of view of linguistic and cultural maintenance, it is true that the
older a colony, the greater the importance of later reinforcement. The earlier
Estonian settlements in Siberia were established by the 1860s. Being founded
by (predominantly male) deportees, those villages survived largely due to later
banishments and especially immigration. The younger villages, founded
around the turn of the twentieth century, have also received replenishment from
the home country. Estonian migration to Russia came to an end in 1920 with
the Peace Treaty of Tartu. In the 1930s the Siberian Estonian villages got new
arrivals from the European part of Russia, as people were fleeing eastward
from collectivization or from repressions. The massive deportations from Es-
tonia in 1941 and 1949 did not add to the colonies, and even the rehabilitations,
beginning in 1956, brought only a few new settlers. Since World War II the mi-
gration has been one-way only - westward to Estonia.
3.3. Geographic concentration
Estonian settlers have lived quite close to each other in several Siberian re-
gions. The Om colony in Jelanka parish included 12 Estonian villages; in the
historical district {kreis) of Marijnsk, seven of the 17 Estonian villages were
situated relatively close to each other (Nigol 1918: 43). Since the introduction
of collectivization several villages have disappeared and their former inhabit-
ants have moved to other settlements. Nowadays the concentration of Esto-
nians is not half what it used to be. We can still reckon five villages in the Om
settlement (Omsk district) and also five villages in the former kreis of Marijnsk
(now Tomsk and Kemerovo districts).
3.4. Proportion of Estonians in their own villages
As for earlier settlements, the West Siberian Ryzhkovo and the East Siberian
Verhne Suetuk have always been villages with a mixed Lutheran population. In
the 1860s separate villages were built for every ethnic group. As a result, in
Ryzhkovo the proportion of the Estonian population (beside the Latvians and
Finns) rose to two-thirds. Today it is about 50%. At the turn of the twentieth
century the number of Finns and Estonians living in Verhne Suetuk was about
equal, then the Estonians gradually began to predominate. As the younger
settlements were founded on the ethnic principle, the problem of other popu-
lations did not exist for them until collectivization set in.
3.5. Isolation
As this factor can favor both language survival and language shift, we prefer to
analyse it in different connections:
a. Isolation from other minorities. The contacts of the Siberian Estonians with
the aboriginal population (Tartars, Koybals, Kets, Khants, and many others)
have always been minimal, as their settlements had already been ousted from
those territories. Nor have they ever communicated much with the Khakass,
Tuvinians, Chinese, and other ethnic groups. In older settlements Estonians
lived side by side with other Lutheran minorities such as Finns, Latvians, and
Germans. In most cases, however, every ethnic group inhabited its own village,
or at least its own village end or street in a mixed village (like Kovalevo of the
Om settlement). Younger settlements were situated far from other Lutheran vil-
lages. We can conclude that at first Estonians lived in considerable, though not
absolute, isolation from other nationalities.
b. Isolation from the majority group. At first the contacts of the Estonian
settlers with Russians as the most numerous ethnos were scarce. They were
recognized, but their majority was not felt directly. Official matters (market,
hospital, lawcourts, etc.) were settled with the help of Russian-speaking medi-
ators. Only in the 1930s did the kolkhoz system bring about closer interethnic
contacts and an increased use of the Russian language. The propagation of
Russian was also enhanced by education being entirely transferred to Russian
in 1937. That was the end of linguistic and cultural isolation for the minorities.
c. Isolationfrom the native country. Isolation from the home country presents a
serious danger to emigrant language survival, whereas close connections, on
the contrary, act as support and sustain resistance to foreign culture. Most Si-
berian Estonians (with the exception of some deported criminals) have never
renounced their native country. Until 1920 letters were exchanged, books and
magazines were subscribed to, people kept in touch with life in Estonia. Ver-
nacular school and religious life also served to maintain a symbolic connection
with the homeland. New immigrants also kept nourishing the vitality of the col-
onies. The isolationist politics, however, adopted by Soviet Russia in the 1920s
and 1930s cut off practically all contact between the Siberian Estonians and
their now independent native country. The political barriers were removed in
1940 when the Republic of Estonia was annexed by the Soviet Union. After the
restitution of the independence of Estonia in 1991 intercourse has again sub-
sided, but this time the reasons are economic rather than political.
130 JüriViikberg
3.6. Educational or language policy of the group
Vernacular education was closely connected with the Lutheran creed. An adult
person, that is, one who had been confirmed, had to be literate, otherwise he or
she could not marry. In older villages most educational matters (as well as other
affairs) were the responsibility of the pastor. In younger villages people also
tackled the education problem as soon as the first adaptation problems had
been overcome. Confirmation classes were held and confirmation rites were
performed by an itinerant pastor. It was also important that, in addition to
national pride, the later settlers had a great appreciation for education. So it was
taken for granted that even in Siberia, the settlers' children would receive their
education in the Estonian language, retain their national identity, and pass their
language and traditions on to the following generations. Vernacular education
persisted until 1937 when it was thwarted in its prime.
3.7. Religious life
Like vernacular education, religious life is one of the most vital supports for the
maintenance of an immigrant group's mother tongue, especially if the faith dis-
tinguishes the group from its national environment. Lutheranism, indeed, dis-
tinguished Siberian Estonians from the Orthodox and Muslims. It also con-
tributed to the maintenance of the Estonian language by means of confirma-
tional obligations. In older villages the pastor was the actual spiritual leader of
village life, taking care of education, morals, temperance, etc. As younger vil-
lages never got around to building their own churches, they were served by itin-
erant pastors, who met people at prayer houses. As vernacular services were in
short supply, people would go to a Russian Orthodox church. In the 1920s pas-
tors were forced to leave the country and in the 1930s churches were turned
into warehouses or village clubs, babtism was considered punishable, wed-
dings were prohibited, etc. And yet, among the older generations religion still
retains its importance today. It is called the Estonian faith (in contrast to the
Russian Orthodox faith). There has always been someone to perform the bab-
tismal and funeral rites, and some religious printed matter, frayed from long-
time handling. Now the church of Verhne Suetuk has been consecrated anew
(1990) and restauration has begun. Twice a year the congregation is visited by a
pastor from Estonia.
Language shift among Siberian Estonians: pro and. contra 131
3.8. Activism (political, cultural, linguistic, etc.)
In addition to an active economic life, the Siberian Estonians stand out for their
cultural activism. Schools were centers of amateur art activities (choral sing-
ing, instrumental music, theatrical performances), directed by schoolteachers.
Local correspondents sent their contributions to the Estonian newspapers
Edasi, Siberi Teataja, and Kommunaar that were published in Leningrad (St.
Petersburg) and Novosibirsk.
3.9. Social configuration of the settler group
By occupation, the majority of settlers were, of course, land tillers. The village
also had its own artisans, shopkeepers, millers, etc., to meet the local needs, but
some services were also bought from representatives of other nations (e.g. Tar-
tar horse dealers and gelders; Russian leatherworkers, builders, pedlars). In ad-
dition to school and church functionaries the civil servants included a village
elder, secretary (often a schoolteacher), night watchman, storekeeper, and
other. As most vital functionaries (except for doctors, lawyers, and policemen)
would be found among the villagers, Estonian persisted in active use in all the
respective spheres. Most later clerks of village soviets, chairmen of kolkhozes,
and team leaders were Estonians as well. The fact that most of the settlers were
cultivators of land, who kept living and working in the same place, had a favor-
able influence on the survival of the language. Sufficient size of the settler
group and low foreign immigration to their villages also had a positive effect,
until the situation changed radically.
3.10. Intermarriages
This factor, which typically works against the maintenance of the native lan-
guage, has partly worked for the maintenance of the Estonian language in Sibe-
ria. As most deportees who founded the earlier settlements were male, some
had to marry females from other ethnic groups. If the couple was of the same
faith, there were no objections, although a partner with the same ethnic back-
ground would have been preferred. According to J. G. Granö's notes of 1902,
the Finnish men of Verhne Suetuk were quite willing to marry Estonian women
and switch over to the Estonian language (Granö 1905: 52). Some similar re-
ports have come down to us from Estonka (Omsk district), which is a younger
settlement. It is likely that there were also many Estonians who came to be as-
132 Jüri Viikberg
similated by other ethnic groups, but no doubt there was a period when inter-
marriages contributed to the Estonian immigrant group in Siberia.
4. Attitudes of migrants
Second, we will discuss the migrants' own attitude toward their mother tongue.
Siberian Estonians used their native language not only at home, but at school,
in church, in local village administration, in societies, and at cultural events.
Estonian also lived on in print. Russian was needed in offices and in communi-
cation with the outside world (documents, correspondents, etc.), while the
other Estonian communities and the home country were condacted in Estonian.
So Estonian served as the medium of thinking, communication, and self-ex-
pression.
4.1. Attitude toward one's native language
The Siberian Estonians are known to have held their own language in high es-
teem. This was especially noticeable with settlers coming from North Estonia.
It often happened that they could not remember where exactly their forebears'
Estonian home had been, but if it was somewhere near Tallinn (i.e. the area
whose dialects provided the basis for the literary standard), it gave the tradition
an almost sacred aura, as it meant that what they speak is purest Estonian. As
for settlers of South Estonian origin, they recognize that the literary language
supports the prestige of the North Estonian dialects, but this does not disturb
their love for their vernacular dialect. They also remember their own literary
standard (the literary language of Tartu), which was used until the early nine-
teenth century. This contradicts statements by Maamägi (probably inspired by
Marxist teachings) that some decline in the national identity of Estonians could
be observed as early as in the 1920s, and in the 1930s people began to prefer
Russian-language to Estonian-language education (Maamägi 1980: 74-158).
Such self-assimilative tendencies could perhaps be manifested in a very small
number of Estonians, who were thinking of nothing but their own future career.
4.2. Attitude toward ethnic pluralism
Although Estonians were aware of the indigenous peoples, contacts with the
latter were few. The fact that a vast territory was inhabited by many different
peoples was taken for granted. Communication was more lively with other ex-
iles (Finns, Germans, and Latvians in older villages) and later with other vol-
untary expatriates (Russians, Ukrainians, etc.). Although the number of differ-
ent ethnic groups was high, most groups lived separately in their own villages,
so that the development of common features, mainly concerning the material
aspects of life, was due to climatic and other enviromental conditions rather
than to any major interethnic influences. Russians were just one of the neigh-
boring nations, not the dominating one. Even if there was some minor friction
on religious and cultural grounds, the differences were perceived as nothing
out of the ordinary, like those between any different nations.
Despite the fact that those nations that showed respect for their own lan-
guage and culture (as well as for those of the others) and were in no hurry to be
assimilated were stamped with the deprecating Sovietist label of "national con-
finement" (e.g. Maamägi 1980: 35), this actual multitude of languages, cus-
toms, and cultural manifestations was a great cultural treasure for Siberia. To
some extent it has helped to resist extensive unification (Russification, Soviet-
ization) even up to the present.
4.3. Attitude toward linguistic purity
In the course of cultural adaptation the Siberian Estonians have enriched their
vocabulary with numerous loanwords. This was necessary to make people feel
better in the new environment, to help them identify with it. As most Estonians
were monolingual, the loans were accommodated to the vernacular system
without any harm to the latter.
The first loans were exclusively dictated by necessity. Parallel to linguistic
borrowing, it was common practice to coin new Estonian terms to fit the new
environment. This applies, for example, to several plant and animal names used
in more recently founded settlements (Viikberg 1992: 86-87). The purity of
Estonian was also promoted by the existence of vernacular education and print
media. Even in the Soviet period the massive advance of new vocabulary
brought along by social revolutions was met by an effort to accommodate the
neologisms to the Estonian linguistic system. The Estonian publications
printed in Russia from 1920 to 1937 ( E d a s i , Kommunaar with its literary
supplement Uus Kiila) carried several articles on how important it is to avoid
unnecessary citation loans and other Russianisms, paying more attention to
genuine Estonian coinages (Päll 1980: 85-86).
134 Jüri Viikberg
4.4. Attitude toward bilingualism
For a long time bilingualism was not a problem for Siberian Estonians. The
majority of resettlers were monolingual, and business and official affairs were
carried on with the help of those compatriots who were better read in foreign
languages. It is likely that Russian was better known in the older settlements, as
the exiles had had contacts with those Russians who were in penal servitude or
just working together with other emigrants. Those people acted as interpreters
as well as loanword mediators. Russian was not widely spoken, though. Bilin-
gualism was rather limited. According to some reports, the inhabitants of
Verhnjaja Bulanka would even have liked their men to marry Russian women,
to help their children learn more Russian. But evidently the plans for intermar-
riages did not work out at that time.
On rare occasions only two languages are involved in interethnic contacts.
While living side by side in mixed villages or mixed marriages, or participating
in public migrant works, some Estonians learned to speak some Latvian or
Finnish. For some reason or other, though, in mixed villages (e.g. Kovalevo and
Orlovka, Omsk district) the Finns, Latvians, and Germans were more active in
using Estonian than vice versa. Both in national and mixed villages the Esto-
nians have persisted in speaking their language and have thus often imposed it
on others too. The settlers themselves explain it as specific Estonian stubborn-
ness or even stupidity. We believe rather that it was the compactness of the
settlement and vernacular social life that evoked the respect of other minorities.
The language contacts of the Estonians were Finnish and Russian for the
older settlements and Russian for the younger villages. Russian-Estonian bi-
lingualism was not widespread. For example, in 1922 in the Tomsk gubernia,
3% of the Estonians knew Russian well, 16% knew it satisfactorily, 26% knew
it to some extent, and 55% did not know it at all (Maamägi 1980: 31).
5. Factors conductive to the language shift: social factors
The following discussion will first focus on certain objective social factors that
forced Siberian Estonian settlers to replace their mother tongue, and then those
factors will be dealt with that are revealed by their own attitudes.
5.1. New social life
With the advent of village soviets in the 1920s and collective farms in the
1930s, Estonians were also drawn into more active communication with other
nationalities. Some of the people moved away. (Some Estonians moved to
other villages in the hope of being able to join an Estonian kolkhoz or to shirk
joining one altogether. Those people who were labeled as kulaks were extermi-
nated.) Some villages disappeared, and settlers of other ethnic backgrounds
started to move in. This was detrimental to the areal compactness of the Esto-
nian settlements.
The importance of Russian, both absolutely and as a mediating language,
increased considerably. Villages entered into close contacts with adminis-
trative authorities over planning, rendering accounts, and receiving inspectors,
and the number of people in contact with the Russian language increased (e.g.
on account of men in military service). At the same time the domain of Esto-
nian was restricted. As churches were closed down and pastors were forced to
repatriate, vernacular religious life was wiped off the public scene (Viikberg
1990).
The fatal blow was dealt by the state policy of repressions culminating to-
wards the end of the 1930s. Thousands of Estonians were exterminated. In
1937 the Estonian schools were liquidated and all education offered since then
has been in Russian. The Estonian press and the publication of schoolbooks
and fiction were stopped as well. All Estonian printed matter found in raids was
destroyed. Rural correspondents were sentenced as spies having contributed to
"German" newspapers. Organized social activities (political, cultural, and
other societies) were banned. All this made the year 1937 a turning-point, para-
lyzing the hitherto active cultural and linguistic life of Siberian Estonians.
Foundations were laid for the development of bilingualism in a diglossic situ-
ation. For minorities it meant obligatory asymmetric bilingualism and the be-
ginning of the decline of their native language.
5.2. Educational and language policies
Here no essential changes have taken place since the end of the 1930s. The cur-
riculum is the same for all Russian schools. Even optional lessons in Estonian
are not taken, for fear of overloading the curriculum with another foreign lan-
guage.
136 Jüri Viikberg
6. Factors conductive to language shift: intrinsic factors
The following intrinsic factors depend on the minority group. As has been
mentioned above, several of them are ambiguous, as depending on the value of
the parameter they may favor either the survival or the replacement of the
mother tongue. Therefore, we would like to focus on them once again now.
6.1. Size of the community
Since World War II the Estonian linguistic enclaves have considerably dimin-
ished. In addition to the human losses suffered during the period of Stalinist re-
pressions and during World War II, the villages of Siberian Estonians have
been drained by migration to bigger centers and especially back to Estonia.
Most emigrants were in their prime.
The fate of the remaining villages depended on the state development pro-
grams of the 1970s. If a village is regarded as having good prospects, that is,
being capable of progress (as was the case with Zolotaja Niva and Ivanovka in
the Omsk district), it meant that construction was carried on and production
was concentrated there. As more labour was imported, the proportion of Esto-
nians dropped and young people were increasingly willing to change their
national identity. If however, a village was regarded as having bleak prospects
(Verhnjaja Bulanka in the Krasnojarsk district, Liliengof in the Tomsk district,
Lillikülä and Estonka in the Omsk district) the villagers could hardly be con-
demned for having no faith in the future.
After the completion of their studies at an institute or after army service,
young people just refused to return to the village. In Zolotaja Niva, for
example, the number of Estonians is only 170, in Lillikülä 180, in Estonka 90
(as of 1983), in Verhnjaja Bulanka 93, in Verhne Suetuk 315, in Kaseküla 327
(as of 1984), etc. The same tendency can be observed in larger areas as well. In
the Karatuzskoe region, Krasnojarsk district (near Minusinsk), for example,
the number of Estonians use to be 1152 in 1959; by 1970 it had dwindled down
to 888 and by 1979 to 565 (see also Table 1).
6.2. Isolation
a. From other Estonian communities. The decrease in areal compactness has

been accompanied by a decrease in mutual interaction of the Siberian Estonian
communities. Communication is carried on between the closest villages only
(e.g. Verhni Suetuk and Verhnjaja Bulanka in the Krasnojarsk district; Lillikülä
and Estonka, the villages of the Om settlement in the Omsk district; Kasekiila
and Liliengof in the Tomsk district, etc.). Communication with more distant
settlements, however, is rather irregular, being practically restricted to contacts
between relatives.
b. From other minority groups. Since the 1930s and especially after World War
II the isolation of Siberian Estonians from other nationalities has dropped dra-
matically, while mixing has increased accordingly. Only a few racial barriers
are perhaps still sensed. In 1983 some Russian town-dwellers from Tara in the
Omsk district mentioned that there was practically no difference between Rus-
sians, Latvians, and Estonians, as they were all white people. Things were said
to be different with "dark" people, that is, with Tartars and others Asians. Then-
character, way of life, and attitudes were considered so different as to impede
working together with them. Some Estonians from Verhne Suetuk who were
interviewed in 1984 and 1990 pointed out that the Tüvinians and Khakass who
lived in the neighborhood were not attractive enough for an Estonian to marry
them. Following the example of Russians, West-Siberian Estonians use the
word kirgissed 'Kirgizhians' to denote Asians and local people with a swarthy
complexion, as they cannot always tell one nationality from another (which is
indicative of infrequent contacts). Evidently the reasons for such an opinion lie
not so much in racial traits as in certain differences in religion and mentality
(customs, attitude to work, manners, morals, etc.).
c. From the native country. Despite the geographic distance between the Esto-
nian Siberian communities and their home country, contacts are maintained
both by letter-writing and directly, as many families have repatriate relatives.
Relatives are visited more in Siberia (i.e. in one's former home) than in Esto-
nia. Despite being impressed by the Estonian living standard, Siberian Esto-
nians tend to dislike the Nordic reserve, individualism, rationality, and some
other traits that are characteristic of indigenous Estonians.
When in Estonia, Siberian Estonians attract attention with their accent,
mindset, and principles of life. Both the Estonians who live in their native
country and those who live in Estonian linguistic enclaves in Siberia lack the
knowledge and desire to accommodate the conditions and traditions of one an-
other. Cultural contacts are too few. People subscribe to very few, if any, peri-
odicals, books, sound recordings, etc., and they are often not even aware of the
existing opportunities.
The reasons may lie in the fact that people with wider cultural interests fell
victim to repressions, repatriated, or (depending on their education and per-
spectives) have by now become oriented to Russian culture. Most Estonians
138 Jüri Viikberg
living in Siberian villages fail to share in the cultural life of present-day Esto-
nia. Without the background given by education in the vernacular, modern lit-
erary Estonian strikes people as a language barrier.
6.3. Urbanization
Urbanization concerns mainly younger generations. Most young people leave

home for bigger centers after having served in the army or having completed
their studies at an institute, etc., as new experience seems to breed dissatisfac-
tion with village life. Evidently working in town seems easier and life more
comfortable.
6.4. Social mobility
This issue associates with the above problem. Education and knowledge of
Russian give young people access to many more (and better) jobs, extends
one's circle of communication, and increase mobility. They have ceased to feel
bound to their village or nationality. Challenges and opportunities for self-real-
ization are more numerous outside one's native village.
6.5. Intermarriages
The proportion of intermarriages has been growing constantly. As peers are

relatively few in number, partners cannot always be found within one's own
ethnic group. The increased interethnic contacts have multiplied the marriage
options, and ethnicity is not the main criterion in partner selection. On the other
hand, those people who have chosen to integrate into the Russian community
(environment) may well find that intermarriage may considerably facilitate
their achievement of this goal. If parents want to interfere, they risk spoiling re-
lations with their children. There is a story from Kaseküla (Tomsk district)
about a mother who objected to a Russian son-in-law. This led to a most serious
quarrel between the mother and her daughter, so that the daughter did not even
attend her mother's funeral.
6.6. Social structure of community
There is a sure correlation between the nationality of village officials and the
local linguistic life. Villages differ in this respect. In Verhne Suetuk and
Verhnjaja Bulanka (Krasnojarsk district), for example, village affairs are con-
ducted by their own upraav "village chairman". The head of the community
center, the librarian, shop assistants, some school teachers, the motion-picture
projectionist, and the mail carrier are also ethnic Estonians.
As soon as the Kasekiila library came to be managed by a Russian, sub-
scription to Estonian books was discontinued. The amount of Estonian allowed
at school (if any) depends on the Estonian teachers. An Estonian film projec-
tionist keeps trying to bring some Estonian films to the screen as well, and that
is important, even if they are dubbed.
Top officials of Estonian descent, who could really have a say in the future
of the village, are a rarity. As the local offices are increasingly filled by non-
Estonians, the domain of Estonian is continuously narrowing down, while the
overall necessity of speaking Russian is ever more pressing.
6.7. Professions and occupations
With time many traditional trades (such as various handicrafts) have receded
and many new differentiated ranks have appeared.
Due to the general centralizing tendency villages have lost their dairies, gar-
dening farms, and other small enterprises that were formerly typical. Farmers
and cattlemen work on their own holdings and produce their own basic food-
stuffs (potatoes, meat, milk, butter, eggs). There are still a few cobblers (mostly
making furry boots from dog hides) and furriers (making traveler's fur coats
and leather mittens). Nobody has such handicrafts as a livelihood but retired
people, but demand creates masters. This also holds for some other walks of
life (healers, ritual masters).
Yet, wherever supply is abundant and choice extensive, the need to make
things falls off. Clothes, furniture, and household appliances have been shop-
ping items for a long time. Most new trades and professions are connected with
technological and cultural innovations. Villagers may include mechanics, ag-
ronomists, milking operators, instructors, laboratory assistants, etc. Their
special knowledge and skills have been obtained at school, together with the
corresponding terminology in Russian. Due to technological advances many
traditional trades such as miller, tanner, tailor, and others are vanishing or gone,
together with the respective terminology.
140 Jüri Viilcberg
6.8. Activism (linguistic, cultural etc.)
In general Siberian Estonians are not too active politically. This is why the cou-
rageous initiative of the Vambola village (Tomsk district) stands out especially
clearly. In 1968 there were plans to unite their kolkhoz that had been named
after Viktor Kingissepp with the Russian sovkhoz Sibir, but the villagers re-
sisted this idea so energetically that it overcame all the explanations and pre-
pared resolution texts.
The decrease in Estonians in absolute numbers as well as in their relative
proportion is accompanied by a similar decrease in their cultural and linguistic
activism. Community singing is done at home parties, mostly by the older gen-
erations. Of traditional festivals, only St. John's Day (June 7, old style) is more
or less commonly observed. Pentecost and Christmas are celebrated at home, if
at all. Traditions have been discontinued. The older generations are too old to
dance or play any musical instrument, whereas the younger ones are not able or
willing to. The new and fashionable that they crave come through the Russian
school and culture. Discs or tapes from Estonia are few (in Verhnjaja Bulanka,
these have all been collected in the club, labeled with the owners' names).
Active use of the mother tongue has been decreasing since 1937, so that
nowadays its functions are limited to home and social intercourse. Estonian has
receded before the advance of Russian in the public sphere, education, and cul-
ture. Whether a child knows the mother tongue or not is entirely up to the home
to decide. As integration into Russian society seems inevitable, more Estonian
children are raised in Russian from the start.
7. Factors conductive to language shift: attitudes
A lot still depends on the self-image of the minority group and on its attitudes
to its linguistic environment. This holds true of the Siberian Estonians as well.
7.1. Attitude toward the mother tongue
This attitude has ceased to be so unequivocally positive as it used to be earlier.

The mother tongue is valued highly by the older generations (born before
1930). As Estonian used to be the language of their education, both their think-
ing and emotions are bound by this language. Also, their children are emotion-
ally bound to their language, for they are proficient in it despite having attended
a Russian schools. Yet they know from experience that their Estonian is not al-
ways sufficient and that modern subjects often require resorting to Russian
terms. They also understand that their Estonian differs from the variety that is
spoken in Estonia and that it is inevitable that they cannot keep up with the la-
test developments.
Postwar generations (born in 1960 and later) do not value Estonian so
highly any more. Russian-language education and communication (plus army
service, etc.) have called their identity into question, unless they have decided
to repatriate, of course. Due to the natural cleavage between generations they
often do not care for the traditions and songs of their parents.
7.2. Attitude toward the majority language
Russian has been regarded as the majority language by Siberian Estonians

since the 1930s. As the advance of Russian was directly connected with social
reorganizations and technological and cultural innovations, its prestige is high.
Russian does not so much serve as an ethnic identifier; it is rather associated
with education and civilized existence. The necessity of knowing Russian is
recognized by all generations.
7.3. Attitude toward bilingualism
Here opinions differ. Nobody questions the necessity of Russian, but younger
people tend to question that of Estonian. Language being a cultural symbol, it
is not always important what language a person is fluent in, it is his or her com-
munity that counts (Lehiste 1976: 8). Consequently, the struggle goes on for
new generations. The older people feel deeply hurt if the young ones give up
speaking Estonian. So the former try to appeal to the latter's conscience by re-
minding them of homeland: "How's that: you'll go to Estonia and you won't be
able to speak the language!?" Sometimes they advise their children to go to live
in Estonia, just in order to preserve their mother tongue. This indicates that the
older people favor bilingualism. As for the younger generation, some of them
go on living in their native villages and manage their affairs in Estonian. Others
have gone to live elsewhere. They use Estonian while visiting their home, but
only when speaking on domestic subjects; everything else is preferably re-
ferred to in Russian. The rest hurry to be Russianized at all costs. For them,
their ethnicity and mother tongue are a scorned memory, something to move
away from and forget about. In this situation linguistic purity has ceased to be a
problem. Under the conditions of isoglossia or massive bilingualism language
switching is preferred to a total abandonment of Estonian.
142 Jüri Viikberg
7.4. Majority group's attitude toward the minority group(s)
The acute public intolerance typical of the 1930s has diminished. At the same
time every measure has been taken to assimilate the Siberian Estonians. Teach-
ers are known to have forbidden children to speak Estonian even at home, ar-
guing that otherwise it would be more difficult for them to acquire Russian
properly. Estonian has repeatedly been an object of disparaging remarks such
as, "This Estonian of yours - what language is it anyhow? In Estonia nobody
could understand you!" Usually complete transition to Russian is offered as
the only remedy. In this connection the Estonian customs and traditions (no
eating at cemeteries, no lamenting at Lutheran funerals, celebration of St.
John's Eve, Christmas, etc.) have sometimes also been looked down on. Some
Estonians, having experienced that in town their conversation attracts undue
attention, have decided that it is better not to use Estonian during their visits to
town.
8. Conclusion
The Estonian villages founded in Siberia in the nineteenth and early twentieth
century have largely survived into the present. As the exiles came from differ-
ent dialect areas, their villages developed their own North-Estonian-based
koine, which has no precise counterpart in Estonia. As for voluntary expatri-
ates, they have mostly retained their dialects. As the resettlers came from a
smaller territory, their dialectal background is more homogeneous, but more
specific dialect traits have receded before the more widespread ones. North Es-
tonians and South Estonians live in separate villages.
The Estonians have been able to retain their language and identity for a
relatively long period because they were numerous enough, they lived in
relatively compact areas, their settlements kept receiving new immigrants and
information from their native country. Active cooperation, tuition in the ver-
nacular, and religious and social life also had their positive effect. Although ac-
culturation proceeded by means of Russian, it took place on an adaptive Estonian
basis. Most resettlers were monolingual; the numerous loans (names of animals,
plants, articles of clothing, foods, etc.) were adapted to the Estonian linguistic
system. Place-names of the closest surroundings (microtoponyms) are Estonian.
A massive advance and breakthrough of Russian started in the 1930s. The do-
mains of Estonian usage began to shrink and the grammar system suffered from
foreign influence. State reprisals were used to oust Estonian from public sphere
in 1937. Since then Russian has been the unrivalled language of education and
culture, as well as the unofficial state language. As a result, mass bilingualism
developed under the conditions of diglossia. All settlers born during 1930 and
1950 are bilingual. In the older generations Estonian still dominates over Rus-
sian, in the younger ones the situation is vice versa. The postwar generations are
on their way from Russian-Estonian bilingualism to Russian monolingualism.
Normally the communities would probably have reached language shift due
to a decrease of population, integrative orientation in younger generations,
mixed marriages, and urbanization; and important factors they have all been in-
deed. But here the primary role was played by the repressive policy of assimi-
lation carried out by Soviet Russia, which either induced the other factors or en-
hanced their influence.
References
Fishman, J. A.
* 1972 Language and Nationalism. Rowley, MA: Newbury House.
1980 Language maintenance. In: S. Thernstrom (ed.), Harvard Encyclo-
pedia of American Ethnic Groups, 629-638. Cambridge, MA: Har-
vard University Press.
Gaarder, A.
1977 Language maintenance or language shift. In: W. Mackey and T. An-
dersson (eds.), Bilingualism in Early Childhood. Rowley, MA: New-
bury House.
Granö, J. G.
1905 Siperian suomalaiset siirtolat [Finnish settlements in Siberia]. Fen-
nia 22 (4): 1-64.
Grosjean, F.
1982 Life with Two Languages. An Introduction to Bilingualism. Cam-
bridge, MA: Harvard University Press.
Haugen, E.
* 1973 Bilingualism, language contact and immigrant languages in the United
States. A research report 1956-1970. In: T. Sebeok (ed.), Current
Trends in Linguistics 10, 505-591. The Hague: Mouton de Gruyter.
Heye, J.
1979 Bilingualism and language maintenance in the two communities in
Santa Catarina, Brazil. In: W. McCormack and S. Wurm (eds.), Lan-
guage and Society, 401-421. The Hague: Mouton de Gruyter.
144 Jiiri Viikberg
Kloss, H.
1966 German-American language maintenance efforts. In: J. Fishman
(ed.), Language Loyalty in the United States. The Hague: Mouton de
Gruyter.
Lehiste, I.
1976 Keelekontakt - keelekonflikt [Language contact - language conflict].
Mana 42: 5-14.
Maamägi, V.
1980 Uut elu ehitamas. Eesti vähemusrahvus NSV Liidus (1917-1940)
[Building a new life. Estonian minority in the Soviet Union
(1917-1949)]. Tallinn: Eesti Raamat.
Nigol, A.
1918 Eesti asundusedja asupaigad Venemaal [Estonian colonies and settle-
ments in Russia]. Tartu: Postimees.
Päll, Ε.
1980 Pool sajandit tagasi [Half a century ago]. Emakeele Seltsi aastaraamat
[Yearbook of the Mother Tongue Society] 24: 81-93.
Viikberg, J.
1988 Vanematest eesti asundustest Siberis [About older Estonian settle-
ments in Siberis], Keel ja Kirjandus 5: 284-288.
* 1990 The Siberian Estonians and language policy. In: D. Gorter, J. F. Hoek-
stra, L. G. Jansma and J. Ytsma (eds.), Fourth International Confer-
ence on Minority Languages, 175-180. (Volume 2: Western and East-
ern European Papers.) Clevedon: Multilingual Matters.
Viikberg, J.
1992 Estonians in Siberia. Fenno-ugrica Suecana 11: 79-90.
1. What kind of "islands" are linguistic enclaves?

2. How are a language shift and "the law" of three generations (in general)
connected?
3. Which choices does a non-native have, if his or her mother tongue has a low
status and prestige?
4. Discuss J. W. Berrry's classification of acculturation (= assimilation, inte-
gration, separation, marginalization) and Siberian Estonians.
5. Compare linguistic purity in linguistic enclaves and purism in general.
On attitudes towards Croatian dialects and on their
changing status
Damir Kalogjera
1. Normativism and the dialects
Croatian philology has been inordinately absorbed in the problems of the norm
of the standard language, so much so that some obvious open sociolinguistic
questions have remained either unnoticed or neglected or treated superficially
and summarily, as if they were of minor importance.
This is particularly true of the following sociolinguistically important
topics: the use of regional dialects in everyday communication, attitudes to-
ward dialects, their status, their (covert?) prestige or the lack thereof, dialects
in contact with the standard, dialects and identity etc. Dialects have remained
the subject of the traditional, historically oriented dialectology, and little has
been said about their social function.
Philologists' absorption in normativism may be ascribed to the fact that
Croatian is a comparatively young standard language (Brozovic 1998b: 64). Its
beginnings, on the Stokavian dialectal basis, have arguably been dated to the
seventeenth or eighteenth centuries, but the actual development and elaboration
of the standard in the modern sense of the term must be linked with the growth
of literacy and national awareness in the nineteenth century when it was replac-
ing German, Italian, and Latin as the language of culture in the Croatian lands.
The discourse about the norm in a general sense has also been stimulated by
the political component in its codification. By the mid-nineteenth century, in
the political climate of South Slavic cooperation and reciprocity (in the face of
the pressure of great foreign languages, cultures and nations: German, Italian
and Hungarian) Croatian and Serbian, being at least declaratively, based on the
same dialect, were proclaimed a single standard language by the contemporary
philological authorities and political groups. But still, there has always been
among the elites a stream that has insisted on keeping the Croatian and Serbian
standards apart.
Thus the question as to what is to be included in or excluded from the standard
language, that is, its codification, covered at least two separable problems: (a) the
selection of phonological, grammatical or lexical forms, especially where written
146 Damir Kalogjera
usage was divided, and (b) the selection of "pure" Croatian items vis-a-vis both
foreign borrowings and the Serbian usage. Interventions along these lines have
more recently been justified by claiming e.g. the Prague School construct of the
autonomy of the standard language (Katicic 1986: 78) for the interference to ap-
pear more linguistic and less political. Orthography has also played a considerable
symbolic role in the latter connection. Writing the future tense as one or two words
contains a symbolic charge (Cr. rad.it cu vs. S. radicu Ί shall work'). An investi-
gation of this process offers an excellent insight into how ideology and selection of
linguistic items can be subtly interwoven.
The preoccupation with the tasks of codification and statements in that connec-
tion have contained an implication that the dialects (except for the selected Stoka-
vian) are of little importance in standardization, since their demise is nigh and
their full replacement by a well streamlined standard language has become inevi-
table. This point of view seems latent (if not explicitly stated) in various contribu-
tions by normativists in guides to correct usage, in the periodical for Croatian lan-
guage culture Jezik ('Language'), in occasional articles in the leading and minor
newspapers and weeklies and in language broadcasts instructing the public in cor-
rect usage of the pure Croatian. Dialects are mentioned there only as sources of
dialectisms, that is, errors a careful speaker should get rid of.
Besides, it is prescribed in advance where and when the standard language
is expected to be used. However, aware that their precepts are in conflict with
reality and that in many cases where standard is foreseen some form of dialect
is used, even by educated speakers, normativists apply the term "parasitic" for
such dialect use. (Brozovic 1966), which says something about their attitudes.
Looking at the literature on Croatian language standardization, it could be
claimed that in the terms of the four phases connected with the process, (Haugen
1966), selection, codification, elaboration (of function) and acceptance, the
least research has been carried out and the least consideration has been given to the
last point, the acceptance of the standard. The interested scholar will find a good
deal of literature on the historical step of the selection of the basis for the standard
and on the disputes raging about that selection (dilemma between rural or bookish
idiom) in both Croatian and Serbian philological and cultural circles (Vince
1978).The problems of codification (around the names of Ljudevit Gaj, Adolfo
Veber-Tkalcevic, Tomo Maretic, Ivan Broz and Fran Ivekovic etc.) and elabor-
ation (e.g., the role of Bogoslav Sulek for scientific terminology) have also been
dealt with. However, neither the question of whether the standard norm is actually
accepted in its spoken and written forms in the various domains where normativ-
ists expect it to be followed, nor the issue of its competition with the regional dia-
lects has been thoroughly and empirically investigated.
On attitudes towards Croatian dialects and on their changing status 147
Occasionally, writers and poets had written about their internal struggles in
switching to the standard (Nazor 1942) but it seems that normativists have not
been interested. They prescribe, or at least authoritatively suggest, solutions to
dilemmas in usage (the better informed ones invoking autonomy of the stan-
dard with regard to the dialects, language system, unnecessary synonyms and
other linguistic concepts as their guides for inclusion or exclusion of items) but
whether such fiats have been accepted by the general public seems to be of little
concern to them.
Moreover, there is a strange belief among some normativists, sometimes ex-
plicitly expressed (e.g., Babic 2000: 172), that when some normative solution
they propose has appeared in print a couple of times, it has, by that very fact,
been generally accepted.They seem to be forgetful of the fact, that before ap-
pearing in print, texts have been diligently scanned by sub-editors, known as
lektors, who are in most cases former pupils of the very authors of the precepts,
who are also sometimes dependent on their recommendation for the job, and
who duly see to it that the precepts are applied. If the normativist "gatekeepers"
had had more direct access to the original texts (before lektors' interventions)
or listened carefully to the impromptu speech of educated speakers in public
communication, they might think twice before jumping to conclusions. Or per-
haps they wouldn't. Because what transpires from some of their statements
(Brozovic 1999) they seem to refer mainly to a kind of "one-style" written lan-
guage in spite of paying lip service to the Prague doctrines about functional
styles within the standard. Little help in this respect can be expected from the
Croatian National Corpus of 30 million words since it consists exclusively of
sub-edited texts from publications since 1990 (Tadic 2000: 26), incidentally,
when that sub-editing became, if anything, more strict. The degree of accept-
ance of innovations or neologisms in language by a speech community surely
requires wider analyses of both the spoken and written language corpora. This
is also true about the acceptance of the canonical standard as a whole.
The acceptance of the standard and attitudes toward it especially in com-
petition with regional dialects will certainly sooner or later open up a wide field
of revealing research. Some perceptive early remarks about this question were
made by Brozovic (1966) who highlighted certain features of colloquial Croa-
tian in Dalmatia, and some work along this line has been done by Magner
(1966), Sojat (1979, 1998), and Bauer (1991) for Zagreb urban use and Ju-
tronic (2000) for Split.
In anticipation of major empirical research a few participant-observer im-
pressions along these lines might be of some interest. We may begin by making
a few claims. In everyday communication at any level from private exchanges
148 Damir Kalogjera
to formal use (e.g., in the Parliament, in serious discussions between highly

educated speakers) regional dialects (in their various forms: local dialects, re-
gional koines, urban dialects) are noticeably present, primarily in accent and
lexis but then also in grammar and style, that is, they carry a considerable com-
munication load (Magner 1966; Sojat 1998). The urban versions of these dia-
lects, (e.g., those of Zagreb, Split and Dubrovnik) enjoy an obvious prestige in
their appropriate areas and they are, as Haugen would have put it, "salon-
fähig". The more localized versions have a strong identity mark for their speak-
ers and thus are kept alive, of course in constantly changing varieties under the
inevitable influence of the standard. It may be said that the situation in Croatia
in that respect is closer to that of Italy and Germany rather than to that of
France (Malkiel 1984; Magner 1966). What is meant here by a dialect as used
in present-day communication (Kajkavian in the north-west, Cakavian along
the coast, and Stokavian in the rest of the country, named according to their re-
spective question words: kaj, 6a, sto meaning 'what') is not the idealized or
most conservative form containing every possible linguistic item historically
belonging to it, but a dialect-colored form of speech that is the result of a cen-
tury-long contact with the standard taught at school and used in the media and
in the literature. To call these varieties simply substandard would be too gen-
eral because they have a strong regional flavor. Thus a non-standard example
like:
Sto budeS tamo radio?
Localised version: Kaj buS tam delal?
Standard: Sto ceS ondje raditi?
'What are you going to do there?'
with the characteristic form of the future tense, belongs to the Kajkavian area
of Northern Croatia. However, a non-standard sentence like:
Dodi had budeS moc.
Localized: Doji ka' budeS moc.
Standard: Dodi had budeS mogao.
'Come when you can make it.'
with the peculiar form of future exact, comes from Southern Croatia, the Ca-
kavian Dalmatian coastal area. Such speech patterns are regularly accompa-
nied by corresponding regional prosody.
Another feature characterising these regional dialects is ample borrowing
from the once dominant languages mostly in the urban areas: German in its
Austrian version in the north, and Italian in the Venetian version along the Ad-
riatic coast. Some type of diglossia with bilingualism, with Austrian and Vene-
tian as High and the local dialects as Low, existed in those areas for a long peri-
od of time and lasted until the beginning of the twentieth century when the bal-
ance was tipped by demographic, educational and political events and Croatian
shifted to Η position with Venetian and Austrian gradually disappearing. What
remains is a post-bilingual situation that could be named metaphorically as
rudimentary bilingualism (Kalogjera 1991). The overt borrowings can be
found in the form of vocabulary items and phrases. The covert loans are to be
recognised in caiques at the lexical and even grammatical level. Proverbs, say-
ings and swear-words, filtered through Croatian phonology, can also be heard.
Thus in the northern dialects, mostly in Zagreb,a formal greeting reserved for
elderly people: Ljubim ruke Ί kiss your hands' is certainly a caique on the Ger-
man model Küss die Hände. The coastal dialectal set-phrase: diniti (l)jubav, 'go
steady' (of girl and boy) is structured on the Venetian model far 1'amor, (literally
'to make love'), standard Croatian: ljubovati. The overt alloglotic vocabulary
items make a good part of any differential dialect dictionary, a number of which
have been compiled.
Influential present-day Croatian philologists engrossed in the task of setting
up the standard norm, have never been outspoken and decisive about this prob-
lem, and while being energetic in their insistence that the public follow closely
the prescribed standard norm (Katiöic 1986: 79), they do not offer equally clear
instructions with regard to what to do with the ubiquitous use of dialects. Thus
it would seem that the dialects have either disappeared or are not worth any at-
tention. Although the situation has improved somewhat in the last few decades
(cf., e.g., Tezak 1977), this state of affairs still creates in local cultural circles
and in the teaching profession a certain amount of confusion: influential phil-
ologists harping about a rigid norm and the educated public taking them seri-
ously only in formal writing, but even there rather loosely, as any lektor pre-
paring the text for publication will tell us. This uncertainty about the status of
dialects in relation to the standard norm, which one acquires with one's edu-
cation in our society, makes the life of the teachers of Croatian in schools es-
pecially difficult and sensitive. Generations leave school without any socio-
linguistic orientation on how to look at their dialect in comparison with the
standard language. Is it really an unwelcome parasite or something else?
Why statements on dialect are avoided or are vague? It may be of interest to
say something about the possible reasons why regional dialects in their pres-
ent-day form have been an attractive subject of enquiry neither for philologists
and linguists nor for traditional dialectologists. Some of the reasons appear to
be general, as they are to be found in a number of European languages, while
others may be unique for Croatian. The first reason that comes to mind is that
the dialect-colored varieties being the result of dialects in contact (the standard
150 Damir Kalogjera
and the local dialects) are "diffused", unstable varieties and cannot be de-
scribed and studied comfortably by applying the methods of traditional dialec-
tology oriented toward "pure" local dialects with the data elicited from a few
idiolects of elderly speakers. In addition, there is this "macaronic" side to the
dialects with alloglotic loans not just as overt lexical items but as covert sub-
stratum and adstratum influences. Traditional dialectologists cannot but see in
such a variety a mixture of two or more historical dialects too chaotic to be
manageable by the application of their concept of dialectology (Ivic 1971: 95).
The second reason is that thus "diluted" regional dialects appear to be a
fleeting transient phase of language in its progress toward the canonical stan-
dard. Normativists seem to imply this replacivist and homogenizing view in
their work, as we mentioned earlier. But the questions of why this process of re-
placement and homogenization is rather slow and why dialects still survive are
not posed. The correlation between the structure and selection of linguistic
items and the structure of society, social groups, and communications networks
is hardly mentioned in Croatian philology.
It seems misguided to model a standard language on historical, logical, and
patriotic bases as an ideal closed system in a vacuum without taking into con-
sideration its speakers, their language repertoire, and their code-switching ha-
bits in speech and writing. It could easily become a standard language whose
rules and selections are observed only by its designers and a few of their pupils
or, in an optimistic scenario, by some fellow philologists, (cf. Brozovic on
"classical" language 1966, 1999: 151-181), a situation which may eventually
lead to unnecessary diglossia.
Regional dialectal expression is going strong in Croatia and there is nothing
negative in this fact. The spontaneous preservation of dialectal expression sug-
gests that speakers need it, probably, as "acts of identity" (Le Page and Tabour-
et-Keller 1985: 14), as a sign of group solidarity, and for other social and cul-
tural reasons. The loss of a dialect, particularly when engineered, must be a
loss for that community's culture (Trudgill 1995). If that is recognized then the
implied "replacivism", according to which the dialects will be substituted by a
perfect standard, has to give way to a serious treatment of dialects and their
present- day structure and function in communication parallel with the build-
ing up of the standard norm. Standard Croatian, perhaps in a less rigid and less
exclusivist form than the one suggested by a model offered by some of the
latter-day normativists, has definitely gained social prestige in Croatian society
and is becoming an asset in the social mobility of its speakers. To help the
speakers (particularly those of highly localized dialects) to use it in the proper
circumstances means to enable them to code-switch consciously.
Educationally speaking, this would mean that the structure of the standard
should be presented in parallel with that of the local dialect utilizing some con-
cepts and method of contrastive linguistics. Instead of reprimanding the stu-
dents for using "corrupted dialectisms", which they ought to forget, it would be
more productive to analyze them in contrast with the corresponding standard
forms. This proposal may sound like a chimera nowadays when language pu-
rity seems to be the most absorbing topic, but if the departments of Croatian at
the universities introduced some changes in their programs along these lines
(contrastive sociolinguistics?), new generations of graduates in Croatian could
apply such an approach and probably welcome it. This would contribute to the
feeling of well-being among a segment of Croatian society as it would rid many
speakers of the conscious or tacit tensions they face as dialect speakers versus
the standard.
2. A boost to the dialects and a misunderstanding
Another participant-observer impression concerns the changing symbolic

status of the three Croatian dialect groups (Cakavian, Kajkavian, and Stoka-
vian) in the lay public eye in connection with certain ideological trends in
Croatian philology. This change has been triggered by recent attempts (since
the 1960s) of a group of influential Croatian philologists to revise and reinter-
pret certain historical aspects of the Standardization of Croatian to emphasize
the view that Croatian and Serbian have had distinct standardization histories.
The debate attracts public attention as it is highly politicized and sometimes
suitably simplified for laymen to take part.
The gist of the argument that interests us here would be as follows. When
Ljudevit Gaj and the Illyrians, a cultural circle that started the movement
known as the Croatian National Revival, decided to take Stokavian as the basis
of the standard Croatian language in 1836, they selected a conservative form of
Stokavian characterized by certain archaic features like the case endings (geni-
tive, locative, instrumental) and a morphophonologically based orthography.
Then at the turn of the century another linguistic reform took place championed
by the philologists Tomislav Maretic, Ivan Broz, Fran Ivekovi6 and others, fol-
lowers of the ideas of the Serbian language reformer Vuk Karadzic, who intro-
duced a "younger" Stokavian dialect (as allegedly spoken in eastern Herze-
govina) as the basis of the standard and a phonological orthography, which
considerably changed even the visual aspect of the standard, which now lost its
archaic patina and became closer to Serbian.
152 Damir Kalogjera
The reformers also prescribed that the only admissible corpus of texts on
which to base the literary language codification should be the works of Ka-
radzic and the folk poetry that he had collected. The crown witness of this "pur-
ism" is the volumes of the Dictionary of the Yugoslav Academy of Zagreb,
which excluded most items coming from non-Stokavian literary sources. Al-
though there were opposing voices to such rigidity even among the supporters
of the reform (e.g., the reputed Slavicist Vatroslav Jagic), not to mention its ad-
versaries (e.g., Antun Radic, politician and writer), the reformed standard has
been functioning ever since with lots of changes and amendments in ortho-
graphy, and at other levels - changes that were most frequently politically
motivated.These changes are an absorbing topic for sociolinguistics where lan-
guage and political ideology intertwine, but that is a wider issue and we are
covering just a segment of it.
Since the 1960s this turn-of-the-century reform has become seriously ques-
tioned. New evidence is being produced or some old evidence recycled to show
that the reform had cut short the normal development of the Croatian standard
by forcing the 'purist' Neo-Stokavian based standard, which went against the
Croatian tradition, namely that the standard language be open to selected
Kajkavian and Cakavian elements, especially lexis. (We have seen that, syn-
chronically, there is less space for dialectisms). Some radical writers saw it as a
subversive act against the Croats, whicht was forged by the Hungarofiles in the
Croatian Parliament of the day as a concession to the Serbs. Others criticized
the "purism" of the reform, which made inadmissible for codification the
works of older Croatian authors.
These arguments from the history of standardization initiated by linguists
have been taken over in political disputes by a wider public. In the process, they
have become simplified and give rise to unexpected results, namely a suspi-
cious image of standard Stokavian, that is, the Standard Croatian. This state-
ment requires some explication.
Now, since Stokavian was selected as the basis for the standard language,
many compliments and much praise have been lavished on its supposed vir-
tues. Thus it has been described by various scholarly and popular sources as the
most intelligible of the Croatian dialects, phonologically transparent, eu-
phonic, possessing an aesthetically unique tonal accentual system, free of "bar-
barisms": that is, mostly features which are linguistically hard to prove. The
two remaining dialect groups obtained considerably less adulation. In this way
a general feeling has developed among the public that Stokavian is a very privi-
leged dialect. And in spite of the normativists' insistence on the autonomy of
the standard, i.e. its freedom from direct dialectal influence, the public has be-
come aware that a Stokavian speaker is closer to the norm than speakers of the
other dialects, and that the latter have to make a considerable effort to master it.
Stokavian speakers have also been vaguely aware of their advantage. And since
the speakers of the other two dialects have been necessarily taught at school
that they cannot use their vernacular either there or in other formal contexts,
they could only conclude that their dialect was a corrupted form of the standard
language. Their teachers, with their inadequate sociolinguistic briefing, could
hardly help them to see otherwise.
The above scenario seems to be more or less general with languages where
one of the existing dialects becomes the basis of the standard language. This
has been noted in the sociolinguistic literature (Haugen 1966), and Fishman
sums the phenomenon up by the (inverted) saying "Heroes are made not born"
(1975: 26).
The present-day critical views on the alleged imposition of Neo-Stokavian by
the turn-of-the-century reform, with the connivance of forces inimical to the inter-
est of the Croats, have obviously produced in the minds of the general public some
confusion, since it may easily be understood as a criticism of the entire present day
Standard Croatian language. This perception may have been reinforced by recent
insistence that whatever is found in the Croatian usage which is more typical for
Serbian has come there exclusively under pressure, while Croatia and Serbia were
parts of the common state of Yugoslavia. And indeed, in the less sophisticated
media and in lay debates, this idea crops up now and then. According to this opinion
Croatian should simply be different. But the recipe for the change remains unspeci-
fied. Thus a journalist queried a Croatian linguist recently about the possibility of
developing a Standard on the basis of the Kajkavian dialect (Vjesnik 5/9 2000:13).
There is another rather positive line of activity that, however, may be mis-
construed. Local cultural clubs and societies cultivating folk art and dances have
recently included among their tasks the preservation of the local dialect. Kajka-
vian and Cakavian festivals have been popular (especially since the 1960s) and
have sometimes become events of national importance. Couldn't not this be im-
plied resentment against the "imposed" standard? Couldn't this lead an average
dialect speaker to conclude that the efforts of mastering the standard, since it was
imposed and based on a practically alien dialect, are not worthwhile? This might
also be interpreted as the small man's backlash.
In view of our criticism of the marginalization of the dialects "that have not
made it" one should welcome this development as a support for our argument,
that is, the necessity of paying more attention to the structure and social func-
tions of the dialects. However, if this lively dialectal activity upgrading the
status of Kajkavian and Cakavian is meant as opposition to the standard lan-
154 Damir Kalogjera
guage, (which the dialects, as it were, could substitute in some way) it can be
only seen as a misunderstanding, as a conclusion (admittedly never made ex-
plicitly) based on false premises.
Standard Croatian on a Neo-Stokavian basis, as has been said, has built its
prestige in its spoken and written versions and consumers of the media have got
used to a smooth communication by means of this variety. Such communi-
cation can be just as easily disturbed by an overly "correct" and puristic version
of the standard as by a version that is broadly regional. Today, however, there is
no way back. Some Croatian linguists have become aware of this misunder-
standing and are trying to disentangle the confusion which, in a way, has been
triggered by a simplification of their own statements since the 1960s (Brozovic
1998a).
The whole dialect-standard complex calls for more empirical research
along psycholinguistic and sociolinguistic lines, and this article is only meant
to outline some problems worth investigating.
References
Banac, Ivo
* 1990 Main trends in the Croatian language question. The Bridge / Most, Vol-
ume 1., 7-96. Zagreb: Croatian Writer's Association.
Bauer, Ivan
1991 Zagreb Kajkavian, a summary of distinctive characteristics. General
Linguistics 31 (1): 26-32.
Brozovic, Dalibor
1966 Ο razgovomom jeziku u Dalmaciji [On the colloquial language in Dal-
matian Slobodna Dalmacija 6767: 7.
1998a Aktualna kolebanja hrvatske jeziöne norme u slavenskome i europskome
svjetlu [Current vacillations in the Croatian language norm in Slavic and
European light], Jezik 45 (5): 161-176.
1998b Uz Skaricev prilog ο standardome hrvatskom refleksu staroga dugog jata
[In connection with Skaric's contribution about the standard Croatian re-
flex of the old long jat], Jezik 46 (2): 62-66.
1999 Ο suvremenoj morfoloäkoj normi hrvatskoga jeziönog standarda i ο
morfoloSkim znaöajkama standardne novoätokavStine uopce. [On
contemporary morphological norm of the Croatian Standard and
about the morphological characteristics of Neo-Stokovian in general.]
In: Marko Samardzija (ed.), Norme i normiranje hrvatskoga standard-
noga jezika, 151-181. [Norms and normativity in the Croatian stan-

dard language]. Zagreb: Matica Hrvatska.
Fishman, Joshua A.
1975 Sociolinguistics: A brief introduction. Rowley MA: Newbury House.
Haugen, Einar
* 1966 Dialect, language, nation. In: J. Pride and J. Holmes (eds.), Sociolin-
guistics, 97-111. Penguin.
Ivie, Pavle
1971 Srpski narod i njegov jezik [The Serbian people and their language].
Beograd: Sipska knjizevna zadruga.
Jagi6, Vatroslav
1948 Izabrani kraii spisi [Selected shorter writings]. Zagreb: Yugoslav Acad-
emy of Sciences and Arts.
Jutronic-Tihomirovic, Dunja
1989 Medudijalektsko sporazumijevanje i utjecaj standarda na dijalekte [In-
terdialectal communication and the influence of the standard on dialects].
NaSe teme 33, 4: 701-702.
Kalogjera, Damir
* 1985 Attitudes towards Serbo-Croatian language varieties. International
1991 / 1992 Rudimentary bilingualism (A report on work in progress). Studia Ro-
manic et Anglica Zagrabiensia 36/37: 63—75.
Katiöic, Radoslav
1984 Novijezikoslovni ogledi [New linguistic essays], Zagreb: Skolskaknjiga.
Kloss, Heinz
1967 "Abstand" languages and "Ausbau" languages. Anthropological Lin-
guistics 9 (7): 29-41.
Lasic, Stanko
1989 Mali predah ili razgovor ο "dijalektalnoj poeziji" i "izvornom i ose-
bujnom jeziku hrvatskog naroda" [A short break or a talk about "dialectal
poetry" and "the original and specific language of the Croatian people"].
Republika XLV, 7 / 8: 55-113.
Le Page R. B. and Andree Tabouret-Keller

* 1985 Acts of identity. Cambridge: Cambridge University Press.
156 Damir Kalogjera
Magner, Thomas F.
1966 A Zagreb Kajkavian Dialect. University Park PA: Pensylvannia State
University Press.
Malkiel, Yakov
* 1984 Revisionist dialectology and mainstream linguistics. Language in So-
ciety 13: 29-66.
Nazor, Vladimir
1942 Na vrhu pera ijezika [On top of the pen and the tongue]. Zagreb: Hrvats-
ki bibliografski zavod.
Sojat, Antun, Vida Barac-Grum, Ivan Kalinski, Mijo Lonäaric and Vesna Zeievic
1998 Zagrebaöki KAJ: Govor grada i prigradskih naselja [The Zagreb Kaj:
The speech of the city and neighbouring settlements]. Zagreb: Institut za
hrvatski jezik i jezikoslovlje.
Tadic, Marko
2000 Hrvatski raöunalni korpusi [Croatian computer corpora]. Vijenac 8
(158): 26.
Tezak, Stjepko and Dragutin Brigljevic
1977 NaS jezik. Zagreb: Skolska knjiga.
Trudgill, Peter
* 1995 Dialect and dialects in New Europe. The European English Messenger
TV/ 1:44-46.
Vince, Zlatko
1978 Putovima hrvatslcoga knjizevnog jezika [On the progress of the Croa-
tian literary language]. Zagreb: Liber.
1. What is the distinction between language and dialect?

