:t

"

"

112

c
Ul

:i"
CQ

iii
::J
CQ

III
co
11I

and disadvantages of 'straight talking'. They must tot up the

social distance between themselves and those they are talking to,
the power relafionship, the cultural norms, and make a decision.
Suppose a colleague was drinking toa much whisky. The
speaker could say:
Stop drinking!
but would be more likely to say tactful1y:
I wonder if we should keep our heads clear for tomorrow's
meeting?
Or they might even make a joke of it:
Even if everybody else goes bankrupt,
manufacturers will survive!

the whisky

And of course, if offending a colleague was really toa much of

a risk, the speaker could just have kept quieto
But suppose someone had an urgent request, and felt obliged to
impose on another person, what happens? There are various
strategies which are used to soothe the situation. For example,
anyone imposing is often pessimistic:
I don't suppose YOlf could lend me a pound, could you?
Or they might try to minimize the imposition:
I won't keep you a minute, but ...
Or they might just apologize:
I'm terribly sorry to bother you, but ...
The various strategies occur worldwide, but they are not all
necessarily found in every language. Each culture has its own
preferred strategies. This type of study therefore overlaps with
sociolinguistics, the topic of the next chapter.

Questions
1 What is pragmatics?
2 What four conversational
principie?

maxims

form the cooperative

3 What is speech act theory?

4 What are frames?
5 Explain what
examples.

is meant

by adjacency

pairs.

6 How might a polite speaker phrase a command?

Give two

This chapter is concerned with

sociolinguistics, which
analyzes variation within a
language. It looks at
differences between speech
and writing, and at variation in
pronunciation between
different social c/asses. It a/so
outlines divergence between
men's and women's language.
It then discusses multilingual
communities and pidgins and
creoles.

Sociolinguistics is often defined as the study of language and

society. Whereas many linguists concentra te on discovcring
unity beneath the diversity of human languages, sociolinguists
try to analyze the social factors which lead to this diversity. In
brief, sociolinguists are interested in language differences, and
especially in variation within a particular languagc.

The notion of a language

Perhaps the first question that a sociolinguist needs to ask is,
'What is a language?' Can the notion of 'a language' be defined
geographically? Can it be equated with nationality? Or should a
language be defined by the mutual intelligibility of its speakers?
The answer to all these questions appears to be 'no'. A
geographical definition of a language would separate
Australian, British and American English, which is obviously
unsatisfactory. Nationality is a vague notion which has little to
do with the language a person speaks. Numerous Soviet Jews,
for example, regard themselves as essentially Jewish, yet speak
Russian. Mutual intelligibility is of little help, since a
Glaswegian and a cockney are likely to find it harder to
understand one another than a Dutchman and a German who
are considered to be speaking distinct languages. And there is no
objective linguistic criterion which can be. applied. Dutch and
German are not only mutually intelligible, they are also
structurally more alike than some of the so-called dialects of
Chinese.
Faced with this dilemma, sociolinguists prefer to start with the
notion of a speech community rather than a 'language'. And
they define a speech community as any group of people who
consider that they speak the same language. Consequently,
Dutch and German must be regarded as separate languages,
since, in spite of their similarities, the Dutch consider that they
speak Dutch and the Germans consider that they speak German.
And all the Chinese dialects must be classified as one language,
because, in spite of far-reaching differences, their speakers all
consider they speak Chinese.

