Towards a Theory of Language Policy1

Bernard Spolsky

Bar Ilan University

This paper, developed as a result of that given at the 2006 Nessa Wolfson
Memorial colloquium, presents the beginning of a theory of language
policy and management. Essential features are the division into domains
(standing for the speech communities to which the policy is relevant);
recognition of language policy as involving practices, beliefs and man-
agement; and a consideration of internal and external influence on
policy in the domain. The paper looks briefly at some domains and con-
cludes with an analysis of school and the complexity of understanding
language education policy.

A
Introduction
nyone using language is regularly faced with choices. A bilingual
must choose which language to use. Many speakers have a choice
of dialects. At a finer level of analysis, a speaker or writer is reg-
ularly faced with a choice of features – sounds or spellings, lexical items,
grammatical patterns – which are significant markers of languages,
dialects, styles, or other varieties of language and which bundled togeth-
er define varieties of language.2 The goal of a theory of language policy
is to account for the regular choices made by individual speakers on the
basis of patterns established in the speech community or communities of
which they are members. One such policy is to maintain the existing sta-
tus of a recognized variety, or more realistically, to resist a tendency of
1 This is very much a working paper, originally a draft of the first chapter of a book I am writing set-
ting out the theory of language policy and management. It has a second claim to be included in this
University of Pennsylvania collection: It was written immediately after I gave the 2006 Wolfson
Memorial colloquium. In that paper, I remarked on the growing relationship between educational lin-
guistics and language policy, and suggested that the former constituted essentially the instruments
available to the latter in the educational domain. Thinking about this after the lecture, I wondered why
there is not a theory of language policy, and why some scholars argue that it is just too complicated. I
recalled that Fishman, whose GIDS model is the closest we have to such a model, argues that complex-
ity should not be an excuse. In the new book, I intend to apply the proposed theoretical model to a
number of levels or domains, starting with the family, and to modify it in accordance with the empiri-
cal evidence. The final model will, I hope, be a refined version of this first sketch, but it may also have
been destroyed by the evidence. In the meantime, I welcome criticism.
2 I agree with (Blommaert 2001) that it is a mistake to deal only with labeled languages in a discussion
of language policy.

