Spolsky - Sociolinguistics

Oxford lntroductions to Language Study
Senes Editor H.G.Widdowson
Published in this series:

Rod Ellis: Second Language Acquisition
Thomas Scovel: P sych olinguistics
Bernard Spolsky: Sociolinguistics
H.G. \)iddo wson: Li n gu ist i c s

George YtIe: P r agmatics
Claire Kramsch Language and Cubure
mm&m&fim
nxfim&ffimm
Bernard Spolsky
OXFORD UNIVE RSITY PRESS
'
Contents
Preface
Author's preface
IX
XII
SECTION I
Survey
The social study of Ianguage

The scope of enquiry
Complementary approaches
The methods of enquiry
hat are the data?

The sociolinguist at work
The approach in this book
8
8
IO
r3
The ethnography of speaking

and the structure of conversation
The structure of conversations
Politeness and politeness formulas
Terms of address
r6
r9
zo
Locating variation in speech

Speech communities and repertoires
z4
r4
Dialect
Styles, gender, and social class
Styles
Specialized varieties or registers and domains
Slang and solidarity
)r
33
35
Language and gender
Social stratification
Accommodation and audience design
36
39
4r
Bilinguals and bilingualism

Language socialization
The description of bilingualism
Bilingual competence
Code switching and code mixing
44
45
47
49
Societal multilingualism
Multilingualism
5r
Language loyalty and reversing language shift

Language and ethnic identity
Language and politics
Language rights
Pidgins and creoles
Diglossia
55
57
58
59
6r
63
Applied sociolinguistics
Language policy and language planning
Status planning
Corpus planning
Normativism and prescriptivism
Language acquisition planning or language
education policy
Language diffusion policy or linguistic imperialism
The spread of English-imperialism or hegernony?
66
68
7o
74
75
76
Conclusions
78
SECTION 2
Readings
79
SECTTON
References
SECTTON 4
III
Glossary
zI
Acknowled{ements
r27
Preface
Purpose
What justification might there be for a series of introductions to
language study? After all, linguistics is already well served with
introductory texts: expositions and explanations which are comprehensive, authoritative, and excellent in their way. Generally
speaking, however, their way is the essentially academic one of
providing a detailed initiation into the discipline of linguistics,
and they tend to be lengthy and technical: appropriately so, given
their purpose. But they can be quite daunting to the novice. There
is also a need for a more general and gradual introduction to language: transitional texts which will ease people into an understanding of complex ideas. This series of introductions is designed
to serve this need.
Their purpose, therefore, is not to supplant but to support the
more academically oriented introductions to linguistics: to prepare the conceptual ground. They are based on the belief that it is
an advantage to have a broad map of the terrain sketched out
before one considers its more specific features on a smaller scale, a

general context in reference to which the detail makes sense. It is
sometimes the case that students are introduced to detail without
it being made clear what it is a detail of . CIearIy, a general understanding of ideas is not sufficient: there needs to be closer scrutiny.
But equally, close scrutiny can be myopic and meaningless unless
it is related to the larger view Indeed it can be said that the precondition of more particular enquiry is an awareness of what, in
general, the particulars are about. This series is designed to provide this large-scale view of different areas of language study. As
such it can serve as preliminary to (and precondition for) the more
PREFACE
IX
specific and specialized enquiry which students of linguistics are

required to undertake.
But the series is not only intended to be helpful to such students. There are many people who take an interest in language
without being academically engaged in linguistics per se. Such
people may recognize the importance of understanding language
for their own lines of enquirn or for their own practical purposes,
or quite simply for making them aware of something which figures so centrally in their everyday lives. If linguistics has revealing
and relevant things to say about language, this should presumably not be a privileged revelation, but one accessible to people
other than linguists. These books have been so designed as to
accommodate these broader interests too: they are meant to be
introductions to language more generally as well as to linguistics
as a discipline.
Design
The books in the series are all cut to the same basic pattern. There
are four parts: Survey, Readings, References, and Glossary.
Survey
This is a summary overview of the main features of the area of
language study concerned: its scope and principles of enquiry, its

basic concerns and key concepts. These are expressed and
explained in ways which are intended to make thenr ars accessible
as possible to people who have no prior knowledge or expcrrise in
the subject. The Survey is written to be readable and is unclurtered by the customary scholarly references. In tlris scnse, it is simple. But it is not simplistic. Lack of specialist expcrtise does nor
imply an inability to understand or evaluate ideas. Ignorance
means lack of knowledge, not lack of intelligence. The Survey,
therefore, is meant to be challenging. It draws a map of the subject area in such a way as to stimulate thought and to invite a crit-
ical participation in the exploration of ideas. This kind of
conceptual cartography has its dangers of course: the selection of

what is significant, and the manner of its representation, will not
be to the liking of everybody, particularly not, perhaps, to sorne of
those inside the discipline. But these surveys are written in the
PREFACE
belief that there must be an alternative to a technical account on

the one hand and an idiot's guide on the other if linguistics is to be
rrrade relevant to people in the wider world.
Readings
Some people will be content to read, and perhaps re-read, the
summary Survey. Others will want to pursue the subiect and so
will use the Survey as the preliminary for more detailed study. The
Readings provide the necessary transition. For here the reader is
presented with texts extracted from the specialist literature. The
purpose of these Readings is quite different from the Survey' It is
to get readers to focus on the specifics of what is said and how it is
said in these source texts. Questions are provided to further this
purpose: they are designed to direct attention to points in each
text, how they compare across texts, and how they deal with the
issues discussed in the Survey. The idea is to give readers an initial

familiarity with the more specialist idiom of the linguistics literature, where the issues might not be so readily accessible, and to
encourage them into close critical reading.
References
One way of moving into more detailed study is through the

Readings. Another is through the annotated References in the
third section of each book. Here there is a selection of works
(books and articles) for further reading. Accompanying comments indicate how these deal in more detail with the issues discussed in the different chapters of the Survey.
Glossary
Certain terms in the Survey appear in bold. These are terms used
in a special or technical sense in the discipline' Their meanings are
made clear in the discussion, but they are also explained in the
Glossary at the end of each book. The Glossary is cross-referenced to the Survey, and therefore serves at the same time as an
index. This enables readers to locate the term and what it signifies
in the more general discussion, thereby, in effect, using the Survey
as a summary work of reference.
PREFACE
XI
Use
The series has been designed so as ro be flexible in use. Each title is
separate and self-contained, with only the basic format in com_
mon. The four sections of the format, as described here, can be
drawn upon and combined in different ways, as required by the
needs, or interests, of different readers. Some may beiontent with
the Survey and the Glossary and may not want to follow up the
suggested References. Some may not wish to venture into the
Readings. Again, the Survey might be considered as appropriate
preliminary reading for a course in applied linguistics or teacher
education, and the Readings more appropriate for seminar discussion during the course. In short, the notion of an introduction
will mean different things to different people, but in all cases rhe
concern is to provide access to specialist knowledge and stimulate
an awareness of its significance. The series as a whole has been
designed to pr:ovide this access and promote this awareness in
respect to different areas of language study.
H. G.I /IDDOl /SON
Author's Preface
The invitation to write this short book is anorher of the many
debts I owe to Henry ) iddowson who, over the years that we
have known each otheq has managed to challenge a'd stimulate
me continually. The special challenge this tirne is to follow in
admired footsteps, for there have been many earlier and more
detailed introducions ro sociolinguistics from which I myself
educational effects of language variation was to continue to be

cncouraged. I was fortunate enough to have a spell living in
Montreal, a city whose people and scholars have made pioneering
endeavours in the realm of multilingualism. Later in my career, an
nvitation to teach at the University of New Mexico landed me in
the midst o an area where students were demonstrating for the
right to remain bilingual. More recentln living in Israel, I have
come to learn and appreciate the complex patterns of language
that make up this county and its surroundings.
My attention to language variation has often had a practical
bent, because I have also been involved in studying language
lcarning and language policy. In spite of this, the central question
I continue to ask, and the one that this volume encourages readers
to ask, pertains to the close intertwining between a language and
the social context in which it is used. Language and society may
not be peculiarly human-how else can one appreciate social
amoebae or the honey bee?-but they are such fundamental
human phenomena that they cry out for better understanding.
Muhammad Amara and Henry'!iddowson read earlier drafts
of this book and made many useful suggestions. In preparing it, I
have benefited from the help of the staff of the English Language
Teaching Division at Oxford University Press who have once
again demonstrated the friendly efficiency that makes an author's
life easy.
This book is dedicated to my grandchildren, busy studying

sociolinguistics in their own way.
BERNARD SPOLSKY
have benefited.
My task, as Widdowson defines it, is to sketch out a conceptual
map for the interested reader of the relations between language
and society. This is, in some respects, bound to be a personal view.
My own curiosity about sociolinguistics grew out of language
teaching. As a young high-school reacher in a New Zealand iown,
my interest was piqued by the bilingualism of some of my Maori
students. \X/hn I naively asked, did boys who spoke Maori at
home write better English essays than those whose pare'ts spoke
to them in a limited version of English? This early interest in the
XII PREFACE
PREFACE XIII
SECTION I
Survey
The social study of language
The scope of enquiry

Sociolinguistics is the field that studies the relation between language and societn between the uses of language and the social
srructures in which the users of language live. It is a field of study
thzrt assumes that human society is made up of many related
patterns and behaviours, some of which are linguistic.
One of the principal uses of language ls to communtcate meaning, but it is also used to establish and to maintain social relationsl-rips. \7atch a mother with a young child. Most of their talk is
devoted to nurturing the social bond between them. Listen to two
'riends talking. Much of their conversation functions to express
rrnd refine their mutual compact of companionship.
'!7hen
you
rreet strangers, the way they talk informs you about their social
and geographical backgrounds, and the way you talk sends subtle
or blatant signals about what you think of them. It is these aspects
of language use that sociolinguists study.
In the thirty years or so that it has been recognized as a branch
of the scientific study of language, sociolinguistics has grown into
one of the most important of the 'hyphenated' fields of linguistics.
This term distinguishes the core fields of historical and descriptive
linguistics (phonology, morphology, and syntax) from the newer
interdisciplinary fields like psycholinguistics, applied linguistics,

neurolinguistics, and sociolinguistics or the sociology of lanEluage. Stranded at times between sociology (one of the field's
putative parents) and linguistics (the other), the practitioners
of sociolinguistics have so far avoided the rigorous bounds of a
single theoretical model, or the identifying shelter of a single
THE SOCIAL STUDY OF LANGUAGE
professional organization. They apply a plethora of methods to a

multitude of subjects that all have in common one single thread:
languages and their use in social contexts.
There are indeed some sociolinguists who wonder how language can be studied in any other way. They believe that the
search
of the formal linguists like Noam Chomsky for
an
autonomous linguistics, with the goal of describing the idealized

competence of an idealized monolingual in an idealized monovariety speech community, is as doomed to failure as was the earlier effort of structural linguists to account for language structure
without taking meaning into account.
\hile Noam Chomsky revolutionized many aspects of linguistics, he followed structural linguists like Leonard Bloomfield in
choosing to study language autonomously, as a seli-sufficient
system. He aimed to find a basic universal grammatical structure
that could account for the similarities in the organization of

languages, without needing to appeal to the social context in
which language is used. For Chomsky, the existence of variation
in language simply confuses, diverting the linguist's atention
from the wonderful abstract system that separates human language from other communication systems. For the sociolinguist,
however, the most important verity is that a language-any language-is full of systematic variation, variation that can only be
accounted for by appealing, outside language, to socially relevant

forces and facts.
Sociolinguistics takes as its primary task to map linguistic variation on to social conditions. This mapping helps understand not
just synchronic variation (variation at a single point of time), but
also diachronic variation (variation over time) or language change.
The close intertwining of linguistic and social facts is crucial to
a sociolinguistic approach. Even before small children can speak
clearly, they develop a distinct style of address to be used when
speaking to anyone or anything smaller. As they grow, they add
more and more variations to their speech, and these come to be
associated with recognizable stytes. As early as the age of five,
children asked to play roles try to imitate the styles of speech of
many different people. These small variations in language that

everyone acquires in normal upbringing can be used to identify
us, or the person we are talking to, or the subject we are talking
S
RVEY
about. There is no single-style or single-variety speaker; no speech
community that does not have a choice of varieties; and many

fewer monolinguals than English speakers might imagine.
The existence of patterned variation in language makes it possi-
ble to identify ourselves and others as belonging to certain
groups. The social prestige or stigma associated with these variations makes language a source of social and political power. Only
by including both linguistic and social factors in our analysis can
this complex but rule-governed behaviour be accounted for. To
do this is the chosen goal of the sociolinguist.
Com plementary approaches

lJschewing the normal acrimony of academic debate, we might
say that the various complementary approaches to the study of
language each find a different aspect of the complex phenomenon
to be of enthralling interest. The formal linguist pursues an
irutonomous universal systeirl, significant elements of which are
to be explained by the very"design and structure of the human
brain. The psycholinguist asks how such a system woks and how
it can be learned or lost. The sociolinguist asks how it is used in a
living and complex speech community. The answers to'each of
these questions is important to the applied linguist, concerned,
arnong other matters, with how to help people learn language and
how to use it effectively to deal with problems of practical everyclay life.
Just as the formal linguist and the psycholinguist focus their
attention on the language as a system with universal features, so
the sociolinguist looks at the complex connections benveen the
variations within a language and the matching variations in the
'S(hy,
the song asks, do l say /tem'olteu/
social groups that use it.
and you say /tem'erteui. \X/hy do I say 'Good morning!' to some
people, and 'Hi!' to others? How did Professor Higgins know

which parts oEliza Doolittle's speech needed to be changed to
rnake her pass for a member of the upper class? Sociolinguistics is
ail about variation, and seeks socially relevant explanations for
regular patterns of variation in language use.
A sociolinguist is interested in the way that members of a
speech community can, and do, identify and respond to fine
TF{E SOCIAL STUDY OF LANGUAGE
differences in language usage that are associated, within a speech
community, with social or economic or political or religious or

cultural or other divisions of the society. At what is often called
the micro end of sociolinguistics, the sociolinguist's goal might be
to show how specific differences in pronunciation or grammar

lead members of a speech community to make judgments about
the education or economic status of a speaker. In New York City,
for instance, pronouncing the word 'this' as /drs/ or pronouncing
'bird' as lbctd/, marks the social class of the speaker. In the same
way, the choice of lexical items (saying 'doctor's surgery' instead
of 'doctor's office') makes clear on which side of the Atlantic a
speaker of English has been living. Over-average use of rag questions would seem to mark young females in New Zealand. A high
proportion of words from Modern Standard Arabic marks the
vernacular speech of educated Arabs. As much as speech itself
communicates content, so the form of speech, the selection
among available socially marked variants, communicates impclrtant social information about the speaker and the listener and
about their relationship to each other. In other words, adapting
Marshall Mcluhan's famous words, the medium (the variety chosen) becomes the message itself.
At the other-the macro-end of the
spectrum, sometimes
labelled the sociology of language as distinct from sociolinguistics,

the scholar's primary attention turns from the specific linguistic
phenomena to the whole of a language or variety (a terrn we use to
include any identifiable kind of language). In macro-
sociolinguistics, we treat language (and a specific language)

alongside other human cultural phenomena. \7e might ask, for
instance, about the significance of a group of imrnigrants shifting
completely to a new language or maintaining their olcl one for

some purposes. \ /hy did most immigrants to the LISA from
Northern Europe drop their home language so fast, while Asian
immigrants to Britain seem to be keeping theirs alive? How did
Welsh survive English occupation, or how dicl Basque and

Catalan stay alive under Franco's policy of enforcirrg the use of
Castilian Spanish? 'S7e might investigate the close boncls berween
language choice and social identity, asking why their language
remained so important to Maoris, Basques, or Frisians that they
have been willing to undertake political acion ro preserve ir. !7e
SURVEY
rnight ask why speakers of certain varieties are influential and

powerful, and why speakers of other varieties are regularly discriminated against. These questions concern the use of a language
or a language variety as a whole rather than individual variations,
rrnd asking them makes the study of language a means to understanding a society.
Some scholars use this difference of perspective to divide the
lield into two. They want to distinguish between sociolinguistics,
which emphasizes the social influences on language, and the

sociology of language, which emphasizes the role of language in
society. In this book, however, I shall examine both aspects
together, arguing for a common concern for the connection
between language and society rather than a need to give priority
ro one or the other.
There are, as we
will see, a large number of discrete but related

phenomena that can be considered from the two perspectives.
Included are topics like language and gender, social stratification'
language planning, language and power, language and ethnicitn
language and nationalism. To capture relevant dimensions of the
patterns under study, the sociolinguist is regularly forced to
broaden the scope of enquiry from the parent disciplines of

linguistics and sociology to other fields like social psychologn
gender studies, public policy management, political science, and
h
istory.
If there is a common theme that emerges from the studies of

sociolinguists, it is that the complex interplay of language structure with social structure means that any user of language is con-
stantly responding
to and signalling social information. My
identity (or rather my various identities) is recognizable from my

choice among all the variants that a language offers. One might
even go so far as to claim that it is this choice of identity that
accounts, better than the other factors we will consider' for the
observable variation in natural speech.
But before we can consider these more specific topics and the
theories held to explain them, it is valuable to make clear what the
data of sociolinguistics are, and what methods are used to collect
these data.
The methods of enquiry

Wbat to study and how to study it are closely related questions. A
theory tells us which facts count as data. A science advances either
by posing new important questions about the data or by finding
new ways of observing data to answer important questions.
Noam Chomsky initiated a revolution in linguistics by asking

how to account for the fact that everyone who learns a language
(a
first language, that is to say) shows evidence ofcontrol ofrules
that are not evident from normal exposure to people speaking.

Sociolinguists ask a different question: how to account for the
variation that exists in every language. Because of what they
study, they are just as concerned with how to answer the question
as with the question itself.
The reason for this concenl is because we need to observe a
dynamic phenomenon in its natural setting. In biologn one common
way of studying frogs is by dissecting them, but of course a dead frog
doesn't do very much. In sociolinguistics, the analogous riddle is
what Labov labelled the observer's paradox, namely: how can we
observe the way people speak when they are not being observed?
The enigma arises because sociolinguists believe that language

use is always sensitive to the social relations among the participants in a speech event. We speak differently to superiors, to colleagues, to friends, and to children. Our speech patterns regularly
change when another person (especially a stranger) enters the
conversation. Even young children quickly develop markedly different styles for talking to different people. Given this sensitivity
of speech to audience, how can a linguist, who is a stranger and an
outside observer, witness and record the vernacular and unmonitored speech patterns that close friends use among themselves?
Doesn't the presence of the sociolinguist kill the naturalness of the
speech just as effectively as the dissector's scalpel kills the frog?
There has been no definitive answer. Sociolinguists have countered the paradox in different ways, as we shall see.
What are the data?

The question of what constitutes interpretable data is a key one in
any discipline. In sociolinguistics, there is a tension between the
S
I]RVF,Y
observers and the quantifiers. Joshua Fishman has recounted that

when he and John Gumperz were working together on a study of
lrilingualism among Puerto Ricans in the Jersey City barrio,

(iumperz would occasionally ask him to produce evidence to
back up an assertion. Fishman, with his background in sociology
ncl psychology, would bring in bundles of computer print-out of
'r
'!7hen
Gumperz himself was similarly
statistical calculations.
challenged by Fishman, Gumperz, trained as an ethnographer,
would describe a conversation he had overheard at a party the
rright before. Note that Fishman used elicited data that could be
nalysed statistically, while Gumperz discovered his data through

observation ofthe use oflanguage in a natural setting.
lf we try to understand the basis for these two approaches' we
rnight say that sociologists seek their evidence in the patterns
rrnderlying the answers of a large number of people to many carelLrlly designed questions. The social forces that control individual
speech behaviour are expressed, sociologists believe, in the statist ically determinable tendencies that can be extracted by analysing
l,rrge quantities of data. Ethnographers, on the other hand,
trrrined to compare the behaviour of individuals in one culture
with the patterns of behaviour observed in many other cultures,
hrve learned to present their intuitions through the interpretation
of single events that they carefully observe. The two approaches
rr
complementary.
\Thatever method they choose, what sociolinguists are looking
ior is evidence of socially accepted rules accounting for variations
in speech. The evidence is in part the speech variation (the differ-
,r
re
in pronunciation or word choice, or in
grammatical
(age,
gender,
the
speaker
choice), in part the characteristics of
of
part
nature
in
the
and
cclucation, place of birth, and domicile),
of
the
relations
role
(its
its
topic'
the
place,
the speech encounter
speakers). Some of these data can be collected by observation,
stlme by elicitation. \hether to trust the observation is of course
e problem: the speaker might be pretending or lying, the observer
rnight be looking too hard or not noticing something critical' All
of these methodological problems are inevitable in the study of a
living phenomenon like language in its social use.
Given these concerns, many studies make use of multiple data
collected in a number of different ways. If the observed encounter
cnces
shows the same pattern as the carefully analysed sample of

speech, and if the attirudinal questionnaire is consistent with the
statistical observations, a sociolinguist can be more confident that

the account of the underlying structure is valid.
The sociolinguist at work

In practice, a number of workable solutions have been found to
the 'observer's paradox' of collecting natural speech samples. To
analyse variation in pronunciation, sociolinguists studying

microlinguistic variation regularly count the frequency of occurrence of a linguistic feature under defined social conditions. For
finer differences, they may make precise instrumental measure-
ments of significant speech sounds. In order to obtain the samples

for this analysis, natural speech must be recorded on tape. This
produces the methodological problem. 'V7on't speakers become
self-conscious, and try to make their speech clearer or more standard in the presence of a tape-recorder? Are the data collected in
this way natural?
Clandestine recording has been tried and largely abandoned.
There are practical reasons, for the tapes are usually noisy and
require very expensive processing. There are also ethical ones.

Fifty years ago, a study of spoken English could be based on surreptitiously recorded telephone conversations passing through a
university switchboard. In a climate of greater respect for the
privacy of their sollrces, sociolinguists now ask their subiects for
permission to use the tapes that they have made. Tape-recorders
are now in the open, and researchers are generally satisfied that
any initial anxiety effecting the level of formality will disappear as
an intervrew contlnues.
The sociolinguistic interview, modelled on the format developed
by \X/illiam Labov for his now classic doctoral study of Ncw York
City English, is one of the most common techniques for g:rthering
samples of language. In the interview, the sociolinguist talks to the
subject, attempting to elicit examples of various kinds of speech.

The normal stylistic level for an interview like this is fairly formal,
for the two people speaking are strangers. There are ways to modify the level of formality. The easiest alteration is in the clirection
of more careful attention to speech, by giving the person being
IO
I'RVF,Y
or a list of words to read. It is more difficult

ro relax the formality. occasional bursts of informal speech occur
when a speaker switches to a topic outside the interview, such as
when a mother being interviewed by a sociolinguist speaks to her
child who has come into the room. Another circumstance is when
the speaker is emotionally involved in an event he or she is narratlng. Taking advantage of this, an interviewer might ask a question
occasio.t when the speaker's life was in danger and so
,,b-oot
interviewed
a passage
"n
obtain less self-conscious speech.