Dialect and accent

Within a speech community, there is considerable language
variation. The speech of its members varies according to many

factors, including geographicallocation, age, occupation, socioeconomic status, ethnic group and sexo
The most obvious type of variety in a speech community is the
use of different dialeets. A dialect is usually associated with a
particular geographical area, such as the Geordie and Cockney
dialects of English, which are spoken in Tyneside and London
respectively. The term 'dialect' refers to far greater difference
than mere pronunciation. The Lancashire dialect differs from
standard British English in sound system, syntax and
vocabulary, with phrases such as I don't want for to go, summat
for 'something', nowt for 'nothing'. American English ranks as
a different dialect from British English, with phonological
innovations such as nasal vowels, and constructions such as 'I
kinda figured maybe' and 'He said for you not to worry'.
Unfortunately, in everyday usage, the term dialect is often
confused with the word accent. An accent refers only to a
difference in pronunciation. A Scotsman and a Londoner are
likely to speak English with different accents. But if the
underlying system and the vocabulary are the same, they will be
speaking the same dialect. In fact, although a considerable
number of local accents are still found in Britain, dialects are
dying out, que to the influence of education, radio and
television.
.

From high to low

More interesting to sociolinguists is variation within a single
geographical area. This is of two main types: variation within
the speech of a single person, and variation between people.
These two interact, and it is not always possible to separa te
them. Let us begin by considering the stylistic variation which
exists in the speech of any one person.
Every native speaker is normally in command of several
different language styles, sometimes called registers, which are
varied according to the topic under discussion, the formality of
the occasion, and the medium used (speech, writing or sign).
Adapting language to suit the topic is a fairly straightforward
matter. Many activities have a specialized vocabulary. If you are
playing a ball game, you need to know that 'zero' is a duck in
cricket, love in tennis, and nil in soccer. If you have a drink with
friends in a pub, you need to know greetings such as: Cheers!
Here's to your good health!

115

116

In some cases, a relatively normal vocabulary is combined with

altered syntax. In newspaper headlines and telegrams, all
surplus words are routinely omitted, sometimes resulting in
unintentional ambiguity:
Giant waves down Queen Mary's funnel (British newspaper)
Dacoits (= bandits) shoot dead policeman (Indian newspaper)
Specialized speech styles are carried to excess in some cultures,
where social situations may follow a high degree of ritual, as
among the Subanun, a Philippine tribe. If you want a drink, it is
not sufficient simply to give the Subanun equivalent of English
'Please may I have a drink'. This utterance might cause a
Subanun speaker to praise you for your fluent Subanun, but you
would not get a drink! Drinking, particularly the drinking of
beer, is a highly ritualized activity which progresses through a
number of stages. At each stage, there is an appropriate style of
speech, and advancement in Subanun society depends on how
well a person copes with this.
Other types of variation are less dearcut. The same person
might utter any of the following three sentences, depending on
the circumstances:
I should be grateful if you would make less noise.
Please be quieto
Shut up!
Here the utterances range from a high or formal style, down to
a low or informal one - and the choice of a high or low style is
partly a matter of politeness (Chapter 9).
But politeness is just one component of a more general skill, the
appropriate use of language. Knowing what to say when is
sometimes known as communicative competence. Native
speakers just 'know' it would be odd to say 'Kindly refrain from
smoking' to a 10 year-old puffing at a stolen cigarette, ar rude
to say 'Put that fag out' to a duchess. Both utterances are
equally inappropriate. Children and foreign learners have to
acquire this skill over a longish period. Lack of this type of
knowledg often makes a speaker sound very funny, so much so
that the use of an inappropriate register is one source of humour
in English, as in:
.Scintillate, scintillate, globule lucific,
Fain would I fathom thy nature specific.
This seems amusing because of the use of a formal style to
'translate' a rhyme associated with an informal nursery setting:

Twinkle, twinkle, little star

How I wonder what you are.
In ~gland, the. use of an inappropriate leveI of formality is not
consldered a senous social blunder in most instances. In any case
there is often a considerable amount of overlap between the use of
the different styles. It would not matter whether you said 'Hallo' or
'Good morning' to your neighbour. ln some other cultures,
however, the social situation requires a far greater degree of rigidity.
An extreme example is found in lava, where society is divided into
three distinet social groups. At the top are the aristocrats. ln the
middle are the townsfolk, and at the bottom are the farmers. Each
of these groups has a distinct style of speech associated with it. The
top leveIof speech is used primarily between aristocrats who do not
know one another very well, but also by a tOWIlsmanif he happens
to be addressing a high government officia!' The middle leveI of
speech is used between tOWIlsmenwho are not dose friends, and by
peasants when addressing their social superiors. The lowest leveI is
used between peasants, or by an aristocrat or tOWIlsmanwhen
talking to a peasant, and between dose friends on any leveI.
Furthermore, it is the form of language used by parents to their
children, so it is the style learned first by all lavanese children.
However, as they grow up, children are expeeted to shift to
addressing their parents in a more formal style, even though their
parents continue to speak to them in the lowest style!
The formality-informality scale overlaps with other stylistic
considerations, in particular the medium used. Let us now
consider this.

Speech versus writing

Speech and writing differ in a number of ways. Consider the
following spoken dialogue:
Speaker A: But the point is she's not such a strong character.
Speaker B: It's not the point she's as str ... she's stronger than
what she makes out ['ll tell you now.
Speaker A: Well maybe.
Speaker B: She's a lot stronger cos otherwise I would have
drived her mad when she lived here but no she's a lot stronger
than what she makes out to you lot ['ll tell you that now.
The talk is shared between two people. They both assume some
mutual knowledge, so we never hear who she is, or where here is
located. It's repetitive: speaker B keeps stressing how strong she is.

117

118

iii
::J

cc
c:
\li

cc
CD

\li
::J

C.

9.

vocabulary
icipant

It's not composed purely of sentences: the fragment 'Well maybe' is

treated as a complete utterance. The verbs are all active ones (telI,
drive, and so on), and the sentence structures are fairly
straightforward. The vocabulary consists mostly of common
words, withsome colloquial phrases (drive mad, ['lI telI you that).
Now look at a passage from a quality newspaper on a similar
theme:
Assertiveness problems are pervasive. For example, marital
discontent can arise from the inability of partners to talk
assertively about their problems. Instead they tend to bottle
up feelings,. which inevitably leads to hostility. Marital
violence also occurs more frequently in men low in
assertiveness and may be explained by their inability to be
assertive as opposed to aggressive.
The uninterrupted flow of words is written by a single author. It
is fully explicit, in that it does not refer to unexplained people
or places. The only repetition is the occasional re-use of key
words, such as assertiveness, inability. The passage contains
only complete sentences. There is a passive verb may be
explained by, and the sentence structures are relatively complex,
with several embeddings (sentences one inside another, Chapter
7), as in inability of partners to talk, tend to bottle up, which
inevitably leads. There is a spate of abstract nouns:
assertiveness, inability, discontent, violence, hostility, and
several lexical items are of fairly low frequency: marital,
pervasive.
These passages contain fairly typical differences between spoken
and written language. They can be summed up in the following
table:
Spoken

figure 10.1

Written
FuI!
sentences
Elaborate
structure
Single
Abstract,
Explicit
Non-repetitive
writer
less
common vocabulary

Several ~f these features ~v~rla~ with the informality-formality

scale, wlth speech contammg mformal features, and written
lang.uage formal o~es. ~onsequently, formal speech has quit~ a
lot m common wlth mformal writing. Readablewriters are
~ometimes said.to ~e those who 'write as they talk' - though this
IS usually an IlluslOn, and apparent effortless spontaneity is
often carefully crafted.
But the imp0r:ta~t ~oint is this: spoken language typically involves
the. charactenstlcs m the left-hand column of Figure 10.1, and
wntten language those in the right-hand column - though each can
borrow from the other. There is no hard and fast divide. A sermon
is likely to have more 'written' characteristics than a chat between
friends in a pub. One is not 'better' than the other, each is
appropriate in certain circumstances. Written language is
sometlmes wrongly thought of as an ideal model for speech. In
praetice, those who reproduce written language when they speak
sound quite odd. Occasionally, recent immigrants are regarded as
pompous pedants, primari1y because they may have painstakingly
learned English from books.
Spoken and written characteristics, then, are another facet of
speech styles which efficient speakers and writers control with
ease.