Working Papers in Educational Linguistics 22/1: 1-14, 2007

 WPEL VOLUME 22, NUMBER 1 TOWARDS A THEORY OF LANGUAGE POLICY

speakers of the variety to shift to the use of another.3 In this essay, a first Fishman, a domain is further distinguished by three characteristics: par-
sketch will be given of a theoretical model of language policy. ticipants, location, and topic. The participants in a domain are
characterized not as individuals but by their social roles and relation-
Language Policy as a Social Phenomenon: The ships. In the family domain, participants are usually labeled with kinship
Domain as Defining Unit terms: father, mother, brother, sister, aunt or uncle, grandfather or grand-
mother, or other appropriate roles such as maid or babysitter. In the
Application of the model to additional cases and data will lead to
school domain, the normal roles are teachers, pupils or students, or prin-
modification and fine tuning of the model. The theory starts with a num-
cipals. In the workplace, they are bosses, employers, workers, employees,
ber of assumptions, which themselves must be open to testing and
foremen, clients, customers. Any individual may fill different roles in dif-
adaptation in the course of the exploration. The first assumption is that
ferent domains, with conflicts sometimes obvious. What variety do I use
while it is intended to account for individual choices, language policy like
with my daughter at school if she is also my pupil? How do I speak to
other aspects of language (as Saussure 1931 pointed out), is essentially a
my son at work if he is also my employer? Secondly, a domain has a typ-
social phenomenon, dependent on the consensual behaviors and beliefs
ical location, usually made obvious by its name. Again, lack of congruity
of individual members of a speech community.
between participant and location – a father switching varieties to a visi-
What is a speech community? From its beginnings, sociolinguistics
tor when he realized this was his son’s teacher and not his friend4 – signal
has avoided a precise answer to this fundamental question. It made a
the existence of norms. Fishman's third component was choice of topic –
clear distinction between a language community – all those who speak a
what is it appropriate to talk about in the domain. Gumperz (1971) has a
specific variety of language – and a speech community – those who share
nice illustration, describing how an employer and employee switch lan-
a communication network, united by agreeing more or less on the appro-
guages when they turn from business to social matters. Essentially, then,
priateness of the use of the multiple varieties used in that community. A
I will be arguing that the regular choices made by an individual are deter-
language community as Hockett (1958), for instance, used the term might
mined by his or her understanding of the language choices appropriate
be the English-speaking world, the complexity of which we realize since
to the domain.5
Kachru (1986) drew our attention to the many varieties which constitute
World English, or the Francophone world (although francophonie is
more a political than a linguistic concept) or at the other extreme, the last Language Policy as Practices, Beliefs and
remaining speakers of a dying language. A speech community, on the Management
other hand, may be a family or a group of people who regularly use the A second assumption, as described in my earlier book (Spolsky 2004),
same coffee shop or work in an office or live in a village or a city (Labov is that language policy has three interrelated but independently describ-
1966) or a region or even a nation (Gumperz 1968). able components: practices, beliefs,6 and management.7 Language
Given this vagueness, although I will regularly talk about speech practices are the observable behaviors and choices – what people actual-
communities, it will be necessary to find a more defined unit. I will start ly do. They are the linguistic features chosen, the variety of language
with the notion of domain, as introduced to sociolinguistics by Joshua used. They constitute a policy to the extent that they are regular and pre-
Fishman in his classic study of the New Jersey barrio (Fishman 1972). dictable, and while studying them is made difficult by the observer's
Although he argued that domains must be empirically defined for any paradox that Labov (1972) identified – for an observer adds an extra par-
specific community, Fishman laid down useful generalizations that I will ticipant and so modifies behavior – describing them is the task of a
adapt. First, a domain is usually named for a social space, such as home sociolinguistic study producing what Hymes (1974) called an ethnogra-
or family, school, neighborhood, church (or synagogue or mosque or phy of speaking. In one sense, this is the real policy although participants
other religious institution), workplace, public media, or government. In may be reluctant to admit it.8 What is critical is that it provides the lin-
building a theory of language policy, I will argue that each of these guistic context for anyone learning language. Children's language
domains has its own policy, with some features controlled internally and
others under the influence or control of external forces. As defined by 4 As Abdeen (2003) recorded in his interviews in an Arab village.
5 This includes the notion of audience design as proposed by Bell (1984) and of accommodation as
3 This desire to maintain the status of a variety (the pattern of those who use it and the functions for explored by Giles, Taylor and Bourhis (1973) and Coupland (1984).
6 Or ideology. I prefer “beliefs” to the political associations of “ideologies.”
which it is used) is commonly matched by a desire to maintain its form, viz. to avoid changes in lexi-
7 Or planning. I prefer “management” as more contemporary than the “planning” so many nations
con, grammar and pronunciation. Fishman (2006) refers to this as the purity dimension (as opposed to
“vernacularity”). adopted in the optimistic days after the Second World War.
8 Try to persuade a literate speaker of French that the /l/ of il is only pronounced before a vowel.