2. What is linguistic normativism? Give examples.
3. How is the changing symbolic status of the Croatian dialect groups related
to the ideological trends in Croatian philology?
4. How is the process of language standardization related to ideological and
social change?
5. What are the positive and negative aspects of language standardization?
Ethnolects - between bilingualism and urban dialect
Wolfgang Wölck
1. Review
During the last decade the definition and status of linguistic varieties has be-
come a focus of interest and discussion, often inspired by the "Ebonics" con-
troversy (Odlin 1999; Wölck 1999). It not only questioned the appropriateness
of the old labels "standard, dialect and language"1, but added some justification
to the introduction of a new label for an emerging linguistic variety, viz. "eth-
nolect". The following is a discussion of some of its uses and of its usefulness
and general relevance to research in contact linguistics and bilingualism.
The term "ethnolect" was coined in the late 1970's in the course of urban
dialect research in a mid-size North American immigrant city to designate "the
English of the descendants of immigrant families long after their original lan-
guage is lost", which then still showed clear traces of their home languages
(Wölck and Carlock 1981: 17). The time of the first introduction of the label
into American sociolinguistics and bilingual research is significant, because the
phenomenon it described had not yet manifested itself for a very long time be-
fore. The study describes how language contact features which survived in the
speech of the residents of several ethnic neighborhoods were used by the city's
population to identify each other's origin and showed how these features could
be isolated. These and comparable instances of the transformation of residual
features from immigrant languages into ethnic urban dialects have been in-
cluded in a textbook on bilingualism (Appel and Muysken 1987:132-133). The
reference to immigrant minority varieties of the host or majority language has
remained the main application of the term, most recently in an excellent dis-
cussion of (im)migrant types of German, particularly that of the many Turks in
German cities (Androutsopoulos 2000). Androutsopoulos adheres to the defi-
nition of an ethnolect as "a variety of the majority language (or host language),
which constitutes a vernacular for speakers of a particular ethnic descent and is
marked by certain contact phenomena". Similarly, Clyne (2000: 86) describes
ethnolects as "varieties of a language that mark speakers as members of ethnic
groups who originally used another language (...)". A large number of appli-
cations of the term in scholarly presentations and publications appears in the
description of Jewish varieties of "host" languages as, e.g., of German (Matras
158 Wolfgang Wölck
1998), of Arabic (Hary 1992), or of Dutch (Hinskens and Jacobs 1997). In the
official report to the Czech government on the situation of the Romani commu-
nity in the Czech Republic, the traces of the Roma varieties in their host lan-
guage are referred to as the "Romani ethnolects" of Czech (Dobal etal. 1998).
In the meantime, and with increasing popularity during the nineties, this usage
of the term has become fairly well established, even in less scholarly appli-
cations as, e.g., in the University of Virginia's advertisement of their summer
courses for teachers of children "whose first language is English, but who speak
with a regional dialect or ethnolect" (www.virginia.edu/cla/flyer98).
In all the above examples, the principal conditions to which the label was
applied were the following: 1. There was a recent history of community bilin-
gualism (no less than third generation immigrants); 2. the communities were
relatively well defined, stable and contiguous; 3. communication among the
community membership was close and frequent; 4. ethnolectal features were
recognized, some stereotyped and stigmatized by outsiders, others (subcon-
sciously) still used as social indices.
2. A typical case2
As two of the world's foremost - and by now oldest - immigrant countries, the
United States and Canada would be good candidates for the development of
ethnolects. Almost 99% of their population came as colonists or immigrants;
their major immigration waves occurred around and after the turn of the (last)
century, i.e. about a century or four to five (statistical generations) ago; and
their major languages were imported in a relatively uniform shape, i.e. without
the strong dialect divisions typical of original linguistic landscapes in pre- or
non-colonial societies as in, e.g., Europe or Asia (Wölck 1999). American "dia-
lects" are still primarily local accents with some notable lexical peculiarities,
but with very little syntactic and morphological variation. There are only very
few popularly recognized "accents", among them Southern, "hill-billy", Black
and New York. The first two might well be properly regional - although the
Southern accent could be traced back to African "immigrant" varieties - the
third is definitely an ethnic variety, and the last is the best known urban dialect
in the best known American city.
Buffalo is the "queen city" of New York State, not as big a New York City, but
with a very similar immigration history. Its main European contributors came
from Germany, Poland and Italy or, rather, Sicily. As in many typical immigrant
cities of the American North-East, the largest numbers came in the early part of
the 20th century, when Buffalo doubled its population from 267,567 in 1892 to
Ethnolects - between bilingualism and urban dialect 159
506,775 in 1920 (see Appendix). As late as the 1960 U.S. Census, only about
15% of the city's residents had been in the U.S. for more than a generation, and
the three European groups were still quite concentrated in ethnic neighbor-
hoods, reaching 40% of the total population in some census tracts of the Stan-
dard Metropolitan Statistical Area (SMSA, see Appendix). In the mid-seven-
ties, when we began to record samples of ethnic speech, this concentration had
already gone down to no more than 10% in some areas, especially for the oldest
settlement group, the German Americans. The identifiability of ethnic origin by
speech, however, was still very high, reaching 95% accuracy for some raters. It
was the highest for Polish English, which was also the most stigmatized, and
lowest for speakers of German descent, who were then still trying hard - and
successfully - to cover up their unpopular origin. At the very top of the diag-
nosticity scale of indexical variables were the prosodic markers of the rhythm of
speech (Carlock 1979). Italian English had an iambic rhythm, as in: "my young-
est däughter älways has a cold (...)"; Polish English sounded anapaestic:
"(...) and I wish they would find a cure for the cold (...)"; Germans maintained
the "suspended final contour", characteristic of their ancestral tongue in state-
ments, which made them sound like questions to outsiders and was reminiscent
of the often ridiculed speech of Californian "valley girls" (Delattre 1965; Bald
1976). Discourse construction was another tell-tale of ethnic origin. A Polish
mother relating an ordinary winter morning's activities would say: "(...) get up
early, let the dog out first, then check for school closings (...)"', quite distinct
from an Italian mother, who would use the corresponding gerundial forms:
"(...) getting up at six, starting breakfast, waking the kids (...)". Note that none
of the above instances violate the norms of Standard (spoken) English. We had
recorded most of our early samples from young to middle-aged middle-class
women. The feature which typifies Buffalo for outsiders is the stereotyped and
often stigmatized "flat A", exemplified in the shibboleth exclamation "fantas-
ticV\ where the second 'a' is even more "flattened" than the first. The phonot-
actics or "inherent" variation of this feature and other phonological character-
istics of Buffalo ethnolects are detailed in Wölck (1984,1989). More interesting
for this discussion is the social variation or "ethnic grading" of the flattening
which distinguished the three studied ethnolects from one another and from
non-ethnolectal "Anglo" speech. Italians had the "flattest A's", Polish the next
strongest, and Germans, with quite a distance, much less "flatness", and Anglos
had least of it. Flattening can be divided into four stages, with increasing stig-
matization: the first stage fronts and raises 'standard' [ae] to [e], a sound only ob-
served in recent English immigrants from Shropshire, who actually start from
Northern British [a]. In the second phase [e] diphthongizes to [ee], a stage still
quite acceptable in Buffalo English (see below); the third stage nasalizes the
160 Wolfgang Wölck
diphthong to [ee] and gives it a definitely ethnolectal flavor; and in the fourth and
final stage it is raised again to a nasalized [ίε], which is most definitely stigma-
tized as ethnic, especially in older uneducated female Italians. While ethnicity is
the prime social grading factor of "flattening", age, gender and education play
important roles, although their hierarchical weighting has not been established.
The two extremes, however, are clear: young educated male "Anglo's" have the
least, old uneducated female Italians the strongest "flattening".3
3. Language change and shift
3.1. 'Chain shifting'
All non-native speakers of English have a problem making the fine distinction
between the vowels in mess and mass, guess and gas, and other similar numer-
ous minimal pairs in English. I do not know of any other language which has to
use that many distinctive features to mark such a basic distinction with such a
high functional load. There is certainly no such distinctions in any of the Euro-
pean languages whose speakers came to Buffalo. The way they adopted and
adapted to that distinction deserves attention: they "reinterpreted" (Weinreich
1953: 18) the distinction by widening it, i.e., by raising, fronting and diphthon-
gizing [ae] to [ee] and lowering [ε] to [A] or even [A]. This is why the affirm-
ative yes now sounds like yus in Buffalo, sex like sucks, eleven and television
like eluvin and tulevision. It has been nick-named the "Buffalo vowel shuffle"
by Pagliuca and Mowrey (1979) and is diagrammed below:
(e)
ae(y)
Κ a
Figure 1. The "Buffalo Vowel Shuffle"
This type of linguistic change fits Labov's description of a process he calls

"chain shifting" (1995: 113) and which is exemplified in the "northern cities
vowel shift" (Labov 1991). Perhaps the narrow realization of the first element
of the diphthong in pine and right, which sound like pain and rate (to outsiders)
is also connected to this ethnolectal feature. Even though it is characteristic of
the whole North-East, and may be mostly regional, its extreme degree of nar-
rowing could well be due to ethnolectal enhancement. Devoicing of final
voiced stops characterized especially the English of Poles and to a lesser de-
gree of Germans, whose original language neutralizes this contrast regularly.
3.2. From ethnolect to urban dialect
Urban American ethnolects are not stable varieties. Just as the ancestral lan-
guage tends to be lost within three or four generations after emigration (Gonzo
and Saltarelli 1983), its ethnolectal representation appears not to remain in the
same form for more than a couple of generations. At least in Buffalo it has
taken no more than twenty years for the weakening and diffusion of some eth-
nolectal features to occur throughout the whole urban area (cf. Tasman's Guide
to Buffalo English). Most noticeable is the destigmatization and general spread
of the "flat A" throughout Buffalo. It has become quite accepted as its linguistic
trade-mark. Less popularly obvious, though quite clear and noticeable, is the
corresponding lowering and centralizing of stressed ("strussed") [ε], as in rest
(=rust) or television, or the devoicing of final stops (have = half, his = hiss,
hers = hearse, cold = colt, and = ant, ha[f]to, woul[t]n't, coul[t]n't)4, or the
generalization of the question-like statement intonation (s.a.) from the Ger-
mans across the entire Buffalo community. Another feature of current Buffalo
grammar which originated in German and Polish ethnolects, is the use of the
incomplete "perfect" tense to express the completed past, as in "I've lived in
New York five years; now my home is Buffalo".
Former bilingualism and its transformation into ethnolects is the immediate
predecessor and major source of the rise of new American English urban dia-
lects. Most of the features mentioned here could not be explained or mapped ac-
cording to the methods of traditional dialectology and linguistic geography. The
fact that cities as far distant as Buffalo, Detroit and New York5 share the 'flat A'
feature is a consequence of their similar ethnic composition and cannot be part
of the ontogenous development of English. Contact linguistics and contact dia-
lectology can better account for such "accommodation" (Trudgill 1986).
The major social correlates of the merger and amalgamation of the ethno-
lects into a mainstream urban variety are the large size of the immigrant popu-
lation, the fact that the two largest groups (Germans and Poles) shared many
ancestral linguistic features and, probably most importantly, that the old ethnic
neighborhoods are fast dissolving. The social advancement of the descendants
162 Wolfgang Wölck
of the European immigrants and their integration into the middle classes af-
fords them a move to new suburban neighborhoods and leads to the disinte-
gration of the old community (Wölck 1989).
4. Phases of contact and diffusion
The following is an attempt to give a brief outline of the stages of linguistic de-
velopment in order to show the place of ethnolects in the process. It is specified
for Western New York, but may ceteris paribus possibly be extended to other
parts of the Americas as, e.g., to Hispanic South America, where the role of
European languages vis-a-vis Spanish in urban centers has been comparable to
their impact on English in North America. The impact of (Native) American
languages on Spanish, however, has been quite different (Muysken 1979).
0. Colonial phase: Native American host languages (Iroquoian) in con-

tact with colonial English.
1. Immigrant phase: Monolingual dialects - North Central, Great Lakes,
Hudson valley varieties - of English (the host lan-
guage) in dominating contact with dialects of immi-
grant languages (rural German, Polish, Sicilian)
2. Multilingual phase: Monolingual dialects of English in contact with ghet-
toized, stigmatized and discriminated bilingual dia-
lects of English in contact with bilingual dialects of
immigrant languages (Pennsylvania "Dutch", Polo-
nian, Italiese (Danesi 1983-1984)
3. Ethnolectal phase: Monolingual monogenetic English in contact with
monolingual polygenetic stereotyped diagnostic
English ethnolects
4. Integrative phase: Monolingual urban English, diffused, destigmatized,
amalgamated
[... and, for a renewed cycle of immigration, back to phase 1 or 2: Monolingual

(urban dialects of) English in contact with new immigrant languages (Puerto-
Rican Spanish, Indo-Chinese) or creolized English]
Note that the time-frame is an important factor in the development of eth-

nolects: in our sample case, as in most (Eastern) American cities, bilingual dia-
lects of English arose through the contact with immigrant languages around the
turn of the (last) century. It took two to three generations for the immigrant
groups to lose their home language and leave their impact in the form of mono-
lingual ethnolects of English - and another generation to amalgamate them into
one urban dialect typical of the entire (host) community. The impact of such
ethnolects on urban linguistic varieties is discussed further in Kallmeyer
(1996).
5. Sociolinguistic diagnostics
We always began the investigation of the linguistic behavior of any community

with the careful preparation of its ethnographic profile ("community profile"),
describing its social structure, membership and boundaries (Labrie and Van-
dermeeren 1996). The fine distinctions among ethnolects, the differences be-
tween them and other (monogenetic) varieties, and the surprising local indexi-
cality of relatively spurious features has forced us to develop special methods
for their detection and description. Inspired by Labov (1966) we developed a
special technique of "subjective reaction" testing that goes far beyond tradi-
tional attitude studies and attitude measurement to determine the actual lin-
guistic carriers of social distinction. Samples of spontaneous speech are re-
corded from members of ethnic communities. The utterances have to be in
spontaneous discourse in order to contain markers on all levels of linguistic
structure. They are played to juries for identification of speakers' ethnic origin.
The most successful jurors are asked to report their criteria for determining the
speaker's ethnicity; they are also invited to produce an imitation of the ethno-
lect, in order to show the stereotypes popular in the wider community. Since
many jurors in their report simply repeat stereotypical features of the ethnolect,
which often did not occur in the sample, samples are replayed with a request
for the juror to stop the tape as soon as (s)he is reasonably sure of the speaker's
origin. It is likely that a highly diagnostic feature has occurred or been repeated
just before the interruption. Through jurors' reports and detailed linguistic
analysis of the tapes a list of features is established. The actual diagnosticity of
features and their rank order is checked through "masking" the recordings,
which consists of deleting or distorting the suspected "diagnostic" variable to
see whether the speaker's origin is still determinable, and how fast (some stat-
istical results are given in Carlock and Wölck 1981).
164 Wolfgang Wölck
6. Ethnolectal situations
As mentioned in the introductory review most authors have used the label "eth-
nolect" for situations characterized by the criteria listed at the end of that sec-
tion, i.e., as a linguistic variety of a majority language whose special structure
has developed through a history of community bilingualism. In our examples
from Buffalo, its speakers were all functionally monolingual and most of their
diagnostic features were not stigmatized. Often the speakers are still bilingual,
as in the case of "Türkendeutsch" (Androutsopoulos 2000) and of relatively
low social status, and the ethnolect is not only an ethnic marker but stigmatizes
its speakers. This would also be the case with Ebonics, a definite candidate for
that label, now that it is in the process of decreolization (Odlin 1999; Wolck
1999). Many of its speakers are, however, still diglossic and are likely to re-
main so in the forseeable future. Diglossia is a special case of bilingualism,
characterized in its original Fergusonian (1959) definition by a close (genetic)
relationship between the varieties in contact. None of our Buffalonians were
diglossic; although it is definitely possible that ethnolects could be used for
'styling', i.e., deliberate posturing or demonstrating ethnic identity, as is often
done by African Americans. Our use of the term also implies a relatively short
lifespan of the variety. It is questionable whether much is gained by applying
the term to varieties like Kashubian (Rokoszowa 1997: 1594), which German
scholars have traditionally treated as a separate language, while Polish scholars
consider it a dialect of Polish. The sociopolitical neutrality of the term has
probably also motivated its application to Macedonian (Herson Finn 1996). Bi-
lingual varieties of immigrant languages have also been labeled as "ethno-
lects", as in the case of the Italian spoken in our neighboring city of Toronto,
Canada (Danesi 1983-1984). By the same token the Pennslyvania German of
the Amish could be called an ethnolect; a German ethnolect of German, an Ita-
lian ethnolect of Italian? Then it might even be justifiable to call American or
Australian English ethnolects of English; and the application to Hiberno-Eng-
lish would make even more sense (O Riagäin 1997: 1105).
Those wider definitions of the term apparently focus only on the fact that
the community, to whose speech it is applied, shares some ethnic or national
characteristic or origin. Adhering to a narrower restriction of its applications
would make its usage clearer and more specific. In our sample case above, eth-
nolects are not only differentiated from dialects of immigrant languages but
also from the English(es) of still bilingual immigrants, whose ethnic accent is
not only stronger but still contains many more interference and transfer fea-
tures and is perceived and rated as a "foreign accent" by the host community.
Host language monolingualism would, therefore, be an added criterion in the

narrow definition. In American immigrant cities, at least, a relatively short life
span of the variety - in the real-time life of a speech community - might be the
last and narrowest of the criteria. Such narrowing of the defining characteristics
is, however, necessary to set our variety apart from 'bilingual dialects' of the
immigrant languages of actual bilinguals and from their bilingual English, as
well as from monogenetic regional dialects. Only this way can we justify the
creation of a new term for a really unique phenomenon. Otherwise it should be
shaved off with "Occam's Razor". 6
Appendix
Table 1. Population statistics for Buffalo, NY: (Source: U.S. Census of Population)
18S0 1875 1892 1920 1950 1990

Total 211,305 269,114 267,567 506,775 580,132 328,123
German 6,803 27,018 38,470 20,898 7,775 20,723*
Italian 35 166 2,502 16,411 14,696 25,531*
Polish 175 1,032 13,053 31,406 15,300 31,671*
[* single ancestry; drastic changes are highlighted]

166 Wolfgang Wölck
City limit*
Polish \\\
Italian /i>y
German ill
Puerto Rican III
Black Ξ
C
VJ " limit*
Map 1. Ethnic Divisions, Buffalo SMS A 1970

Notes
1. For an earlier attempt at definitions of linguistic varieties, cf. Wölck 1978.

2. This and the following chapter are revised versions of a presentation to the Ameri-
can Association for Applied Linguistics in Vancouver on March 12, 2000.
3. Sample tapes of recordings can be made available to readers (wwolck@acsu.
buffalo.edu).
4. Note that this phenomenon also occurs word-internally at morpheme boundaries.
5. For New York City, cf. Labov (1966: 31); information for Detroit from Dr. I. C. Cat-
ford (personal communication).
6. 'Lat.: 'Entia non sint multiplicanda praeter necessitatem.'
References
Androutsopoulos, Jannis
* 2000 From the streets to the screens and back again: on the mediated diffu-
sion of ethnolectal patterns in contemporary German. Paper presented
to the International Conference on Language Variation in Europe,
Barcelona, June 29, 2000. [On the web at: rzuser.uni-heidelberg.de/
~iandrout/papers/iclavedraft.html.]
Appel, Rene and Pieter Muysken
1987 Language contact and bilingualism. London/New York: Edward Ar-
nold.
Bald, Wolf Dietrich
1976 Contrastive studies in English and German intonation. Papers and
Studies in Contrastive Linguistics 4: 37-47.
Carlock, Elizabeth
1979 Prosodic analysis of two varieties of Buffalo speech. In: W. Wölck and
P. Garvin (eds.), The Fifth LACUS Forum, 377-382. Columbia: Horn-
beam Press.
Carlock, Elizabeth and Wolfgang Wölck
* 1981 A method for isolating diagnostic linguistic variables: the Buffalo eth-
nolects experiment. In: D. Sankoff and H. Cedergren (eds.), Variation
Omnibus, 17-24. (Current Inquiry into Language, Linguistics and
Human Communication 40.) Edmonton, Alberta: Linguistic Re-
search, Inc.
Clyne, Michael
2000 Lingua franca and ethnolects in Europe and beyond. Sociolinguisti-
ca 14: 83-89.
168 Wolfgang Wölck
Danesi, Marcel
1983-1984 Canadian Italian as a marker of ethnicity, NEMLA Italian Studies 7-8:
99-105.
Delattre, Pierre
1965 Some characteristics of German intonation for the expression of con-
tinuation and finality. Phonetic a 13: 134-161.
Ferguson, Charles A.
1959 Diglossia. Word 15: 325-340.
Goebl, Hans, Peter H. Neide, Zden6k Stary and Wolfgang Wölck (eds.)
1996-1997 Contact Linguistics - An International Handbook of Contemporary
Research. Berlin/New York: Walter de Gruyter.
Gonzo, Susan and Mario Saltarelli
1983 Pidginization and linguistic change in emigrant languages. In: R. An-
dersen fed.), Pidginization and creolization as language acquisition,
181-197. Rowley, MA.: Newbury.
Hary, Benjamin
1992 Multi-glossia in Judeo-Arabic. Leiden: E. J. Brill.
Herson Finn, Victoria S.
1996 What is Nas? Toward a theory of ethnolect in the South Slavic dialect
continuum. Dissertation, Ohio State University, Columbus OH.
Hinskens, Frans and Neil Jacobs
1997 Jewish Dutch: historical and sociolinguistic perspectives on an extinct
ethnolect. Paper presented on Oct. 26, 1997 at NWAV 26, University
Laval, Quebec.
Hinskens, Frans
1998 18e eeuws 'liturgisch' Negerhollands: hoe creools zijn de Deense en
Duitse lecten? [18th century liturgical negro Dutch: how creole are the
Danish and German lects?]. In: A. Bruyn and J. Arends (eds.), Men-
gelwerk voor Muysken [Collected works for Muysken], 201-210. Am-
sterdam: University of Amsterdam.
Kallmeyer, Werner
1996 Plurilinguisme dans les agglomerations urbaines. In: Hans Goebl,
Peter H. Neide, Zdengk Stary and Wolfgang Wölck (eds.), Contact
Linguistics - An International Handbook of Contemporary Research,
450-458. Berlin/New York: Walter de Gruyter.
Labov, William
1966 The Social Stratification of English in New York City. Washington
D.C.: Center for Applied Linguistics.
1991 The three dialects of English. In: P. Eckert (ed.), New Ways of Analyz-
ing Sound Change, 1-44. New York: Academic Press.
1995 Principles ofLinguistic Change: Internal Factors. Oxford: Blackwell.
Labrie, Normand and Sonja Vandermeeren
* 1996 L'analyse du profil de la communaute. In: Hans Goebl, Peter H. Neide,
ZdenSk Stary and Wolfgang Wölck (eds.), Contact Linguistics - An
International Handbook of Contemporary Research, 764-770. Berlin/
Matras, Yaron
1999 Lecture on February 16, 1999 at the Center for Jewish Studies, Uni-
versity of Manchester.
Muysken, Pieter
1979 La mezcla de quechua y castellano. El caso de la 'media lengua' en el
Ecuador [The mixture of Quechua and Spanish. The case of 'half-
language' in Ecuador]. Lexis 3: 41-56.
Odlin, Terence
* 1999 An army, a navy and Ebonics. In: J. Niemi et al. (eds.), Language Con-
tact, Variation and Change, 195-233. (Studies in Languages 32.)
Joensuu, Finland: University of Joensuu. [On the web at www.cohums.
ohio-state.edu/english/people/odlin. 1/papers/ebonjoyl.htm.]
Ο Riagäin, Pädraig
1997 Ireland. In: Goebl et al. (eds.), Contact Linguistics - An International
Handbook of Contemporary Research, 1097-1106. Berlin/New York:
Walter de Gruyter.
Pagliuca, William and Richard Mowrey
1979 The Buffalo vowel shuffle. Presentation to the New York State Coun-
cil of Linguists at Cornell University, Ithaca, NY.
Rokoszowa, Jolanta
1997 Poland. In: Goebl et al. (eds.), Contact Linguistics - An International
Handbook of Contemporary Research, 1583-1594. Berlin/New York:
Walter de Gruyter.
Tasman, David
1999 Guide to Buffalo English. On the web at www.verinet.com/~tasman/
h_guide_to_buffalo_english.html and at www.geocities.com/~libma
ry/bfloslang.htm.
Trudgill, Peter
1986 Dialects in contact. Oxford, Blackwell.
170 Wolfgang Wölck
Weinreich, Uriel
1974 Languages in contact: Findings and problems. The Hague: Mouton de
Gruyter. First published New York: Linguistic Circle of New York
[1953].
Wölck, Wolfgang
* 1978 Towards a classificatory matrix for linguistic varieties: the dialect / lan-
guage problem revisited. In: M. Paradis (td.), Aspects ofbilingualism,
211-219. Columbia SC: Hornbeam Press.
* 1984 Sounds of a city: types and characteristics of the speech of Buffalo and
its ethnic groups. New York Folklore 10: 7-22.
1989 The linguistic resolution of urban ethnic conflict. Plurilingua 7:21-29.
1999 Natives, immigrants and language conflict in the New World. Pluri-
lingua 21: 185-199.
1. What, if any, is the justification for introducing the new label "ethnolect"
into the catalogue of sociolinguistic varieties?
2. What is the difference between the narrow and the wider definition of "eth-
nolect"; which is preferable and why?
3. What role did/do ethnolects play in the development of American dialects?
4. How are ethnolects different from bilingual and/or regional dialects?
5. Which are/were some of the structural characteristics of Buffalo ethno-
lects, and which survived into the general English of the city?
The development of Navajo-English bilingualism
Bernard Spolsky
1. Introduction
Writing only a decade or so ago, Baetens Beardsmore (1988: 579) was sur-
prised to find that the study of language in urban contexts was "only inciden-
tally the focus of attention". He believed, and his work has amply demon-
strated, that all the interesting patterns and permutations could be found in the
complexity of sociolinguistic factors affecting urban conglomerations. There
has in fact been a growing interest in the city. The move from concern with the
pristine simplicity of rural speech forms to the complex intertwined patterns of
city variation marked the essential transition from dialectology, handmaiden of
historical linguistics, to sociolinguistics. Paralleling the recognition of the
world trend towards urbanization and the growth of huge metropolitan areas in
underdeveloped as well as developed countries, sociolinguists now regularly
tackle the study of cities as "not necessarily an urban melting pot in which code
and variety differences get smoothed into a unique urban vernacular," but as
areas marked by the continuing social and geographical coexistence of a host
of varieties. Studies like Baetens Beardsmore's own, or Labov's (1966) pion-
eering investigation of New York City, or of Norwich by Trudgill (1974), have
shown the continued existence of language variation in the city reflecting the
complexity of religious, demographic, social, political, cultural, and edu-
cational patterns and pressures. Truly, the city has started to become a favored
site for the study of language and variety contact and of bilingualism.
A number of studies have looked at the city as the focus of immigration
from rural areas. Blanc (1964) showed that the three principal communal dia-
lects of Baghdad reflected not just the religious makeup of the population, but
also the fact that most of the Muslim residents were comparatively recent ar-
rivals to the urban setting, attracted by the general economic pull of a town al-
ready populated by Christians and Jews. Fishman, Cooper and Ma (1971) in
their study of the Puerto Rican barrio of Jersey City showed how an immi-
grant community was able to maintain its social identity though transported to
a new environment.
172 Bernard Spolsky
The modern city is in fact turning out to be too powerful for its increasingly
blurred boundaries to shut in its influence and sociolinguistic effects. With im-
provements in roads, mass transportation, and the communication media, areas
adjacent to the city are undergoing a process of semi-urbanization that deserves
careful sociolinguistic scrutiny. For example, in recent studies of Palestinian
villages in Israel, Amara and Spolsky (1996) and Amara (1999) have traced the
sociolinguistic reflexes of several competing pressures, including the pen-
etration of Hebrew into the local speech repertoire partly as a result of the vil-
lages starting to function as suburbs for nearby Greater Tel Aviv.
A related pattern appears to help explain the sociolinguistic transformation
of the Navajo Nation. Fifty years ago, the Navajo were characteristically mono-
lingual in their own language, with a small but highly influential group of
people whose Navajo-English bilingualism gave them an important role as
brokers for the majority culture. Nowadays, bilingualism is the rule among
adult Navajos, and younger people are monolingual in English, albeit a dis-
tinctly Navajo variety of English. A significant proportion of the explanation
for this major language shift has been the effect of urbanization and suburban-
ization, not just the fact that many Navajo now live or work in the towns bor-
dering the Reservation, but also in the changing demographic patterns of the
Reservation itself.
In this paper, I wish to explore the process of Navajo language loss es-
pecially over the last quarter century, attempting to identify the contribution of
the various factors involved. There is, I must acknowledge at this stage, a per-
sonal interest. Twenty-five years ago, I published a paper (Spolsky 1975) in
which I assessed the prospects for the survival of the Navajo language. That
paper, based on two surveys of six-year-old Navajo children beginning school,
revealed increasing knowledge of English. It recorded, for some parts of the
Reservation and especially for the border towns, the first reports of Navajo
children whose knowledge of Navajo was considered to be nil or poor. In the
years that followed, observers noted continued language attrition. Whereas in
1970 about 90% of Navajo children who came to boarding schools met English
there for the first time, by 1990 the situation seems to have reversed, with six-
year-old children beginning Head Start said to have very little if any knowledge
of the language of their people. Platero (1992) and Holm (1996) agreed that
only about half of Navajo children enter kindergarten speaking any Navajo at
all, and maybe a sixth are considered by their teachers to be reasonably com-
petent five-year-old speakers of Navajo. What contribution, the paper will ask,
did urbanization (or urban influence) make in the process?
The development of Navajo-English bilingualism 173
2. Navajo country - background1
The Navajo Nation occupies an area of about 25,000 square miles in the South-
western United States; portions are in Arizona, Colorado, New Mexico, and
Utah. The 1990 United States Census recorded 219,198 Navajos, 148,993 of
them living on the Reservation. Population density in 1993 was estimated at six
people per square mile. Whereas once most Navajo subsisted through sheep,
livestock, and farming, now half depend on wages, a quarter receive general as-
sistance, and 15% depend on Social Security. Unemployment is high, and over
half the population is below the poverty level. Most Navajo still live in small
houses forming small family camps and separated from others by miles. Few
homes have plumbing; water usually needs to be hauled. More recently, small
clusters of 30 or so houses with electricity and running water are starting to be
built, typically not far from a school or other government institutions. Many
Navajo live in the rapidly growing small towns on the Reservation or in off-
Reservation towns. About a quarter of the population has completed high
school, but only 2% have college degrees.
The changing demographic pattern has been summarized by Lee and
McLaughlin (2000):
Only the most isolated and remote communities remain unconnected to the reserva-
tion highway system. Telephones, cell phones, electricity, cable and satellite televi-
sion, and solar-powered twelve-volt electricity systems that support radio, televi-
sion, VCRs, the internet, and the like are all widely available. Trading posts, which
used to serve as information hubs in the communicative economies of reservation
communities, have been replaced by 7-11 convenience stores owned by off-reserva-
tion, non-Navajo, English dominant interests. Navajos' intermarriage to non-Nava-
jos has increased dramatically. Simply put, the Navajo Nation is no long isolated
from the rest of the world.
These changes have also been associated with the serious loss by younger Na-
vajo of knowledge and use of their language.
3. Current state of the language
Young Navajo children do not speak Navajo, but their knowledge of English is
far from adequate. Phonology is the most obvious feature showing that they are
speaking a Navajo variety of English, with the result that when they do start to
learn Navajo they have very little problem with pronunciation. At the same
174 Bernard Spolsky
time, the Navajo Nation remains widely Navajo speaking. Navajo is still re-
ported to be the main language spoken between adults in the family context,
while Navajo and English appear to be used equally when speaking to young
people and outside the home (Lee and McLaughlin 2000). While local chapter2
meetings continue to be conducted mainly in Navajo, tribal governance (coun-
cil and committee meetings, judiciary proceedings, and regular staff work) is
increasingly bilingual or in English. The biggest change has been the loss of
natural intergenerational transmission of Navajo.
Lee and McLaughlin (2000) has been the latest to document these effects by
gathering data from 200 Navajo adolescents from five reservation schools.
Adults at home still speak mainly Navajo and some English, but when speaking
to the adolescents or outside the home, Navajo and English are more or less
equal. Young Navajo themselves also use both languages about the same. The
adolescents sensed an increase in English around them, but they themselves
think that they speak more Navajo as they grow older. Half of them take part in
traditional or Native American Church activities, where they report hearing
mainly Navajo. But they also reported hearing more English in these contexts
than when they were young.
Parsons-Yazzie (1995) lists the six main reasons parents give for not speak-
ing Navajo with their children:
1. They commonly use English anyway.
2. Children spend their time in settings where English is more common and
has higher prestige.
3. The children are ambivalent about or ashamed of Navajo.
4. It helps children at school if their parents speak English with them.
5. Knowing Navajo is not important
6. English is associated with academic success. The only Navajo-speaking
adults at school are in more menial occupations.
There is, Lee and McLaughlin conclude, cause for alarm about the prospects
for the survival of the language. How did this come about?
4. Early contact with English
The anthropologist Sapir, who was one of the first to make an extensive study
of Navajo language and oral traditions, remarked how little the language had
been influenced by several hundred years of contact with Spanish. English was
to have a much greater effect, in large measure as the result of a determined
Government policy. The Indian Bureau opened the first boarding school for
Navajo children at Fort Defiance in 1882, with the goal of integrating the vari-
ous Native American tribes into a homogenous group with a single language
(English) and open to US control (McCarty 1998: 31). Discipline was severe,
and there was Navajo opposition to boarding schools culminating in violence at
Round Rock when in 1892 attempts were made to carry children off to school
(Aberle 1966: 39).
Many stories bear witness to the traumas of those experiences. Among the
texts collected by Sapir (1942) and his student Harry Hoijer in 1929 is an ac-
count by one Navajo of his early schooling and his satisfaction at learning Eng-
lish:
Twenty years have passed since I came among [the Americans]. When I was eight
years old, I came among them at the place called Fort Defiance in order to learn the
American language. It is a very difficult language.
It is not pleasant for one who first comes (to school), for boys are very mean to
me (...) When I became bigger, it came to be that they did not treat me so. And then
when I had learned the American language, I was pleased. So now I can talk to you
by means of it. At Fort Defiance I spent nine years going to school. And then I went
also to Albuquerque. There I spent three years. At that point I was through with
school. (Sapir 1942: 436-437; told and translated by John Watchman)
The narrator worked for three years as a blacksmith, and then returned to the
hooghan and busied himself with sheep. Sheep, it seems, constitute a fitting
conclusion to a Navajo biography: Old Mexican finishes up the account of his
life that he gave to Dyk (1947) with the statement "Well, I haven't anything else
to tell. Nothing more that is interesting. All the rest is sheep" (Dyk 1947: 168).
Bennett (1964) who was born in the mid-1920s talks about the frightening
visit of the superintendent from the boarding school at Toadlena who came to
take the girls to school. In a conversation reminiscent of Russian Jews afraid of
their sons' conscription into the Czarist army, the mother anguishes over which
child should be sacrificed to the law. McCarty (2002) has collected stories
about the impact of the early boarding schools, built in Fort Apache, Fort Win-
gate, Phoenix, Albuquerque and Santa Fe (all towns on the periphery of the Na-
vajo Reservation) and in other Western States. The military discipline (army
posts turned into schools), the poor and inadequate food, the forced labor, and
most of all, the strict banning of Navajo and enforced speaking of English all
have left an indelible memory of schooling in the early years. Well into the
20th century, punishment for speaking Navajo continued.
176 Bernard Spolsky
School was thus established as the principal weapon of the linguistic colo-
nialism in the government's efforts to control and civilize the Navajo. Attend-
ance at boarding schools interrupted many aspects of traditional Navajo life.
Children who were away at school missed the traditional grandfather's tales
told only in winter. Girls missed the correct timing for the traditional Kinaaldä
puberty ceremonies (Frisbie 1967: 83). They lost thus the opportunities pro-
vided for linguistic enrichment.
Only the lack of universal schooling and the failures of curriculum and
teachers slowed the process of Navajo language attrition and loss. Day schools
built during the Roosevelt administrations allowed some children to return to
Navajo speaking homes at the end of the day, but the scattered traditional resi-
dential patterns and the absence of roads meant that even day schools needed
dormitories, removing children from a Navajo environment.
In the 1930s, a government program of forced stock reduction destroyed the
poor economic base of Navajo life, making the effect of the economic depress-
ion even greater and building lasting enmities within the communities over the
implementation of the stock reduction. Destruction of the local economic base
also greatly encouraged Navajo willingness to seek work off the Reservation.
Some 3,700 Navajos joined the armed services when the war began in 1941;
more than 10,000 went to work in war-related industries in peripheral towns or
in distant States.
During the war, funding for Navajo schools was reduced and many schools
were closed. Only in 1950 was a development program funded to try to pro-
vide education for all Navajo children. By 1960, the number of children in
schools had doubled, reaching six times the number of Navajo children en-
rolled in 1939. The bulk of these children were in boarding schools on
(13,000) or off (6,000) the Reservation. Another large group (about 7,500)
were in the elitist public schools build on the Reservation for children of BIA
employees, traders and missionaries. These schools admitted only the most
highly qualified Navajo children. Other children were in mission schools or
day or trailer schools.
Traditional residential patterns and the existence or quality of roads had an
important influence on the type of school a Navajo child would attend. For cul-
tural and economic reasons, Navajo families generally lived in a hooghan miles
away from their neighbors. As long as most children lived, according the offi-
cial criterion for boarding school enrolment, more than a mile and a half from
the nearest road along which a school bus serving a public school went, board-
ing school was the most common. Essentially, this meant that starting at the age
of 6, Navajo children would leave home for a non-Navajo environment. Physi-
cally, a high wire fence usually surrounded the school compound, enclosing
school and dormitory buildings for the children and a small cluster of separate
houses for the English-speaking teachers who staffed the school. Until the
1970s, there were only a handful of Navajo teachers working in Reservation
schools. The result was English-speaking teachers who had to communicate
with and educate Navajo-speaking pupils, pupils who came from traditional
homes where only Navajo was spoken. The study of six-year-olds in 1969 and
1970 (Spolsky 1970; Spolsky and Holm 1977) captured this pattern. Some
90% of the children attending BIA boarding schools were reported by their
teachers to have known no English before they came to school.
The public schools were however different. The first was established at Fort
Defiance; in the 1950s, others were added at Ganado, Crownpoint, Kayenta,
Shiprock, Tohatchi, Tuba City and Chinle, all semi-urban settlements sur-
rounding Bureau of Indian Affairs Agency offices. Attended by the children of
English-speaking employees and teachers (and so often located alongside a
boarding school to cater for its staff), they admitted only those Navajo children
whose English made them capable of handling an English only curriculum. By
the 1970 survey, most children in these schools were reported to know at least
some English by the time they started school. The children living off the Res-
ervation in towns like Gallup of course knew the most English; some knew
little or no Navajo.
The 1969-1970 studies thus produced evidence of two factors accounting
for the spread of English, both directed related to residential pattern and pro-
ximity (or better, accessibility) to a source of English. Locally, schools were
one such focus: the closer a family lived to a school, the more likely it was for
its children to know some English. Public schools, whose children lived
relatively close by, had much more English than boarding schools (even when
located side-by-side in the same community) whose families lived much
further away. Secondly, the accessibility of a community to an off-Reservation
two correlated significantly with the percentage of bilingual children.3
These analyses gave a vivid and sensitive picture of the penetration of Eng-
lish into the Navajo Nation as a result of access to work and English off the
Reservation and as a result of the development of English-using schools. At the
same time, they presented an over-simplified view of the complex factors lead-
ing to language shift. They showed the changes of access to English, either at
work or shopping or at school, but do not show the attitudinal, motivational or
ideological changes that accounted for a situation where most Navajo parents
seem superficially unconcerned that their children are growing up not speaking
the language, or at least powerless to take action to resist language loss.4
178 Bernard Spolsky
5. Ideology and loyalty
Fishman (1966) was one of the first to analyze the notion of language loyalty,
although latterly he has made his concern more specific by talking about efforts
to reverse language shift (Fishman 1991, 2001). The concept is related to lan-
guage or linguistic ideology (Schieffelin, Woolard and Kroskrity 1998; Silver-
stein 1979), which might be defined as set of beliefs of about how language
should be and how it should be used. In a pioneering typology of language use,
Stewart (1968) proposed four key factors: autonomy, standardization, vitality,
and historicity. Fishman (1970) reinterpreted these as attitudes or beliefs, so
that for him vitality became not the fact of natural intergenerational trans-
mission, but the belief that parents and caretakers should speak to young
children in this language. In all considerations of language maintenance and
loss, this issue becomes the key factor. Our question here becomes not what
were the changing conditions that gave Navajos access to English, but what
were the changing attitudes that led so many Navajos to believe that it was
better for their children to speak English.
From the answers of Navajo adults to questions about this matter, the most
common references are to their own or other people's school experiences, the
way they suffered or were punished for not knowing English and for speaking
Navajo, and the desire to spare their own children this discomfort. This reaction
of parents to the forced imposition of another language in schools, reported
regularly in the cases of peripheral minority languages in Europe or in the cases
of immigrant languages, probably most closely fits the term "linguicide"
coined by Skutnabb-Kangas (2000). While the first generation suffers, it at
least had some chance to acquire the heritage language before or outside
school. When future parents have been successfully co-opted to supporting the
imposed language, the critical feature of natural intergenerational transmission
of the language is fatally compromised.
Reinforcing the negative reinforcement of punishment, there was of course
also the prestige of English. Other references make clear the vastly greater
status of English. In the schools, English was the language of the principal and
the teachers; only the cleaning staff and the cooks and the drivers could speak
Navajo with the children or their parents. English was also the language of the
rapidly intruding outside modern world, of the nearby towns and of the cities
where relatives had sought work, of literacy, and, as it was introduced, of
television and computing. What purpose could be served in resisting its spread,
or even in attempting to maintain Navajo alongside?
6. Literacy
Literacy as a factor in language loss deserves more careful consideration, for it

is associated with the second critical factor in Stewart's and Fishman's list,
standardization (the existence of a standard form and a writing system). The
main effect of schooling for the Navajo has been an acceptance of English lit-
eracy alongside (at least until recently) Navajo oracy (Spolsky and Boomer
1983; Spolsky and Holm 1973). Public oral discourse (Tribal Council and
Chapter meetings, Tribal courts, radios) continues to be largely in Navajo, but
public written activities (the minutes of the meetings, the records of court
sessions, the Tribal newspaper) have always been and still are in English.
This may be illustrated by the case of a Navajo woman elected to serve as
secretary of her local Chapter. Her English (she had been raised partly off the
Reservation and her father was English-speaking) was proficient enough for
her to write down the minutes of the meeting, although she needed help in Na-
vajo from her husband to make sure she hadn't missed anything. But she then
needed to spend hours before the next meeting improving her spoken Navajo
and practicing so that she could "read" (that is, translate into acceptable Nava-
jo) the minutes aloud.
In spite of this general rule, there were and continue to be uses of literacy in
the vernacular. Young (1977) has reported on his work with Morgan and Har-
rington in developing what is fairly considered to be a highly successful ortho-
graphy for Navajo in the late 1930s (W. Holm 1996).5 The new system was im-
mediately adopted for three important functions - for schools (in a limited
way), for government propaganda (urging Navajo to reduce their livestock and
help in the war effort), and by the Protestant missionaries working on a trans-
lation of the Bible. It was not taken over for other official functions, for officials
were either non-Navajo speaking Anglo-Americans or English-literate bilin-
gual Navajo. Nor was it seen to fill any indigenous tasks - for most Navajo, lit-
eracy remained an alien skill for alien functions. 6
At the same time, it must be noted that Navajo literacy also provides a
chance for resistance. McCarty (2001) presents a vignette of Navajos who ac-
quired literacy for religious purposes (see next section), but went on to use it,
when they became bilingual teachers, for indigenous goals. The existence of
an accepted orthography, the high level of the dictionaries and grammars re-
cording the language, and the respectable body of printed material provides a
basis for the continued teaching of the language in schools and at the tertiary
level.
180 Bernard Spolsky
7. Religion
For many languages, religion served as a major force of language maintenance

and spread. The existence of sacred texts in Hebrew, Sanskrit and Arabic
played an unquestioned role in the continued knowledge of these languages. In
the case of Hebrew, it was the religious and related educational system that pre-
served the language for some 1,700 years, providing a strong basis for modern
revitalization (Spolsky 1991a, 1991b). The Arabic of the Qur'an remains a
unifying factor for the many countries to which Islam spread (Wagner 1993).
But, conversely, the religious spread of a language could just as easily work
against indigenous languages, as evidenced by the way Arabic replaced Ara-
maic in the Middle East.
A Navajo may be associated with one or more of the four religious group-
ings - traditional Navajo religion, Native American Church, a variety of Chris-
tianity, and Mormonism. None of the religions is an unalloyed supporter of the
Navajo language.
Traditional Navajo religion, with its carefully prescribed rituals conducted
by highly trained "singers" or "medicine men," its long traditional poems and
ceremonies, remains on the face of it a major force for language preservation.
The rituals and ceremonies are in Navajo, and form the principal body of Na-
vajo verbal tradition (Reichard 1963; Spencer 1957; Wyman 1957). Singers ac-
quire their knowledge by several years of apprenticeship to another singer,
often a paternal kinsman (Aberle 1966: 45). The singers who are guardians of
the tradition express some concern that it might be changed or lost. Frank Mit-
chell, whose conduct of a Kinaaldä (girl's puberty ceremony) in Chinle was
recorded by Frisbie (1967), explained to people present why he thought it was
all right for non-Navajos to record the event:
That is the reason that these people who are non-Navaho are interested. They are re-
cording to preserve these things. You younger people are not sincere in trying to
learn it to carry it on in the future. So it is a good thing white people record it to pre-
serve it for us. You ask why they are here to take pictures, make records, and to pre-
serve the whole ceremony. In the future, someone may be interested in inquiring
about it; then he will be able to see it. (Frisbie 1967: 53)
Ceremonies continue to be conducted in Navajo, but from time to time now pa-
tients respond in English. The influence on language status is not as strong as
might be expected. Because of their sacredness, there is some resistance to
committing the texts to writing. Frisbie (1967: 101) tells about two singers who
were willing to record some of the Kinaaldä songs, but not songs that were part
The development of Navajo-Etiglish bilingualism 181
of the Blessing Way Ceremony which were "too special." Because of this con-
cern to protect sacred things, traditional Navajos are not automatically sup-
porters of the teaching of Navajo language and literacy in schools, institutions
that they tend to distrust. A second negative influence is that the association of
Navajo language and culture with traditional religion leads many Christian Na-
vajos to object to teaching the language in school. Efforts to maintain tradi-
tional Navajo religious practices then do not lead automatically to Navajo lan-
guage preservation.
Many Navajo are Christians. One major result of the efforts of Protestant
missionaries in Polynesia in the 19th century was the rapid indigenization of lit-
eracy in the vernacular (Spolsky, Englebrecht and Ortiz 1984). Among the Na-
vajo, the effects were more complex and less favorable. Austin-Garrison (1996)
include a number of interesting accounts of the way that Catholic and Protes-
tant and Mormon missionaries introduced Navajo to literacy and encouraged in
them an interest in the language. The existence of the Navajo Bible and its use
of the standard orthography were cited as a major reason for not revising the
writing system in the 1970s. Positive as this effect has been, it is often more
than counter-balanced by the strong opposition expressed by some church
members to the teaching of Navajo in school for fear that it might be associated
with the teaching of non-Christian traditional Navajo religious views and be-
liefs. Reichard (1963: 45) points out the fundamental conflict between Navajo
and Christian beliefs about the purpose and end of human life. For the Navajo, it
is a state of non-personal and non-individual perfection in a spiritual union with
nature, which is directly opposed to the Christian view of individual survival
and resurrection as a result of faith or works. As Navajo religious ceremonies
express this concept, it is understandable that most churches are deeply resis-
tant to anything that preserves Navajo religious observance or belief.
Many early Christian missionaries appeared to share the common belief
that Navajos would need to give up their language to be modernized. The
Roman Catholics at St Michael's were a marked exception, developing an or-
thography at the beginning of the 20lh century and publishing the first Navajo
dictionary (Franciscan Fathers 1910). Father Berard Haile, who worked on the
Reservation from 1930 to 1960 and published a good deal of traditional materi-
al, was considered by Navajos to know their language and culture remarkably
well:
Father Berard Haile is a good example. Was he a proselytizing Christian? Who