Charting phonological variation

Speakers vary not only their vocabulary and syntax, but also the
sound structure. Phonological variation, both between speakers
and within a single speaker, is immensely important as a
reflection of various social factors. Speakers of a language alter
their phonology to suit a particular situation, often without
realizing it. For example, someone from Devon is likely to
pronounce the [r] in a word such as farm when chatting with
friends at home, but would probably attempt to suppress it in a
formal interview in London. In this case, the speaker may well
be aware of the change in pronunciation. On the other hand,
few speakers of standard British English realize that in informal
situations they often omit the [t] at the end of words such as last
in phrases such as last thing.
At one time, it was thought that such variation was fairly
random, and that no precise statements could be made about it.
But an American sociolinguist, William Labov, showed that this
was not so. In a piece of work which has now become famous

119

iii
::J

(Q

c:

III
(Q
CD

III
::J

Q.

~
C)

120

iii
::;,
lQ
cIII
lQ
CD

i
III

g
~

-.a.

<:)

he examined the pronunciation of words such as car, park in

New York. New YOIkers sometimes pronounce an [r] in these
words and sometimes do noto Although Labov was unable to
telI wich words weIe likely to be pIonounced with [r], and
which without, he found that he could predict the percentage of
[I] sounds which each socio-economic class and each age gIoup
would use in any given type of speech.
Labov started bis work on [r] in a highly amusing way. First, he
found out which departments were on the fourth fIoor in three
New York department stores. He then asked as many shop
assistants as possible a question such as 'Excuse me,. where are
children's coats?' The answer to each of these quenes was, of
course, 'On the fourth (loor', which included two words that
could each contain an [r]. It is welI-known that sales-staff tend
to mimic the speech of their customers, a':ld, as Lab?v pr~dict~d,
he found that in the store that was consldered socIalIy mfenor,
the number of [r] sounds was low, under 20%. In the middleranking store, [r] was inserted about 50% of the time, and in the
store considered socially superior, [r] was used over 60% of the
time. These pre1iminary results clearly showed that the use of [r]
in New York was a useful guide to social status.
After this preliminary survey, Labov ~hen examined ~he speech
of each class of person in more detal!. Perhaps predIctably, he
found that [r] was inserted much more frequently in careful
speech and in the reading of word lists than in casual speech.
This was true of alI social classes. There was, however, one
unexpected finding. When reading word lists, low~r-middleclass speakers inserted [r] more often than upper~mId?le-class
speakers - even though in casual speech, the. S!tuatIon was
reversed, with lower-middle-class speech contammg fewer [r]
sounds. This suggests that lower-middle-class speakers are more
consciously aware of speech as an indicator of social class, and
are making efforts to improve their status.

Phonological variation in British English

At first, one might assume Labov's results to be unique, in that
they possibly reflected an American soci~l situation tha~ ~as
unlikely to be paralleled elsewhere. But m England, a. s~ml1~r
state of affairs has been found in the speech of people hvmg m
Norwich. Consider the differing pronunciations of words
ending in -ing. Sometimes, Norwich inha?itants pronounce ~h,e
-ing as in Standard English, and at other trmes they say walkzn ,