2 3
 WPEL VOLUME 22, NUMBER 1 TOWARDS A THEORY OF LANGUAGE POLICY

acquisition depends in large measure on the language practices to which management decisions: The different values assigned to standard lan-
they are exposed. For example, immigrant parents are often upset to find guages and to heritage languages regularly explain decisions of parents
that their children do not know certain words in their heritage language, as to what language to speak and encourage in the home, just as they
not realizing that they themselves regularly replace them with words bor- explain government decisions on national language policy. Management
rowed from the new language.9 accounts also for some language choices, but it is not automatically suc-
The second important component of language policy is made up of cessful.10 Management, I will also argue, presupposes a manager: The
beliefs about language. The beliefs that are most significant to our con- pressures produced by language practices and beliefs are different in that
cerns are the values assigned to the varieties and features. For instance, they may be authorless.11
given the role played by language varieties in identification, the variety I So far, this theory accounts for language choices within a domain on
associate with my most important membership group – whether it is my the basis only of internal forces, derived from language practices, lan-
nation, my educational class, my region or my ethnic heritage – is likely guage beliefs, and language management within the domain itself. But it
to have the highest value for me, while certain other varieties will be stig- quickly becomes clear that there are forces outside the domain. First is the
matized. Of course, beliefs are not practice: It may well be that I myself fact that any individual is a participant in several levels of his or her com-
use stigmatized forms. munity, that is to say, any individual has different roles in different
The third component is language management, the explicit and domains. The fact that I am at once a parent, a neighbor, a congregant, an
observable effort by someone or some group that has or claims to have employer, and a citizen means that I am familiar with the language prac-
authority over the participants in the domain to modify their practices or tices and beliefs of a number of different domains, and so I may well have
beliefs. The most obvious form of language management is a constitution reason to favor the values of one domain when I am in another. The men
or a law established by a nation-state determining some aspect of official from the Papua-New Guinea village who came home from a year or more
language use: a requirement to use a specific language as medium of in the plantation and chose to speak Tok Pisin provide an example
instruction or in business with government agencies, for example. (Kulick 1992).12 Second, language management provides many examples
Another example is the decision of the Roman Catholic Church at Vatican of efforts to impose language practices on what we might call a lower
II to change the centuries-old policy of requiring Latin for the mass. In the domain, as when an ethnic language revival movement or a school lan-
family domain, efforts by immigrant parents to maintain their heritage guage policy tries to influence home as well as public behavior. This
language or to persuade their children to learn the new language consti- multilevel analysis helps explain some of the problems of centralized lan-
tute language management. guage management, which has to overcome practices, beliefs, and
sometimes management at the lower levels. In an exploration of the sig-
Internal and External Influences on Language Policy nificant domains, it is appropriate to start with internal forces affecting
of a Domain the domain but also necessary to note probable external pressures.
The theory I am exploring will hold that each of these three compo-
nents within (and, as we shall see, others outside) the domain produces Exploring the Domains
forces that account for language choices by participants. Strongest of all The model I am exploring entails a number of defined speech com-
is language practice, for in its absence there is no available model of lan- munities, social levels, or domains, ranging from the family through
guage to learn. As no one in my home ever spoke Yiddish, I missed the various social structures and institutions up to and including the nation-
early opportunity to learn it. The child brought up in a monolingual envi- state and supranational groupings, each of which has pressure for
ronment is denied the possibilities open to a bilingual. Migrants who no language choice engendered by internal and external language practices,
longer speak or hear their language suffer from attrition (Feldman 1997).
Proficiency in a language, whether spoken or written, sets a necessary 10 Just as King Cnut demonstrated that his kingship did not give him control over the waves, so puta-

limit for language choice, and provides a strong instrument for implicit tive language managers find similar powerlessness over language use patterns (Spolsky 2006b).
11 Consider, for example, the argument as to whether the spread of English is the result of demograph-
language management. The other two components also account for sig- ic and economic pressures or the planned activity of an identifiable imperialist conspirator. As a rule, I
nificant forces. My beliefs about the varieties of language from which I will take the position that it is management only when we can identify the manager. A number of
may choose, based on my perceptions of their use inside and outside the scholars cite rhetorical statements in favor of a language as though they proved the existence of lan-

family domain, help account not just for language choices but also for
guage managers.
12 Weinreich (Weinreich 1980) offers this as an explanation of how German was introduced into