Interviews provide a good deal of data, and are therefore
invaluable in studying in depth the language variation of the subjects. By selecting a sample of subjects carefully, the researcher
can make sure that significant social types (for example, by gender, age, education, occupation) are represented' One way to do
this is to draw the subjects from a sample previously studied in a
sociological survey. But interviews are expensive in time and
cffort, so that the number of different people studied is limited'
For studying larger populations, one technique is the covert col-
lection of non-intrusive esponses. Labov used this technique when

he asked salespeople in three New York department stores a question to which the answer was 'Fourth floor'. This utterance gave
him two post-vocalic 'r's and a 'th'to record. By asking for a repetition, he obtained a second set of data, this time with added stress.
A related approach has been tried using telephone calls made in
different langrrages in a multilingual city, to find out whether there
was prejudice against speakers of one of the languages'
O... natural speech sample has been collected, it must be
"
analysed. usualln the linguist chooses a variable, a specific feature
that previous observation suggests is likely to prove of social significance. In New York English, the pronunciation of kl aftet a
vowel is such a feature, and in Palestinian vernacular Arabic it is
the use of /tJ/ rather than /k/ in the word for dog, kalb or chalb'
Each occasion where the feature could occur is counted, and the
percentage of each vaiant under specific conditions (for example,
ty sp.nk.. or by style) provides statistical data which can be compared with other factors.
Because the interviewer usually has some clear preconceptions
about what phenomena are likely to be interesting, a sociolinguis-
tic interview follows a prepared protocol. After an opening

I1
carry on a conversation?').
of objects. An interview provides extensive statistical data on a

large number of linguistic variations.
The approach in this book
For sociolinguists working in the ethnographic tradition, the

main technique is the recording (either at the same time with a
nate discussion of the field.
In Chapter z an approach to language study is introduced
derived from anthropology called the ethnography of speaking or
communication. This approach, which starts with language in

nse rather than with its abstract structure, will provide us with a
also described in that chapter. Social factors are turned to in
those features thar are to be analysed. one of the clifficulties with

direct questionnaires is that they focus attentio. on the issue so
Questionnaires are particularly useful
in
gathering demo_
graphic data ('How old are you?' .How

-"'y y... did yu spend
at school?' 'lWhat is your occupation?'). They are also ur.d to
IZ
SURVEY
rnultilingual societies, the forces leading to shifts in language use,

and the establishment of stable bilingualism or diglossia' In
Chapter 7 ways in which the knowledge gained by sociolinguistics is applied to social problems are discussed, such as language
policy and planning and language education.
The ethnography of speaking and

the structure of conversation
scntences uttered by the speaker) a register of a language presumrrbly understood by the congregation, and the topic some appro-
Sociolinguists believe that the study of lang
the sentences that are the principal focus of
retical linguistics. It must go beyond langua

context. It must deal with the 'real' texts that make up human
communication and the social situations in which they are used.
The focus of attention shifts from the sentence to the act of com_
munication, the speech event.
a
Jakobson.
use of language o
factors, each asso
Building on
Roman
proposed by
ommunicarive
seven distinct
The first two
are the speaker-writer and the hearer-reader; the third is the

message form passed between them which is drawn from the
resources of a speech code. The message form expresses a topic,
hich is passed by some physical

in the case of writing or sign
of speech. The speech event is
It is possible to study rhese one ar

a time, but each must be included to understand the working of
the system. In particular cases, one or more of these factors ."r. b.
emphasized. The emotive or expressive function stresses the

speaker-writer's attitude. The referential or denotative function
stresses the topic. Some messages just check that the char-rnel is
open ('Can you hear me?') or assure that it is (,Uh huh').
_ Dell Hymes proposed that this model should provide the

basis for an ethnography of speaking (sometimes also callecl an
r4
SURVEY
t'rhnography of communication), which is an approach to the,

,l...ri!tin of speech events that ca1ls for an analysis of each of
thc relevant factors. Each of them may be studied independently,
brrt all are closely interrelated in forming the structure of the
wl.role event. For each genre or kind of speech event, the factors
,r re realized and related in appropnate ways'
Consider for instance how we might describe a sermon' The
a group of people conr y pical speaker is a clergyman, the listeners
a church or-sllasetting
the
siit.tth.g religious congregation,
"
amplified by
or
voice
voice
direct
g,rg.r. J. -orqo., the channel
by radio or
transmitted
version,
i,,.idrp."k.. (or in a modified
words, and
souncls,
(i.e.
actual
the
form
rcleviiion), the message
p ri
ate religious content.
Underlying the event is a complex set of socially recognized
l'r-rles,
whichian
be most easily recognized by considering possi-
ble breaches of them. Imagine for instance a preacher who stands
there and says nothing, or reads out an account of a football

rnatch. Or think what might happen when the preacher uses for
his sermon a liturgical language like Latin or Coptic or Sanskrit
that is still used for prayer but that the congregation may no
longer understand. Oia congregation that sings a hymn while the
precher is speaking, or goes to sleep. Or when a 1ecturer Starts to
give a sermon on morality in a physics class' These breaches of
ilo.*ul conditions for a sermon suggest how one might build a set
of rules to describe a typical sermon as a speech event' One might
do the same for other recognizable speech events, like a chemistry
lecture, or a dinner-party conversation' or a bargaining session,
or a lovers' quarrel.
The majoi value of the ethnography of speaking to sociolinguistics *s i., setting up an approach to language that went far
"b.yo.rd
the attempt to account for single written or spoken senIt
,..r..r. widened the scope to include all aspects of the speech
event. This proves invaluable in considering the structure of one
of the commonest of speech events' the conversation' when two
or more people speak to each other.
THE ETHNOGRAPHY OF SPEAKING
r5
The structure of conversations
Linguists have as a general rule focused on the smaller units o

language like sounds, words, and sentences, leaving it to other
scholars to examine the larger units that make up ,p"...h events.
In the case of the written language, the study of i.*i, and genres
has been the task ofthe literary scholar. There is also a crosover
field of stylistics and poetics where occasionalry linguists and literary scholars study the same objects, and more occasionallv talk
to each other about their different views. oral literary texts .\ /ere
also studied by folklorists and lirerary scholars, b,.rt ih. study of
natural chunks of spoken language was generally ignored by lin_
guists until ethnographers' socio1ogists' and sociolinguists strted
to explore its structure.
Because so much earlier linguistic analysis was based on the
written language, it is understandable that the sentence should

have been considered as the important unit to study. But sentences
are less useful in the study of speech, for if you look at the
verba_
tim transcription of a normal conversation. you will see how few
sentences are finished. More usefully, the conversational
inter_
change is the basic unit of the spoken language. Its structure was
first teased out in some innovative studies of ielephone conversa-
tions, where it was shown that a normal telephone conversation
has a number of distinct parts:
IJtterance
Comment
Caller
(dials; phone rings)
Other
This is the summons
Hello?
Ansuer
Caller
Hello, this
is
Joe. Is that Bill?
Other
Identification
Identity stagc
Caller
The meeting is still on?
Message
Other
Yes.
Acknowledgcnrcnt
Caller
OK. Bye.
Caller
16 suRVEy
I'11 see
you there.
Closc
Hangs ult
low do we know that telephone conversations are rule-
li,,verned behaviour? One quick way is to imagine what is likely

to happen if someone doesn't follow one of the rules we propose.
ll when you make a telephone call, you hear the receiver being
liltcd but no one speaks, the conversation is usually stuck. You
will usually issue another summons, saying something like

'l lcllo! Are you there?'If, on the other hand, like many children
wlro have not learned the rules yet, you start speaking as soon as
tlre receiver is lifted, not waiting for the answerer to say something, there is a moment of confusion. If you call a number and
hcar a voice saying 'I'm busy at the moment. Please call back' or
'l'rn not here at the moment. Please leave a message after the
beep', you assume you are talking to a machine and behave
:rccordingly. To hang up without a formal close is considered

rrbrupt and insulting behaviour.
There is a great deal of culturally and socially determined variation in the possible choices within the pattern set out here. In
lrngland, people commonly answer the phone by reciting their
telephone number. Telephone operators in offices are trained to
answer by identifying their employer: 'English Department. Good
Lnorning!' Intercom calls are answered with 'Yes' rather than
'Hello'. Asking'Is X there?'is interpreted as asking to speak to X'
There are national differences in these rules. In some countries, it
is considered impolite to ask to speak to someone else before initiating a series of polite social interchanges with the person answering. The development of answering machines and of voice-mail is
adding new structures to the rules, setting new challenges for the
novice or the conservative.
The important notion from our point of view is that there

is a formal structure to conversations, in part determined by
the nature of the event (until the answerer says something, the
caller has no one to talk to), and in part determined by social
rules (what it is appropriate to say to specific people in defined
circumstances).
This quick analysis of the telephone conversation demonstrates
the existence of socially structured rules for conversational interchanges. There have been studies of various aspects of cotlversation, such as the nature of service encounters (stlch as between a
customer and a selier), the rules for turn-taking and illterrupti()r1'
rHE ErHNocnnpwd6Ti:',
the organization of invitations, and the normal patterns of social

intercourse in casual conversarions.
Service encounters also have a fairly straightforward underlying structure. The first element is similar to the bell ringing in a
telephone conversation: engaging the attention of the person

meant to give the service. The task here is to establish the channel
between speaker and hearer. This varies according to the social
situation and cultural pattern. In some cultures and situations, it
leads to a preliminary social exchange; in others, it involves
simply catching the eye of the clerk or ticket-seller and making a
gesture without saying anything. Shop-keepers in the Middle East
remark on the brusqueness of tourists who start their conversation by pointing to something and saying 'How much?' Their
normal conversational interchanges starts with a formal set of
greeting, including enquiries after the health of the parties and
their presumed families, comments on the weather or some other
neutral subject, and finally a mention that there is some commercial purpose to the event. In other words, they require that the
topic be broached only after a social exchange.
One kind of service encounter that has been highly developed
in some societies is the process of bargaining, where the two parties seek to arrive at a price that satisfies both. In Middle Easrern
usage, this often involves appeals to a neutral bystandeq who initially is expected to agree with the buyer that the first price is too
high, but is ultimately expected to confirm that the seller's last
price is a fair one. This kind of negotiation is even more highly
developed in industrial bargaining and in diplomatic exchanges.
Turn-taking, the question of who speaks, is one of the most
intriguing aspects of conversational interchange. The physical
constraint is obvious. If two people are speaking at once, they and
others find it difficult to understand everything said. In various
formal situations, there are clear rules on the order of speaking. In
a classroom, teachers claim the right to control turn-taking. The
teacher speaks more or less when he or she wants, and grants per-
mission to students to talk. In a parliament or other public meeting, a chairperson is given the authority to determine who can
speak and for how long. In trials, there are clear rules on who
speaks first, who has the last word, who may ask questions, and
who must answer them. Lay witnesses are often confused and
r8
SURVEY
usually at a disadvantage in their lack of understanding of these

rules.
In informal conversations and informal meetings' the issue of
turn-taking is often quite complex, depending on power and

'!7ho
has the floor (the right to talk at any given moment)
status.
varies according to rules of the social group. Once someone has

the floor, it is possible to try to interrupt, but a speaker can ignore
this. Silence sometimes leaves the floor open, but there are tunholders-ways of signalling that the speaker intends to continue
after a break-like 'umm ...' or avoidance of a final intonation
pattern.
Much of the study of discourse has been carried out with a

focus on the meaning or topic rather than the other more social
factors, and is therefore better considered under the rubric of
pragmatics. Flere, our focus is the social, and we will analyse in
some detail two matters concerning the influence of social aspects
of the relation of the speaker-writer and the listener-reader in the
limitations set in choice of the message form. First we ask, what is
politeness and how does it control speech? Second, we will look at
the socially controlled choice of forms involved in selecting an
appropriate term with which to address the person to whom you
are talking.
Politeness and politeness formulas

Because speech events regularly include both a speaker-writer and
a listener-reader, it is not surprising that language is particularly
sensitive, in the rules for speech use, to the relations between the
two parties. Just as a good actor can utter a single sentence
expressing a wide range of emotional states of the speaker, so the
choice of an appropriate message form can be modified to express
a wide range of attitudes of the speaker to the listener. Given the
same general situation, I can pass information or make a request
or simply greet in a whole set of different ways that will define my
attitude to the listener and the importance I give him or her.
In its simplest terms, politeness consists of this recognition of
the listener and his or her rights in the situation. Requests, which
are an imposition on the listener, are mitigated by being made
indirectly, as questions ('Could you possibly pass me the salt?') or
T9
as statements ('I think that is the salt beside your plate'), or by

adding formulas like'Please'and'if you would be so kind'. Social
relations are eased by complimenting ('I do like your new car!' or
'Congratulations'). In some languages, there are elaborated sets
of politeness formulas, like in Arabic saying mabruk to someone
who has just bought something new, or na?imanto someone who
has just had a haircut or a bath or a short nap. For each formula,
there is an appropriate repln 'allah ybarik fik (may God bless
you) to the first and'allab yn?am ?alek (may God refresh you) to
the second. In American English, the equivalent is saying ,you're
welcorne'in reply to 'Thank you'.
The most common kinds of politeness formulas are involved
with greetings. Greetings are rhe basic oil of social relations. To
fail to greet someone who expects to be greeted signals either
some unusual distraction or a desire to insult the person. Each
social group has its own set of rules about who should be greeted,
who should greet first, and what is an appropriate form of
greetrng.
English greetings range from an informal 'Hi!' through a

neutral'Good morning'to a slowly disappearing formal .How do
you do!' It is common to add a second part of the greeting, a
purely phatic 'How are you!'to which no reply is expected. Arab
greetings Lrse an elaborate set of paired greetings-plus-responses,

depending on time of day or orher social aspects of the situation.
The study of greetings, therefore, provides a first useful method
of exploring the structure of a social group. A second area showing patterned variation in speech and similarly studied within the
ethnography of communication has been the conditions on the
use of terms of address.
icrman) was used to address someone of higher status' who

return the T form (tz in French, duin German). Thus, a
scrvant would use V to a master who would reply with T. In
rrrc'dieval French, this moved through the whole chain of being,
lvith God speaking to the angels as T, or men using T for animals.
V could also be used between equals. There developed over time a
tcrrdency to switch from V-V use to T-T use, as a sign of intimacy.
l'he French verb tutoyer, to use T to someone, refers to this
tlrange of familiarity. At one stage, in German student society
llcfore the First 'Wrlrld \ar, the switch was accompanied by a
lormal cerernony. In early twentieth-century French society, two
males who had served in the army together would use T to
'rrlr-rlt
t'rrch other. While the pattern has been relaxed, it remains impolirc to use T to a stranger. There have been complaints that stutlcnts in Canadian-French immersion programmes are sometimes
only with the T form, and so have not learned the
'rrldressed
rr ppropriate use of the V form with adult strangers.
\X/ith the growing egalitarianism of modern life, there has been
rr slow breakdown in the formality of address systems. The ideoIogically based switch from V to T associated with communism
has been documented in Russian novels of the period. French chilrlre n return T to their parents. Many speakers of Swedish now use
(
wrrr-rld
'l even to strangers.

A related phenomenon in languages that do not have the T/V
clistinction is the use of address terms. English once had the
th<>ulyou distinction and still offers a range of address terms'
rarrging from Title Alone (Sir, Your Majesty, Madam, Constable)
rhrough Title + Last Name (Mr Jones, Dr Smith, Lord Clark, Miss
ones, Mrs Jones, perhaps Ms Jones) to First Name to Multiple
f
Names (including Nicknames). The conditions for choosing vary
The choice of second-person pronoun and the related phenomenon of terms of address in \ /estern European languages in particu-
socially. Increasingly, in North American and British academic

circles, people who have just been introduced as 'Professor X,
rneet Dr Y' move immediately to first names. There are still intercsting cases of uneven usage. American doctors and dentists use
irst names to all their patients, but expect Dr X in retufn.
lcachers in many societies receive Title or Title + Last Name, but
the plural uous.
return first name (or in some schools, last name).

ln Arabic, there is an elaborate set of address patterns. One
interesting feature is the custom of addressing friends and
Terms of address
lar shows the formalization of politeness and sratus in a language.

A number of languages offer the same sort of choice as French of
addressing a single person using either the singular pronoun tu or
In earlier usage, the plural V form (uous rn French, Sie in
SURVEY
THE ETIINOGRAPHY OF SPEAKING
ZI
acquaintances by the name of their eldest son: a woman is

addressed as um Ahmed, mother of Ahmed, and a man as abu
Ahmed, father of Ahmed. The custom is even extended to people
without children.
Title plus first or last name is a common pattern in many languages. Non-relatives may also be addresseJ with terms oi ,.lutionships, as in an English pattern of training children to address
adults of the parent's generarion as Uncle John or Auntie Mary. In
Tongan, on the other hand, while there is a complex hierarchical
system with almost feudal ranks, people are never addressed by
kinship or other titles, but only by rheir name.
Military usage relared to address sysrems shows special patterns. Peace-time armies with strict discipline and emphasis on
ceremonial are likely ro have strict rules for addressing superiors.
In the US Marine Corps, senior officers were addressed in the
third person ('!ould the General like me to bring him a cup of
coffee?') and other officers received 'sir' from their inferi,ors.
with which sociolinguistics is concerned, the existence and nature
,rf regularly patterned variation in language and the role of social
iirctors in accounting for this regularity of pattern. In the ethnography of speaking, the setting is usually defined socially. But setting can also be defined in geographical terms, and can also be
placed in terms of the patterning of social class.
Non-commissioned officers were addressed by rank (.yes,

sergeant.') In a different setting, such as under battle conditions,
things changed. An officer was addressed directly, often by a

regular nickname. Company commanders, for instance, were
addressed as 'skipper' and sergeant-majors as ,Gunny'. More
democratic armies often make
point of dropping special address
rules along with saluting. These changes parallel the changes
from V to T under similar circumstances.

The ethnography of speaking moved the focus of analysis from
the sentence to the speech event, and offered a first approach to
the analysis of natural speech, by showing patterns that could be
understood if social information were included. In this, as in any
other study of language in use, the aspect that became more
obvious to the sociolinguist was the existence of regularly
patterned variation. It provided a wider canvas on which to paint
the complexity of language behaviour in its social setting, nd a
technique for capturing some of the ways in which each may
reflect the other. It opened up rhe way to the study of language in
use, to the importance of different channels, to the critical importance of relations between speaker and hearer, and to the social
context of language.
Against this wider background, we return to the central issue
zz
SURVEY
jrrst a
single language but a repertoire of languages or varieties. For
interlocking
nctwork of communication whose members share knowledge
the sociolinguist, the speech community is a complex
Loc ating variation in speech
;rbout and attitudes towards the language use patterns of others
well as themselves. There is no theoretical limitation on the

location and size of a speech community, which is in practice
,lcfined by its sharing a set of language varieties (its repertoire)
'rs
norms for using them.

'fhe idea that the members of a speech community share norms
rrbout the selection of varieties is important. Though they might
rrot all know and use each of the varieties' they recognize the contlitions under which other members of the community believe
tlrrt it is appropriate to use each of them. Londoners recognize
( )ockney and Mayfair varieties of English though they may themsclves use neither. A small social network (such as regular patrons
of a coffee shop) forms a speech communitn and so does a large
rlrctropolis or a country, a region, or a communication network
the goal of sociolinguistic study of
i lil<e the Internet). In each case'
srrch a community is to relate the significant language varieties to
tlrc significant social groups and situations.
ln small Israeli Palestinian villages, there are commonly four
significant varieties making up the linguistic repertoire: the village
vcrnacular (a dialectal variant of Palestinian Arabic), Classical or
Modern Standard Arabic (the fus-Ha taught in school and used
ior writing and public speaking), Modern Israeli Hebrew (learned
by those who worked outside the village or went to high school),
:rncl a set of
Speech communities and repertoires

The non-hyphenated fields of linguistics like phonology, semantics, and syntax focus on the language system ideally abstracted
from all social context. Psycholinguistics deals with the individr-ral speaker's acquisition and use of language, and relates this to
mental processes. Sociolinguistics is concerned with Ianguage in
situ and in uiuo, alive in its geographical and social setting and
space. ril/hat this space is like, we now consider. To start with,
because our main interest is in social matters, we will deal with
social space, and look for the location of varieties of speech
wirhin definahle social unirs.
Some of the units with which we are concerned are already
familiar and established social groupings. Thus, we can study the
language
of families, neighbourhoods, villages, cities,

T
x::t
Jn :n i*li;::
stares,
:;*:ir;l:
a speech community.
For general linguistics, a speech communitlr is all the people who
speak a single language (like English or French or Amharic) and so
share notions of what is same or: different in phonology or grammar. This would include any group of people, wherever they might
be, and however remote rnight be the possibility of their ever want-
ing or being able to communicate with each other, all using the
same language. The notion is preserved in such a concept as .la
francophonie', the French-speaking world, and it can serve as the

basic slogan for political co-operation. Underlying it is the idea of a
group of people who could, if they wanted, speak to each other.

Sociolinguists, however, find it generally more fruitful to focus
on the language practices of a group of people who do in fact have
the opportunity to interact and who, it often turns out, share not
z4
SURVEY
'.'rdsomeschool-rearned'ixllil;lT;1ffi :"ti1iT:;1;T:;*::1
nal or religious or formal public functions;

e outside the village; and English for educa-
tional use.
In a city, the pattern is likely to be even more complex. Inside

rhe walls of the Old City of Jerusalem, there are thirty or more
l:rnguages used by different residents in different settings. In
Hong ong, while most local Chinese use Cantonese most of the
time, many have learned and use Putonghoa (Standard
Mandarin), whose status changed with the reversion to Chinese
sovereignty, and all who go to school learn and (less often) use
lrnglish. In Toronto, Melbourne, New York, or London, English
LOCATING VARIATION IN SPEECH
25
is the common
langu
larly found sharing

immigrant language
clear division betwee
dominant.
u_
of
a
re
The speech repertoire may be divided functiglally as well as

spatially. On the Naval'o Reservation in the'\)Tesiern United
States, most people spoe Navajo up to a few years ago, but most
uriting was done in English. The rnany FM radio stations
communications media will involve different stylistic choices, and

the
the variants can be considered as making up the repertoire. In
s.rme way, international organizations might, like the European
Oommunity, have a formal policy governing choice of language
lrom the repertoire available'
The speech community is, therefore, the abstract'space' studied in sociolinguistics, the location in which the patterned variations in selection from the available repertoire takes place'
Dialect
to the issue of variation
ll before sociolinguistics
ts of language gave seri-
loca
dialects
ished in
iuterest
rhe
have been srarting to speak English.
was
the
the
(ove
age that correlated with
nal
ur-
the
the
any
at
system
concern for synchronic description of a language
half, something added to the older North-south diarectal variations. New York or London studied as speech communities show
not just regional variation but also social variation.
Smaller
othe
might, for
the comm
each
with
One
ke up
.n..L
phone ifyou have a Black or Southern accent)'
z6
SURVEY
LOCATING VARIATION IN SPEECH
The earliest scholars concerned with regional dialectal

varia_
tion were the philologists who set our first to explain
differences
ding
re the manuscripr
r
socia
u
se.{
the
;';":l.,.T:':,'.':T"
ect variation. The

fil.: i: that all languages change over rrme, as new words
are
added to deal with new concepts or as contact
with other lan_
guages anci 'phonetic drift' leads to modifications
in phonology.
The second is that people who communicare with
each oth.. t.ii
P
from
'
;1:;
aphical boundaries from other
placewiuchange.rheronge-iB:1r?::J?:ff r"i:',i:";i::
their varieties will have changed. ltrith'th. breakdown

of isolation in the modern world, as roads are built and
as radio
and television enter more and more homes, dialectal

variation
tends to diminish and languages become more
and more
homogenized.
Dialectology is the search for spatially and geographically

deter_
mined differences in various asp..t, oi larrgr',"g.. po.
.".h viliage
or region that they study, dialectologists *"'',
know the typit
local vocabulary or pronunciatic.,. A, result, th.l.
.uby..i, oi
" lived
choice are usually older people who have
all their iives in
one location and who have had a minimum of education.
Once
found, they are qtizzedby the fieldworker for names
of ob;.ects or
pronunciations of words or strange expresslons.
In the popular wisdom, there is recognition
of the ,broad /a,/, of
_
the Yorkshireman, the glottal ,top
o"f
the Cockney, the Texas
drawl, the /r/-less dialect of upp.r-.in.s Boston. These

are ste eo_
types, fixed and prejudicial paterns of thought
p."pl.1t'",
"t "",
may be mistaken, but they focus on the most obvious
feature of
the local accent. There are also obvious differences
in lexicon.
Peanuts may be called groundnuts or goobers or pindersin
differ_
ent parts of America. There are place in the
USA where you buy
potatoes in a bag and they put your groceries in
n sack; others
where you find potatoe s in aiack and tk. you, grocer:ies
home irr
z8
SURVEY
.t
ag. \Vhere older dialect terms remain' you might hear tote o
::: fiTi',i:5::J;
both grease and greasY

ere pronounced with an lsl , a transitional zone where grease is
*ith un lsl and greasy with a lzl , and a southern where both are
lil . Diferences may be quite striking: Texas English, for instance,
lras one fewer vowel than general American usage' making no difierence between the vowel of pin and pen, so that a Texan is careiul to distinguish between a writing pen and a sticking pin'
In long-settled European countries, dialect atlases show the

cffects of earlier settlement patterns and of contact. one can trace
which areas were originally celtic, or see evidence of the limits of
a more recently settled country like the
the differences in original settlement on
owing from what part of England the settlers came and what other linguistic groups they were later mixed
with. The US atlases also permit plotting the \estern movement
of pioneers from the Eastern seaboard along the different pioneering trails.
grist for the socioGeographical differ
as the
rJomplex
li,'g,.ris mill, but the
ofthe
study
recent
i,ruence ofother fact
distance
that
indicates
instance,
for
Mexican-American border,
from the border is indeed one of the explanations of Spanish language maintenance among people who have crossed into the
"ited States, but that it needs to be set against other sociological
factors such as education and mobility. Geographical space, in
other words, is not enough to account for language variation'
This becomes clear if we look at the regular discussion of the
clifference berween a language and a dialect. From a linguistic
point of view, regional dialects tend to show minor differences
?rom their immediate neighbours, and greater differences from
clistant varieties. Thus, one can demonstrate the existence of a
chain of dialects from Paris to Rome. At the Franco-Italian border, however, although there is no linguistic break in the chain, the
poiitiral distinction is enough to make it clear that one has moved
from dialects of French to dialects of Italian.
LOCATINC VARIATION lN sPEtcH
z9
Styles, gender, and social class
Styles
s a good beginning when we want to explain
Dialectology is able to account for many of

otherwise play havoc with those who seek a
pure, unified language with a single t of correct forms. The differences between diued and doue, between footpatb and pauement,between /b,rte/, /F.t'?el , and /b'rde/ set difficult quandaries
for someone trying to describe theEngIshlanguage. Being able to
add regional labels to variations helps a great deal. Thus' dictionaries can label forms as British, American, or Australian, implying the existence of unmarked correct forms.
But even if this is accepted, there remains the issue of variations
within individual speakers who come from a single location'
Speakers of English sometimes use 'don't' and sometimes use 'do
not'. Some Londoners sometimes say lbntel and at other times say
lbt?el .If you carefully record anyone speaking, you will find that
there is still patterned variation in the pronunciation of a single
phoneme, in the choice of words, and in grammar.
A first useful explanation is provided by the notion of style and
the related dimension of formality. At times, we are more careful,
and at times we are more relaxed in our speech or writing, just as at
times we are more careful or more relaxed in other kinds of behav-
iour, like how we dress or eat. This varying level of attention to variety forms a natural continuum' the various levels of which can be
divided up in different ways. Each language has its own way of
doing this: some, like Javanese or Japanese, have a finely graded set
of levels, marked specifically in morphological and lexical choice'
30
SURVEY
STYLES, GENDER, AND SOCIAL CLASS
3T
How many distinct points there are on what is really
a contin_
uum is not important, but most accounts of language (such as

those in good dictionaries and complete grammars) now make
some reference to levels of stylistic variation. The cautious writer

or speaker is warned in this way how others might react to possi_
ble choices, iust as etiquette books advise ,eaders how to avoid

embarrassment in social settings.
In the sociolinguistic inrerviews that Labov conducted in the

