\-:'.

i:,"1'

_,r*-,

16

.>

Lexical variation and change

Variation in language is multi-dimensional. In sections 3 and 4, we

have looked at how v6riation in social structure is reflected in the sound patterns of language and how this variation is often indicative of language
change in progress- We have also seen how geographical variation in language
is caused by different levels of contact between different peoples at different
times. In this section we are interested in variation in words and in their
origins, meanings and contexts of use. We'il also examine change in both the
choice ol words and the meanings of those words.

Borrowing words
What is the origin of words like shampoo, pizza, alcohol and curry?
When did they enter the English language? And why? Almost certainly, you
will be able to answer these questions for at least some of these words, but we
can ask the same questions with respect to words which are much less 'exotic'.
According to published counts of word frequencies, the items listed in (208)
are arnong the most lrequently occurring nouns in English:

(208)

people, way, water, word, man, day, part, place, things, years, number,
name, home, air, line

All these words have been part of the English language for centuries, and
while most of them date back to Germanic languages which preceded the separate development of English, some had their origins in Latin (part, place and
air, for example). Throughout its history, English has been adding to its
lexicon by acquiring new words from other, often unrelated, languages.
Risotto and pizza come from Italian, vodka from Russian, goulash from
Hungarian, cofee and yoghurt fromTurkish, alcohol and sherbet from Arabic,
sago from Malay, ketchuplcatsup frorn Chinese and tomato from Nahuatl (a
254

Borrowing

words

255

central American language and the language of the Aztecs)- These new words
are known as borrowings- Of course, as well as having borrowed thousands of
words, English has been a great provider too, much to the annoyance, lor
example, of language purists in France who strive to find native French words
to replace le parking, le hamburger and le walkman.
Why do speakers borrow words from other languages? Perhaps the most
obvious reason is sheer necessity. People need to develop words for new and
unfamiliar concepts - new technology, new plants and animals, and in the
examples above, new and unfamiliar foods. Note that the model of lexical representations discussed in section 14 supposes that there is a distinction
between concepts and lexical entries, and from this perspective, there is
nothing odd about the suggestion that we have concepts lor which we lack
words. Another reason is prestige. If certain cultures are associated with particular prestigious activitieq it is common for the words associated with that
activity to come from the language of that culture. Continuing with the food
theme, France was at one time considered the centre of world gastronomy, and
hence English has words llke cordon bleu, gourmet, cuisine, restaurant, menu,
mousse and soup which it has borrowed from French.
When a word is borrowed, it is olten gradually changed so that it fits the
phonological and morphological structure of the borrowing language or
dialect. So whilst Frangoise and Ricardo might go to a cale [kafe] for a croissant [kr waso] and a cappuccino [kaputtJino], Mavis and Vic, in London,
would go to the [kaf] for a [kwason?] and a,[kapetJrineu]. Similarly, whereas
the plural of pizza is pizze in Italian, English now applies its own plural morpheme to tlre borrowed word, hence pizzasSometimes when new concepts are introduced from other societies, the speakers of a particular language may use their own native linguistic resources to coin
a new word. These are known as calques- Let's look at some examples of this.
Comanche, an American language of the southern USA, has a word ?ohaplti?ataka-sikikamatl, which literally means 'orange's brother tastes sour'. It is the
word used for a lemon! In Irish Gaelic the words sciathfearthanna translate as
'rain shield', and refer to an umbrella. In New Zealand, it is the job of the Maori
Language Commission to create new words by using words already in the language. As a consequence, we find examples such as those in (209):

(209)

New

Word:

papa

patopato wai mangu roro hiko

Literal Meaning: board knock water

ink
Idiomatic Meaning: keyboard

black

brain electric
computer

English tends to resort to Latin and Greek when new words are devised,
particularly for referring to new technology. Examples appear in (210):

7l

256

Lexical variation and change

(210)

television: Greek rele ('far')

Latin visro ('sight, thing seen')

+ Greek skope in ('observe closely')
photograph: Greekphoto ('light') * Greek graphos ('written')
microscope: Greek mikro

(' small')

Borrowings, then, are words which originated in one language (or dialect), but
which have come to be used in another, even by people who don't speak the
'lending'language. These borrowings are very often assimilated to the phonological and morphological structure of the new host language (exercises I, 2 and 3),