9 This point I take from Kopeliovich (2006). medieval Jewish communities by men with external contacts.

4 5
 WPEL VOLUME 22, NUMBER 1 TOWARDS A THEORY OF LANGUAGE POLICY

language belief systems and ideologies, and language management (such as the Roman Army, the French Foreign Legion, the British con-
efforts. Exploring the domain of the home first, the practice of the partic- trolled Indian Army, or the post-independence Israeli army) have
ipants, their language beliefs, and their attempts to influence the encouraged an assortment of management policies, and the desire to
practices and beliefs of other members of the home speech community communicate with the enemy or with the inhabitants of occupied terri-
are critical. Immigrant parents who maintain heritage languages are tory has led to elaborate military language policies.14
obviously more likely to have children who know them, while those who
abandon them are encouraging shift (Kopeliovich 2006). To add profi- The School Domain
ciency in another language, one of the most effective methods is to hire a
Of all domains, school proves to be one of the most complex. Its par-
nanny who speaks that language.13 But the home language ecology is
ticipants bring with them the practices and beliefs of a complex and
quickly influenced by external domains – the Papua-New Guinea men
increasingly multilingual society. Schooling is by its very nature a
who brought Tok Pisin back to the village and the Palestinian fathers
domain committed to language management. The two main categories of
whose working in Hebrew-speaking environments modified their Arabic
participants are students whose language practices and beliefs are to be
(Spolsky & Amara 1986) are examples. Once the children are exposed to
modified and teachers charged with the process of modification.
the language practices and beliefs of their peers in the neighborhood or
Students15 vary of course on a number of critical dimensions: age, gender,
in school, a new conflict is established. Thus, even the family, the pre-
ability level, and motivation, for example. They vary also in the variety
sumably simplest and most basic domain for its effect on natural
or varieties of language that they know and in their level of proficiency.
intergenerational language transmission, turns out to be open to the
The younger they are, the more likely their language pattern is to reflect
influence of other domains.
the language pattern of their home. Their experience in the home
Adding new domains – religious institutions, workplaces, the legal
domain, in the neighborhood, and elsewhere will have introduced them
domain, the health and medical domain, schools, the military, local and
to various language practices, have developed in them beliefs about lan-
regional government, national or federal government, and supranation-
guage and values that they assign to language varieties, and exposed
al organizations – the relevance of external forces adds to the
them to various attempts to modify their language practices and beliefs.
complexity of the model. The test to be applied is how well it appears
Thus, we are by no means dealing with a tabula rasa, for children come to
to account for the forces that should predict language choices within the
school with established language abilities, behaviors, and values.
domain.
The second group of participants in the school domain is made up of
The domains that I am currently exploring provide the evidence.
the teachers. Again, there is variation on such criteria as age, gender,
Religious institutions have their own language policies, especially
training, experience, social status and, of course, language proficiency.
influenced by an established belief about the importance of maintaining
Here too, there may be relative homogeneity or diversity. An addi-
the original language of the sacred texts – Islam and pre-Vatican II
tional factor to be considered is the social, economic, and linguistic
Roman Catholicism resisted use of the vernacular; Judaism allowed and
similarity or dissimilarity between teachers and students. When I first
Protestant Christianity favored translation (Spolsky 2003). In the work-
visited schools on the Navajo Reservation in the late 1960s, 100% of the
place, the languages of managers and of customers turn out to acquire
students were Navajo-speakers with limited if any exposure to English
extra value, and globalization buttresses international languages and
before they came to school, while over 90% of the teachers were English
especially English (Coleman 1985; Dicker 1998). In the legal domain,
speakers with virtually no knowledge of Navajo (Spolsky 1970). This
there has been growing pressure to permit the increasing numbers of
situation, not uncommon in developing societies or in communities
immigrants to understand the process of their trials, adding a new par-
with large numbers of immigrants, reflects the fact that teachers are
ticipant (the legal interpreter) to the domain. In the health domain,
commonly hired only from among those who have successfully com-
where there are pragmatic reasons for professional participants (doctors
pleted many more years of schooling than minority students can yet
and nurses) to be able to communicate with their patients, the provision
of qualified interpreters (as opposed to the common use of patient’s 14 The Canadian effort to establish a bilingual defense force is an exception, driven completely by a
bilingual children) has depended on civil rights pressures in Europe government language policy on bilingualism and finally unsuccessful. In the US, the ASTP was an

and the US. In the military, the complexities of multilingual armies

effort in the Second World War to make up for the failures of language teaching in US schools (Iglehart
1997; Spolsky 1995). The National Security Language Initiative (Spolsky 2006a) is the latest manifesta-
tion of this.
13 US newspapers reported on high wages paid to Mandarin-speaking nannies by families with expec- 15 I use this term generally for the learners, whether they are in pre-school, elementary or high school,
tations of growing trade relations with China. university, or adult education programs.