talkin', singin', with [n] instead of [1)]. When the distribution of

-ing was examined more closely, a number of interesting facts
emerged. First of alI, and predictably, the proportion of -ing
forms was much higher in careful speech than in casual speech
for alI social classes. For example, those classified as lowerworking class used -jng around 70% of the time when they were
asked to read word lists, but hardly ever in casual speech. On
the other hand, middle-middle-class speakers used -ing 100% of
the time in word lists, but only around 70% of the time in
casual speech.
Second (and perhaps surprisingly), upper-working-class Norwich
inhabitants were found to behave in a very similar way to lowermiddle-class New Yorkers. For this social group, there was an
enormous discrepancy between the type of speech used in word
lists (-ing occurred 95% of the time), and that used in casual
speech (-ing occurred only 13% of the time). Once again, speakers
with a re1atively low social status appeared to be attempting to
'better themseIves' by speaking in a style they regarded as superior
to their normal speech. The Norwich situation is illustrated in
Figure 10.2 (p. 122).
A further analysis of the use of -ing by upper-working-class
speakers revealed an unexpected sex difference. Women were
found to uSe -ing more often than men. This suggests that
'changes from above' (Labov's term for changes of which
speakers are consciously aware) may well be initiated mainly by
women. There is further evidence that other changes are taking
place 'from below', that is, below the levei of conscious
awareness. These changes appeared to be initiated by workingclass men. For example, in Norwich, the vowel sound in words
such as night, rye, side, is moving towards the sound oy [;)1].
This change is furthest advanced in the speech of working-class
men. It has been suggested that, perhaps unconsciously, people
admire working-class men and associate them with strength and
virility, and, without realizing it, adopt features found in their
speech. This pattem seems not to be unique to Norwich, but a
general phenomenon found in other areas of the world also.
These examples of phonological
variation are highly
informative. They provide an objective reflection of various
social factors such as socio-economic class, ethnic group, age
and sexo The effect of any one of these factors on language can
be analyzed, and so can the interaction between them. For
example, in one study of the interaction between ethnic group
and age in Boston, a particular pronunciation of [o] was found

121

iii
::;,
cIII

lQ

lQ
CD

i
III
1II

g
~

-.a.

<:)

122

to he associated with eIderly Jews) and with Italians of all ages.

The younger memhers of the Jewish community) who were
mostly highly educated) had abandoned it) perhaps hecause they
regarded it as non-standard. The Italians, on the other hand)
tended to favour it as a mark of Italian identity. Such studies
can shed interesting light on the pressures and attitudes within
particular communities.
Use of -ing
%
100
Word
Iists

Word
lists

50

Word
lists

Middle
middle
class

Upper
working
class

Lower
working
class

figure 10.2

function. But, as with many surveys, the result is oversimplified.

In practice, people do not norma1ly live in such dearcut layers:
someone &om the so-called working dass might well have
middle-elass friends and neighhours.
In fact, human heings tend to duster into social networks,
groups of people who regularly interact with one another. A
detailed study of the social networks within one particular
speech community can provide a useful corrective to Lahov-type
studies) which tend to suggest humans are rigidly stratified.
Network studies can provide a more realistic picture of the way
people interact in real life. Furthermore) if a sociolinguist
manages to he introduced into a network, its members are more
likeIy to chat in a natural way than in a Lahov-type survey in
which it is sometimes difficult to observe people speaking
'normally) .
The British linguists Jim and Lesley Milroy pioneered the
linguistic study of social networks with a study of three
communities in Belfast. Lesley was introduced into each group
as 'a friend of a friend'. This ensured that she was accepted, and
that people would talk relativeIy normally in front of her: when
one youth tried to show off hy talking in a somewhat affected
way, his friend punched him and shouted: 'Come on) you're not
on teIevisiori now) you know'.
Networks can he of high density) when the same people tend to
work, play and live together. On the other hand, they can he of
low density) when people only have a small amount of contact
with any one network, in that they may live in one area, work in
another) and traveI eIsewhere for their sociallife (Figure 10.3).
When the Milroys examined their data, they discovered a
number of things which would not have been detectable in a
Lahov-type survey. For example) Lahov's work suggested that
men and women's speech tended to differ, with women on the

/e

Social networks
Labov-type surveys rely on collecting data from a random sample
of individuaIs. Their speech is analyzed for various key
characteristics, which are then correlated with their socioeconomic background. The result, perhaps not surprisingly,
suggests that human society is somewhat like a layer-cake, with
different socio-econornic layers stacked up on top of one another.
In one respeet, this is a useful insight into the way societies

/ ./e
\__e/'.