New York City study referred to earlier (see Chapter r ), he found
evidence of the informal style (the vernacular he was most interested in) being
rupted to speak
cup of coffee to
he or she was t
a
obta
an emotionally
formal
use by
passage or read a list of

asked the subject to tell
e him three or four lev_
els, an d the possibility of comparing changes in cerrain features at
each of them.
In bilingual communities, these stylistic levels may be marked
by switching from one variety ro anorher. officials in switzerland
words. To
who use Swiss German in intimate and casual circumstances

move to High German for inforrnal and formal speech.
Paraguayan city-dwellers switch to Guarani for casual and inti_

r.rse the vernacular
in norrnal conversation shift to Modern standard Arabic when
they are giving public speeches.
The commonly accepted explanation for this stylistic variation
is,the care that speakers and writers take with their expression.
The more formal the situation, this explanation goes, ih.
-u..
mate speech and jokes. speakers of Arabic who
Attention or care is a good explanation as f:rr as it goes, but it

leaves open the question of where the norms c.rm. from, and it
does not deal with the possibility of conscio's ch.ice.r:r ress or
more formal style. one explanation for these cases is thc idea of
SURVF,Y
audience design. A speaker who can control more than one variety
chooses a level of speech according to the audience he or she is
'V7e
might consciously choose an informal style when

speaking to strangers in order to seem friendly. Related to this is
runconscious accommodation; we automatically adjust our speech
to be more like that of our interlocutor' Both of these approache s
offer some idea of the importance of language in establishing
social relations and in representing a speaker's sense of identity, a
topic we will explore later in more detail.
It should be noted that this recognition of stylistic levels as
being appropriate to specific social situations is in opposition to
normativism, the approach taken by purists who clairn that there is
oue 'correct' version and that all variation is incorrect and
bad. When'il/ebster's Dictionary in its fourth edition introduced
stylistic labelling and listed such informal usages as 'ain't', there
were many who criticized its admitting the barbarians into the
qates of pure English.
rrddressing.
Specialized varieties or registers and domains

Dialect concerns variations that are located regionally or socially.
Style refers to differences in degree of formality. A third set of
variations concerns the special variety (or register) especially
rnarked by a special set of vocabulary (technical terminology)
associated with a profession or occupation or other defined social
group and forming part of its ia gon or in-group variety. People
who work at a particular trade or occupation develop new terms
for new concepts. Phrases lke hacking an surfing the nethave no
obvious meaning to those who are not keeping up with the computer revolution. Terms llke a sticky wicket and hit for a six are
understood best by people with some experience of cricket.
A specialized jargon serves not iust to label new and needed
concepts, but to establish bonds between members of the ingroup and enforce boundaries for outsiders. Ifyou cannot understand my jargon, you don't belong to my group. (You might
have noticed how in this series of books, the writers are careful to
identify new terms by putting them in bold, and to explain them in
a glossarn all to make it easier for the novice reader to ioin the
group of experts.)
33
Australian aboriginal secret societies developed their own

special forms of language. Thieves and underworld jargons
(sometimes called cant) are another example. The goal of these
was often to make it hard for the outsider to understand con-
versations. This is not limited to rhe underworld. In Alsace, where

the fact that so many people knew German made yiddish more
wi
ha
of
were reported to
numbers and parts

the course of time,
these terms might get known by all professionals in the field, and
form part ofthe register ofhorse-traders. Gangs and other closed
peer groups often develop their own forms of jargon to serve as
markers of group membership and also to make their speech less
intelligible to outsiders.
Dialects, styles, and registers as we have presented them are
ways of labelling varieties of language. The starting point of our
classification is the linguistic variation, which we attempt to
explain by associating it with a specific set of social features. !7e
might choose to work in the reverse direction, by classifying

social situations, and then naming the variety that is suitable for
it. A register is a variety of language most likely to be used in a
specific situation and with particular roles and statuses involved.

Examples might be a toast at a wedding, sports broadcast, or talking to a baby. A register is marked by choices ofvocabulary and of
other aspects of style.
A useful way of classifying social situations is to analyse them
into three defining characteristics: place, role-relationship and

topic. Together, these make up a set of typical domains. One common domain s home. Domains are named usually for a place or
an activity in it. Home, then, is the place. The role-relationships
associated with home (the people likely to be involved in speech
events) include family members (mothe5 father, son, daughter,
a suitable set of topics
X'il','iiT'::f
domain. In
;'"1T:Y:
multilin-
gual community, different languages may well be considered
appropriate for different domains. In a multilingual family, different role-relationships might involve different language choice.
14
SURVEY
For instance, husband and wife might use one language to each
other, but father and children might use another.
Another common domain is work. The place might be aactory
<lr an office or a stoe. The role-relationships include boss,
worker, colleague, customer, foreman, client, to mention just a
few. The topics are work-related. Now we can understand some
of the sociolinguistic complexity that occurs when two people
who have one role-relationship at home (such as father and son)
have another at work (boss and worker, for instance). !7hen they
speak, they can choose a register or language variety to show
which relationship is domin'ant at the time.
Slang and solidarity

The importance of language in establishing social identity is also
shown in the case of slang. One way to characterize slang is as special kinds of intimate' or in-group speech. Siang is a kind of iargon
rnarked by its rejection of formal rules, its comparative freshness
and its common ephemerality, and its marked use to claim solidarity.
Solidarity, or common group membership, is an important

sociai force that has a major impact on language. The solidarity
relations (the claims that we belong to the same group) underlie

rhe notion of accommodation mentioned above. $7hen we are
ralking to someone, most of us unconsciously move our speech
closer to theirs (which explains why our accents change after we
have lived in a new place for a long time). Simiiariy, by choosing
the form of language associated with a specific group, we are
making a claim to be counted as a rnember of that group.
This contrasts with the power relation, in which a person's
speech carries a claim to be more or less powerful than the other.
Slang is primariiy speech claiming group membership, and it
rejects the power dimensions associated with formal language"
Often, slang is associated with peer group and gang speech,
intentionally used to obtain some degree of secrecy. It may be
compared to the secret languages found in some tribes. In one
Australian aboriginal language, there is a men's society with a
secret language in which every word means its opposite. Pig Latin
is a children's secret language in which a meaningless vowel is
inserted after every syllable. Canay uyay unayderaystanday
35
tbisay? In southern Arzona, the Spanish-American young people

developed a secret variety called Pachwco in which they used
idioms translated literally from English to Spanish, which
couldn't be understood by either their Spanish-speaking elders or
their English-speaking fellow students. Cockney rhyming slang

(for example, 'titfer' for 'hat,' abbreviated from 'tit for tat') has
also been widely publicized.

Slang regularly transgresses other social norms, making free
use of taboo expressions. The use of words like 'fuck' and 'shit' in
public media has become a mark of liberation or a sign of revolt,
depending on one's point of view. But slang also sets up its own
norms, the norms of the in-group, so that the gang is easily able to
recognize a lame or outsider, who does not understand or who
misuses slang terms. Slang thus serves social functions, setting
and proclaiming social boundaries and permitting speakers to

assert or claim membership of identity or solidarity groups. Slang
is a feature of the speech of the young and the powerless. Its
dynamic nature is partly an effect of the need to develop new ingroup terms when slang terms are adopted by other speakers.
Language and gender
All
these cases have started to show how language reflects,

records, and transmits social differences, so we should not be surprised to find reflexes of gender differences in language, for most
societies differentiate between men and women
marked ways.
in
various
Observations of the differences between the way males and
females speak were long restricted to grammatical features, such

as the differences between masculine and feminine morphology in
many languages. In earlier usage, the word Eender was generally
restricted to these grammatical distinctions. They cause problems
for speakers of languages like English, where grammatical gender
is marked mainly in pronouns, when they learn a language like
French, where non-sexed items like table (la table) can be gram-
matically feminine.
It was ethnographers who first drew attention to distinct
female and male varieties of language, often with clear differences
in vocabulary. The famous anthropologist Levi-Strauss noted
16
SURVEY
how an Amazonian father laughed at his young daughter for
word for 'hunting'. Other ethnographers have provided cases of marked differences in the language of men and
women. American servicemen in Japan who learned Japanese
from the women with whom they associated were thus a source of
amusement to people who knew the language.
Historically, these differences sometimes seem to have arisen
from customs encouraging marriage outside the community. If
rhere is a regular pattern of men from village marrying and
bringing home to their village women from village B, then it is
likely that the speech of women in village will be marked by
rnany features of the village B dialect. The preservation of these
introduced features depends on the maintenance of social differcntiation in occupations, status, and activities.
Children soon pick up the social stereotypes that underlie this
cliscrimination. They learn that women's talk is associated with the
home and domestic activities, while men's is associated with the
outside world and economic activities. These prejudices often
r,rsing the male
remain in place in the face of contrary evidence. Thus, while there is

,r popular prejudice that women talk more than men' empirical
studies of a number of social situations (such as committee meerings and Internet discussion groups) have shown the opposite to be
true.
There is some intriguingly suggestive evidence of differences in

neurophysiological process of aspects of language between males
and females. In a recent set of studies using functional magnetic
resonance imaging, phonological processing in males was shown
to be located in the left half of the brain and in females to involve
both left and right parts of the brain. No difference in efficiency
was shown, nor is there any evidence so far that any neurophysiological difference accounts for differences between male and
female language. The causes are social ather than biological.
Of the social causes of gender differentiation in speech style, one
of the most critical appears to be level of education. In all studies, it
has been shown that the greater the disparities between educational opportunities for boys anc! girls, the greater the differences
between male and female speech. This can be illustrated with
American ultra-orthodox Jewish communities. Males in these
communities are expected to spend longer studying traditional
sryl-Es, GENDER, AND SOCIAI- CLASS
37
Jewish subjects. Linguistically, this results in their srronger competence in Yiddish and Hebrew, and their weaker control of English.
Females on the other hand spend more time on secular studies.
\il/hile their Hebrew knowledge is
much less, rheir English is much
clos
of differences between the speech of
Ara
provide evidence that the major cause
of d
In one village, we found greater differences between male and female speech in the half where girls had
less education than boys than in the half where both boys and girls
had more or less equal opportunity for schooling.
\7hen offered an equal educational opportuniry, there seems
to be a tendency for women to be more sensitive than men to the

status norms of the language. The tendency has been noted jn
some cities for lower-class males to have much tighter social networks (their neighbours are male relatives, alongside whom they
work, and with whom they share leisure hours) and to find their
norms within the tight network. The women in these cities have
looser multipie nerworks; they mix more with people outside

their community, and so their speech is influenced by the social
norms of the wider society.
Studies of gender differences have shown the power of stereotyping. A poet s taken more seriously than a poetess; women's
status is lowered by references to rhe girls. In Hebrew, only the
lower ranks in the army (up to the rank of lieutenant) have feminine forms. The use of generic masculine ('Everyone should bring
is lunch, we need to hire the best man avaIable'), however wellmeaning and neutral the speaker's intention may be, reinforces
the secondary status of women in many social groups. With the
growth of social awareness in this area over the past decades,
there have been rnany attempts to overcome this prejudicial use of
language.
In contrast to the words of the popular saying that .Sticks and
stones may break my bones, but words will never hurt me', it has
been shown that anthropoeent ic speeoh lvhich assumes that men
are more important than
dices and actions that do
reflect and record current
ted, reinforcing the lower
in a society. Many publishers and journals now adhere to guide38
SURVEY
lines to avoid gender stereotyping and gender-preiudiced langllage use. Everyone should take care with their language.
Exploring the correlations between gender-related linguistic
clifferences and social differences between the genders is another
way to see how closely language and social variation are related.
lJut modern societies are divided in other ways too' one of the best
studied being social stratification or division into social classes.
Social stratification
\X/hile note had been taken earlier of the effect of social class on
speech, it was the work of !illiam Labov in New York that establi.shed social stratiication' the study of class distinction in speech,
ils a majo topic in sociolinguistics. Labov himself started out
with a purely linguistic question. He wanted to know how, in the

terms of the structural linguistics that was in vogue when he was a
graduate student, to set up a phonological analysis that included
features that were sometimes zero. \7hat were you to do, he
asked, in New York City, where speakers sometime pronounced
the lrl after a vowel (post-vocalic kl) and sometimes didn't? The
notion of free variation, the notion that the choice of variant was
uncontrolled and without significance, was widely used for such
cases, but it seemed an unsatisfactory dodging of the question.
He wondered next whether there was any scientifically observable explanation to the variation. In a clever pilot study (see
above, page r r ), he found that the shop staff (socioeconomically
similar in level, but finely varied by the differences in customers
and prices) showed regular and predictable variation' The percentage of r-coloration (any tendency to pronounce post-vocalic
lrl\, he found, correlated closely with the social level of the

customers of the store. In fact, in one store, he found a higher
percentage of use of the prestige feature among salespeople on the

higher, more expensive floors of the store.
In later studies, using extensive interviews with subjects
selected on the basis of their socioeconomic classification, the
relevance of sociolinguistic evidence to socioeconomic stratification was firmly established. In cities, variations in speech provide
clear evidence of social status.
There are historical explanations for social differentiation' The
39
coming of a group of Tewa speakers to rhe Hopi villages in

Arzona explains why the people in the village of Hano were
bilingual in Hopi and Tewa, but it was sociocultural and religious
differences that accounted for the maintenance of this cleavage
for two hundred years. There were similar reasons for the three
distinct dialects of Baghdad Arabic, one Christian, one Jewish,
and the third Moslem. The different religious groups lived in the
str:iking observation was a tendency in the upwardly mobile and

socially insecure lower-middle class to over-use (relative to the
normal slope) socially desirable features in very careful speech
!7hile historical differences may also be the original cause of

social differentiation in large cities (and this is certainly true now
as increasingly large groups of immigrants arrive in most cities),
there can develop socially marked stratification within a single
language. New York is the classic case. Leaving aside the special
minority groups (the Blacks and the Hispanics), New Yorkers
speak a kind of English that includes the same features, but with
certain crucial and socially relevant differences in their distribution. Certain salient phonological variables (such as the r-colouring or the pronunciation of [th] or the height of the vowel s in bad
or off ) vary in all speakers in various situations, with a more
standard or prestigious version appearing more often in more
formal speech. Thus, the pattern for lower middle-class speakers
in New York was to use the stigmatized /t/ or lt\l pronunciations
only occasionally in very careful speech reading word lists, to use
it about zo%" of the time in careful speech, and to use t 3o" of
percentage of r-coloration) with social or demographic factors

(gender, educational level, socioeconomic status). As far as it
goes, the explanatory power of these correlations appears good'
l)ut correlation and causation are not the same thing. \7e obtain
Each sociai level (as determined on the basis of income, occupation, and education) had a similar gradation according to style or
degree of formality. But there were also marked differences
between the social levels. In casual speech, for instance, the uppermiddle class would use a stigmatized form about roTo of the time,
the lower-middle class about zo", the working class about 8o7o,
and the lower class about 9oo ' Thus, the same feature differentiated the stylistic level and the social level.
In practice, these fairly fine differences, which affect only a
small part of speech and do not interfere with intelligibility, help
relatively unaffected by that of other groups. It was because the

religious groups in Baghdad had limited contact with each other
that the Christians, Moslems, and Jews there maintained distinctive dialects. It was the social distance between the castes in an
tndian village that led to differences in their speech.
An alternative suggestion is to consider the driving force as
audience design, a concept mentioned earlier on page 3 3. In this
view, the speaker, consciously or not' chooses a stylistic level
appropriate for the audience he or she wishes to address. The
notion comes from radio announcers' who suit their style to their
audience. The same announcer will be found to have distinct
styles when reading a news item on a national station and when
introducing a song on a popular music station. By selecting a style
appropriate to a particular audience, the announcer is identifying
same city while maintaining social and cultural isolation.
the time in casual speech.
New Yorkers to identify themselves and each other socially.
Sometimes they do this even more subtly and sensitively than do

more obvious socioeconomic markers like income and education.
There are social forces leading to or delaying the diffusion. One
4o
SURVEY
and reading. This hypercorrection suggests that the feature is

recognized as a stereotype rather than still serving as an unconscious social marker.
The analyses we have discussed to date of these variations
have depended on associating linguistic features (for example, the
ll lore powerful account of what is involved if we add social

psychological factors like attitude and accommodation, and consider them as causes.
Accommodation and audience design

l{ow is it that dialect differences and stylistic differences emerge?
The simplest solution is that people tend to talk like the people
rhey talk to most of the time. The physical isolation of villages
explains why their dialects are different from the dialects of

neighbouring villages, and even more different from those of
rnore distant villages and towns. Similarly, the social isolation
of specific groups explains why their languages or dialects remain
4T
himself or herself with the audience or claiming membership of

the group that it constitutes.
Adding this social dimension increases the explanatory power.
One speaks most like the people with whom one regularly associates, but one may also choose, in appropriate circumstances, to
allow one's speech to move in the direction of another group. As
we noted earlier, many speech sounds are not always pronounced
in the same way by the same speaker, but their realizations form
rather a pattern not unlike the patterns of bullet or arrow hits on a
target. I7hile there may be a rare bull's-eye, the shots as a whole
form a more or less consistent group. If one moves one's aim, the
whole group moves, with the centre changing.
In conversations between people with differing pronunciation,
it has been noticed that there is a common tendency for the pronunciation of the two ro move slightly closer together. This
process, called accommodation, explains the way that a person
who moves to a new part of the country gradually modifies his or
her speech in the direction of the new norm. Because we are talking about changes in probabilities and percenrages, the change
need not be immediately obvious to the speaker or the listener.
But if we record a conversation between two speakers of differing
varieties, we find that their percentage of use of some features
often converge. It is common to find that your speech-choice of
vocabulary, grammatical forms, and even pronunciation-moves
towards that of your interlocutor.
The opposite effect also occurs, when a speaker chooses not to
The same factor also accounts for the tendency to speak like
one's friends and peers, and to modify one's speech either in their
clcsire to move to the mainland.
The existence of variation in language' therefore, is not acci-
c{ynamism of change.
The possibility of using variation in language to identify group

rnembership can have harmful effects, when it is associated with
convege but to diverge, by moving his or her speech away from the
other party. Rather than converging, one may choose to stress fea-
tures that connect one not to the other person present, but to an
absent but valued hypothetical audience, such as a peer group or an
admired outsider. \7e have already mentioned the same phenomenon
in the use of non-standard slang for showing in-group membership.
This powerful sociolinguistic phenomenon would seem related
to the most fundamental linguistic features involved in social
bonding. Just as two speakers talking together tend to be moving

in the same rhythm, so they unconsciously adapt their speech to
accommodate to each other. It is this sympathetic movement and
its absence that enable a speaker easily to pick out which members of his or her audience are not listening.
4z
SURVEY
speaker. The more stratified a society' the more likely it is that

speaking a prestige variety will be rewarded, and that speaking a
non-standard variety will lead to preiudicial treatment.
'!hile it is possible, as we have Seen' to recognize factors like
these through the study of variation with a single language vari
etn they ni. .u.., more salient when two or more languages
a involved. In the next chapter we will look at bilinguals and
bilingualism.
43
realize which people can be addressed in a mixture of the two

languages. In this wan bilingual children can be said to develop
.o.r-,.o1 over three distinct varieties of language' The study of
bilingualism provides an excellent laboratory for learning how a
chilcan learn to be a member of two (or more) distinct societies.
The descriPtion of bilin$ualism

!hile it is the case that even speakers of a single language (putative monolinguals) control various styles and levels of that
Language socialization
Children acquire language and social skills together. Their sensitivity to the social uses of languages is already apparent in their
early learning of different varieries. Even while they are still in the
babbling stage' many children have a different way o addressing
small objects (animals, toys, other babies) from the way they
address adults. If they do this, they are showing that they have
learned that babies are talked to using a different variety. This
register that is used to speak to babies is called baby talk, and has
been shown to occur in many languages. From an early age,
children learn that there is more than one variety of language.
There are in fact a vast set of social rules about language that a
child must acquire to be successfully socialized. One is the rule for
conversational organization. Knowing when to speak and when
to be silent, how to enter a conversation, when to speak quietln
and when clearly, are ail part of the conversational rules that
children have to learn. Equally confusing at first are the pragmatic
ruies, such as comprehending that a question may be a request.
We may be frustrated when speai<ing to child on the phone. 'yes!'
he answers when we ask 'Is your mother there?', making no effort
to fetch her. Children have ro learn the social conventions for
language use. Learning these social conventions is a key component of socialization.
Cne of the most revealing opportunities for studying ianguage
socialization is in the case of children growing up bilingually, for
they manage not just to keep the two languages separate, but to
learn quickly which language to use to which person. They also
44
SURVEY
language, it is very common that people develop some knowledge

in a second language and so become bilingual' The
".r "bility
simplest definition of a bilingual is a person who has some functional ability in a second language. This may vary from a limited
rbility in one or more domains' to very strong command of both
lungo"g.s (which is sometimes called balanced bilingualism)' The
that there must be a single definition leads to confu"rr.r-ptio.t
sion, s-uch as when one person is talking about the highly skilled
multiple-domained balanced bilingualism of an expert translator
and interpreter, and the other the uneven skills of a recent immigrant. AJditional confusion is caused by the common use of the
term bilingual to refer to a socially disfavoured minority group: in
Texas, for instance, it is restricted to Mexican-Americans'
Rather than worrying about definition, it is more useful to

consider what is needed to describe rhe nature of an individual's
bilingualism. Clearly, the first (and not necessarily easy) element

is to]dentify each of the languages"!e will often need to clarify
which variety is involved: to distinguish between Cantonese and
Putonghoa, or between Egyptian and Moroccan Arabic, or
between High German and Swiss German.
A second important feature is the way each language was
acquired. It is useful to distinguish between mother (or native)
tongue learning, second (or informal) language learning, and
fore"ign (or addiiional) language learning. Each of these suggest dif-
fereni possible kinds of proficiency. It is useful also to note the age of

learning and the time spent using the language' S7e describe two
bilinguls in this way: 'X is a native speaker of Cantonese and

trearned English in school.' 'Y grew up speaking Moroccan Arabic,
but was edocated in French and has lived in Paris since the age of r 5 ''
BILINGUALS AND BILINGUALISM
45
with High German used in the work domain and Swiss German in
tlre home and neighbourhood. In his study of Puerto Ricans in
Amharic were more
1'
recep-
ronger
eaking
a lan-
tive skills of reading a

in a learned language
and writing. Many
guage at school, but cannot speak it.
In describing the bilingualism of an individual, anorher set of

differences is often evident in the performance of certain internal
functions. Bilinguals usually prefer one language for functions

such as counting, doing arithmetic, dreaming (some people
dream in language, others don't), cursing, or praying silently.