knows? By the time he was well into studying Navajo language and documenting
Navajo ceremonial life, he may well have had so much wisdom about Navajo ways
that he might have been more Navajo than Anglo. (Austin-Garrison etal. 1996: 356)
182 Bernard Spolsky
Other missionaries however were assumed to use Navajo and literacy only in
order to convert Navajos who would then give up their language and ways of
thinking.
Even mission schools shared the mentality of the times, which led to the prohibition
of the use of the mother tongue and the exaltation of the white man's language. Chris-
tianity and progress were often equated with embracing Anglo-American speech and
dress. This superficial brand of conversion often result in disillusionment and disap-
pointment for the missionary and frustration for the Indian. (Wallis 1968: 37)
Faye Edgerton, whose work Wallis describes, was one of those missionaries
who believed in the importance of learning Navajo and of developing a version
of the Bible in it.7 She started active work as a translator in 1944 and spent the
next ten years, with advice from the Wycliffe linguists and the collaboration of
Navajos and other missionaries, working to complete the translation of the
New Testament. One important decision she made was to switch from the dif-
ficult semi-phonetic orthography developed by the Franciscan Fathers and con-
tinued by Edward Sapir to the new "government" alphabet developed by Ro-
bert Young and William Morgan. This choice, as Holm (1996) points out, was a
critical factor in standardizing Navajo - the agreement of Church and state pro-
duced a powerful partnership in teaching Navajo literacy.
As McLaughlin (1987; Austin-Garrison et al. 1996) has demonstrated,
church-related literacy in Navajo could easily develop into an empowering
tool, permitting Navajos to take even greater control of their modernization.8 In
the decade after the publication of the full translation of the New Testament in
1954, there was evidence of a boom in Navajo literacy. But because of the criti-
cally serious gap between Christianity and traditional Navajo religious beliefs,
the overall effect of the Christian churches was to work against language main-
tenance. The quandary was expressed in the 1970s by Navajo Christians who
said they could accept bilingual but not bicultural education. Regularly, it is
members of Navajo Christian Churches who oppose the teaching of Navajo in
school for fear that it will bring with it the teaching of Navajo religious prac-
tices and beliefs.
The third major religious force, the growing Native American Church of
Navajoland now includes Navajo songs in the all-night peyote ceremonies.
Originally, the songs were in languages that Navajos did not understand, but
the prayers were all in Navajo. The Peyote religion, "the most popular, and one
of the most durable of all the religious movements created by American Indian
groups suffering from the effects of domination by American society," (Aberle
1966: 3), reached the Navajos in the 1930s. By 1950, when Aberle was em-
ployed by the Bureau of Indian Affairs to help them decide whether to accede
to a request from the Navajo Tribal Council to wipe out the cult, it had reached
some 20% of Navajos and continued to grow. As Aberle describes the Church
in the 1950s and 1960s, its appeal came for a number of reasons. The all-night
ceremonies are less rigidly prescribed than the Navajo traditional ceremonials
and the training of the leaders is less demanding. There are differences in ideo-
logy and belief, reflecting what Aberle sees as the depressing social and econ-
omic effects of white domination and especially of the stock reduction cam-
paigns. Prayer, it appears, is more spontaneous and may be in English.
It is the language issue that Aberle (1991) now cites as one of the main rea-
sons for the appeal of the Native American Church.
Navajos today want to maintain their identity. Religion is one way of doing so. Al-
though traditional religion is an obvious way to tie oneself to a Navajo or Indian
identity, it faces two problems: a decline in the number of chanters, which will re-
duce the number of chants and the availability of ceremonies; and the loss of the Na-
vajo language among the young. Transmitting Navajo religion without the language
in which its ideas are embodied will be difficult. Today peyotists refer to themselves
as traditionalists and respecters of Navajo ceremonies. They see an alliance between
Peyotism and Navajo ceremonies, contrasting them with Christianity viewed as
alien. (Aberle 1991: vii)
Lacking a rigid language ideology, the Native American Church has been able
to adapt itself and take advantage of the shift to English, and so encourage it.
Another important religion is Mormonism. As a large part of the Navajo
Reservation spreads into the State of Utah, contact between the Mormons and
the Navajos has been continual. Some of this involved missionary activity,
possible even before 1978 when Church doctrine was amended to permit per-
sons with dark skin into the priesthood. Mormon missionary activity among the
Navajos took a new direction in the 1950s with a plan to have Navajo children
adopted as foster children by Mormon families. This, it was hoped, would rem-
edy the fact that many earlier conversions had turned out to be superficial and
impermanent. The linguistic effect of this policy was of course to encourage
shifting to English.9 As a whole, then, Mormonism too is likely to have been a
force for the shift to English.
Overall, then, the general effect of religion has been to weaken ideological
support for Navajo and strengthen the status of English.
184 Bernard Spolsky
8. Political organization
Often, language revival and resistance activities have been integral parts of
nationalist attempts to gain political independence for an ethnic group. Irish
and Hebrew at the beginning of the 20th century, Quebec, Catalonia and the
Basque region, Maori in New Zealand all involved simultaneous political and
linguistic action. It is worth considering how this influences the Navajo situ-
ation.
There is no clear consensus about the nature of Navajo political organiz-
ation before white contact. Young (1978) considers that the institution of Head-
man (naat'άαηη), a local man (or woman) "of high prestige, eloquent and a
natural leader capable of governing by persuasion" (25) has always existed.
There is however dispute about the nature and size of the naachid, a regular (or
occasional) gathering of local (or tribal) headmen for religious or political or
war purposes. The Spanish at the end of the 17lh century attempted to appoint a
single Navajo governor, following the model they imposed successfully on the
Pueblo tribes, but this had little effect. Spaniards and, later, Americans were
mistaken in their belief in the coercive powers of the headmen with whom they
tried to make agreements or treaties.
Young (1967) sees the Navajo attitude to coercion as basic to understanding
the cultural conflict that arose after contact:
Navajo culture does not have a heritage of coercive religious, political or patriarchal
family figures, and in the Navajo scheme of things one does not usually impose his
will on another animate being to the same extent, and in the same ways as one does
from the English point of view. (Young 1967: 7)
It is, he notes, very difficult in Navajo to express notions like "ought to, must,
duty, responsibility." This clearly had a major effect on the development of Na-
vajo political organization under white pressure:
The Navajo parent is likely to ask a child if it wants to go to school, rather than issue
a mandate to the effect that it must go. By the same token, coercive laws are distaste-
ful from the Navajo point of view, and Tribal leadership has long preferred persua-
sion to force, even in applying 'compulsory' education laws on the Reservation.
As a result, acceptance of the Anglo-American police and court system, based as it is

on the principle of compulsive behavior - the enforcement of coercive laws - has
still not been comfortably accommodated within the Navajo cultural framework,
despite the fact that the Tribal government supports a system of Tribal courts and a
well-equipped police force. (Young 1967: 8)
The imposition on the Navajo of a system of governance suitable to the needs

of the US government on the one hand and the cultural preferences of the Na-
vajo people on the other was a complex matter. The government wanted a local
system that would maintain order and be able to give away mineral and land
rights but not at the same time demand independence.
The Americans did no better than the Spaniards had in setting up a puppet
government. After the military campaign of Kit Carson in 1863, most Navajo
were taken into captivity in Fort Sumner where a plan was developed to divide
them into 12 villages each headed by a chief appointed by the Army. This prin-
ciple, Young says, was the basis for the model proposed by the Special Indian
Agent after the Navajo were permitted to return to Navajo Country. In fact, for
the next decades, there were various patterns of recognition of local chiefs and
ad hoc councils of headmen summoned by the government agents. At these
meeting, Navajo leaders would freely express their opinions, but all power re-
mained with the US government representative, and the tribe appeared from
outside as a community without an organized government. This situation lasted
well into the 20th century.
From the beginning, language became a source of power, as bilingual inter-
preters were needed to convey information between Navajo-speaking chiefs
and council members and English-speaking Indian agents.10 The status and in-
fluence of Chee Dodge, while in his early twenties, appointed in 1883 by the
Indian Agent as Chief of Scouts and a year later as Head Chief of the Navajo
Tribe, was premised on his being "the only man known who can transmit the
speech of an ordinary English speaking person into fluent Navajo" (Young
1978: 44).
At the same time, as Young (1967: 16) notes, interpreters were underpaid
and selected without any formal testing of their abilities. It was assumed any
school child, any janitor or cook could serve as a reliable interpreter between
parents and school authorities or between Navajo patients and Anglo-Ameri-
can doctors. Some of this explains the misconceptions about Navajo knowl-
edge of the human body shown by the Cornell health study and later exposed
by Werner's work on Navajo ethnoanatomy. George Boyce, who joined the In-
dian Service in 1938 and later became director of Navajo schools, reported on
the problems in the stock reduction campaign:
Not one of the government administrators spoke or understood the Navajo language
sufficiently to know what the interpreter may actually have said in attempting to
translate such ideas into Navajo (...) Beatty suspected that the Navajos who were
being used as interpreters were not sufficiently skilled in English to translate accu-
rately. (Boyce 1974: 34)
186 Bernard Spolsky
But, as Neundorf (1987) has shown, the key position that skilled interpreters,
since Chee Dodge, have played in negotiating contacts between government
agencies11 and Navajo political bodies served as a stepping-stone to power for
Navajos who were bilingual. As a result, Navajos who were fluent in Navajo
and English played a key role in communication between Anglo-American
rulers and their Navajo subjects.
The pressure for formalization of Navajo political organization came from
the mining companies who sought in the early 1920s some way to obtain Na-
vajo approval of their exploitation of mineral rights. To handle the legal prob-
lems produced by issues of whether the authority was local or regional, the
Commissioner of Indian Affairs in 1923 drew up regulations establishing a Na-
vajo Tribal Council, a body with which the government could negotiate agree-
ments (Young 1978: 59). Each of the six Superintendencies into which the res-
ervation has been divided was to elect delegates, who in turn were to elect a
permanent chairman and vice-chairman. The Council had neither legislative
nor executive power, but was there to discuss any issues brought before them
by the government. Chee Dodge was chairman of the Tribal Council estab-
lished in 1923, and it had twelve delegates and two interpreters.
A few years later, a further step was taken with the institution of local meet-
ings of what were sometimes called Chapters or Livestock Associations or Far-
ming Associations, with elected officers, that became a forum for discussion of
local issues. While there was no formal connection by the mid-1930s, the
100 community associations had become an important element in tribal organ-
ization.
One task the Bureau had for the Tribal government was to approve the sale
of Navajo mineral and land resources.12 A second task was to take responsibil-
ity for the stock reduction policy starting in the 1930s (Young 1978: 71) and
continuing for thirty years, which, Aberle (1966) says, produced enormous suf-
fering and led to depression and disillusionment. Not surprising considering its
key role as a tool of US policy in these areas, Tribal government remained a
center of disagreement and dispute.
During the 1930s, there was continuing dispute over the constitution of the
Council, and finally, in 1937 a Constitutional Assembly met, appointed itself
Tribal Council, and prepared a draft constitution defining the organization and
powers of a self-governing Tribal authority. Instead, a year later, the Depart-
ment of the Interior drew up rules for an elected Council the powers of which
were not defined. All control of money, including Tribal funds, remained in the
hands of the government. A movement for self-determination developed again
after the war, when, with the help of an appointed Tribal Attorney, the Council
began to gain greater control over Tribal funds.
In all this long process, a key sociolinguistic effect was the growing role of
English. As noted the first chairman of the Council was selected for his bilin-
gualism. The council in its early years could only meet at the call of the Com-
missioner of Indian Affairs to discuss topics he proposed and in the presence of
Federal officials for whom all debate needed to be translated into English
(Young 1978: 69). From early on, there was preference for council members
who had gone to school; of the twelve original delegates selected for the 1933
Council, only two had not gone to school. While council members were then
generally bilingual, it was for a long time accepted that debates be conducted in
Navajo and translated into English for the Federal officials.13
By the late 1980s, however, there had been a change and the Tribal Council
now practices regular code shifting. Neundorf (1987: 163) notes that "so many
Councilmen are used to code switching informally, and it takes some effort to
stay entirely within either English or Navajo for extended excursions". The
session I attended in 2000 showed a pattern of switching. Formal parliamentary
statements were made in English ("Referring to section 5B of the bill tabled
yesterday"), but many speeches would then move into culturally sensitive and
rhetorically Navajo presentations, with full and appropriate kinship references.
Only bilinguals can follow this kind of debate; those who do not speak Navajo
(like the Tribal Legal Advisor) must depend on the simultaneous interpretation
now provided through headphones. Neundorf concluded: "Today in Navajo so-
ciety about everyone is bilingual, but still the power belongs to those who are
fully coordinate bilingually and biculturally" (1987: 183).
9. Changes in a limited industry community
Lisa Ann Lane (1999, 2000) has been studying identity and sociolinguistics
changes in maritime communities undergoing a drastic change as a result of
major alterations in their single industry, fishing, because of global influences
setting quotas for environmental protection. These communities typically con-
fined to serving a single economy have evolved their own local patterns of life
and identities. The changes in economy have, she has shown, had marked ef-
fects in changes of identity and local loyalty. The Navajo Nation is certainly not
maritime,14 but looking over its economic history, it can clearly have been con-
sidered at one stage to be an economically limited industry community, as
sheep and shepherding provided the major source of wealth and led to a pattern
of living that by the early 20th century characterized social and economic or-
ganization.
188 Bernard Spolsky
Realizing this, it is not at all far-fetched to consider the stock reduction plan
of the middle of the 20th century as a central factor in the collapse of Navajo
traditional life and a strong contributor to the loss of Navajo identity and lan-
guage. The arguments put forward by the Bureau of Indian Affairs for stock re-
duction, like the arguments put forward by those setting fishing quotas, were
based on the need to maintain environmental stability. The strict imposition of
these principles on a people with limited ability to resist applies as well to the
case of Navajo sheepherders as it does to the Danish fishermen that Lane
studied.
The effect of the changes of economy with the reduction of sheepherding
was demographically important. Once children were no longer needed for
looking after stock, they could be allowed to go to school. Once a family no
longer needed to live as close as possible to the range where its stock grazed,
they could move to the growing semi-urban settlements at the agency head-
quarters, or to a town off the reservation where they could be employed, or even
to one of the small settlements being built along the roads.
The loss of Navajo can then be traced to a misguided government economic
policy - by attacking the economic basis of Navajo life, the policy of stock re-
duction led to major demographic changes, with a shift from scattered resi-
dence patterns to a semi-urbanization that supported and encouraged shift from
Navajo monolingualism to Navajo-English bilingualism as a first stage, and to
Navajo English monolingualism as the next.
10. The media
In the section on literacy, it has already been noted that while there has been
some vernacular literacy, the key pressure from literacy has been towards Eng-
lish. Another communication medium, the radio, for a long time provided a
significant counter-force. For many years, the small FM radio stations that sur-
rounded and dotted the Reservation broadcast almost entirely in Navajo, an-
nouncing the Country Western music that was their staple, and presenting all
news and advertisements in Navajo. Only in recent years has a tendency been
noticed to present certain announcements and news (most recently, basketball
results) in English, in recognition of the growing number of young Navajos
who do not easily understand Navajo. Television is of course almost entirely in
English, apart from an hour or two of local broadcasting of Tribal Council
sessions. The development first of repeater stations and then of satellite televi-
sion on the Reservation made a significant change in accessibility. Before that,
access to English was by roads that were often hard to travel, but television
connected remote Navajo homes to the global village, or rather city.
11. School language policy and practices
For many years, the only formal language policy step that the Tribal Council
had taken was a decision to spell Navajo with a "j" and not an "h." There have
been decisions however about school language policy. While the Tribal Council
of the Navajo Nation is on record as favoring the maintenance of Navajo-Eng-
lish bilingualism through the use and teaching of both languages in schools, im-
plementation of this policy is constrained by the fact that the Tribal Division of
Education has no authority over the curricula or teaching practices of the vari-
ous school systems. Boarding schools continue to be under the control of the
Federal government though the Bureau of Indian Affairs. Public schools come
under the aegis of local district school boards that are answerable to four dif-
ferent State educational systems (Arizona, New Mexico, Colorado and Utah),
each with its own set of regulations. Charter and community schools also have
their own boards. All schools are strongly influenced by regional and national
accreditation agencies and by pressure to meet state and national standards.
The Division of Education does have direct (if somewhat filtered) authority
over the Head Start programs that it administers directly under a contract with
the Federal Government. In these programs, spread throughout the Navajo Res-
ervation, it is able the carry out the Navajo immersion curriculum required in
an Executive Order issued by the Navajo Nation President (W. Holm 1996:
404).15 In these programs, all instruction (during a school day limited by the
pressure of long school bus routes and the detailed schedules requiring atten-
tion to providing food and health care for the children) is given in Navajo.
These Navajo curricula now assume that children start with little or no knowl-
edge of the Navajo language. The curriculum starts by teaching basic vocabu-
lary and the grammatical structures needed for the routine of the Head Start
program and for the content that is to be covered.
Schooling and education have played a major role in the destruction of
tradition and language among the Navajo. Bureau of Indian Affairs and local
State-controlled public schools have, with the exception of the brief period in
the 1970s when bilingual education had some Federal support, been concerned
to switch Navajo children to monolingual speakers of English. Even the inde-
pendent Navajo schools have failed to correct the trend. One thinks naturally of
the major force that schooling was in the revitalization of Hebrew and Irish and
190 Bernard Spolsky
Welsh, and of its growing role in the revival of Maori in New Zealand and Yid-
dish among Hasidim in Israel. In each of these cases, while the mainstream
schools continued to work towards the official language, these community-
controlled independent schools had an ideological commitment to the mainten-
ance of their ethnic language.
Thirty years ago, the first schools on the Reservation came under the control
of the local communities, usually by contracting with the Bureau of Indian Af-
fairs. Now, Lee and McLaughlin (2001) note, some 10% of schools have this
status. In some of them, there was an early commitment to Navajo language,
and both Rock Point and Rough Rock schools, the pioneers of community con-
trol, developed a program where the teaching of Navajo literacy preceded Eng-
lish and where Navajo served as language of instruction at elementary and sec-
ondary levels (W. Holm and A. Holm 1990; Rosier and Holm 1980). In spite of
the evidence of their academic (as well as Navajo) success, the model did not
spread (A. Holm and W. Holm 1995). Recent observation of these schools
makes clear that the Navajo programs in them are declining and weakening,
unable to stand up to the pressures of accreditation and funding and external
curriculum demands.
Lee and McLaughlin (2001) sum up the results. Only 10% of Navajo pupils
receive any Navajo language courses, almost always presented as supplemental
programs (in Arizona, schools may teach Navajo to fill the foreign language
requirement). Teaching at the high schools is even weaker, with isolated teach-
ers receiving little support for their work. Schooling, then, the original insti-
gator of English, has given little if any support to activities intending to reverse
language shift.
12. Prospects for the survival of the Navajo language
Looking at the current situation, it seems that my paper a quarter of a century

ago may well have been over-optimistic about the prospects for the survival of
Navajo. The enthusiasm in the 1970s that accompanied the development of
community schools, the growth of the Navajo Teachers Association and its
transformation into the Tribal Division of Education, the burgeoning of bilin-
gual and Navajo medium education, the major campaign to train one thousand
Navajos as teachers, all seem to have faded in the face of the relentless pressure
of changing economic and demographic circumstances. The destruction of
traditional life and the semi-urbanization have had their sociolinguistic reflex-
es. In spite of the presence of token Navajo political and educational autonomy,
the educational policies of the Federal government and of the four State gov-
ernments have continued to work towards the shift from Navajo to English that
reflects the general tendency in the United States as a whole.
Any remedy will only come, I suspect, as part of a major regeneration of
Navajo identity and power. To an outside observer, the contrast with what has
happened to the Maori language in New Zealand over the same 25 years is
striking. Twenty-five years ago, there were a bare handful of Maori children
who knew their own language. Currently, in the wake of the major political, so-
cial and educational revolution, over 20,000 Maori children are receiving bilin-
gual or immersion Maori medium instruction in New Zealand schools, a sig-
nificant number of which are now under direct Maori control. This is not the
place to analyze this further, nor is it appropriate to come up with a list of sug-
gestions for action. The comparison, however, makes clear the difficulties that
any program to rescue the Navajo language must face.
Having lasted longer than most autochthonous languages in the US, Navajo
too is now endangered. The clash of cultures took a century to have its effect,
the final blow being the move from rural isolation to semi-urban and urban den-
sity of communication. The city has had its damaging effect.
Notes
1. This section is largely based on Lee and McLaughlin (2001).

2. "Chapter" is the term used for local community associations. See below.
3. The 1969-70 survey was crude in many ways. The data were collected from school
teachers, few of whom knew Navajo. They were asked to classify their six-year-old
children as they were at the time they began school by their proficiency in Navajo
and English, each on a simple three-point scale (knew it well, knew some, knew
none). We checked the reliability of the judgments by asking a number of teachers to
fill out the same form a couple of months later (Spolsky 1970) and the validity by
sending judges to test a selection of classrooms (Spolsky, Murphy, Holm and Ferrel
1975). The scores for schools were averaged. We next estimated the accessibility of
the school to the nearest town off the Reservation by multiplying road mileage by a
factor representing road quality and condition. The correlation between English
score and accessibility was significant. For one school, we calculated the accessibil-
ity of each child's home, correlating this with his or her English score. Again the
correlation was significant.
4. McCarthy (personal communication) notes that parents express concern, but feel
helpless to reverse the shift. Perhaps this relates to the Navajo notion of authority
discussed later in this paper.
192 Bernard Spolsky
5. Before that, Sapir and Haile had created systems for scientific linguistic use.
6. An exception proving (that is to say, testing) the rule was one isolated community
studied by McLaughlin (1987) where literacy introduced by the church and school
was spread and used for other purposes as a result of developing local control and
community empowerment. Similarly, in a number of school related literacy pro-
grams in the 1970s and 1980s, many individuals developed literacy, forming a tiny
cadre of Navajo writers. But as a general fact, it is reasonable to say that literacy has
been a force for English and language shift.
7. She came to the Navajo Reservation in 1924 as a Presbyterian missionary. Portions
of the Bible had been translated earlier, but there was no continuing translation work
under way when in the 1940s, she discovered the Wycliffe Bible Translators, a
group of missionary-linguists who conducted the Summer Institute of Linguistics
and were involved in Bible translation all over the world.
8. In Tonga, vernacular literacy combined with indigenization of the churches (Spols-
ky etal. 1984) played an important part in preserving the language.
9. George P. Lee was a Navajo in the adoption program. In 1975, he was called to the
First Quorum of Seventy, an exceptionally high rank in the Mormon Church. Lee
was excommunicated for heresy in 1989.
10. Ability to speak with government officials was commonly treated as evidence of
power; thus, as Dozier (1966) reports, the chief of the Tewa-Hopi bilingual village
Hano was mistakenly assumed by President Theodore Roosevelt to be the senior
rather than the most despised of the Hopi village chiefs who came to visit him in
1906.
11. Robert Young is the only Anglo-American I know who during his years as a Bureau
of Indian Affairs official was able to speak, read and write Navajo at the level the
State Department would classify as 5 (the highest) for a diplomat. I suspect that
there were very few who would reach the level of 3, defined as minimal proficiency
for professional activity. It is interesting to note that Young, the world-renowned ex-
pert on Navajo language and culture, ended his bureaucratic career assigned to
Pueblo Indian affairs.
12. Whose interests were generally served is suggested by a recent court decision
(No. 93-763L February 4, 2000 in the US Court of Federal Claims) as reported in
The Arizona Republic on February 8,2000. It found that "Reagan administration In-
terior Secretary Donald Hödel 'violated basic fiduciary duties' to the Navajo Nation
when he killed a plan to boost royalty payments from Peabody Coal Co."
13. Robert Young (personal communication) told me that Council members used to
complain loudly if any speaker mixed the two languages.
14. Traditionally, fish was not considered appropriate food for Navajos, and its con-
sumption was one of the major changes attributable to boarding schools.
15. Working at the other end of the educational ladder, the Tribe also requires students
receiving a Navajo Tribal scholarship to include some Navajo courses in their pro-
grams.
References
Aberle, David F.
1966 The Peyote religion among the Navaho. New York: Wenner-Gren
Foundation for Anthropological Research Inc.
1991 The Peyote religion among the Navaho. (Second edition). Norman,
Oklahoma: University of Oklahoma Press.
Amara, Muhammad Hasan
1999 Politics and sociolinguistic reflexes: Palestinian border villages. Am-
sterdam/Philadelphia: John Benjamins Publishing Company.
Amara, Muhammad Hasan and Bernard Spolsky
1996 The construction of identity in a divided Palestinian village: socio-
linguistic evidence. In: Yasir Suleiman (ed.), Language and identity in
the Middle East and North Africa, 81—100. Richmond, Surrey: Cur-
zon.
Austin-Garrison, Martha, Bemice Casaus, Daniel McLaughlin and Clay Slate
* 1996 Dine Bizaad Yissohigii: the past, present and future of Navajo literacy.
In: Eloise Jelinek, Sally Midgette, Keren Rice and Leslie Saxon (eds.),
Athabaskan Language Studies: Essays in honor of Robert W. Young,
349-390. Albuquerque, NM: University of New Mexico Press.
Baetens Beardsmore, Hugo and Piet Van de Craen
* 1988 Research on City Language. In: Ulrich Ammon, Norbert Dittmar and
Klaus J. Mattheir (eds.), Sociolinguistics: An International Handbook
of the Science of Language and Society (Volume 1), 579-584. Berlin/
Bennett, Kay
1964 Kaibah: Recollection of a Navajo girlhood. Los Angeles: Westemlore
Press.
Blanc, Haim
1964 Communal dialects in Baghdad. Cambridge, MA: Harvard University
Press.
Boyce, George A.
1974 When Navajos had too many sheep: the 1940s. San Francisco: The In-
dian Historian Press.
Dozier, Edward P.
1966 Hano, a Tewa Indian community in Arizona. New York: Holt, Rinehart
and Winston.
Dyk, Walter
1947 A Navaho autobiography. New York: The Viking Fund.
194 Bernard Spolsky
Fishman, Joshua A.
1966 Language loyalty in the United States: The Maintenance and Perpetu-
ation of Non-English Mother Tongues by American Ethnic and Relig-
ious Groups. The Hague: Mouton.
1970 Sociolinguistics: A Brief Introduction. Rowley: Newbury House.
Fishman, Joshua A. (ed.)

2001 Can threatened languages be saved? Reversing language shift, Re-
visited: A 21s'century perspective. Clevedon, Avon: Multilingual
Matters Ltd.
Fishman, Joshua Α., Robert L. Cooper and Roxana Ma

1971 Bilingualism in the barrio. Bloomington, IN: Research Center for the
Language Sciences, Indiana University.
Franciscan Fathers
1910 Ethnologic dictionary of the Navaho language. St. Michaels, Arizona.
Frisbie, Charlotte Johnson

1967 Kinaaldä: a study of the Navaho girl's puberty ceremony. Middle-
town, Connecticut: Wesleyan University Press.
Holm, Agnes and Wayne Holm

* 1995 Navajo language education: retrospect and prospects. Bilingual Re-
search Journal 19: 141—167. .
Holm, Wayne
* 1996 On the role of "YounganMorgan" in the development of Navajo liter-
acy. In: Eloise Jelinek, Sally Midgette, Keren Rice and Leslie Saxon
(eds.), Athabaskan Language Studies: Essays in honor of Robert
W. Young, 391-406. Albuquerque, NM: University of New Mexico
Press.
Holm, Wayne and Agnes Holm

1990 Rock Point: A Navajo way to go to school: A valediction. Annals,
AASSP 508: 170-184.
Labov, William
1966 The social stratification of English in New York City. Washington,
D.C.: Center for Applied Linguistics.
Lane, Lisa-Ann
1999 Levels of contact: linguistic and social change in a Jutlandic fishing
community. Paper presented at the Languages in Contact Conference,
Groningen.
The development ofNavajo-English bilingualism 195
2000 Glocalization and the reconstruction of geosocial identities in local

linguistic communities. Paper presented at the Sociolinguistics Sym-
posium 2000, Bristol.
Lee, Tiffany
1999 Sources of influence over Navajo adolescent language attitudes and
behavior. Unpublished Ph.D. dissertation, Stanford University, Stan-
ford CA.
Lee, Tiffany and Daniel McLaughlin
*2001 Reversing Navajo language shift, revisited. In: Joshua A. Fishman
(ed.), Can threatened languages be saved? Reversing language shift,
Revisited: A 21s' century perspective. Clevedon, Avon: Multilingual
Matters Ltd.
McCarty, Teresa L.
1998 Schooling, resistance and American Indian Languages. International
* 2002 A Place to be Navajo: the Struggle for Self-Determination in Indigen-
ous Schooling. Mahwah, NJ: Lawrence Erlbaum Associates.
McLaughlin, Daniel Jay
1987 When literacy empowers: an ethnography of English and Navajo lit-
eracy. Unpublished Ph.D. dissertation, University of New Mexico,
Albuquerque.
Neundorf, Alice
1987 Bilingualism: a bridge to power for interpreters and leaders in the Na-
vajo Tribal Council. Unpublished Ph.D. dissertation, University of
New Mexico, Albuquerque.
Parsons-Yazzie, Evangeline
1995 A study of reasons for Navajo language attrition as perceived by Na-
vajo-speaking parents. Unpublished Ph.D. dissertation, Northern Ari-
zona University, Flagstaff AZ.
Platero, Paul
* 1992 Navajo Head Start Language Study. Window Rock: Navajo Nation:
Division of Dine Education.
Reichard, Gladys
1963 Navaho religion: a study of symbolism. (Second edition.) New York:
Pantheon Books.
Rosier, Paul and Wayne Holm
1980 The Rock Point experience: a longitudinal study of a Navajo school
program. Washington, D.C.: Center for Applied Linguistics.
196 Bernard Spolsky
Sapir, Edward
1942 Navaho texts. Iowa City, Iowa: Linguistic Society of America.
Schieffelin, Bambi B., Kathryn A. Woolard and Paul V. Kroskrity (eds.)

1998 Language ideologies: practice and theory. New York/Oxford: Oxford
University Press.
Silverstein, Michael
1979 Language structure and linguistic ideology. In: Paul R. Clyne (ed.),
The elements: a parasession on linguistic units and levels, 193-247.
Chicago: Chicago Linguistic Society.
Skutnabb-Kangas, Tove
2000 Linguistic genocide in education, or worldwide diversity and human
rights? Mahwah, NJ: L. Erlbaum Associates.
Spencer, Katherine
1957 Mythology and values: an analysis of Navaho Chantway Myths. Phil-
adelphia: American Folklore Society.
Spolsky, Bernard
1970 Navajo language maintenance: six-year-olds in 1969. Language
Sciences: 19-24.
* 1975 Prospects for the survival of the Navajo language. In: M. Dale Kin-
kade, Kenneth Hale and Oswald Werner (eds.), Linguistics and An-
thropology, in honor ofC. F. Voegelin, 597-606. Lisse: The Peter de
Ridder Press.
1991a Control and democratization of sacred literacy. In: Samuel Rodin
(ed.), Encounters with Judaism: Jewish studies in a non-Jewish world,
37-53. Hamilton: Waikato University and Colcom Press.
1991 b Hebrew language revitalization within a general theory of second lan-
guage learning. In: Robert L. Cooper and Bernard Spolsky (eds.), The
influence of language on culture and thought: essays in honor of
Joshua A. Fishman's sixty-fifth birthday, 137-155. Berlin/New York:
Mouton de Gruyter.
Spolsky, Bernard and Lorraine Boomer

1983 The modernization of Navajo. In: Juan Cobarrubias and Joshua
A. Fishman (eds.), Progress in language planning: international per-
spectives, 235-252. The Hague: Mouton.
Spolsky, Bernard, Guillermina Englebrecht and Leroy Ortiz

1984 Political, religious and educational factors in the development of bi-
literacy in the Kingdom of Tonga. Journal of Multilingual and Multi-
cultural Development 4: 459-470.
The development ofNavajo-English bilingualism 197
Spolsky, Bernard and Wayne Holm

1973 Literacy in the vernacular: the case of Navajo. In: Ralph W. Jr. Ewton
and Jacob Ornstein (eds.), Studies in language and linguistics,
1972-1973, 239-251. El Paso: University of Texas at El Paso Press.
*1977 Bilingualism in the six-year-old Navajo child. In: William Mackey
and Theodore Andersson (eds.), Bilingualism in early childhood,
167-173. Rowley, MA: Newbury House Publishers.
Spolsky, Bernard, Paul Murphy, Wayne Holm and Allen Ferrel
1975 Three functional tests of oral proficiency. In: Bernard Spolsky and
Leslie Palmer (eds.), Papers in language testing, 1967-1974, 75-90.
Washington, D.C.: TESOL.
Stewart, William
1968 A sociolinguistic typology for describing national multilingualism.
In: Joshua A. Fishman (ed.), Readings in the sociology of language,
531-545. The Hague: Mouton.
Trudgill, Peter
1974 The social differentiation of English in Norwich. Cambridge: Cam-
bridge University Press.
Wagner, Daniel A.
1993 Literacy, culture and development: Becoming literate in Morocco.
Cambridge/New York: Cambridge University Press.
Wallis, Ethel Emily
1968 God speaks Navajo. New York: Harper & Row Publishers.
Wyman, Lyland C. (ed.)
1957 Beautyway: a Navaho ceremonial. New York: Pantheon Books.
Young, Robert W.
1967 English as a second language for Navajos: an overview of certain cul-
tural and linguistic factors. Window Rock, AZ: Navajo Area Office,
Division of Education.
* 1977 A history of written Navajo. In: Joshua A. Fishman (ed.), Advances in
the Creation and Revision of Writing Systems. (Contributions to the
Sociology of Language 8.) The Hague: Mouton.
* 1978 A political history of the Navajo Tribe. Tsaile, Navajo Nation: Navajo
Community College Press.
198 Bernard Spolsky
1. By charting over time, and according to locations, functions, and demo-

graphic groups, the penetration of English into the Navajo sociolinguistic
repertoire, trace the change from monolingualism to bilingualism of the Na-
vajo Nation. What do you learn from your charts about the "cracks" through
which English entered?
2. Can you derive any conclusions about the stability of the current situation?
3. Does any one factor appear to be the major cause of the change from Navajo
monolingualism to Navajo- English bilingualism to incipient English
monolingualism?
4. What activities to reverse language shift appear to have the best prospects to
help maintain Navajo?
5. From other cases you know, how typical is the Navajo situation? What fea-
tures appear special?
Language ideology, ownership and maintenance: the
discourse of the Academia Mayor de la Lengua
Quechua
Tim Marr
1. Introduction
Language maintenance efforts are sometimes thought of as being in essence

either a matter of top-down state planning or the result of "grass roots" ini-
tiatives by speakers themselves (see Hornberger 1999 for a discussion of this
for Latin America). The present paper considers a case which appears to fall
into neither camp. It concerns the ongoing Peruvian project of status and cor-
pus planning being carried out by a language academy, the Academia Mayor de
la Lengua Quechua (based in the former Inca capital of Cusco), which displays
an ambivalent attitude towards both the nation-state which it claims to serve
and the speakers whose aspirations it claims to represent. It will be argued that,
given the peculiar linguistic, social and political agenda espoused by many of
the senior members of the organisation, a serious Quechua language mainten-
ance project on the part of either state planners or grass roots activists would be
likely to find in this body at best an unreliable ally, and at worst an implacable
enemy.
The paper examines the perspectives and ideology of the Academia Mayor
de la Lengua Quechua (henceforth, "the Academia"), drawing in part on pub-
lications by and about it and its members, and in part on interviews conducted
in 2001 (by a proxy interviewer) with the current president of the Academia,
DrLeandro Herencia Fernandez, and with a teacher of Quechua who works for
the institution. The principal source of data to be analysed, though, is an hour-
long interview with one of its senior officials, a professionally-trained aca-
demic, carried out in Cusco in 1996 as part of my research on Quechua lan-
guage shift (see Marr 1998). For reasons of confidentiality this individual is not
named here (and is generally referred to as "the academician"). However, it
should be noted, first, that he was interviewed in his capacity as a senior,
elected office-holder of the Academia rather than as a private individual; and
second, that he agreed enthusiastically to the interview being recorded on tape.
The opinions I solicited from him were expressly those of the Academia as an
200 Tim Marr
institution. It is also perhaps worth noting that the person interviewed has been
a leading member of the Academia for many years, and by dint of his seniority
and authority, might be supposed to exert some influence within the institution;
certainly his views seem to accord on many questions with those of the current
president.
2. Language and legitimate authority
Few academic linguists, in Peru or abroad, would regard the Academia as a

serious scholarly institution; in such circles its position tends to be at best mar-
ginal (see Cerron-Palomino 1997). The fact remains, though, that it is is recog-
nised by the Peruvian government as the major representative body for the
Quechua language; Article 3 of its constitution (reproduced in AMLQ 1995)
lists as one of its six fundamental tasks the preparation of an approved Quechua
version of the Constitution of the Republic, and the President of the Republic is
automatically elected as an honorary senior member of the institution (miem-
bro honorario protector). As will be seen, though, the Academia's attitude to
the Lima-centred Peruvian state is rather more complex and conflictive than
this formal relationship might suggest. As holder of the state franchise, as it
were, for Quechua, the Academia sees itself as the defender and protector of
the language, and sometimes, by extension, as the voice of the "authentic" lan-
guage as opposed to the linguists and educationalists it regards as outsiders or
"foräneos" (Itier 1992b).
The Academia in its present state was instituted by congressional decree in
1990 out of the former Academia Peruana de la Lengua Quechua. It has,
though, existed in various forms for very much longer; its immediate intellec-
tual lineage might be traced back to the politico-cultural indigenista movement
which flourished in Peru (and particularly in the Cusco region) in the early part
of the 20th century. (For a critical account of this historical background, see
Nino-Murcia 1997). Much of the ideological discourse of the Academia,
though, lies squarely within a tradition that has existed in Cusco since at least
the 17th century, whereby local social and political elites have sought to portray
themselves as the legitimate heirs of the Incas, in part through the appropri-
ation of the supposed language of the Incas (Itier 1992a; Godenzzi 1992; Nino-
Murcia 1997). Traditionally, such groups - composed almost invariably of bi-
lingual mestizos (those of mixed Spanish and indigenous descent) - have
claimed that their own Quechua sociolect preserves the "purity" and "nobility"
of the Inca tongue. Members of the Academia like to refer to the language as
"quechua imperial" (see e.g. Manya 1992); or as qhapaq simi or apu simi (that
The discourse of the Academia Mayor de la Lengua Quechua 201
is, as something like "language of the nobles" or "language of the lords") in an

unambiguous attempt to differentiate it from runa simi - "people's language",
the term by which the language is generally known by runa, the Quechua-
speaking peasants of the Andes.
Along with this insistence on the prestige and legitimacy of the Academia's
"own" variety of the language goes an insistence on the display and recognition
of duly-sanctioned authority, often deployed in order to fend off the perceived
threat from other, competing authorities in the world of Quechua. The current
state of "authority" in the field appears muddled and prone to strife. The emerg-
ence of locally-based Academias in other historically Quechua-speaking areas,
such as the Callejön de Huaylas in the central Andes, tends to be regarded with
a mixture of satisfaction at the resilience of the language and an intense suspi-
cion that these bodies constitute a potential threat to the hegemony of the aca-
demicians of Cusco. A short while prior to the main interview discussed here,
my interviewee informed me, the Lima branch of the organisation (or certain
members of it) had taken to styling themselves "Academia Mayor de la Lengua
Quechua del Peru", hence implicitly disavowing the authority and pre-emi-
nence of the Cusco centre. The president of the Cusco centre was due to travel
to Lima that month to call a meeting of the branch and re-impose the authority
of the Academia - "cuya sede es la ciudad del Qosqo, Peru, cuya sede es la ciu-
dad del Qosqo, Peru (...)" [whose headquarters is in the city of Cusco, Peru,
whose headquarters is in the city of Cusco, Peru] as I was told with marked rep-
etition. (The word Qosqo is the Quechua original of the hispanicised Cusco or
Cuzco. It is to be assumed that its use in a Spanish-speaking context, by a bi-
lingual, inevitably carries ideological weight: here, as elsewhere where it oc-
curs in this paper, it is doubtless intended to convey a sense of the historical
legitimacy of the city where matters of language and culture are concerned).
The problem of the proliferation of competing self-appointed authorities on
Quechua, added to the ideological imperative - to be discussed below - of pro-
moting a particular sociolect, leads the Academia to adopt at times a quite dis-
tinctive mode of discourse, in which for example questions formulated pri-
marily in terms of language are answered in terms of authority to pronounce
upon language, as in the following extract from the interview. The academician
was asked if the Academia's dictionary (AMLQ 1995) attempted to record the
way everyday Quechua is spoken (that is, whether the dictionary is in essence a
descriptive or a prescriptive work). The reply came:
El diccionario se ha elaborado con la participation de los miembros de numero, que

es la maxima categoria de los maestros del idioma quechua. Son personas que estän
ahi 30,40, 50 anos, estudiando, investigando (...)
202 Tim Marr
[The dictionary was put together with the participation of the miembros de
nümero, which is the highest rank of expert in the Quechua language. These are
people who have been studying and researching for 30,40, 50 years (...)]
A lengthy exposition followed on the professional and personal eminence
of the senior members of the Academia. In seeking to understand the import (at
times, it seems, only at the level of a semi-conscious subtext) of what the aca-
demician has to say, it is necessary to appreciate that his linguistic Weltan-
schauung is shaped by a distinct conception of power and authority: all tends
towards the justification of the proposition that he and his fellow academicians
are uniquely able to pronounce upon Quechua, upon Cusco, upon the Andean
region, upon the Republic as a whole, and even beyond. Itier's (1992b) critique
of the institution concludes that its activities are designed ultimately to demon-
strate that "(...) el Cusco, su clase media y sus intelectuales estän legitima-
mente llamados a representar la supuesta cultura andina y, por ende, la nacion"
(Itier 1992b:90). [Cusco, its middle class and its intellectuals are legitimately
called upon to represent the supposed Andean culture and, consequently, the
nation.] That this is a substantially accurate assessment will, it is hoped, be-
come apparent in due course.
3. Cuscocentrismo and the cult of the Incas
As Itier's comment implies, the Academia's institutional attitude to language is

conditioned by a heavily ideologised worldview which is essentially Cusco-
centric (that is, determinedly regionalist), and, within this, class-based. The
academicians' claim to linguistic, cultural and political pre-eminence depends
upon the generalised recognition, firstly, that the Inca past represents the moral
and cultural heart of Peru; and secondly, that control - in both the linguistic and
political senses - of the supposed qhapaq simi affords them an indisputable
claim to the heritage of the Incas. For this section of the mestizo elite, then, it is
axiomatic that Quechua originated in Cusco under the Incas. Thus Juan Anto-
nio Many a, a former president of the Academia, writes: "El hablar del idioma
quechua es hablar del Qosqo, que es ciudad milenaria, area sagrada, cubierta
con el denso velo del misterio; emporio, en epoca fabulosa, de riqueza, ciencia
y poder (...)" (Manya 1992:49). [To speak of the Quechua language is to speak
of Cusco, which is an age-old city, a sacred ark, covered with a thick veil of
mystery; a treasure-house, in that fabled epoch, of wealth, learning and
power (...)] And so on for several paragraphs in the same vein. The identifica-
tion of the language with the city is absolute, as is the identification of the city
with the Incas, and with the heart of the modern Peruvian nation. Much of the
discourse of the Academia relies on this kind of romanticism and mysticism to

make its case. Its president since December 2000, Dr Herencia Fernandez, can
claim that: "El sonido del idioma quechua es como una piedra que toca el agua
del rfo y el sonido va hasta el infinito". [The sound of the Quechua language is
like a stone hitting the water of the river, its sound echoing into infinity.]
Like this equation of the sound of the language with the very sounds of na-
ture, the notion that the history of Cusco is "cubierta con el denso velo del mis-
terio" owes more to wishful thinking - or wilful obfuscation - than reality. The
well-established complex of modern historico-linguistic research showing that
Quechua developed not in Cusco but in the central part of Peru, including its
Pacific coastal section (see e.g. Torero 1974; Rojas 1980; Cerron-Palomino
1989; Mannheim 1991) tends to be ignored or dismissed out of hand by the
Academia, incompatible as it is with the cherished myths of Incaic Cusco.
The determined propagation of the cult of the Incas seems to necessitate a
near-total disregard for any other Peruvian or South American culture. At no
point in the interview with the academician was reference made to any pre-Inca
civilization in Peru: the history of the country is virtually understood to begin
with the Inca empire. Degregori's (1994) account of discussions with a group
of Cusco residents shows how deeply-rooted this mode of thinking is; he notes
the marked tendency of his informants to reduce the whole of Peru's precolum-
bian history to "(...) su ultimo momento de desarrollo, el relacionado con los
Incas" (1994: 448-9) [(...) the last moment in its development, the one associ-
ated with the Incas.] Within this style of discourse, as Degregori rightly notes,
virtually nothing of any significance is attributed to pre-Inca cultures - not
even the cultivation of such ancient Andean crops as maize and tubers. In simi-
lar fashion, the academician's vision of America at the time of the Conquest is
an extraordinarily limited one. The Spaniards, he says: "(...) a nivel del nuevo
mundo, ο sea entre el norte de Argentina, Chile, Peru, Bolivia, Ecuador, etcete-
ra etcetera, han matado a aproximadamente 200 millones de personas". [(...) in
the New World as a whole, that is to say, in northern Argentina, Chile, Peru,
Bolivia, Ecuador, etcetera etcetera, killed approximately 200 million people.]
One might or might not wish to take issue with the figure cited. However, the
most striking point here is surely that this definition of "New World" contrives
to ignore entirely most of the continent, which is dismissed with an "etcetera
etcetera": the whole of America is presented as being basically synonymous
with the limits of the Inca empire. The same selective vision of culture is pro-
jected into the present. If the true culture of Peru is self-evidently an Andean
Quechua one, and Quechua-speaking (or, rather, bilingual) Cusco self-evi-
dently the very essence of it, then any Peruvian who is less than fervently in-
terested in Quechua is simply suffering from cultural dislocation. Indeed, so
204 Tim Marr
strong is the belief that the Inca culture defines Peru that the Academia - ac-
cording to the teacher interviewed - is now demanding that the national Min-
istry of Culture be moved from Lima to Cusco. This is clearly not going to
happen; but the evident lack of sustained governmental support for the Aca-
demia's cultural and linguistic project is rationalised thus by the academician:
"Y el Estado pues no ha (...) no ha brindado su apoyo total, sino un apoyo es-
porädico, un apoyo asi circunstancial (...) mas eran hombres no totalmente
identificados con su cultura, con su mundo andino, etcetera". [And the State
hasn't (...) hasn't lent total support, only sporadic support, support in certain
circumstances (...) but they were men who didn't identify totally with their cul-
ture, with their Andean world, etcetera.]
Their culture - their Andean world: every Peruvian, from wherever in the
country, speaking whatever language, of whatever extraction or orientation, is
expected willy-nilly to acknowledge the cultural and linguistic pre-eminence
of Cusco and the Andes and accept it as the defining mark of his or her peruani-
dad (or "Peruvian-ness"). By extension, then, Quechua is not just one of the es-
timated 44 languages extant in Peru (the number cited by the linguist Gustavo
Solis at the first Encuentro Internacional de Peruanistas at the University of
Lima, 3 - 6 September 1996). Rather, Quechua is the true "native" tongue of all
Peruvians. If they refuse stubbornly to recognise this, it can only be because
they are ashamed of their nationality:
Que es asi como el nino yanqui ο norteamericano se siente orgulloso de su pais, de su

cultura, de su historia, nosotros tambien queremos que el nino peruano, el joven pe-
ruano, se sienta orgulloso £no? de su historia, de su cultura ^no? de su lengua materna.
[It's like the way Yankee children, North American children, feel proud of their
country, their culture, their history; we want Peruvian children too, Peruvian
young people, to feel proud, you see? Of their history, of their culture, you see?
Of their mother tongue.]
Within the logic of this worldview, the Academia, faithful to the supposed
mother tongue, represents the true patriots. Indeed, it seems to be suggested that
South Americans (Latin Americans? Americans?) in general should accept this
definition of cultural identity, whereby the Tawantinsuyo (the "four quarters",
the historic area of the Inca empire at its brief apogee) comes to stand as em-
blematic of the whole continent: "El quechua del Qosqo. Qosqo como capital de
la nacionalidad continental del Tawantinsuyo". [The Quechua of Cusco. Cusco
as the capital of the continent-wide nationality of the Tawantinsuyo.]
The academician thus defines the agenda of his organisation in the most
grandiose terms; it amounts to the reduction of the modern republic (and even
beyond) to a sphere in which his and his colleagues' influence - as guardians of

the true language and culture - might be supposed to hold sway.
4. Quechua language and the discourse of Quechua superiority
The cult of admiration for the Incas is founded on a series of givens. One is the
superiority of their system and philosophy of government to any other. If, ac-
cording to the former (to 2000) president of the Academia, Inca society was
"casi un paraiso" [almost a paradise] (Pacheco 1994: 13), then it must have
been based on extraordinary principles. Hence the academician lights upon
a - frankly rather banal - pensee attributed to Pachacutec Inca, and invests it
with enormous significance (and note the curious conflation of the Inca empire
with the modern notion of the "Republic" - presumably of Peru):
Ha aqui algun pensamiento de Pachacutec, y su traducciön es: el hombre que no sabe

gobernar su casa y sufamilia, menos sabrä gobernar la Republica. Un pensamiento
de profundo sentir y contenido filosofico. Ningun pensador griego, romano ο orien-
tal ha tenido este pensamiento por ejemplo (...)
[There is a theory that Pachacutec had, and it translates as: the man who cannot
govern his house and his family, still less will he be able to govern the Repub-
lic. A thought of profound feeling and philosophical content. No Greek,
Roman or Oriental thinker has had such a thought, for example (...)]
The presumed superiority of Inca science and cosmology is also singled out
for praise, at the expense of the supposed scientific underdevelopment of
16th century Europeans, who had "(...) una conception filosofica atrasada,
retrograda" [a retarded and backward philosophy]. The present president of the
Academia remarks without further elaboration that Pachacutec Inca was "el
mejor estadista y el mejor astronomo" [the greatest statesman and the greatest
astronomer], while the Quechua teacher insists: "nadie actualmente ha llegado
al nivel de desarrollo al que llegaron los incas" [no-one in the present day has
reached the levels of development that the Incas reached], and even suggests
airily that the Incas engaged in space travel.
The recurring leitmotif of Inca cultural superiority, though, is more tan-
gible: the grandeur of their building. The archaeological sites of Incaic Cusco
(and only Incaic Cusco - there is no mention of Incaic sites elsewhere, or of
pre-Incaic sites such as, say, Tiwanaku in modern Bolivia, or Chavin de Huän-
tar and Chan Chan in Peru) are deployed as prima facie proof of the perfection
of that civilization, without any further explanation being considered necess-
206 Tim Marr
ary: "Sentimos orgullosos de esa historia gloriosa de los incas. ^Por que? Por-
que ahi estän sus obras, como Machu Picchu ο Sacsayhuamän, que es motivo
de admiration de parte de toda la humanidad". [We feel pride in the glorious
history of the Incas. Why? Because there we have their great works, such as
Machu Picchu or Sacsayhuamän, which are admired by all humanity.]
Later in the interview it is suggested that these sites were actually destroyed
by the Spaniards: "Han destruido las portentosas obras, unas obras excep-
cionales que habian de los incas, como Machu Picchu - bueno, Machu Picchu
no, sino Sacsayhuamän, Pisaq, Ollantaytambo". [They destroyed the majestic
creations, exceptional creations of the Incas, such as Machu Picchu - well, not
Machu Picchu, but Sacsayhuamän, Pisaq, Ollantaytambo.] The demonstrable
continuing existence of these sites sits ill with the simultaneous claim that the
Spaniards destroyed them (and the assertion that Machu Picchu - undiscovered
by the Spaniards - was destroyed is indeed withdrawn in timely fashion). How-
ever, the underlying intention of the discourse, as it appears to the listener, is to
establish not facts, but the thoroughgoing malevolence and philistinism of the
conquistadores. The "Incas" (by which term one is presumably supposed to
understand the ruling class of the Inca empire) are conceived of as mystical,
noble, and spiritual - the symbolic moral antithesis of the Spaniards.
Within the framework of this discourse, all facets of Inca / Andean culture
(the two are never formally distinguished) are superior to all manifestations of
Spanish/European culture (again, never formally distinguished). Whatever is
Spanish is corrupted, backward, barbarous; whatever is Incaic is, in the most
real sense, perfect: consummate and unimprovable. Most importantly, for our
present purpose, this must hence be true in the matter of language. As was
noted earlier, for the Academia it is inarguable that the Incas, in Cusco, were
the first Quechua speakers: the language is hence perceived - as if it were a
kind of verbal Machu Picchu - as yet another unsurpassed achievement of Inca
culture. The interview includes a commentary on the ways in which Quechua is
structurally superior to all other languages:
Sinceramente el idioma quechua por ejemplo ha superado, este (...) el articulo. No

existe como una categorfa gramätica el articulo. i,No? Para decir por ejemplo: yo voy
a ir a tu casa. Yo - voy - a - i r - a - t u - casa. Siete palabras serfan. En quechua serfa
wasiykita risaq. Dos palabras. Es un idioma polisintetico, que sintetiza, ^no? que
con unas solas palabras se puede expresar toda una oracion, todo un juicio.
[Quite honestly, the Quechua language for example has overcome, er (...) the
article. The article does not exist as a grammatical category. You see? To say,
for example: I ' m going to go to your house. I ' m - going - to - go - to - your -
house. That would be seven words. In Quechua it would be wasiykita risaq.