Register: words for brain surgeons and soccer players,

hairdressers

an

d lifesavers

A register is the specialised vocabulary common to a particular trade,

occupation, topic or activity. Hairdressers, soccer players, brain surgeons and
undertakers all have specialised words or uses of words which refer to concepts
particularly common or specific to their activity or profession. As a soccer
player you might nutmeg your opponent (kick the ball between their legs) or
play a one-two; you might ask a hairdresser for aflat top or a bob, or need a
surf lilesaver to rescue you from a rip (a dangerous backcurrent on a surf
beach), but you are unlikely to ask a brain surgeon for a lobotomy.It is of some
interest that occupations, interests, etc. can have some impact on the important idea of a basic level of categorisation introduced in section l3- Thus,
whereas for many of us, dog corresponds to a basic level category, for those of
us preoccupied with dogs, the basic level shifts down to that of particular
breeds. Similarly, while all of us are familiar with such words as beech, ash and
elm,many of us are not in a position to distinguish these different types of treeFor those of us who are arboreally challenged in this way, it is plausible to
suggest thal tree appears at the basic level in our categorisation systemsHowever, this is not the case lor botanists and tree surgeons. For them, the
basic level of categorisation will be that of beech, ash and elm or, indeed, the
more specific level of copper beech, mountain ash, etcOften, people consider that the registers of doctors and lawyers (and even
linguists) hinder communication and understanding. The term Jargon' is
sometimes used to refer to the confusing registers of particular occupations.
In some senses, registers are 'in-group' varieties, which lead to accurate and
speedy communication of information among those that know and use them,
but confuse those who don't. It is obviously important that the doctor tells the
nurse that you have had a coronary infarcture or a stress fracture of your left
tibia, but what you want to know is that you've had a heart attack or broken

Biscuit or

cookie?

257

your lelt leg. In this medical example, the use of a special register is clearly a
necessity - the leg, for example, has several major bones and it is vitally important for the nurse tO know which one you've broken. Some registers, however,
are deliberately confusing so as to hinder understanding by outsiders. This
may be because the group speaking the particular register wants to maintain
a sharply contrasting identity, or maybe has something to hide (exercise 4).

Biscuit or cookie? Variation and change in word choice

Consider (21l):
(2t

l)

ConcePt

Word to refer to concept

Britain:
. U.S.A.:
Australia:

biscuit
cookie
biscuit/cookie

The thin flat, often round, usually sweet, hard but crumbly thing we eat during
our coffee break is called different things in different English-speaking areas.
In Britain it is usually relerred to as a biscuit and in the USA it is a cookieAustralia is experiencing the initial stages of language change with the word
biscuit gradually losing out to the word cookie.
Such geographical differences in word choice are well known. Most people
are familiar with the US-UK contrasts between sidewalk and pavement, gas
and petrol, pants and trol$ers, elevator and lift, vacation and holiday- Jusl as
borrowing is frequent in situations of language contact, as we saw earlier, it is
also very common when dialect contact arises. In the past century, within the

English-speaking world, most interdialect borrowings have come irom

American English, with the newly borrowed words pushing out or beginning
to push out older words, usually of British English origin. Thus, we find examples such as those in (212), where in each case, the American English lorm is
replacing or has replaced the British English equivalent:

(212)

British English

American English

houseY

bingo

bakery
minerals
pictures

baker's shop

soft drinks
movies

lorry

truck

chips

fries

crisps

potato chips

7!

258

Lexical variation and change

An interesting study of lexical shift from older 'British'-type words to

American borrowings has been conducted by Miriam Meyerhoff in New
Zealand. As is indicated in figure 39, she found that while some .British'
English words were being retained, manyAmericanisms were being borrowed,
a finding which reflects both the increase in sociocultural contact between
the

uS and New Zealand, and the reduction in such contact

Znaland and the UK-

between New

'British'words
'American'words

tr6U

.=

70
d

360
Eso
E

'6

644
O
bo

6
E30
o

ff20
t0
0

FqF-o./)o\!u
x!trtoc/

EiJir

634-EirrFPe
.^o42-n
;a\,)
pq6()

&

Lexical pairs

Figure 39 Reported use of lexicar pairs in New Zearand Engrish

A number of studies have suggested that people are able to acquire new
lexical items rather more quickly than they can acquire new phonological fea_
tures. For instance, Jack Chambers has compared the rate at which a group of
canadian children whose parents had moved to southern England adopted
British English lexical and phonological features (see section 3, exercise 7). He

selected twenty-five British,/canadian lexical pairs (including nappyldiaper,

pushchairlstroller and bootltrunk) and five pronunciation pairs (including
ibenone]/[benrena] and [tamotnu]/[tamerdrru]) and analysed the extent to