Another useful way to describe bilinguals is by describing the
external functions they can perform in each language. These
might be expressed as 'can-do' statements: X can read a daily
newspaper, can carry on an informal conversation, can give a lecture. One special ability (nor rrue in rhe case of all bilinguals) is
the skill of translation from one language to the other. Another
useful approach to describe
bilingual's language use
is
by
domains rather rhan by funcrions.

A domain, as we have discussed above on page 34,is an empirically determined cluster consisting of a location, a set of rolerelationships, and a set of topics. Just as this notion was useful for
identifying the use of registers, so it is useful for considering bilingualism. For each of the domains, a bilingual is likely to have a preferred language. Some examples of domains are shown in Table r.
5.
Bilinguals have a repertoire of domain-related rules of language choice. The home-school or the home-work switch is
probably the most common, with one language learned at home
from parents and the second learned at school and used at work.
\ hen there is a language shift in progress, certain traditional
domains may remain favoured for the use of one language. For the
Maori people, before rhe recent language revival activities began,
the marae where traditional ceremonies ar-rd meetings took place

remained the strongest bastion of Maori language use. The bilin-
gualism we menrioned earlier in Swiss adults is domain-related.

46
RVF,Y
New Jersey, Fishman noted strong Spanish maintenance in home,

rreighbourhood, and church, and strong English usage at school
,rnJ at work. It is normal for immigrants to continue to use their
original language in the home and in religious domains, while
rrsing the new language in work, education, and public domains'
l,ocation
Role-relationships ToPics
llome
Mother, father, son,
Domestic, personal, etc.
daughteq etc.
Neighbourhood Neighbour,shoPSchool
(
keePer, street-cleaner
social greetings
Teacher,student,
Socialgreetings,
educational
Priest, parishioner,
Sermons, prayers,
confession. sociai
principal
lhurch
etc.
'[-AB
LE
5.
\eather, shopping,
Because dornains are cornposite concepts' there is the possibil-
ity of conflict and therefore rnarked choice between languages'

Thus, two people who normally speak the standard language at
work mighl use their home language there to signal either a
change of role-relation (famiy.member or friend rather than co-
worklr) or topic (a home or neighbourhood topic) while still
shall take this up again later when

about switching' At this point, the important notion is
that a bilingual's use of his or her two languages is likeiy to vary
considerably according to domain.
being in the sorne location.
*. lk
BE
Ei
1 7e
riguiaE ortpe [el{.E
We have been describing so far the language use or performance o

a bilingual. The description makes clear why it is that it is rare to
find equal ability in both langr-rages. Assume a biiingual immigrant who grew up speaking l-anguage A, but was educated for-
47
mally in Language B. Such a person might well have all the
conversational skills in Language A, but be quite weak at dealing

with academic matters in it. Misunderstanding of this possible
difference in competence often leads to educational problems:
teachers might assume, for instance, that a child who has reasonable conversational ability also has the full basis for academic
work in the language.

The nature of bilingual competence is a topic of considerable
interest and importance for the psycholinguist as well as the socio-
linguist. How are the two languages organized in the bilingual
brain? For a number of years, there was an attempt to distinguish

between compound bilinguals whose two languages were assumed
to be closely connected, because one language had been learned
after (and so through) the other, and co-ordinate bitinguats who
had learned each language in separate conrexts, and so kept them
distinct. Over-simplifying, co-ordinate bilinguals were assumed
to have two meaning systems each with its own set of words,
while compounds had a single system with two sets of words.
Co-ordinate
ish concept'table
English word 'table'
Navajo concept'table'
Navajo word bik adani
Compound
Mixed concept'table'
English
word'table'
Navajo word bik adani
The problem that a compound bilingual would face is thar the

two words in fact refer to a different set of concepts. Underlying
this is the question of whether for the bilingual the knowledge of
the two languages develops independently or together. The
notion of domain difference suggests the different kinds of experiences most bilinguals have in each of their languages, implying a
common core of knowledge with subsequent differentiation.
Recent neurolinguistic research suggests that paired words are
stored in the same place in the brains of those who are bilingual
from infancy, br-rt in non-overlapping places in those who develop
bilingualsm later.
48
SURVEY
However explained neurophysiologically' the phenomenon of

bilingualism is the prime example of language contact, for the two
languages are in contact in the bilingual This contact can lead to
interference. A compound bilingual who has learned the meaning
of words in another ianguage by attaching them to the words of
his or her first language demonstrates semantic interference.
There can be interference in all aspects of a language, from the
sound system (having an'accent') to conversational rules (interrupting or saying'please' in the wrong way).
The phenomenon of interference, especially when it involves
using the two languages together, has led to the study of code
switching.
Code switchin$ and code mixin$

Bilinguals often switch between their two languages in the rniddle
of a conversation. These code-switches can take place between or
even within sentences, involving phrases or words or even parts of
words. The switching of words is the beginning of borrowillg.
which occurs when the new word becomes more or less integrated
into the second language. One bilingual individual using a word
from language A in language B is a case of switching, but when
Lnany people do, even speakers of B who don't know A are likely
to pick it up. At this stage, especially if the pronunciation and
rnorphology have been adapted, we can say the word has been
borrowed.
There are various kinds of code switching. Immigranrs often use

many \Mords from their new language in their old language,
because many of the people they speak to know both languages.
In situations like this, bilinguals often develop a mixed code. In
such a case, we might want to distinguish between code switching
of the two languages and the mixed variety. The history of English
shows many such mixed codes, as first Danish and later Norman
words were added by bilinguals. The various contemporary

Englishes, such as Jamaican English or New Zealand English, can
be seen as mixed codes, with the addition of local iexicon as their
obvious feature.
For a bilingual, shifting for convenience (choosing the available
word or phrase on the basis of easy availabiiity) is commonly
11ost
49
related to topic. Showing the effect of domain differences, a

speaker's vocabulary will develop differentially for different
topics in the two languages. Thus, speakers of a language who
have received advanced education in a professional field in a
second language will usually not have the terms in their native
Soc ietal multilingualism
language. Scientists trained in an English-speaking country giving
university lectures in their own language often mix in English
words or even switch to English phrases and sentences.
More interesting effects are achievable
by shifts concerned
with
role-relationships. It is important to note that each of a bilingual's

languages is likely to be associated not just with topics and places,
but also with identities and roles associated with them. In the
midst of speaking about work matters in Language A, a sentence
or two in Language B will be able to show that the two speakers
are not just fellow-ernployees but also fellow members of an ethnic group. The use of tags and expressions from Language B while
speaking Language A enables a speaker to make this kind of identity claim easily. This kind of shift, called metaphoricat switching, is
a powerfui rnechanism for signalling social attitudes or claiming
group membership or solidanty.
The selection of a language by a bilingual, especially when
speaking to another bilingual, carries a weaith of social meaning.
Each language becomes a virtual guise for the bilingual speaker,
who can change identity as easily as changing a hat, and can use
language choice as a way of negotiating social relatlons with an
interlocutor.
The bilingual individual thus provides a rich field for sociolinguistic study. A full understanding of bilingualism, howeveq
depends on a deeper understanding of the natllre of the speech
communities in which they operate. In the next chapter we will
look at societal multilingualism.
Multilingualism
The discussion of speech communities and repertoires in Chapter
z foreshadowed ou detailed consideration of the interest that

sociolinguists take in bilingual and multilingual societies.
Bilingualism and multilingualism, whether in an individual
speaker or in a social group, are the most obvious and salient
Vith stylistic or dialectal variation,

is harder and open to dispute, but
cases of variation to observe.
identifying each variety
with distinctly recognized languages, there is generally agreement

'We
can study how two or more
on the varieties and their names.
languages intertwine and separate without first being forced,
as we are when we talk about stylistic variations within a single
language, to establish the criteria for difference. It is both the
salience and the commonness of multilingualism that has led to its
being so well studied.
Monolingual speech communities are rare; monolingual countries are even rarer. Even a country as linguistically homogeneous
as Japan has its linguistic minorities like the Ainu and the
Koreans, marginalized as they might be. True, many countries
have developed an explicit or implicit language policy as though
they were monolingual, but it is rare (and becoming rarer) for lin-
guistic and national borders not to overlap in various complex
ways. Most countries have more than one language that is spoken
by a significant portion of the population, and most languages
have significant numbers of speakers in more than one country.
Historically, multilingual communities evolve in a number of
ways. One is as a result of migration, the voluntary or involuntary
5o
SURVEY
SOCIETALMULTILINGUALISM
5I
movements of people speaking one language into the territory of

people speaking another. \7hen the Hopi Indian permitted or
encouraged a group of Tewa Indians to move from the Rio
Grande area to the Arizonan mesas (each group has a different
version of the story), they produced a bilingual village, Flano,
among nine that were Hopi-speaking. Mutual distrust and a ban
on intermarriage that lasted into the beginning of the twentieth
century kept the villages so

villagers of Hano added Spa
repertoires, and after the i
Affairs schools, joined the rest of the Hopi in shifting rowards
English use.
Involuntary migration or forced movement of population was
common in the ancient Middle East, as is recorded in the Biblical
account of the Babylonian exile, and has continued to be a significant force accounting for multilingual communities. In the nineteenth century, the British policy of bringing indentured Indian
workers to the Fijian sugar plantations led to Fiji's currenr division between speakers of the indigenous Fijian dialects and
Hindi-speaking descendants of the original plantation workers.
The African slave trade moved large numbers of native speakers
of different languages into the East and I7est Indies, and led to the
formation of the pidgins and creoles (to be discussed in a later section). In the twentieth century, the Soviet policy of forced movement of populations assured that many of the newly independent
post-Soviet countries are saddled with a challenging multilingual
problem. In the Baltic states, it is the Russian immigrants, once
the rulers, who are faced with the challenge to learn the now
dominant Estonian, Latvian, and Lithuanian.
In the years after the Second'World \Vaq Northern European
countries, too, enhanced their multilingualism by encouraging
guest workers from the Mediterranean areas. There are significant Turkish minorities in many parts of Europe, and Greek,
Spanish, Algerian, and Italian immigrants moved north in the
same way. In a response to the social and linguistic problems
produced, a new Norwegian multilingual policy is intended to
cope with (and encourage rhe maintenance of) nearly a hundred
languages.
Voluntary migration has produced major changes in the
5z
SURVEY
inguistic make-up of many countries in the world. !7hile some of
ils multilingualism was produced in other ways, rhe United
Srrltes, as the world's foremost receiver of voluntary immigration,
lircw quickly into a multilingual society, constantly assimilating
l:r rge numbers of the immigrants through a melting-pot policy. In

the nineteenth and early twentieth century, the United States
,r bsorbed large communities of speakers of German, Norwegian,
( ireek, Italian, Yiddish, Polish, Ukrainian,
Japanese, various
(lhinese languages, and Spanish. The rate of absorption was
down after 1923, when strict immigration laws were

policy in the post-war
pcriod, including an influx of South East Asian speakers of
Vietnamese, Cambodian, Laotian, and other languages, and a
r.ccent wave of immigrants from the former Soviet Union. Most of
sl<rwed
prrssed. There was some relaxation of this
these groups have acquired English, and many have given up on

t
heir traditional languages. Throughout this period, continued
rrnrnigration, legal and illegal, especially of Spanish speakers, and

the rise of ethnic awareness have been threatening to upset this
r o mfortable monolingual trend.
Migration from the countryside or from small towns to the
lrr rge metropolitan cities that have grown everywhere in the twent ie th century is another ma jor cause of multilingual communities.
ln the Third World as much as in the developed countries, this
nrovement to the cities is creating huge megalopolises, conurba-
tions with populations in the millions, attracting complex

l)atterns of multilingualism, and producing major problems
ior social, economic, and political development. As African
cities expand at an ever-increasing rate, they too become highly
nrultilingual.
Multilingualism has also historically been creared by conquest

rrnd the subsequent incorporation of speakers of different languages into a single political unit. The incorporation of Brittany,
Alsace, and Provence into France submerged the languages of

these regions. The spread of English power over the British Isles
produced multilingualism and lead to the loss of some Celtic languages. The growth of the Russian empire under the Czars, conrinued under Soviet rule, made the Soviet Union a multilingual
country. The conquest of Central and South America by the

Spaniards and Portuguese eventually produced countries with
53
large indigenous minorities, some still speaking many Indian languages. The occupation of New Mexico and Texas and the incorporation of Puerto Rico by the growing United States included
new Spanish-speaking populations within territorial limits.
Colonial policies also led to multilingual stares. I7hile
the
Moslem Empire largely replaced the indigenous languages with

Arabic, pockets of multilingualism remained-the Kurds in
Turkey, Iraq, and Syria, the Aramaic speakers in Syria, the Copts
in Egypt, the Berbers in Algeria and Morocco, to mention a fewand the language mixes led to the great variations in the spoken
Arabic dialects held together by the general acceptance of an overarching Classical Arabic. !hen Spain conquered Latin America'
it created countries where Spanish dominated a mixture of marginalized indigenous varieties, including some, like Mayan, that
had previously had their turn as the dominant language in a multilingual empire.
IThen the ma;'or European powers divided up Africa in the
nineteenth century, they drew boundaries that left most postindependence states without a single majority language, and usually with languages that had many speakers outside as well as
inside the new state borders. They thus opened the wan wittingly
or not, for a tendency to adopt the colonial government's metropolitan language as a needed lingua franca. Newly independent
states like India, Pakistan, Indonesia, and Singapore also faced
complex language policy decisions that were heavily weighted
with effects of colonial policies.

Many of these former colonies might be considered cases of
forced federation. More rarely, there has been voluntary federation. One classic case is Switzerland where speakers of French,
German, Italian, and Romansch formed a multilingual state.

Another is Belgium, where Waloon speakers of various French
dialects, Flemish speakers of various Dutch dialects, and some
speakers of German dialects added to the package, now form an
uncomfortable but working French-Dutch bilingual state. Other
federations, like the Serbo-Croatian union in Yugoslavia, or the
Czech-Slovak union, brought into existence in the halcyon days
of the post-First l7orld \Var spirit of tolerant democracy and held
'World
in place in the post-Second
\Var period by Soviet power,
have proved to be unstable after the collapse of the Soviet empire.
54
SURVEY
These diverse historical circumstances have produced many

rlifferent kinds of multilingual mixes, sometimes stable and sometimes volatile and short-lived. The most common result of this
language contact has been language conlict, producing pressure
lrom one language on speakers of other languages to adopt it.
I'his pressure, whether the conscious result of a planned policy or
the effect of a multitude of unplanned factors, has produced challcnges to social structure that many people have begun to worry
rr lrout. The study of language maintenance and of language shift has
t h r-rs become a central concern of sociolinguists interested in mulrilingual societies.
Language loyalty and reversing language shift

Many people nowadays have become troubled by the extinction
of various species of animals and birds, and lists of endangered
species are regularly publicized. Linguists have noticed that lanquages too are in danger of dying, and for some time have been
srudying language loyalty, the ability (or lack of it) of speakers of a
lrr nguage to stand up to the pressure of more powerful languages.
'I'hey have
expressed distress at the threatened fate of endangered
tanguages, languages that are no longer being passed on to chil-
tlrcn as native languages, but are spoken by a contracting and

group of adults.
One early major study looked at what happened to the immirrant and indigenous languages of the United States when faced
by the inexorable power of English. Most American immigrant
languages were slowly sapped of their strength as younger speakt'rs shifted to English not just in the public domains, but also in
rheir own community and homes. Some factors appeared to speed
rrp the process or slow it down. The greatest resistance to language shift was found in groups that chose to isolate themselves
both linguistically and culturally from the mainstream. Two clear
'rging
cases were isolationist Mennonite Christian groups (especially
llutterian and Old Order Amish) and the ultra-orthodox
l{assidic Jews, both of whom rejected not just the language but
elso the dress and social conduct of their new country. In these
cases, the isolation was self-imposed.
A second group that maintained their languages were those
55
who were segregated and isolated by the outside society and

whose access to the easy social mobility that other immigrants
enjoyed was obstructed by social discrimination. The clearest
cases here were the indigenous Native Americans and the various
Spanish-speaking indigenous and immigrant groups. I hen they

were denied access to jobs, housing, and education, they were at
the same time cut off from easy access to the English that was
assumed to be the way to assimilation.
Language shift has been studied in many parts of the world.
There are groups that have worked actively to reverse the seem-
ingly inevitable language shift that occurs when small weak
languages, or the languages of marginalized groups, come into

contact with large powerful languages used and favoured by the
majority or dominant group. There have been many attempts to
correct this loss of linguistic diversity. A commonly cited case is
the national effort to revive the use oflrish in lreland, a nationalistically inspired and state-supported initiative to preserve Irish in
the western areas (the Gaehacht)where it was srill spoken, and to
teach it through the schools in the other areas where there were
few speakers left. In the English-speaking areas, students continue to learn Irish at school, but to use it very little outside school
or afterwards. Even inthe Gaeltacbt there has been a continued
loss, largely because of the failure to combine social and economic planning with linguistic. At first, the continuing poverty of
the area led Irish speakers to move away to the cities or emigrate,
in both cases switching to English; later, economic development
plans brought in English speakers looking for jobs.
More successful was the revitalization of Hebrew, a strong ideologically based process realized between r 89o and r9 r4, mainly in
Ottoman Palestine, by returning Zionists who were looking to

build a new nation using an old language. In agricultural settle-
ments, in new towns like Tel Aviv, and in communal settlements,

Hebrew was revitalized. It had the component of vitality or narural
intergenerational transmission restored after some rToo years
during which it had only been learned as an additional language.
Building on the widespread knowledge of the conrinually enriched
written language, and driven by the force of their ideological com-
mitments, the revivers were successful in establishing modern

Hebrew as the language used for all purposes in Israel today.
56
SURVEY
Activities aimed at reversing language shift are sometrmes prlvrte and small (as with the few hundred enthusiasts working to
rt'vive the Cornish language) but often public and political. The
r'lftrrts to save French language, culture, and identity in Quebec
t h reaten to divide the province from the rest of Canada. In Spain,
thc post-Franco policy of granting semi-autonomy to the regions
lrrrs led to strong government-supported campaigns for Basque
.rnd Catalan. In the Baltic States, the collapse of the Soviet Union
lurs permitted the restoration of the power of Estonian, Latvian,
,rncl Lithuanian.'$7e shall return to discuss this issue later when
wc talk of language planning and policy.
Language and ethnic identitY

!hy does multilingualism and
language contact entail so much
cnr<ltional reaction? The answer lies not in the practical commu-
ricative realm, but in the symbolic function of languages and

v;r
rieties. One of the most common ways of identifying
person is
lry his or her language. Because language is inherently involved in
socialization, the social group whose language you speak is an
rnportant identity group for you. There are other markers of

thnic identity, such as food or clothing or religion. But language
special role, in part because it organizes thought and in part
lrccause it establishes social relations.
Multilingual societies inevitably face conflict over language
lr,rs a
ehoice. Some aspects of concern for language choice can be

e xplained practically, politicalln or economically. The speakers of
language are in a stronger position when their language is used
'r
rlr national or international communication, or for government.
,rr for trade and commerce, or for education. But the role of
language in establishing social identity adds an additional, nonnraterial dimension to the conflict.
Ethnic groups regularly use language as one of their most
significant identifying features. There are some groups, like the
l;risians in the north of the Netherlands, who are hard put to find
other features that distinguish them from their neighbours.
(lommonly, the name of an ethnic group and its language are the
same. Most ethnic groups believe that their language is the best
ruredium for preserving and expressing their traditions.
f
57
One of the paradoxical effects of this connection of language

and ethnicity may be understood by looking at rhe case of postFranco Spain. I7ith the granting of some degree of autonomy ro
the provinces, Catalan and Basque have once again been recognized as official languages in their own autonomous regions. The
result of this new territorial policy has been to create problems for
people who are ethnically Basque or Catalan but live outside the
regions, and for people who are Castilian speakers but live within
them.
As we will note in the next section, conflict over choice of lan-
guage often accompanies the development of a new nation.
Language and politics

Language is regularly used in the exercise of political power. A
government can attempt to control its minority groups by banning their language, as Turkey bans the use of Kurdish by one its
larger minorities. By requiring that voting material be made available in Spanish and other minority languages, the US Federal
Voting Act tried to increase minority participation in government. By offering extra pay to federal Civil Servants who knew
both English and French, the Canadian government attempted ro
weaken the demand for Quebec separatism. By requiring all its
citizens to pass examinations in Estonian, Latvian, or Lithuanian,
the newly independent Baltic states attempted to redress the balance of power for indigenous citizens over the large Russian
minority populations that were dominant during the period of
Soviet rule. The issue of language choice is most critical in the case
of a newly independent state, as will be discussed in the next
chapter.
There are more subtle uses of language in politics. The use of a

regional or a social dialect by a political leader is often a claim to a
specialized ethnic identity. South American politicians sometimes
claim greater regional identity by using more Indian features in
their Spanish. Labour Party politicians in England have sometimes used regional accents to mark a dissociation from middle-
class speech and values. Anwar Sadat backed away from

Pan-Arabism by using more Egyptian vernacular in his speech
when the norm for Arab public speech is the Classical language.
58
SURVEY
Language rights
l'he issue of language or linguistic rights provides an opportunity
to attempt to take an ethical rather than a scientific view of lan-
Buage contact and corrflict. There are a number of possible

