Fishman, Joshua A. (ed.) (2001): Can threatened languages be saved?
Multilingual Matters, xvi + 503 pages. Paperback ISBN 1-85359-492-X,
GBP24.95/US$39.95, hardcover ISBN 1-85359-493-8 (price not listed on
Reviewed by Angela Bartens, University of Helsinki.
In the course of an extremely productive academic career which spans
approximately half a century Joshua A. Fishman has made substantial
contributions and even founded several subfields of sociolinguistics,
e.g. language maintenance and language shift, language and ethnic
identity, language and nationalism, language planning and the sociology
of bilingual education. (cf. Preface, p. xiii). With the publication of
Fishman (1991), the field of Reversing Language Shift (RLS) came into
The volume under review here constitutes an assessment of the first
decade of RLS studies. Firstly, the contributions attempt a critical
evaluation of Fishman's RLS theory as outlined in his 1991 publication.
Second, they examine the effects of a decade of RLS efforts on the
language communities chosen by Fishman 1991 for his case studies.
Language communities dealt with in Fishman 1991 are Navajo (Ch. 2 of the
present volume, pp. 23-43, by T. Lee and D. McLaughlin), New York City
Puerto Rican Spanish (Ch. 3, pp. 44-73, by O. Garc�a, J.L. Mor�n and K.
Rivera), both Secular and Ultra-Orthodox Yiddish in New York City (Ch.
4, pp. 74-100, by Joshua A. Fishman), French in Quebec (Ch. 5, pp.
101-141, by R.Y. Bourhis), Irish (Ch. 8, pp. 195-214, by P.�. Riag�in),
Frisian (Ch. 9, pp. 215-233, by D. Gorter), Basque (Ch. 10, pp. 234-259,
by M.-J. Azurmendi, E. Bachoc and F. Zabaleta), Catalan (Ch. 11, pp.
260-283, by M. Strubell), Modern Hebrew (Ch. 15, pp. 350-363, by B.
Spolsky and E. Shohamy), both Immigrant and Aboriginal Languages in
Australia, as a revision to the treatment in Fishman 1991, the
differences that necessarily exist between the two scenarios are duly
recognized in devoting separate chapters to them (Ch. 16 and 17, pp.
364-390 and 391-422, by M. Clyne and by J. Lo Bianco and M. Rhydwen,
respectively), and Maori (Ch. 18, pp. 423-450). Newcomers to this set of
concrete language communities which through case studies inform RLS
theory (cf. p. 22 and Ch. 19) are Otom� (Ch. 6, pp. 142-165, by Y.
Lastra), Quechua (Ch. 7, pp. 166-194, by N.H. Hornberger and K.A. King),
Oko as an example of a threatened language in Africa (Ch. 12, pp.
284-308, by E. Adegbija), Andamanese (Ch. 13., pp. 309-322, by E.
Annamalai and V. Gnanasundaram), and Ainu (Ch. 14, pp. 323-349, by J.C.
Fishman outlines the theoretical framework in the first and final
chapters of the volume (Ch. 1 and 19, pp. 1-22 and 451-483,
respectively). In addition, the volume contains lists of contents (pp.
v-vi) and of contributors (pp. vii-xi), a Preface (pp. xii-xvi), and an
Index (pp. 484-503). The stages conceived by Fishman 1991 for RLS
efforts and central to the discussion in this volume are reproduced on
In Fishman's own words, RLS "is the linguistic part of the pursuit of
ethnocultural self-regulation ... concerned with the recovery,
recreation and retention of a complete way of life" (p. 452). Just as
RLS constitutes both an addition and an alternative to globalization (p.
459), the plea of most RLS movements is for a compromise of
multilingual and multicultural coexistence, not the total rejection of
the language and culture in power (p. 7) -- the local ethnicity it
purports as a part-identity constitutes a healthy counterbalance to
civil nationalism (p. 460). As a subcategory of language, especially
status planning, "RLS theory seeks to be directive or implicational
vis-�-vis social action, rather than merely descriptive or analytic of
the sociocultural scene" (p. 464). The cornerstone of RLS theory, the
Graded Intergenerational Disruption Scale (GIDS), constitutes the aspect
of RLS theory which has been most criticized, both in the contributions
to this volume and elsewhere, whereas Fishman intended its stages to be
"nothing but a logical set of priorities or targets to guide RLS-efforts
toward a derived goal" (p. 465). Instead of following the sequencing of
stages slavishly, a common misconception of Fishman's GIDS, it is the
linkage of stages, i.e. language functions, which matters. Only the
compartmentalization of languages can assure the maintenance of a
threatened language. Stage 6 on the bottom up scale from 8 to 1 has been
found to be the crucial stage of "the intergenerational and
demographically concentrated home-family-neighborhood-community: the
basis of mother tongue transmission" (e.g. p. 466). "... if this stage
is not satisfied, all else can amount to little more than biding time"
(Fishman 1991:399). If measures targeting other stages of the GIDS are
not linked to stage 6, RLS efforts are doomed to fail in the long run.