6 7
 WPEL VOLUME 22, NUMBER 1 TOWARDS A THEORY OF LANGUAGE POLICY

hope to attain. It is the basis for what I have called the home-school lan- governments may select policies considered unsuitable in certain regions;
guage gap (Spolsky 1974), the fact that teachers use a language which State and Federal governments may vary; religious groups may disagree
their students do not understand. The first problem in such situations is with government policy. One of the special features of school is that the
(or should be) establishing communication between students and teach- teaching takes place in a closed room, difficult for outsiders to observe.
ers. This, of course, increases the power of teachers, so that complex systems
There are other potentially significant participants in the school of control (classroom visits, centrally controlled microphones and video
domain. The first among these are the professional administrators – prin- cameras, or most commonly, externally administered tests and examina-
cipals and department heads in schools, provosts and deans and chairs in tions) are needed.
universities – who may be selected from the same group as the teachers All of these are participants whose beliefs need to be taken into
and who may be responsible to authorities outside the school for man- account and who might function as managers of language education pol-
agement of its educational and language policies. A second significant icy. Add to this the existence of activist groups – groups of parents or
group may be the non-academic support staff – the bus drivers, secre- community members – attempting to influence the school or school
taries, cleaners and cooks. In the Navajo schools in the 1960s, these were authorities at any level, and one readily realizes the underlying structural
some of the only people on the staff who could speak Navajo and so com- explanation of the multiplicity of language education patterns that we
municate with the students and with their parents. find in practice.
Each of these categories of participant brings significant language
practices and beliefs to the school domain, but our critical question is Language Education Policy
what determines the language instructional policy of the school. Here,
The complexity of possible patterns was captured by Mackey (1970) in
the variation is once again enormous. There are schools where manage-
his pioneering typology of bilingual education.18 The crucial dimensions
ment is essentially internal, with the school staff (principal, teachers, and
of this typology were named varieties of language, year or level of
other relevant professionals) determining their own educational and lin-
instruction, amount of time allocated in the weekly schedule, medium,
guistic goals and choosing their own appropriate method of achieving
topic or subject. The broadest categories were transitional programs
them. More commonly, there is some individual or group external to the
(starting in one language and gradually moving to another) or mainte-
school domain with the authority to establish goals and methods. In
nance programs (starting in one and moving to two). The varieties of
some cases, this may be the parents of the students, working as members
language in competition are commonly the various home varieties (ver-
of an elected school board or through their financial power to influence
naculars or dialects) and the official national language.
school policy. In other cases, the school will be under the authority of a
There are a number of possible points in the system for transition from
religious leader or religious organization. In others, there may be a demo-
the home or local variety to the national school variety. One model,
cratically elected school board with authority over several schools in the
adopted in the British Empire after failures of the English-only program
region. Sometimes, this authority is assigned to a local body such as a city
in 19th century India (Evans 2002), was to provide initial education in the
council.16 In other cases, the authority is centralized and under the con-
vernacular with gradual transition to English no later than the beginning
trol of the central government or, in a federal system, of state or
of high school. The number of years of vernacular instruction has varied,
provincial governments. In colonial systems, authority was commonly
although a consensus from recent research in Africa and elsewhere sug-
placed on the metropolitan home government.17 In normal practice,
gests that six years is needed to achieve good educational results (Heugh
authority over the school programs is divided among these various lev-
2005; Walter 2003). There are systems which follow the French and
els. Each arrangement is likely to have different effects on the
Portuguese colonial models and that assert that education must be in the
establishment and implementation of school language programs.
standard metropolitan language from the very beginning.
Conflicts among the levels are common. Parents may favor one lan-
This controversy over the educational value of instruction in the home
guage and teachers another; the principal may attempt to implement a
language remains one of the most basic issues in language education pol-
policy that varies from the desires of higher and lower levels; national
icy. It depends ultimately on contrasting beliefs about the ability of
16 Recently, the Washington City Council has been studying how the New York City Council managed children to learn language. One position holds that it is just as easy for
to take over control of schools from the education board. 18 Because of the great variation in the meaning and application of this term, and because of the great
17 Phillipson (1992) explores this phenomenon as an explanation for the spread of English. The influ-
deal of political emotion that it engenders, I shall do my best to avoid using it without careful defini-
ence of French colonial policy (Bokamba 1991) and of Spanish (Mar-Molinero 2000) are perhaps even
tion.
clearer examples.