,rpproaches. One, favoured by some linguists, puts emphasis on
rhe right of a language, like any other endangered species, to survive. Because every language incorporates some unique features
derived from the rich and varied experience of human beings,
language loss (i.e. the loss of all its speakers) is held to be as serious
the loss of an animal or bird species. There are two possible

ways of dealing with this. Most commonly, anthropological linguists have worked to preserve, in a grammar, dictionary, and
:rs
collections of texts, as much of the language as possible while

there is still one speaker alive. More recently' linguists have provicled support to the speakers of the language in their efforts at
lt'versing language shifr.
The second approach is to focus not on the rights of the lanllLrage as an abstraction but on the rights of the speakers of the
lrrnguage. Here, we may distinguish between the rights of the
speakers of a language to use it, and their rights to maintain it by
tcaching it to their children.
The first of these issues concerns the rights of linguistic minori-
tics or of individuals who do not speak the national or official

lenguages of a political unit. To the extent that a state recognizes
the right of its citizens and other inhabitants to access to work,
health care, housing, education, iustice, and democracy, so it
must take care to deal with the potential lessening or blocking of
these rights for those who do not speak, read, or write the official
or national language or languages. There are several ways this
right may be recognized. One is the provision of adequate instrucrion in the official or national language or.languages to all who do
not control it-not just children, but new immigrants and temporary foreign workers. A second is the provision of interpreting
rrnd translating services to those who have not yet had the opportunity to learn the national language. This first language right,
therefore, is the right to learn the national language, and in the
rneantime, to be assisted in dealing with those situations where
lack of control of it leads to serious handicaps.
59
A second right is not to be discriminated against, in access to

work, education, justice, or health service, on the basis of being
identified as a member of a group speaking another l".rg.r"g. oi
variety. This refers to the way in which linguistic
-irrority ,i.-_
bers, however competent they may be in the standard l"r.gr.rag.,
are often classified as 'bilinguals' and afforded lo*., ,t""t.,r."I
also refers to the way in which speakers of an unfavoured dialect

or accent or other variety of a language are automatically recog_
nized as 'different' on linguistic grounds, and discriminat
against. It should be noted that this right is part of the larger right
not to be discriminated against on the basis of group mem"bersh"ip,
religion, gender, ethrric group' or other factors irrelevant to te
matter being decided.
A third righr concerns the right of a group of speakers of a lan_
guage to preserve and maintain their own favoured language or
variety, and to work to reverse any language shift to th. t"ti,s o.
prestige variety. Here, there are some more complex issues. One
is the potential conflict berween the rights of lndividuals and
groups. A group may wish ro preserve its language, but individual
members may prefer to shift to the dominant language, which is
generally a language more able to deal with modein life and eco_
nomic success. Another is the issue of who should pay for the
reverse shift efforts. Should it be the language community, and
should
it be
provided outside the regulr school
'y.t"-t
Examples of this are rhe Greek and Chinese afternoon schtols
in
the USA and some orher countries, the
Jewish Day School move_
ment that has grown up in the USA, Canada, Australia, Latin
America, and elsewhere, and the international schools thai oper_
ate rn many countries. Or should it be the state, in programmes ro
provide bilingual education ro as many minority groups as possi_
ble? In this issue of linguistic minorities, it is generaliy
that indigenous minorities, like the Native Americans"...pt.d

in the
United States, have a higher claim to maintaining language, reli_
gion, and culture, than do immigrant groups who cu-. by.hoi...
Considering language rights takes us into major issues of lan_
guage policy, which will be dealt with in Chapter
7.
6o suRVEy
Pidgins and creoles

A second aspect of language contact is the development of distinct
varieties of language. A pidSin language is one that evolves in

circumstances where there are limited relations between the
speakers of different languages, such as a market, or where there
is a special situation of power relations, being typical of the kind
tlf master-slave relation on a plantation. It is a variety of language
that is marked by the fact that it is not a native language of
iulyone, but is learned only in contact by people who normally
continue to speak their own language inside their own community. The complexity of a pidgin varies according to the commurricative demands placed on it; as there are increasing functional
tlemands, there is a growth in the power and complexity of the
pidgin to meet them.
A pidgin is a social rather than an individual solution. There are
c,rses where individual speakers acquire only a limited control of
rr language in which they need to do business. Such, for instance,
was the limited knowledge of Navajo developed by white traders.
lrach speaker made his own mistakes and compromises. The term
'pidgin' is better kept for social varieties with established norms'
A pidgin involves the mixture of two or more languages.
Sometimes, the grammatical system is based more or less on one
language and the vocabulary is largely taken from another. In all
cases, the grammar is simplified, that is to say certain features of
the base language are dropped. Many different pidgins have been
identified and described, including, to name a few, Nigerian

Pidgin English, Papuan Pidgin English, Vietnamese Pidgin
l.rench, New Guinea Pidgin German, Kenya Pidgin Swahili,
Fanalago (a pidgin based on ZuIu), and Chinook Jargon.
In many social circumstances, pidgins have become quite stable
over time. Spoken only as second languages, and functioning in
limited domains as languages of wider communication' they are

learned informally in contact and used especially as trade languages. In multilingual areas where each of the existing language
groups maintain their distinctiveness and do not intemarry, the

pidgin continues.
In many cases, there is a further development. This occurs
when, as a result of intermarriage of a couple whose native
SoCIETALMULTILINGUALISM
6T
languages are different but who both speak pidgin,

the pidgin is
spoken at home and learned by children as a first
tongue'
oi -th".
ory, this leads
to some
do so in
and it is believed that this
invot
.the-language
ir,H*m:
guistic capacity and universals that accounts for first

l".rg,r"g.
acquisition. New fearures emerge as a result both
of tt is Jrra 3r
the growing complexity of the social circumsrances

in which the
language is used. It is no longer just a contact language,
with lim_
ited social functions, but is called on ro deal *itt
"r]r...;;;"gly
needs. The process is called creolization,
as t"he
nd develops, displaying grearer phonoiogical

mplexity.
Some of the better-described creoles are Haitian
Creole, Tok
Pisin (a creolized version of a New Guinea pidgin English),
and
Hawaiian Creo
include Berbice
Spanish). These
cal complexity
course show many of the characterisrics of their original

pidgin
status, such as the blended phon tlogy, and the e*istence
of t*Jo.
more grammatical and lexical bases.
A third stage of development can occur when speakers of a
creole or pidgin are introduced, usually by education,
to the standard language on which the creole or pidjin was originally
based.
There can ensue what has been labelled Jpost_creole
continuum, in
which the various levels of social and stylistic variation
may be
filled by a version of the standard language at the upper
end and
of the creole or pidgin'at the lower ..r. A";"-"ican_may,
in various.social situations, choose the creole caild;amaica
Tak or one
of the various intermediate revers, or a standaid
Jamaican version
of English, or may switch from one to the other as
in other kinds
of code switching.
One controversy in sociolinguistics has been over the
origin of
.
the variety of English associated with the speech of "Af.o_
Americans. For many years, it was assumed to be a non_Standard

social dialect, similar ro and based on the Southern
regional
dialects of American English, and reflecting the social
isolation
6z
SURVEY
'rnd
inadequate education of the former slaves. Some psycholo-
gists pointed to certain features, such as the absence of the copula

(the verb 'to be') in the present tense, or different rules of verb
:rgreement, or the use of the double negative, as evidence of lin-
guistic inferiority and therefore justification for discrimination

rrgainst speakers of the variety. Linguists pointed out that these
catures afe common in standard languages (Russian and Hebrew
tlo not use a copula in the present tense, French negatives are usu,rlly double), and show that these and other features support a
rl-reory that Black English, as they labelled it, derives from an original creole like the Gullah still spoken in some communities.
l;rom this, others argued for its status as a separate language, and
c,rlled for its recognition and maintenance. The controversy over
what is variously called Black English, Afro-American Vernacular
l,.nglish, or Ebonics has raged in American educational situations
for at least thirty years.
Because of their lack of formal recognition, pidgins and creoles
,rre often treated just as a local jargon and linguistic aberration. It
is only recently that they have become an area of great interest to
linguists interested to learn about universal tendencies in languages and to study language status, attitudes to language, and
the importance of language to group identity. There are still many
controversies about how to describe them and how to explain
their creation and development. But it is their very marginality
rhat makes them interesting to sociolinguistics, for they are most
open to social influences and, lacking academies and educational
cstablishments, the least likely to be formalized and restricted by
normarivistic ru le-mak i ng.
Diglossia
A third aspect of language contact relates to the issue of functional allocation.
'!7ith
a handful of languages, two distinct varieties of the same language are used, side by side, for two different
sets of functions. The term diglossia (modelled on the word
bilin-
gwal, and using Greek rather than Latin forms) was coined origi-
nally to label this phenomenon. In the Arabic-speaking world,

there is the contrast between the Classical language and regional
dialect. The same pattern, more or less, occurs in the German-
SoCIET'AL MULTILINGUALISM
63
I
rland with High German as the
Cerman as the vernacular, in Haiti
ole, and in Greece with the literary

vernacular' dh imotik. !7hile there
iety are rnarked by the use of grammatical suffixes. There are also
major differences in the vocabulary of the rwo varieries.
situations in which diglossia often occurs, with the H language

rrssociated with power. Educational pressure is normally in the
clirection of the H varietn and those who cannot master it are usuelly socially marginalized. At the same time, the L variety rnaintirins value as a marker of membership of a peer or ethnic group.
'S7hile
the classical diglossic cases have been stable for a long
time, sociopolitical changes are starting to have their influence.
l(eference has been made to the possible emergence of an intermecliate variety of Arabic, a kind of Educated Standard Arabic. In
many countries, too, the globalization of English has introduced a
rhird significant language, so that triglossia or polyglossia is starting to emerge. This tendency confirms our central theme, the close
intertwining of social and linguistic structure, so that changes in
orle are reflected in changes in the other.
system. It is also likely to be used over a wider region and thus can
serve some unifying purpose. The L varieties are more localized
and show dialectal variation and the tendency to change of
unwritten dialects.
cate the boundaries, and offering two clear identities to the mem-
bers of the community. It is important also to note rhe political
64
SURVEY
SoCIETALMULTILINGUALISM
65
W
I
ffi
Language policy and language planning
In this chapter, we will look at a number of kinds of language

planning or policy activities. These terms represent fashion rathr
than other differences. s7hen sociolinguists started to be involved
in the 195os and 196os, they preferred the term tanguage ptanning
as the term for any effort to modify language form or use. In the
late r98os, the regular failure of national planning activities
seems to have encouraged the more neutral-seeming term,
language policy.
Exactly where these activities arise depends in large measure on
the perceived language situation of the social organizaton
involved. For instance, in a situation where there are seen to be

two or more languages available, any attempt to set up norms or
rules for when to use each is what is called status planning. A
decision to make one language official, or to ban another from
use in school, or to conduct church services in a third, are cases of
status planning. The most studied cases are in deciding on official
or national languages for a newly independent state.
Once a language has been fixed as appropriate for use in a specific situation (i.e. as the official language, or in printing books, or
in schools), any effort to fix or modify its structure is called copus
planning. The coining of new terminology for languages coping
66
SURVEY
with modernization, or the Young Turk policy to remove Arabic

words from Turkish, or the French efforts to rid the language of
English words, or the Dutch decisions to change spelling, are all
cases of corpus planning.
One aspect of corpus planning is the process of language standardization, which consists of attempting to standardize grammar
and pronunciation towards some norm that is discovered or
invented by some officially appointed or self-proclaimed group
of language guardians. This process may be called normativism or
prescriptivism by linguists who study it, or 'keeping the language
pure'by those who carry it out.
A language status decision often produces a situation where
some people need to learn a language that they do not normally
speak. In Finland, for instance, the decision to recognize both
Finnish and Swedish as official languages means that Finns must learn
Swedish and Swedes Finnish. Sometimes called language acquisitaon planning, this process of language education policy is also
involved when a government decides which foreign languages are
to be taught in school or through other means. Similarln
national policy to develop literacy in a language might be considcred a kind of language acquisition policy.
For various reasons, a country or other social group may wish
to encourage other people to learn their language. Language diffusion policy is sometimes associated with religious missionary
work, as Islam spread Arabic, or with the national concerns of
imperialist powers, as in Soviet activities to spread Russian
throughout the USSR and Soviet-dominated Eastern Europe, or
the French policy to spread la francopbonie.
In countries where there is clear recognition of the existence of
two or more respected languages and associated ethnic groups,
such as Belgium, Switzerland, or Canada, status planning is an
important activity. This is also the case in newly independent
states where there exists a myriad of languages that must be chosen between, as in post-colonial India, Indonesia, or Nigeria. In a
country where there is believed to be only one important language, and where other indigenous languages tend to be marginalized, the principal activity tends to be some aspect of corpus
planning, such as the purification of the standard language.
APPLIED SOCIOLINGUISTICS 67
Status planning
Status planning typically becomes an important activity when a
country becomes independent, but it has probably already been a
central concern of the nationalistic activities that preceded
the actual independence. As mentioned earlieq nineteenth- and

twentieth-century nationalist movements generally included a
choice of a national language in their ideological package. Thus,
newly independent Norway proclaimed its freedom in moving
away from Danish power and language influence, but became
involved in a bitter and continuing struggle as to what form of
language to use; its eventual compromise has been to recognize
the official equality of two varieties, called Riksml and Nynorsk.
The Irish nationalist movemenr took the restoration of Irish as its
goal. The Zonstmovement at the turn of the century proclaimed
its interest not just in peoplehood and territorial regeneration, but
in the restoration of Hebrew as a national language.
The decision is particularly harsh in a post-colonial state with a
selection to be made among a number of indigenous language. In
India, the costly and still uncertain solution was ro recognize
seven languages in the constitution. In many new African states,
with the lack of congruence between language boundaries and the
political borders that had been drawn as the colonialist powers
carved up the continent, the choice among competing ethnic and
tribal languages was part of a struggle for central power.
In these cases, the decision has often fallen on the colonial
language as the official language for the new state. This is especially likely to be true in those cases where colonial policy was
most successful in imposing belief in the value of the metropolitan
language, and where the colonial language is only spoken, perhaps imperfectly, by a small educated elite. Such a policy was
ollowed by France in all its territories and by Portugal in its
colonies.
Issues concerning status planning continue to make headlines.
New Zealand has finally agreed that the Treaty of l7aitangi by
which it obtained sovereignty over the Maori in r84o requires it
to recognize Maori as an official language alongside English.
Post-apartheid South Africa ponders a language policy that will
provide appropriate status alongside Afrikaans and English for
68
SURVEY
tlrc many languages spoken by the African majority, recalling no

tloubt that it was a language issue-riots in Soweto by African
se hool children who were upset by a government decision to have
thcrn learn in Afrikaans rather than English-that may be consiclcred the beginning of the successful campaign to destroy
, r
partheid.
l.anguage-status policy is by its very nature a political activity.
l,inguists are sometimes consllted, but decisions are made by

ll()vernment or elected parliament and sometimes form part of the
torrstitution. The status decision determines which language or
l:urguages are to be used in various public functions, by governnrcnt, the legal system, the media, and the educational system.
Sometimes there is an explicit policy spelled out by constitution
,rr law. As mentioned earlier, the Indian constitution lists the
scven constitutional languages and lays down their use for official
;r rrd educational purposes. A recent French law (ruled unconstitutional by the constitutional court) attempted to make it illegal for
pLrblic meetings (including international scientific congresses) to
than French. The Soviet Constitution prolaimed that all languages were equal. The Quebec government
prssed a law requiring that all public signs and advertisements
should appear only in French, and another that laid down that
child whose parents had not themselves gone to an English'rny
language school in the province must have French-medium
rrse languages other
ilrstructron.
In other cases, questions of language status are determined by
n:rtional, regional, or local law, or are left to local practice.

l)resent United States policy is a mixture of some local laws, some
court-inspired recognition of language rights (see earlier discussion), and various local practices. Recently, however, a political
campaign has begun to make English the only official language.
In Israel, while Hebrew and Arabic are formally recognized as
official languages for certain purposes, there is de facto recognition that most public signs are in English as well.
\il/hile the term 'official language' suggests governmental recognition, it is worth determining what precisely this status implies.
In Quebec, the official status of French determines language use in
signs and in education. In New ZeaIand, the official status of
Maori effects mainly a requirement that government bodies
APPLIED SOCIOLINGUISTICS 69
adopt a Maori translation for their name and include it in their

otherwise English announcements. In the respective Spanish
autonomous regions, the new official status
of Basque
and
Catalan involves a wider range of activities, including support for

major government agencies concerned with the diffusion of the
language. In the United States, the Voting Rights Act required the
use in electoral ballots of major minority languages. In the constitution of the State of New Mexico written after its incorporation
into the USA, there was a requirement for pharmacists to be bilingual in Spanish and English. In countries with r:ecognition of the
need for second language knowledge, there is often a requirement
or a salary supplement for officials with the favoured competence.
Religious bodies often have significant language status policies.

The decision of the Roman Catholic Church to change the language of the Mass from Latin to the local vernacular echoed a
decision made four centuries earlier in the Reformation by the
Protestant Churches. Hinduism, Orthodox Judaism, Islam, and
Greek and Russian Orthodox Christianity, on the other hand, all
have language policies which support maintenance of the status
of a sacred language.
A political decision on the status of a language, if it is in fact

to be implemented, usually leads to other activities. Often, a
language whose status has changed needs to be modified in some

way. This is corpus planning. And often it will need to be taught
to people who do not speak it: this involves various kinds of

language-diffusion policy, language-acquisition planning, or
language-education policy.
Corpus planning
When it has been determined that the status of language is ro be
moved to a more elaborate level of standardization or to an
expanded set of functions, the task of corpus planning begins.
One of the most common processes is the need for modernization
and elaboration of vocabulary. The rapid increase of concepts
associated with the modern world, and the expansion of terminology.needed to label ali the new objects involved in modern
technology, set a major challenge for all languages.
7o
SURVEY
One obvious example is the computer I am working at noq

which is a laptop or, to be more precise , a sub-noteboo. I just had
ro have a new motherboard put in, as the old one would not
PCMCIA card that is supposed to be used to connect
ny diskette driue. Bt my dowble-spaced hard driue ofers me a
ntegabyte of memory (not to mention the r6-K RAM that my pro'qrdms can access) and my trackball o ?nouse makes it easy to
control my pctssiue matrix screen. Most of the words I have italici.ed arc ones that were not needed in English a decade ago, or
lrave taken on new meanings quite recently. The problem facing
language that wishes to deal with the modern world is that it
'rny
rnust keep up with the new developments.
There are some obvious choices. A language can simply take an
rrccept the
word (Iike driue or screen) and give it a new meaning. A comltuter rnthe Oxford English Dictionary (r933 editionl rs aperson
who does calculations. A mouse still seems a quaint word to most
rrld
oltus for a pointing device. To say that a computer has a memory

rrs a storage device is a pretty obvious metaphor (and, so I suspect,
is stordge deuice).
Another technique is to coin a new term, like trackball,

or megabyte, by combining existing words or mor-
ttiskette,
phemes into more or less transparent forms. For many languages,
rhe simplest technique would seem to be borrowing from another
lrrrrguage where the term is in use because the concept or object

has already been invented.
English is one language that has borrowed freely. From its

beginnings in regular contact with first Danish, then French, its
lexicon was constantly being enriched not just by coinages but by
borrowing. During the periods of scientific and technological
clevelopment, English freely coined new words based on Latin or

(ireek (or even' to the horror of Classical purists, with elements o
both).
As part of his policy of unifying France, Cardinal Richelieu

cncouraged the establishment of the Acadmie Frangaise, one of
whose principal charges was to be to maintain the unity and
purity of the French language. This it still continues to do,
protesting regularly the tendency towards Franglais and the use
<rf words hke le weekend and le biftek.
One of the earliest kinds of corpus planning, called for as a
APPLIEDSOCIOLINGUISTICS
7T
language takes on official, standard, and educational functions as
a result of changed status, is the developing of an
orthography.
riting has not been invented very often, but more commonly it
has been borrowed and adopted from one language to another.
Most recent orthographies are slight modifications of other
alphabets. The roman alphabet is most commonly used, under
the influence of European languages. The Stalinist policy of linguistic centralization involved also changing the orthographies of
many languages in the Soviet area of influence from Roman or
Arabic orthography to the Cyrillic in which Russian and related
languages are written.
'Westernization
major component of the Turkish
movement was to change from Arabic to Roman
script. Romanization has been proposed for Hebrew and
Chinese, but with no success, for the weight of tradition has been
too strong. The cost of maintaining a non-Roman alphabet is not
small, as those who tried to develop a typewriter for them discovered, but computers are simplifying things.
But to develop an orthography is only a first step in the process
of standardizaton and modernization; later, we will look at the
related task of developing mass literacy.
Normativism and prescriptivism

\7hen in r97o aprogramme was started to teach Navajo children
to read their own language, one of the first inconveniences was
the absence of a typewriter that could produce the special diacritics and letters (a French acute accent, a Czech cedilla, and a
Polish slashed L) that the developers of the Navajo orthography
chose. (Incidentally, one of the few corpus planning decisions the
Navajo Tribal Council made was to spell the name of the tribe in
English with a 'j' rather than an 'h'.) A new problem soon turned
up.
The first reading book one of the assistants wrote was a story of
a Cat, fo which the writer used the Navajo word m si. A litt]e
later, another writer included a cat in a book, but chose rather to
write it msi. The Navajo dictionarn which had been written

some thirty years before, listed both spellings. The Navajo linguist who had collaborated in writing the dictionary and was now
a lecturer in the language at the Navajo Community College
SURVEY
backed the dictionarn saying that he always told his students to

write words as they and their family pronounced them. The
spelling msi was therefore used in the second book, with a note
ftlr teachers in the back of the book that some people used msi,
while others sad m si. The teachers were unhappy with this decision. They had been trained in English where there is usually only
()ne correct spelling, and considered it wrong to be called on to
tcach rules that allowed too many choices.
In this attitude, they reflect the point of view of most'Westerncducated people who assume and expect that language rules are
set, clear, unambiguous, and to be enforced. Some teachers of
lrnglish as a foreign language complain in the same way about difierences between British and American usage. This idea of 'correctness' is a mark of developed literate societies. In pre-literate
societies, one regularly finds notions of'a good speaker', by
which is meant someone who has power to speak persuasively in
public, but seldom the notion of'correct' speech and spelling.
If you look at an Elizabethan book, you will note that printers
were not concerned about 'correct' spelling, varying the spelling
of the same word on a single line if it made the words fit in better.
As printing and education spread, however, the notion of correctrress became increasingly important. In the absence of an
Academy legally charged with the task, dictionary and grammarbook writers took it upon themselves to describe and define what
rhey considered standard and correct usage, and to prescribe

these standards as required. Even when the description was
clearly labelled as an arbitrary choice of one out of a number of
varieties, it was generally accepted in a society seeking methods of
gate-keeping as the mark of education and acceptability. George
Bernard Shaw's Pygrualion (or the musical adaptation, My Fair
Lady) is a touching and accurate account of how changes in
speech and dress permitted a Cockney flower-girl to move into the
high society from which she was otherwise barred.