This is the case of Irish: thanks to the school system, the geographic
bias encountered in the early 20th century has been leveled out to a
national standard; at the same time, competence drastically declines in
the population according to the number of years they have left the
school system. As a matter of fact, stage 6 is so pivotal on the GIDS
that Fishman recognizes the necessity of differentiating within it just
as stage 4 has been divided into a and b according to whether the public
school system is controlled by members of the minority group or not.
Stages 8 to 5 constitute the "program minimum" of RLS (Fishman 1991:400)
for which speakers of the minority language do not need the cooperation
and approval of those in power while stages 4 to 1 are stipulated to
constitute the "high power" stages (p. 473) less willingly relinquished
by the dominant group. However, the applicability of all stages,
especially 2 and 1, i.e. "local/regional mass media and governmental
services" and "education, work sphere, mass media and governmental
operations at higher and nationwide levels", is questioned e.g. in the
contributions on Maori and on Australian Aboriginal Languages while it
is suggested in the chapter on Australian Immigrant Languages that the
mass media should at least in certain cases occupy a lower level of the
GIDS as their contribution to RLS does not always merit such high
ranking. While physical, demographic, social and/or cultural dislocation
causes most cases of language shift (Fishman 1991:55-65), the
contribution on Andamanese exemplifies the case where a language is
threatened by simple biological extinction. At present, there are 35
Andamanese speakers, three of them infants under the age of five years.
Additions/modifications to RLS theory in general and to the GIDS in
particular proposed or implied by the contributors to this volume
include the variables vitality (Bourhis on Quebec), dialectal
homogeneity of the language (Hornberger and King on Quechua), economic,
social and spatial variables (Riag�in on Irish), qualitative vs.
quantitative recovery (Azurmendi, Bachoc and Zabaleta on Basque),
community activism (Strubell on Catalan), restriction to use as language
of sanctity and/or literacy (Spolsky and Shohamy on Hebrew), new
technologies and media (Clyne on Australian Immigrant Languages) as well
as the incursion of the television into stage 6 (i.e., as a member of
the family; Benton and Benton on Maori)! Measures targeting several
stages of the GIDS at a time are found to be most effective by
Hornberger and King (p. 189). On the other hand, in the specific case of
New York Puerto Ricans, RLS is not an issue of great interest since
there is little attachment to language as a marker of identity. The
history of New York Puerto Ricans as a colonized minority has turned
what Garc�a et al. call "linguistic vaiv�n" into the linguistic
idiosyncrasy that characterizes them.
In spite of the emergence of a general climate more favorable to RLS
efforts it has to be stated in a global verdict that no dramatic
successes were encountered among the communities surveyed both in 1991
and a decade later, not even in the case of three "success stories"
Hebrew, Quebecois French and Catalan. As a matter of fact, the current
passiveness of the Catalan community may contribute to renewed language
shift to Spanish as societal bilingualism necessarily leads to language
shift while only boundary maintenance and compartmentalization
guarantees language maintenance as exemplified by the relative success
of RLS in the case of Ultra-Orthodox Yiddish. However, as an answer to
the question formulated in the title of the book reviewed Fishman
rightfully concludes "Yes, more [languages] can be saved than has been
the case in the past, but only by following careful strategies that
focus on priorities and on strong linkages to them, and only if the true
complexity of local human identity, linguistic competence and global
interdependence are fully recognized." (p. 481).
This review has focused more on the overall issues addressed in this
volume than with the contents of individual chapters. Suffice it to say
that those are written by the authorities on RLS in their respective
communities and/or academic subdisciplines and that the methodological
and scientific standard of virtually all papers is outstanding. By
consequence, the volume under review constitutes an important
contribution to the young and still forming theory of RLS as well as to
the study of the language communities in question.
Fishman, Joshua A. (1991): Reversing Language Shift. Clevendon:
Dr.phil. Angela Bartens is Docent of Iberoromance Philology
at the University of Helsinki. Her research interests include
language contact, pidgins and creoles, sociolinguistics and applied
sociolinguistics including language planning. She is currently working
on a project financed by the Finnish Academy "A Contrastive Grammar
Islander (San Andr�s and Old Providence Creole English) Caribbean