Biscuit or cookie?

ffil

F1

tr

o7(l

60

2s9

lexical items
phonological items

60

tal

fr50
f40
o
g30
o

Eza
bo

Ero

o
o
A

o
9

years

14

Years

l7

Years

Age of child

Figure

40

The adoption

of Britislt English by Canadian children:

lexicrsn and phonology

which the Canadian youngsters adopted the British lorms. The graph in figure
40 presents the findings lor three of the children he studied. Each child had
acquired more of the lexical items than of the pronunciation features. We have
therefore seen two examples ol dialect contact leading to change in lexical
choice: sociocultural contact with North America has led to the adoption of
American English words in other English dialects, and contact with British
English has led a number of Canadian children to shilt away from their indigenous lexical patterns to those of their new home. This dialect contact also
has a considerable effect on lexical variation witltin individual E,nglish speaking countries. In England, the urbanisation of rural areas has had a devastating effect on the survival ol traditior-ral rural dialects. Urban varieties are
increasingly being diffused into the surrounding rural areas, the effects of
which are particularly visible in the lexicon- Traditional dialect words are
losing ground in competition lrom words from urban or standard dialects. An
example ol such lexical attrition is presented in the map below. A century ago,
the word dwile (meaning'floorcloth') was widely used in the Eastern Counties
of England. Today it is restricted largely to the adult populations of Norlolk
and parts of Suffolk. In a recent study, as indicated in figure 41, the word
dwile was barely recognised by any of the children surveyed, which strongly
suggests that it is unlikely to survive long into the twenty-first century (erercise 5)-

260

Lexical variation and change

The area of 'dwile'use in late

twentieth century

The area of 'dwile' use in late

nineteenth century

Map

The lexical attrition

g >ESoo
E E.E
9C
=-

of the word dw|Ie in East Anglia

100
80

E E:* +
=
-- 3*i
.='E
UE
e
diE
o o-5
boQ c

ggo

60
4O

).o

a?n

o-v

adults

children

Age group

Figure

4l

The lexical attrition

Same word

of dwlle in East Anglia

new meaning

A'nice'example to begin our discussion of the way word meanings

change is presented by the word nice itself. This word entered the English lan-

Same word

new

meaning

261

guage around the thirteenth century from Old French, a descendent of the
Latin word nescius meaning'ignorant'. By the fourteenth century its meaning
had already changed to mean 'silly'or'wanton': a nice person was one from
whom favours might easily be obtained. In the fifteenth century, nice carne to

mean 'coy' or 'shy', by the sixteenth

eenth century

it meant 'subtle', and only in the eightdid it reach its present meaning of 'agreeable' or 'good'.

Nowadays, the meaning of nice appears to be weakening: it has such a bland,

general, quality of 'goodness'that in some contexts, such as that illustrated in
(213), it means little more than'OK':

(213)

[conversation between father and daughter]

HErDr: Hey, dad, I've just bought a new Golf GTi convertible. What
do you reckon?
ALBERT: Mm. It's nice-

For a more contemporary example of semantic change, consider the word

gay. Originally, gay meant 'full of joy and mirth, light-hearted'. In the middle
of the twentieth century, however, it also came to mean 'homosexual', and
now this later meaning has aimost completely taken over lrom the former one.
Just like linguistic change in phonology, which we discussed earlier (section 4),
semantic change is always preceded by semantic variation - in other words, at
some stage in the shift from meaning A to meaning B, both meanings will be
current within a community- At one time, therefore, both Joyful'and 'homosexual' were meanings of the word gay. Gradually, over time, one meaning has
begun to be used much more than the other to such an extent that the older
meaning is dying out.
If we look back into the history of English, many thousands of words have
changed their meaning in the same way that the word gay is changing today.
In an attempt to establish regularities of semantic change, historical linguists
tend to classify rneaning changes according to the nature of the semantic shift.
Some changes are due to semantic broadening: here the word takes on a
wider, more general meaning than it had previously. The word thing is a classic
example of such broadening. In Old English and Old Norse, this word meant

'a public assembly'. In

present-day lcelandic, a language with similar

Germanic roots to English, it still does. In Modern English, however, it has
now been extended so much that it simply means 'an entity of any kind'. The
word companion provides another example. It used to mean 'someone who
eats bread with you' (see Italian con'with' + pane 'bread'); now it means
'someone who is with you'- The word broadcast, which only a couple of centuries ago meant 'to sow seeds', has now, in this technological age, been
extended to include the spreading of information on television and radio.