8 9
 WPEL VOLUME 22, NUMBER 1 TOWARDS A THEORY OF LANGUAGE POLICY

pre-adolescent children to pick up a new language as it is assumed to be Conclusion

for babies to acquire their first language. For some children, this appears Despite these constraints, it will be rare for linguistic or educa-
to be true, but on average, the evidence of published research suggests tional considerations to determine school language policy. More
strongly that the majority suffers educationally in such situations (Walter commonly, schools reflect the ideological position of those who con-
2003). trol them. Normally, their policy will be driven in part at least by the
Another consideration is the state of the varieties of language being policy of the national government. Thus, recent major changes from
looked at. The issue is the state of development of each potential lan- Bahasa Melayu to English and from Urdu to English for teaching sci-
guage of instruction. There is a scale (a typology proposed by Stewart ence and mathematics in Malaysia and Pakistan respectively, or the
1968), ranging from an unwritten and unstandardized vernacular, intention to provide six years initial instruction in the vernacular in
dialect, creole, or pidginto a standard or classical language, the upper South Africa, or the intention to restore English as medium of instruc-
end implying the availability of a writing system (a critical feature in tion in the Philippines have all been announced as central
schooling which assumes literacy as a first goal) and of a dictionary government policy; the insistence on instruction in Kannada in
and grammar (also demanded by school teachers) as well as of modern Karnataka is a decision of the state government; efforts to establish
terminology. This quickly becomes apparent when you visit schools English-only programs and ban bilingual education have been
beginning to use home languages. When we worked on developing focused on state governments in the US and in particular on those
elementary education in Navajo, one of the big gaps that needed to be states with popular referenda. Language instruction in the school
filled was lexicon for mathematics (Navajo had no simple way to label then is a key component in national language policy and is recog-
fractions, for instance). I also have vivid memories of a Maori first nized as a key stage in Fishman’s (1991) GIDS model.
grade teacher discussing with her aunt (acting as language assistant), The present language policy model, then, suggests that the school
with the school inspector who was my escort, and with the students domain is the one most likely to be influenced externally, whether
what Maori word to use for some new concept. It is this that sets the from “below” (home, religion, neighborhood) or “above” (levels of
inevitable bridge between two kinds of language management, man- government), and to be most often the target of activist intervention
aging speakers (what Kloss 1969 labeled status planning) and in support of one variety or another. The study of language education
modifying a language (what Kloss called corpus planning, but perhaps policy is thus perhaps the most difficult and challenging field of all,
better labeled using the Prague School term cultivation [Prague School and deserves the thorough attention it receives in this special issue of
1973]). Fishman (2006) has argued that these two processes commonly WPEL.
share the same motivational dimensions, but there is an even earlier
connection in that the assignment of a function to a variety regularly
entails that it must be modified to fill that function, either by develop- Born in New Zealand in 1932, Bernard Spolsky was educated at Victoria University
ing a writing system or by modernization. and the University of Montreal. He taught at McGill University, Indiana
What this means is that programs that aim to use the local or home University, the University of New Mexico and Bar-Ilan University, retiring in
variety as language of instruction commonly incur the extra expense 2000. He has written and edited two dozen books, including Conditions for Second
and trouble of needing to add a language cultivation component. It has Language Learning (1989), The Languages of Jerusalem (1991), Measured Words
been suggested that one of the problems with implementing the South (1995), Sociolinguistics (1998), The Languages of Israel (1999), Concise
African constitutional recognition of nine languages alongside English Encyclopedia of Educational Linguistics (1999) and Language Policy (2004), and
and Afrikaans is the weak provision of resources for language develop- about 200 articles and chapters. He was founding editor of three journals: Applied
ment (Alexander 2004). The 1970s Navajo educational programs were Linguistics, Journal of Asia TEFL, and Language Policy. He lives in the Old City of
supported by projects concerned with writing or reprinting material in Jerusalem where he is writing a monograph on fundamentals of language manage-
Navajo. Maori education has been similarly hampered by the absence of ment.
contemporary writing in the language, apart from material prepared for
school use and translated government documents. Earlier European E-mail: [email protected]
language maintenance and revival programs were fortunate in that
their leaders were often literary figures who had begun writing in the
language. The absence of (or need for) language cultivation can then
serve as a reason or excuse for not using home varieties in school.