Two words. It's a polysynthetic language, a synthesising language, you see? So
with just a few words you can express a whole sentence, a whole opinion.]
The linguistic absurdity of this comparison (like that of the blithe assertion,
cited below, that Quechua is "more perfect than Latin and Greek") goes un-
examined: indeed, it is irrelevant. Within the terms of the discourse of Quechua
superiority, every facet of its phonology, lexicon and morphosyntax, even the
fact that it happens to lack articles (that it has "overcome" them is the proud
boast) is simply further proof of its innate perfection.
5. The influence of the European in the discourse of the

Academia
The irony concealed at the heart of the discourse of the Academia is its para-
doxical embracing of foreign, colonial and metropolitan influence in language,
thought and behaviour. The foundation of this entire ideological edifice is the
glorification of the Incas: yet the Incas are never thought of or understood in re-
lation to contemporary or earlier Andean cultures; still less are they thought of
in objective isolation. They seem to gain shape only when they are compared
with outsiders. Throughout the interview, the speaker's style of argument rests
on the use of contrast and comparison: nothing can be judged on its own terms.
As seen above, he cannot be satisfied with the quoting of Pachacutec Inca's
philosophy of government but must compare it to the ancient philosophers of
Europe and the East (finding it, of course, superior). Most tellingly of all, the
academician contentedly cites the interest of a handful of traditionally-minded
foreign linguists in Quechua as if this were the final, triumphant proof of the in-
herent virtues of the language.
Entonces tenemos nosotros la suerte de contar con intelectuales de mucho prestigio,

como por ejemplo con este doctor David Weber. Tambien habia un lingiiista Honorio
Mossi, italiano. El ha dicho por ejemplo que el idioma quechua es mäs perfecto que el
latin y el griego. Hemos tenido a un alemän. Ernst Middendorf. Tiene varios libros.
[So we are lucky enough to have intellectuals of great prestige, like for example
this Dr. David Weber. There was also an Italian linguist, Honorio Mossi. He, for
example, has said that the Quechua language is more perfect than Latin and
Greek. We have had a German, Ernst Middendorf. He has written several
books.]
208 Tim Marr
In a similar vein, the teacher of Quechua describes a conference in May

2000: "Han venido desde lo mäs lejos hablando quechua, desde Canada, Japon,
Austria, Dinamarca. Todos hemos estado de acuerdo de que el quechua del
Cusco debe ser el que se difunda uniformemente en todas partes". [People
came from far away speaking Quechua: from Canada, Japan, Austria, Den-
mark. We were all agreed that it must be the Quechua of Cusco which is spread
uniformly everywhere.]
For all their determined championing of their home ground, there appears to
be an uneasy feeling amongst many of the members of the Academia that real
recognition can come only from outside. The frustration and resentment felt by
a provincial elite towards the real seat of power, the capital, is everywhere
manifest in the discourse of the academician. First the importance of Lima as
an attraction in itself is scorned, and again, the opinions of foreigners - in the
government-sponsored "Year of 600,000 Tourists" - are invoked to prove the
point: "Se quiere hacer por ejemplo turismo de 600,000 personas, ο sea la veni-
da de 600,000 turistas al Peru, al Cusco. Pero ^en funcion de Lima, en funcion
del turismo? [No! En funcion del Qosqo y sus riquezas culturales". [They want
for example to have 600,000 tourists, that is, to have 600,000 tourists coming
to Peru, to Cusco. But is this tourism happening because of Lima? No! It's be-
cause of Cusco and its cultural riches.]
But while Lima is thus to be dismissed out of hand, it is simultaneously the
source and measure of real success and influence. The presence of Quechua in
Lima universities (a largely illusory presence; not for nothing does the list peter
out after the mention of San Marcos University) is spoken of with ingenuous
pride: "Por eso yo veo con mucha admiration que tenemos la suerte de que en
la mayor parte de las universidades del Peru por ejemplo, jen Lima misma! en
San Marcos, en (...) en todas las universidades de Lima, se ensena quechua".
[So I admire greatly the fact that we have the good fortune to have Quechua
taught in most of the universities in Peru, for example, in Lima itself! In San
Marcos, in (...) in all the universities in Lima.] In the terms of this discourse,
Spanish colonialism can have brought nothing of any benefit to Peru. The
Spaniards who came to Peru are condemned in the harshest terms, as for
example "(...) gente sacada de las cärceles (...) grandes eliminates, y en el caso
de las mujeres, eran mujeres sacadas de los peores prostibulos de Europa"
[people plucked from prisons (...) complete criminals, and in the case of the
women, women taken from the worst brothels in Europe]. Certainly the Cusco
elite would wish to disassociate itself from such people. And yet the influence
of Spanish culture, thought and language in the universe of the academicians is
all-pervading. One scarcely needs to be a practising Freudian to suspect that
this perhaps haunts their unconscious thoughts, given the bombastic way in
which it is denied. The former president of the Academia writes:
[T]enemos y debemos abandonar la mentalidad, actitud, position, parämetros, cat-

egorfas occidentales, europeo-hispanistas, que son total y completamente incompat-
ibles con la realidad y la estructura biosiquico-social del habitante andino-Inka, y
por ende, con su mentalidad y manifestaciones conductales. (Pacheco 1994: 109)
[We must and we have to abandon the Western, European-Hispanic mentality,

attitude, position, parameters and categories, which are totally and completely
incompatible with the environment and the biopsychological-social structure
of the Andean-Inca, and also, therefore, with his mentality and behaviour.]
Unable to free themselves from the constraints of the Spanish paradigm,
and often seemingly unconscious of this, the mestizo elite are condemned to
recreate it, in endless inferior variations, in the Andean world. The model for
the Academia - indeed the very notion of a language academy - has, of course,
been taken from Spain. In the interview the task of the Academia was justified
thus: "Asi como la Real Academia Espanola, £lo cierto? ejerce esa actividad de
normar, ^lo cierto? de igual manera la Academia Mayor de la Lengua Quechua
tambien tiene esa obligation de normar (...)". [Just as the Real Academia Es-
panola - does it not? - carries out normative functions - does it not? - so in the
same way the Academia Mayor de la Lengua Quechua also has that obligation
to set norms (...)]
Itier (1992a) points out, acutely, that the Academia-approved use of five vo-
wels in Quechua orthography, rather than the three favoured by most academic
linguists (see Itier 1992b, Samanez 1992 for a taste of this long-running and
bitter argument) is no more than a demonstration of the extent to which Spanish
has permeated the Quechua of Andean bilinguals. The academician's defence
of pentavocalismo provides substantial (if unwitting) support for this assertion:
El uso por ejemplo de las cinco vocales no es capricho de uno ο dos intelectuales: es
producto, es determinaciön de congresos internationales, como el congreso interna-
tional todavia de 1950 en La Paz, Bolivia. El congreso international de quechua acä
en la ciudad del Qosqo en 1987. El congreso international en Lima en 1991. Son
congresos internationales.
[The use of the five vowels, for example, is not the whim of one or two intel-
lectuals: it is the product of, it has been determined by, international con-
gresses, like the international congress in La Paz, Bolivia, as long ago as 1950.
The international Quechua congress here in the city of Cusco in 1987. The in-
ternational congress in Lima in 1991. These are international congresses.]
210 Tim Marr
Just as the Academia perceives Quechua orthography through the distorting

grid of the Spanish phonological system, so does it perceive the nature of its
task through a Hispanic mindset. While one would not perhaps wish to argue
that it is or was a trait peculiar to the Hispanic world, certainly a degree of
legalism in argument has been a marked element of the culture bequeathed by
Spain to Peru. With his concern for laws and decrees and his insistence on the
rights of duly constituted authority - not to mention his determined regional-
ism and resentment of the capital - the academician is almost a caricature of
the provincial lawyer in 19th century Europe. And yet he insists that he and his
colleagues are the cultural heirs of the Incas, and that the Europeans brought
nothing of any lasting significance to Peru. In his lament over the failure of the
government to legislate anew for the compulsory teaching of Quechua in
schools, the legalism inherited from Spain is faithfully reproduced even as the
influence of Spanish culture is deplored: "Sigue entonces esa mentalidad, to-
davia occidental, esa mentalidad espanola (...) Y lo que nos falta es, precisa-
mente, de que se tiene que oficializar el idioma quechua. ^Para que? Para ense-
nar a esa nueva generation". [So there still continues that Western mentality,
that Spanish mentality (...) And precisely what is needed now is that Quechua
be declared an official language. Why? So that the new generation will be
taught it.]
There is no hint of doubt here that language can - should, even - be con-
trolled and regulated by decree. The securing of the future of Quechua is under-
stood by the Academia to be synonymous with the achievement of a single
goal: official status, and hence compulsory teaching - of the Academia's qha-
paq simi, it goes without saying. As the teacher explains: "Queremos que se di-
funda obligatoriamente en los colegios (...) Que se haga obligatorio, pero que
lo maneje la Academia Mayor de la Lengua Quechua". [We want it to be an ob-
ligatory subject in schools (...) We want it to be obligatory, but it must be di-
rected by the Academia Mayor de la Lengua Quechua.]
In this determinedly top-down, legalistic scheme of things, the complexity
and delicacy of real-world language maintenance efforts go entirely unac-
knowledged. School language teaching is enough, and will succeed in produc-
ing new generations of Quechua speakers. The very substantial body of experi-
ence and literature derived from bilingual education projects in the Andes (see
e.g. Weber 1994; Hornberger 1994; Hornberger and King 1996) is never once
mentioned in the course of any of these interviews, any more than is the relative
failure of the Velasco government's attempt in 1975 to establish Quechua as a
national tongue by making it official and a compulsory element of the school
curriculum (see e.g. Cerron-Palomino 1989; Paulston 1992, 1994; von Gleich
1994).
An outsider with an interest in language maintenance might justifiably

wonder whether the attentions of these traditionally-minded quechuistas would
not be better directed away from congressional decrees and towards the main-
tenance of intergenerational transmission at grass roots level (as urged by Fish-
man 1991, and see also Fishman 2000) - that is, towards the millions of speak-
ers who have acquired a variety of Quechua as their sole or principal code with-
out the intervention of a single senator or schoolteacher. As will now be seen,
however, such speakers have not always been regarded by the Academia as a
resource to be cultivated.
6. Whose Quechua? The provincial elite and ideology in

language
The attitude of Peru's ruling classes to the country's historical culture has tra-
ditionally been ambivalent and selective: it is summed up with admirable suc-
cinctness in the title of Mendez's (1996) article "Incas si, Indios no". The mes-
tizo bourgeoisie of Cusco wish to see themselves (and be seen) as the inheritors
of Inca glory, not as the kin of impoverished peasants, and the symbolism of lan-
guage must be pressed into serving this end. Seeking to remake the model of pe-
ruanidad in their own image, they hence start from the assumption that the
"best" Quechua is spoken by themselves. Rene Farfän Barrios, another leading
member of the Academia, insists: "Cuando se habla del quechua-hablante no
hay que pensar en el indio que estä en la puna, ο en el indigena, el nativo, como
se quiera llamar. Es el mestizo, a ellos hay que dirigirnos" (cited in Itier 1992b:
91). [When we talk about Quechua speakers we should not think of the Indian
on the high Andean plateau, or the indigenous person, or native, or whatever
you wish to call him. It is the mestizo, that is who we must address ourselves to.]
In fact, as Godenzzi (1992: 63) remarks, the supposedly Incaic Quechua
preferred by the Academia (he points out that it is actually heavily influenced
by Spanish) is a class-specific sociolect. There exists amongst some of the gov-
erning elite of the Academia a patronising and disdainful view of the language
used by the great majority of monolingual speakers, which is seen as having
fallen away from the classical model, become coarsened and degraded, a kind
of lingua romana rustica to the lingua latina of the academicians. At one stage
in the interview with the academician the "problem" of the peasant speaker was
broached:
212 Tim Marr
El problema fundamental con el campesinado es que la mayor parte de ellos no

saben leer y escribir ^no? en idioma quechua. ^Como hacemos? ^Cömo hacemos?
Fundamentalmente haciendo la correction de su pronunciation. i,No? En su pronun-
ciation. i,No? Conversamos asi, ^no? (...) ^Que hacemos nosotros? Nosotros esta-
mos ahi tratando de corregir, £no?
[The fundamental problem with the peasants is that most of them do not know
how to read and write Quechua, you see? So what do we do? What do we do?
Fundamentally we correct their pronunciation. You see? Their pronunciation,
you see? So we talk to them, you see? (...) What is it that we do? We are there
trying to correct them, you see?]
Perhaps the most remarkable leap of logic made here is that, as most mono-
lingual Quechua-speaking peasants do not know how to read and write, so they
do not know how to pronounce their own language correctly. Significantly, this
disdain for varieties other than the supposed standard is projected on to other
languages and cultures, and the foreign interlocutor is assumed to be support-
ive and understanding of this: "Lamentablemente por ejemplo digamos en caso
de Estados Unidos, ese negro que habla atropellando el ingles, ^lo cierto? (...)
Ya no podrias casi influirlos, ^no? Pero la preocupacion fundamental es la ju-
ventud". [Regrettably, for example, say, in the case of the United States, those
blacks who babble their English, you know? You couldn't really influence
them now, could you? But what we are concentrating on fundamentally is the
young people.]
This ingrained sense of racial and class superiority, itself perhaps a compen-
sation for the resentment and sense of inferiority felt by the provincial mestizo
bourgeoisie, has deep historical roots in Cusco. The regionalist Federico More,
writing in 1925, commented: "[E]n la sierra actua el quechua, lengua noble y
lirica, mientras que en la costa apenas suenan los monosflabos de las plebes de
Pekin y las guturaciones de aquellos negros que fueron esclavos (...)" (cited in
Itier 1992a: 41). [In the Andes is spoken Quechua, a noble and lyrical lan-
guage, while on the coast you can barely make out the monosyllables of the Pe-
kinese underclass and the hoarse sounds of those blacks who used to be
slaves (...)]
The echo of More's words some 70 years later is striking; this modern
member of the Academia, too, considers Black speech to be degenerate - and
one suspects that he, too, might have little time for the speech of Chinese-Peru-
vians.
Meanwhile, those academic linguists and quechuistas who take issue with
the Academia's stance on the language are bitterly attacked, and their motives
maligned (Itier 1992b; Cerron-Palomino 1997): and so, within the framework
of discourse established by the Academia, must they be. Far from indulging in
objective academic debate, they are challenging the very authority of the insti-
tution, its right to pronounce upon "its" language. This right was re-asserted in
the interview in the strongest terms, with these linguists being accused of writ-
ing books on Quechua for financial gain, rather than in the interests of scientific
research ("Para hacer fortuna [...] No con esa buena intention de hacer ver-
daderamente ciencia"). Similarly, the president of the Academia, Dr Herencia,
states:
Hay algunos intelectuales, entre comillas, que estän tergiversando el idioma que-
chua (...) Aquellos que manejamos la ciencia lingüistica del idioma quechua ten-
emos razon, y tarde ο temprano van a caer esas personas que trafican con el idioma.
[There are certain intellectuals - in inverted commas - who are twisting the
Quechua language (...) But those of us who deal with the linguistic science of
Quechua are in the right, and the downfall will come, sooner or later, of those
people who make a living by trading in the language.]
These are harsh words. And yet the level of "scientific" rigour demanded by
the Academia appears, unfortunately and all too often, to be low. Cerrön-Pa-
lomino's withering review (1997) of the Academia's long-awaited dictionary
(AMLQ 1995) lists a seemingly endless series of errors, omissions, inconsist-
encies and prejudices to be found within its pages. Nor does one need to be as
learned in Quechua as Dr Cerron-Palomino to find it an inadequate work: this
expensively-produced hardback book has its bibliography arranged by alpha-
betic order of first names. A similar level of scholarship seems to have in-
formed the book on Inca society written by DrHerencia's predecessor as presi-
dent of the Academia, Dr Juvenal Pacheco (Pacheco 1994). In the section of the
book which purports to show that Quechua is superior to all other languages,
the author asserts that languages such as Thai and - incredibly enough - Viet-
namese use ideographic writing systems. The same is claimed, too, for Arabic,
language of the "bloque arabe" [Arab bloc] - in which for good measure is in-
cluded Iran (Pacheco 1994: 120). Given such ignorance of matters linguistic at
the most senior levels of the Academia, it comes as scant surprise that Itier
(1992b: 86), in describing the heated debate on pentavocalismo that broke out
at a 1986 workshop in Cusco, can remark drily that dialogue was made difficult
because "los academicos de la lengua quechua no entendian lo que es un fone-
ma" [the academicians of the Quechua language did not understand what a
phoneme is].
214 Tim Marr
7. Conclusion
The discourse of the Academia often seems to display a tension, a marked am-
bivalence, which is experienced at several levels. This is seen in the question of
the institution's relationship to the Peruvian state. Acutely conscious of the
prestige and authority which state legitimation affords them, guarding jeal-
ously in the face of all competitors and critics their official endorsement, the
academicians (or at least those senior members who purport to speak for the
academicians) are nevertheless resentful of the power exercised in Lima. The
capital is regarded as a usurper of Cusco's ancient privileges; Cusco remains
the rightful symbolic centre of the nation, and those who do not understand or
accept this supposedly self-evident truth are considered to be lacking in self-
knowledge, or even in patriotism.
The Academia's attitude to the majority of Quechua speakers is likewise an
ambivalent one. Quechua is exalted as the most noble and perfect of tongues,
and yet the speech of most of those who speak it every day as a mother tongue
is regarded as debased and flawed. Quechua is not only to be maintained, then,
but must also be corrected and purified. This process of correction and purifi-
cation is viewed as necessitating, in essence, the imposition of the language of
the academicians themselves - middle-class, literate bilinguals - on current
and (especially) future speakers of the language, preferably through compul-
sory teaching in schools.
Much ambiguity derives, of course, from the contradictions inherent in the
way Quechua is perceived and represented in Peru; it is simultaneously the
stigmatised language of an oppressed minority and a state-legitimated symbol
of former national glory (that is, what Fishman [1972: 44] calls "the link with
the glorious past"). Any body (indeed, any individual) campaigning for the
maintenance or revitalisation of Quechua in Peru, then, does so within a social
context where language attitudes have become heavily ideologised in some-
times contradictory ways: discourses upon Quechua tend to dwell heavily on
the notions of shame and pride, often together.
What are the implications of all this for Quechua language maintenance ef-
forts? My own research amongst migrant communities undergoing rapid lan-
guage shift (Marr 1998) suggests that the constant identification of Quechua
with the Incas, which forms the core of the Academia's political and linguistic
discourse, is - at best - effective only at the level of reinforcing national or re-
gional pride. In terms of attitudes to "real" language it is negative, having the
effect of demeaning the speech of present-day monolinguals and of locking the
Quechua language into an idealised and remote past. As Nino-Murcia (1997:
157) quite rightly concludes: "The purist discourse in Cuzco, although it ap-
pears on the surface to legitimize indigenous culture (...) in reality contributes

to the marginalization (...) of the indigenous language and ultimately of its
rural speakers, whose language one sees marked by the stigma of poverty and
equated with a lack of culture".
Genuine grass roots movements for language maintenance are hence un-
likely to find an ally in the shape of the Academia as presently constituted. It is
to be suspected that much of the leadership of the institution, with its roots in
the bilingual Cusco elite, would have little interest in, and perhaps a decided
hostility to, any language maintenance project (whether it arose from state
planning or from grass roots aspirations) that it did not itself legitimate and
control. Perhaps the most striking point to emerge from a consideration of the
tradition which gave rise to the Academia is that, quite obviously, it could be
hostile to any project which proceeded from the assumption that the language
of rural monolinguals was in itself good, whole or representative.
It might of course be argued that the Academia's illustrious European mod-
els, such as the Spanish Real Academia, take a not dissimilar line: such insti-
tutions tend, after all, to be in the business of formulating and imposing (or at-
tempting to impose) norms, not that of dispassionately describing language
use. However, the nature of the task that the Peruvian institution sets itself must
inevitably be coloured by the hard fact that Quechua is not an expanding world
language, but an embattled minority language which has relatively few speak-
ers outside its rural Andean heartland, and a powerful competitor in the shape
of Spanish. Simple maintenance, rather than anything more ambitious, must
surely be the primary goal (and indeed, maintenance of the language is one of
the Academia's primary stated aims); the question is whether its approach to
the maintenance of Quechua is an effective one or not.
What is at issue here is the perceived "ownership" of Quechua: the authority
to speak about, and on behalf of, the language. However, the Academia can
only with difficulty be regarded as a language academy or language mainten-
ance organisation as commonly understood. Some of the senior academicians
appear to have as their chief aim not the normalisation and/or maintenance of
Quechua, but public ratification of their own view of themselves as the su-
preme embodiment of a culture, a region, a nation and perhaps even a conti-
nent. If the principle of self-determination for linguistic minorities is to have
any real value, it is vital that this subtext to the discourse of Peruvian language
maintenance be recognised, and that the opinions and aspirations of Quechua
speakers as a whole not be confused with those of their self-appointed repre-
sentatives. The Academia potentially has a great deal to contribute to the
struggle for Quechua language maintenance: however, the fact that its support
for the language has become entangled with a complex of dubious ideological
216 TimMarr
positions means that its efficacy and authority are, for the present at least, se-
verely compromised.
References
Albo, X.
1979 The future of the oppressed languages in the Andes. In: W. C.
McCormack and S. Wurm (eds.), Language and Society, 309-330.
The Hague: Mouton.
AMLQ (Academia Mayor de la Lengua Quechua)
1995 Diccionario Quechua-Espafiol-Quechua / Qheswa-Espafiol-Qheswa
Simi Taqe. Cusco: AMLQ / Municipalidad del Qosqo.
Cerron-Palomino, R.
* 1989 Language policy in Peru: a historical overview. International Journal
of the Sociology of Language 77: 11-33.
1997 El Diccionario Quechua de los academicos: cuestiones lexicogräficas,
normativas y etimologicas [The Quechua Dictionary of the Academi-
cians: lexical, normative and etymological issues]. Revista Andina 29:
151-205.
Degregori, L. N.
1994 Una aproximacion al cusquenismo [A view on Cusquenismo], Allpan-
chis 43 / 44: 441-476.
Fishman, J. A.
1972 Language and Nationalism. Rowley: Newbury House.
* 1991 Reversing Language Shift. Clevedon: Multilingual Matters.
Fishman, J. A. (ed.)
* 2000 Can Threatened Languages be Saved? Reversing Language Shift, Re-
visited: a 21s' century Perspective. Clevedon: Multilingual Matters.
Godenzzi, J. C.
* 1992 El recurso lingüistico del poder: coartadas ideologicas del castellano y
el quechua [The linguistic recourse of power: ideological alibis of
Castilian and Quechua]. In: J. C. Godenzzi (ed.), El Quechua en De-
bate: Ideologia, Normalization y Ensehanza [Quechua in debate:
ideology, standardisation and teaching], 51-77. Cusco: Centra de Es-
tudios Rurales Andinos Bartolome de las Casas.
Hornberger, Ν.
1994 Whither bilingual education in Peru? Quechua literacy and empower-
ment. In: P. Cole, G. Hermon and M. D. Martin (eds.), Language in
the Andes, 74-89. Newark: University of Delaware.
* 1999 Maintaining and revitalising indigenous languages in Latin America:

state planning vs. grassroots initiatives. International Journal of Bi-
lingual Education and Bilingualism 2: 159-165.
Hornberger, Ν. and Κ. King
1996 Language revitalization in the Andes: can the schools reverse lan-
guage shift? Journal of Multilingual and Multicultural Develop-
ment 17: 427-441.
Itier, C.
* 1992a Lenguas, ideologia y poder en el Cusco: 1885-1930 [Languages,
ideology and power in Cusco]. In: J. C. Godenzzi (ed.), El Quechua en
Debate: Ideologia, Normalizacion y Ensenanza, 25-48. Cusco: Cen-
tra de Estudios Rurales Andinos Bartolom£ de las Casas.
1992b "Cuzquenistas" y "Foräneos": las resistencias a la normalizaciön de la
escritura del quechua ["Cuzquenistas" and "foreigners": resistance to
the normalisation of the writing of Quechua]. In: J. C. Godenzzi (ed.),
El Quechua en Debate: Ideologia, Normalizacion y Ensenanza, 85-93.
Cusco: Centra de Estudios Rurales Andinos Bartolome de las Casas.
Mannheim, B.
* 1991 The Language of the Inka since the European Invasion. Austin: Uni-
versity of Texas Press.
Manya, J. A.
1992 El quechua imperial del Qosqo [The imperial Quechua of Qosqo]. In:
J. C. Godenzzi (ed.), El Quechua en Debate: Ideologia, Normaliza-
cion y Ensenanza, 49. Cusco: Centra de Estudios Rurales Andinos
Bartolome de las Casas.
Marr, T.
1998 The language left at Ticlio: social and cultural perspectives on Que-
chua language loss in Lima, Peru. PhD thesis, Institute of Latin
American Studies, University of Liverpool.
Mendez, C. G.
* 1996 Incas si, indios no: notes on Peruvian Creole nationalism and its con-
temporary crisis. Journal of Latin American Studies 28: 197-225.
Nino-Murcia, M.
* 1997 Linguistic purism in Cusco, Peru: a historical perspective. Language
Problems and Language Planning 21: 134-161.
Pacheco Farfän, J.
1994 Filosofia Inka y su Proyeccion al Futuro [Inca philosophy and its pro-
jection into the future]. Cusco: Universidad Nacional de San Antonio
Abad.
218 TimMarr
Paulston, C. Β.
1992 Linguistic minorities and language policies. In: W. Fase, Κ. Jaspaert
and S. Kroon (eds.), Maintenance and Loss of Minority Languages,
55-79. Amsterdam/Philadelphia: John Benjamins.
1994 Linguistic Minorities in Multilingual Settings. Amsterdam/Philadelp-
hia: John Benjamins.
Rojas Rojas, I.
1980 Expansion del Quechua [The expansion of Quechua]. Lima: Edi-
ciones Signo.
Samanez Florez, D.
1992 Pentavocalismo vs. trivocalismo [Five vowels versus three vowels].
In: J. C. Godenzzi (ed.), El Quechua en Debate: Ideologia, Normal-
ization y Enserianza, 97-105. Cusco: Centro de Estudios Rurales An-
dinos Bartolome de las Casas.
Torero, A.
1974 El Quechua y la Historia Social Andina [Quechua and Andean social
history], Lima: Universidad Ricardo Palma.
Von Gleich, U.
* 1994 Language spread policy: the case of Quechua in the Andean republics
of Bolivia, Ecuador and Peru. International Journal of the Sociology
of Language 107: 77-113.
Weber, D. D.
1994 Mother-tongue education for speakers of Quechua. In: P. Cole,
G. Hermon and M. D. Martin (eds.), Language in the Andes, 90-115.
Newark: University of Delaware.
1. In terms of language maintenance, in what ways might the Academia's at-

tachment to the historic Inca culture of Peru be a strength, and in what ways
a weakness?
2. Quechua has been called an "oppressed" language (Albo 1979). If this is the
case, does it help to explain some of the Academia's ideology?
3. Should Quechua language maintenance efforts in Peru concentrate around a
recognisably prestigious dialect - the supposed qhapaq simi - or around the
variety of regional dialects used by ordinary speakers, or around a mixture
of the two? What problems do you think each of these approaches might en-
counter?
4. Why do you think the Academia seems to set such store by the question of
official status for Quechua, and compulsory teaching of the language in
schools? What are some of the risks and benefits inherent in this approach?
5. Quechua-speaking migrants to urban areas, and especially to Lima, typi-
cally are reluctant to use the language in public, and sometimes even in pri-
vate. Their children almost inavariably grow up speaking only Spanish
(Marr 1998). What role might organisations like the Academia have to play
in stemming or slowing this language shift? What would they need to do,
and what other individuals or organisations would need to be involved?
Xhosa as a "home appliance"? A case study of
language shift in Grahamstown
Vivian de Klerk
1. Introduction
A number of scholars (Fasold 1984; Aitchison 1991; Denison 1977; Dorian

1980; Gal 1979) have examined the issue of language maintenance and shift,
trying to discover why certain languages (or language variants) sometimes re-
place each other among some speakers, particularly in certain domains of lin-
guistic behaviour under some conditions or intergroup contact. This article
provides an overview of the main factors which have been identified as playing
an important role in influencing language shift, and then reports on the relative
importance of these factors in a survey which examined the experiences and at-
titudes of Xhosa-speaking parents who have recently chosen to send their
children to English-medium schools in Grahamstown (Eastern Cape, South Af-
rica). The research was carried out during 1998, and the project was a multi-fa-
ceted quantitative and qualitative longitudinal study involving responses to a
postal questionnaire sent to all non-English parents at English-medium schools
in the town, and follow-up interviews with 26 parents. The aim of the project as
a whole was to observe whether there is any evidence of a process of language
shift taking place from Xhosa to English, both on an individual level and on a
broader societal basis, to assess the linguistic and psycho-social effects on in-
dividuals who move to English-medium schools, and to monitor changing per-
ceptions, language loyalty and attitudes over this period. This paper attempts to
assess the relative importance of a range of variables influencing the rate of
possible language shift.
2. The local context
"The study of language maintenance and shift is concerned with the relation-
ship between change and stability in habitual use, on the one hand, and ongoing
psychological, social or cultural processes on the other hand, when populations
differing in language are in contact with each other" (Fishman etal. 1966:424).
222 Vivian de Klerk
It has frequently been documented that when the mother tongue (MT) is dif-
ferent from the economically dominant language of a given region, shifts in
usage and attitude are observed. These shifts are accelerated when a child
moves to a school with a MOI other than his / her MT, for although the child
may continue to speak the MT at home, the language of school and peers in-
fluences the child's language preferences, and "during this period there is a no-
table increase in the percentage of persons who make the definite break with
the language of daily use" (Veltman 1983: 20). Such trends have been observed
world-wide: McKinnon (1978) and Taighde (1990) report on the decline in
Gaelic usage in Ireland in favour of English; Hofman and Cais report on re-
search in Israel (1984:151); Holmes etal. (1993) report shift to English in New
Zealand, as does Wen Lang Li's (1982) study of assimilation to English among
Chinese Americans; Gal's (1979) work describes conditions in Austria; Pütz
(1991) writes of German-speaking people in Australia and Smith (1977) pro-
vides a detailed account of language shift in Teherne, Friesland. In the South-
em African context, studies on language shift include the following: Schuring
and Calteaux's (1997: 17) report on the increasing trend among speakers of in-
digenous African languages of seeing English as "the language of prestige and
something to be aspired to"; Mesthrie's (1996) comprehensive documentation
of the slow replacement of Indian languages by English in Natal; and Smeija's
(1998a and 1998b) study in Botswana, which reports a steady language shift
from Setswana and other minority languages to English, (which is seen as a
powerful economic and educational tool, the language with higher status and
prestige).
Language shift occurs when linguistic communities find themselves in con-
tact with a language which offers greater practical and economic rewards or
carries higher prestige. "What begins as the language of social and economic
mobility ends, within three generations or so, as the language of the crib as
well, even in democratic and pluralism-permitting contexts" (Fishman 1989:
206).
South Africa is just such a democratic, pluralist context: a great deal of ef-
fort has recently gone into promoting its recently declared policy (1996) of par-
ity between all of its 11 official languages. In addition to legislative measures
to strengthen local languages and provide special support for the nine indigen-
ous languages, to ensure that the national language plan can be successfully
implemented, various measures have been implemented: after wide consul-
tation, a Language Plan Task Group (LANGTAG) was established, which for-
mulated a national language plan, and brought into being the Pan-South Afri-
can Language Board (PANSALB), a Senate subcommittee, charged with a
Xhosa as a "home appliance" ? A case study of language shift in Grahamstown 223
watchdog role in ensuring that the country's new language policy is carried out,
a function shared by statutory language committees in each province.
However, despite de jure parity, there is still a long road ahead before de
facto linguistic parity is achieved, and in particular the appeal and power of
English is increasingly evident: English is showing an increasing tendency to
monopolise many areas of public administration in South Africa, it is perva-
sively used in government contexts (where first language (LI) English speakers
are few and far between) and in parliament, where the majority of members are
fluent in 3 or 4 African languages. Pandor (1995: 75) reports that 87% of all
parliamentary speeches in 1994 were in English, and English is the most popu-
lar default language in other multilingual contexts such as business, schools,
university campuses and government institutions such as military camps and
prisons (de Klerk 1996; de Klerk and Barkhuizen 1998; Bowerman 2000; de
Klerk and Barkhuizen 2001). There are also interesting patterns of English
usage on South African television, where, on English programmes such as the
news, non-English reports are avoided, or translated in cases where they are
utilized; however, in programmes in all other languages, English inserts are fre-
quent, and translations are not provided (Kamwangamalu 1998: 284; Bower-
man 2000: 63).
PANSALB recently complained that the "tendency in government tiers to
use English as the sole medium of communication was showing a disturbing
rise" (Eastern Province Herald 6/6/97), and in its most recent report, LANG-
TAG (1996) explicitly denounces the steady drift at provincial and national
level to the use of only English, and declares its intention to institute even
stricter measures to counteract its insidious effect. As Albie Sachs put it (1994:
1) "the omnipresence of English can be inconvenient and suffocating and in-
duce a sense of disempowerment and exclusion. In a sense, all language rights
are rights against English, which in the modern world is such a powerful lan-
guage that it needs no protection at all (...)"
Yet, despite the growing awareness in South Africa of the possibility that
elitism, domination and social injustice, as well as personal language loss
could result from the spread of English, and despite concerted attempts to uplift
formerly disadvantaged languages, there seems to be some evidence of lan-
guage shift to English occurring in several regions.
The current study focuses on Xhosa speakers in the Eastern Cape province,
who are not in a numerical minority, but who find themselves in a world where
English offers the greatest financial and political rewards. Because of South
Africa's linguistic diversity and its democratic constitution, the issues of mi-
nority language rights and language in education are particularly important,
224 Vivian de Klerk
and, as Dow (1987: 5-6) puts it, "[SJociolinguistics has continued to challenge
linguists to come out into the field and look even more closely at the dynamics
of language loyalty, conflict, spread, erosion, shift, and in some cases, mainten-
ance against dominant languages".
3. Factors influencing the pace of language shift
While hardly ever abrupt, and invariably preceded by widespread bilingualism

(Weinreich 1979; Appel and Muysken 1987: 40), language shift is the long
term collective result of language choice of a particular language in certain do-
mains, which used to be for the old one, and a desire to give up membership of
an identifiable sociocultural group in favour of "identity as a part of some other
community" (Fasold 1984: 240).
Sociolinguistic studies in the area of language shift and maintenance con-
stantly seek to discover the intervening variables, and to determine what factors
influence shift or maintenance. These studies are particularly important for lan-
guage planners, who could use such information to predict future patterns and
institute preventative measures. While sociopolitical and socio-economic fac-
tors usually predominate in language shift (Gal 1979), a number of additional
factors also play a part in determining linguistic behaviour, and these factors
are not always easily identifiable. There are ten primary factors which have
been identified in the literature, in terms of which the data in this study will be
explored. These include economic factors, levels of institutional support, the
educational environment, education and literacy levels, existing linguistic net-
works, language attitudes, language status and functions, mass media and
gender.
4. Methodology
Lieberson (1980: 11) lists four goals of sociolinguistic surveys: description of

the language situation, measurement of sociolinguistic change, estimation of
consequences of policy and considerations of the impact of language on social
conditions. These goals are difficult to attain in the best of linguistic worlds, es-
pecially in view of the problems in determining the criteria for adequate de-
scription. Can one really define which dimensions have to be included in an ad-
equate description of a language situation? In Goiter's opinion (1987: 43-44)
there is no such thing as a 'real' or 'final' description, there can only be inter-
subjective consensus on what makes an adequate description. "Obviously the

ideal sociolinguistic survey will be one in which there is a sizeable but purely
random sample of the entire population of the universe (...) it is equally ob-
vious that such samples will be rare" (Lieberson 1980: 22).
For this reason, in this study a stratified sample was found to be more pro-
ductive and manageable, since literacy is a serious problem in the Eastern
Cape, and a general random mailing would probably have failed to elicit a rep-
resentative response. This study aimed rather to identify instances of potential
language shift and to monitor them.
In the first phase of this research, statistical information was obtained from
all of the seven English-medium schools in Grahamstown, (a major edu-
cational centre in the Eastern Cape province) regarding the home-language /
mother-tongue of all pupils from the pre-primary level to Grade 8. From this
initial survey, all families of children who spoke Xhosa as home language were
identified and sent questionnaires. In addition, the parents of two Xhosa-speak-
ing pupils from each school year (who had indicated their willingness in their
questionnaire returns) were selected for in-depth interviews. Here, the aim was
to investigate personal linguistic histories and attitudes, changes in language
practices and subjective experiences of the process of language shift. Selected
teachers were also interviewed for their perspective on the experiences of such
students. (See de Klerk, in press.)
The study used both questionnaires and interviews in order to get in-
formation about age, levels of education, parental occupations, language back-
grounds, language usage (in the family, worship, TV, reading, shopping etc.),
attitudes and proficiency. "Not only does the surveyor want to obtain cross-
tabulations between parents and children's mother-tongues, but also additional
characteristics about the parents, so that it is possible to determine the social
characteristics which distinguish those bilinguals who pass the acquired lan-
guage to their children from those bilinguals who do not" (Leiberson 1980:
17).
The difference between reported and observed language behaviour is the
classic problem of reliability and validity in survey research: as far as this re-
search is concerned, it is acknowledged that informants' reports of their own
behaviour (both retrospective and synchronic) may not necessarily be abso-
lutely accurate, since interviewees may want to please the interviewer, or be
unaware of their own language use or attitudes. This poses a problem for valid-
ity and reliability (Gorter 1987: 46), but in such surveys one has limited op-
tions, since observing behaviour is problematic. While the problem cannot be
solved, attempts were made to build in consistency checks by repeating ques-
tions from the questionnaire in the interview, and by using the same question in
226 Vivian de Klerk
more than one way during the interview. This runs a slight risk of redundancy
and obtuseness, but is important for reliability.
In addition, Smith (1977) makes the point that populations which tend to be
more reliable in interviews and questionnaires tend to be those living through
times in which language has been politicised or ideologised; since the language
issue in SA has been a topic of heated debate during the last decade, and com-
munities are regularly confronted with language-related problems and issues, it
is highly likely that these parents, who have personally come to grips with the
thorny issue of the language of learning and teaching of their own child, were
well suited to a study which makes use of self-report, both in questionnaires
and interviews.
5. Results
Altogether there were 194 questionnaire responses from Xhosa-speaking par-

ents, and 108 of them indicated their willingness to be interviewed. 26 of these
parents (representing 2 children from each grade) were selected for in-depth
interviews. Results from the questionnaire responses and the interviews will be
reported alongside each other, with interviews used in each case to offer more
detail regarding the statistical trends obtained from the questionnaires.
5.1. Evidence of language shift?
A major precursor of language shift, according to Lieberson (1980), is bilin-

gual ability, and this ability seemed to be growing in the informants. While the
vast majority of respondents were Xhosa-speaking, only 70% claimed to use
only Xhosa when speaking to each other; 26% reported using a mix of English
and Xhosa and only 4% English (although many admitted to using far more
English with their child). 20 of the 26 interviewees unambiguously declared
that Xhosa was their home language, with only two adding English as an "ad-
ditional" language. With only 3 exceptions, parents interviewed had been edu-
cated in Xhosa (12), or a mixture of Xhosa and English (18), with English as a
second language and Xhosa as the primary medium of instruction. In rating
their own fluency in English, respondents rated themselves fairly positively
overall, as is evident in Table 1:
Table 1. Parents' self-rating of fluency in English
Father (n=156) Mother (n=189)

Excellent 29% 13%
Good 47% 62%
Fair 22% 23%
Poor 2% 2%
Bad 0 0
Another sign of incipient language shift is increased usage of the second lan-
guage in the domestic context. Questionnaire responses regarding the use of
English to the child show that 39% of parents used only Xhosa to their child,
12% used only English, and the rest used a mixture of Xhosa and English. In-
terviews confirmed this tendency, with 12 interviewees reporting that in speak-
ing to their children they made a special effort to use English more, one ex-
plaining that "there is a tradition between me and my children that if they catch
one another speaking Xhosa, they use two clicks to remind the other to speak
English." In the words of one parent "We use Xhosa most of the time but since
Μ was told at school that she must try to speak English at home, so we are try-
ing our level best to speak English with her." Some admitted that since moving
the child, their use of English with each other had increased: "we try to use
Xhosa, but we are using more and more English"; "at first we tried to speak
Xhosa, but you can't help it, the children know the English words, and their life
at school it is English".
It was evident from the interviews that all of these parents strongly believed
in early exposure to English for their children, and made concerted efforts to
speak the language themselves: "I wanted to give her a good background of
English"; "we had to help him very much so, we stopped speaking Xhosa to
him, 75% of the time was English." Several parents had made efforts to lay the
foundation for moving to the English school, either by sending their children to
English preschools, by increasing the use of English at home (8), or by describ-
ing what lay ahead (3): "before she went to school, in the mornings sometimes
I'm just asking her to do something with English, like "Stand up"; "We talked a
lot to him in English, we tried to teach him a few sentences, how to express
himself to the teacher, how to express himself to the other children, how to be-
have, and we tried to collect as much English vocabulary as we could for him".
Seven parents reported deciding to use only English, one mostly English,
and 12 a mixture of Xhosa and English. As one parent put it, they had aimed for
228 Vivian de Klerk
"both English and Xhosa, 50-50, but it is getting to be closer to 75% Eng-
lish - it is a problem, because the little ones, now that their vocabulary is so
good in English, often they cannot understand it in Xhosa, so we must talk so
that they can understand." Nine parents reported a steady increase in the use of
English at home, six noticing "better" English accents and increased confi-
dence in their children: "Her accent is that of a first-language speaker, when we
hear her on the phone we are so proud, we cannot believe it"; "his accent is a
first-language person's accent. Sometimes we struggle to catch what he says";
"They hardly speak Xhosa (...) all the time they like English. I don't encourage
them to speak Xhosa, not at all"; "P chides me not to speak Xhosa, but rather
English"; "L is more open to English than Xhosa"; "the children only speak
Xhosa when we have visitors"; "we want the children to hear us speaking Eng-
lish. So we put a few English phrases so that the children can experience it".
Another indicator that children are moving from one language towards an-
other is the rate at which they themselves use the new language in the domestic
context, instead of reserving it for more formal contexts. It is interesting to note
in Table 2 the percentages of Xhosa children who were reportedly using Eng-
lish at home with parents and siblings.
Table 2. Language used by Xhosa child to talk to.
Mother (n=191) Father (n=149) Siblings (n=185)

Xhosa 26% 34% 36%
English 20% 20% 14%
Xhosa/Eng 54% 46% 50%
In the interviews, it also emerged that English was on the increase in inter-sib-
ling communication, with only six families using only Xhosa, and the rest mix-
ing Xhosa and English or using only English (6): "They speak English even to
their black friends because they struggle with some words in Xhosa"; "Since
she knows more English than her sisters, she teaches them how to speak it
when playing".
6. Factors which played a role
6.1. Economics
Socio-economic factors are mentioned in nearly all studies on language shift:

for most, the instrumental potential of the other language offers upward social
mobility, and patterns of language shift tend to follow patterns of industrial-
isation and economic change (Gal 1979; Fasold 1984; Appel and Muysken
1987). It is only the affluent who are permitted the luxury of language main-
tenance (Wen Lang Li 1982: 111).
It must be remembered that the respondents in this study are economically
advantaged: their ability to pay school fees which are significantly higher than
those at formerly black schools is proof of that; in addition, Table 3 shows that,
on the basis of reported occupations, these parents were from the higher social
levels of South African society, members of the more "elite" classes, especially
when one bears in mind the legacy of apartheid, which precluded black people
from rising much higher than blue collar jobs until the mid 1980s.
Table 3. Occupations of questionnaire respondents
Father (n=181) Mother (n=180) Overall

Business 21% 8% 15%
Teacher/lecturer 19% 37% 28%
Professional 7% 24% 16%

Blue-collar 26% 15% 21%
Labour 3% 6% 5%
Absent/no work 25% 9% 17%
Responses to the questionnaire revealed that their decision to enrol their

children at English schools had been strongly motivated by economic factors of
one kind or another. In supplying reasons for their decision (altogether 360 rea-
sons were offered), 26% (95) of respondents felt that English-medium schools
offered their children better educational opportunities, more sports and cultural
facilities. The next most popular reason, offered by 19% (70) of informants,
was that English is an international language, "the most important world lan-
guage" and would prepare the child to be competitive in the modern world.
Others (15%) (56) mentioned the important role played by English in South
230 Vivian de Klerk
Africa per se, and said that English would open the door to more job opportun-
ities and equip the child with a competitive edge, since it is the "language most
used in the workplace" and the language of science and technology. For this
reason it was vital to be able to speak and write English fluently (16%) (56).
Several mentioned wanting to give the child a better chance in life than they
had had themselves, a chance that some, for financial reasons, could only offer
to the first-born child in the family (3%) (9), often in order to have "at least one
person in the family speaking English".
Parallel to the trends reported in the questionnaires, the majority of the par-
ents interviewed had made the decision in order to provide their children with
better opportunities in life (7), and especially with chances for a better edu-
cation (8) because of the better discipline and facilities at these schools: "The
opportunities are so much greater if one is educated in English. It is an inter-
national language, the language of business and government, of education of
everything. I want my children to have more chance than I did".
6.2. Institutional support
The significance of societal institutions in linguistic and cultural maintenance

has long been recognised (Fishman etal. 1966; Holmes etal. 1993: 15), and
studies have shown that the extent to which the language of a minority group is
represented in various institutions of a nation, region or community, and is ac-
tively used in government, church or cultural organisations is a powerful factor
in determining the pace of language shift.
It was clear from many comments that part of the reason for choosing Eng-
lish schools was dissatisfaction with conditions at local Xhosa schools. While
the legislation is in place to uplift and support Xhosa, no noticeable difference
is evident in most formerly black state schools, where the legacy of apartheid
lingers on, and inefficiency and incompetence exacerbate an already inad-
equate educational system.
In contrast, religious habits reported by interviewees suggest some stability
among the parents, but the shift to a preference for English seems well under
way in the second generation: while 15 of the 26 interviewees reported that
their families worshipped in Xhosa, and 11 in a mixture or English only, the
following comments show resistance from some children: "But when we pray
for food, she prays in English - only her, the rest of the family prays in Xhosa";
"She struggles to understand the Xhosa they use in church"; "she doesn't go to
Sunday school because she doesn't like singing in Xhosa"; "we are having
problems at church. You know these children learning English they don't know
the songs that we are singing. They can't read. When we are singing we used to
sing Xhosa so they just hear from us but they can't read from the hymn book".
6.3. The educational environment
Studies (e.g. Veltman 1983; Genesee 1989; Harres 1989; Pütz 1991; Holmes
et al. 1993) frequently report the vital role played by education (especially
medium of instruction) in language shift, and it is therefore important to inves-
tigate the formal linguistic experiences of the children in this study. In 22 of the
26 families interviewed, the child had first acquired Xhosa as mother tongue,
with English encountered later through creches and preschools (18), or at
school (4). The other four learned some English earlier, often from foreign
neighbours or friends. Most acknowledged that the child's use of and under-
standing of English at the point of starting formal schooling was fair to weak,
with only six claiming that by then the child's competence was "good". Typi-
cally, parents attempted to expose their children to English as early as possible,
often by sending them to a creche or pre-school prior to the commencement of
formal schooling (which is usually at age 6). Overall, only 16% of all respon-
dents sent their children to English schools after the age of ten (see Table 4).
Table 4. Average age when the child was sent to an English school (n=191)
Age 2 3 4 5 6 7 8 9 10 11 12 13 14
% 3 9 10 17 22 10 6 6 6 6 2 0 2
6.4. Education and literacy levels
Although many studies report a link between levels of literacy and rate of lan-
guage shift (McKinnon 1978; Wen Lang Li 1982; Fishman 1989), with edu-
cational level positively related to retention of MT, there is also a view (see
Pütz 1991) that higher educational levels might actually increase the pace of
language shift. According to Clyne (1985: 31), a "lower educational standard
promotes isolation from the dominant community, and maintenance is likely; a
higher educational background favours assimilation".
This latter view is confirmed in this study: the higher the educational level
the higher the likelihood of a shift to English, and Table 1 reveals a relatively
high professional status among the parents. Responses to questions also
showed a high level of commitment to improving literacy among their children:
232 Vivian de Klerk
when the children started at an English school, their proficiency in English was
reported (n=194) as generally weak (54% weak, 34% moderate, 12% good),
and parents took the following measures to improve their child's English: 41%
encouraged the reading of English, by buying materials, getting the child to
join a library or actually reading to them; 31% arranged extra English lessons
of various kinds for their children, helped with homework and provided more
practice in writing and spelling; more opportunities to speak English at home
or with mother-tongue speakers was seen as desirable by 19% of respondents;
only 1% mentioned using the MT to get concepts across. Interviewees men-
tioned parallel practices, 20 of them reporting an active interest in and partici-
pation in the school life of the child, indicative of a high level of commitment to
success ("just to show that you care and it gives an impression to the school. It
also sends a message to the teacher").
6.5. Linguistic networks
The demographic distribution of speakers of the MT and target language (TL)

are usually good indicators of the likelihood of language shift (Lieberson and
McCabe 1982: 83), which explains why migration often leads to language shift
(Timm 1980; Dorian 1980; Tabouret-Keller 1968). However, numerical
strength within communities is an ambivalent factor, depending very much on
network strength: dense and frequent social interaction with MT members is
vital to language maintenance (Lieberson and McCabe 1982: 83; Wen Lang Li
1982; Clyne 1985; Schumann 1986; Pütz 1991; Holmes etal. 1993), and simi-
larly, strong TL networks are likely to speed up shift, especially in cities, due to
urban dwellers' greater immersion in the host environment and consequential
declining MT network strength (Fishman 1972: 126).
In this study, it became clear that the establishment of networks in the TL
was proving very difficult, but that MT networks were also suffering. Question-
naire responses suggest minimal levels of contact with English outside of
schools, despite the fact that several informants (2%) (7), obviously wanted
their children to mix with an English world and TL speakers naturally and
easily: "she must learn English from first-language speakers"; "I didn't want
my child to be embarrassed when speaking to an English person".
Results showed that the language of the children's friends was overwhel-
mingly Xhosa (64%), or mixed Xhosa and English (14%) with only 22% of
them claiming to have English friends. Most of the parents interviewed also re-
ported that their children's friendships were mainly with Xhosa (5) or mixed
Xhosa and English (10) children: "many of her friends at school are English,
she has Xhosa friends in the location"; "she has friends that are both English
and Xhosa (...) all play together. But the white children don't come to our home
and she doesn't go to their's - except for birthday parties"; only six claimed
that their children's friends were primarily or only English-speaking, and these
tended to be children who did not live in the township: "he has no Xhosa-
speaking friends, his life is very much in the school"; "Most of his friends are
English-speaking, from the school that he goes to. He doesn't have any Xhosa-
speaking friends, and none in the township where he lives." Such reports sug-
gest considerable social conflict and isolation. One mother had even made a
concerted effort to prevent her child from talking Xhosa: "she must spend as
many weekends in town with her English friends, even if she stays there all
weekend".
Parents' friends were also predominantly Xhosa-speaking, with English ac-
quaintances mainly colleagues at work ("just people that you phone, you phone
each other, not really visits"). However, seven reported occasionally using
English in interactions with their Xhosa friends "when we discuss 'is-
sues' - political, educational, articles - we switch to Xhosa when we make
some jokes." It would seem from these results that, although the desire for in-
tegration is strong, some difficulty is being experienced in making sufficient
contacts with MT speakers. Part of this problem lies in the changing de-
mography of the formerly white state schools, in which, as Table 5 shows, Eng-
lish children are becoming a numerical minority in some schools.
Table 5. Average proportion of non-English children in local schools
1997 1998
State Schools: School A 28% 46%