262

Lexical variation and change

Pudding, which today is usually sweet and eaten for dessert, comes from the
French word boudin which means a sausage made with animal intestines.
The opposite of semantic broadening is semantic narrowing, with the word
taking on a more restricted meaning than before. In Middle English, a girl was
a young person of either sex, a boy was a male person of any age and lust
simply meant 'pleasure'. A number of words with similar meanings have
undergone shifts in different directions of generality. For examplg the word
hound was once the generic word for a canine- This word's meaning has narrowed and the generic canine term is now dog, which once referred to a par-

ticular breed of dog.

These changes in word meaning have often obscured the Germanic roots of
the English language, with many originally Germanic words either changing
in meaning or dying out. Consider table 30 below:

Modern English

old English

Frisian

Dutch

German

meat

flesh

fleis

vlees

Fleisch

animal

deer

dier

dier

Tier

dog

hound

houn

hond

Hund

cloud

wolcen

wolk

wolk

Wolke

die

steorfa n

stjerre

sterven

sterben

bird

fugol

f0gel

vogel

Vogel

smoke

reek

rikje

roken

rauchen

Poor

earm

earm

arm

arm

air

lvft

Iucht

lucht

Luft

take

niman

nimme

nemen

nehmen

In this table, note the similarities between the Old English words and the
equivalents in the modern day varieties of the closest cousins ol English.
Words such as steorfan (Modern English: starve) and reek have been semantically narrowed in the transition from Old English to Modern English, and
many of the other words have died out in the face of competition from other
English words, or from words borrowed from other languages. For example,
poor is a word borrowed from Old French.
It is also common to contrast changes involving amelioration with those due

to pejoration. Pejorations involve the development of a less favourable

meaning or connotation for a particular word. Villains were formerly farm-

Same word

new

meaning

263

dwellers, but are now criminals; people who were crafty and cunning in medieval times were strong (see German Kraft) and wise, but now are deceitful and
evasive. Grotesque meant 'resembling a grotto or cave', but now means 'distorted and ugly'. The word dunce is taken from the name of a thirteenthcentury scholaE John Duns Scotuq whose thinking was discredited long after
hisdeath, Ameliorations, orthedevelopment of more favourable meanings for
words, are fewer in number- Some of the more notable examples are constable,
the meaning of which has shifted from 'an attendant at the stable' to 'a police
officer' and knight which in Old English referred to a boy or servant but now
has a much more prestigious meaningWe have now seen a number of examples of semantic changes. But what is
it about'meaning'that allows such changes to take place? How is it possible
for the meanings of words to alter so radically? April McMahon has suggested
three possible reasons:
a. Most words are polysemic - they have a range of meanings * and over time
marginal meanings may take over from central meanings (possibly because a
borrowing has invaded the semantic space of the central meaning). Note that
polysemy must be distinguished lrom ambiguity. An ambiguous word such as
match or bankcorresponds to two (or more) distinct lexemes and normally has
two (or more) distinct entries in a conventional dictionary. A polysemous word
has only a single lexical entry with a rarrge of ciosely related rreanings. An
example illustrating the takeover of central meaning is the word sloth, which
once had a central meaning of 'lacking in speed'. This central meaning was
taken over by the word slowness and so the central meaning of sloth shifted to
what was formerly a more peripheral meaning, namely 'laziness'.
b. Children do not receive a fully formed grammar and lexicon from their

parents, but, with help from lJniversal Grammar, have to figure it out for
themselves. The child may therefore acquire a slightly different meaning for a
word than that understood by its parents. Earlier we saw that children, in the
very early stages of language acquisition, sometimes seem to use certain
words with broader meanings than the adults around them e-g. dog to mean
'any hairy animal with four legs'(see section l3). As the child gets older, it
gradually restricts the meaning of the word more and more. It is not too difficult, however, to imagine that slight semantic shifts may emerge at the end of
this restriction process. We did, of course, express some reservations about
the extent of such overextended lexical use by small children in section 13, but

these reservations need not rule out what we are contemplating hereConsider, for instance the broadening of Old English dogge, relerring to a
specific breed of dog to the current situation where dog is the generic term for

264

Lexical vari'ation and change

canines- We suggested in section 13 that children are 'tuned in'to the basic
level of categorisation, and we can suppose that lor the case in question t.his
is the level of Modern English dog- All we need to suppose, then, is that lor
whatever reason at some point a child was exposed to examples of dogge and
interpreted them as referring to the basic level generic category. For such a
child at this point, the semantic broadening has occurred- Of course, it is still

necessary

to understand how such a child's 'non-standard' interpretation

became established and spread throughout the community, but we do at least

have a plausible account of the first important step in semantic change-

Overall, the suggestion that children are crucially involved in language

change is a very attractive one.
c. The relationship between concepts and the words which conventionally
refer to those concepts is arbitrary (see section 14) and so either can vary or
change fairly freely through time and across space. Just as different geographical areas may have different words to represent different concepts (lexical variation), so also different words may, through time, evolve so as to be associated
with different concepts (semantic change) (exercise 6).