10 11
 WPEL VOLUME 22, NUMBER 1 TOWARDS A THEORY OF LANGUAGE POLICY

References Hymes, D. (1974). Foundations in sociolinguistics: An ethnographic approach.

Philadelphia: University of Pennsylvania Press.
Iglehart, F.N. (1997). The short life of the ASTP. Baltimore, MD: American Literary
Abdeen, W. (2003). Sociolinguistics of Silwan. Unpublished doctoral dissertation, Press.
Bar-Ilan University, Ramat-Gan. Kachru, B.B. (1986). The alchemy of English: The spread, functions and models of non-
Alexander, N. (2004). Socio-political factors in the evolution of language policy in post- native Englishes. Oxford: Pergamon Institute of English.
apartheid South Africa (LAUD Paper 605). Essen, Germany: University of Kloss, H. (1969). Research possibilities on group bilingualism: A report. Quebec:
Duisburg-Essen. International Center for Research on Bilingualism.
Bell, A. (1984). Language style as audience design. Language in Society, 13(2), 145- Kopeliovich, S. (2006). Reversing Language Shift in the Immigrant Family: Case study
204. of a Russian-speaking community in Israel. Unpublished doctoral disserta-
Blommaert, J. (2001). The Asmara Declaration as a sociolinguistic problem: tion, Bar-Ilan University, Ramat-Gan.
Reflections on scholarship and linguistic rights. Journal of Sociolinguistics, Kulick, D. (1992). Language shift and cultural reproduction: Socialization, self and syn-
5(1), 131-142. cretism in a Papua New Guinean village. Cambridge and New York:
Bokamba, E.G. (1991). French colonial language policies in Africa and their lega- Cambridge University Press.
cies. In D.F. Marshall (Ed.), Language planning (Vol. III of Focusschrift in Labov, W. (1966). The social stratification of English in New York City. Washington,
honor of Joshua A. Fishman on the occasion of his 65th birthday, pp. 175- DC: Center for Applied Linguistics.
214). Amsterdam & Philadelphia: John Benjamins. Labov, W. (1972). Some principles of linguistic methodology. Language in Society,
Coleman, H. (1985). Talking shop: An overview of language and work. 1(1), 97-120.
International Journal of the Sociology of Language, 51, 105-129. Mackey, W. (1970). A typology of bilingual education. Foreign Language Annals,
Coupland, N. (1984). Accommodation at work: Some phonological data and their 3(4), 596-608.
implications. International Journal of the Sociology of Language, 46, 49-70. Mar-Molinero, C. (2000). The politics of language in the Spanish-speaking world: From
Dicker, S.J. (1998). Adaptation and assimilation: US business responses to linguis- colonisation to globalisation. London: Routledge.
tic diversity in the workplace. Journal of Multilingual and Multicultural Phillipson, R. (1992). Linguistic imperialism. Oxford: Oxford University Press.
Development, 19(4), 282-302. Prague School. (1973). General principles for the cultivation of good language
Evans, S. (2002). Macaulay's Minute revisited: Colonial language policy in nine- (P.L. Garvin, Trans.). In J. Rubin & R. Shuy (Eds.), Language planning:
teenth-century India. Journal of Multilingual and Multicultural Current issues and research (pp. 102-111). Washington DC: Georgetown
Development, 23(4), 260-281. University Press.
Feldman, Z. (1997). Language attrition among former Soviet immigrants to Israel. Saussure, F. de. (1931). Course de linguistique generale. Paris: Payot.