)High density network

figure 10.3

Low density network

123

Ql

1Il

on
~

124

iii
::::I

(Q

C
1lI

cg
1lI
::::I

a.
~
9.

whole being doser to the prestige norm. The Milroys :ound t!'lis
pattem also in Belfast overalI, but with some mteres~l~g
subtIeties. When Lesley examined the three commumtles
separately, and charted the occurrence of the way [a] was
pronounced by people of different ages and both sexes, she
found some modification of the overalI pattem. 1n the oldest,
most tightIy-knit, and most traditional com.munity, she found
that the predicted pattem of male-female dlfference was rn.0st
prominent. But in the other two, which were newer, and falrly
loose-knit, this pattem was less evident, and was even reversed
among the younger women of one community. This su~gests
that a blurring of sex differentiation in language may be lmke.d
with the break-up of dose-knit networks. Findings such as thlS
indicate that linguistic variation needs to be considered from at
least two angles: from the point of view of a broad Labov-type
survey based on a random sample of people, but also from a
dose-up view of a number of social networks.

Language and sex

Possible sex differences in language
attracted a lot of attention.

usage have recentIy

First we need to sort out whether women realIy do speak

diffe~entIy from men. People's impressions are not necessarily
correct: it is often assumed, for example, that women talk more
than men, whereas almost alI research on the topic has
demonstrated the opposite, that men talk more than women.
Similarly, it is sometimes daimed that women u~e 'empty'
adjectives, such as divine, charming, cute, yet thlS type of
description is possibly more usualIy used by (presumably male)
writers in popular newspapers to describe women.
The most consistent difference found between men and women
within the westem world is a tendency for women to speak in a
way that is doser to the prestige standard. 1n colIoquial terms,
they speak 'better' than men. No one is quite sure why this is so,
and severaI explanations have been proposed, which ma.y alI be
partialIy right. For example, women may be pressunze? by
society to behave in a 'lady-like' manner, and 'speaking mcely'
may be part of this. Or because they are the. maio; child-rearer~,
they may subconsciously speak in a way WhlChwllI enable t?elr
children to progress socialIy. Or they may tend to have Jobs
which rely on communication, rather than on strength. AlI these
factors, and others, appear to be relevant. Moreover, in recent

years, particular1y among employed women, the differences

between men's and women's speech appear to be diminishing.
Furthermore, some characteristics attributed to women tum out
to be far more widespread. For example, women have been
daimed to use more hedges, tentative phrases such as kind of,
sort of, in place of straight statements: 'BilI is kind of short',
instead of 'Bill is short'. They have also been accused of using
question intonation in response to queries: 'About eight
o'dock?' as a reply to: 'What time's dinner?' Yet this insecure
style of conversation seems to be typical of 'powerless' people,
those who are somewhat nervous and afraid of antagonizing
others. Powerless people come from either sexo
But there is an altemative explanation: such speech may be
supportive. A question intonation promotes the flow of
conversation. A comment such as: '1t's cold today, isn't it?'
encourages an easy-to-make response, such as: 'Yes, I even put my
winter boots on'. 'Powerless' speech can therefore be viewed as
friendly and cooperative, and powerful speech as insensitive and
authoritarian.
Friendly speech may also reflect the setting. At a meeting, fairly
formal speech is the norm. At home, or in the shops, informal
conversation is more likely. TraditionalIy, men are more likely to
be at business meetings, and women at home, though this is
partIy changing.
Supportive speech is more often associated with women than
with men. Friendly females are likely to help the conversation
along by saying 'mmm', 'aha', 'yes' - so calIed 'minimal
responses'. These encourage the speaker, by showing that she is
being listened to. Simultaneous speech can aIso be supportive,
when the speaker's message is reinforced by the listener, as in the
folIowing overlap about going to funeraIs:
Speaker A: Perhaps they would want you to go, you know .
Speaker B:
yeah for their comfort .
Such supportive speech contrasts with its opposite, power
talking, whose characteristics are outIined below.