Prescriptivism, therefore, is an understandable development in
ir mass education system where successful learning of prestige
speech styles is a first step in social upward mobility. It is, however, unfortunately accompanied by a mistaken belief that speakers of non-standard varieties of a language are less intelligent, or
less inherently capable than standard speakers. When linguists
73
argue that all languages are equally good, they are arrempring ro
fight the common prejudices that assume that standard languages
and their speakers are inherently superior to non-standard languages and their speakers. Every variety, they maintain, has the
potential to handle all tasks, and there is no evidence that people
who speak a non-standard language are intellectually inferior to
people who speak a standard language. At the same time, the normal association of the standard language with literacy and with
formal education means that a key goal of many systems is to provide access to the standard language to the largest possible section
of the population.
Language acquisition planning or language

education policy
Teaching the standard language to all is one of the first tasks of
most educational systems. In traditional religious education, the
task of the religious school is usually to develop literacy in the language of the sacred texts: Hebrew for Jews, Classical Arabic for
Moslems, Sanskrit for Hindus, Old Church Slavonic for Russian
Orthodox, Giiz for Ethiopian Christians and Jews. In secular
education, the equivalent first task is developing control of the
written standard language.
The task is regularly complicated by the fact that the spoken
language of the home is commonly not the standard written language of the school. In the case of languages of religious rexts, this
was usually obvious. Teachers in ultra-orthodox Jewish religious
schools find it natural to teach Yiddish-speaking children in

Yiddish to read texts written in Hebrew. In diglossic situations
like those mentioned earlier, the problem is made more difficult,
for the teachers and pupils sometimes have no respect for the L
variety they actually speak, and ciaim or pretend to be speaking

the H variety.
The first task of a formal educational system is usually, therefore, the teaching of the national standard language, with emphasis on literacy in it. Depending on social and political pressures,
the system aims at students acquiring other varieties too.
Especially in countries which recognize more than one language

as part of the national tradition, there may be a programme to
74
SURVEY
reach the other official language or languages. As noted earlie5

for example, in Finland, all Finnish speakers are expected to learn
Swedish and all Swedish speakers to learn Finnish. Similarly, in
Quebec, all English speakers learn French and all French speakers
learn English, and in Israel, Arab children learn Hebrew, and
Arabic is in turn compulsory for Hebrew speakers.
Language diffusion policy or lin$uistic

imperialism
Political and military conquest have been major causes of language spread. Aramaic in the ancient world, Greek in the Eastern
Mediterranean, Latin in \estern Europe, Arabic in the Middle
llast and North Africa, Mayan in Central America, Manding in
West Africa are all cases. In these cases, the rulers did not follow
an explict policy of requiring the conquered to learn their
languages, but essentially left the choice open. Similarln when a
language has been spread by trade (as for instance Swahili in
Africa), the diffusion has been more or less unplanned. Even languages spread by missionary activity have not necessarily been
the result of direct planning, for missionary groups commonly
accept that the sacred texts will need to be translated ifthey are to
be understood. It is for this reason that missionary activity so
often leads to the development of vernacular literacy.
It is important to distinguish between this kind of unplanned

language intervention, where altered circumstances encourage
conquered or converted or commercialtzed populations to learn
the language of their conquerors, missionaries and traders, from a
planned language diffusion policy. In the first, we are dealing with
the kinds of language shift we have talked about earlier; in the
second we have a deliberate policy of a government or other institution to change language acquisition and use.
Language diffusion policy may be external or internal. !7hen a
country decides that all its inhabitants, whatever their home language, should learn and use the national language, we have a case
of internal diffusion. !'hen New Zealand started teaching all its
Maori pupils in English in r87o; when the British required
English in Welsh schools; when France would not allow Occitan,
Breton, or Basque in its schools; when Stalin pushed for the use
75
r
and higher starus of Russian in all Soviet schools; or when China
assumes that all pr,rpils speak and should learn putonghoa, these
are all examples of language education policy that involve
conscious and planned internal diffusion of a standard or official
language.
Policy in conquered lands or colonies is both internal and exter-
nal diffusio'. Because the territory concerned is now under the
control of the imperial rulers, it is not unnatural that they want to

make governing easier by encouraging some, at least, olthe colo_
nial subjects t. learn their language. colonial language ecrucation
policies have varied in their commirment to language diffusion.
There have been, and continue to be, policie, uil"rrg.rnge diffu_
sion beyond national and even imperial boundaries. Orre of the
earliest and strongest of these has bee' a French tradition of
encouraging the spread
of the French language beyond
its
national and colonial borders.

More recently, Germany has taken up language diffusion activ_
ities in its support for the Goethe Institute. A number of other
countries work through the cultural attachs in their embassies or
through semi-governmental bodies to encouage the diffusion of
their national languages. Teaching the nationailanguage, as well
as teachi'g about the national culture, is sometimes seen as an
effective way of spreading influence and developing international
interesr in rrade and rourisrn.
has continually exceeded the supply. Language diffusion efforts
of English-speaking countries have tended to be attempts to
exploit world-wide desires to learn the language. There has been

little need to fan the interest. The association of English with
rnodern technologn with economic progress, and with internationalization, has encouraged people all over the world to learn
English and to have their children learn it as early as possible. The
rnore this has succeeded, the greater the reason for others to want
to have access to the power and success assumed to be a result of
l<nowing English.
The demand is not new: as early as the r9zos, Japanese and
Chinese business leaders were starting to value knowledge of
English. And it is deeply entrenched: even at the height of the
Cultural Revolution in China, when there was deep suspicion of
rnodernity and foreign influences, Chairman Mao continued to
speak in favour of teaching English.
From the point of view of rnany observers, this growing linguistic hegemony of English is dangerous and harmful, and it is not
unnatural to seek someone to blame. But whatever the cause, tbe
spread of English is producing a new sociolinguistic reality, by
threatening to take over: important functions from other major
languages, and by furthering language shift. It is an important
rask of sociolinguistics to understand this process.
The spread of English-imperialism or
hegemony?
The recent world-wide diffusion of English, so thar it now looks

set to become a world language, has raised not just concern
among speakers of other languages, but controversy among
socioli'guists. To what extent, they argue, is it the .es,-rlt of conl
scious planning by rhe governmenrs and experts of English_
speaking countries like the United Kingdom, the United Siates,
Canada, South Africa, and Australia, and to what extent is it the
result of a i:rrge array of factors connected with modernization
and globalization?
A closer look at the process by which English has in this cenrury
developed into a global language Sugge sts that in fact the demur'
76
SURVEY
AI'PLIED SOCIOLINGUISTICS 77
SECTION 2
Conclusions
Readings
Because it is deeply concerned with language in society, socio_

linguisrics has from its start been equally inulued in social rnatters. It is no wonder, therefore, that sociolinguists have had to
come to grips with issues of social inequality, linguistically

encoded and enforced discrimination, and the banning of
languages, and the punishing of those who speak them. Haid as
many sociolinguists try to describe dispassionately and scientifi_
cally what they see, they are regularly entangled in the efforts of
minority groups to resist forced assimilation, or in defending
speakers of stigmatized varieties from being considered ineducal
ble. It is this regular ability ro help thar provides an extra incen-
tive lor so many sociolinguisrs
s been
the fiel
rious p
evelop
explain and predict sociolinguistic behaviour within the area of
interest, no over-arching model or u

lished to encourage the development
result is the freedom the sociolinguis
interrelationship between language and society, and to continue
to use increased understanding of the structue of one of the components to understand the other.
78
SURVEY
Chapter 1
Text 1
RoNA LD wARDHAUG u: An [ntrctduction to Sociolinguistics

(znd edn.) Blackwell r992, pages ro-r r
in the relationships betwcen

language and society. If we stdrt off by defining'sctciety' as an
organized group of people and 'language' as the way tbey
communicate uith eacb other, we haue already set up one
relationship between the two concepts. A society speaks a language. But there are otber, more complex, relationsbips that
might result. In the following pdssdge, Ronald Wardhaugh
suggests four possible relationships between the two.
Sociolinguists dre interested
'fhere is a variety of possible relationships between language and

society. One is that social structure may either influence or deterrnine linguistic structure and/or behaviour. Certain evidence may
be adduced to support this view: the age-gradlng phenomenon,
whereby young children speak differently from older children
and, in turn, children speak differently from mature adults; studies which show that the varieties of language that speakers use
reflect such matters as their regional, social, or ethnic origin and
possibly even their sex; and other studies which show that particr-rlar ways of speaking, choices of words, and even rules for conversing are determined by certain social requirements. We will
also find that 'power' is a useful concept that will help explain
rnuch linguistic behaviour. Power, as both something to achieve
READINGS
79
and something to resist, exerts considerable influence on the lan-
guage choices rhat many people make.
A second possible relationship is directly opposed ro the firsr:

linguistic structure and/or behaviour
-"y .ilh.. in]uence or
determine social srrucure. This is the view that is behind the
many of those
hese languages
rhe influence is
ce each other.
one variant of this approach is that this influence is dialectical in
nature, a Marxian view put forward by Dittmar, who argues that
'speech behaviour and social behaviour are in a state of constanr
interaction' and that'material living conditions'are an important
factor in the relationship.
A fourth possibility is ro assume that there is no relation ship at
all between linguistic structure and social structure and that each
is independent of the other. A variant of this possibility would be
to say that, although there might be some such relationship, pre_

sent attempts to charactertze tt are essentially premature, given
what we know about both language and society.
structure on languagc?
ma
either
hand
In
other
s' one uses the masculine
female (for example,,Eue

in time.') Is tbis 'sexist'?
to
td
of
ship'position?
Text 2
JAMES MrLRoy: LinguisticVaridtion and Change. Blackwell

rggz,pages 5-6
Presumably, tbe best udy to obserue the relationship between
language and society is when language is being used in a social
situation. For tbis reason, Milroy argues that we sbould

first
look at language being used in a conuersation.
Speech is a social activity in a sense that writing is not, and the

primary locus of speechs conuersation. Conversations take place
between two or more participants in social and situational con-
texts, and linguistic change is one type of phenomenon that is

passed from person to person in these situations. The first principle for a socially based model of language change therefore concerns the obseruation of language in use: it is the principle that
speech exchanges can be observed only within social and situational contexts-they can never be devoid of such a context. To
xpress this more fully:
Principle I
As language use (outside of literary modes and laboratory

experiments) cannot take place except in social and situational contexts and, when observed, ts always observed in
these contexts, our analysis-if it is to be adequate-must
take account of society, situation and the speaker/listener.
This first principie carries with it a number of implications, the

rnost important of which is that generalizations about languagc
structure depend on a process of abstracting'language'from the
'We
do not actusituational contexts in which it naturally occurs.
ally observe 'the language' or 'language' in the abstract: we
observe people talking. In a social account of language change,
therefore, we have to explain how changes get into this abstract
structure that we call language (which we cannot observe
directly) as a result of the activities of people talking (which we
can observe more directly). Furthermore, unstructured observations of very selective phenomena will not be enough here: our
descriptions of sociolinguistic patterns will depend on observing
recurrent patterns and will have to be systematic and accountable
to the data. . . . It also follows from this first principle that close
attention to methods of data collection and analysis (and the relation of one to the other) is crucial. . . .
\(hereas Principle I concerns the impossibility of obseruing

language independently of society, Principle z concerns the
impossibility of desuibing language structures independently of
society. This is not as controversial as it may seem.
Principle z
A full description of
8o READINGS
the structure of a variety (whether it is
READINGS
8T
ll
'standard' English, or a dialect, or a style or register) can
only be successfully rnade if quite substantial decisions, or
judgements, of a social kind are taken into account in the
description.
The word 'social' here does not mean social class or prestigethe decisions (or jr-rdgements) we are talking abour are decisions
(or judgements) abour the 'norms' of the variety concerned, and
these norms are social in the sense that they are agreed on
socially-they depend on consenslls among speakers within the
cornmunity or communities concerned and will differ from one
community to another.
wbat sense?
needed in order to obtain reliable information about language
uariation?
on'fudgements of a social kind'?

Text 3
wILLTAM LABov: The Social Stratification of English

in New York City. Center for Applied Linguistics r996,
pages
r-2
If ue accept tbe argument tbat our primary data should come

from language in social use, the question remains of how to
coLlect it. Tbe problem is tbat the obseruer (the data collector)
adds anotber pdrty to the conuersation.
For accurate information on speech behavior, we will eventually
need to compare the performance of large numbers of speakers.

Furthermore, we will want to study a sarnple which is representative of a much larger goup' and possibly of the New York speech
community as a whole. This cannot be done without random

sarnpling. Yet to complete random sampling, and to make the
data for many speakers comparable, we need structured, forrnal

interviews. Here is the paradox which we sensed: the formal
interview itself defines a speech context in whrch only one speaking style normally occurs, what we may call cAREFUL spEECH.
8z
READINGS
The bulk of the informant's speech production at other times may

be quite different. He may use careful speech in many other contexts, but on most occasions he will be paying much less attention
to his own speech, and employ a more relaxed style which we
may call cASUAL sIEECH. \i e can hear this casual speech on the
streets of New York, in bars, on the subway, at the beach, or
whenever we visit friends in the city. Yet anonymous observations
in these contexts will also be biased. Our friends are a very special
group, and so too are those New Yorkers who frequent bars, play
stickball in the streets, visit public beaches, or talk loud enough in
restaurants to be overheard. Only through a painstaking method
of sampling the entire population, and interviewing speakers
chosen at random, can we avoid serious bias in our presentation.
The problem is now to see what can be accomplished within the
bounds of the interview.
l> How
does the presence of a stranger with a recording instrument define the situational context?
Hou, tuould this interfere with obseruing casucll speech?

produce a biased picture?
Text 4
JosHUA A. FISHMAN:'The sociology of language' in Joshua

A. Fishman (ed.): Aduances in the Sociolctgy of Language.
Mouton t97r,YoI. t,page zzr
One of the important contributions of sociolinguists to the

study of language has been the effort to determine tbe social
ualue that is accorded to any uariety. Tbe writer argues that it
is not just tbe existence of difference that is important, but the
'symbolic ualue' giuen to a language or uariety that determines
whether people use it or not. tust as different kinds of clothing
can keep us equally ruarm, but a uniform symbolizes the
group ue belong to, so different uarieties also come to carry
udlue cts symbols of group membership.
All in all, the sociology of language seeks to discover not only the
societal rules or norms that explain and constrain language
READINGS
8,,
Chapter 2
The ethnography of speaking and the structure
of conversation
Text 5
RALpH FASoLD: Tbe Sociolinguistics of Language.

Blackwell r99o, pages 4o-r
.Just ds languages can uary, there can be fundamental differ_
ences from one community to another in the rules of how lan_
guage is used. Tbere can be different rules about when to

speak and when to be silent.
'il understand what the ethnography

of communication is all
rrbout, it is nece
It is one of Hym
damental concepts.
aking can vary substantially from
in the most fundarnental ways. For example, it has been pointed out (for instance,
Schegloff t97z) that most middle-class white Americans (an
1lossibly members of other !estern societies as well) have a .no
Many people argue that all

wbich'thely *ri,
Hou does Fishman's
t. r tueu
idea
prrrr-,
idea?
';f?
this
with it? Does this help one understand botu

national moue_
ments pick a language?
set
oth
Pedn
onds' duration, the participants become extremely uncomfort_

eble. Someone will start talking about somerhing unimportant
just to fill the 'gap' or the group will break up.
this rule for speakers who have it
a conversation being carried on in
GSzd found that ir was quite the
to carry on discussions with more
than one speaker speaking simultaneously. On the other hand,
Saville-Troike (1982) reports that there are American Indian

groups where it is common for a person to wait several minutes in
a speaking turn.
experiences in a
nversational gaps
Obviously, an ethnography of communication for middle-class
84 READINGS
READINGS
85
white Americans would include the ,no gap, no overlap' con_

versational rule- The corresponding description of Antiguan
speech rules would not include the ,no overlap' rule. And a
description of the American Indian groups savilie-Troike refers
to, or the Lapps that Reisman lived near, would not include the
'no gap'rule.
Hou
tr
would people in your community redct to Antiguan

speech style, or to the American Indian speech style tbat
S auille-Tr oike talks ab out ?
Do you baue tbe 'no gap' or the ,no
speech commwnity?
rule?
ouerlap' rule in your

Hou do people interpret breaches of the
Text 6
pENELopE BRowN: 'How and why are women more polite:

some evidence from a Mayan community'in Sally
McConnell-Giner, Ruth Borker, and Nelly Furman

'Women
(eds.):
and Language in Literature and Society. praeger

198o, pages r4_5
Most speakers are ctware that in certain situations, wben
talking to certain people, they need to take special care of their

speech. Being polite, Broun argues, is taking care not to hurt
otber people. But, she goes on, there is more than one uay of
being polite.
r7hat politeness essentially consists in is a special way of trearing

people, saying and doing things, in such a way as to take into
account the other person's feelings. on the whole that means that
what one says politely will be less straightforward or more com_
plicated than what one would say if one wasn'r taking the other's
feelings into account.
Two aspects of people's feelings seem to be involved. One arises
when whatever one is now about to say may be unwelcome: the
addressee may not want to hear that bit of news, or be reminded
ofthat fact, or be asked to cooperate in that endeavor. A request,

for example, or anything that requires a definite response diiectly
imposes on rhe addressee. One way of being polite in such situa_
tions is to apologize for the imposition and to make it easy for the
86 READINGS
rrddressee to refuse to comply. So we try to give the most interactional leeway possible, and this, in one sense, is what it is to be
polite.
Our long-term relations with people can also be important in

reking their feelings into account. To maintain an ongoing relarionship with others, one greets them on meeting in the street,
inquires about their health and their family, expresses interest in
their current goings-on and appreciation of the things they do and
like and want.
These two ways of showing consideration for people's feelings
can be related to a single notion: that of FACE. Two aspects of
people's feelings enter into face: desires to not be imposed upon
(negative face), and desires to be liked, admired, ratified, related
to positively (positive face). Both can be subsumed in the one
notion of face because it seems that both are involved in the folk
notion of 'face loss.' If I walk past my neighbor on the street and
pointedly fail to greet him, I offend his face; and if I barge into his
l.rouse and demand to borrow his lawnmower with no hesitation
rlr apology for intrusion (for example, 'Give me your lawnnlower; I want it') I equally offend his face. So blatantly and without apologies imposing on and blatantly and without apologies
ignoring the people with whom one has social relationships
rrre two basic ways of offending their faces. Three factors seem to
be involved in deciding whether or not to take the trouble to be
polite:
r One tends to be more polite to people who are socially

superior to oneself, or socially important: one's boss, the vicar,
the doctor, the president.
One also tends to be more polite to people one doesn't know,

people who are somehow socially distant: strangers, persons
from very different walks of life.
In the first situation politeness tends to go one way upwards
(the superior is not so polite to an inferior), while in the second
situation politeness tends to be symmetrically exchanged by both

partres.
3 A third factor is that kinds of acts in a society come ranked
as
more or less imposing, and hence more or less face threatening,
READINGS
87
and the more face threatening, the more polite one is

likely to
be.
These three factors appear to be the main determinants

of the
overall level of politeness a speaker will use.
o
y
How
bow they should be
all
is politeness related to social structure?
'How are you?'
to be more polite than men?
iour not to greet someone?
Text 7
ELATNE cnArKA: Language: The Social Mirror (znd
Newbury FIouse 1989,page 44
'l7hether
response 'Fine' can properly end the greeting sequence.

or not the person is truly fine is immaterial. Phatic communicarion has been completed with its utterance. If the greeter wants to
know more, such as why 'fine' was uttered glumly, he or she can
stop and ask for more information. At this juncture, it is proper to
go into details. Greeting, therefore, fulfills two functions: first, the
requirements of phatic communication, and second (if desired on
rhe part ofthe greeter), opening further interaction.
ed.n.)
Hou
many different forms of greeting do you use? Can you

expldin tbe circumstances under which you use them?
Chapter 3
Locating varaation in speech
Text 8
JoHN GUMppnz: 'The
speech community'in David L. Sills

lnternational Encyclopedia of the Social Sciences.
Macmillan r968, Vol. 9
(ed.):
Tbe writer argues for the double function of language, bcttb
communicating information by referring to extralinguistic

reality, and communicating social information about
the users. The system operates in uhat Gumperz defines
cts ct speech community, for which be proposes a linguistic
definition.
Although not all communication is linguistic, language is by far

the most powerful and versatile medium of communication; all
lcnown human groups possess language. Unlike other sign systems, the verbal system can, through the minute refinement of its
grammatical and semantic structure, be made to refer to a wide
variety of objects and concepts. At the same time, verbal interaction is a social process in which utterances are selected in accordance with socially recognized norms and expectations. It follows
88
READINGS
READINGS
8g
that linguistic phenomena are analysable both within the context

of language itself and within the broader context of social
behav-
y human
ag
by reguof verbal
fferences
ction by me
imilar aggre
the sociological concept of community is interaction. This much

is evident in the quotation from Gumperz's work cited above. It
encompasses a view of things reminiscent of the social dialectology of the nineteenth century with the emphasis on boundaries
determined by language which in turn is determined by social factors. The important features of Gumperz's definition are interaction and language use although we should not disregard his
reference to verbal signs. The concept once again appears to be
the product of linguistic rather than social factors since the

boundaries are defined by linguistic features. However, it would
be a sterile concept if it was discussed simply in terms of linguistic
uniformity. Thus we have the possibility of 'differences' in language use which relate to interactional factors. The attributes of
difference which serve to define the speech community are linguistic rather than social and they have the result of producing
'aggregates' rather than social groups in the sociological sense.
Thus, it would appear that the speech community is an aggregate
of individuals in interaction.
cctmmunity has in common with tbe sociological idea of com-
munity?
communication?
Does cumperz allow for multilingual and non-geograpbical
and language use' different from dialect'boundaries defined

by language features'?
Does this definition set dny size for a speech community?

D9 any non-linguistic features enter into Gumperz' definition
of a speech community?
Text 10
speech communities!
Cambridge University Press r98o, page
Text 9
c Lr'N wr L Lreu s : Sociol inguisti
Routledge 1992, page 7z
cs :
S o
ciolo gical Cr iti que.
disagrees tuith
terid. In tbe res
s for their failur
One feature which the concept of speech community shares with

9o
READ INGS
J.K cHAMBEns and rETER TRUDGTLL: Dialectology.

5
How does one distinguisb between a language and dialect?

Not iust on linguistic grounds. Nor is it simple to distinguish
between a dialect and an accent, although generally rae use

'accent' to medn a difference in pronunciation.
It seems, then, that while the criterion of mutual intelligibility

may have some relevance, it is not especially useful in helping us
to decide what is and is not a language. In fact, our discussion of
the Scandinavian languages and German suggests that (unless we
READINGS 9T
want to change radically our everyday assumptions about what a

language is) we have to recognise that, paradoxically enough, a
'language'is not a particularly linguistic notion at all. Linguistic
features obviously come into it, but it is clear that we consider
Norwegian, Swedish, Danish and German to be single languages
for reasons that are as much political, geographicl, historil'
sociological and cultural as linguistic. It is of course relevant that
all three Scandinavian languages have distinct, codified, standardised forms, with their own orthographies, grammar books,
and literatures; that they correspond to three separate nation
states; and that their speakers consider that they speak different
languages.
The term 'language' is thus from a linguistic point of view a

relatively non-technical term. If therefore we wish to be more
rigorous in our use of descriptive labels we have to employ other

terminology. One we shall be using in thrs book is venrp,rv. .We
shall use 'variety' as a neutral term to apply to any particular kind
of language which we wish, for some purpose' t consider as a
single entity. The term will be used in ad hoc manner in order to
be as specific as we wish for a particular purpose. !e can' for
example, refer to the variety 'Yorkshire English', but we can
equally well refer to 'Leeds English' as a variety, or .middle class
Leeds English'-and so on. More particular terms
will
be
a
ACCENT and nrRrpcr.'Accent'refers to the way in which
speaker pronounces, therefore refers to a variety which is phonet_

vrieties.
'Dialect', on the other hand, refers to varieties which are gram_
matically (and perhaps lexically) as well as phonologically ifferent from other varieties. If two speakers say, respectiv ely, I done it
last night and I did it last night, we can say that are speaking
different dialects.
ically and/or phonologically different from other
intelligible' when a person who speaks one cdn understand a

person speaking tbe other. Are there dialects of English or
your ou.)n language tbat you cdnnot understand?
9L
READINGS
American Englisb uith a British accent?
Chapter 4
Styles, gender, and socialclass

Text 11
cREGoRy R. cuy: 'Language and social class' in FrederickJ.

Newmeyer (e.): Linguistics: The Cambridge Suruey.
Cambridge University Press r988, Vol. tv, page 37
One obuictus effect of social structure on language should be
that groups that dre diuided socially should also sbow linguistic differences. Three obuious kinds of social diuision are by
gender, by age-grading (as mentioned inText
aboue), and by
social class.
In all human societies individuals will differ from one another in
the way they speak. Some of these differences are idiosyncratic,
but others are systematically associated with particular groups of
people. The most obvious of these are associated with sex and
developmental level: women speak differently from men, and
children differently from adults. These two dimensions of social
variation in language are in part biologically determined (for

example, differences in laryngeal size producing different pitch
levels for adult men and women), but in most societies they go
beyond this to become conventional and socially symbolic. Thus

rnen and women differ by far more in language use than mere
pitch. (In fact, even their pitch differences are more pronounced
than can be anatomically explained.) Such sociosymbolic aspects
of language use serve an emblernatic function: they identify the
speaker as belonging to a particular group, or having a particular
social identity.
In many societie s some of the most important of these sociolinguistic divisions are associated with differences in social prestige,
wealth, and power. Bankers clearly do not talk the same as busboys, and professors don't sound like plumbers. They signal the
social differences between them by features of their phonology,
grammar, and lexical choice, just as they do extralinguistically by
READINGS 93
their choices in clothing, cars, and so on. The social groups at

issue here may be harder to define than groups like .inen' and
'women,' but they are just as real. They are the divisions of a
society along lines of s ocrer cLASS.
based on status and power in a
eople are respected and deferred

conversely, looked down on or
social and material resources a
person can command, the ability (and social right) to make deci_
sions and influence events.
way tbey
innate?
How do differences in speech sbou group membership?