School Β 39% 41%
SchoolC 65% 65%
Private Schools: SchoolD 22% 28%

School Ε 27% 22%
SchoolF 20% 19%
Further evidence of problems in building up networks came from reports from

seven of the parents interviewed, who said that they or their children had ex-
perienced social problems, ranging from a lack of integration at schools to re-
jection in the local community: "Children separate themselves into black and
234 Vivian de Klerk
white, Xhosa and English"; "the problem is that she has no friends here who
she speaks English to. The ones she does have, who speak English, live far
away, so it's very difficult for her to get experience and the accent and to be
fluent"; "our children have problems: the township children they say things,
they don't like those who speak English, they reject them. So our children,
when they are in the township, they stay at home, they don't go to friends
there - their friends are at the school"; "[the community] have a lot to say about
us, so I have decided to stay in the house. They are jealous, but I have decided it
is because of the ignorance (...) they are not nice to me (...) they say: "She
thinks that she is better"; "She sometimes speaks English to her Xhosa friends,
they [she and her other English friends] try to put on accents to brag to them.
They try to be 'mistresses'. But, she mostly speaks Xhosa to them"; "she
doesn't feel like speaking English to her Xhosa friends because when she does
they think she is bragging."
6.6. Language attitudes
Language shift is closely linked to language attitudes, since these influence an

individual's motivation to learn a second language (Spolsky 1989:149; Stauble
1980:43). Learners must want to become members of the new community (Fa-
sold 1984: 240). This requires motivation, a positive attitude and ego-permea-
bility (Schumann 1986; Gardner 1985:10). Acquiring new linguistic habits in-
volves learning the beliefs, attitudes, values and other behavioral patterns of
the TL group, and can be facilitated by congruence and similarity between two
cultures (Schumann 1986; Clyne 1985; Spolsky 1989: 143). On the other hand,
political or government pressure (Lieberson etal. 1982: 83) can inspire more
efforts towards language maintenance.
The fact that 78% of respondents reported that making the decision to send
their offspring to English schools had been "very easy", 15% rated it as mod-
erately difficult, and only 7% admitted to some soul-searching, suggests a very
positive attitude to the TL, confirmed in the following comments: "we had no
worries at all about educational problems, or about losing Xhosa - not at all!";
"the advantages were so obvious, it was exciting."
The children of 24 interviewees were also reportedly positive, happy and
excited at going to English schools, with only two feeling scared or nervous;
the following comments speak volumes: "When V refuses to do what I want, I
threaten to send her to one of the township schools"; "her mother joked with
her and said she was going to [Xhosa school] and she cried and she said she
wanted to go to [English school]." Confirming this generally positive attitude is
the fact that 90% of questionnaire respondents (n=167) reported that they were
satisfied with their child's progress and 84% said they did not regret their deci-
sion, 4 of them taking the trouble to explain that the opposite was the case: their
offspring had blossomed and were happy and confident, and that they were de-
lighted. A similarly high proportion (22) of the parents interviewed had no re-
grets whatsoever, and reported general satisfaction with their children's prog-
ress and general confidence with English (15 strongly positive, eight fairly
positive and only three worried): "his English is just fine - wonderful, we can't
tell him apart from an English speaker"; "he is more confident in English than
in Xhosa"; "she sounds just like an English speaker all the time." Confirming
these high levels of satisfaction, 25 of the 26 interviewees said that if they were
consulted by other parents concerning a similar school change for their
children (eight already had been) they would strongly encourage it, the sooner
the better ("I said she must start young, even at pre-school to get a good start";
"start speaking English with the child at home"; "the sooner the better - you
can only win").
Although 81% (n=187) of all the questionnaire respondents, and 17 of the
26 interviewees reported that they believed that it was important to support and
maintain the MT, they certainly were not prepared to lead by example: re-
sponses to other questions revealed that many of the 81%, despite their prin-
cipled support for Xhosa were making every effort to increase their child's ex-
posure to English, not Xhosa. The remaining 19% did not believe it was worth
the effort, and when one considers that these are the parents of the more privi-
leged classes of the Xhosa community, 19% is not an insignificant number. In
investigating positive responses to the role of Xhosa more deeply in the inter-
views, it emerged that parents felt it was important to cultivate Xhosa only for
limited functions, specifically for use at home and for cultural activities (men-
tioned by 15 interviewees): "We have those feasts and things, that we have to
keep cultural identity"; "We must try to keep up, we must try to keep doing our
culture"; "I like my language - my cultural identity"; "as the mother I want her
to know our language, that is Xhosa as well because of our culture"; "I think
it's very important to make the child to be proud of what she is (...) her culture,
where she comes from, she mustn't leave it behind"; "You must know where
you belong, you must know your roots, where you come from"; "To keep the
heritage because it's important for children to know their culture".
This theme of concern about language loss involving cultural loss also came
up among those parents (16%) who admitted to having some regrets, in retro-
spect, about having sent their children to English-medium schools. 60% of
them mentioned their fear of losing their language and culture, which could re-
sult in marginalisation: "they tend to lose identity"; "It's sad that she won't be
236 Vivian de Klerk
able to read and write Xhosa"; "it's sad that her own mother-tongue, she misses
out on that." Cost was also a factor for 19% of the parents. The remaining 13%
of respondents with regrets admitted that their children were experiencing aca-
demic or social difficulties which had not been anticipated.
8% of the respondents reported that people had actually tried to persuade
them that they were doing the wrong thing. Of those interviewed, five families
had experienced resistance of some kind prior to their decision: "people in the
township say it is letting down the culture, that it is bad, that you are a traitor.
They don't like it"; "neighbours, friends, they always talk of confusion and cul-
ture and losing Xhosa (...) they actually came and talked about it".
6.7. Language status
The status of a language rises and falls with the status of its speakers. With the
USA at the zenith of its economic and cultural hegemony, English thrives as the
world's lingua franca and threatens the status of minority languages, increas-
ing the likelihood of shift to English (Clyne 1985).
The prestige of English was recognised by several informants (3%) (9) as an
important factor: "English is a better language", "English is never inferior";
others believed it would bestow certain social advantages on their children
(3%) (9): "Children who learn English become assertive and confident"; "they
develop self-esteem and self-discipline". However, this status had its down-
side, as the following comments reveal: "a white farmer told me 'you think
you're white because you have children at [English] school'"; "others labelled
me as someone who has money, saying I can afford to send my child to that
school - just because you are trying to improve your life." In contrast, parents
also reveal their deep awareness (often tinged with regret) of the low status of
Xhosa, despite recent efforts to improve the status of indigenous languages.
Xhosa carries with it the negative associations born of over a century of racial
discrimination which had inevitable linguistic side-effects.
6.8. Language Functions
If a language has a wide range of societal functions, its instrumental appeal will
outweigh that of a language whose functions are restricted to only a few do-
mains, retaining only a symbolic or ritualised function (Hofman and Cais
1984: 151).
The view that Xhosa had limited uses in South Africa was a frequent theme
in the interviews (10 out of 26). Many wanted Xhosa only for basic communi-
cative purposes within the family: "some of my relatives can't speak English, it
is important for him to be able to talk to them"; "they have to speak Xhosa to
their grandparents"; "She must have her language to speak to me and the
family." These respondents were not strongly enthusiastic about the need to
preserve and build up their language, often because they felt that simply speak-
ing Xhosa (sometimes rather badly) was sufficient: "on the Xhosa side it
doesn't matter if she has hiccups when she speaks it because people would still
understand her. She doesn't have to be that good in Xhosa but being able to
speak English is our main goal"; "There's nothing wrong with not reading
Xhosa, if she can speak it at home it's fine. It is enough"; "Xhosa is right for the
sake of culture, but wrong for the sake of communication".
Nine of the interviewees were actually of the firm opinion that the language
was doomed, five of them expressing some sadness over this, and others being
unemotional and pragmatic: "it's fine to let it [Xhosa] die. We have never teach
our son any Xhosa"; "I don't think there will be a need to be a Xhosa-speaker
later on"; all of these interviewees felt that it was not worth the effort, since it
was ultimately restrictive in South Africa: "I am sad for Xhosa. We will need to
make special acts to uplift the language and yet, I don't know if it is worth it - if
you get educated in Xhosa, up to tertiary level - what then? Who can you com-
municate with? nobody else will understand you"; "It's like a cultural iden-
tity" - not much more than that"; "Xhosa is just a language. I'm not really at-
tached to it."; "It's dying. Xhosa, it can't help you most of the time"; "It will be
a language people speak in their homes"; "It is good for home use but it is not
that good for being used elsewhere. It's more like a home appliance"; "I am
glad I am a Xhosa speaker (...) I don't regret, but at the same time I don't see it
taking me any further. I am a Xhosa, but I can't use it anywhere else; Xhosa, it
cuts you off".
Some had given up the battle: "I would like her to see herself as a Xhosa-
speaking, but already I can say she sees herself as an English-speaking person
and there is nothing I can do"; "she can't speak her mother tongue and she's
going to be a disgrace to her nation".
6.9. Mass media
The powerful role of the mass media in language shift has recently attracted in-
creasing attention (Baetens Beardsmore 1984; Renz 1987; Pütz 1991), and
there is a growing awareness of the all-pervading potential of cultural and lin-
238 Vivian de Klerk
guistic values transmitted by the media. Availability of written materials in the

MT, especially newspapers and magazines, is also recognised as a very import-
ant factor in influencing the pace of language shift (Holmes etal. 1993: 15).
Yet, despite the availability of Xhosa magazines and of Xhosa TV and
radio, 13 of the parents interviewed said they "never" bought any Xhosa read-
ing material ("If I buy Bona I buy the English one, I find it much easier to
read"), and as far as TV viewing was concerned, only 4 preferred Xhosa, and
the rest expressed a preference for English only (14) or a mix (8). There are
several possible explanations for this: firstly, the lack of quality newspapers
and magazines in Xhosa is genuine; these parents, being educated and more
discerning, would therefore probably prefer English reading material; in addi-
tion, international films and programmes are in English; choosing the Xhosa
channel on radio or TV usually means choosing a chat show or poor quality
local music. Until better-quality Xhosa media become available, English media
will continue to have strong appeal.
6.10. Gender
According to Bratt Paulston (1994: 13) language shift often starts with women,
because of their sensitivity to issues of power, especially the language of
power. In contrast, there is a view (Harres 1989: 398) that women (probably be-
cause of their role as housewives and mothers) maintain the LI longer. In this
study, mothers (42% of the 190 parents) came out strongly as being keenest to
make the change of schools and influential in making this decision, with only
4% of the fathers reported as the "keen" one, and 54% reporting a joint enthusi-
asm. Among the parents interviewed, mothers (seven versus three fathers)
again played a leading role in guiding the decision (four couples claimed to
have made a joint decision, four had been persuaded by relatives and 6 by
teachers); the conservatism of the males is evident in the following comments:
"my husband was hesitant; he didn't want them to lose their language"; "my
husband is very indigenous, he doesn't have very many ambitions for his
children"; one mother reported that although she tried to speak as much Eng-
lish as possible to her children, her husband refused to speak it, "not because he
can't speak English, his English is very good".
This trend is completely contrary to that reported in the literature, in which
the females, because of their housebound status, play the more conservative
role; it is more in line with Labov's (1972: 303) view that "it is likely that the
rate of advance and direction of language change owes a great deal to the
special sensitivity of women to the whole process", hence their position in the
vanguard of social change. We are seeing in this particular study, an elite group
of women (many of them single parents), nearly all of them working, ambitious
for their children, who see English as the key to social advancement.
7. Conclusions
Most of the studies available on language shift have focussed on immigrant

communities, who find themselves immersed in a community speaking a lan-
guage (usually English) which is economically and numerically dominant. The
South African situation is somewhat different: speakers of indigenous lan-
guages have not necessarily relocated anywhere (although a few may have
moved from rural to urban areas), but their environment has changed, and Eng-
lish, while far from being a numerically dominant language, holds other kinds
of power and attractions, which have consequences in terms of the language
choices parents have begun to make.
Lieberson (1980) warns of the danger of over-interpreting surveys, and of
basing causal inferences on such studies. Basic epistemological factors make it
extremely difficult to measure sociolinguistic change from a survey and to infer
causal patterns - they tend to be one-off affairs, and must be regarded as such.
The closest one can come to comments on change would come from close at-
tention to differences between age cohorts within the sample: examining the
sociolinguistic attributes of age, and cross-tabulating language characteristics
of parents with children to make a rough estimate of the rate of MT shift be-
tween generations.
On the basis of the evidence in this study it would seem that, among the
children of these parents, shift to English is well under way, and is almost ir-
revocable; in terms of the ten variables which were explored in this paper, some
have emerged as more prominent than others, but it would be foolish to venture
to rank them categorically. However, they divide themselves into two groups,
which influence the decision makers to differing degrees.
The first set of factors which get the ball rolling down the slippery slope of
language shift seem to be primarily economic and functional factors, in tandem
with high educational levels among parents: language shift has to start from
somewhere, and in the Grahamstown context, the seeds are sown in the minds
of the parents, not in the children whose language will ultimately shift. The
parents who made this crucial decision tend to come from the better-educated
and wealthier sectors of society, where they have seen the dividends which
come from an "investment" in English. Economic and educational factors, all
tied in to the relative statuses of the languages concerned, featured prominently
240 Vivian de Klerk
in their decision-making, as did prospects for future employment for their

children. All were ambitious parents, whose children are likely to be members
of the elite in this country in the future. The choice of English by these parents
is clearly related to socio-economic and political processes, and to the distribu-
tion of knowledge and power vis a vis English, and this will have a significant
effect on the future function, status and development of Xhosa (Dua 1996).
Once the decision has been made, the next set of factors comes into play in
determining the relative 'success' of the shift, and these are socio-cultural.
Prime among these is the educational context in which the child is placed.
While formerly whites-only English government schools have rapidly changed
as far as racial and linguistic profile goes, and pupils from a very wide range of
linguistic and cultural backgrounds are now in the same classes, the teachers
are still predominantly English-speaking, the MOI is English and the prevail-
ing ethos is still western, Christian, English and white. For many, 'open school-
ing' simply means "allowing black students into formerly white schools with-
out adapting the syllabuses, the teaching methods or patterns of classroom in-
teraction in order to facilitate them" (Gordon and Barkhuizen 1994: 74). The
aim of teachers and parents alike is assimilation and integration (Baker 1993:
253), in the belief that educational success is only possible through mastery of
English, which is seen as giving access to social and educational mobility and
advancement to native and non-native users who possess it as a linguistic tool.
Such a context accelerates the pace of language shift. (As Table 5 suggests,
however, in state schools non-English speakers are becoming the majority, and
the situation is likely to deteriorate as the number of highly trained MT English
teachers in South Africa is rapidly declining and there are simply not enough of
them to go around any more).
Further acceleration of language shift comes from positive attitudes to the
target language and its speakers, attitudes which are shared by parents and off-
spring, and which relate to more than the functional value of English. Com-
ments have shown indisputable strong support for English from these parents,
whose power to influence their children's attitudes is likely to be significant.
Further impetus comes from the practical measures taken to facilitate the lan-
guage shift, both in terms of educational assistance and media support. The
prediction that inadequate institutional support for the MT will speed up the
pace of language shift has been borne out in this research: despite the elaborate
measures taken by local and national administrations to support Xhosa (dis-
cussed earlier), despite the high level of awareness of the importance of the in-
digenous languages, and despite the availability of Xhosa schools and other in-
stitutional facilities such as churches etc., the practical wherewithal is lagging
way behind: there is a severe lack of well-trained Xhosa-speaking teachers, an
appropriate curriculum and textbooks (see Crawhall 1994; Ffolliott and Stear
1995). The old and inappropriate language syllabuses and the unacceptable
teaching methods of apartheid have survived. In addition, quality literary re-
sources in Xhosa are limited and provincial funding is at an all-time low. While
there is a plethora of Xhosa schools available, most of these schools offer a
third-rate education as a result of decades of under-resourcing and appalling
provision of teacher training for black students, as well as political unrest. Par-
ents who can, are seeking alternatives, alternatives which at present necessitate
abandoning the principle of MT education. By the time institutional support for
Xhosa is sufficiently impressive and reliable, the elite among this community
will probably already have shifted allegiance to English completely.
The final factor which is commonly associated with rapid language shift has
not played its expected role in this research: the establishment of social net-
works to support new linguistic habits and speed up cultural assimilation is
highly problematic in the local Grahamstown context for various reasons
which have been explained earlier. Yet this does not seem to matter too much to
these informants at this stage; mind prevails over matter, a determination to
master English and a will to succeed seem to be sufficient. More to the point,
the commitment to maintain the MT is notably absent in most cases, and there-
in lies the rub: if the home language and culture are strong enough, bilin-
gualism and biculturalism will result, but loss to English monolingualism will
result from insufficient maintenance of the MT (Harres 1989; Grosjean 1982).
Many of these families, prior to their decision, were virtually monolingual
families, which, instead of becoming bilingual, are in the process of replacing
Xhosa with English in more and more contexts.
The major task of sociolinguistics is to reconcile the essentially neutral or
arbitrary nature of linguistic difference and linguistic change with the social
stratification of languages and levels of speech unmistakable in any complex
speech community (...) we might term this the study of how social processes
seize upon linguistic disparities that are intrinsically symmetric and upon lin-
guistic processes constantly operating in any language, and systematically ma-
nipulate them into a highly structured system of speech varieties which mirrors
and reinforces social class and power distinction (Sankoff 1980: 5).
It is hoped that this research throws some light on the factors influencing the
process of linguistic change in one small group of Xhosa speakers in the East-
ern Cape. While this small elite group appears to be going counter to the cur-
rent ethos of the African renaissance, and could well be accused of being trai-
tors to their own cause, one needs to consider what 'democratic language
rights' are. For some, this implies the need to recognise and promote the loca-
lised languages as equally viable and effective as any other language. For
242 Vivian de Klerk
others, it implies the right to acquire competence in a model of English that has
international currency and will afford them the advantages that others, the
world over, enjoy. While these parents recognise the intrinsic value of Xhosa in
meeting the immediate communicative needs of local speakers, they also see its
potential to restrict them in wider linguistic contexts, limiting their comprehen-
sibility and their opportunities to participate in the global village on an equal
intellectual and economic footing with speakers of English which has currency
world-wide.
Repeated longitudinal studies on a larger scale than this survey will be
necessary to establish the rate and future pattern of language shift in the prov-
ince.
References
Aitchison, Jean
1991 Language Change: Progress or Decay? (Second edition.) Cambridge:
Appel, Rene and Pieter Muysken

1987 Language Contact and Bilingualism. New York: Routledge and Chap-
man and Hall.
Baker, Colin
1993 Foundations of Bilingual Education and Bilingualism. Clevedon:
Beardsmore, H. B. (ed.)
1984 Language and television. International Journal of the Sociology of
Language 48: 5-10.
Bowerman, Sean
2000 Linguistic imperialism in South Africa: the unassailable position of
English. Unpublished Masters dissertation, University of Cape Town.
Clyne, Michael
1984 Multilingual Australia. Melbourne: River Seine Publications.
Crawhall, Nigel
1994 Serving the people's interest. Bua! 9 (2): 8-10.
de Klerk, Vivian
1996 Use of and attitudes to English in a multilingual university. English
World-Wide 17 (1): 111-127.
* 1997 Encounters with English over three generations in a Xhosa family: for
better or for worse? In: E. W. Schneider (ed.), Englishes around the
World 2, 97-118. Amsterdam: John Benjamins.
in press Language issues in our schools: whose voice counts? Part 2: The
teachers speak. Perspectives in Education.
de Klerk, Vivian and G. Barkhuizen

1998 English in the South African Defence Force: a case study of 6SAI.
English World-Wide 19 (1): 33-60.
2001 Language usage and attitudes in a South African prison: who calls the
shots? International Journal of the Sociology of Language 152:
97-115.
Denison, N.
1977 Language death or language suicide? International Journal of the
Sociology of Language 12: 13-22.
Dorian, Nancy
1980 Language shift in community and individual: the phenomenon of the
laggard semi-speaker. International Journal of the Sociology of Lan-
guage 25: 85-94.
Dow, James R.
1987 Introduction. International Journal of the Sociology of Language 68:
5-6.
Dua, H. R.
1996 The politics of language conflict: implications for language planning
and political theory. Language Problems and Language Planning 20
(1): 1-17.
Eastern Province Herald

6/6/1997
Fasold, Ralph
1984 The Sociolinguistics of Society. Oxford: Blackwells.
Ffolliott, E. and N. J. Stear

1995 Motivation for urgent attention to be given to addressing the problems
and anomalies existing in the present state-aided schools with respect
to implementing the new language policy. Unpublished manuscript.
Fishman, Joshua Α., V. C. Nahirny, J. E. Hofman and R. G. Hayden

1966 Language Loyalty in the United States: The Maintenance and Per-
petuation of non-English Mother Tongues by American Ethnic and
Religious Groups. The Hague: Mouton.
244 Vivian de Klerk
Fishman, Joshua A.
1972 The Sociology of Language. Rowley, Mass.: Newbury House.
*1989 Language and Ethnicity in Minority Sociolinguistic Perspective.
Clevedon, Avon: Multilingual Matters.
Gal, Susan
* 1979 Language Shift. Social Determinants of Language Change in Bilin-
gual Austria. New York: Academic.
Gardner, Robert C. and Wallace E. Lambert
1959 Motivational variables in second language acquisition. Canadian
Journal of Psychology 13: 226-72.
Gardner, Robert C.
1985 Social Psychology and Second Language Learning: The Role of Atti-
tudes and Motivation. London: Edward Arnold.
Gardner, Robert C. and R D. Maclntyre
1992 A student's contributon to second language learning: part 1: cognitive
factors. Language Teaching 25: 211-220.
Genesee, Fred
1989 Early bilingual development: one language or two? Journal of Child
Language 16: 161-177.
Gordon, F. and G. Barkhuizen
1994 Alienation and conflict in an open primary school: a multilingual set-
ting. Southern African Journal of Applied Language Studies 3 (2):
60-76.
Gorter, D.
1987 Surveys of the Frisian language situation: some consideration of re-
search methods on language maintenance and language shift. Inter-
Grosjean, Francis
1982 Life with Two Languages. Cambridge, Mass.: Harvard University
Press.
Harres, A.
1989 Being a good German: A case study analysis of language retention and
loss among German migrants in North Queensland. Journal of Multi-
lingual and Multicultural Development 10 (5): 383-399.
Heugh, Kathleen
2001 The case against bilingual and multilingual education in South Africa.
PRAESA occasional papers. University of Cape Town.
Hofman, J. E. and J. Cais

* 1984 Children's attitudes to language maintenance and shift. International
Holmes, J., M. Roberts, M. Verivaki and A. Aipolo

1993 Language maintenance and shift in three New Zealand speech com-
munities. Applied Linguistics 14(1): 1-24.
Kamwangamalu, Nkonko
1998 Identities of English and codeswitching in post-apartheid South Afri-
ca. Multilingua 17 (2 / 3): 277-296.
Labov, William
1972 Sociolinguistic Patterns. Philadelphia: University of Pennsylvania
Press.
Lambert, Wallace E. and D. M. Taylor

1996 Language in the lives of ethnic minorities: Cuban American families
in Miami. Applied Linguistics 17 (4): 47-64.
Li, Wen Lang

1982 The language shift of Chinese-Americans. International Journal of
the Sociology of Language 38: 109-124.
Lieberson, Stanley
* 1980 Procedures for improving sociolinguistic surveys of language main-
tenance and language shift. International Journal of the Sociology of
Language 25: 11-27.
Lieberson, Stanley and E. McCabe

1982 Domains of language usage and mother-tongue shift in Nairobi. Inter-
McKinnon, Κ. M.
1978 Language shift and education: conservation of ethnolinguistic culture
amongst schoolchildren of a Gaelic community. Social Sciences Oc-
casional Papers no. 2. School of Sciences, Hatfield Polytechnic.
Mesthrie, Rajend
1996 South African Indian English. In: V. de Klerk (ed.), Focus on South Af-
rica, 79-98. Amsterdam: John Benjamins.
Pandor, N.
1995 Constitutional multilingualism: problems, possibilities, practicalities.
Proceedings of the Southern African Applied Linguistics Association
Conference 15: 57-74.
246 Vivian de Klerk
Paulston, Christina Bratt

1994 Linguistic Minorities in Multilingual settings: Implications for Lan-
guage Policies. Amsterdam: John Benjamins.
Pütz, Martin
1991 Language maintenance and language shift in the speech behaviour of
Germain-Australian migrants in Canberra. Journal of Multilingual
and Multicultural Development 12 (6): 477-492.
Renz, Β. Β.
1987 Portuguese-language broadcasting in linguistic and cultural mainten-
ance in northern California. International Journal of the Sociology of
Language 68:23-40.
Sachs, Albert
1994 Language Rights in the New Constitution. South African Constitution
Studies Centre, University of the Western Cape.
Sankoff, Gillian
1980 The Social Power of Language. Philadelphia: University of Pennsyl-
vania Press.
Schumann, John H.
* 1986 Research on the acculturation model for second language acquisition.
Journal of Multilingual and Multicultural development 7: 379-92.
Schuring, G. and K. Calteaux

1997 Are black languages dying in urban areas? Bulletin. News for the
Human Sciences 4(1): 16-17.
Smeija, Birgit
1998a Language shift, cultural identity, and language loyalty in Botswana. A
research (sic) on language use and language attitudes. Unpublished
manuscript.
1998b Codeswitching and Language Shift in Botswana: indicators for lan-
guage change and language death? A progress report. Unpublished
manuscript.
Smith, J. F.
1977 Language use and language attitudes in a bilingual community: Te-
herne, Friesland. PhD, Stanford University.
Smolicz, J. J.
1984 Minority languages and the core values of a culture: changing policies
and the ethnic response in Australia. Journal of Multilingual and
Multicultural Development 5(1): 23-41.
Spolsky, Bernard
1989 Conditions for Second Language Learning. Toronto, Ontario
Stauble, A. M.
1980 Acculturation and second language acquisition. In: R. C. Scarcella
and S. D. Krashen (eds.), Research in Second Language Acquisition,
43-50. Rowley Mass.: Newbury House.
Tabouret-Keller, Andree
1968 Sociological factors of language maintenance and language shift: a
methodological approach based on European and African examples.
In: J. Fishman, C. Ferguson and J. Das Gupta (eds.), Language Prob-
lems of Developing Nations, 107-18. New York: John Wiley.
Taighde, Y.
1990 Language Shift among Schoolchildren in Gaeltacht Areas 1974—1984.
The Linguistic Institute of Ireland.
Timm, L.
1980 Bilingualism, diglossia and language shift in Brittany. International
Veltman, Calvin
1983 Language Shift in the United States. Berlin/New York: Mouton de
Gruyter.
Weinreich, Uriel
1979 Languages in Contact. Berlin/New York: Mouton de Gruyter.
1. Language shift occurs when linguistic communities find themselves in con-

tact with a language which offers greater practical and economic rewards or
carries higher prestige. The parents in this study could be said to have"in-
vested" in English, rather than Xhosa, which one parent said "is good for
home use but it is not that good for being used elsewhere. It's more like a
home appliance". Discuss the relative economic value of English and
Xhosa, and the role this factor is likely to play in language shift.
2. Language rights have achieved much prominence in South Africa over the
past 5 years. To what extent could it be argued that the language rights of the
children in this study are being disregarded by their parents and teachers?
248 Vivian de Klerk
3. The following comments were made by interviewees. Discuss each one in

turn, seeking a deeper understanding of what lies behind the words:
a. "At first we tried to speak Xhosa, but you can't help it, the children know
the English words, and their life at school it is English."
b. "Her accent is that of a first-language speaker, when we hear her on the
phone we are so proud, we cannot believe it."
c. "She has friends that are both English and Xhosa (...) all play together.
But the white children don't come to our home and she doesn't go to
their's - except for birthday parties."
d. "When V refuses to do what I want, I threaten to send her to one of the
township schools."
e. "You must know where you belong, you must know your roots, where
you come from."
f. "I am a Xhosa, but I can't use it anywhere else; Xhosa, it cuts you off."
4. It is frequently stated that language shift is invariably preceded by wide-
spread bilingualism (Weinreich 1979; Appel and Muysken 1987: 40), yet
this is not the case in this study. Comment on the possible reasons why this
might be so.
5. Parents in this study have chosen English for their children. However,
Heugh (2001) provides copious evidence of the failure of most children
who have plunged too quickly into English without strong support in the
school for their home language, especially those whose home language has
a lower status than English. In a country where 83% of South African pupils
are African-language-speaking, come from poor socio-economic back-
grounds and are effectively in monolingual (African language) schools
(Heugh 2001 quotes a figure of 92%), bilingual education would seem to be
the obvious solution, in order to ensure a good education and access to Eng-
lish. What do you think the ideal solution for the majority of South Africa's
children would be?
Japan's nascent multilingualism
Florian Coulmas and Makoto Watanabe
1. Introduction
Japan has been well-known for its ethnolinguistic homogeneity. Ever since the
Meiji reforms, ethnolinguistic homogeneity has been cultivated as a pivotal part of
Japan's self-image. As a result, the unity of nation, state and language has been
long taken for granted. The state functioned exclusively in the national language.
However, Japan's ascent to become an economic power after the Pacific War
brought in its wake more social diversity than anticipated or desired by the politi-
cal establishment. Though Japanese is overwhelmingly dominant in the society
bilingualism has become an issue in recent years. So far, bairingaru is a term with
positive connotations. A bairingaru is thought to be a person who speaks English
or, perhaps, another European language. The real bilingualism of resident Koreans
and Chinese, of the speakers of Ryukyuan and Ainu, is not what the man in the
street means by bairingaru. Yet, the socioliniguistic landscape of Japan is chang-
ing, especially since an influx of foreigners in the 1980s has reinforced resident
minority communities in Japan and created new ones.
According to conservative figures by the Japanese Ministry of Justice,
about 1.55 million registered foreigners are living in Japan as well as more than
300,000 illegal residents. All told, 1.8 to 2 million non-Japanese nationals are
living in Japan, some 1.5% of the population. Increased immigration to Japan
in the 1990s has turned Japan into a society with growing individual bilin-
gualism and nascent societal multilingualism.
These figures are still relatively low, but the issues raised by what we call
"multilingualization" are beginning to have an impact on communication pat-
terns, institutions and questions of identity. Some politicians sport multilingual
home pages. Radio stations in Tokyo, Osaka, Fukuoka and other cities broad-
cast news programs in several languages. City halls in major industrial areas
employ interpreters for English, Chinese, Korean, Portuguese, and other lan-
guages. All of this is new.
Therefore, at the present time, Japanese society offers an opportunity to
study the transformation of a society operating largely under monolingual as-
sumptions into one which has to come to terms with greater linguistic plurality.
In this paper, we give an overview of the language situation of modern Japan
250 Florian Coulmas and Makoto Watanabe
since the Meiji period (1868-1911); examine the causes of increasing linguistic
variety; and outline a research strategy for assessing Japan's nascent multilin-
gualism.
2. Language and modernization in Meiji Japan
The history of modern Japan is a history of language change. The unification of

literary and vernacular forms of Japanese (Twine 1991), the establishment of a
unified standard (Sanada 1987), and the spreading of the "national language" (ko-
kugo) throughout the country and, later, the colonies (Lee 1996) were important
aspects of Japan's leap into modernity. Language reform and the implementation
of compulsory education were effective measures to supplement, where not to re-
place, the predominantly vertical structures of communication in feudal society by
horizontal structures more suitable for the requirements of industrialization and
the creation of a national public.
2.1. Narrowing the gap between spoken and written language
For many centuries literary and spoken registers and styles developed separately.
By the mid nineteenth century, a number of vastly different function-specific
written styles coexisted all of which were far removed from any variety of spoken
Japanese. This situation was perceived by many writers and educators as an im-
pediment to the dissemination of knowledge considered so important for Japan to
catch up with the West. They recognized the importance for a modern society of a
modern language, one that is suitable for mass literacy and national unification.
Thus the genbun itchi movement for the "unification of the spoken and written
language" came into being. The abolition of Chinese characters wasfirstproposed
by Maejima Hisoka, a scholar and government official, who in 1866, just before
the collapse of the ancien regime of the Tokugawa Shogunate, submitted a petition
to this effect to the government. Other suggestions such as Nishi Amane's 1874
call for the adoption of the Latin alphabet followed. Fukuzawa Yukichi's more
moderate proposal of 1873 to reduce the number of Chinese characters in com-
mon use carried the day. In conjunction with attempts by writers such as Futabatei
Shimei, Yamada Misa and Ozaki Köyö to develop a style that was closer to ver-
nacular speech it led to a notable reduction of the differences between spoken and
written Japanese.
The new style took shape during the three decades before and after the turn of
the century. At roughly the same time this new language was given a new function
Japan's nascent multilingualism 251
as a means of advancing national cohesion. It was during that period that the no-
tion of kokugo or "national language" gained currency. Ueda Kazutoshi
(1867-1937) who had studied linguistics in Germany introduced linguistic
nationalism to Japan (Twine 1991: 163). While linguistic differences corre-
sponded very strongly with the stratification of feudal society in Japan, what he
had gotten to know in Europe opened his eyes to the advantages of languages that
could be used to bridge class differences and create a bond of national community.
His influential book Kokugo no tame [For a national language] published in two
parts in 1895 and 1903 marked the beginning of the political instrumentalization
of Japanese and of a deliberate language policy. The kokugo became the sole
medium of education, even though teacher in many parts of the country were
hardly more proficient in it than their pupils. The new variety developed as it was
disseminated throughout the country. Standardization and language spread went
hand in hand in a process which was remarkably swift and effective. It met with
little resistance, although at a price. Repeating as it were a course of events that
had accompanied the emergence of national languages in European countries, the
kokugo was promoted at the expense of dialects and minority languages. Hogen
fuda (dialect tags) were employed at school to discourage the use of local speech
forms. Unwritten Ainu and Ryukyuan were particularly hard hit, as their speakers
were subjected to strongly assimilationist policies. At the time, however, language
maintenance and language rights were not a part of public discourse on language
which, following European models, was conceptualized first and foremost as a
modern language which united speech and writing and could serve the functions
of a national language.
2.2. Status and prestige of foreign languages in modern Japan
In the course of Japanese history, a limited number of foreign languages played a

special role. First, there was Chinese, for centuries Japan's prestigious written lan-
guage (kanbun). Literacy in kanbun was a prerequisite for elite participation.
Japan was at the periphery of the sphere of China's superior culture absorbing and
transforming many of its accomplishments. The Chinese language was both the
most important medium of gaining and processing information and an object of
cultivation. Buddhism and Confucianism entered Japan in the guise of the Chinese
language. Calligraphy, after the Chinese model, was one of Japan's most highly
regarded art forms. Kanbun was enculturated in the Japanese context and took on
various forms unknown in China, but it continued to function as the preferred lan-
guage of law, administration, and scholarly writing until the upheavals of the Meiji
Restoration. Rather than a foreign language it was considered a necessary part of a
proper education. The high status and prestige Chinese had enjoyed for close to a
thousand years only eroded when China fell victim to Western imperialism in the
nineteenth century and the conservative and fragmented Chinese leadership
showed itself unable to face the challenges of modernization. When Japan's intel-
lectual elites turned away from weakened China towards the powerful countries of
Europe and America, Chinese culture lost its function as a model and kanbun re-
ceded ever more into the background of Japan's literacy practices. A medium
more suitable for mass literacy was needed. For this, Dutch provided a model.
In the sixteenth century, some Japanese converts learned Portuguese from
Christian missionaries, the first Europeans to come to Japan. The first Western lan-
guage that was systematically studied, however, was Dutch. Throughout the Edo
period (1600-1868), especially after the last Portuguese had been expelled in
1639, the Dutch trading factory in Nagasaki served as Japan's only window to the
Western world. Like the Chinese, the Dutch were expected to report to the Japan-
ese government on events in the rest of the world. Dutch studies, rangaku, became
an important endeavour, the oranda tsuji or "Dutch interpreter" became an estab-
lished profession. Towards the end of the Edo period, a polarization took shape
pitting Dutch learning against Chinese learning. Many members of the intellectual
avant-garde studied Dutch, the language which, they thought, held the key to
Japan's future as a modem nation. Chinese, by contrast, was linked to outmoded
ways and a stagnant culture incapable of reforming itself.
It wasn't just the information available through Dutch on medicine, astron-
omy, military technology, and other Western sciences that made it attractive.
There was also a social aspect which Japanese reformers recognized: The
Dutch wrote much like they spoke. Enlightened scholars and politicians such
as Fukuzawa Yukichi (1835-1901) and Mori Arinori (1847-1889), the first
Minister of Education, early on laid a link between Western advances in
science and technology and vernacular literacy. They believed that their own
literacy practices were a stumbling block on the way to successful diffusion of
new knowledge and social reform. Translation from Western languages into
any of the written styles was difficult and gave rise to a sense of language crisis
so acute that even the adoption of a Western language instead of Japanese was
seriously discussed (Coulmas 1990).
The importance of Dutch was subsequently reduced. By the mid nineteenth
century the Dutch had lost their trade monopoly and Japan was forced to inter-
act with other Western powers. As a consequence, other European languages
quickly gained importance, English, German and French, in particular. For a
while Dutch continued to be an important language of international relations,
but it wasn't lost on the Japanese that its standing in the world was not nearly as
high as it had been in their secluded country for the past two hundred years.
English, the language of the country which had pried Japan open with its gun
boats, rapidly became Japan's foremost foreign language. In 1872, at the height
of the Westernization fever which raged through Japan during the early Meiji
years, it was even considered as a possible alternative to Japanese as a national
language. This bizarre discussion foreshadowed other proposals, such as that
for the adoption of Esperanto in the early Showa period (1926-1989) (Hirai
1948: 173-175) and the adoption of French as Japan's national language after
the defeat in the Pacific War (Shiga 1946).
During the war, Japan pursued an active policy of language spread in her
colonies (Kawamura 1994). Some authors anticipated a grand future for Japan-
ese and saw it take a place among the leading world languages (Ishiguro 1941).
After the defeat, this kind of megalomania gave way to an image of Japanese as
an exotic language of a small island country inaccessible and of little signifi-
cance to outsiders. At the same time, seven years of American occupation ac-
corded English the status of Japan's undisputed first foreign language. Its im-
portance has continued to grow ever since (Haarmann 1987; Hildebrandt and
Giles 1980). Science, the media, commerce, and foreign relations rely heavily
on English. More Japanese are proficient in English today than have ever been
able to speak any other foreign language. This development culminated in a
proposal the late prime minister Obuchi Keizo advanced in 2000 to install Eng-
lish as Japan's second official language. Like the earlier suggestions to change
Japan's language regime of 1872 and 1946, this one came at a time of acceler-
ated social change induced by developments both inside and outside the
country.
3. Divergence: forces of change
3.1. Internationalization: English as a second official language of Japan?
Language tends to become a focus of discussion in Japan whenever crisis

looms. While Japan is a very language conscious culture and language is,
therefore, considered very important, it is also commonly used as a stage to
play out dramas with a different content. The downturn of the Japanese econ-
omy following the Asian financial crisis was perceived by many as the second
national defeat after the Pacific War. It is against this background that former
prime minister Obuchi's call for giving English official status must be under-
stood. Some argue that in the absence of public discussion and a clear defini-
tion of what an official language is no second official language can be installed
(Tanaka 2000). This can hardly be denied, and it will take time before any
changes will be effected. In the meantime, however, recommendations to
change the language regime are not meaningless.
On the contrary, they have profound symbolic significance. They articulate
a persistent feeling of isolation in Japan, which, in this age of globalization, is
rightly regarded as a serious problem. They also indicate that language is con-
sidered important and a sense that altering the language regime may be part of
the solution to nonlinguistic problems. Japan is not proficient in English. Some
political and economic leaders see this as a competitive disadvantage that
weakens Japan's international position. The world's No. 1 aid donor and sec-
ond largest contributor to the United Nations' ordinary budget, Japan is badly
underrepresented in major international agencies. In other areas where global-
ization is advancing, particularly international finance, e-commerce, science
and technology, this deficit is painfully felt. It isn't clear that assigning English
official status would be the solution, but the idea nevertheless serves a useful
purpose by provoking debate. And debate is needed because Japan's communi-
cation requirements are changing.
3.2. Migration issues: forces from within
There are two sides to Japan's language regime, an external and a domestic
one. Both need to be reconsidered. At issue is communication rather than lan-
guage. Society at large must be enabled to meet contemporary communication
requirements.
At the domestic front, increased labour migration from abroad poses new
problems. In the June 2000 edition of Monthly Keidanren, the newsletter of the
Japan Federation of Economic Organizations, Keidanren President Takashi
Imai pointed out that it "is indispensable for this country to keep a leadership
position in the global community. It is also hoped that the government will take
up a serious review of immigration policy soon, with a view to securing ad-
equate manpower in new growth sectors to prepare for an inevitable decline in
Japan's population in the coming decades". As this statement indicates, in-
fluential representatives of Japan's economy are clearly aware of the inevitabil-
ity to further open Japan's labour market. Thus, Japan will have to accommo-
date more labor migrants and their dependents. If "adequate manpower" is to
be secured, fluency in Japanese can hardly be taken for granted. Communities
with limited Japanese proficiency are springing up in a society that so far has
operated under monolingual assumptions. Obuchi's proposal is highly relevant
in this connection.
Notice that this is not a problem of the future but of the present time. Con-
sider Kanagawa, a prefecture adjacent to Tokyo, as an example. It is a major in-
dustrial center with two big cities, Yokohama and Kawasaki. According to the
prefectural government, the number of registered foreigners has increased
from 41,266 to 116,535, and the number of nationalities has almost doubled
from 85 to 155 (1975-1999). Nation wide the trend is the same, the number of
registered foreigners having doubled from 774,505 to 1,556,113 during the two
decades from 1979 to 1999 (figures: Ministry of Justice). The major countries
of origin are Korea 636,548 (40.2%), China 294,201 (18.9%), Brazil 224,299
(14.4%), and the Philippines 115,685 (7.4%), followed by the US and Peru.
There is a tendency for foreign workers to live in industrial areas in and around
big cities. Ten prefectures, which include Tokyo, Osaka, Nagoya, Kyoto, Fu-
kuoka and some other big cities share over 70% out of the registered numbers.
In Osaka 2.35% and in Tokyo 2.32% of the population are foreign nationals.
The main cause of the increase in foreign labor was the booming economy of
the 1980s. However, labour migration into Japan continued after the bubble
burst in the early 1990s.
Japan's guest workers constitute the youngest social minority. There are sev-
eral others. Generally speaking, there are two types of minorities: old minorities of
autochthonous groups and newcomers from outside Japan. The Ainu in Hokkai-
do, at the northern periphery, and the Ryukyuans on the islands of Okinawa in the
south, belong to the former group. On the other hand, the latter group are recent
migrants coming from various Asian, and Latin American countries. This division
between autochthonous groups, on one hand, and migrants, on the other, is, how-
ever, not clear with respect to the two largest communities, the Koreans and the
Chinese. Both have a long history of migration to Japan. Overseas Chinese have
settled in Japan in compact communities (chuka-gai or China towns) in cities
such as Nagasaki, Kobe and Yokohama for many generations. Nowadays, these
communities also function as initial footholds for newcomers from Mainland
China, Taiwan and Hong Kong. Initial contact with Koreans also goes back many
centuries when they acted as transmitters of Chinese culture and technology, but
the present-day community of resident Koreans are a remnant of Japanese colo-
nialism. Many of the older generation came to Japan as conscripts or forced labour
during the Pacific War. 1.35 million Koreans were resident at the time of Japan's
surrender of whom some 900,000 were repatriated. Those who remained were
stripped of their Japanese citizenship and turned into alien residents. The Chinese
and Koreans came to and remain in Japan for various reasons and backgrounds.
Among the Chinese there are generational differences. Before the war, they came
from China, while post-war immigrants are mainly from Taiwan, and hikiage or
Japanese "returnees", and recent newcomers are foreign workers and students.
Linguistic and social divisions within the Chinese minority are a consequence of
these migration minorities. Accordingly, there is a broad social spectrum in the
Chinese communities reaching from wealthy merchants and industrialists with
their own school and temples to impoverished day-laborers.
A classification such as that which distinguishing autochthonous and migrant
groups is gradual rather than clear-cut. Immigrants from Vietnam and neighbour-
ing countries now have a second generation presence with their children born in
Japan. Even second generation "war orphans" (zanryu koji) - Japanese children
left behind in Manchuria during the tumultuousfinalweeks of World War Π - have
appeared. In these groups, the voluntary element of membership is evident, but it
also plays a role in other groups. Many find it easier nowadays to be more assert-
ive. Young Koreans, for example, although even more assimilated than their par-
ents, are more inclined than they to use their Korean name rather than the Japanese
name they were forced to adopt as Japanese subjects. In Okinawa and Hokkaido,
various action groups promote local languages and cultures supporting revivalist
movements. On local and regional levels these movements have met with toler-
ance and some support, the central government, however, has shown little interest
in their concerns. Many civil servants in the national administration including the
Ministry of Education, Culture, Sports, Science and Technology are still com-
mitted to the idea of Japan's ethnolinguistic homogeneity.
3.3. The state lags behind
Following a well-known pattern, the Japanese government is more conser-

vative than the society, which is quicker to face the reality of Japan's growing
internal diversity. The same can be said of local administrations. Many local
governments have recently adopted a flexible and pragmatic stance in dealing
with non-Japanese residents and their communication requirements. City halls
and ward offices have established interpretation and counseling services in sev-
eral languages besides English, including Korean, Chinese, Portuguese, Taga-
log, Indonesian and Farsi. Local governments are far ahead of the national gov-
ernment, and understandably so, because this is the level at which community
issues have to be dealt with. The national government, despite paying repeated
lip service to internationalization and globalization, frequently relapses into an
isolationist mentality.
For example, the prefectural government of Kanagawa operates a multilingual
consulting service in English, Korean, Chinese, Spanish and Portugese. In and
around Kobe and Osaka where more than 80,000 foreigners were living when a
devastating earthquake hit the area in 1995, the local administration and NGOs
have developed manuals with specific instructions and information to assist

people with limited Japanese ability in case of emergency and disaster (Sato
1999a, 1999b). The Center for Multicultural Information and Assistance, an NPO
which has grown out of an organization formally called jishin joho senta [Center
for earthquake information] offers services in 15 languages, providing in-
formation about everyday life, medical assistance, legal consultation and edu-
cation. It maintains branches in Osaka, Kyoto, Kobe, and Hiroshima.
In stark contrast to such accommodating measures, the Ministry of Justice
and the Ministry of Home Affairs feel compelled, time and again, to execute
organized deportation campaigns, because illegal immigrants pose new prob-
lems for the government: work injuries without insurance, legal conflicts
among foreign nationals, sham marriages and so on. The government would
prefer to eliminate these problems rather than deal with them. But they won't
go away. Japan's growing diversity is irreversible.
Linguistic diversity calls for an active language policy that makes room for
languages other than Japanese. There is no need to question the predominant
position of Japanese, but many in Japan feel that the fact that there is a resident
population of speakers of a variety of other languages ought to be acknowl-
edged. This is particularly important at the elementary-school level. Right now,
there are youngsters in Japan who do not receive regular education. Compul-
sory education applies to Japanese nationals only. As a result, there are children
of refugees and labor migrants who are growing up without attending school
because the state's responsibility does not extend to them. At the same time,
their parents lack the time and the financial resources to make sure that they re-
ceive at least basic education. Many of them are incapable of guiding their
children through the Japanese educational system even if they wanted to. These
problems have been acknowledged by local school administrations and con-
fronted by informal circles and a number of NGOs, but the national govern-
ment has been slow to acknowledge them. The compulsory-education regu-
lation will have to be updated sooner or later.
One important change was effected in 1999. The government decided to
suspend its long standing policy of refusing to grant Korean-run schools offi-
cial status, thus making Korean senior high school students eligible to sit for
entrance exams of national universities. This was an important step for the Ko-
rean community but has no implication for other groups. The authorities still
find it hard to officially depart from the idea of Japan's homogeneity. Yet the
traditional self-image of Japan as a closed, homogeneous society has begun to
break down, and the push for more pluralism is welcomed and supported by
many Japanese. People's attitudes have changed because socioeconomic con-
ditions have changed and because international currents of political thinking
have changed. The monoethnic nation state is out of fashion, at least in ad-
vanced countries that can afford to be generous with their minorities. Japan
cannot afford to be an exception. As a consequence, there is heightened aware-
ness of Japan's nascent multilingualism which manifests itself, among others,
in more sociolinguistic research.
3.4. A review of the literature
Labour migration has been a recognized research topic among sociologists for
many years, but Kajita (1994: 102-111) was one of the first to address the im-
portance of the language issue in this connection. Social problems of inte-
gration and communication in the wake of an increased influx of unskilled
foreign labour have also been investigated by Miyajima (1993), Kajita and
Miyajima (1996), Komai (1995), Herbert (1996), and Douglass (2000). A
major project about unguided language learning by labour migrants and their
communication problems has focussed on the Brazilian-Japanese community
in Fujisawa city, Kanagawa prefecture (Furuishi and Hirataka 1998). Of the
many recent publications dealing with societal multilingualism in Japan or in-
volving language contact with Japanese the following deserve mentioning.
Maher and Yashiro (1995) must be credited for presenting the first overview
and general discussion of Japan's linguistic minorities. Loveday (1996) has put
language contact research about Japanese on a scientific footing. A number of
publications, such as Miura (1997), Inoue (1999) and Miura and Kasuga
(2000) are largely receptive, introducing the Japanese scientific community to
recent developments in multilingualism research. Bilingualism, too, has been
receiving more attention, as more Japanese children grow up in bilingual en-
vironments in Japan and abroad (Noguchi and Fotos 2000).
A new journal Kotoba to Shakai [Language and society], takes issue with so-
cietal multilingualism inside and outside Japan. In a series of reports on Ainu,
Ryukyuan, Korean and Japanese Sign Language it provides up-to-date in-
formation in these fields. Inoue Fumio investigates the linguistic market place in
multilingual Japan testing a number of parameters that determine the material
value of languages (Inoue 1993,1995,1997a, 1997b, 2000). Tamura's focus is on
foreign workers and the media (Tamura 1997), while Miyajima (1995) studies lin-
guistic diversity in public announcements and advertisements. Several scholars
and journalists conduct research on what they call "ethnic media", that is, radio,
TV programmes, newspapers, magazines, and leaflets. Moriguchi's comprehen-
sive guide to ethnic media is a useful research tool to investigate the degree of
multilingualism in audience specific media (Moriguchi 1997), while Shiramizu
continues the discussion from the point of view of globalization and identity (Shi-
ramizu 1988,1996,1998,1999, 2000).
Minority protection and language rights stand for another set of topics dealt
with in recent publications (Gengoken kenkyükai 1999). Language revitaliz-
ation, especially with respect to Ainu (Nakagawa 1995, 1996, 1997; Sato
1999), and language maintenance among the Korean communities (Fujii 1998,
1999-2000; Ryang 1997) have likewise attracted scholarly attention.
Against the backdrop of increasing interest in Japan's internal linguistic diver-
sity, the ideological role of Japanese as Japan's national language (kokugo) during
and after the colonial period is critically reexamined by Kawamura (1994), Lee
(1996), and Yasuda (1997, 1998, 1999a, 1999b), among others. Sakai (1996,
1997) offers a critique of the discourse on the Japanese language as a means of
cultural nationalism, a theme also discussed in conjunction with the discourse on
Japan as a homogeneous nation by Oguma (1995). At the same time, the need is
perceived to make Japanese more accessible to speakers of other languages
(Mizutani 1995; Kato 2000). Sanada (2000) diagnoses a tendency of destandardi-
zation in Japanese, which can be interpreted as yet another manifestation of a cul-
ture of opening up and the softening of rigid conceptions of linguistic uniformity
and homogeneity.
Taken together, the publications referred to in this section are indicative of a
trend. They show that, while in the 1960s and 70s linguistic research was pre-
occupied with the Japanese language, its purported uniqueness, and its signifi-
cance for Japanese identity, the 1980s and 1990s have witnessed scholarly at-
tention for the linguistic environment in which Japanese exists and for the pres-
ence in Japan of speech communities other than the Japanese. This trend sug-
gests, however indirectly, a growing willingness in Japanese society to ac-
knowledge their existence and to accommodate them.
4. Perspectives
As we have argued above, Japan is undergoing a number of structural changes

at the present time which will bring about, among other things, changes in its
linguistic composition and in its language regime. While these changes are per-
ceived by some as a threat to the integrity of Japanese culture and social har-
mony, others welcome them as harbingers of a more pluralistic and less self-
centred society. Greater linguistic diversity is already apparent, further devel-
opments in the same direction can be expected. Nascent multilingualism in
Japan is driven by two factors, the demographics of a rapidly ageing society
and the labour market. The particular nature of multilingualism in Japan thus
both reflects socio-economic conditions and has a bearing on them. Each lan-
guage present in Japan has its own profile, its own history, and its own life ex-
pectancy. In order to investigate the dynamics of Japan's nascent multilingual-
ism two explanatory tools can be put to use, ethnolinguistic vitality theory
along the lines developed by Giles and his associates (Giles, Bourhis and Tay-
lor 1977; Giles 1978), and linguistic market analysis as suggested by Coulmas
(1992), Grin (1999), and others. Another important aspect of incipient multi-
lingualism can be captured by examining institutional responses to communi-
cation patterns and requirements.
4.1. Linguistic vitality and the economics of multilingualism
Ethnolinguistic vitality is an aggregate measure of a language's strength in a

given environment. Three parameters commonly used to determine ethnolin-
guistic vitality are (1) the status of a language - e.g. national language, official
language, recognized minority language, minority language without official
recognition - (2) the demographic strength of a language, and (3) institutional
support for a language, that is, its use by government agencies, school, church,
and other organizations. In Japan, the only language which enjoys official
status is Japanese, although a discussion about granting English official status
has been initiated. In legal terms all other languages are minority languages
without official recognition at the national level. Assessing the demographic
strength of the languages present in Japan poses problems well-known in the
sociology of language. Government statistics contain no language data. Alien
registration and immigration statistics are broken down for nationality, but in
many cases this allows no conclusions about the languages spoken by the
groups in question. Even the available information about Koreans and Chinese,
the largest minority languages, is both scarce and unreliable. Many second and
third generation Koreans have only limited if any proficiency in Korean. Lan-
guage maintenance studies on Korean in Japan are small in scale and few in
number. As for institutional support, there are a number of schools which use
Korean and Mandarin as medium of instruction or teach these language as
mother tongues. Both languages are also used in religious services in some
communities, but there are no systematic records. Statistical data on the other
groups is even more fragmentary. To gather the relevant information on the
various languages large scale surveys would be required. However, because of
the large number of unregistered foreigners even instruments such as inter-
views and questionnaires wouldn't necessarily yield sound language data.
Other avenues must therefore be explored to investigate the vitality of minority

languages.
As mentioned above, the market of ethnic media has been rapidly expan-
ding in recent years. This market offers a promising opportunity for multilin-
gualism research not much exploited in the past. Moriguchi (1997) lists more
than 160 newsletters, journals, gazettes, and other publications in some 15 lan-
guages. Shiramizu (2000) examines the circulation figures of various "ethnic"
publications considering supply and demand statistics. Data of this sort are
relatively easy to come by, and in conjunction with interviews with opinion
leaders of the communities in question can be interpreted with regard to what
role their languages play in oral and written communication. Moreover, print
runs are not the only interesting aspects of these publications. Advertisements
are indicative of services and products available in the respective languages as
well as of the target readership. Following Shiramizu's suggestion ethnic
media market analyses must be expanded to include electronic media. Satellite
and cable TV and the Internet with all its stationary and mobile access gadgets
have dramatically enhanced the possibilities of even small communities to use
their languages for everyday communication. Brazilian TV programmes in
Portuguese are freely available. Of the 200,000 or so Brazilians in Japan at the
end of 1999, some 30,000 were registered viewers, the actual number of
viewers has been estimated to be more than three times as many. For this com-
munity the media help to build an infrastructure enabling its members to pur-
sue their affairs efficiently without having to speak Japanese much (Shiramizu
2000: 111-115), although many of these descendants of Japanese emigrants
have a fair command of Japanese.
In addition to ethnic media such as radio and TV programmes and news-
papers, other language-specific commodities (LSCs) can be included as a con-
stituent factor of the economic assessment of linguistic vitality. By this term
Grin (1999: 39) refers to "consumption goods and services, non-material com-
modities, or production factors that embody some language-related character-
istics." Printers such as, for example, the company KBS in Osaka which
specializes on multilingual printing offering services in Chinese, Korean, Rus-
sian, Thai, Greek, and Vietnamese, are paradigmatic, but there are many other
LSCs, especially in the software industry, in education, and in the service sec-
tor which can be taken into account.
School attendance and newspaper circulation are good indices both of the
vitality of a language and of the economic power of its speech community. The
Chinese community in Yokohama maintain schools for students from Taiwan
and Mainland China, an obvious indication of its economic potency. That
politics can be a significant intervening factor is even more clearly evidenced
by the Korean community which is split along political lines. Those whose al-
legiance is with the Democratic People's Republic of (North) Korea have
shown greater language loyalty and a far higher degree of organization than
those who maintain ties with the Republic of (South) Korea. The vast majority
of Korean schools in Japan, including a university, are managed by Northern
loyalists. Fujii (1999:195) cites the figure of 140 North Korea oriented schools
with a total student population of about 18,000. In contrast, there are only four
South Korea oriented schools enrolling about 2000 students (Maher and Ka-
wanishi 1995: 89).
4.2. Institutional support
Turning now to how Japanese institutions react to the new linguistic pluralism,
it must be noted that there is a marked difference between national and local
administrative agencies. On the whole, local authorities are more accommodat-
ing than the national government. This not only reflects different political atti-
tudes and the influence of the traditionally conservative national bureaucracy,
but also different levels of involvement. It is local governments which have to
meet changing communication requirements. Social welfare, healthcare, edu-
cation, and immigration while following general guidelines set by the national
government are put into practice at the local level. It is not surprising, therefore,
that local governments have in many cases acted pragmatically rather than in a
principled way.
Of the institutions which have attracted attention in conjunction with the
problematique of linguistic pluralism two stick out, schools and courts of law.
Having to deal with students of limited Japanese proficiency poses entirely new
problems for Japanese schools which most teachers have no experience or
preparation to deal with. So far, no general policy is discernible. Extra efforts
by motivated teachers is the only support such students can expect to receive.
While private initiative has led to the establishment of many culture exchange
circles and support groups, the school system as an institution has been slow to
respond to the new situation. While non-Japanese children have the right to at-
tend Japanese schools, the practical conditions to do so are lacking in many
cases. A 1999 survey in Mie prefecture tried to determine where the resident
non-Japanese children attended school. Japanese schools were attended by less
than half of the respondents and "mother tongue schools" by 13%. The re-
mainder responded that their children were educated at home or that they
couldn't answer the question (Shikama 2001). Findings such as these indicate
that education authorities are only just beginning to assess the situation and are
yet to design effective responses to deal with the new situation.
This can also be said of courts of law. Dozens of languages have found their
way into Japanese court rooms in recent years, but rather than setting general
standards for court interpretation and translation, the legal system has contin-
ued to operate on a case by case-basis. There is a serious shortage of court in-
terpreters, especially in uncommonly taught languages such as Farsi, Bengali
and Tagalog. While this is a practical problem of finding qualified personnel, a
more general issue has to do with defendants' rights. Article 181 of the Japan-
ese criminal procedure law stipulates that legal costs including interpretation
and translation fees should be bom by the defendant if found guilty. However,
the International Covenant on Civil and Political Rights which the Japanese
government signed in 1979 establishes that translation and interpretation are
essential prerequisites of proper procedure and fair trial and that, therefore, the
government has to carry the cost. Thus, if Japan is to live up to her international
commitments the criminal procedure law must be amended to secure defen-
dants' rights. Japan is a member to many international contracts and covenants
which establish obligations not just for inter-state relationships, but for the
treatment of foreign nationals at home. In conjunction with the increasing mo-
bility of labour these conventions undermine the effects of geographic insular-
ity which Japanese politicians have often invoked in order to justify Japan's ex-
ceptionalism when it came to matters of immigration, the acceptance of refu-
gees, etc. This is to show that the Japanese government does not have unlimited
discretion to shape institutional responses to incipient multilingualism, but, as
a consequence of progressing integration into the international community, is
increasingly bound to comply with regulations not of its own making.
At the domestic front, too, the national government has rarely taken the lead
in formulating active policies of accommodating and integrating foreigners.
More commonly, local governments have been at the forefront of paving the
way for non-Japanese nationals to participate in social and political life. For
example, in 1997 Kawasaki, an industrial port city south of Tokyo, created a
representative assembly for foreign residents that delivers policy suggestions
to the mayor. It was also the first major municipality to employ foreigners in
administrative government jobs. The Tokyo Metropolitan Government also es-
tablished a council of foreign residents. Other local governments are following
suit. While members of these bodies are typically long term residents fluent in
Japanese, non-discrimination and recognition of cultural and linguistic diver-
sity are among the concerns most frequently articulated by them.
From the above observations it appears that at this time Japanese institu-
tions are in the process of searching for proper ways to cope with a new situ-
ation. In language sensitive domains such as education and law the need to
allow for greater variety and to meet the communication requirements of resi-
dents with limited or no Japanese language abilities has been recognized. How-
ever, although more immigration is seen as inevitable by many, no constructive
immigration policy or social policy designed to accommodate newcomers is in
sight. As has often been the case in Japan, the State continues to lag behind
large sections of society.
5. Conclusion
Japan's linguistic unification was a declared political objective in Meiji Japan.