Power talking
Speaker A: Now tell me what you're going to do.
Speaker B: Yes, well, first ...
Speaker A: Louder, please, we all want to hear.

125

iii
::::I

(Q

c:
1lI

(Q
lD

:la.
1lI

~
()

126

iii
::J
r.a
c
DI

~peaker .1): 1a star,; vy cutung uns nere.

Speaker A: What do you mean 'this here'?
Speaker B: The place where ...
Speaker A: Have you washed your hands?

connnued to address one another as tu. This situation is shown

in Figure 10.4. This linguistic situation reflected the social
situation. There existed a feudal society in which the power of
one class over another was all-important.

r.a
CD

DI

5.
~
~

tu lower
To
vos class
To upper class

'Powerful' speakers typically control the topic, interrupt others,

and demand explicit explanations. Occasionally, this may be
Lowerclass
justified if someone is chairing a meeting, or in
someUpperclass
teaching
situations. Yet quite often, as perhaps in the example, above, the
'controller' goes over the top, and tries both to domina te and
flatten the confidence of other participants.
Power talking may be used by either sex, though it is more
typically male. Male speakers not only talk more, they also
interrupt more, even though they may not perceive themselves
as doing so.
Men also issue more direct orders. In a study of doctor-patient
interaction in the United States, men used explicit commands in
about one third of the directives, as: 'Lie down', 'Take off your
shoes and socks'. Women preferred to phrase commands as joint
actions: 'Maybe we should just take the top of your dress off?',
'Maybe what we ought to do is stay with the dose you're on',
and so on.

Change in language styles

The social situation is not necessarily static. Any change in the
social relationships is likely to be mirrored in changing language
styles. An example of a change of this type occurred in the gradual
meaning change in the two forms of the pronouns you in
European languages.
Originally, in Latin, there was a singular form tu and a plural
form vos. For some reason (the cause is disputed), the plural
carne to be used as a polite form of address for speaking to a
single person in authority. One theory is that when there were
two Roman emperors - one in Constantinople, the other in the
west in Rome - it became customary to address each of them as
vos, since both emperors were implicitly being addressed at the
same time. This began a general trend for using vos to anyone
in authority. It gradually became customary for a working-class
person to address a member of the, aristocracy as vos, while the
upper classes still used tu to a lower-class individual.
Meanwhile, as a mark of respect, the aristocracy began to
address one another as vos, although the lower classes

127
iir
::J
r.a

I:
DI

r.a
1Il

III

~
~

figure 10.4

However, as feudalism died out, so did this strueturing of tu and

vos. Gradually (according to one theory), people ceased to feel
such respect for those in power, and instead, they merely felt
remote from them. Vos (it is claimed) carne to be not so much a
mark of respect, as one of non-intimacy. Tu carne to be thought
of as indicating intimacy, companionship and solidarity. People
involved in friendships or close relationships started to call one
another tu irrespective of the power situation. And this is the
state of affairs today in the numerous European languages which
have two fOftllS of the word you.
The 'power' to 'solidarity' switch is possibly only one of several
factors involved in the change from vos to tu, and some have
disputed its importance. However, a similar phenomenon seems
to be occurring in other parts of the world also: in India, in the
Hindi and Gujarati languages, there was formerly a power
pattern shown in the non-reciprocal nature of the forms of
address between husband and wife, and older and younger
brother. Nowadays this is dying out. Reciprocal relations are
gradually becoming more important than the power of one
person over another, and members of Indian families are
beginning to address one another with the intimate you forms.

Multilingual communities
'I speak Spanish to God, Italian to women, French to men, and
German to my horse', is a saying attributed to the Holy Roman
Emperor Charles V. As this quotation suggests, in some cultures
a changed social situation is marked by a change in the actual
language spoken, a pheno~enon known as code-switching.