Do you dgree that tbe choice of phonology, grctmmct, and
lex_
icon is Like tbe choice of clotbing or a cdr? If not, bou is it dif_
ferent?
ally well explained as the result of differences in sta_

wer?
TextL2
EDMUND A. AUNGER: 'Regional, national and official
languages in Belgium' in International
Journal of the
Sociology of Language ro4, 993,pages
44-5
French, and
Dutch, German, or French. This policn characterized more by
neglect than by outright repression, has contributed to the decline
of the regional languages. There have been occasional, but
'Walloon, Picard, and Lorrain

belated, exceptions to this pattern.
may now be taught in the French language schools-by virtue of a
r98z decision of the Executive of the French Community-but,
in fact, very few schools have ever offered courses in these languages. In a neighboring state, the Grand Duchy of Luxembourg'
Luxemburgish is recognized as the national language, and it is the
language of instruction for the first year of elementary school.
Belgium's linguistic identity has been shaped by its recognition
of three standardized languages-French, Dutch, and Germanas national languages. This recognition, dating from r866, has
had a profound effect on the choice of official languages. After
decades of protracted debate, the Belgian State has recognized
French, Dutch, and, to a limited extent, German as its official languages. French and Dutch are the languages of the Belgian parliament, and all laws have been published in these two languages
since r898. Public employees in the central administration are
listed on a French roll or a Dutch roll, and all government positions are classed by language. Justice is conducted in the language
of the region, and judges must be proficient in that language.
Similarly, school instruction is in the language of the regionFrench, Dutch, or German. These, therefore, are Belgium's public
tongues, the languages of political power and public affairs.
Belgium?
le speak in the
dialect suggested in Text r o aboue?

three official languages ?
Text 13
JANET HoLMES : An Introduction to Sociolinguistics.

Longman 1992, pages 17 _z
Among other differences between the speech of men and
Luomen, one that bas been the source of considerable debate is
94
READINGS
READINGS
95
the fact tbat uomen tend to use more standard

forms of lan_
guage tban do men with tbe same educational leuel.
Some linguists have suggested that women use more standard

speech forms than men because they are more status-conscious
than men. The claim is that women are more aware of the fact
status groups than the older women who were working at home.
This evidence thus throws some doubt on suggestions that

women without paid employment are more likely to use standard
forms than those with jobs, and so indirectly questions the social
status explanation for women's speech patterns.
tr
Do you think tbat any of the suggested explanations haue anything to do uith biological differences behueen men and
uomen?
conscious than men?

The fact that women interviewed in New york and in Norwich
use more standard forms tban tbose
uho stayed home?
Ghapter 5
Text 14
indirect evidence which throws doubt on this expranation. It is
cEoRGE sAUNDERS: Bilingual Cbildren: Guidance for the

Family. Multilingual Matters Ltd ry83,page 9
likely to claim high social status by using more standard forms.

This implies that women in the paid workforce should use fewer
standard forms than women working in the home. But the little
For some people, a bilingual is a person with equal control

of two langwages, for otbers, it is someone u.,ith full control of
one and limited control of a second. Giuen this uncertainty,
suggested that womcn who are not in paid employment are most
evidence that we have, in fact suggests that just the opposite may
be true. An American study compared the speech o]-women in
s and hotels, for instance,
the home. Those in paid
than those working in the

rst group of women were
interacting with people who used more standard forms, and this
interaction had its effect on their own usage. By contrast, the
women who stayed home interacted mainly with each other, and
this reinforced their preference for vernacular forms.
Exactly the same pattern was found in an Irish working_tlass
community. The younger women in Ballymacarrett, a suburb of
Belfast, found work outside the community, and used a much

higher pecentage of linguistic features associated with high
g6 READINGS
the issue of classification is imprtrlant.
How proficient does a person have to be, then, to be classed as
bilingual? Haugen, an American linguist who has worked extensively in the field of bilingualism, suggests that bilingualism
begins 'at the point where a speaker of one language can produce
complete, meaningful utterances in the other language'. Diebold
considers that a type of bilingualism has even commenced when a
person begins to understand utterances in a second language
without being able to utter anything him- or herself.
Bilingualism, therefore, simply means having two languages
(and bilingualism is often used in the literature to mean the same
as multilingualism, that is, having more than two languages).
Bilinguals can be ranged along a continuum from the rare equilingual who is indistinguishable from a native speaker in both
READINGS
97
languages at one end to the person who has just begun to acquire
a second language at the other end. They are all bilinguals, but
possessing different degrees of bilingualism. A monolingual (also
called a unilingual or monoglot) is thus someone who knows only
one language. (In this book monolingual is used, for the sake of
convenience, to refer also to persons near the extremity of the

bilingualism continuum, namely to persons who are minimally
bilingual, that is, who have very little proficiency in more than
one language.)
A bilingual's
degree of bilingualism can be assessed in the four
skills of listening comprehension, speaking, reading comprehension and writing. There are many possible combinations of abilities in these skills. Many children of immigranrs, for instance,
possess all four skills solely in the official language of their country of residence (for example, English in Australia), whilst they
may be able to understand only the spoken form of their parents'
language (for example, Italian) and barely be able to speak it.
Haas would class such children as'receiving oral bilinguals', since
they are bilingual only in receiving the spoken form of two
languages, in listening comprehension. Someone who is bilingual
in all four skills would, using this system, be classified as a 'receiving sending oral visual bilingual'. Again, within each skill there
could be differing abilities in each language, for example, an
English Chinese bilingual educated through English could be
much more proficient at writing English than Chinese, whereas
his spoken Chinese could bc better than his spoken English, and
so on.
The term'balanced bilingual' is frequently encountered in the

literature on bilingualism. Whilst some writers (for erample,
Haugen) use it as a synonym of equilingual, most researchers use

'balanced bilingualism' in a different sense which does not imply
perfect mastery of both languages (for example , Peal & Lambert).
Balanced bilinguals in this sense are bilinguals who are roughly
equally skilled in their two languages, i.e. a balance exists
between the two.
a foreign language at scbool a bilingual? Is it useful to
How might you determine if someone is bilingual?
98 READINGS
do so?
How
migbt someone become

common is tbis likely to be?
a 'balanced bilingwal'?
Horu
Text 15
cARo L MyERs- sc orroN : Social Motiuations for

Codeswitching: Euidence from Africa. Clarendon Press r993,
pages
r-2
The mixing of two languages is often seen
ness, and as the cause
cts
the resuh of lazi-
of ueakening of a
language.
Codeswitcbing is uery common ctmong bilinguals, and MyersScotton suggests that it serues importdnt social functions.
Everyday conversations in two languages are the subiect-matter
of this volume. All over the world bilinguals carry on such conversations, from Hispanics in Texas, who may alternate between
Spanish and English in informal in-group conversations, to'West
Africans, who may use both Wolof and French in the same conversation on the streets of Dakar, Senegal, to residents in the Swiss
capital of Berne, who may change back and forth between Swiss
German and French in a service exchange. Contrary to some popular beliefs, such conversations are not mainly a transitional stage
in a language shift from dominance in one language to another. It
is true that many immigrants who are in the process of language
shift do engage in codeswitching, but this form of conversation is
also part of the daily lives of many 'stable' bilingual populations
as well. Neither is codeswitching only the vehicle of social groups
on the socioeconomic 'margins' of society; for example, in every
nation, successful business people and professionals who happen
to have a different home language from the language dominant in
the society where they live frequently engage in codeswitching
(between these two languages) with friends and business associates who share their linguistic repertoires. Consider P.unjabiorigin physicians in Birmingham, England, Lebanese-origin
businessmen in Dearborn, Michigan, or Chinese-origin corporate
executives in Singapore.
Codeswitchiag is the term used to identify alternations of linguistic varieties within the same conversation.'!7hile some prefer
to discuss such alternation under two terms, employrng codemixing as well as codeswitcbing, the single term codeswitcbing rs
READINGS
99
used here. ... The linguistic varieties participating in codeswitching may be different languages, or dialects or styles of the same
language.
proceedings in the higher courts. Besides being the language of

literate professions (law, medicrne, engrneenng, accountancy,
etc.) as well as the nation's library and archival language, English
is also a medium of interethnic and international communication.
a bilingual codeswitcbes when talking to a monolingual?
How might codeswitcbing account for all the Norman Frencb

words that are now part of Englisb?
Ghapter 6
Text 16
F. NIYI KINNAS o: 'Vernacular literacy in modern Nigeria'

in International lournal of the Sociology of Languctge r19,
r996, pages 46-7
When a country has a uery large number of languages spoken

in it, eacb comes to fill a different role in the national linguistic
repertorre
Nigeria's complicated sociolinguistic landscape reveals three
ma;'or
three e
a relat
4oo indigenous languages; (z)

ic, English, and French; and (3 )
mely, Pidgin English. The lant
guages form a hierarchy, charactertzed by a .six language-formula,' in which the languages are stratified according to the
degree of official recognition, prestige, range and context of use,
extent of development, population of speakers, and so forth. ...
ln keeping with its colonial legacy, it remains as the nation's de

facto'officral' language, being the language of the constitution
and of legislation (Constitution of tbe Federal Republic of

Nigeria [henceforth Constitution] r989, Section 53). English,
therefore, carries the highest symbolic value. ... As the language of

mainstream institutions and activities, it is the most desirable language that parents want their children to learn in school. ...
The four hundred or so indigenous languages occupy various
positions below English, with three-Hausa, Igbo, and Yorubabeing officially recognized as 'major' languages, to be learned in
school in addition to the mother tongue and English, and
accorded (at least in principle) almost equal status with English in
that they are recognized in the constitution as legislative lan-
guages, to be so used 'when adequate arrangements have been

made therefor' (Constitution, Section 53). The three languages
are also used in the communications media for network news
brcladcasts on radio and television, for newspaper publication,
and for environmental print (billboards, notices, graffiti, etc.). ...
Of all Nigerian languages, these three are the most developed and
have the longest history of involvement in literacy use in and outside the mainstream.
The most outstanding feature of Nigeria's language-planning
model, therefore, is the adoption of a system of stratified rational-
tbis is so?
while the right of other languages ro exisr and be developed is also

respected. This favorable attitude to language rights underlies the
language education policy and, especially, the attitude toward the
development of local languages and their use in literacy.
English is clearly at the rop of the language hierarchy, carrying
the heaviest functional load as the language of much of administration, education, mass communication, commerce, and judicial
groups?
ization in which one or more majority languages are accorded

some special status at federal, regional, state, and local levels,
IOO
READINGS
How is the notion of language status related to the notion of

symbolic ualue (see Text
l?
READINGS
TOI
V
Text 17
P ETE R u H rHu s Lr'n : P idgin and
Blackwell r986, pages 84-5
Cr eole
Linguistics.
Pidgins start off filling important but limited roles as lingwe

franche (tbe plural of lingua franca) or languages of wider
communication, bwt often deuelop a growing range of functions and increasing complexity.
Continued use of a stable pidgin by speakers from many different

language backgrounds, parricularly when transported from the
plantation to a larger multilingual society such as that of Papua
New Cuinea or Vanuatu, led to further functional and structural
expansion. On the one hand, the permanency of contacts
required the encoding of personal feelings; on the orher, social
norms of politeness such as small talk were needed. The addition
of the expressive function typically goes hand in hand with a
widening of domain. In many instances, rhe use of a pidgin for
religious purposes triggered this new functional use. Phatic communion (a term introduced by Malinowski ry4) is associated
with new domains (meetings, social gatherings, etc.) as well as
new media. In using a pidgin for telephone conversations, for
instance, certain devices referring to this channel of communication and its functioning are required.
It has to be kept in mind, at all times, that pidgins are second
languages and that any new functions have to be seen in relation
to the functional use of a speaker's first language. The use of a
jargon-that is, a language for propositional and heuristic functions only-typically has little effect on a speaker's vernacular. It
is simply added to it. The same is true with beginning stabilizing
pidgins and many older stable pidgins: they are additional ro traditional vernaculars, which continue to remain intact and to be
used in all functions relevant to the traditional society in which
they are spoken.
'S7ith
the ongoing functional and structural expansion of a
pidgin, however, its relationship with the speaker's first language

tends to become changed. Instead of being added to, traditional
languages tend to be replaced in an increasing number of domains
and functions. This is particularly striking with the expressive
function in the domains of religion and abuse. Thus the religious
oz READINGS
cxperiences transmitted by expatriate mission bodies to the

indigenous population tend to be incompatible with traditional
modes of religious expression and pidgins, or mission lingue
franche are preferred. Prayer, services and discussion of religious
matters thus are associated with speakers' second languages. The
same is true of insults, expletives and other forms of strong language. For Tok Pisin among the Kwoma, Reed observed:
We found that youngsters not only counted and sang in pidgin
but also used it in the new game of football-especially in

angry altercations. Their own language was not lacking
in terms of abuse, but those in pidgin were preferred. (Reed
194:286)
Text 18
pETER TRUDGILL: On Dialect: Social and Geograpbical
Perspectiues. Blackwell r983, page
rz7
You can't be an X unless you speak Xisb. Replace X by the

name of dn ethnic group and Xish by the name of the associated language, and you haue ct stdtement ruith wbich many
will agree. But Trudgill points owt that the situation is more
complex.
It is well known that language can act as an important defining
characteristic of ethnic group membership, and in many communities the link between language and ethnicity is strong, and obvious. It also has to be recognized, however, that a simple equation
of ethnic and language group membership is far from adequate.
There are, obviously, many examples of situations where a

separate ethnic identity is maintained even though a distinctive
language has been lost (see Fishman, ry68\. Examples include the
British Jewish community, and the Catholic community in

Northern Ireland. (In some cases of this sort, dialect or accent
differences within a single language may serve as identifying characteristics-Trudgill,
r9
7 4b, chapter
3 .)
More puzzling, perhaps,
READINGS
ro3
are cases where language distinctiveness appears not to be accompanied by any awareness of a separare ethnic identity. To what
extent, for example, do Gaelic-speaking Scots form a separate
ethnic group within Scotland?
It is not easy to determine what factors are involved in the
establishment of these varying attitudes to language and ethnic
group membership.
are distinct despite language similarity, or uhere language dif-
ferences do not mark ethnic differences?
group besides language?

Text 19
pETER LADEFocED: 'Another view of endangered

languages' n Language 68/ 4, t99z, pages 8 r o-l r
In a paper published in early r992, Ken Hale and a nwmber of

colleagues, all of them actiue in the study of American Indian
languages, drew attention to the fact that a large number of

the 6ooo or so languages still being spoken are in seriows danger of disappearing in the next roo years or less. They took the
position tbat linguists should belp preserue these endangered
languages. Peter Lddefoged expresses a somewhat different
urew.
So now let me challenge directly the assumption of these papers

that different languages, and even different cultures, always
ought to be preserved. It is paternalistic of linguists to assume that

they know what is best for the community. One can be a responsible linguist and yet regard the loss of a particular language, or
even a whole group of languages, as far from a 'catastrophic

destruction'(Hale et aI. r99z:7). Statements such as'just as rhe
extinction of any animal species diminishes our world, so does the
extinction of any language' (Hale et aI. r99z: 8 ) are appeals to our
emotions, not to our reason. The case for studying endangered
languages is very strong on linguistic grounds. It is often enormously strong on humanitarian grounds as well. But it would be
self-serving of linguists to pretend that this is always the case. \7e
ro4
READTNGS
rnust be wary of arguments based on political considerations. Of

course I am no more in favor of genocide or repression of minorities than I am of people dying of tuberculosis or starving through
'$e
should always be sensitive to the concerns of the
ignorance.
people whose language we are studying. But we should not

assume that we know what is best for them.
'We
may also note that human societies are not like animal
species. The world is remarkably resilient in the preservation of

diversity; different cultures are always dying while new ones
arise. They may not be based on ethnicity or language, but the differences remain. Societies will always produce subgroups as variecl as computer nerds, valley girls, and drug pushers, who think
and behave in different ways. In the popular view the world is
becoming more homogeneous' but that may be because we are
not seeing the new difference that are arising. Consider two
groups of Bushmen, the Zhulosi and the !X who speak mutully unlntelllgible languages belonging to different subgroups of
the Khoisan family, but otherwise behave in very similar ways'
Are these two groups more culturally diverse than Appalachian
coalminers, Iowa farmers and Beverly Hills lawyers? As a linguist, I am of course saddened by the vast amount of linguistic
and cultural knowledge that is disappearing, and I am delighted
that the National Science Foundation has sponsored our UCLA
research, in which we try to record for posterity the phonetic
structures of some of the languages that will not be around much
longer. But it is not for me to assess the virtues of programs for
lang.rage preservation versus those of competitive programs for
tuberculosis eradication, which may also need government funds'

In this changing world, the task of the linguist is to lay out the
facts concerning a given linguistic situation. ...
Last summer I was working on Dahalo, a rapidly dying
Cushitic language, spoken by a few hundred people in a rural district of Kenya. I asked one of our consultants whether his teenaged sons spoke Dahalo. 'No'' he said. 'They can still hear it, but
they cannot speak it. They speak only Swahili.' FIe was smiling
when he said it, and did not seem to regret it. He was proud that
\X/ho
his sons had been to school, and knew things that he did not'
am I to say that he was wrong?
READINGS
ro5
languages? Does he think tbat these reasons arequally Jalicl?

to his or her cbildren?
\yhyl
Chapter 7
Text 2O
L.
z3r_z
Colon
mong the most important
spite of the predominance of the English language, has been a

nation of induced diversity. Diversity was an inescapable consequence whatever attempts were made to ignore it; so that the
American problem has always been to ensure unity while accommodating diversity. On the other hand, Russia and now the Soviet
Union has always set out to acquire diversity. It was always able
to enforce Russian political unity and prepare for linguistic dominance by its military strength and by the cult of the Orthodox
Church, and now by the promotion of a single party, a uniform
ideology. From its point of view the problem has been to

acknowledge (without sustaining) the inescapable linguistic
diversity within a uniform political sysrem. Britain artempted to
suppress linguistic diversity, and the Soviet Union began by seeking to preserve it. The different processes of colonization have
r:esulted in different emphases on aspects of the intractable problem of the relation of linguistic unity and diversity, which is at the
root of language spread.
i!i,j,earenotsimpleforces,as
language policy?
linguistic unity and diuersity to be intractable? Do you knou

of anyuhere that it has been solued?
In the neu independent countries like Ukraine and Estonia

tbere are now strong mouements for establishing the dominance of the national language ouer Russian. Would tbis
deuelopment haue surprised Lewis?
Text 21
RoBERT L. coopER: Language Planning and Language

Cbange. Cambridge University Press r989, pages
rr8-r9
\y{hile tbe notion of actiue policies to change the status of ldnguages is not new, the study of language planning started witb
tbe problems of netuly independent states in the post-Second
'World
War period, many of which had to cboose between a
number of indigenous languages and one or more colonial
spread to the European west and sought out the diverse Asian
narions of the south and east. To all intents the united states, in
ro6
READINGs
languages. Canada
is one country u,bere, more tban huo
hundred years after tbe political situation uas establisbed tbat

READINGS
IOA
brought French and English into conflict, tension continues to
surr ound language status.
in the hands of the
Anglophone minority and foreign

Anglophones. Although Francophones have entered middle man_
overwhelmingly domarge business firms.
ranks of top managerial incentives to learn

English have inspired non-Anglophone, non-Francophone immi_
grants (termed 'Allophones') to learn English rather than French
promotion and defense of language maintenance. They also provide clear recognition that status planning refers ultimately to the
status of those who use the language.
Hotu didBill
tot strengthentbis?
Text 4) and the usefulness of a language?
Tert22
cALVIN vELTMAN: 'The English language in Quebec
r94o-r99o'inJoshua A. Fishman, Alma Rubal-Lopez, and

(eds.): Posr-lmperial Englisb: Status
Changes in Former Britisb and American Colonies,
rglo-r9oo. Mouton de Gruyter r996,pages z3z-j
Andrew \V. Conrad
Summing up the effect of the language status planning undertaken in Quebec (see Text zr), Veltman looks at tbe situation
more than a decade after the two Bills were enacted.
The demographic situation of the English-speaking group has

also undergone marked changes, changes which make it difficult
to imagine that the trends which we have examined in this chapter
can be reversed or even arrested. First of all, the French language
group is now attracting about two immigrants in every three

among third language groups, leaving the English language group
with substantially fewer new recruits than in the past. This has
government contracts. In r977, the provincial government, under
the Parti Qubcois,adopted Bill ror, the 'Charter of the French
implementing these provisions, including coercive measures to

Bill ror provide an unusually clear
recognition of the importance of commercial incentives for the
ensure compliance. BIl zz and
ro8
READINGS
already had a marked effect on the English language school system and will undoubtedly reduce the clientele of all the institutions of the English-speaking community in the future.
Second, there has been a substantial reduction in the number of
English-speaking immigrants from the United States, the United
Kingdom, and frpm English Canada itself over the past thirty
years. Quebec is ho longer seen as a preferential destination for
English-speaking immigrants. Further, recent data suggest that at
least ro percent of anglophone immigrants from outside Canada
during the decade of the 19Sos have integrated into the French
language community, testimony indeed to the new attractive
power ofthe French language.
READINGS
ro9
In addition'it would seem

likely that the rrend roward
greater
workpl'..
*il
coninue' While the Eng|ish
,the
language schools t
.n,J. o;.;;.
francisation of
o th.
sECTroN
d..'','ti"u.
e ; d*:J :T;T
:;T Ti,:T..'i::ffi:
rional srru*ure,and
r..ialntial neighborhoods makes
-r"
"t
it
difficult for anglopho"..r'ia.."]l
t
be
hi
comfortable
..k;
l""r,'
e;'.".-r
ro
""J,rglr
.'Lorrr.ou.rrtly, the pressure
ro
ve Quebec
-r'""
,r'"y
ii'r'
well,
'r'.li
rive.the English language
'es the school system and
eration of children'
potr
,tt..o--.rn'liy
In short, whil
enloy a greater
p'''.t, d;;g
from a ;-;;,
extremelypain
group of its
e in Quebec continue to
|s' and services than do
Canada, the transition
Y grouP has been
an
eople can still be served
m E"gfl.fr i.r
contrnue to "l
live
French is hardl
ernment services
to English scho
life; most anglophones

l envrronments where
rough they can obtain gov_
h and their children can go
suffi
cient," o[T:iT:
;T'*l^",. j
king people is nor as much

appreci"i.a
escape from the pressures
i t.".r' ",
"r'a
.cularly in the workplace, _""uirr*fi.f,
e view migration to
an Englisir ,;;;il;r#i.";,
an rmporranr option. This choice
has been ,t. p.in.ip?i.r"r..f
",
rhe decline of rhe englirh_.p.nk;;;
;
and it is likely to
presage significant decline
i,, th. f,rtur.lroup'
" y:::T' ::;f!:ct

Did the Bills lead
Hou
of the Bills on tbe English-speaking

popu-
to greater integration of
tbe huo populations?
does the. Q,uebec example

shou tbe potential and limitations of political efforts
n
"iorg;)'rociolinguistics situation?
TIO READINGS
References
talize on their educatio"l

dies continue to show ,ir", ".tl*.-.rr..
-".ry'yotrrrg
vho rep.ortedly speak French
indicate that
.do.utlo.'. NJt
>
The references which follow can be classified into

introductory
level (marked rnl), more advanced and consequently
more tech'_
nical (marked rll), and specialized,, very de-"r'dr'g (--k"l
rrr).
Chapter 1

!!!
RALpH
FAS o LD: The

,Sociolinguistics of Langwage:
Introduction to Sociolinguistics, II. Blackwell r9!o
Among a number of useful general introductory

textbooks that
now cover the field, this book stands out for it,
effect of social class on patterns of language
use.
.orrerag. of tlr.
t!!
IoSHUA A. FrsHM_AN: Sociolinguistics: A Brief
Introduction Newbury House publishers r97o
l'ritten
as the field of sociolinguisrics was taking

shape, this short
book by one of the leading tlreorists in the fiel
su;ariz; th;
main concerns that are still debated. It is written

clearly and introduces many basic concepts.
!Il
JoSHUA . FIsHMAN: ,Putting the ..socio'' back into the
sociolinguistic enterprise, in Inlernational ournal

of the
J
Sociology of Langwage 9z, 199) pages tz7_78
REFERENCES
III
More than anyone else, Joshua Fishman has worried about the
difficulties thar the field has had in maintaining its rerationships

with one of its parent disciplines. This paper dicusses th. ,'nt,.r..
of the problem and argues for a closer,.l",lon with sociology.
ALLEN D. cRTMSHAw (ed.): ConflictTalk: Sociolinguistic
Inuestigations of Arguments in Conuersation. Cambridge

University Press r99o
This collection of papers shows how the field of conversational
-t!!
JANET HoLMES : An Introduction to Sociolinguistics.