The fact that it could be accomplished to a very large extent is widely seen as a
contributing factor to Japans rapid modernization. Language reform, literacy in
a single national language and compulsory education were instrumental in
transforming Japan from a feudal, largely agrarian country into an industrial-
ized modern state. Homogeneity and reliability in communication accorded
well with the requirements of industrial mass production. Nowadays, Japan is
faced with new challenges. More strongly integrated with the rest of the world,
it attracts more foreign residents than ever before whose presence, moreover, is
indispensable for the continuing development and prosperity of a rapidly
ageing nation. The once favoured ideals of uniformity and the identity of state,
people and language are losing strength, however slowly, as are the prospects
of realizing them. Increasing linguistic pluralism in Japanese society is a fact.
The advent of multilingual pockets in a society which still operates under
largely monolingual assumptions has only begun to have an impact on public
discussion and Japan's self-image, but there can be no doubt that minorities in
Japanese society are both more visible and meet with greater readiness to ac-
knowledge their existence than only a couple of decades ago. To advance our
understanding of the dynamics of societal multilingualism it will be of major
interest to observe how Japan manages in the coming decades to reinvent itself
as a more pluralistic and open society.
References
Coulmas, Florian (ed.)

1984 Linguistic Minorities and Literacy: Language Policy Issues in Devel-
oping Countries (Trends in Linguistics: Studies and Monographs 26.)
Berlin/New York: Mouton de Gruyter.
Coulmas, Florian
* 1990 Language adaptation in Meiji Japan. In: B. Weinstein (ed.), Language
Policy and Political Development, 69-86. Norwood, NJ: Ablex.
* 1992 Language and Economy. Oxford: Blackwell.
Douglass, Mike
2000 Japan and Global Migration: Foreign Workers and the Advent of a
Multicultural Society. London: Routledge.
Fasold, Ralph
1984 The Sociolinguistics of Society (Language in Society 5 / Introduction to
Sociolinguistics vol. 1). Oxford: Blackwell.
Fujii Konosuke
1998 Tagengo jöhösäbisu no gendankai: 'jöhö jakusha' wo tsukuranai tame ni
[Present stage of multilingual services: to avoid creating 'informational
handicapped']. Kenkyükiyo (Sekai jinkenmondai kenkyü sentä) 3:
253-316.
1999 Zainichi chösenjin no gengo seikatsu to chösengo kyöiku: chösengo wo
shiranai sedai ga fueru naka de [Language life of Japanese Korean and
Korean language education: in the era of decreasing fluency of Korean
language among younger generations]. In: Shinpuro 'nihongo' kenkyO-
han 1 and gengoseisaku kenkytlkai (eds.), Sekai no gengo mondai 4:
39-76. Tokyo: Kokuritsu Kokugo Kenkyüjo.
1999-2000 Chösengo [Korean language], Kotoba to shakai 1,2, and 4.
Fukuzawa Yukichi
1959 Moji no oshie [A lesson on writing]. In: Fukuzawa Yukichi (ed.), Zen-
sha, volume 3,555-611. Tokyo: Iwanami Shoten.
Gellner, Ernest
1994 Nationalism and modernization. In: J. Hutchinson and A. D. Smith
(eds.), Nationalism, 55-63. Oxford/New York: Oxford University Press.
Gengoken Kenkytlkai (ed.)
1999 Kotoba eno kenri: gengoken towa nanika [Arightto language: what are
linguistic rights?]. Tokyo: Sangensha.
Giles, Howard
1978 Linguistic differentiation in ethnic groups. In: H. Tajfel (ed.), Differenti-
ation Between Social Groups, 361-393. London: Academy Press.
Giles, H., R. Y. Bourhis and D. M. Taylor
1977 Toward a theory of language in ethnic group relations. In: H. Giles (ed.),
Language, Ethnicity and Inter-group Relations, 307-348. London:
Academic Press.
Grin, Francois
1999 The notions of supply and demand in the economic analysis of language.
In: A.Breton (ed.), Exploring the Economics of Language, 31-61
(New Canadian Perspectives series). Ottawa: Canadian Heritage.
Haarmann, Harald
1989 Symbolic Values ofForeign Language Use: From the Japanese Case to
a General Sociolinguistic Perspective. Berlin/New York: Mouton de
Gruyter.
HenshO Iinkai (ed.)

1999-2000 Tagengo shakai nippon [Multilingual society Japan]. Kotoba to shakai:
tagengo shakai kenkyü 1 to 4.
Henshubu (ed.)
2000 Köyögoron no shiten: 21 seiki nihon no gengoseisaku wo kangaeru [The
official language discussion in focus: a language policy for 21st century
Japan], Gengo 8 (29): 22-63.
Herbert, Wolfgang
1996 Foreign Workers and Law Enforcement in Japan. London/New York:
Kegan Paul International.
Hildebrandt, Nancy and Howard Giles

1980 The English language in Japan: A social psychological perspective.
Japan Association of Language Teachers Journal 2: 63-87.
Hirai Masao
1998 Kokugo kokuji mondai no rekishi [History of the national language and
writing]. Tokyo: Sangensha.
Inoue Fumio
1993 Kotoba no chiteki kachi to jöteki kachi [Intellectual value and emotional
value of language], Gengo 22 (12): 24-31.
1995 Nihongo no shijö kachi no hendö [Fluctuation of Japanese language's
market value]. Gengo 24 (9): 68-75.
1997a Gendai nihon no gengo shijö [Language market of contemporary Japan].
In: Shinpuro 'nihongo' kenkyühan 1 and Gengoseisaku kenkyOkai
(eds.), Sekai no gengo mondai 3, 21-41 [Language problems of the
world 3]. Tokyo: Kokuritsu Kokugo Kenkyüjo.
1997b Market value of languages in Japan. Nihongo kagaku 2: 40-61.
2000 Köyögoka no hitsuyö keihi: eigo daini köyögoron no gengo keizaigaku
[The cost of an official language: the language economy of English as a
second official language]. Gengo 29 (8): 30-37.
Inoue Teruo et al. (eds.)

1999 Tagengo shugi no kanösei: kokuminkokka wo koete [Possibilities of
Multilingualism: beyond the nation state]. KEIO SFC REVIEW 5 (3 / 2).
Ishiguro Osamu
1941 Nihongo no sekaika: kokugo no hatten to kokugo seisaku [The global-
ization of Japanese: the expansion of the national language and the
national language policy]. Tokyo: ShQbunkan.
Kajita Takamichi
1994 Gaikokujin rödösha to nihon [Foreign workers and Japan]. Tokyo: Nip-
pon Hoso Shuppan Kyokai.
Kato Hidetoshi (ed.)
2000 Nihongo no kaikoku [Opening Japanese]. Tokyo: TBS Buritanika.
Kawamura Minato
1994 Umi wo watatta nihongo: shokuminchi no 'kokugo' no jikan [Japanese
overseas: 'National language' lessons in the colonies], Tokyo: Seidosha.
Komai Hiroshi
1995 Migrant workers in Japan. London/New York: Kegan Paul Inter-
national.
Lee Yeon-suk
1996 "Kokugo" to iu shisö: kindai nihon no gengo ninshiki [Thinking the
"national language": conceptions of language in modern Japan], Tokyo:
Iwanami Shoten.
Loveday, Leo J.
1996 Language Contact in Japan: A Socio-linguistic History. Oxford: Cla-
rendon Press.
Maejima Hisoka
1956 Kanji gohaishi no gi [A proposal for abolishing Chinese characters].
Maejima Hisoka Jijoden, 153-159. Hayama: Maejima Hisoka Denki
Kankukai.
Maher, John C. and Yashiro Kyoko (eds.)
1991 Nihon no bairingarizumu [Bilingualism in Japan], Tokyo: Kenkyusha
Shuppan.
*1995 Multilingual Japan. Clevedon/Philadelphia/Adelaide: Multilingual
Matters.
Maher, John C. and Yumiko Kawanishi
1995 On being there: Korean in Japan. In: Maher and Yashiro (eds.), Multilin-
gual Japan, 87-101. Clevedon/Philadelphia/Adelaide: Multilingual
Matters.
Miura Nobutaka (ed.)

1997 Tagengoshugi towa nanika [What is multilingualism?]. Tokyo: Fujiwa-
ra Shoten.
Miura Nobutaka and Kasuga Keisuke (eds.)
2000 Gengo teikoku shugi towa nani ka ? [What is a linguistic imperialism?].
Tokyo: Fujiwara Shoten.
Miyajima Takashi and Kajita Takamichi (ed.)
1996 Gaikokujin rödösha kara shimin e: chiikishakai no shiten to kadai
Icara [From foreign workers to citizens: perspectives and problems of
local society], Tokyo: Yohikaku.
Miyajima Takashi
1993 Gaikokujin rödösha to nihonshakai [Foreign workers and Japanese so-
ciety]. Tokyo: Akashi Shoten.
Miyajima Tatsuo
1995 Tagengo shakai eno taio [Dealing with the multilingual society]. Handai
nihongo kenkyü 7.
Mizutani Osamu
1995 Nihongo no kokusaika [Internationalization of Japanese]. In: BunkachO
(ed.), Shin 'kotoba' shinzu 1 kokusaika to nihongo, 34-47. BunkachO.
Moriguchi Hideshi
1997 Esunikku media gaido [Ethnic media guide]. Tokyo: Japan Mashinisu-
tosha.
Nakagawa Hiroshi
1995 Ainugo wo firudowaku suru: kotoba wo tazunete [Doing fieldwork in
Ainu: visiting the language], Tokyo: Taishükan Shoten.
1996 ShOsOminzoku to gengo no hoji [Ethnic minority and language mainten-
ance]. In: Miyaoka Hakuto (ed.), Gengojinruigaku wo manabu hito no
tame ni, 263-280. Tokyo: Sekaishisösha.
1997 Ainugo wa shigo ka? Nihon ni okeru shösügengo hukkouundo no genjö
[Is Ainu a dead language? The state of minority language revitalization
in Japan]. In: Shinpuro 'nihongo' kenkyühan 1 and gengoseisaku ken-
kytlkai (eds.), Sekai no gengo mondai 3,43-72. Tokyo: Kokuritsu Ko-
kugo Kenkyojo.
Nishi Amane
1874 Yöji ο motte kokugo ο shosuru ron [An argument for writing Japanese
with Western letters], Meiroku Zasshi 1.
Noguchi, Mary G. and Sandra Fotos
* 2000 Studies in Japanese Bilingualism. Clevedon: Multilingual Matters.
Oguma Eiji
1995 Tanitsu minzoku shinwa no kigen: 'nihonjin' no jigazO no keifu [Origin
of the myth of the homogeneous nation: genealogy of self-portraits of
'Japanese']. Tokyo: Shinyosha.
Ryang, Sonia
1997 North Koreans in Japan: Language, Ideology, and Identity. Colorado:
Westview Press.
Sakai Naoki
1996 Shizan sareru nihongo nihonjin: 'nihori no rekishi-chiseiteki haichi
[Stillborn Japanese language and the nation: historical and geopolitical
arrangement], Tokyo: Shinyosha.
1997 Translation and Subjectivity: On Japan and Cultural Nationalism.
Minneapolis/London: University of Minnesota Press.
Sanada Shinji
1987 HyOjungo no seiritsu jijo [The formation of the standard language].
Tokyo: PHP.
2000 Datsu hyojungo no jidai [The post-standard language era], Tokyo: Shö-
gakukan.
Sato Kazuyuki (ed.)

1999a Saigaiji ni tsukau gaikokujin no tame no nihongo anbun: rajio ya
keijiban nado ni tsukau yasashii nihongo hyögen [Disaster information
for foreigners: Easy Japanese for radio broadcasts and notice boards
(draft)]. Tokyo: Kokuritsu Kokugo Kenkyüjo.
Sato Kazuyuki
1999b Saigai ga okotta toki ni gaikokujin wo tasukeru tameno manyuaru (Hi-
rosaki ban) [Disaster assistance manual for foreigners (Hirosaki ver-
sion)]. Hirosaki: Hirosaki Daigaku Jinbun Gakubu Kokugogaku Ken-
kytlshitsu.
Sato Tomomi
1999 Ainugo no fukkou to Ainugo jiten [Revilatlization of Ainu language and
Ainu dictionary], Gekkan Hyakka 442: 58-66.
Shiga Naoya
1946 Kokugo mondai [The national language issue], KaizO.
Shikama Ayako
2001 Tagengoshakaika suru nihon no gengoseisaku [Japan's language pol-
icy for increasing societal multilingualism]. Unpublished Master's
thesis, Faculty of Policy Studies, Chuo University, Tokyo.
Shiramizu Shigehiko
1988 Komyunikeshon to bunkahendö: media, komyunitl, fukyü [Communi-
cation and cultural change: media, community and the diffusions].
Tokyo: Hakuto Shobö.
1996 Esunikku media: tabunkashakai nihon wo mezashite [Ethnic media: to-
ward the multicultural Japan], Tokyo: Asahi Shoten.
1998 Esunikku bunka no shakaigaku: komyunitl, ndä, media [The sociol-
ogy of ethnic culture: communities, leaders and media]. Tokyo: Nippon
Hyöronsha.
1999 Esunikku media to gengo [Ethnic media and language]. In: Shinpuro 'ni-
hongo' kenkyOhan 1 and gengoseisaku kenkyükai (eds.), Sekai no
gengo mondai 4,77-112. Tokyo: Kokuritsu kokugo kenkyüjo.
2000 Hitobito no idö to esunikku media [Global migration and ethnic media].
Kotoba to shakai 4:109-119.
Slawey, Charles M.
1993 The Island Dependencies of Japan. London: Morice.
Tamura Norio
1997 'Kokkyo naki rödösha' to media: nihon nimo yattekita tagengoshakai
[Borderless workers and media: multilingual society reaches Japan],
Tokyo: Nitchu Shuppan.
Tanaka Katsuhiko
2000 Köyögo towa nani ka [What is an official language?]. Gengo 29 (8):
40-49.
Twine, Nanette
* 1991 Language and the Modern State: The Reform of Written Japanese,
London/New York: Routledge.
Ueda Kazutoshi
1895 Kokugonotame [For a national language], volume 1. Tokyo: Füzambo.
1903 Kokugo no tame [For a national language], volume 2. Tokyo: Füzambo.
Yasuda Toshiaki
1997 Teikoku nihon no gengo hensei [The linguistic composition of imperial
Japan], Yokohama: Seori Shobo.
1998 Shokuminchi no naka no 'kokugogaku' ['National language studies' in
the Japanese colonies], Tokyo: Sangensha.
1999a 'Gengo' no köchiku: Ogura Shinpei to shokuminchi chösen [The con-
struction of 'language': Ogura Shinpei and colonial Korea], Tokyo: San-
gensha.
1999b "Kokugo" to "högen" no aida: gengo köchiku no seijigaku [Between

"national language" and "dialects": the politics of language construc-
tion]. Kyoto: Jinbun Shoin.
1. How does the difference between spoken and written language relate to
multilingualism, in Japan and in general?
2. Standardization is a process that affects many social sub-systems. How
does it come to bear on language and modernization in Japan?
3. What were the major forces effecting changes in the functional distribution
of languages in Japan during the twentieth century?
4. Japan is often cited as a paradigm case of a monolingual state. What lessons
can be learned from the inclusion of Japan among the countries discussed in
this volume?
5. What does 'internationalization' mean for Japan's language regime? Dis-
cuss the implications of this concept for language policy in Japan and in
general.
III. Multilingual management and education
Managing multilingualism in Singapore
Xu Darning and Li Wei
1. Introduction
The majority of sociolinguists, as we know, do not endorse direct control of

language by the government or similar political bodies. Even among students
of language planning, "intervention" is not always a favorable term. The strong
interventionist approach, or linguistic dirigisme, is criticized for ignoring the
ecological complexities of language and society and is often connected with in-
effective measures or negative results (Schiffman 2002). The recent surge of
theoretical characterization of language as a commodity circulated in a market-
place of symbolic values motivates some analyses of language situations and
cases of language planning (Bourdieu 1982; Li Wei 2000, among others). In
such a perspective, the parallel of the control of linguistic market and the con-
trol of market in economy certainly leads to its negative reception in the gen-
eral approval of the free-market spirit and practices.
Our objective in this chapter is to evaluate the activities of the Singapore
government in its attempts to change the linguistic realities and charter the fu-
ture development of the languages of the country. On the one hand, we examine
the effectiveness of the measures implemented and actions taken; on the other,
we analyze the nature of the action to see how much it can be characterized as
acts of dirigisme or as manipulation defined perhaps more appropriately as
"language management" (Jernudd 1982; Kuo and Jernudd 1994; Pakir 2000).
2. Language planning in Singapore
The language situation in Singapore has been surveyed in Afendras and Kuo
(1980), Anderson (1985), Kuo (1976,1985), Kuo and Jernudd (1994); updated
community-based studies of the language situation with respective foci on the
major ethnic groups of Singapore include Abdullah and Ayyub (1998), Sara-
vanan (1998), and Xu etal. (1998); and studies on Singapore's language pol-
icies include Gopinathan (1979, 1998), Pendly (1983), and Pakir (2000). Nu-
merous articles and books have been published on the teaching and learning of
the four official languages in Singapore, Malay, Chinese, Tamil and English,
276 Xu Darning and Li Wei
and among them with English receiving most of the attention. From a different
angle, descriptions and analyses of Singapore English as a form of New Eng-
lishes are also abundant. Publications about education and its relationship to
language planning and bilingualism, as well as those focusing on specific as-
pects of bilingualism in Singapore are also too numerous to enumerate. An up-
dated representative sample of these would be the articles included in the vol-
ume edited by Gopinathan etal. (1998). The government reports of population
census contain important information about the language situation in Singa-
pore, and these also include interpretive summaries such as Tay (1980) and
Tham (1990).
From these sources, it is clear that the language situation in Singapore has
changed tremendously since its independence in the late 1950s. More import-
antly, all the changes in language can be directly or indirectly traced to certain
government planning initiatives. In fact most can be unequivocally seen as
reactions to specific institutionalized moves of intervention in the linguistic,
educational, and socio-economic lives of Singaporean residents. Let us first
look at the scale of change in the linguistic composition of the population.
Table 1 gives data taken from the 1957,1980, and 1990 census reports.
Table 1. Percentages of speakers of different languages in Singapore in 1957,1980, and

1990
1957 1980 1990

(Mother Tongue) (Home Language) (Home Language)
Mandarin 0.1 7.9 23.5
Non-Mandarin 75.3 62.9 39.8
Chinese Dialects
Malay 9.3 15.5 14.4
Non-Malay 4.3
Malayo-Polynesian
Languages
Tamil 5.6 3.3 3.0

Non-Tamil 3.3
Indian Languages
English 1.8 8.9 18.7

Others 0.5 1.5 0.6
(Based on Chua 1957 and Tham 1990)

Managing multilingualism in Singapore 277
The 1957 census report gives the figures for mother-tongue self-claims,
which serve as the starting point for comparison. The 1970 census was the first
after the independence, but did not include questions on languages spoken. The
succeeding two censuses (1980 and 1990) included questions on the use of lan-
guages at home. Although the mother-tongue claims in 1957 and the home-lan-
guages self-reports from the two more recent censuses do not make up exactly
the same indices, they nevertheless show clear trends of language shift. If one
argues that mother-tongues are essentially different from home languages, the
contrast between 1980 and 1990 in the same indices still holds. Besides, the as-
sumption that the mother-tongue claims have close connections to the use of
languages at home is supported, especially with regard to the trends of shift,
from independent language surveys in 1972 and 1978, which would fill the gap
in the censual information (Kuo 1985). The independent surveys, which were
conducted for the purpose of market research, show the same trends as did the
censuses, that the population as a whole has been shifting from the original
mother tongues of Chinese dialects, Malay, Tamil, and various Indian lan-
guages towards English and Mandarin, with the ethnic Chinese being basically
the Mandarin acquirers (Xu et al. 1998). If we can reasonably assume that
home is usually the last fortress of the threatened ethnic language (see Li Wei
etal. 1997; Gupta and Siew 1995), the apparent relinquishment of it by an in-
creasing proportion of the population is a distinct sign of language shift of the
usual kind - ethnic languages yielding to an imposing standard language of
wider circulation.
However, what is unusual about the Singapore situation is that the replacing
languages are not the inherited vernaculars of the majority ethnic group, the
Chinese, which makes over 75% of the total population. In Singapore it is the
majority ethnic group that has become the most bilingual and multilingual
among the population. No less remarkable is the rapid and sweeping alteration
in language affiliation as shown by the figures in the surveys. The increases of
English and Mandarin speakers from 1.8% and 0.1% in 1957 to 18.7% and
23.5% in 1990 respectively are amazing. Apparently, for over 40% of the Sin-
gapore population the process of language shift is telescoped into one gener-
ation, while the familiar pattern is the "three generation change" (Fishman
1964).
The Singapore miracle of language shift is a miracle in language planning,
where under government guidelines the speakers of English and Mandarin in-
creased over 100 folds and 200 folds respectively in a short span of three dec-
ades. In the process of industrialization and modernization, the government of
Singapore has consciously tried to transform its population and to create a cul-
tural environment conducive to the achievement of its economic and political
goals. The mastery of the designated varieties of its official languages among
the population is an essential part of the cultural transformation. In fact, the
cultural transformation has been well-interwoven with the economic success
and the political consolidation.
Behind the census statistics is the multilingual-multifunctional use of lan-
guages in realities (cf. Gupta 1994; Xu and Tan 1997; Xu and Tham 1997; Li
Wei etal. 1997; Xu etal. 1998, among others). It is rare to find a contemporary
Singaporean who is completely monolingual, without some working knowl-
edge of at least one other language. However, the general situation is post-dig-
lossia, a diglossia that is in the process of dissolvement. As defined by Xu and
Tham (1997) with reference to the Singapore situation, post-diglossia is char-
acterized by serious "leaking" both ways, multiple standards of language use,
rising trends in code-switching practices, and conflicting language attitudes
found in the society. What is not reflected in the census report is the indecisive-
ness of many Singaporeans, when asked, about what is the most frequent lan-
guage they use in a specific type of situation, e.g., at home; neither does the re-
port reflect the practice of code-switching among the young people (Xu etal.
1998; Chew etal. 1997).
We can define a native language as "one that is first learnt and in which one is
still conversant". In that sense, the census figures for home language are in the
same range as the native languages of the population. This assertion is in fact
corroborated with many other estimates from different sources (Kuan-Terry
1991; Gupta 1994; Chan 2000, etc.). Given that all prejudices about what is the
proper kind of a language are forsaken, the number of speakers who can identify
with English or Mandarin is even larger than the census figures for home lan-
guage (defined in the census as "predominant household language"). As
pointed out by Li Wei etal. (1997), in Singapore, as in many other places, many
first-learnt languages are no longer the most fluently used ones. With the rec-
ognition that many of the speakers who claim to be native speakers of English
and Mandarin or are regarded as such do not speak the standard varieties as in-
ternationally recognized and that they may tend to mix them with words of some
other varieties of language in actual usage, we grant them the status of native
speakers for no other varieties they know come closer to meet the criteria. The
essential facts are that they depend on these varieties for daily communication
to practically the same extent as their parents did or even as they themselves did
when younger on the varieties of those receding languages in the shift situation.
With the bubbles removed from the surface, the stark realities of change are
exposed. As pointed out above, not only is the change a change for its own sake
but part and parcel of the general facts of industrialization and modernization,
but the facts also look real when examined in their raw form of state. Conse-
quently, the miracle of language planning is essentially true, albeit not in an

idealistic form. The next step we will take is to analyze how the miracle is cre-
ated, with a focus on whether it is brought about with strong drives of interven-
tionism, means of dirigisme, or persistent but flexible efforts in utilizing all
available resources and opportunities to achieve a general goal.
3. Language management
Jernudd and Neustupny (1986) used the concept of language management to

avoid the strong interventionist connotation of the term "language planning".
We find that the Singapore practices of language planning are best described
and explained in this perspective. In this perspective, the language planners
take account of the economic characteristics of language. They assume the role
of management, whose major responsibilities are maintaining, utilizing, and
developing available assets. They do not create from nothing and they do not
ordain out of belief. This approach tends to be pragmatic rather than idealistic,
flexible rather than dogmatic. They organize a process of development, em-
phasizing feasibility and efficiency, rather than decreeing and expecting instan-
taneous changes. Perhaps the key concept of language management is the treat-
ment of language and the sociolinguistic conditions of languages as resources.
Proper identification of these resources, reasonable allocation of the resources,
and careful coordination of the development of the resources are perhaps all
that language planners can do in "planning" and changing the course of the so-
ciolinguistic development of a society.
In speaking of resources for language planning, we refer to all favorable
conditions the language planners can rely on, including such wide-ranging fac-
tors as the existing speakers of certain languages, the state power for imple-
menting language policies, the society's support for the planning programmes,
the physical facilities for carrying out the programmes, etc. These should in-
clude all material and non-material assets the planner possesses or otherwise
can utilize to initiate, facilitate, and complete any actions of planning.
3.1. Identifying resources
In setting the goals of language planning, the Singapore government has pro-
ceeded with what it has available as resources. Among the resources, the socie-
tal consensus seems to be the most important. Equality in the treatment of the
three major ethnic groups in language matters and in education was agreed on,
as early as 1956, by a consultative committee representing all political parties

and major interest groups of the time and this was documented in All Party Re-
port on Chinese Education (or All Party Report in shorthand) (cf. Xu and Li
Wei, in press). This was to serve as the earliest formal attempt to set the goal of
language planning for the incubating republic then. When Singapore achieved
internal self-government in 1959, the PAP-dominant government introduced a
Five-Year Plan in education by which the principle of the equality of the ethnic
languages with English was to be realized. However, due to the political agenda
of merging with the Federation of Malaya, the PAP government also empha-
sized the special status of the Malay language and designated it as the "national
language". Malay was to be taught in all schools, including the English, Chi-
nese, and Tamil-streamed schools. A Malay-knowing bilingual and trilingual
education policy was formulated. The goal was to unify the nation with Malay
while preserving the traditional streams of education in the four languages. The
Malay-knowing bilingualism was to be replaced by the English-knowing bilin-
gualism in time. The four streams of schools would also see their end in the
eighties. However, the bilingual program with the preservation of the lan-
guages of the major ethnic groups has remained unchanged ever since the All
Party Report. Even when Singapore became part of Malaysia in 1963, it did so
with the provision that its education policy was protected. In 1965, when it was
forced out of Malaysia and became a fully sovereign state, the official status of
the four languages was formally declared.
An ideal resolution of linguistic diversity would be to unify the country
with one language while eliminating others but this was deemed unfeasible by
the government of the new state. Although communicative and political inte-
gration was recognized as a top priority, the government decided to take a step-
by-step approach. The colonial mode of communication, characterized by ter-
ritorial bilingualism and ethnic-enclave monolingualism (c.f., Fishman 1972;
Xu and Tham 1997), was not to be continued. With the hope that Singapore
could be merged with Malaya, the PAP government wanted to gradually re-
place English with Malay as the administrative and interethnic communication
language. This looked feasible at the time when Bazzar Malay, a pidginized
Malay vernacular variety was already circulating in Singapore, mainly for pur-
poses of interethnic transactions.
The political rationale for the four-official-language policy is to uphold the
principle of pluralism. However, from a practical point of view, the decision
was based on some business grounds with languages being treated as commod-
ities with values. English had both true economic value and symbolic value. It
had served as an important trade language in the entrepot trade economy. It had
been the administrative and legal language of the country as well as being the
school language and frequently also the native language of an elite group in so-
ciety. To downgrade it to one of the four official languages, on a par with the
three ethnic languages, is to distance it from the colonialist position in lan-
guage policy. While being kept as an official language, English continued to
hold its special position as the administrative and legal language by default.
The elite group and their English-knowing followers thus continued enjoying
their linguistic advantage in the public domain, albeit being cleared of the co-
lonialist association. Therefore, with clever manipulation, English kept its so-
cial prestige in post-colonial Singapore. Malay had important symbolic value,
both as an intended symbol for an orientation towards the Malay-speaking
countries of the region, and as a "national language" implying nationalism con-
trasted with colonialism. Mandarin Chinese had been an Η variety in a com-
munity-wide diglossia for decades (Kuo 1985), and thus enjoyed prestige and
was accepted as a symbol for the Chinese nationality as a whole. Tamil's value
lay mainly in the relatively large number of its speakers, in consideration both
of its proportion in the total population and in the Indian sub-population. Al-
though the Tamils were the largest ethnic group among all Indian nationalities
in the Singapore population, using their language to unite all Indians in Singa-
pore, as has been attempted in language education, has not been so successful.
The equation of "Tamil" with "Indian" in education policy1 raised resentment
among the non-Tamil Indian population because of their non-identification
with Tamil yet at the same time gave them some consolation that Indian culture
and Indian minorities were being given consideration and recognition.
In fact, the number of speakers was less important compared to the lan-
guages' realized value in the symbolic market, which concerns social prestige,
social symbolism, and the social perception of the values of the languages. The
four languages chosen all have a long tradition of literature and standardization.
In Singapore, they were also school languages each with its tradition. Compar-
ing with other language varieties found in the communities, they had the ad-
vantages in "codification", "standardization", "historicity", and cultural "au-
thenticity" (Stewart 1962). The then current Bazzar Malay and Hokkien,
though each commanding much larger numbers of speakers than Standard
Malay and Mandarin respectively, would not compare in these aspects. In hind-
sight, the success of the popularization of English in the country, Mandarin in
the Chinese community, and in leveling the dialectal differences in the Malay
community all depended crucially on the original choice of the norms. The
choice was correct because it was a rational one weighing the market values of
the language varieties. The relatively unsuccessful spreading of Tamil can also
be traced to its lower symbolic value (as well as economic value). To do justice
to economic determinism, we have to admit that the faring of the four official
languages did correlate with the economic values of these languages. The value
of a language thus should be a composite measure of its socio-symbolic value
and its economic value.
We can examine the Singapore case of language status planning with
Haugen's (1966) criteria of efficiency, adequacy, and acceptability. We can see
that the planner had little concern about adequacy and their primary concern
seemed to be acceptability. The All Party Report and symbolic value of the lan-
guages as reviewed above reflect the acceptability of the chosen languages. The
transfer of the colonial administrative language to the newly-independent state
was certainly an efficient move. Apart from that, the Singapore government
certainly considered the education legacy of the four languages and the support
for them in society. None of the chosen official languages was a major lan-
guage of the country by the criterion "spoken as a native language by more than
25% of the population" (Stewart 1962; Ferguson 1971). It might have appeared
that the government of the newly-independent country just ordered four ready-
made standard languages as its official languages. However, it now becomes
clear that the choice was made on the basis of resources that provided the foun-
dation for development - two of the chosen have by now become major lan-
guages of the country by any standard.
The apparent outsourcing practice then was in fact an effective use of inter-
nal resources. Although English and Mandarin were the mother tongues of less
than two per cent of the population in 1957, they were understood and spoken
by large portions of the population as a second language. While only 1.8% of
the population claimed English as a mother tongue, 23.5% claimed the ability
to speak it; the case of Mandarin was 0.1% and 19.9%. These were chiefly the
results of education. In Singapore English schools and Chinese schools both
dated back to the nineteenth century while Malay and Tamil schools had a
shorter history. The early Chinese schools used Chinese dialects as the medi-
um of instruction. But after Mandarin was chosen as the "National Language"
by the Republic of China in 1917, they successively switched to Mandarin in
the following years. The impact of the English and Chinese education was also
the strongest among the population. The most prosperous time for the Chinese
education was the post-war and pre-independence period, during which time
the enrolment in Chinese schools exceeded that of the English schools and
when the first university in Chinese education outside of China was established
in Singapore. While the colonial government laid emphasis on education in
English, Chinese education, which depended entirely on initiatives from the
Chinese community, became dominant in the educational market. For instance,
in 1950 enrolment in Chinese schools was 53% of the total enrolment, while
enrolment in English schools was only 33%. However, the English school en-
rolment had caught up with that of the Chinese in 1954. By 1960, the English
school enrolment reached 51%, while the Chinese school enrolment dropped
to 42%. We can leave the contest between English and Chinese education for
later discussion. Here it is to be noted that, since the late fifties, the sum totals
of English and Chinese school enrolments have always accounted for over
90% of total enrolment, until the disappearance of Malay and Tamil schools in
the seventies and the complete win-over of English schools in the eighties.
Therefore, it was only sensible for the government of the new republic not to
forsake these educational and linguistic legacies in English and Mandarin in
spite of the extremely small proportion of native speakers among the popu-
lation. The only language variety with the number of native speakers exceed-
ing 25% of the population then was Hokkien, a variety of South Min dialect of
Chinese, which accounted for 30% of the mother tongue claims of the popu-
lation by the 1957 census. It was perhaps also the most widely spoken as a sec-
ond language among the population. The 1957 census did not include Hokkien
in the enumeration of languages spoken other than mother tongue. However, a
market research survey in 1972 revealed that Hokkien was the most widely
understood, by 73% of the population while the next in rank was Malay at
57%. The choice of Mandarin over Hokkien as an official language was politi-
cally justified as imparting impartiality among the different Chinese dialectal
groups as the choice of English was impartial at a higher level of ethnic affili-
ation, but with the addition of some well-rooted cultural legitimacy (cf. Chen
etal. 2000).
To summarize, the decision to adopt English, Mandarin, Malay, and Tamil
as official languages of the republic was a rational choice for political reasons.
We may or may not attribute futuristic insight to the decision-makers then for
the later-shown economic values of these languages. However, the fact that
these are well-codified standard languages with local and international prestige
and that they are the educational languages of the population was certainly
within their purview. Seen from a market perspective, these varieties were all
circulating in Singapore then with well-received symbolic as well as purely
economic values to different degrees. Therefore, the language planner in this
move was not insensitive to languages and their sociolinguistic conditions as
resources to utilize, with all their interventionist intentions.
3.2. Utilizing resources
Once the resources were identified, the next step was to utilize them and am-
plify them. The Five-Year Plan in education by the post-independence govern-
ment can be seen as the first move to utilize educational resources to achieve
their goal in language planning. As mentioned above, the specific program was
bilingual and trilingual education with Malay being taught in all schools to be-
come the common language for all students. Here we see that although eager to
spread the Malay language, the government depended on the use of existing
educational facilities which were molded in separate language and ethnic tradi-
tions. At the same time the Plan included an emphasis on the study of the sub-
jects of mathematics, science and technology. This was of course in line with
the new government's economic development strategies with a professed goal
of "the economic survival" of the nation. The economic value of English in the
world market and its predominance in facilities of scientific and technological
education thus gave English education an additional edge. Consequently, in
spite of the fact that the Five-Year Plan gave an official guarantee that all four
streams of education would be treated equally, we see a sharp decline in edu-
cation in ethnic languages in the following years. The government initiatives
then could be seen as a response to the changing economic environment and
thus they were only partially responsible for the upward trend of English skills
in the labor and educational market, which had already started before the gov-
ernment move.
Promising equal treatment of different schools and letting them compete on
their own in the market was to make them fall into the control of economic
forces. In sharp contrast is the neighboring Malaysian government's differen-
tial education policy favoring education in Malay, which counter-balanced
much of the market forces (Wong and James 2000). The ideological motivation
for sending children to ethnic schools as a reaction to colonial government's
unequal education policy was no longer there in post-colonial Singapore, since
all streams of schools were now officially equal. The practical concern for em-
ployment and the desire for socio-economic ascendancy influenced the par-
ents' decision in choosing a school for their children. As the rapid industrializ-
ation process began and an influx of foreign companies were settling in, a
knowledge of English became essential for well-paid jobs in the private sector
whereas it was exclusively in government services previously. With the trans-
formation of the entrepot trade economy into one with modernized industries,
scientific and technological education also became essential for employment
and this came most readily also in English.
Educational reform unifying all streams of schools with a common curricu-
lum and with the British system, and an increasing preference for English edu-
cation among the population led to the decline of ethnic schools. After the
Malay and Tamil schools' Primary-One enrolment became nil in 1983, the Chi-
nese schools' new enrolment dropped to less than one per cent of the total en-
rolment in 1983. The Ministry of Education thus decided that all schools would
teach English as the "first language" in 1987, officially ending the four streams
of education. Gopinathan commented (1998: 28):
The political ascendancy of English was already there at the colonial time, but it was
made irrelevant for the majority of population for lack of access to the political pro-
cess. It became relevant when the political association of the language became popu-
larly accessible after the independence. The efficient transfer of the colonial lan-
guage to the government language of the independent state, as mentioned above, in
fact, rooted its political and social privileges in the population to the extent that the
former colonialists would not have dreamt. The economical values of the language
apparently had much to do with foreign investment and the presence of multi-
national corporations, and the overall outward-orientated economy. The educational
supremacy of English in Singapore can be seen as a response to the political and
economical advantages of the language worldwide. However, its realization took a
prolonged term of careful management of the government.
Apart from managing the educational resources, the government also tried to
steer linguistic development in singaporean society. Three phases (or four, see
below) of government action in language planning in Singapore can be distin-
guished, with reference to acts of explicit intervention. The first phase of gov-
ernmental acts in language planning focused on the promotion of Malay as the
national language. This lasted from 1959, the beginning of self-government, to
1965, shortly after Singapore's separation from the Malaysia Federation.
Among the important measures to promote Malay was the establishment of
special government departments for it, requiring civil servants to pass Malay
language tests, starting to use Malay in government announcement, and setting
up "National Language Action Council" involving both government and non-
government public figures from all ethnic groups to scheme and carry out pro-
grams of promotion, which included "National Solidarity Week", "National
Language Month", "National Language Week", etc. This Malay-promoting
policy grew out of the political goal of the government to merge with Malaya
(and also effectively into Malaysia after the official merger) because the leaders
then believed that Singapore as an independent country would have no econ-
omic viability.
The second phase was steering towards English as the promoted language.
The annual "National Language Publicity Week" stopped in 1967 and at about
the same time government officials started issuing statements about the import-
ance of English for the economic development of the country. The leaders fin-
ally realized that Singapore had to stand on its own and any hope of merging
into the Malay econo-cultural entity had to be put aside if not totally aban-
doned. Opportunities had to be sought in a wider market, where the working
language was predominantly English.
The contradiction between the need for English and the lack of it in the
population was resolved mainly by an expansion in education with an emphasis
on English. The dominance of English in economic domains established itself
most rapidly in the seventies and eighties. At the same time, its political domi-
nance was further strengthened. Since the eighties English has been called Sin-
gapore's "de facto national language" (Fishman 1972) because of its predomi-
nance in the society. Being known as an English-speaking country accounts for
much of Singapore's international competitive edge and it is also one of its in-
ternally-recognized achievements.
The most strategic utilization of resources in language planning in Singa-
pore, in our view, is the use of the second language speakers for the creation of
native speakers. This is of course explicitly and even didactically done in the
case of Mandarin; while it was much subtler in the case of English. Before the
launch of the "Speak Mandarin Campaign" in 1979, government officials sel-
dom called the people to use English for daily communications, as they later did
in urging them to adopt the school languages at home. However, the planner's
confidence that using Mandarin at home would bring about a generation of Man-
darin speaking children was apparently attained through the observation of the
practices and consequences of parents doing so with English. Since the govern-
ment did not claim to be responsible for the shift to English, we must attribute it
to the masses' self-motivated emulation of the linguistic practices of the social
elite, the bulk of which were made up of government officials. According to
Gupta (1994:19), 70% of the current generation of children in Singapore may be
native speakers of English; moreover, these young native speakers are either
from socially prestigious backgrounds, as it had traditionally been, or from other
backgrounds as a result of the social spread of domestic use of English in more
recent years (1994:14). As seen from the census data, the natural breeding of the
1.8% of native speakers in the population in 1957 could not possibly have cre-
ated the 1990 results of 18.7% of native speakers, let alone the 70% by Gupta's
estimation. Therefore, the second language speakers of English, which made up
approximately one fifth of the adult population, would have to be chiefly re-
sponsible for this fantastic creation of a new generation of native speakers.
Gupta also noted two categories of adult native speakers of Singapore English,
one of which includes those "who have had their education in English from an
early age up to a high level, and who continue to use English in adulthood in all
major domains to the extent that English is their dominant language" and the
other includes those who "acquired English in the home from birth, not subse-
quent to any other language'' (1994:14). She noted, in addition, that members of
the community were not able to distinguish between these two groups (ibid.).
Therefore, many of the native child speakers of English might have come from
families where the parents were already using their adopted tongue in a native
way. There surely were others whose parents had acquired English at a second
language level. Since, after 1979, parents were explicitly told that using Man-
darin, which is a second language for them, to address their children would make
them proficient in this school language and pave the way to a brighter future, they
would easily put two and two together, figuring out that using another second
language at home would bring similar results, and in fact ones pertaining to a
more important school language, one bringing more promising careers for their
children.
The large number of native speakers of Mandarin were created in similar
ways, albeit on a greater scale and with more conscientious involvement of the
government. From the available sources of data, we know that the native speak-
ers of Mandarin "from birth" and those created by education (parallel to those
English ones defined by Gupta 1994) make a much smaller base than their Eng-
lish counterparts if compared in the pre-"Mandarin Campaign" years. There-
fore, we can define the third phase of the Singapore government's explicit lan-
guage planning activities as starting in 1979, with the "Speak Mandarin Cam-
paign" as the indicator. This phase is different from the previous phases in that
it does not only spread a language as its stated goal, but also aims at language
shift as a specified target.
At its inception, the objectives of the "Speak Mandarin Campaign" were
clearly stated the use of Mandarin to replace the Chinese Singaporeans' dia-
lects in daily communication among members of the ethnic group. Interest-
ingly, a major argument for doing so was that using Mandarin instead of dia-
lects would lessen the school children's learning burden. Since they had to learn
English and Mandarin at school, continued use of dialects at home would make
the children "find their work in school very burdensome" (Lee 1979: 10). An-
other argument given was that, since English is indispensable for interethnic
communication, the ultimate choice for the Chinese Singaporeans would be
"English-Mandarin", or "English-dialect" (Lee 1979: 12). The annual month-
long campaign, which included many publicity activities in the media, the or-
ganization of forums, seminars and other events, and provision of Mandarin-
learning services, etc., has continued till today, although the campaign rationale
and foci have been modified and shifted from time to time. Accompanying the
campaigns, the government implemented a series of measures to assist the lan-
guage-shift goal, which include the phasing out of dialect programs over radio
and television, the requirement of Mandarin proficiency for some government
service personnel and other public service personnel such as counter clerks and
taxi drivers, and using Hanyu Pinyin in translating place names, etc., all aiming
at "altering] the language environment" (Lee 1979: 11).
Gopinathan (1998: 24) points out some political implications of the appar-
ent policy-shift in 1979 as indicated by the "Speak Mandarin Campaign".
Openly targeting dialects as the objects of attack had the effect of deflecting
concerns over the taking over of English education, although the claim that dia-
lects fragmented the Chinese community is justified on its own right. The hope
that spreading the use of Mandarin would enhance the cultural strength of the
community certainly gave comfort to the Chinese-educated majority members
of the community who were suffering from a westernization of society. Apart
from those who opted for "English-Mandarin" from a current state of "English-
dialect" and those who decided for practical reasons to become completely
anglicized, both together still made up a minority of the community, the major-
ity members of the community, with quite limited access to English despite the
post-independence education expansion of two decades, had no choice but to
lay much hope on a rapid transfer into only one of the school and official lan-
guages. To engage them in using the classroom language for daily communi-
cation and in creating a new generation of Mandarin native speakers is a very
efficient way of utilizing a surplus labor force and unused linguistic resources.
The encouragement of linguistic transfer, present already in the Prime Min-
ister's first campaign speech (see Lee 1979: 10), had much linguistic and cul-
tural legitimacy. By motivating the people, the creation of a large native popu-
lation of Mandarin did result in a transfer within and across generational lines.
However, we have to take into account the basis of the second language speak-
ers of Mandarin in 1957 (19.9% of the adult population) for the later develop-
ment of the native population. The 1990 achievement of Mandarin native
speakers of approximately a quarter of the population has to do with many a
second-language Mandarin speaking parent switching to Mandarin at home, to
draw a parallel with the English case as mentioned above. Looking at the data
of English and Mandarin, we can figure out that most of those proficient in
English as a second language in 1957 had a counterpart in the native population
of English in 1990 while the 1990 Mandarin native population should have ab-
sorbed all impact of the 1957 second-language speakers of Mandarin and more.
The Chinese population of Singapore is now well spread out in their lin-
guistic distribution as a consequence of the language planning efforts by the
government. The language repertoire of the community now include English,
Mandarin, and the (non-Mandarin Chinese) dialects. A community-wide sur-
vey in 1996 revealed that about 10% of the population use English only as a
predominantly-used language variety, 20% use dialects, and 30% use Manda-
rin as a predominantly-used variety (Xu etal. 1998). The "English and Manda-
rin" bilingual mode of language use, being the ideal of the education policy and
originally stated goal of the "Mandarin Campaign" was reached by 10% of the
population as indicated by the survey. While the "English and dialects" mode
and "English, Mandarin, and dialects" mode was each reduced to less than 5%,
in line with the targeted elimination as the original goal of the campaign, about
15% of the population were in the "Mandarin and dialects" mode. While the in-
crease of Mandarin users in monolingual and bi/trilingual mode of use is im-
pressive (nearly 70% of Chinese population now using it in daily communi-
cation), the majority of the population (also approximately 70%, including
some non-Mandarin users) still hardly use English. The survey also revealed
that the people who were non-English-users tended to be from socio-economi-
cally disadvantaged groups. Consequently, we see the creation of the native
Mandarin population, though as impressive as the creation of the English
native population, had a quite different social basis.
The spreading of English and Mandarin in Singapore has different political
and economical impetuses. The government weighed the political and econ-
omic factors, when taking a language planning action, differently at different
times. The spread of English has been given the reign of economic forces while
channeling through a carefully safeguarded colonial heritage of sociolinguistic
stratification. The spread of Mandarin has been strongly politically driven and
strategically managed. Since 1981, the "Speak Mandarin Campaign" has been
explicitly given an additional purpose as preserving traditional Chinese cul-
tural values. Since 1985, the promotion of Mandarin has also been associated
with its economic values, responding to the opening-up and rapid growth of the
Chinese economy. Only in more recent years has the spread of Mandarin been
consciously raised to the level of defusing the spread of English. This is re-
flected in the demand for greater use of it in formal and government domains; it
also reflected in the recent campaign themes of urging the English-educated to
speak more Mandarin.
3.3. Maximizing resources
The Singapore government never ceased developing its pool of second lan-
guage users of English and Mandarin, apart from producing from it first lan-
guage speakers of these languages. At the same time, it attends to the favorable
conditions for its goals of language planning, never satisfied with the results it
has attained but striving for better ones. The use of second-language speakers to
breed first-language speakers in a creolization-like manner (cf. Gupta 1994) is
itself an effort to maximize resources. However, in the following we concentrate
on an analysis of maximizing the resources of a less tangible nature.
The policy of bilingual education was created as a language-planning ac-

tion. But once accepted, it was used to motivate further actions of language
planning. The All Party Report can be seen as a societal consensus on bilingual
education. The PAP government developed it into an integrative bilingual edu-
cation policy and put it into practice. Although the original Malay-knowing bi-
lingualism was replaced with English-knowing bilingualism in the education
policy, the essence of bilingualism has been kept till today. From the above
analysis, we already see that bilingual education is used to justify the "Speak
Mandarin Campaign". In the following we will see how the planning effect is
maximized within bilingual education itself.
An important move in the adjustment of bilingual education is indicated by
the Report on the Ministry of Education 1978 (published in 1979 and com-
monly referred to as the Goh Report in Singapore). This was an effort at re-
examing the educational policy and practices in Singapore for improvement.
Following the Goh Report, the bilingual education system has been reformed
into a differential-treatment system to address "the problem of education wast-
age", "ineffective bilingualism", while taking account of "variation in school
performance" (cf. Goh 1979). Following the recommendations of the Goh Re-
port, a "streaming" system was implemented, by which pupils were put into
different "streams" according to pupils' prior performances. The different
streams have different curriculums with different requirements of bilingualism.
The lowest-requirement system was called "monolingual" system, where pu-
pils were basically exempted from studying the "second languages" (the stu-
dents' ethnic languages). From the reformer's point of view, the educational re-
sources were efficiently used, and the students would perform better with more
realistic goals set according to their abilities. What was unnoticed in this re-
form is the institutionalization of differentiation of bilingualism which event-
ually paved the way to elitist bilingualism.
From the above example we can see that the Singapore government did not
simply build up on language education for its official languages. It took every
opportunity to maximize the conditions for its ultimately integrative goal in
language policy. While the old streams of education following the ethnic lines
were abolished, the new streams in differential bilingualism were established.
No political rationale was given as a consideration for the move. Instead it
was done in the pure form of education reform, with efficiency of education
and students' needs as the articulated purposes to serve. On the surface, it
looks like retrogression in scraping some of the requirements of bilingual edu-
cation. In fact, the integrative purpose of bilingual education was better
served. At the same time, streaming did bring more efficient and effective re-
sults in bilingual education, albeit at the sacrifice of its previously owning uni-
versal nature. At the same time the position of English in education was
further strengthened.
In Singapore we can see that the language planner always tried to further
develop what had been achieved, to strengthen, and to expand it. Perhaps lan-
guage planning is by nature a constant process of maximizing language and re-
lated resources.
Language planning in Singapore has perhaps already entered a fourth
phase. In this phase, the essence of bilingual education has changed from being
a means unto a goal. A new identity of Singaporean is being forged in the
image of the elitist English-knowing bilinguals while at the same time a new
diglossia with English as the Η variety and the other three official languages as
the L varieties is being pushed more explicitly (Pakir 2000; Chan 2000). At the
same time, bilingualism constructed on the new educational basis is indeed
suited to the country's economic development and its strategic position in glo-
bal and regional markets. The new dynamism is yet another example of how
language planners can build on what they have achieved. There is no end as to
how resources can be maximized.
4. Conclusion
Language planning in Singapore is successful mainly because the planner has

taken a management approach. While language issues are treated basically as
political, economic, and educational problems, the Singapore government has
given them persistent and deliberate attention, and consistently solved prob-
lems in accordance with its overall governing goals. This gradual, flexible,
and comprehensive approach exhibits the art of management. The depoliticiz-
ation and bilingual-education measures exemplify more specifically the strat-
egies which can be taken in such a approach. For the purposes of more general
applications, the most important lesson we can learn from the Singapore case
is perhaps that languages as well as their socio-economic value should be
treated as resources to manage and to develop in the language planner's per-
spective.
Note
1. For a number of years, all students categorized as being of "Indian race" had to take
Tamil as an obligatory "mother tongue" course, just as the Chinese students had to
take Mandarin and Malay students Malay.
References
Abudulla, Kamsiah and Bibi Jan Mohd Ayyub

1998 Malay language issues and trends. In: S. Gopinathan et al. (eds.J, Lan-
guage, Society and Education in Singapore: Issues and Trends,
179-190. Singapore: Times Academic Press.
Afendras, Evangelos A. and Eddie C. Y. Kuo (eds.)