128

Sociolinguistical1y, this is not very ditterent trom styl1StlC

variation within a single language. Sociolinguists have therefore
become interested in studying code-switching in bilingual and
multilingual communities.
For example, in Sauris, a small community of north-eastern Italy,
high in the Carnian Alps, a quite remarkable linguistic situation
exists. The 'inhabitants use three different languages in the course
of their everyday life: a German dialeet, Italian and Friulian (a
Romance dialect). Italian is the la~guage of organized religion,
and also that used in schools. Friulian is the language used by men
in the local bars. And German is the language in the home. It is
highly unusual to hear German outside the home, though it was
observed on one occasion when a furious woman burst into a bar
and upbraided her husband for not having returned home at the
time he was expected!
A study of the ways in which these multiple languages are used is
particularly important for language planning, a situation in which
a governrnent or education authority attempts to manipulate the
linguistic situation in a particular direction. This is more likely to
be successful if existing uses of a language are gradually extended,
since the sudden imposition of a particular language by decree
may well result in failure.
However, multilingual societies in which all the speakers are
proficient in alI the languages spoken are something of a rarity.
Quite often, one language, or simplified language, is adopted as
a common means of communication. This can happen either
naturalIy, or as a result of language planning. A common
language of this type is sometimes known as a lingua franca. A
couple of millennia ago, Latin spread around the Mediterranean
countries in this way. In India today, English tends to be a lingua
franca: Hindi speakers from the north are likely to communicate
in English with people from the south who mostly speak one of
the Dravidian languages. The artificiallanguage, Esperanto, is
sometimes proposed as a candidate for a world lingua franca.

Pidgins and creoles

Adopting a lingua franca is not the only solution to the problem
of communication between groups of people speaking different
languages. In some cases, a pidgin develops.
A pidgin is a restricted language system which arises in order to
fulfil essential communication needs among people with no

common language. It is no one's first language, and is used at

first in a limited set of circumstances. Such a system typicalIy
develops on trade routes and in coastal areas.

129

A pidgin is usually based on one language, though it soon

acquires an admixture of other languages, as well as
independent constructions of its own. For example, Tok Pisin
(also known as Melanesian Pidgin English and Neo-Melanesian)
which is spoken in Papua New Guinea, is based on English, and
many of the words sound somewhat like English ones:
Mi go long taun. 'I go/went to the town'.
Yu wokabaut long roto 'You walklwalked along the road'.
But there are plenty of others, which cannot be predicted from
English, such as lotu 'church', diwai 'tree', susu 'milk'. In
addition, it has acquired syntactic constructions which do not
figure in English. For example, there is a consistent distinction
between verbs with an object ('transitive' verbs) which take the
ending -im, as with bagarapim 'wreck', and those without
('intransitive' verbs) as in bagarap 'collapse', 'break down':
Mi bagarapim ka bilong YU. 'I crashed your car'.
Ka bilong mi i bagarap. 'My car broke down'.
Another innovation is the particle i which sometimes has to be
placed before;the verb (as in the second sentence above).
The phonology, syntax and lexicon are simpler in a pidgin than
in an ordinary language. There are fewer phonemes. In Tok
Pisin, [p] and [f] are often merged, so are [s] and [f], and there
are only five vowels. English 'fish' was borrowed as pis, and
English 'ship' as sip. In order to avoid confusion, 'piss' (urinate)
became pispis, and 'sheep' became sipsip. There are very few
word endings, the sentences have a simple structure, and there
is a smalI vocabulary. One or two items stretch over a wide area,
as with the following uses of the word pikinini 'child':
pikinini
pikinini
pikinini
pikinini
pikinini
pikinini

man 'son' (lit. child man).

meri 'daughter' (lit. child woman).
hos 'foaI' (lit. child horse).
pis 'minnow' (lit. child fish).
bilong rais erice kernels' (lit. child of rice).
bilong diwai 'fruit of tree' (lit. child of tree).

Sometimes, pidgins die out of their own accord. At other times

they increase in importance, and become used in more and more
areas of life. If someone then acquires a pidgin as a first language
- perhaps because of intermarriage between people whose only