Longman r99z
This is
, lt!
another introductory textbook, with good general coverage and a strong section on language and gender.
a
RoNALD r /ARDHAUGrr: An Introduction to Sociolinguistics

(znd edn.) Blackwell r99z
This standard introduction to the field is now in its second edi-
tion. It covers the main issues clearly and soundly.
Ghapter 2
The ethnography of speaking and the structure
of conversation
lIl
ALAN BELL: 'Language style as audience design' in Language

in Society r3lz, r984,pages r45-zo4
The level of formality is generally considered a result of how
much attentior n ,p."L.. is payirrg io his or her speech. Basing
his
analysis first on radio announcers, Bell suggests that it can

b."b.tter understood by considering the .p."k.i, i.rtended audience.
TIT
ERVTNG GoFFMAN:,Replies and responses, in Language in
analysis has developed. It concentrates on disagreement.

It!
JoHN J. cuMpERZ,
ed.: Languages and Social
Cambridge University Press r98z
ldentity.
This is a collection of papers that consider, from an ethnographic

perspective, the way language is used to communicate a speaker's
choice of identity.
ll!
DELL Hr'ME s: Foundations in Sociolinguistics: An

Ethnographic Approach. University of Pennsylvania Press
r97 4
This is an approach to sociolinguistics from the point of view of
an anthropological linguist who proposed the key notion of communlcatlve competence.
lr!
TAMAR KATRTEL: 'Brogez: Ritual and strategy in Israeli

children's conflicts' rn Langwage in Society t414, r98 5,
pages 467-9o
This article is a good example of the way an ethnographer investigates the language use patterns of young children. It analyses the
way that young Israeli children formalize conflicts and disagreements, and how they make peace after such fights.
Society 5/3, r976,pages ;-57_3r3
This pioneering essay investigated the structure of telephone con_

versations and became the model for studies of conversational
structure.
II2
REFERENCES
REFERENCES
II3
!F
Chapter 3
Locating variation in speech
lt!
J.K. cHAMB
r98o
Bn s:
Dialectology. Cambridge University press
A short and useful introduction to the field, this book summarizes

basic principles.
tr!
CHARLES A. FERGUSON : Sociolinguistic perspectiues:
Pdpers on Language and Society, 19 59_994.

Oxford University Press r996
This is a collection of brilliant pioneering observations on the

nature of language in use by a scholar who has had great influence
on the development of sociolingulstrcs.
LEy MrLRoy: Obseruing and Analyzing Natural

Langwage. Blackwell r9 8 7
This is a detailed study of the narure of social variation in lanLES
guage use and structure.
IseslL cRoucH:'The
question of
'-89-94
There are many stereotypes about the way that women use language. This is an empirical study of the use of tag questions that
shows that often the stere otypes are wrong.
l!!
JANEr HoLMES :'W men, Men dnd Politeness. Longman

19g 5
Are men or women more polite? If so, why? This book investi-
gates this key question about differences between women's and

men's language.
Ro B IN LAKo FF: 'Language and woman's place' in Language
in Society zlr, r973, pages 45-8o
This is a pioneering essay on women's language. I7hile not all of

its generalizations have held up, it was influential in opening a
new field of study.
!!!
pErER TRUDGTLL: The Dialects of English. Blackwell r99o

This general description of the dialecrs of English is written with
full appreciation of the relevance of sociolinguistics.
Chapter 4
Styles, gender, and socialclass
ttt
L LIAM LAB o v : So ciolingui

Pennsylvania Press r97z
wI
sti c P
atterns. University of
This book, by one of the most important and influential scholars

in the field, studies the relation between social and linguistic
structure, showing how by adding social information one can
account more fully for language variation.
lD!
lt!
ersnnr
6TLMAN: 'The pronouns of

power and solidarity' in Thomas A. Sebeok (ed.): Style in
Language. MIT Press r96o, pages ;-53-76
This is a pioneering study that started a whole series of investigations into systems of address.
TI4 REFERENCES
BETTv Lou DUBoIs and
tag questions in women's speech: they don't really use more

of them, do they?' rn Language in Society 413, r97 5, pages
t!!
Ilt
RocER BRo\ /N
ll!
and
DEBoRAH TANNEN: Gender and Discowrse. Oxford

University Press r996
This is a collection of essays on gender and language which provides a good picture of current concerns.
REFERENCES
II5
Chapter 5
This book present a theory of language choice, with extensive

illustration from codeswitching in several African contexrs.

TT!
rrE H o F FMAN : An lntro

Longman r99r
c HAR L o
du
ction to
Chapter 6
ilingualism.
This book surveys the major issues concerning the nature of bilingualism and the studies that have been made of the phenomenon.
ITT
JoHN EDrvARos :'Monolingualism, bilingualism and
identity: lessons and insights from recent Canadian
experience' n Current Issues in Language and Society
rt!
E. cLvN LEwrs : Multilingualism in
MoutonrgTz
The former Soviet Union was marked by a complex pattern of

rnultilingualism. This book describes it, and shows the changes in
]overnmental attitude and policy through the Soviet period. It
zlt,
shows linguistic imperialism in an extreme form.
tgg5,pages 5-38
Canada has been the location of many studies of bilingualism.
This article shows the way language choice intersects with social
and political identity.
ltr
JosHUA A. FTsHMAN (ed.): Aduances in the Study of

Societal Muhilingualism. Mouton r978
Tl-ris is a collection of papers dealing with
ll!
ous parts of the world.
CHARLEs A. FERGUsoN: 'Diglossia' n'Word I5,1959,

pages 325-4o
The phenomenon of two varieties of the same language with
TTT
This is
ltt
JoSHUA A. FTSHMAN, RoBERT L. coopER, and noxaNR
v'e,: Bilingualism in tbe Barrio. Research Center for the
Language Sciences, Indiana University r97r
its methods and theories. It is a detailed study of a urban bilingual
community, showing how an immigrant minority interacts with

its new environment.
trt
cARoL MyERS-scorroN: Social Motiuations for
multilingualism in vari-
EINAR HAUGEN: 'Language and immigration'in NorwegianAmerican Stwdies and Records ro, t938,pages r-43
separate sets of functions was first set out in this paper.
A pioneering study of a stable bilingual communiry, this book had

a major impact on the development of the sociology of language,
tbe Souiet [Jnion.
a pioneering study of
Norwegian in America showing the
effects of immigration on patterns of language use.
I
{
lr!
JoSHUA A. FTsHMAN: Reuersing Language Shift:
Tbeoretical and Empirical Foundations of Assistance to
Threatened Languages, Multilingual Marters Ltd t99r
Social and demographic changes have led his century to the
growth of major languages and a strong tendency for less powerful languages to be lost. This study looks ar recenr efforrs by
speakers of these endangered languages to resist language shift,
and presents a theoretical model of how such resistance can work.
Codestuitching: Euidence from Africa. Oxford Studies in

Language Contact, Clarendon Press r993
rt6
REFERENCES
REFERENCES
TT7
It!
CHRTSTTNA BRATT pAULSToN (ed.): International
Handbook of Bilingualism and Bilingual Education.
Creenwood Press r988
This volume includes descriptions of multilingualism and of the

associated bilingual and multilingual educational patterns in a
wide variery of countries.
TID
BERNARD sPoLSKY and RospRt r. cooppR: 7e
Languages of Jerusalem. Clarendon press r99 r

This is a study of the history and currenr sociolinguistic pattern of
a multilingual city.
Chapter 7
T!D
CoLrN BAKER: Foundations of Bilingual Education and
Bilingualism. Multilingual Maters Ltd ry93
This book summarizes the possible educational responses to the

existence of multilingual societies.
Il!
RoBERT L. coopER: Language planning and Social

Change. Cambridge University press r9g9
This is a highly readable survey of the field of language planning
and policy which sers out clearly the main issues in thefield.
tIt
JoSHUA A FISHMAN, ALMA RUBAL_LopEZ,andANDREw
w. coNRAD (eds.): Posr-imperial English: Status Changes in
Former British and American Colonies, r940_r99o.Mouron
deGruyter 1996
This book includes a number of empirical studies of the spread of
English in countries that were at one time British or American
colonies. It provides a basis for answering the question of how
and why English has spread.
rr8 REFERENCES
t!!
FRED GENE sEE; Learning throwgb Ttuo Languages: Studies

Immersion and Bilingual Educatictn. Newbury House
Publishers r987
Including mainly empirical studies of bilingual educarion in
Canada, this book investigates an educational policy in which
ctf
speakers of one language start their education in another.

ltI
TIN HICKEv and 1nNNv wILLIAMS (eds.): Language,

Education and Society in a Cbanging'World.
IRAAL/Multilingual Maters Ltd ry96
This is a collection of papers on various aspecrs of educational linuuistics' showing current attempts to deal with multilingualism.
TT!
RTcHARD D. LAMBERT
(ed.): Language planning
around
tbe World: Contexts and Systematic Cbange. National
Foreign Language Center r994

This book collects a number of articles dealing with language and
language education policy in various parts of the world.
lt!
B ERT pHr L Lr p S oN : Linguistic
Oxford University Press r99z
Ro
Imp erialism.
This book presents evidence and arguments in support of the case

that the spread of English is largely the resulr ofconscious plan_
r ing on the part of those English-speaking countrie s that ,t"r.d to
gain by its international use.
Ilt
TovE S KUTNAB B - KAN GA s : Bilingualism or Not: Th e

Education of Minorities. Trans. L. Malmberg and D. Crane.
Multilingual Mafters Ltd r98r
This book documents a large number of cases where educational
and other policies have forced minorities to give up their own
languages.
REFERENCES TI9
SECTTON 4
Glossary
Page references to Section
r, Survey, are given at the end ofeach entry.
accommodation Adjusting one's speech to converge with or

diverge from the speech of one's interlocutor. 133,421
of the masculine (man, he) to

include the feminine. See accommodation theory. [38]
audience design Adjusting one's speech to be similar to that of a
anthropocentric speech Use
real or imagined listener. t33,
baby talk
411
egister regarded as appropriate
for addressing
babies. taal
balanced bilingualism Very strong (almost equal) command of

two languages.
[45]
bilingual A person who has some functional abilitv in a second

language. [45]
Black English The vernacular variety used by Afro-Americans in

the US, sometimes called Ebonics or Afro-American
Vernacular English.
1os1
borrowing The integration of a word from one language into

another. [49]
cant Thieves' and underworld jargons. [34]

clandestine recording Collecting data surreptitiously. [10]
code switching Changing from language to language in the
midst of an utterance.
[49]
competence Underlying knowledge of a languagel cf. performance. [48]
compound bilinguals Bilinguals who learned one language after

(and so through) another; cf. co-ordinate bilinguals. [48]
GLOSSARY
ZI
'
conversational interchange The basic unit of the spoken language, where two or more speakers take turns to speak. See
grammatical class; (z) a term for socially marked
sexual variation. 136l
turn-taking. [16]
generic masculine Use of the grammatically masculine form

(mdn, he) to include the feminine. See anthropocentric speech.
pound bilinguals. [48]
genre Kind of speech event, or kind of literary form.
co-ordinate bilinguals Bilinguals who have learned each language in separate contexts and so keep them distinct; cf. comcorpus planning An attempt to fix or modify the structure (writing system, spelling, grammar, vocabulary) of a language. 166l
creole A pidgin once it has native speakers. See creolization. [62]

creolization Changes in a pidgin as a result of adding vitatity or
mother tongue speakers.
language over time. See histori-
cal linguisticsl cf. synchronic variation. [4]
dialect A variety of a language used recognizably in a specific

region or (a social dialect) by a specific social class.
127l
dialectology The search for spatially and geographically determined differences in various aspects of language. [28]
diglossia A situation when two distinct varieties of the same

language are used, side by side, for two different sets of functlons.
[63]
domain Typical social situation with three defining characteristics: place, role-relationship, and topic. [34]
endangered language Language that is no longer being passed
on to children as a native language, but is spoken by a contracting and aging group ofadults. Seetanguage loss. [55]
ethnographic observation The recording of natural speech
events by a participant-observer. [12]
of speaking Sometimes also called the ethnography of communication, an anthropological approach to the
study oflanguage use which is based on rhe actual observation
ofspeech. lr+1
ethnography
floor The right
to talk at any given moment in a conversation. See

turn-taking. [19]
formality Degree of care taken with speech. See styte. [31]
free variation The notion that the choice of variant is uncontrolled and without significance.
GLOSSARY
t38l
[39]
[15]
historical linguistics The study of language change over time.

See diachronic variation. [27]
hypercorrection Tendency to over-use socially desirable features

in careful speech and reading.
[62]
diachronic variation Changes in
IZZ
gende (r) A
[41-]
interference A feature of one language appearing when speaking or writing another. l49l
jargon Speech used by
a marked group of people such as a trade
or occupation; cf. cant, slang. [33]
language acquisition planning, or language education policy

Policy determining which languages should be taught and
learned.
1oz1
language conflict Situation where two or more languages compete for status. [55]
language contact Situation where two or more languages are
brought into contact by virtue of bilingualism. t49, 551
language diffusion policy Policy to spread a language to people
who do not speak it. loz, zs]
language loss A process by which speakers of a language slowly
stop using it, resulting in its dying out. See endangered language.
t5el
language loyalty The ability (or lack of it) of speakers of a language to stand up to the pressure of more powerful ones. [55]
language maintenance A situation where speakers continue to
use a language even when there is a new language available. [55]
language of wider communication Language chosen by speakers
of several different languages to communicate with each other.
t61l
language planning, or language policy Any effort to modify language form or use. [66]
GLOSSARY rz3
shift Changes in the degree of functional use from one

language to another. [55]
language standardization The imposing of a norm on the gram_
mar and pronunciation of a language. t67l
language
macrosocioringuistics or the sociology
of ranguage Area of
loss. [57]
revitalization The restoration to
a language of vitality. [56]
service encounters Occasions in which one person (client, customer, patient) seeks help from another (seller, clerk). 1rz1
of largon marked by its re;'ection of formal rules. See
cant. [35]
sociolinguistic study which concentrates on the use of a variety

and its social significance. [6]
metaphorica! switching Changing from one language to another
to signal a change in role relationship; cf. code switching.
[50]
microsociolinguistics Area of sociolinguistic study which concentrates on linguistic variables and their significance.
[6]
mixed code A variety with extensive code switching used by
bilingluats to talk ro each other. [49]
intended to collect speech samples. [10]

sociology of language See microsociolinguistics. [6]
solidarity Feeling of shared experience and common group mem-
networks Groups of people who communicate with each other
speech community (r) All the people who speak a single lan-
regularly. [26]
non-intrusive responses A research technique involving asking

strangers short non-personal questions. [11]
normativism The claim that there is one 'correct' version and all
variation is deviant. 133, 671
observer's paradox How can we observe the way people speak

when they are not being observed?
1s1
orthography l7riting (and spelling) system.
[72]
performance Linguistic behaviour (as opposed to linguistic
knowledge); cf. competence. [48]

pidgin A variety of language that is not a native language of any_
one, but is learned in contact situations; cf. creore, creolization.
[61]
politeness Recognition of other's rights in a social situation.

[19]
post-creole continuum !hen speakers of a creole or pidgin are
introduced, usually by education, to rhe standard language on
which the creole or pidgin was originally based. 1oz1
register Variety associated with a specific function. [33]

repertoire Collection of varieties or languages spoken by a group
of people. See speech community. [25]
rz4
reve sing language shift Efforts by a group to resist lanEuage
GLOSSARY
slang A kind
social stratification The study of class distinction in
speech.
t3el
sociolinguistic interview Technique of a recorded conversation
bership. [35]
guage (like English or French or Amharic); (z) a complex inrerlocking network of communication whose members share
knowledge about and attitudes towards language use. [24]
speech event Act of human communicarion. 1f+1

status planning Any attempt to set up laws or norms for when to
use a language. [66]
stereotypes Fixed and prejudicial paterns of thought about
kinds of people that are often mistaken. t28, g7)

styles Varieties of language used by an individual appropriate to
a
level of fomality. [4,31]
synchronic variation Variation at

diachronic variation. [4]
a single point of time;
cf.
terms of address Second-person pronouns, or names, or titles,

used when speaking to someone. [20]
turn-holders \ays of signalling that the speaker intends to continue after a break, intended to keep the floor. [19]
turn-taking Rules for determining who speaks when in

sational interchange.
a conver-
See floor. [18]
variable A linguistic feature (a sound, word, or grammatical

form) which has more than one variant, each of which has a
sociolinguistic significance.
[11]
GLOSSARY rz5
variant A form (word, sound, or grammatical form) which alternates with another under definable conditions. See variable. [6,
1,1,,291
variety A term used to denote any identifiable kind of language.

t6l
vernacular A variety used in common or popular speech; not the

educated or standard variety.
[8]
learned as a mother tongue).
1so1
Acknowledgements
vitality Natural intergenerational transmission (the language is
The author and publisher are grateful to the following for permission to reproduce extracts from copyright material:
Addison'!esley Longman for an extract from Janet Holmes: n
Longm an r 9 9 z).
I ntr o du cti on t o S o cio I inguisllcs
(
Blackwell Publishers for extracts from Ralph Fasold The
Sociolinguistics of Language (Blackwell r99o); James Milroy:
Linguistic Variation and Change: On tbe Historical
Sociolinguistics of English (t99z); Peter Miihlhusler: Pidgin and

Creole Linguistics (ry86); Peter TrudgilI: On Dialect: Social and
Geograpbical Perspectiues (r983); and Ronald !7ardhaugh: n
lntroduction to Sociolinguistics, znd
en.
(t99z).
Cambridge University Press for extracts from J. K. Chambers and

Peter Trudgill: Dialectology (r98o); Gregory R. Guy: 'Language
and social class' in Linguistics: Tbe Cambridge Suruey edited by
FrederickJ. Newmayer (r988); and Robert L. Cooper: Language
Planning and Language Change (tg9g).
Center for Applied Linguistics for extracts from'William Labov:
Tbe Social Stratification of English in I'tretu York City (196611' and
E. Glyn Lewis: 'Movements and agencies of language spread:
'il/ales and the Soviet Union compared' in Language Spread:

Studies of Diffusion and Langwage Change edited by Robert L.
Cooper (Indiana University Press, r98z).
Greenwood Publishing Group, Inc. for an extract from Penelope
Brown: 'How and why are women more polite: some evidence
from a Mayan community' in 'V/omen and Language in
Literature and Society edited by Sally McConnell-Ginet, Ruth

Borker, and Nelly Furman (Praeger r98o).
rz6 cLossARy
ACKNOWLEDGEMENTS r27
Indiana University Press for an extract from E. Glyn Lewis:

'Movements and agencies of language spread: Wales and the
Soviet Union compared' rn Language Spread: Studies of

Diffusion and Language Cbange edited by Robert L. Cooper
(r982\.
Linguistic Society of America for an extract from Peter Ladefoged:

'Another view of endangered languages' n Language 68 (t99z).
Macmillan Library Reference USA, a Simon and Schuster

Macmillan Company, for an extract from John J. Gurnperz:
'Linguistics: the speech community' in International Encyclopedia of the Social Sciences edited by David L. Sills, Vol. 9,
Copyright @ ry68 by Crowell Collier and Macmillan, Inc.
Copyright renewed @ r996.
Mouton de Gruyter, a division of !alter de Gruyter & Co,

Publishers for extracts from F. Niyi Akinnaso: 'Vernacular literacy in modern Nigeria' in International Journal of tbe Sociology
of Language, rr9 (!7. de Gruyter ryg6); Edmund A. Aunger:

'Regional, national and official languages in Belgium' in
International Journal of the Sociology of Language, ro4 ('S . de
Gruyter r993 ); Joshua A. Fishman: 'The sociology of language' in

Aduances in tbe Sociology of LanguageYol. r edited by Joshua
A. Fishman (Mouton r97r\; and Calvin Veltman: 'The English
language in Quebec t94o-r99o' rnPost-lmperial Englisb: Status
Cbanges in Former Britisb and American Colonies r94o-r99o,
edited byJoshua A. Fishman et al. (Mouton ry96).
Multilingual Matters Ltd for an extract from George Sanders:

Bilingual Cbildren: Guidance for the Family GgSZ).
Oxford University Press for an extract from Carol Myers-
Scotton: introduction to Social Motiuations for Codeswitcbing:

Euidence from Africa (Clarendon Press r993).
Routledge for an extract from Glyn silliam s:, Sociolinguistics: A
So ciolo gical Cr iti que (, g g r).
Despite every effort to trace and contact copyright holders before
publication, there are cases where this has not been possible. If
notified the publisher will be pleased to rectify any errors or

omissions at the earliest opportunlty.

Spolsky - Sociolinguistics

Uploaded by

Copyright:

Available Formats

Spolsky - Sociolinguistics

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

Spolsky - Sociolinguistics

Uploaded by

Copyright:

Available Formats

Oxford lntroductions to Language Study

Senes Editor H.G.Widdowson

Published in this series:

H.G. \)iddo wson: Li n gu ist i c s

OXFORD UNIVE RSITY PRESS

The social study of Ianguage

hat are the data?

The ethnography of speaking

Locating variation in speech

Language and gender

Accommodation and audience design

Bilinguals and bilingualism

Language loyalty and reversing language shift

an advantage to have a broad map of the terrain sketched out

before one considers its more specific features on a smaller scale, a

specific and specialized enquiry which students of linguistics are

language study concerned: its scope and principles of enquiry, its

ical participation in the exploration of ideas. This kind of

conceptual cartography has its dangers of course: the selection of

belief that there must be an alternative to a technical account on

issues discussed in the Survey. The idea is to give readers an initial

encourage them into close critical reading.

One way of moving into more detailed study is through the

educational effects of language variation was to continue to be

This book is dedicated to my grandchildren, busy studying

The social study of language

The scope of enquiry

interdisciplinary fields like psycholinguistics, applied linguistics,

THE SOCIAL STUDY OF LANGUAGE

professional organization. They apply a plethora of methods to a

of the formal linguists like Noam Chomsky for

autonomous linguistics, with the goal of describing the idealized

that could account for the similarities in the organization of

accounted for by appealing, outside language, to socially relevant

many different people. These small variations in language that

about. There is no single-style or single-variety speaker; no speech

community that does not have a choice of varieties; and many

ble to identify ourselves and others as belonging to certain

Com plementary approaches

people, and 'Hi!' to others? How did Professor Higgins know

differences in language usage that are associated, within a speech

community, with social or economic or political or religious or

to show how specific differences in pronunciation or grammar

At the other-the macro-end of the

labelled the sociology of language as distinct from sociolinguistics,

sociolinguistics, we treat language (and a specific language)

instance, about the significance of a group of imrnigrants shifting

completely to a new language or maintaining their olcl one for

Welsh survive English occupation, or how dicl Basque and

rnight ask why speakers of certain varieties are influential and

which emphasizes the social influences on language, and the

will see, a large number of discrete but related

broaden the scope of enquiry from the parent disciplines of

If there is a common theme that emerges from the studies of

to and signalling social information. My

identity (or rather my various identities) is recognizable from my

THE SOCIAL STUDY OF LANGUAGE

The methods of enquiry

Noam Chomsky initiated a revolution in linguistics by asking