1980 Language and Society in Singapore. Singapore: Singapore University
Press.
Anderson, Edmund A.
1985 Sociolinguistic surveys in Singapore. International Journal of The
Sociology of Language 55: 89-114.
Bourdieu, Philippe
1982 Ce que parier veut dire. Paris: Fayard.
Chan, Chiu Ming

2000 The 21s' century Challenge: The present situation and the future of the
Chinese Language in Singapore. Singapore: Times Media (in Chinese
with an Introduction in English).
Chen, Songcen, Xu Darning and Tham Wai Mum

2000 A research report on language use and attitudes of the Singapore Chi-
nese. In: Li Rulong (ed.), Studies on Southeast Asian Chinese, 48-90.
Beijing: Beijing Language and Culture University Press (in Chinese).
Chew, Cheng Hai, Xu Darning and Chen Songcen

1997 A Survey of Language Use and Language Attitudes in the Singapore
Chinese Community. Singapore: Centre for Chinese Language and
Culture, Nanyang Technological University.
Fishman, J. A.
1964 Language maintenance and language shift as a field of inquiry. Lin-
guistics 9: 32-70.
Ferguson, C. A.
1971 National sociolinguistic profile formulae. In: C. A. Ferguson (ed.),
Language Structure and Language Use, 157-70. Stanford: Stanford
University Press.
Goh, Keng Swee

1979 Report on the Ministry of Education 1978. Singapore: Ministry of
Education.
Gopinathan, S.
* 1998 Language policy changes 1979-1997: Politics and pedagogy. In:
S. Gopinathan et al. (eds.), Language, Society and Education in Sin-
gapore: Issues and Trends, 19-44. Singapore: Times Academic Press.
Gopinathan, S., Anne Pakir, Ho Wah Kam and V. Saravanan (eds.)
* 1998 Language, Society and Education in Singapore: Issues and Trends.
Singapore: Times Academic Press.
Gupta, A. F.
1994 The Step-tongue: Children's English in Singapore. Clevedon: Multi-
lingual Matters.
Gupta, A. F. and P. Y. Siew
1995 Language Shift in a Singapore Family. Journal of Multilingual and
Multicultural Development 16 (4): 301-314.
Haugen, Einar
1966 Linguistics and language planning. In: William Bright (ed.), Socio-
linguistics. The Hague: Mouton.
Jernudd, Β. H.
1982 Language planning as a focus for language correction. Language
Planning Newsletter 8 (4): 1-3.
Jernudd, Β. H. and J. V. Neustupny
1987 Language planning: for whom? In: Laforge (ed.), Proceedings of the
International Colloquium on Language planning, 71-84. Quebec: Les
Presses de 1'Universite Laval.
Kuo, Eddie C. Y.
1976 A sociolinguistic profile. In: Riaz Hassan (ed.), Singapore: Society in
Transition. Kuala Lumpur: Oxford University Press.
1985 Language and Society in Singapore. Taipei: Zhengzhong (in Chinese).
Kuo, Eddie C. Y. and B. Jernudd
* 1994 Balancing macro- and micro-sociolinguistic perspectives in language
management: The case of Singapore. Language Problems and Lan-
guage Planning 17: 1-12.
Kwan-Terry, Anne (ed.)
1991 Child Language Development in Singapore and Malaysia. Singapore:
Singapore University Press.
Lee Kuan Yew
1979 Mandarin or dialect? In: Speak Mandarin Campaign Launching
Speeches (1979-1989). Singapore: Ministry of Communications and
Information.
Li Wei
2000 Towards a critical evaluation of language maintenance and language
shift. In: U. Ammon, K. J. Mattheier, and P. H. Neide (eds.), The Fu-
ture of European Sociolinguistics, 142-147 (Sociolinguistica 14).
Tiibingen: Max Niemeyer Verlag.
Li Wei, V. Saravanan and Julia Ng Lee Hoon
* 1997 Language shift in the Teochew community in Singapore: A family do-
main analysis. Journal of Multilingual and Multicultural Develop-
ment 18: 1-21.
Pakir, A.
2000 Singapore. In: Ho Wah Kam and Ruth Y. L. Wong (eds.), Language
Policies and Language Education, 65-84. Singapore: Times Aca-
demic Press.
Pendley, Charles
1983 Language policy and social transformation in contemporary Singa-
pore. Southeast Asian Journal of Social Science 11 (2): 46-58.
Saravanan, V.
1998 Language maintenance and language shift in the Tamil community. In:
S. Gopinathan et al. (eds.), Language, Society and Education in Sin-
gapore: issues and Trends, 155-179. Singapore: Times Academic
Press.
Schiffman, H. F.
2002 French language policy: Centrism, Orwellian dirigisme, or economic
determinism? In this volume.
Stewart, W. A.
1962 An outline of linguistic typology for describing multilingualism. In:
F. Rice (ed.), Study of the Role of Second Languages in Asia, Africa
and Latin America, 15-25. Washington D.C.: Center for Applied Lin-
guistics.
Tham, Seong Chee

1990 Multilingualism in Singapore: Two decades of development (Census
Monograph 6.) Singapore: Department of Statistics.
Tay, Mary Wan Joo

1980 Trends in Language, Literacy and Education in Singapore (Census
Monograph 2.) Singapore: Department of Statistics.
Wong, Ruth Y. L. and Joyce E. James

2000 Malaysia. In: Ho Wah Kam and Ruth Y. L. Wong (eds.), Language
Policies and Language Education, 209-240. Singapore: Times Aca-
demic Press.
Xu, Darning, Chew Cheng Hai and Chen Songcen

* 1998 Language use and language attitudes in the Singapore Chinese com-
munity. In: S. Gopinathan et al. (eds.), Language, Society and Edu-
cation in Singapore: Issues and Trends, 133-154. Singapore: Times
Academic Press.
Xu, Darning and Pack Ling Tan
1997 Trends of English use among Chinese Singaporeans. In: A. Brown
(ed.), English in Southeast Asia 96. Singapore: National Institute of
Education.
Xu, Darning and Li Wei
in press The coexistance of languages: Language policy in Singapore. In:
Zhou Qingsheng (ed.), Nation, Ethnicity and Language: Language
Policy Studies. Beijing: Yuwen Press (in Chinese).
Xu, Darning and Tham Wai Mum
1997 Post-diglossia in Singapore. Paper presented at the 1st International
Symposium on Bilingualism. University of Newcastle-upon-Tyne,
UK.
1. What may be the benefits from the government's point view of language
planning and intervention?
2. What active steps has the Singaporean government taken in its language
planning since 1950s?
3. How are the Chinese-, Malay- and Tamil-speaking communities in Singa-
pore differentially affected by the government's policies?
4. How well does the notion of "diglossia" apply to the present sociolinguistic
situation in Singapore?
5. How should language planning be evaluated?
Managing languages at bilingual universities:
relationships between universities and their language
environment
Björn Η. Jernudd1
One avenue towards understanding issues of bilingualism is to relate interac-

tional-level communication practices to social, economic and political pro-
cesses (Li Wei 1994: 184). Social-linguistic analysis must account for "actual
language choice practices of the language user" and these are "constitutive of
social reality", reproducing and potentially changing social norms (Li Wei
1994: 186). Speakers deploy resources and are active participants in communi-
cation (Li Wei 1994: 168-169). They manage their speaking. One site which
allows study of the relationship between speaking and social processes is the
bilingual university.
In this paper, I will discuss how language management theory can diagnose
and suggest solutions to language problems that arise in a bilingual university.
1. Theory
1.1. The concept of language problem
Language problems arise both in and outside of discourse. According to lan-

guage management theory (see e.g., Neustupny 1994), language problems that
arise in discourse can be conceptualized as deviations from the norm that gov-
erns that discourse. The norm provides the behavioral standard against which
the speaker/writer notes, evaluates, adjusts deviations, and implements these
adjustments in his/her speech or writing. Implemented discourse adjustments
normally occur within a few turns of noting while speaking. Listeners and
readers, too, note, evaluate, and adjust deviations, and may communicate these
reactions to the interlocutor.
People may discuss what has been noted, evaluated or adjusted, i.e., they
make "on-line" discourse management a topic of deliberation "off-line". This
discourse about discourse (meta-discourse) may lead to problem solving. To
distinguish this overt, "off-line" reflection and discourse about language from
298 Björn Η. Jernudd
on-line discourse management, we can call this off-line mode 'directed lan-
guage management'. Negative evaluations of inadequacies in the flow of dis-
course which receive this kind of overt attention can be called language prob-
lems.
Problems may have solutions.2 Off-line, people may refer their concern
with language to specialist managers. Probably all speech communities have
their designated or self-proclaimed language specialists.
Regrettably, much decision-making about languages in education is made
without benefit of consultation with language management specialists. This
can be explained by the fact that all people practice language management
on-line, and possess a rudimentary concept of how language works. People
therefore think they can solve language problems. Specialists are surprisingly
confined to a few areas, such as, for example, language teaching, or to roles as
arbiters of grammatical "correctness".
Language problems also arise in other ways. People may, for example, pro-
ject language problems onto discourse on the basis of social, economic or
political interests (see Jernudd and Neustupny 1987). Some people may also
rely on mistaken lay theories about language and opine, for example, that
shorter words are better than longer words, that spellings should be "regular",
or that "mixing" of languages is harmful, and so forth.
People may also adopt a solution in principle but not in their own discourse.
For example, adults may endorse a standard variety of a language, even the use
of another language, but not practice it themselves. They do not accept the stan-
dard norm or use of the other language as valid for them. They continue to
speak in their own way (whether dialect or language). Yet at the very same time
they implement it for the upcoming generation through their control of edu-
cation. Pupils may be pliant enough to replace a home dialect norm with the
school's standard norm and to acquire another language if the learning environ-
ment is conducive enough to enable them to do precisely that. Other people be-
sides pupils who do not share a prediction of own future communicative fail-
ure, equivalent to evaluating an actual inadequacy in the relevant future com-
municative context, are unlikely to take the time to learn a new language.
People who do not note a deviation from the norm in their own use of language
are not likely to respond to suggestions that there is something the matter with
their language.
Managing languages at bilingual universities 299
2. Application
2.1. Relationships between language problems and the bilingual university:

the case of Hong Kong
Universities in the Americas and Asia operate in very different communicative

environments. For example, universities in the United States operate in an en-
vironment that foregrounds language problems associated with minorities is-
sues and issues of social justice, especially equal opportunity issues. Univer-
sities in Canada and especially in Quebec manage language problems that arise
because of political action by an assertively mobilizing French-speaking ma-
jority in Quebec and Francophone communities elsewhere in Canada. Minority
issues, ethnic issues, and regional integration issues are currently not official
concerns of the universities in Hong Kong. There are however some signs that
minority-ethnic issues are coming to the fore in public life in Hong Kong as
well, including language selection as a problem. Issues of discourse and power
and of gender equity in educational organizations also stand out as language
problems in a global search of literature on the nexus between language and uni-
versities - in addition to the obvious concerns with foreign language offerings.
There is a huge international literature that deals with the postcolonial de-
velopment of language policies of newly independent states but very few pub-
lications that deal specifically with the university's postcolonial transition. An
exception is Douglas 1986. There are some papers on more specialized con-
cerns as well, such as admissions policies and objectives for English depart-
ments in post-colonial societies (Zughoul 1985; Oman and Zughoul 1986).
What language problems actually arise in the communicative acts that com-
prise each category of situations at universities in, for instance, Hong Kong is
an empirical matter.3 To gather such data seems to me a new endeavour in uni-
versity language management and therefore in university management. The
data should be gathered as an exploratory fact-finding exercise. Data can be ob-
tained in quite a straightforward manner. Some of the more obvious approaches
to collecting such vital data are to observe and record actual communications,
to play back video-recordings of actual communicative acts to participants and
conduct so called follow-up interviews, to discuss texts (of any which kind)
with their writers and readers, and to use think-aloud and protocol methods.
Communicative acts in bilingual university settings can be classified into
the following broad and admittedly ad hoc categories:
Teaching acts between students and teachers,
- Study acts by students,
- Administrative acts between students, members of faculties and adminis-

trator representatives of university departments and administrative of-
fices,
- Research acts,
- Writing and other presentation acts,
- Service acts by members of faculties in communication with many differ-
ent audiences,
- Governance acts between representatives of the university and represen-
tatives of government offices and the public.
In the following I shall discuss language management in the first three cat-
egories of acts, with a focus on Hong Kong. Hong Kong universities operate
under a trilingual (English, Cantonese and Putonghua) and biliterate (Chinese
and English) policy. The use of Putonghua is largely confined to communi-
cations with visitors from the PRC and it is taught as a subject. Cantonese over-
whelmingly dominates spoken interaction at universities except when it is not
available as a shared language in which situations English is used. English
dominates as a textbook language, is therefore occasionally used in class, and
remains as a language of governance and therefore of board meetings.
2.1.1. Teaching acts and language management
Hardly any attention has been paid to the processes of use of English as a medi-
um of instruction in Hong Kong's bilingual universities. Attention to partici-
pants' noting of deviations in any and all communicative acts that define stu-
dent and teacher roles will likely reveal that participants experience a variety of
inadequacies in teaching acts. These could derive from any instructor's (lack
of) skills of using language for particular purposes of enabling learning, from a
student's (in)ability to comprehend and to seek clarification or a student's (lack
of) background knowledge to enable meaningful communication in the time
periods available, and/or from discontinuities of language use in the classroom
with language use outside it, not the least the use of English by instructors, and
subject instructors' inability to help students to manage English.
Equivalently, research will likely reveal a series of practices that by de-
clared interests of Hong Kong's trilingual and biliterate language policy would
have to be regarded as deviations but that serve the communicating parties well
in teaching acts. One such practice is the continuity of language use in the
classroom with language use outside it, not the least dominant use of Canto-
nese but also use of a Hong Kong English norm. Other such practices arise
from a variety of actual adjustment strategies, including in-class student-to-stu-
dent consultations in Cantonese, also the instructors' management of vocabu-

lary by efficient incorporation in discourse of technical/scientific terms, and,
generally, many instructors' skilled management of English to achieve student
comprehension and participation.
I claim that regrettably many teachers supply "corrections" that fall short of
what students themselves plausibly perceive to be deviations. Those correc-
tions are red-penned prescriptions that have little impact because they are ap-
plied to language that students themselves evaluate positively. My claim ap-
pears logically unassailable. Nevertheless, merely assume that my claim is
valid. If it is, teachers would have to engage the student in an interactive pro-
cess of a very different kind than a "corrective" one. The students would have
to be enabled to manage the generation of discourse according to norms that
they value as valid and they would have to be enabled to themselves make these
norms their own. After all, in the red-penning exchange I describe, the teach-
er's and the student's norms are in conflict, and it is the status differential rather
than a communicatively cooperative relationship that is being transacted. The
student has no basis on which to prefer the teacher's correction other than the
latter's authority. Teaching by sanction of authority produces compliant aping;
teaching by demonstrating the existence and validity of another norm through
directed language management techniques that aim at self-adjustment by the
student produces active learning.
Interestingly, face to face communication or follow up with little delay for
purposes of noting and evaluating discourse characterize good management of
language acquisition. For this process to work, students' communications have
to be discussed not 'next week' but within memory of their formulation, and
even then paradoxically not with a focus on noting deviations from language
norms but with a focus on possible, alternative expression of content.4 A cre-
ative focus in teaching acts on alternative expression of content implies attend-
ing to three types of deviations. There are deviations from rules of grammatical
competence, from rules of sociolinguistic competence, and from conventions of
the sociocultural context of communication5. The importance of focussing on
expression form has been noted (see Doughty and Williams 1998); however, far
too little explicit attention is given to sociolinguistic competence and sociocul-
tural context. An example of attention to sociolinguistic competence is teaching
the in-house 'research paper'. There is obviously ample scope for radical re-
thinking of directed management of communication teaching at universities.
The language management approach goes far beyond the conventional
grammatical proficiency thinking that dominates management of the use of
English at universities. The implications are as profound as they are obvious:
- involve many more participants who engage in diverse communicative

acts at a university in language management than at present,
- enable participants in a broad range of teaching / learning situations to
work together on improving communication as an ongoing activity that
constitutes the situations,
- especially harmonize university teaching acts with realities of partici-
pants' norms of communication.
Hong Kong universities have been under continuous pressure from outside in-
terests to train students to know (more) English. As far as the universities are
concerned, this is an interest-based request for university action in favor of
English. One of the effects of this pressure is to uphold the policy of English-
medium instruction in universities.
2.1.2. Teaching acts: managing the medium of instruction
From the point of view of language management, the difficulty with imple-
menting English at universities is that the existing and valid norm of interaction
between students in Hong Kong is Cantonese and its associated written variety
of Chinese.6 Students do not perceive a deviation from their norm of language
selection and use when they use Cantonese. They do perceive a deviation from
their norm of language selection and use when they use English among them-
selves. The only acts for which they accept English as a norm are acts in which
foreigners participate. This simple fact renders the English medium policy an
imposition on students. Students reasonably require ample justification for
them to consider adopting English to replace Cantonese in interaction for
which Cantonese is the norm. The English medium policy is a hegemonic pre-
scriptive norm unto students.
The English-medium policy upholds a prescriptive norm because as a
matter of fact at universities in Hong Kong most students do not practice Eng-
lish other than when no other language is available. Students use English in
some communications: they have no choice in many subjects in which they
have to read assigned English texts (which they return in English in the written
examinations)7, when lectures are delivered in English, and so on. These pre-
scriptions do not invalidate the fact that an English selection norm is not valid
for them. Were a student to select English, that selection act would itself be a
sociocultural and sociolinguistic deviation which would oblige evaluations by
other participants. This is obviously a burden on communication.
The prescriptive norm is not given the push into a possibly existing one for
the students because as a matter of fact most instructors readily speak Canto-
nese when so addressed by students (or Putonghua if they do not have Canto-
nese). Many lecturers offer students a choice of language of lectures and stu-
dents normally select Cantonese.
An additional and perhaps paradoxical complication is that although stu-
dents do not practice English they endorse the prescriptive norm in principle,
despite the fact that they practice Cantonese. Individuals react to this contra-
diction in different ways. Some may feel inadequate and devalue their Chinese
even though that is the language they practice.
Although more people know more English than ever before in Hong Kong
(Bacon-Shone and Bolton 1998), the 1997 transition sidelined English in pub-
lic communication. Privately the mass of the mobilized and mobilizing popu-
lation of Hong Kong speak Chinese; for foreign-marked communications they
recognize the value of English and many practice it, in and out of the university.
English has been and is used in order for instructors to be able to communicate
with pupils / students in Hong Kong: either the teachers learn Chinese or stu-
dents learn the teachers' languages.8
Government policy unambiguously favors Chinese while recognizing the
value of English. For universities to limit the use of Chinese under these cir-
cumstances would be strange indeed, yet, that is what universities continue to
do, contrary to participants' norms. The Hong Kong policy of trilingualism
(lately with the addition "biliterate") mediates the seeming impasse between
use of English and Chinese in universities. In my opinion, the ambiguity has to
be resolved and it can only be resolved by putting English in place, in the right
places. There is every reason to think that participants would readily welcome
specification and learn and use English in those specific contexts.
One might object that being a student means to comply with prescription. I
counter that such pedagogy is costly indeed. Justifying a suggested norm to
motivate its acceptance (Bartsch 1987: 148) requires an investment cost but in
the total analysis is less costly than prescription. Students signing up to seek
expert help with adjustment of the inadequacy that they themselves have
noted - that they do not know enough English - is the criterion of efficient
management here. This means that English would be taught as a foreign lan-
guage in specially arranged acquisition classes to those who require grammati-
cal competence beyond their secondary school achievement. Such teaching
would be supplemented by the use of English in specially designated subject
matter classes to enable acquisition of sociolinguistic rules and subject-specific
languages and to enhance fluency by the select number of students specialising
in the subjects for which English is motivated.
A course for which the use of English would seem motivated is the English
language and literature major. The university could issue certificates to any stu-
dent who demonstrates that s / he knows English (by varied means of assess-
ment). The devolution to departments or even individual teachers of the deci-

sion in which language to teach, a "privatisation" of language policy, will free
members of the university to manage language problems of their own noting.
Some learning could also be placed overseas in a community that uses the lan-
guage of demand, as is already the practice for European Studies, students of
German and French and Translation and Chinese Studies' students for Putong-
hua at Hong Kong Baptist University.
It is not necessary to impose English on university students. Students
readily respond themselves to cross-community opportunities of foreign travel,
study and work. However, a foreign language should not get in the way of the
first business of a university which is to develop a spirit of inquiry and an in-
terest in exploring new knowledge. This business is best carried out in a lan-
guage which is already shared and effortlessly used by students' and most ad-
ministrators' families, indeed, by most families in Hong Kong. That language
is Cantonese. The university should be responsible for producing people who
can cope with communication in an international society. Hence, the university
should position topics, subtopics, foreign teachers, international students and
foreign settings for study by its own students in a way that obliges students to
master communication in English.
2.1.3. Study acts
A similar approach must be taken to the category of study communications.

Within students' ranges of proficiency in various languages and varieties of
languages, what do we know about the individual student's process of manag-
ing reading and writing for different study purposes? More specifically,
- What norms do students by their own choice rely on in self-adjustments,
e.g., when writing an essay?
- Do existing university support systems know about students' own norms?
- What norm authorities do students refer to for solutions to language prob-
lems in either English or Chinese?
Ceteris paribus, I advocate a pedagogy that aims at enabling students to note
deviations themselves and therefore to themselves explore evaluations and ad-
justments - thus empowering the students. This pedagogy is a cost-effective
one to enable learning. Happily, university leaders in Hong Kong advocate
change that supports a pedagogy that involves the learner in discovering possi-
bilities and opportunities that explains choices to the learner and that promotes
self-realization and student initiative in the study process. A university study
system along these lines would build support systems to enable learning ac-
cording to student choice. This is compatible with the language management

approach.
2.1.4. Administrative acts

I have few facts about inadequacies in administrative acts of communications
between administrators and faculty members. Clearly, descriptive research is
much needed.
Administrative acts: Terminology

A recurring problem much in need of solution is in-house standardization of
vocabulary usage. An example is the use of "subject" and "course", in English,
across the university. In written documents, "subjects" constitute a "course".
The use of course with reference to the three-year program and subject with
reference to its semester-long components would become apparent from the
development of the text. Yet, this specialization of meaning conflicts with gen-
eral usage. In committees, this distinction is often not maintained in speech,
which results in temporary confusion. Discourse management normally re-
solves the problem as is usually the case in spoken interaction, but at a cost.
Therefore, raising awareness of specialized usages, i.e., of terminologies, in
university administration would be beneficial. There is room for initiatives to
evaluate university vocabularies, and to compile and distribute term lists.
I suspect that the terminological problem is even greater for the now esca-
lating use of Chinese in in-house documents and work-related discourses than
for English. Terminological problems would include finding suitable words in
Chinese for English terms, in addition to negotiating agreements on preferred
terms and their definitions. Terminological solutions should be shared system-
wide in Hong Kong, and should include the University Grants Commission, the
Education Department and the Legislature.
Administrative acts: bilingual information

In regard to administrative communications with students, the Hong Kong
Baptist University is now adjusting for the absence of subject descriptions in
Chinese "for China-related subjects". The Registrar is acting on a student's
suggestion which apparently mobilized widespread shared noting of the ab-
sence of Chinese subject descriptions in the Calendar I Bulletin. The adjust-
ment furthermore partly aligns a major university document with the univer-
sity's policy to be bilingual. I predict extension of bilingual descriptions to
other subjects. Whether this language problem is motivated by the opinion that
Chinese ought to be used more or by a concern to make the descriptions easier
to read for some group of readers who would otherwise struggle in English is
moot.
Administrative acts: internationalisation

The presence on Hong Kong campuses of English-speaking international
students is a positive new development. They do not know normally Cantonese.
Therefore, they now reveal the complexities of language use associated with
the policy and belief that English is used as a "medium of instruction" in
classes where teacher and students share Chinese. Some international students
at the Hong Kong Baptist University noted Chinese language usage in subjects
they took to the extent that they had to withdraw. These classes were designated
as "English medium" subjects. The problem came to the attention of the Aca-
demic Registry and the International Student Office.
I suggest that English is rarely exclusively used. This may not however be
commonly understood. Even if administrators realize the reality of language
use, they believe that English could be exclusively used. There is obviously
room here for description of actual language use in the supposedly "English
medium" classes.
In light of such descriptive findings, and aided by reasoning based in the-
ories of language use, there is room for rethinking how class discourse should
be managed. An absolute condition is that in classes with international stu-
dents, these students feel included in the teacher's communications. The prob-
lem now overt, since teacher and student face each other in class, the teacher
should be able to manage his/her language to that extent at least. To insist on
English only (and what kind of English?) would appear to be an unnecessarily
restrictive solution.
It is a different matter and an open question whether international students
will negatively evaluate their local peers' norm that Cantonese be used in per-
son-to-person interaction in discussion groups and asides in class. Do they feel
excluded by such language use by classmates? Might they want to drop out of
those subjects for that reason, thus widening rather than narrowing the social-
communicational gap between local and foreign participants in the university?
I am myself true to the declaration of contents for the subjects I teach as Eng-
lish medium because I do not speak Chinese; yet students nevertheless deprive
me of essential background information for achieving optimal communication
in teaching by their exclusionary practice. One obvious educational cost is that
I do not know whether discussion groups in class are on topic. And the students
miss a chance to practice English. What is a greater cost, however, is that the
classroom interaction reproduces the sharp social division between the locally
Chinese and the foreign others in Hong Kong social structure. The exclusion-
ary use of Cantonese is but one exponent of a bilingualism which is not recip-
rocal in Hong Kong.
2.2. Evaluation
There is much room for clarification of language use in all categories of com-
municative acts to bring about sound discourse management to meet the bilin-
gual university's educational goals. I recommend audits of language use to in-
form directed language management. Language management specialists would
work with university administrators and faculty members and students with the
goal of removing the likelihood of systematic discourse inadequacies across all
communicative acts in the university. Such joint language management action
would necessarily have to make explicit the motivated relationships between
language behavior and interaction in the university to the surrounding social
and economic structures. Not only would such study affect the very ways in
which the university conducts its educational business but it would also reveal
the complex realities of bilingualism in practice.
Notes
1. An earlier version of this paper was presented at the International Seminar on

Changing Patterns in University Management, Tsinghua University, Beijing,
April 12-14, 2000.1 want to thank the following individuals who attended a depart-
ment seminar in which a draft of this paper was discussed for their helpful com-
ments: Ed Anderson, Gillian Bickley, Stuart Christie, Clayton MacKenzie, K. C. Lo,
Sui Sang Mok, Chew Kheng Suan, Linda Wong, and Suying Yang. I am indebted to
Professor J. V. Neustupny who commented on a draft of this paper in some detail.
2. It follows that solutions to language problems do not exist as adjustments according
to anybody's norm until someone uses them to generate on-line discourse. They re-
main prescriptive until they are implemented in online discourse.
3. Analysis of power hegemony, or a discourse ethical approach may arrive at similar
results. For alternative approaches, I recommend Corson 1995.
4. I give my views on suitable organization of teaching in Jernudd 1994.
5. For elaboration and application of this distinction, see Neustupny 1987.
6. This written variety is classified as locally valid, as in some sense Cantonese, and
not classified as Putonghua.
7. This latter claim that students return English because they 'learnt' in English is hy-
pothetical. I maintain that students would write in Chinese if students were to accept
an approach to studies that implies that examinations reflect cognitively creative
processing of content, say, as problem-solving. My claim can be tested by investi-
gating particular students' approaches to learning and predicts different preference

of selection of language according to learning style.
8. Gillian Bickley, oral communication, 28 February 2000; see also Bickley (1997:
280).
References
Bacon-Shone, J. and K. Bolton

1998 Charting multilingualism: language censuses and language surveys in
Hong Kong. In: Martha C. Pennington (ed.), Language in Hong Kong
at Century's End, 43-90. Hong Kong: Hong Kong University Press.
Bartsch, R.
1987 Norms of Language: Theoretical and Practical Aspects. London:
Longman.
Bickley, G.
1997 The golden needle. The biography of Frederick Stewart (1836—1889).
Hong Kong: David C. Lam Institute for East-West Studies, HK Bap-
tist University.
Corson, D. (ed.)
*1995 Discourse and power in educational organizations. Cresskill, NJ:
Hampton Press, Inc.
Douglas, D.
1986 From school to university: language policy and performance at the
University of Khartoum. International Journal of the Sociology of
Language 61: 89-112.
Doughty, C. and J. Williams (eds.)
1998 Focus on form in classroom second language acquisition. Cambridge:
Jernudd, B.
1994 Improving tertiary learners' professional writing skills. Journal of
English and Foreign Languages 12: 34-39.
Jernudd, B. and J. V. Neustupny
* 1987 Language planning: for whom? In: L. Laforge (ed.), Proceedings of
the International Symposium on Language Planning, 69-84. Ottawa:
Les Presses de l'Universite Laval.
Li Wei
*1994 Three Generations, Two Languages, One Family. Language Choice
and Language Shift in a Chinese Community in Britain. Clevedon/
Philadelphia/Adelaide: Multilingual Matters.
Neustupny, J. V.
1987 Communicating with the Japanese. Tokyo: The Japan Times.
1994 Problems of English contact discourse and language planning. In:
Thiru Kandiah and John Kwan-Terry (eds.), English and Language
Planning: A Southeast Asian Contribution, 50-69. Singapore: Times
Academic Press.
Oman, Khalid and Muhammad Raji Zughoul
1986 Towards formulating an educational admission policy for English ma-
jors at Arab universities: a case study. Report. ERIC Document
#ED271947.
Zughoul, Muhammad Raji
1985 Formulating objectives for the English departments in Arab univer-
sities: rationale and assessment. Guide. ERIC Document #ED271943.
1. What is language management?

2. What are the differences between language management and language
planning?
3. Meta-discourse may lead to problem solving. Discuss.
4. What communicative acts can be identified in your own bilingual edu-
cational setting and how are they related to language management?
5. How should we evaluate language management in bilingual education set-
tings?
Using descriptive inquiry to transform the education
of linguistically diverse US teachers and students
Ofelia Garcia and Cecelia Traugh
With the faculty of LIU /Brooklyn Teaching and Learning Department and Cy-
press Hills Dual Language Community School1
1. Introduction
Education in the United States in the first decade of the 21st century will be in-
creasingly characterized by the polarization of two opposing trends. On the one
hand, the linguistic heterogeneity of the school population is increasing. On the
other hand, standard academic English is being used in standardized assess-
ment as the only measure of school success and of the achievement of high
standards for both children and their teachers.
In the 70s and the 80s, as the multilingualism of the country started to ex-
pand through immigration, the use of languages other than English (LOTE) by
students and teachers in bilingual education classrooms was prevalent in the
nation's schools (For more on bilingual education during this era, see Arias and
Casanova 1993; August and Hakuta 1998; Baker 2001; Crawford 2000; Garcia
1991; Garcia and Baker 1995; Ovando and Collier 1998). But as the bilin-
gualism and bidialectism of U.S. citizens started to increase, state education
systems found ways of halting the use of LOTEs in schools, sometimes by pas-
sing strict regulations such as Proposition 227 in California that forbids the use
of a LOTE in instruction, but most often by imposing that students and teachers
pass high stakes standardized tests in standard English that require sophisti-
cated use of decontextualized language skills. These high-stakes tests increas-
ingly determine who gets promoted, who graduates from high school, who
enters college, who teaches.
The focus on passing standardized tests and meeting high standards, always
assessed in English only, has had a debilitating effect on the use of children's
native languages in education. Bilingual teachers, fearful that their students
will not pass the English-only assessments and that they will receive low marks
on their teaching, increasingly teach in English only. Parents with children in
bilingual classrooms have become fearful that their children will not be pro-
moted or graduate. And college programs to prepare bilingual teachers, once
312 Ofelia Garcia and Cecelia Trough
numerous and growing, have shrunk in enrollment, as the bilingual teachers

fail standardized tests that put their existence in jeopardy. Bilingualism in
schools in the United States in the 21st century is seen as a threat to the high
educational standards of an "advanced" society.
This chapter first describes the increased sociolinguistic heterogeneity and
socioeducational homogeneity of the United States and of New York City in
particular. It then looks at how Descriptive Inquiry as an innovative research
mode has enabled us, at this threatening juncture, to transform a university pro-
gram to continue to educate urban teachers, most of whom are bilingual or bi-
dialectal, and support the efforts of a dual language school to educate for bilin-
gualism.
2. Sociolinguistic heterogeneity in the United States and NYC

in the 21st century
The United States has become increasingly multilingual in the last decade. Ac-
cording to the 2000 Census, a language other than English (LOTE) is spoken in
approximately 18% of all U.S. households. In states like California, New
Mexico, Texas, New York, Hawaii, Arizona and New Jersey, well over 25% of
the population live in households where a LOTE is spoken (NCBE Newsline
Bulletin 2001).
School age children increasingly live in LOTE-speaking households. From
1990 to 2000, the number of children ages 5 to 17 living in homes where
LOTEs were spoken grew approximately 55%, making up a total of almost
10 million children (9,700,000). Children in Spanish-speaking homes showed
the greatest increase from 1990 to 2000, an increase of 60%. Children in homes
where Asian and Pacific languages were spoken grew by 49%. It is interesting
to note, however, that the percentage of children living in LOTE speaking
homes who also speak English well or very well increased to 86% in 2000, in-
dicating a growing generation of bilingual North Americans. At the same time,
the total number of school-aged children who are English language learners
has increased dramatically, mirroring the growth of immigration in the last dec-
ade of the 20th century (NCBE 2001).
The linguistic diversity of the United States reflects not only its multilin-
gualism, but also the complexity of the English varieties spoken, both non-
natively, as well as natively. The English spoken includes African American
varieties, as well as the English spoken by the many recent anglophone immi-
grants - Indian English, African English, and West Indian English.
Using descriptive inquiry to transform US teacher and student education 313
In New York City, the setting for the teacher education program and the
dual language program that will be described, almost half of households (46%)
spoke a LOTE in 1990 (Garcia and Fishman 1997). From 1990 to 2000, the
foreign born population increased from 28% to 40% (Lambert 2000). A half a
million of these immigrants arrived in the city from 1990 to 2000 (Kaplan
2001).
What distinguishes New York City from other urban U.S. contexts is its lin-
guistic complexity (Garcia and Fishman 1997). In 1990 there were more than
1,000 speakers of fifty-two different languages, and one of five New Yorkers
spoke Spanish at home (Garcia and Fishman 1997). A decade later, the city's
largest group of LOTE speakers is still Spanish speaking. In 1999, Puerto Ri-
cans account for 744,000, of whom 290,000 were born in the island. In a dec-
ade, Puerto Rican New Yorkers went from constituting 50% of the population
to 37% (Navarro 2001). Dominicans numbered 387,000 in 1999, and the
number of Mexicans quadrupled in one decade, reaching 133,000 in 1999. Im-
migrants from the former Soviet Union numbered 229,000 in 1999, while
South Asians from India, Pakistan and Bangladesh, more than doubled in one
decade to 146,000 (Lambert 2000).
3. Educational homogeneity in the United States and NYC in

the 21st century
The United States educational system has always distinguished itself by its
lack of centralization and the absence of a centralized national curriculum.
Educational policy has traditionally been set by each state's Education depart-
ment within federal guidelines. But the move towards high standards that are
assessed through high-stakes standardized tests increasingly homogenizes edu-
cational policy and practice (AERA 2000). As schools step up their students'
preparation for standardized tests that determine whether the students will be
promoted and /or graduate, and whether the school itself and its administration
and teachers will be put on probation, instruction has become standardized
(National Research Council 1999). The educational publishing industry and
the testing industry are meanwhile making substantial profits in what has be-
come a huge industry of material to prepare for and to take the standardized
tests.
Upon admission to New York City public schools, LOTE speaking children
are tested with a standardized test. Those who do not meet appropriate English
language proficiency are categorized as English Language Learners (ELLs).
But it now turns out that these same ELL students are being asked to pass as-
sessments for promotion and graduation that require English language profi-
ciency.
English Language Learners in NYC High School must now pass a six hours
exam in English, given by the State Regents, in order to graduate. It has been
predicted that in 2005 when all the provisions of the new graduation policy will
be enacted, there will be neighborhoods in New York City with few high school
graduates. And after three years, elementary and middle school students who
continue to be categorized as ELLs will be required to pass standardized as-
sessments in English reading and writing in order to be promoted. Those test
scores, bound to be lower in schools in which there are large bilingual, immi-
grant, and poor populations, are used to evaluate the school's performance in a
public "Report card".
In this climate of higher standards assessed through English only, English
Language Learners and bilingual and bidialectal students have become a liabil-
ity. Teachers would rather not teach them since their standardized test scores
are likely to be lower than those of native English speakers in wealthier com-
munities. Schools would rather not have ELLs since the school's performance
is going to be evaluated through the lower scores that English Language
Learners will likely receive. But fortunately, education is compulsory until the
age of 16 in the United States, leaving teachers of bilingual students and
schools with large numbers of English Language Learners scrambling for sol-
utions; and leaving those committed to bilingual education and to teaching
English Language Learners unsure and doubtful that their students can be suc-
cessful.
Higher education, however, does not have to find solutions to these prob-
lems and can simply deny access to students who fail to meet these high stan-
dards on standardized tests. The New York State Education Department has
started requiring passing scores in new standardized teacher certification
examinations, which many bilingual and bidialectal teachers fail to pass. The
three written standardized exams - one in Liberal Arts and Science, the other in
the area of specialization, and the other in teaching skills - consist of multiple
choice questions that test ability to solve content-based problems embedded in
reading passages, as well as a written essay. The New York State Education De-
partment also requires that 80% of those graduating from teacher certification
programs pass the examinations (NYS Board of Regents 1998).
There are differences in performance in the New York State Teacher Cer-
tification Examinations among different ethnolinguistic groups. In 1996 to
1997, the last year for which data was publicly released, 92% of white teacher
candidates passed, but the pass rates of African Americans and Latinos was
only 50% and 47% respectively. One third of teacher candidates in New York
City fail the examinations that qualify them to teach, as compared to one-sixth
overall in the State (NYS Board of Regents 1998).
The high failure rate of bilingual and bidialectal students make them a lia-
bility for admissions in teacher education programs. Colleges, even those with
policies of open admissions, have started to close the door on the very students
who are deeply knowledgeable of language minority communities and their
ways of using languages, literacies and cultures, and who are committed to
educating children in those communities (Garcia and Trubek 1999).
4. Descriptive inquiry in educational research
To protect the gains made during the later part of the 20th century in educational
access and the use of the child's native language in instruction, education pro-
grams, both at the university and the school level, have had to undergo trans-
formations that enable them to continue their commitment to bilingual and bi-
dialectal students, while increasing their rates of success in standardized tests.
But meaningful change cannot be imposed. In the case of the university pro-
gram and the dual language program that we describe below, descriptive in-
quiry as a research mode has enabled both faculties to work collaboratively to
spur, support and imagine change. Descriptive inquiry has helped both groups
create a generative tension between what could be two sets of warring val-
ues - those of heterogeneity and those of homogeneity.
Descriptive inquiry is a disciplined process of research in teaching and
learning derived from the work of the Prospect Center for Education and Re-
search (Carini 1993; Carini 2001; Himley and Carini 2000; Prospect 1986) and
based on ideas drawn from phenomenology. Driven by story, image and detail,
descriptive inquiry grounds itself in the works of both students and faculty and
description of students and teachers. Through disciplined description, a group
can collaboratively cut through generalities and abstractions, make the com-
plexity of the lived reality more visible, and so enlarge understandings that can
generate ideas for action (Traugh 2000).
The descriptive inquiry process that we have used in the transformation of
the LIU/Brooklyn teacher education program and in the support of the Cypress
Hills Dual Language School engages the entire faculty and staff over a long
period of time in disciplined descriptive inquiry into language, texts, children's
and teacher's works, and teaching practices. Each participant describes fully,
working to withold judgment or interpretation and to be respectful of the maker
of the work, and of each other (Carini 2000, 2001). A chair pulls together the
main threads of the descriptions through the course of each session. The pro-
cess enables the group to collectively learn new ways to talk with each other
and to generate questions and ideas about the focus at hand. Over time, a body
of knowledge about children, teaching, and learning is developed among
members of the group.
Descriptive inquiry is multi-faceted, and inquiry groups chart their own
courses depending on their interests and the questions they pursue. The pro-
cesses, as described below, were developed by the Prospect Archive and Center
for Education and Research (Prospect 1986).
- Descriptive inquiry into language engages the group in describing im-

ages, metaphors, associations, that the individual holds of a word and the
ideas it embodies. Words are chosen by the chair of the group for the in-
sights into the group's questions they may provide.
Descriptive inquiry into texts begins with literal paraphrasing of each
line of text, as each of the readers tries to gain deeper understanding of the
author's words. It then moves to a more interpretative level, with each
member of the group adding a layer of meaning. Texts can be ones written
by teachers themselves, e.g., a mission statement, or articles or books
chosen because of their relevance to the inquiry's focus.
Descriptive inquiry into children's works requires that members of the
group start out by paying close attention to detail and describing the
child's work literally. After exhaustive literal description, the group
moves to figurative descriptions.
- Descriptive reviews of students engages teachers in describing students
fully under five headings - physical presence and gesture, disposition and
temperament, connections with others both children and adults, strong in-
terests and preferences, and modes of thinking and learning (Carini
2000). Parents are also often engaged in this process.
Descriptive reviews of teaching practice helps a teacher, in the com-
pany of colleagues, take an inquiry stance to her/his work by asking a
framing question and showing "the rough edges of work" through de-
scription (Traugh 2000). Colleagues grapple with understanding the per-
spective of the person presenting and imaginatively responding to the
question.
The chair is responsible for writing and disseminating descriptive notes of the
inquiry sessions. These richly detailed notes provide a valuable historical rec-
ord of the development and the movement in the group and of the body of
knowledge developed by the group over time.
As we will see below, descriptive inquiry has proven to be an important

means of making a transforming and sustaining space for bilingual teachers
and students as the education system has become more homogenized.
5. Transforming the education of urban and bilingual teachers
The Brooklyn Campus of Long Island University is a private university in

downtown Brooklyn with an 80% minority student body, mostly poor and im-
migrant, and a mission of access. Although the university does not have an 80%
pass rate on the New York State Teacher Certification Exams, the institution is
committed to strengthening its teacher education programs to increase the like-
lihood that their urban minority students, most of them bilingual and bidialec-
tal, pass the tests. One of us (Garcia) has been the Dean of the School of Edu-
cation since 1997. The second one (Traugh) is the Director of its Center for
Urban Educators (CUE).
To transform the teacher education program, we all had to go beyond tradi-
tional academic discourse, discourse that encourages intellectual generaliz-
ations, answers, and solutions, and discourages questions that problematize
and bring out complexity. We also had to cut through intellectual generalities
and together understand the societal and educational challenges and imagine
pedagogical and curricular solutions.
In 1998, with Traugh as consultant, the faculty and staff started to use De-
scriptive Inquiry on a monthly basis to study our own reality. Almost immedi-
ately, questions about the language differences of the teacher candidates
emerged, questions that parallel the ones teachers in classrooms have about
their own language minority students. For example, during the third inquiry
session a faculty member offered an extensive description of her practice and
raised an essential issue. The descriptive notes of that session record the faculty
member's voice:
When I began teaching at LIU/Brooklyn, I was overwhelmed by the academic, social

and personal struggles of my students. As a result, I did not know how to facilitate
their intellectual development. In acknowledging their weaknesses, I also had to ac-
knowledge my own. I realized how little I knew about teaching linguistically and cul-
turally diverse students how to teach (...) My students need to be fluent in Standard
English in order to be successful in college and in a career as an elementary school
teacher. They also need to retain their pride and commitment to their native languages
and cultures (...) I need to facilitate their entrance into Standard English, while I allow
them to give voice to their own literacy backgrounds. (November 19, 2001: 5)
Collaboratively we grappled with how to pay close attention to the language

and cultural differences of our teacher candidates. Through the process, we
slowly discovered the strengths embedded in those differences, and we learned
how to help students meet society's expectations without standardizing their
intellectual development.
The inquiry sessions have generated tensions. However, through the ques-
tions it has raised, descriptive inquiry has enabled us to transform the pedagogy
and the curriculum used to educate teachers, and, in the process, transform our-
selves as scholars and researchers.
Since our bilingual and bidialectal students of teaching need intense lan-
guage and literacy development, the new teacher education courses engage stu-
dents in doing close readings of complex texts using descriptive inquiry of texts
and words. To help the students read critically and write intensively, we've de-
veloped a transformative pedagogy that we have called "of the borderlands." In
her 1987 book, Borderlands I La Frontera, Gloria Anzaldua reminds us: "In the
Borderlands / you are the battleground / where enemies are kin to each
other/ (...) To survive the Borderlands/you must live sin fronteras /be a cross-
road". Our pedagogical practices attempt to create a generative third space that
joins faculty, students and their students in the shifting urban landscape. For
example, students use double-entry journals to describe their multilingual and
multicultural experiences alongside excerpts they copy from their theoretical
readings. This is a way for students to grapple with large ideas as they construct
meaning from their own particularity. This is also a way for faculty, mostly
white and not bilingual, to gain a different access to the students' experience
and thinking, and so develop deeper understandings of familiar traditional the-
ories of teaching and learning and generate new ideas.
Alongside pedagogy, the teacher education curriculum has undergone
major transformation. The traditional starting place of teacher education pro-
grams has been the foundational disciplines of history, philosophy, sociology
and psychology of education. Through our descriptive inquiry work, the fac-
ulty shifted the starting place of the program to focus on the persons who are
the object of the education, that is, the teacher and the children. This is done
through two initial foundational interdisciplinary courses which also introduce
students to the processes of descriptive inquiry, developing in the students the
habits of mind of noticing, of observing and describing, of remaining "wide
awake" to differences (Greene 1995). The entire curriculum has also been
grounded in work in the work of schools and classrooms by requiring fieldwork
from the very beginning.
We're in the process of developing an additional facet of the on-going in-
quiry into our program and the opportunities it provides into our students'
thinking and language. We are creating longitudinal collections of our stu-

dents' work, and faculty are collaboratively engaged in descriptive inquiry into
what these collections hold.
All courses in the curriculum address the teaching of diverse learners, in-
cluding bilingual and ESL students. In an increasingly diverse country, all
teachers, whether specializing in the education of English language learners or
not, need to know about the interdependence of the first and second language in
cognition and learning, to be ready to use the child's native language to further
the child's thinking and imagination, and to know how to adapt instruction
when teaching in a child's second language. In an increasingly interdependent
world, all teachers need to encourage English speaking and LOTE speaking
students to use other languages, other literacies, other sources, to seek plural
and multiple sources of data and information to understand our complex glo-
balized world.
Our inclusive vision has also had an impact on the way we educate special-
ists to teach English language learners. Besides taking specialized courses,
these teachers, whether ESL or bilingual teachers, are now engaged in a year-
long collaborative inquiry, along with other students of teaching, as together
we enlarge our understandings of the complexity of all school children.
Descriptive inquiry created sufficient generative tension in the group, so as
to open up spaces for creativity, imagination, and regeneration. And, it contin-
ues to support the faculty and staff as a new curriculum and new pedagogy is
implemented, through the on-going development of ways of teaching and
learning that will ensure that more bilingual and bidialectal teachers achieve
New York State certification.
6. Supporting the education of bilingual and biliterate children
Educational scholarship and university-generated research has traditionally

been either conforming to, or critical of, the educational system, but has paid
little attention to noticing the child or the teacher-to-be as active learners and
teachers in the making. Traditional scholarship on bilingualism in schools has
rarely systematically studied works done in school by bilingual children and
bilingual teachers as complex landscapes deserving of intellectual attention.
In September of 2000, one of the public bilingual schools in New York City,
Cypress Hills Dual Language School in the East New York section of Brook-
lyn, received funding for staff development. They requested that LIU faculty
become involved in this effort. Rather than traditional staff development activ-
ities, the two of us started collaborating with the school faculty and staff in
320 Ofelia Garcia and Cecelia Traugh
whole-school descriptive inquiry, a process that Traugh had used successfully

in many other schools (Traugh 2000).
Cochran-Smith and Lytle (1998) have argued that the distinction between
theoretical knowledge and practical knowledge in teacher education has main-
tained the hegemony of university-generated knowledge by giving practical
knowledge low status. This serves, they say, "to reify divisions that keep teach-
ers in their place - the separation of practitioners from researchers, doers from
thinkers, actors from analysts, and actions from ideas" (Cochran-Smith and
Lytle 1998:289). Descriptive inquiry has allowed us to set this distinction aside
and so to start inhabiting a space of questions and possibility with bilingual
teachers, alongside teachers, to be companions, as we remain attentive to the
changing landscape of their bilingual classrooms.
The Cypress Hills Dual Language Community School was started in 1997
through the efforts of a group of parents in the local community based organ-
ization. Parents, both Latino and African American, wanted the school to de-
velop the bilingualism of their children, regardless of whether they were Eng-
lish monolingual students, Spanish monolingual students, or bilingual. Cy-
press Hills is the first public school to have a parent co-director who is bilin-
gual, Maria Jaya-Vega, with a teacher co-director, Sheryl Brown, a bilingual
Anglo. The school was reorganized two years ago with an innovative five-day
cycle. Linguistically heterogeneous groups of children are instructed in one
language for five days in a row, with the second language used for the next five
days. Every five days the children switch language, classroom and teacher. Al-
though the children are linguistically mixed, languages are strictly separated by
classroom and teacher in the five-day language cycles. There are approxi-
mately 100 students in Kindergarten through second grade at the time of this
writing, with a grade scheduled to be added annually until sixth grade. Twen-
ty-five percent of the students are African American and 50% of the Latino
children speak English only. All teachers, whether Latino or Anglo, are bilin-
gual. As in all bilingual schools in an era of higher standards through English,
there are questions and tensions about remaining accountable to the value of bi-
lingualism and biliteracy that underlies their structure and practices.
Big ideas about bilingualism and learning started to emerge from the group
itself immediately. The issue of language in instruction was raised not as a
structural or administrative issue, but as an issue of language springing directly
from the work of the child. And it was addressed by Garcia not as the expert,
but collectively through the teacher's experience and the children's work. In
fact, the process called into question Garcia's expert knowledge of bilin-
gualism, by confronting specific situations that did not respond to theoretical
generalities.
For the third inquiry session, the English and Spanish first / second grade
team, worked with Traugh in preparing a Descriptive Review of a Latino Eng-
lish-speaking first grader who presented challenges. In preparing their review,
they realized, for the first time, that neither of them knew of the child's work in
the other language. The collective discussion that ensued after the review en-
abled the school's faculty to share the different stories of the child they held,
and the multiplicity of views and diverse images they had about him. They
began to see him across languages, across disciplines, across grades, across
home and school cultures, as a being in the process of making.
Emerging from our description of children's work, the question of the ef-
fects the language of the class might have vis-ä-vis the language dominance of
the child and the work she or he produces was raised almost immediately. To
pursue this question, we chose an English Language Learner for the next de-
scriptive inquiry session. The two kindergarten teachers introduced a five-year-
old girl who came to school "with no English at all." The teacher of the English
part of the kindergarten, described her:
X is definitely a talker in Spanish. She follows stories really well and has a lot to say
about them. She is very outgoing (...) She is very interested in learning and seems
very happy with things. Now she seems to be speaking a lot of English. (April 18,
2001: 1)
It is interesting to reflect on all this quote says about bilingualism in instruction

and second language acquisition. Because of the teacher's bilingualism, she
knows that the student is a talker, even when she teaches in English only. It may
be because the teacher is able to understand the child that the child also seems
to understand the teacher's stories when she tells them in English. The teach-
er's bilingualism makes the child interested in learning and happy to be in the
classroom. And in seven months, despite the fact that more than half of the time
is spent in Spanish, the child is speaking English.
The teacher continued to comment on children's language use:
It seems, in general, that the kids who are just speaking English love to speak English
in my class because it is a new thing. The ones who are just learning Spanish, they
can show off with the Spanish they know. It is funny to have the English speakers
speaking to me in Spanish, and the Spanish speakers speaking to me in English.
(April 18, 2001: 1)
This description shows a level of novelty and excitement in using the second
language which traditional research literature has not commented upon. In a
safe setting such as that of this dual language school where teachers and direc-
tors, whether Latinas or Anglos, are bilingual, everyone wants to speak the
other language. Because the teacher is bilingual, kindergartners can "try" the
other language regardless of the language of instruction.
The teacher returned to the child's process in learning to speak and write
English:
Her English is broken. Sometimes she speaks in English; sometimes she speaks in
Spanish. Not in complete sentences. And, she copies. When MsX came in and read,
she copied what MsX said. I think that is the way she is learning language. She is a
mimic. It is the way she is learning. (April 18, 2001: 1)
The teacher described the child's use of Spanish to fill in for the gaps that she
has in English, a common language use in the borderlands, as Zentella (1997)
has pointed out. And in much the same way we have written this paper, as we
copy others' words to make them meaningful for us, the child makes language
meaningful for herself by copying.
After the presentation of the child, we took an initial look at her work from
the English and Spanish class that the two teachers had collected. Our first col-
lective impressions drew us immediately into language. The teacher who
teaches in English is extremely interested in what the child is doing in Spanish.
She questioned:
I wonder if you could say that in English, she is copying more. Like in November she
copied the whole alphabet, and the alphabet was on the table. Later on, there are
whole words that she copies. I don't think I see much invented spelling in English. In
Spanish, it is happening. (April 18, 2001: 2)
The group then decided to select one drawing done in the Spanish class for
close description, given here as Figure 1. Members of the inquiry group first
described it literally, noting the colors, the shapes, the numbers in the middle of
the picture, the writing, how the figure was made, the pencil pressure in differ-
ent places. After the literal description, the group joined in further thought, re-
produced below from the descriptive notes:
- I am fascinated by the first line of letters - "me casa". The first "a" looks
like a "g" but it is supposed to be an "a". The "a" at the end is fine, and the
"a's" she wrote below are fine.
- It is very interesting how she drew her house. I did a home visit, and her
house is normal white, straight rectangular building, with stairs going up.
She drew it more creatively.
- It seems to me like how someone would draw the Empire State Building,
if they were on the sidewalk looking up.
- Does her house have a flag on the top?