Unpublished master’s thesis, Bar-Ilan University, Ramat-Gan. Spolsky, B. (1970). Navajo language maintenance: Six-year-olds in 1969. Language
Fishman, J.A. (1972). Domains and the relationship between micro- and macroso- Sciences(13), 19-24.
ciolinguistics. In J. Gumperz & D. Hymes (Eds.), Directions in sociolin- Spolsky, B. (1974). Linguistics and the language barrier to education. In T. A.
guistics (pp. 435-453). New York: Holt Rinehart and Winston. Sebeok, A.S. Abramson, D. Hymes, H. Rubenstein, E. Stankiewicz & B.
Fishman, J.A. (1991). Reversing language shift: Theoretical and empirical foundations of Spolsky (Eds.), Current Trends in Linguistics: Linguistics and adjacent arts
assistance to threatened languages. Clevedon, England: Multilingual and sciences (Vol. 12, pp. 2027-2038). The Hague: Mouton.
Matters Ltd. Spolsky, B. (1995). Behind the ASTP Myth. In G.H. Richins & R.K. Belnap (Eds.),
Fishman, J.A. (2006). Do not leave your language alone: The hidden status agendas Selected Papers from the Proceedings of the Twentieth Annual Symposium of
within corpus planning in language policy. Mahwah NJ: Lawrence Erlbaum the Deseret Language and Linguistics Society, 3-4 March 1994 (pp. 119-124).
Associates Publishers. Provo Utah: Deseret Language and Linguistics Society.
Giles, H., Taylor, D.M., & Bourhis, R. (1973). Towards a theory of interpersonal Spolsky, B. (2003). Religion as a site of language contact. Annual Review of Applied
accommodation through language: some Canadian data. Language in Linguistics, 23, 81-94.
Society, 2(2), 177-192. Spolsky, B. (2004). Language Policy. Cambridge: Cambridge University Press.
Gumperz, J. (1971). Language in social groups. Stanford: Stanford University Press. Spolsky, B. (2006a). Does the US need a language policy, or is English enough?
Gumperz, J.J. (1968). The speech community. In D. L. Sills (Ed.), International Language policies in the US and beyond. In A. Heining-Boynton (Ed.),
Encyclopedia of the Social Sciences (Vol. 9, pp. 381-386). New York: The ACTFL 2005-2015: Realizing our vision of languages for all (pp. 15-38).
Macmillan Company. Upper Saddle River NJ: Pearson Education.
Heugh, K. (2005). Language education Policies in Africa. In Keith Brown (Ed.), Spolsky, B. (2006b). Language policy failures - why won't they listen? In M. Pütz,
Encyclopedia of langauge and linguisticss. (Vol. 6, pp. 414-422). Oxford: J. A. Fishman & J. Neff- van Aertselaer (Eds.), 'Along the Routes to Power':
Elsevier. Explorations of empowerment through language (pp. 87-106). Berlin & New
Hockett, C.F. (1958). A course in modern linguistics. New York: Macmillan. York: Mouton de Gruyter

12 13
 WPEL VOLUME 22, NUMBER 1

Spolsky, B.& Amara, M.H. (1986). The diffusion and integration of Hebrew and
English lexical items in the spoken Arabic of an Israeli village.
Anthropological Linguistics, 28, 43-54.
Stewart, W. (1968). A sociolinguistic typology for describing national multilin-
gualism. In J. A. Fishman (Ed.), Readings in the sociology of language (pp.
531-545). The Hague: Mouton.
Walter, S.L. (2003). Does language of instruction matter in education? In M. R.
Wise, T. N. Headland & R. M. Brend (Eds.), Language and life: essays in
memory of Kenneth L. Pike (pp. 611-635). Dallas TX: SIL International and
the University of Texas at Arlington.
Weinreich, M. (1980). History of the Yiddish language (J.A. Fishman & S. Noble,
Trans.). Chicago: University of Chicago Press.

14