- That's where the "casa" and the "esqela" become the same thing. The
school would have the flag.
- The teacher wrote, "^Donde estä tu casa?" And she wrote, "Es cerca de la
esqela".
- The teacher then looked up her address and it was 239, so she reversed the
first two numbers. (April 18, 2001: 5)
This kindergartner displays in her drawing the coming together of her house
and her school. She draws it not only near it, as her words express, but as a
blend of both worlds, as that third space which all children, coming to school
for the first time, create for themselves to inhabit. But educators rarely stretch
their school conception to embrace and contain that of the home and the com-
munity. And we're left wondering what would have to be done to shape schools
as to blend in the world of the home, a world which especially language minor-
ity children must connect to in order to make meaning for themselves. The Cy-
press Hills Dual Language School seems to be an effort in that direction.
The group then turned to a drawing that the child had done around the same
time in the English class, given here as Figure 2. The group noted:
- We noticed the layering and the depth of that layering. There is layering
up and down, for example, the shape of the big figure is layered and inside
there is the layering of the patterns - the zigzag on the bottom, the re-
peated triangles which merge into the circles which merge into the crosses
with the circles. Because of the erasure you have layering that is almost
3-D, the layering of what is kept and what is erased (...) (7).
- The words. All those female words - sister, her, she, daughter. Then, the
"I" and something started and erased. Lines connect sister and her and her
and she. The dot is really worked; it is not just placed. The parallels be-
tween the patterns in the writing and the drawing are very visible (...) It
feels like there is a story here. One is so plain and one is so decorated. One
seems crying and the other is so happy. The words don't give us any hints
about the story. Unlike the first piece where the words are connected to
the drawing (8).
In discussing the drawing, we learned that indeed the words had been copied
from a typical ESL lesson on "her" and "she" as female. But the group was sur-
prised that the child herself added the word I, which for her is female, suggest-
ing that there is serious thinking in the copying and that original expression in
writing in a second language is nicely emerging, "a feminist poem" as one of
the teachers suggested. Another member of the group saw the child as a mathe-
matics thinker, as well as a language learner. She added: "There are math ideas
in this work - patterns, shapes, pairings" (9). And then, we turned to language
use again, as the use of the English word "me" for the Spanish "mi"was noted
in the first drawing.
The Spanish kindergarten teacher explained:
I notice that they do that a lot. "Me" is a sight word. It has been a sight word all year.
All the kids write "mi" as "me." Only now have we started doing "mi" and they see
there is another way to write it. This is one of the things that really interests me with
the dual language (...). (April 18, 2001: 9)
It seems that despite the strict language separation of this dual language school,
for this English language learner, inhabiting a space that merges home and
school, home language and school language, literacy emerges for the first time
as a system with overlaps, connections, coincidences, blendings. The child's
bilingualism and biliteracy, as well as herself, are in the making.
Descriptive inquiry has enabled the entire Cypress Hills school faculty to
continue to find the intellectual and creative energy, passion and space necess-
ary to continue their efforts to develop the children's bilingualism in the face of
mounting attacks. It has also provided a space and time that allows them to con-
sider their teaching practice and school structures in the light of work of indi-
vidual children. The descriptive inquiry process has also enabled the faculty to
build relationship not only with children through their own work, but also with
each other. The process has kept the complexity of teaching and learning, and
especially of developing bilingualism and biliteracy, alive and visible in the
face of standardization and homogenization.
7. Conclusion
This chapter describes the efforts educators in two settings are making to con-
tinue to support linguistically diverse teachers and learners in US schools. Spe-
cifically, this chapter discusses how descriptive inquiry as a research method-
ology has held the possibility of regeneration of one teacher education program
and of deeper understandings of bilingualism and the role it plays in education
in a dual language program. As such, descriptive inquiry as a research mode
can be an important tool to regenerate bilingual teaching and learning, particu-
larly in societal contexts that offer resistance. Descriptive inquiry is also im-
portant to keep difference and heterogeneity generative of educational oppor-
tunity in the face of global socioeducational forces which homogenize.
Appendix
W*C 5 k
ό P o n ^ e es4o' t-
4u COSA > β SC |_3/4
^Qtlo,
@ ©
(8) &>
JL
Figure 1. Figure 2.
Note
1. The following faculty members of the Teaching and Learning Department at LIU
and of the Cypress Hills Dual Language School have been engaged in the Descrip-
tive Inquiry sessions that are described in this chapter. As such, they're collaborators
in the development and writing of this chapter. For LIU/Brooklyn: Gurprit Bains,
Rebecca Dyasi, Linda Jacobs, Carole Kazlow, Valerie Lava, Nancy Lemberger,
Laurie Lehman, Sonia Murrow, Robert Nathanson, Klaudia Rivera, Judith Singer,
Jessica Trubek, and Susan Zinar. For Cypress Hills: Laura Ascenzi-Moreno, Carrie
Barnes, Barbara Rossi, Christian Denese, Shelley Rappaport, Marjorie Suärez,
Irene Leon, Maria Jaya-Vega, Sheryl Brown.
References
AERA/APA/NCME
1999 Standards for Educational and Psychological Testing. Washington,
D.C.: American Psychological Education.
Anzaldua, G.
1987 Borderlands / La Frontera: The new mestiza. San Francisco: Spins-
ters/Aunt Lute.
Arias, M. and U. Casanova (eds.)
1993 Bilingual Education: Politics, Practice, Research. Chicago: National
Society for the Study of Education / University of Chicago Press.
August, D. and K. Hakuta
* 1998 Educating Language Minority Children. Washington, D.C.: National
Academy Press.
Baker, C.
2001 Foundations of Bilingual Education and Bilingualism. Clevedon/Buf-
falo: Multilingual Matters.
Crawford, J.
*2000 At War with Diversity. Clevedon: Multilingual Matters.
Carini, P.
1993 Images and immeasurables II. Prospect Center Occasional Paper.
*2000 Prospect's descriptive processes. In: M. Himley and P. Carini (eds.),
From Another Angle: Children's Strengths and School Standards. The
Prospect Center's Descriptive Review of the Child, 8—20. New York:
Teachers College Press.
2001 Starting Strong. A Different Look at Children, Schools, and Standards.
New York: Teachers College Press.
Cochran Smith, Marilyn and Susan Lytle
1998 Relationships of knowledge and practice: teacher learning in commu-
nities. Review of Research in Education 24: 249—305.
Garcia, O.
1991 Bilingual Education. Amsterdam/Philadelphia: John Benjamins Pub-
lishing.
* 1997 New York's multilingualism: World languages and their role in a US
city. In: O. Garcia and J. Fishman (eds.), The Multilingual Apple. Lan-
guages in New York City, 3-52. Berlin/New York: Mouton de Gruyter.
Garcia, O. and C. Baker
1995 Policy and Practice in Bilingual Education: A Reader. Clevedon:
Garcia, O. and J. A. Fishman
1997 The Multilingual Apple. Languages in New York City. Berlin/New
York: Mouton de Gruyter.
Using descriptive inquiry to transform US teacher and, student education 327
Garcia, O. and J. Trubek

* 1999 Where have all the urban minority educators gone and when will they
ever learn? Educators for Urban Minorities 1: 1-8.
Greene, M.
1995 Releasing the Imagination. San Francisco: Jossey-Bass Publishers.
Himley, M. and P. Carini (eds.)
2000 From Another Angle: Children's Strengths and School Standards. The
Prospect Center's Descriptive Review of the Child. New York: Teach-
ers College Press.
Kaplan, Fred
2001 Tales from a multilingual melting pot. A babel of media finds many
niches in New York. Boston Globe, July 25, 2001.
Lambert, Bruce
2000 40 percent in New York born abroad. New York Times, July 24, 2000.
National Research Council (NRC)
1999 High Stakes: Testing for tracking, promotion, and graduation. Wash-
ington, D.C.: National Academy Press.
Navarro, Mireya
2001 Puerto Rican presence wanes in New York. New York Times, Febru-
ary 28, 2001.
NCBE Newsline Bulletin
2001 Census survey suggests greater linguistic diversity and English profi-
ciency. August 21, 2001. (http://www.ncbe.gwu.edu/newsline/2001/
0821.htm).
New York State Board of Regents and the New York State Education Department
1998 New York's commitment. Teaching to Higher Standards. Albany, NY:
New York State Education Department.
Ovando, C. and V. Collier
* 1998 Bilingual and ESL Classrooms. Teaching in multicultural contexts.
Boston: McGraw Hill.
Prospect Archive and Center for Education and Research
1986 The Prospect Center Documentary Processes. N. Bennington, Ver-
mont: Prospect Center Archive.
Traugh, C.
2000 Descriptive notes of inquiry sessions with TAL faculty. Unpublished
manuscript.
* 2000 Whole-school inquiry: values and practice. In: M. Himley and P. Ca-
rini (eds.), From Another Angle: Children's Strengths and School
Standards. The Prospect Center's Descriptive Review of the Child,

182-198. New York: Teachers College Press.
Zentella, A. C.
* 1997 Growing Up Bilingual. Puerto Rican Children in New York. Maiden,
MA/Oxford: Blackwell Publishers.
1. What is the contradiction between the sociolinguistic and socioeducational

reality of New York City?
2. What is descriptive inquiry and why has it been effective in the trans-
formation and support of teachers and students in the two programs de-
scribed in this chapter?
3. What is the difference between traditional research on bilingual education
and the work done by Garcia and Traugh in the Cypress Hills Dual Lan-
guage School? Discuss advantages and any limitations you might see.
4. What are the implications of an inclusive teacher education curricula? What
are the societal and pedagogical advantages and disadvantages?
5. How is the model of bilingual education used in the Cypress Hills Dual
Language school different from more traditional models? What do you
think are the advantages and disadvantages?
Coda
Changing paradigms in the study of bilingualism
William F. Mackey
If one had sought the word "bilingualism" in any glossary of linguistic terms a
half century ago, one would have found definitions like "the equal mastery,
choice and use of two languages". In the mid-forties, for example, a director of
the Harvard University department of comparative philology had assigned re-
search on bilingualism to doctoral students, resulting in a long and futile search
which was later abandoned for lack of "bilinguals".
1. Bilingualism as a problem
It had become evident that each discipline in which the "problem" of bilin-
gualism could be relevant treated it as a peripheral phenomenon. Articles could
be found in the journals of abnormal psychology where the greatest concern
was to find out how "bilingualism" affected the testing of intelligence. In the
journals of education, the concern was with the correlation of bilingualism
with backwardness in the basic subjects. Research results based on standard
unilingual tests were unclear or contradictory. This was not surprising since all
articles on bilingualism were not dealing with the same phenomenon. In jour-
nals of comparative philology, the existence of bilingualism was evoked to ex-
plain phenomena which could not otherwise be accounted for by the phonetic
or morphological laws of language change.
It was the accepted view that language, as a basis for social communication,
depends on the existence of linguistic signs and that these in turn depend on a
system of mental representation. Such a system, best manifested in its written
forms, must be the basis of all language study.
Predominantly, the written language embodied the repository of national
culture, the history, laws, religion and lore of the nation, the only texts (new or
ancient) worthy of study, as literature or philology. This was the received wis-
dom current in the forties. There were no departments of linguistics.
330 William F. Mackey
2. Bilingualism and linguistics
Although European philologists had already been working for more than a half
century on a comprehensive science of language, in America, it was the
achievement of cultural anthropologists in the description of unwritten lan-
guages that promoted linguistics as an academic discipline. Field workers de-
veloped a rigorous analytic method for describing oral languages coupled with
a linguistic theory (structuralism) which dominated most of the fledgling de-
partments of linguistics supplanting the courses in comparative philology. In
this context "bilingualism" had more to do with speaking two languages. In the
early sixties, however, the analytic and data-based methods of structuralism
began ceding ground to the speculative paradigms and agendas of revolution-
ary, young "generative" grammarians who were redefining the study of lan-
guage at the same time as departments of linguistics were proliferating in
American universities. In their search for paradigms based on the "ideal speak-
er", the study of bilingualism did not rank high on their agenda.
The paradigms of structuralism having evolved from the phoneme to the
sentence and from the sentence to the conversation, were now going to proceed
beyond, to the analysis of communication as such. The more analytically
oriented linguists turned to the study of the minutiae of social interaction, so
detailed that no generalizations were possible. Psychologically-oriented lin-
guists and social psychologists added to a growing collection of studies labeled
"sociolinguistics" with measures of language-related attitudes and values. In
this way, a repertoire of "bilingualism" could be put together from products ap-
pearing on the fringes of such disciplines as social psychology, social dialec-
tology and social anthropology.
3. Bilingualism and society
Meanwhile the world in which many of these paradigms of the study of lan-
guage and society had been based was rapidly evolving. In every quarter of the
globe, primitive societies were becoming features of the past. Rural, stable and
simple family-oriented societies had been rapidly evolving into urban, mobile
and complex social structures - even in those parts of the world where one used
to go to study the workings of primitive societies.
One result of this evolution from the isolated traditional society to the mo-
bile, urban, industrial megapolis was that people were now becoming polyso-
cial, belong to two or more groups for work, trade, profession, recreation, cul-
Changing paradigms in the study of bilingualism 331
ture, schooling or religion. To the extent that any of these groups functioned in
a different language or language variety, people by becoming bi-social had to
become bilingual and diglossic in language use.
The phenomenal increase in the urbanization, mobility and possibilities of
interpersonal communication had transformed some societies into groupings
of multisocial individuals. In the process many had, of necessity, become
multilingual. In some parts of the world, the very act of attending school had
supposed the acquisition and use of a new and often unrelated language; to be-
come educated meant to become bilingual.
Since the possibility of linguistic intercommunication is the primary pre-
requisite for the existence of any social group, any increase in the availability
of communication facilities would be followed by an increase in the number
and types of social groups, including bilingual ones. This vast and growing
population making daily use of two or more languages did not obviously speak
and write them equally well and equally often.
4. The redefinition of bilingualism
For all these reasons, it eventually became evident that in order to become the
object of research the phenomenon of bilingualism would have to be redefined.
Discussing the impasse with people engaged at the time in the study of minor-
ity languages, child language and language contact, notably Einar Haugen,
Werner Leopold and Uriel Weinreich, one gradually obtained enough assur-
ance to suggest that the whole concept of bilingualism be redefined (Mackey
1956). In doing so, one was simply stating what people engaged in some aspect
of language contact had simply decided to take for granted. For example, Fish-
man's basic and often quoted question "who speaks which language to whom"
engendered studies of bilingualism and diglossia of different types and degrees
in a number of different countries.
Bilingualism gradually ceased to be regarded as a peculiar either/or trait,
but rather as a phenomenon that was widespread, relative, multidimensional
and multidisciplinary. It embraced many language-related questions already
found in the vast literatures of the humanities and the social sciences. As rel-
evant studies multiplied into the thousands (Mackey 1972, 1982), the concept
had become so broad as to include the use of two varieties or registers of the
same language (Pap 1982).
By the 1980s, having been appropriated by the young sciences of sociolin-
guistics, psycholinguistics and applied linguistics, the field had incorporated so
much material that it began to generate a life of its own - in the process, en-
gulfing such sub-disciplines as ethnolinguistics, demolinguistics, glotto-
politics, language planning and others. During this period it evolved from
being programmatic to data-oriented, from sociographic to quantitative and
theoretical. It became much more comprehensive than "linguistics" had been.
At the same time, formal linguistics tended to become more pragmatic and in-
clusive, tending toward a general theory of language by taking into account its
uses and the role of context. Since context depends on social relation, the line
of demarcation between what constitutes bilingualism and what does not had
become even fuzzier at the end of the decade.
5. Discipline-oriented paradigms
The study of bilingualism was still beholden to the changing concepts and ter-
minologies of the different disciplines and to the use of captive audiences in the
field of language teaching (Mackey 1965-1990). The problems of bilingualism
attracted the attention of the developing interdisciplines of sociolinguistics and
psycholinguistics. Sociolinguistics as a science had been based neither on lin-
guistic theories of society nor on social theories of language. Its underlying as-
sumption was that there exists a relationship between the groups to which
people belong and the way they express themselves. The groups to which
people belong, by choice or by necessity, vary in size and nature and range
from the close-knit family to work-groups, voluntary organizations, profes-
sional societies, ethnic groups, nation-states, and multinational societies. The
determinants of these groups, may be genetic, regional, religious, economic,
cultural or some combination of these and other social characteristics. The life
of such groups and their very social existence depends on the possibility of in-
terpersonal communication, and this, to such an extent that the modes and
means of such intercommunication themselves become characteristic of the
group. Such characteristics of expression, as language, dialect, accent or lexi-
con, may sometimes be considered as the sole determinants. It is as if the con-
sensus of the group were based on such reasoning as: "Here is someone who
speaks like us. Therefore, this person must be one of us".
Yet within the frame of reference covered by the study of language in so-
ciety, it was difficult to arrive at a generalized conceptual framework for the
study of bilingualism when the frame itself was continually changing, along
with the paradigms that fit into it. A science that could tolerate two dozen defi-
nitions of bilingualism (van Overbeke 1972), two dozen definitions of 'ethnic-
ity' (Isajiw 1974), three dozen definitions of 'culture' (Mackey 1988), and a
score of definitions of 'primitive' (Akhmanova 1985) was still a long way from
rating as a unified field of knowledge (Sartori etal. 1975).
6. The problem with academic disciplines
If sociolinguistics did not fit into a unified theory, what about each of its com-
ponents - linguistics and social psychology? During the past half-century, lin-
guistics had evolved, or rather devolved, from a unified theory of comparative
philology with its various laws, into a diversified collection of theories, each
with its own focus. Most of these had been inward-looking, self-centered the-
ories of language, concerned with sound-symbol, object-thing or person-thing
relations, while ignoring both the dynamics of human interaction and most of
the important functions of language, which are social ones.
Each of these theories had defined what "linguistics really means", accord-
ing to the focus it elected as being fundamental. Since terms are theory-bound
and there are so many theories, it is not surprising that the terminology of lin-
guistics had proliferated.
Problems first arose when the everyday lexicon of one language provided for
linguistic distinctions, while that of another language did not. French, for
example, has two words for the English word 'language', each of these French
words (langue and langage) has a different meaning in any linguistic theory
written originally in French. This has been the means for elaborating an import-
ant differentiation between the functions of language (Fr. langage) and the func-
tions of a particular language (Fr. langue) in the same society. (Compare: "school
language": la langue scolaire and "child language": le langage enfantin.)
One of the first to describe the latter function (in the late thirties) was the
British ethnographer, Branislow Malinowski. His notions surfaced again (al-
beit without acknowledgement) in the mid-seventies in the context of theoreti-
cal psychiatry where language in society was described as a situation of per-
manent "tetraglossia" separating the functions of culture, communication,
communion and myth (Gobard 1976). For one of these functions of language
(Fr. langage) in society, Malinowski had coined the term 'phatic communion'.
For the rest of the functions, we still lack good terms. Yet, they could become
especially important in the description of multilingual settings, since each
function could be appropriated by a different language.
Then came differences in what each theory accepted as "language". Since
most theories of formal linguistics seemed standard and static, it was possible
to group their terminologies into a single metalanguage (language about lan-

guage). The distinction "language-discourse", "code-message" and "compet-
ence-performance", for example, could be considered as belonging to three dif-
ferent meta-dialects (or metalects) of the same metalanguage of linguistics
(Brann 1990). Yet, even though it could be argued that these three sets of terms
conveying the same dichotomy, users of any of these were reluctant to admit
that the distinctions were equivalent. Since these did not fit their paradigms,
they would claim that language is more than a code or that knowledge of the
code is not the same thing as competence, or that message is not discourse, and
so on. All this within a single discipline.
Between disciplines, paradigms were even more confused. The term "code"
for example was used in formal linguistics and also in sociolinguistics; but not
with the same denotation. In formal linguistics, it denoted the system of lan-
guage symbols used to communicate something (code vs. message). In a so-
ciolinguistic metalect it denoted verbal behaviour which was situation appro-
priate, a verbal repertoire composed of ranges of meaning and style; in sum, a
principle of semantic organization which generates options in each social situ-
ation likely to occur in the speakers experience. This code, pre-determined by
the culture, provides the configurational roles which determine social identities
within a given environment (Bernstein 1971). Within this context, the term
code was seldom given the denotation of code as used in formal linguistics.
Yet, both denotations had been used indifferently in studies on bilingualism.
7. Confusion of terms, symbols and concepts
Later it became fashionable to avoid such distinctions by resorting to symbols.

While it might be salutary to replace a technical term or a vague, polysemic
emotion-loaded everyday word (like "mother-tongue"), by a symbol, the one-
for-one replacement did not of itself improve the clarity of the concept (Skut-
nabb-Kangas and Phillipson 1989). By way of example, take the much used
L, / L2 dichotomy. By using these symbols instead of terms like "native-lan-
guage", "first language", "second-language" or the like, one made it appear as
a difference of the same type that holds between integers (Mackey 1983).
While the first appears before the second, this dichotomy did not fit the com-
plex reality of language acquisition, language learning and language use
among bilinguals. The L, had been used to mean such different things as "lan-
guage first learned" (origin), "language most used" (function), "language best
known" (competence), and the "language of belonging" (identity). Even when
we restricted the symbol to one of these meanings, it could still say nothing
about the several types, dimensions and variables of language origin, function,
competence or identity that one found within a plurilingual population.
Another stumbling block was the concept of the "speech-community". It
was still defined as "a group of people who use the same system of speech sig-
nals". This had been interpreted to include anything from an Hispanic barrio to
the entire Spanish-speaking world, from a Francophone minority to the entire
Francophonie. Since linguists had been incapable of establishing an objective
language boundary between one dialect and the next, or between one language
and another, what in fact constituted a speech community was often quite ar-
bitrary.
Similar inconsistencies had appeared within sociology and its paradigms
for the study of society. We had social theories which considered society as an
all encompassing system. All of human life must be social, since it is based on
learning and not on instinct. The basic biological needs for food, shelter, pro-
creation and protection have engendered basic social functions, social roles
(based on age, sex and kinship), values (belief systems), affective and cognitive
learning behaviour, authority and motivation (with rewards and penalties) and
ways of distribution of goods and services. This is what might be called the
container model of society. For it was based on the existence of communities
(communities based on language and language based on communities), of ter-
ritoriality, and of face-to-face interaction.
In contradistinction to these conventional static models of society, more dy-
namic construction-production models appeared whereby each individual
holds a role and certain functions that can be described in only one way. It was
a process-oriented model: one thing at a time. The model was clear and con-
sistent, but probably limited to military societies, bureaucracies and the like. It
hardly fitted the all-encompassing uses of language in a multilingual society,
where language is not production-oriented but outcome-oriented. Its objective
can be attained in different ways. Many things happen at the same time; many
ways lead to the same objectives. The use of different languages reaches the
same objective in different ways.
8. Defining bilingual identity
There was a failure to distinguish between the language determinants of social

grouping, and the social determinants of language behaviour. Each grouping
may vary along the dimensions of space (territoriality) and time (stability), the
variations being affected by such social forces as mobility, exogamy and accul-
turation, the latter ranging from incipient to residual bilingualism. The nature
of such groupings, however, depends on the structure of the society, which may
vary in complexity from the nomadic tribe to the industrial state. In primitive
societies individuals could belong to a single and only language group for all
their needs. At the other extreme, in a highly complex urban society, persons
could belong to a different language group for each social need.
Not only did people belong to a number of different groups for different
purposes but they belonged to each with a different degree of intensity, both as
regards commitment and investment. Identity became dependent on the struc-
ture of the community, its stratification, degrees of differentiation and group
solidarity. And language became an indicator of social status. If the identity
was ethnic, status was also achieved through various types of language laws.
Belonging to any language group however, could simply be a matter of de-
gree, - since language minorities, far from being isolated entities, were becom-
ing part of a wider social system. This was especially true in industrial and
post-industrial societies, where the necessity of interaction with the majority
presupposed the use of another language.
One assumption of sociolinguistics had been that such bilingual groups are
identifiable. This itself was debated. The groups depended on the structure of
the society to such an extent that it became difficult to generalize from one so-
ciety to the next. After all, society itself is only an abstract category. People do
not depend on this abstraction. They depend on other people and this depend-
ence may affect their language behaviour. It is therefore such social variables in
behaviour that had to be observed and explained. Such a grouping of inter-de-
pendent people has been conceived as a network whose workings and inter-
communication may be analysed by appropriate methods (Gulliver 1971; Li
Wei 1996).
Conventions for the workings of plurilingual groups are not always the
same as those governing mono-ethnic societies. In linguistically monolithic so-
cieties like those created by long-stabilized nation-states any deviation from
the norm is either not understood or not tolerated. Strangers are expected to
conform. In such societies, language standardization constitutes a form of be-
haviour control in the use of the written word, where spelling mistakes may be
universally condemned. Contrariwise in multilingual societies there may be a
rapidly evolving norm, permitting the individual to adjust to each situation to
achieve certain objectives, speaking one way to give one impression (like want-
ing to conform to outside job requirements) and another way to give the im-
pression of solidarity with the people of the area (LePage 1972, 1974). Thus,
pluralistic societies could survive through tolerance of free variation. In these
societies interintelligibility depends on a wide range of differences in compet-

ence between each individual's idiolect and the capacity for understanding
those of others. A high degree of non-reciprocal multidialectalism is the norm.
Such observations seemed to confirm a theory of language accommodation
based on similarity-attraction, whereby bilinguals can induce others to evaluate
them better by reducing dissimilarities (Giles and St.Claire 1979). In interlin-
gual communication, this reduction means an expenditure on the part of one or
of both parties for which a return is expected. The relationship has been ex-
pressed in terms of exchange theory. The extent of accommodation required to
make one's speech acceptable in any given type of situation also depends on
variables of person, place and type, related to age, sex, education and ethnicity
(Greenbaum 1977). Accommodation in interlingual interaction is also go-
verned by a sort of economy of communication (Mackey 2000).
In the past decades, many have also theorized on the bilingual determinants
of ethnolinguistic identification. Three determinants were put forward: pater-
nity, patrimony and potential. In other words, both ethnic origin and cultural
maintenance can determine ethnic identification, which becomes functional
only if the potential can be manifested through existing supportive institutions.
In addition to trying to isolate the determinants of ethnic belonging, social psy-
chologists examined the process of language identification through the use of
variants of regression analysis (path analysis), starting with the ultimate de-
pendent variable and working back to possible predictors. Secondly, sociolin-
guistic theorists studied the social determinant of bilingual language behaviour
in the context of communications theory. Much attention was paid to the ability
to make oneself intelligible and to the notion of communicative competence.
Such competence implies that speakers can function in the language; it does
not however assume grammatical or phonetic correctness, but rather the ability
to initiate conversations, to argue, to convince, to refuse politely and the like,
according to situation, topic and participant. Social behaviour in language use
was expressed in terms of norms and rules. Such paradigms were called deter-
ministic - as if an abstract construct (society) made "rules" to which all had to
conform.
After observing language behaviour in plurilingual societies, one also cast
doubt on the ethnic or language-centered paradigm assumed in most policy-
oriented studies. For in situations of contact between two language groups it
was necessary to take the existence of a third group into account, - namely the
bilinguals (Miemois 1979). It seemed that some of the problems in the imple-
mentation of language policy therefore stem from the presence of ethnic bilin-
gualism maintained through a functional distribution of language uses in the
community- but seldom taken into account.
9. The components of conflicting paradigms
From all these investigations, what can one gather about the changing para-
digms in the study of bilingualism. What do they have in common? Most dis-
ciplines that have studied bilingualism have analyzed the phenomenon by cat-
egory, dichotomy or scale.
By category, bilingualism has been studied according to proficiency and ac-
cording to function. From the point of view of proficiency, such categories as
"complete bilingualism", "perfect bilingualism", "partial bilingualism", "in-
cipient bilingualism", and "passive bilingualism" have become current. From
the point of view of function, we have heard of "home bilingualism", "school
bilingualism", "street bilingualism", and similar terms, denoting the use to
which bilingualism is put. The disadvantage of such categories in the study of
bilingualism is that they are often impossible to delimit.
Bilingualism has also been studied by the use of dichotomies. For example,
there has been coordinate versus compound bilingualism, individual versus
national bilingualism, stable versus unstable bilingualism, balanced versus un-
balanced bilingualism, pure versus mixed bilingualism, simultaneous versus
sequential bilingualism, comprehensive versus limited bilingualism, organized
versus incidental bilingualism, general versus specific bilingualism, regressive
versus progressive bilingualism, personal versus institutional bilingualism, and
so on. Some of these dichotomies became standard models for a generation of
researchers. Presumptions like, for example, the compound / coordinate dis-
tinction or language lateralization in bilinguals turned out to be only concep-
tual artifacts, later discredited by research in neurolinguistics (Paradis 2000).
The difficulty in cutting bilingualism into dichotomies has also been that the
slices are rarely mutually exclusive, and individual cases are rarely an either/or
proposition. If they are made up of variables, we have to ask ourselves whether
some of these could not be converted into scales and whether a number of such
scales could not interlock to provide profiles.
Bilingualism has indeed been pictured in scales. They have appeared as
dominance configurations, profiles of bilingual background, bilingual sem-
antic differentials and the like. The difficulty here was that most scales presup-
posed standard units of measure which did not exist, while procedures for their
delimitation were not indicated. They affect the function, stability, and dis-
tribution of the languages involved, in relation to their location, origin, and
dominance.
Most measures presuppose the creation of units. In the measurement of bi-
lingualism, this is rendered difficult by the fact that these units are not self-evi-
dent. They were, therefore, often simply measures of indices which were as-
sumed to reflect certain variables of bilingualism, such as dominance, skill, and

regression. Before being used as basis for units of measurement, such indices
required validation. The problem was threefold. First, there was the validation
of the most suitable indices. To what extent, for example, was word-association
skill an index of individual language dominance? Secondly, the obtaining of
valid samples had often been difficult. For example, questionnaires were used
to measure language proficiency. Recorded samples were obtained for the
measurement of language function. Thirdly, the elaboration of units of
measurement posed a number of sampling problems. Type-token ratios were
used as measures of interference. Percentages of loan-words were used as
measures of language dominance.
The first step in the examination of the developmental paradigms of bilin-
gualism had to do with the sampling and measures used in case studies. In this
type of inquiry, researchers had used written questionnaires, oral recording,
and descriptive introspection. Most studies, however, had taken into account
only the bilingual's output. Few of them, if any, had also recorded the relevant
input; that is, what the developing bilingual hears and reads. The relation be-
tween input and output had therefore not been established. Secondly, one had
to check the indices, sampling, and measures of bilingual proficiency, capacity
and performance. Most measures of this psycholinguistic dimension of bilin-
gualism had been based on tests ranging from conventional, skill-based lan-
guage tests through time-reaction tests on rapid word translation; oral reading
of homographs and mixed word lists; word detection and word completion
tests; reading of colour names written in colours other than the name indicated;
and a number of tests invented for specific experiments.
What needed to be studied was the extent to which such tests were indices
of bilingual proficiency or capacity. Did a standard word-translation test, for
example, measure only language proficiency, only translation skill, only
switching facility, or all three skills - and to what extent?
Then there were the purely linguistic paradigms, including the measure-
ment of interference and inter-language distance. One of the main problems
here was that of distinguishing between code and message, that of identifying
each item in the bilingual's chain of speech and allocating it to one language or
the other in the idiolect of the individual. The fact that the bilingual used an
item from the other language was not necessarily an indication of interference
if this item had already become part of that individual's or the family's lan-
guage norm. Among measures used, there were word associations and type-
token ratios in running texts.
After studying the measurement of how well the bilinguals knew their lan-
guages, it was necessary to study what they did with them. This sociolinguistic
dimension had to do with language function. Various models and configur-

ations of dominance, bilingual background profiles, and other measures had
been used, most of the data coming from questionnaires. One of the main prob-
lems was making sure that the measures gave an accurate picture of the dis-
tribution of both languages throughout the entire behaviour of the bilingual.
From the individual, one passed to the sociocultural aspects, and considered
the bilingual group as a group. How could one measure the behaviour of bilin-
gual groups? The first problem here was the isolation of behavioral features of
the bilingual group which were distinct from those of each unilingual group
with which there was contact. This behaviour may be both linguistic and nonl-
inguistic. It may reflect group attitudes and prejudices, group values and cus-
toms. The description of these ranged from the anecdotal to the statistical.
Measurements included direct tests of group attitudes and indirect tests based
on questionnaires and pre-recorded voice guises.
Finally, there were the demographic aspects: the measurement of bilingual
populations. In many countries, this was a practical problem of great import-
ance. Some countries kept language statistics, but these either said very little or
were largely meaningless to bilinguals. The problem was that of collecting
large masses of data with simple questionnaires, and second, that of finding out
how to classify people as bilinguals from the evidence of their answers to a
question which was meant for monolinguals and, for practical reasons, had to
be simple.
In the study of these paradigms, we find that we are faced with different
types of data normally dealt with by such diverse disciplines as psychology,
linguistics, sociology, anthropology, and perhaps politics, law, and govern-
ment. Each discipline has been in the habit of categorizing and representing
these data according to its own paradigms and dealing with the results in its
own way to arrive at its own theory of bilingualism.
10. Toward a framework for the future
To succeed in effectively measuring bilingualism, we have to elaborate com-

mon categories and common or related types of representation of the same phe-
nomena to describe such inherent differences as may exist between the bilin-
gual and the unilingual, the differences among bilinguals themselves, the inter-
personal and intergroup relations among and between bilinguals and unilin-
guals. A great help would be a general and unified theory of bilingualism with
paradigms of problem-oriented research independent of any discipline. This
would require a standard terminology. But a standard terminology has to be
structured according to a system of classification. The system links the rela-

tionship between concepts. Although there is not any single, unique method of
classifying these links, a distinction must be maintained between concepts of
equal significance to the system and concepts linked to other concepts in a re-
lation of dependency. Links of dependency are structured as hierarchical,
causal, and functional. One could distinguish, for example, relations which are
logical, ontological and sequential (Felber 1981). Yet, if one has to overcome
entrenched usage, intractable dogma and magic symbols, it becomes extremely
difficult to implant even the most reasonable of terminological changes.
Referring back to the link between theory and terminology, we may con-
clude that, while some linguistic theories now account for degrees of bilingual
language variation, these cannot be adequately correlated to societal variables,
since the social theories have yet to accommodate the many degrees of belong-
ing to what rates as "society". Nor have they accounted for the fact that some of
these societies are more societal than others (Mackey 1979). Since a social the-
ory is only an explanation of how society functions, it is valid only so long as
the society functions in the same way the theory supposes it does. In the mid-
sixties, when Intelstat III went into orbit over the Indian Ocean, it became poss-
ible, for the first time in human history, to have a network of communications
whereby every spot on earth could henceforth be in immediate touch with each
and all of the others.
Since that time the multiplication of wide-ranging satellite networks has
created extraterritorial societies relying on the transmission of speech and im-
ages regardless of distance. Even though many of these seemingly haphazard
images have been pre-selected to attract, the authority of the image has created
pseudo-communities which can now become dominant. Masses of people now
share the most private experiences, witnessed as if they had actually been privy
to them, regardless of where or how far away. Since there are more distant
events than local ones, the far drives out the near in the mind of the individual.
This vicarious and expanded context of society has changed the grammar of
life. Indeed, everything in society has changed except our concept of what it is.
So that some of the old paradigms may no longer apply; we now need new
ones.
Almost everywhere, we can already observe new social groupings and re-
groupings claiming the right to be different, the right to autonomy, often based
on a common language (Mackey 1975). Since most existing language laws lack
sociolinguistic underpinning, lacking even the concepts and corresponding
terms, they have often failed to solve problems of language rights, irredentism
and language conflict, since they do not adequately account for such questions
as: whose language, does it exist, what is it, who speaks it and who wants it?
New paradigms for new concepts could lead the way to new thinking about lan-
guage in our increasingly diverse and complex societies.
References
Akhmanova, Olga S.
1985 Terminology and terminologies. Terminologie und benachbarte Ge-
biete 1965-1985, 86-91. Vienna: Hermann Böhlau.
Bernstein, Basil
1971 Class, Codes and Control. Volume 1: Theoretical Studies: Towards a
Sociology of Language. London: Routledge and Kegan Paul.
Brann, Conrad Max Benedict
1990 Pour une metalangue du multilinguisme [For a metalanguage of multi-
lingualism]. La Linguistique 26 (1): 43-51.
Edwards, John R.
* 1994 Multilingualism. London: Routledge.
Felber, Helmut
1981 The Vienna School of Terminology: fundamentals and theory. Info-
term Series 6: 69-86.
Fernandez, Mauro
* 1993 Diglossia: A Comprehensive Bibliography. Amsterdam: John Benja-
mins.
Giles, H. and R. St. Claire (eds.)
1979 Introducing Language and Social Psychology. Oxford: Blackwell.
Gobard, Henri
1976 L'alienation linguistique: analyse tetraglossique [Linguistic alien-
ation: tetraglossic analysis]. Paris: Flammarion.
Goebl, Hans, Peter H. Neide et al. (eds.)
* 1996 Contact Linguistics: An International Handbook of Contemporary
Research. Berlin/New York: Walter de Gruyter.
Gulliver, P. H.
1971 Neighbors and Networks. Berkeley: University of California Press.
Haugen, Einar
1972 The Ecology ofLanguage. Edited by Anwar S. Dil. Stanford: Stanford
University Press.
Isajiw, Wserolod
1974 Definitions of Ethnicity. Ethnicity 1 (2): 111-124.
Kelly, Louis G. (ed.)

* 1969 The Description and Measurement of Bilingualism. Toronto: Univer-
sity of Toronto Press.
Le Page, R. B. et al.
1972 Preliminary report on the sociolinguistic survey of multilingual com-
munities. Part I. Language in Society 1: 155-72.
1974 Preliminary report on the sociolinguistic survey of multilingual com-
munities. Part II. Language in Society 3: 2-32.
Li Wei
* 1996 Network analysis. In: Hans Goebl, Peter H. Neide et al. (eds.), Con-
tact Linguistics: An International Handbook of Contemporary Re-
search, 805-811. Berlin/New York: Walter de Gruyter.
Mackey, William F.
1956 Toward a redefinition of bilingualism. Journal of the Canadian Lin-
guistics Association 2 (1): 4-11.
1965-1990 Language Teaching Analysis. London: Longman (with updated trans-
lations into French, Russian, Japanese, and Chinese).
* [1972] 1982 International Bibliography on Bilingualism / Bibliographie inter-
nationale sur le bilinguisme. Quebec: Presses de l'Universite Laval.
1975 Trends in multinationalism. In: W. F. Mackey and Albert Verdoodt
(eds.), The Multinational Society, 330-353. Rowley: Newbury House.
* 1976 Bilinguisme et Contact de langues [Bilingualism and language con-
tact]. Paris: Klincksieck.
1979 Toward an ecology of language contact. In: W. F. Mackey and J. Orn-
stein (eds.), Sociolinguistic Studies in Language Contact: Methods
and Cases, 543-460. Berlin/New York: Mouton de Gruyter.
1983 Six questions sur la valeur de la dichotomie L, / L2 [Six questions on
the L, / L2 dichotomy]. Rassengna italiana di linguistica applicata 15:
23-58.
1988 Texte, contexte et culture [Text, context and culture]. Traduction et
cultures 1, 1 1 - 2 0 .
2000 Prolegomenes ä l'analyse de la dynamique des langues [Prolegomena
to the analysis of language dynamics], DiversCite Langues 5.
(http://www.teluq.uquebec.ca/diversite)
2001 Bilingualism and Multilingualism. In: Ulrich Ammon, Norbert
Dittmar and Klaus J. Mattheier (eds.), Sociolinguistics: An Inter-
national Handbook of the Science of Language and Society, 699-713.

Berlin/New York: Walter de Gruyter.
Miemois, Karl Johan
1979 Bilingual Self-Identification (Report of the University of Helsinki Re-
search Group in Comparative Sociology 22.) Helsinki: University of
Helsinki.
Pap, Leo
1982 Bilingualism in need of a conceptual overhaul. Language Sciences 4,
71-84.
Paradis, Michel
2000 Language lateralization in bilinguals: enough already! In: Li Wei
(ed.), The Bilingualism Reader, 394-401. London: Routledge.
Paulston, Christina and Richard Tucker (eds.)
1997 The Early Days of Sociolinguistics. Dallas: SIL (International Lin-
guistic Institute).
Sartori, G. F., W. Riggs and H. Teune
1975 Tower of Babel: On the Definition and Analysis of Concepts in the So-
cial Sciences (Occasional Paper 6.) Pittsburgh: International Studies
Association.
Skutnabb-Kangas,Tove and Robert Phillipson
1989 "Mother-Tongue": the theoretical and sociopolitical construction of a
concept. In: Ulrich Ammon (ed.), Status and Function of Languages
and Language Varieties, 450-477. Berlin/New York: Mouton de
Gruyter.
Van Overbeke, Maurice
1972 Introduction au probleme du bilinguisme [Introduction to the bilin-
gualism issue]. Brussels/Paris: Didier.
1. How can the language / discourse distinction explain the basic difference
between integration (bilingual borrowing) and bilingual interference?
2. How can a person's bilinguality vary over a life-span?
3. How many differences can the L, /L 2 dichotomy carry?
4. How has bilingualism been measured?
5. What aspects of bilingualism have been studied within disciplines other
than linguistics? Which disciplines? Do they overlap?
Index
Africa, 27,56 ecology, 10,26-31,57,275

Africaan, 5, 56 education, 45,46,48-52,55, 135,
Arabic, 5, 6, 157 189-190, 231, 276, 281, 282, 283,
Asia, 27, 158 284, 286, 287, 288, 290, 291,
attitudes, 132-134, 140-142 297-309,311-324, 329
Australia, 4, 11, 27, 56, 74-76, 77 Edwards, John, 10,25-44
English, 4 , 6 , 7 , 8 , 1 6 , 1 8 , 2 1 , 2 5 , 4 5 ,
49,56,159,162,163,164,173,174,
Baetens Beardsmore, Hugo, 12, 52, 177, 179, 185,221, 222, 226-242,
57, 171, 237 249, 253, 256, 275-291, 300-307,
Belgium, 1,6,12,106,111,112,116 311-324
bilingualism Europe/European Union, 1,5,12,27,
causes of, 1,6,9 56, 69,73, 105-124, 158
challenges of, 3-8,9-10
opportunities of, 2-3, 9-10
problems of, 1,4, 7, 8, 262, 329 Ferguson, Charles, 18, 164, 282
Fishman, Joshua Α., 10, 15-24,46,
56, 78, 79,125, 171, 178, 214, 221,
Canada, 27,48, 56, 77, 158, 164, 299 222, 231,280, 286,313,331
China, 8,20 French, 3,5, 6,7,11,25, 33, 89-104,
Chinese, 2,5,6,17,45,78,129,162, 252, 304
249, 250, 251, 252, 255, 256, 260,
261,300-307
Clyne, Michael, 76,77,231,232,236,
275-291 Garcia, Ofelia, 11
Coulmas, Florian, 11, 249-271 German, 5, 6, 7, 16, 18, 25, 45, 107,
culture, 3, 5, 9, 16, 17,78, 79, 137, 108, 109, 111, 112, 113, 116, 117,
153, 237, 278, 281, 288, 329, 332, 134, 145,157,159, 162, 164, 252,
333,334 304
Greek, 17,18,76,77
dialects, 145-156, 157-170

diglossia, 23, 164, 278, 281, 291 Haugen, Einar, 29,40, 56, 125, 146,
dirigisme, 89-104 282,331
Dürkheim, Emile, 10,16, 17 Hebrew, 16, 19, 20, 22
Dutch, 5, 6,12, 19, 25, 157,162, 252 Η (High) variety, 23, 281, 291
346 Index
identity, 3,4, 9, 10, 19, 26, 27, 69, Mühlhäusler, Peter, 30,31,32,33,37,
264, 291,334, 335, 336 38, 39,48, 56
ideology, 5,10, 11, 71, 178,183,
199-219, 259,284
Indian, 2, 5 Neide Peter H., 11, 105-124
Irish, 4,8 norm, 16,281,297
Italian, 5, 78, 145, 160
Polish, 33,77,159,162
Japan, 11,249-271
Japanese, 4, 19,45 Russian, 20,45,56,69,129,130,132,
Jernudd, Björn Η., 11, 275, 297-309 133, 135, 141, 143
Schiffman, Harold F., 11,89-104,275

language maintenance, 11, 21, 27,
Singapore, 3,11,275-295
105-124,199-219
Skutnabb-Kangas, Tove, 10,45-67,
language rights/linguistic rights/lin-
69,76, 178
guistic human rights, 1,7, 8,10,11,
Smolicz, J. J., 11,69-85
45-67,52-55, 259
Spanish, 4, 5,6, 7, 19,45, 162, 256,
language shifl/loss, 11, 21, 30,
312-324
105-124,125-144, 160-162, 224,
Spolsky, Bernard, 11, 171-198, 233
226, 277
Latin, 3,18,21,22,145
Li Wei, 1-12, 275-295, 297, 336 Urdu, 4,6
linguistic diversity, 2,3,7,8,9,10,27, USA, 12, 15, 27, 48, 77, 158, 299,
30, 32, 34, 35, 55,257, 280, 311-324 311-324
literacy, 179,231
L (Low) variety, 23,291
value, 4,11,69-85
Mackey, William F., 12, 33, 329-344 Xhosa, 11,221-248

majority, 2, 4, 5, 6, 9, 37,277, 288
minority, 2,5,6,9, 37,47,55,77,78,
255,259, 288,331 Yiddish, 18,19, 190

(9783110852004 - Opportunities and Challenges of Bilingualism) Opportunities and Challenges of Bilingualism

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(9783110852004 - Opportunities and Challenges of Bilingualism) Opportunities and Challenges of Bilingualism

Uploaded by

Copyright:

Available Formats

Opportunities and Challenges of Bilingualism

© Printed on acid-free paper which falls within the guidelines of

Library of Congress — Cataloging-in-Publication Data

Opportunities and challenges of bilingualism / edited by Li Wei,

Die Deutsche Bibliothek — Cataloging-in-Publication Data

Opportunities and challenges of bilingualism / ed. by Li Wei

Introduction: Opportunities and challenges of bilingualism

"Holy languages" in the context of societal bilingualism

II. Bilingualism worldwide

French language policy: centrism, Orwellian dirigisme, or economic

The development of Navajo-English bilingualism

III. Multilingual management and education

Managing multilingualism in Singapore

Changing paradigms in the study of bilingualism

Dr. Alex Housen

Dr. Jiiri Viikberg

1. What are the opportunities?

2. What are the challenges?

languages (e.g. Panjabi, Urdu, Gujarati) are usually described as "minority"

It is difficult to form a complete picture of numbers of academic staff in Chinese lan-

3. Opportunities and challenges for the research community

second language. Some researchers have wondered whether the emphasis on

4. About this volume

1. Times and tides in the study of bilingualism

2. Bilingualism as a sociocultural norm

We owe to the sociolinguistic enterprise the realization that bilingualism is not

ferred to as "holy" we must, strictly speaking, investigate the semantic (rather

4. Types of holy languages1

4.1. When one of the two languages is considered inherently holy

gualization of a sociocultural aggregate, one of whose languages is strongly

5. When both languages are holy

6. When neither language is holy

7. The consequences of language sanctity beliefs

Language sanctity convictions do not operate in a vacuum. Whether or not they

a complementary allocation of functions falls apart (e.g., when both languages

8. The staying power of sanctity

Questions for discussion

1. Why would sociologists or sociolinguists who are followers of Emile Dürk-

From Talavera to Waterloo, Richard Sharpe's tumultous military career has

slim likelihood of success has become a hopeless or unwarranted one: a foray

2. The ecological thrust

world's languages (1992: 7-8). More traditional linguistic documentation is

in human affairs and, in the eighteenth century, Montesquieu built an elaborate

ones, since the former often acknowledged a Darwinian sort of linguistic

further attests to its unexceptional character: bilingualism is not rocket science.

linking it with a stronger cultural pillar (Catholicism). This is an analogy that

We have forbidden the importation of materials whose removal damages the

possibilities for long-term bilingualism are more propitious. Indeed, isolation

Questions for discussion

1. What are the constraints on the feasibility of long-term bilingual accommo-

2. From linguistic genocide to linguistic human rights

What is linguistic genocide? I see it as an end point on a continuum where the

2.2. Linguistic genocide - first scenario

likelihood of success in attempting to teach pupils a known local language,

This stands in sharp contrast to the enriching outcomes of additive forms of

2.3. Can present linguistic human rights counteract linguistic

The European Convention on Human Rights and Fundamental Freedoms of

One can ask what constitutes "appropriate measures", or "adequate opportun-

In areas inhabited by persons belonging to national minorities traditionally or in sub-

educational clauses do not oblige states to more than non-discrimination on the

3. Work to counteract minorisation and promote the survival of

4. Implications for language ecology

A comparison of the world's linguistic and biological megadiversity countries