Variation and change in morphology

As mentioned in section 10, English verbs have few inflectiong but
one which is found is that which marks present tense and agreement with the
third person singular subject. This is not the case in all dialects of Engiish,
however, and in some dialects this suffix has been lost. Speakers of Alrican
American Vernacular English (AAVE) in the USA and the English of East
Anglia in the UK produce examples such as those in (214):

(2t4)

a. this dog chase rabbits

b. this cat rniaow all night
c- he spend a lot
d- she dance well

This contrasrs with the situation in southwest England, where people would
not only say he spends a lot,buL also produce examples such as those in (215):

(2rs)

a. they spends a lot

b. I dances every night

In this area, the -s suffix does not mark present tense and agreement (with third
person singular subjects) but only present tense. Around the English-speaking

Variation and change in

morphology

265

world, therefore, there is variation both in the presence or absence of the -s

suffi.x and in its gramrnatical function.
Older versions of English, and most other Germanic languages (apart from
Afrikaans) have far more extensive systems of inflection than present day
Standard English. In Old English, there were four different present tense forms
still are today in German, although they are distributcd differently),
with two in Modern Standard English. This is illustrated for
comparison
in
the verb help and its equivalents in table 3l:
(as there

tc

Modem English

Modern German

Old English

helpe

lch

helfe

help

You (sing)
He,/She
We
You (pl)
They

help

hilpst
hilpth

Du

hilfst

Er/Sie

hitft

helpath

Wir

helfen

ge

helpath

lhr

helft

hi

helpath

Sie

helfen

thu

he/heo
we

helPs

help
help
help

Similarly, Modern Standard Engiish has lost the three noun genders of Old
English illustrated in (216):

(216)

tha stanas
tha giefa

tha scipa

the stones (masculine)

the gifts (feminine)
the ships (neuter)

Over the centuries, then, morphological change in English has largely been in
a direction of radical reduction and simplification of inflections to an extent
not seen in most other Germanic languages.
The reduction of two former Old English inflections -inde and -ingel-ynge
to Modern Standard English -ing has had a considerable effect on present-day
variation in English. In most English speaking countries, there is social variation in the pronunciation of (ing), some pronouncing it [rg], which is the standard form, and others [rn], the wideiy used non-standard form. Sociolinguists
have found variation in (ing) particularly interesting for a number ol reasons.
First, a number of studies from around the English-speaking world have
found that, all else being equal, women use a higher proportion of the standard [r4] form than men. Some representative results appear in figure 42:

266

Lexical variation and change

100

90

Fso
h70

;60
o

tr50
o

9ao
a
o

S?o
E

820

A.

10

Sidney,
Australia

Wellington,

Norwich,

New Zealand

England

City

Figure 42 Speaker sex and lhe use of (ing) in casual speech in three
English-spe aking c i t ies
Secondly, variation in (ing) appears to be fairly stable over the entire speech
community of English. In other words, neither lorm seems to be replacing the
other, but there is a pattern of stable variation, with [rg] being the acrolectal
form (used in higher social classes and in more formal contexts) while [tn], the
basilectal form, is used among working class groups and in more informal contexts. Figures 43a and 43b support this assertionFinally, research has shown that people use different proportions of [t4] and
[rn] at different stages ol their life. A study in Norwich in Eastern England, for
example, lound that young people predominantly used the non-standard [tn]
form, but changed their behaviour in middle age to use a greater proportion
of the standard form, before reverting to a greater use of the non-standard
form again in retirement (see figure 44).
Peter Trudgill, who conducted the Norwich study, has suggested that people
come under the pressure of the standard variety more in their economically
active years than in their youth or in their retirement and that this would
account for the variation in (ing) use across a person's lifespan.
Synchronically, (ing) can be regarded as a phonological variable, the alternation of velar and alveolar nasal realisations of the final segment (ng).

Variation and change in morphology

267

100

90

q80
!

o870

;60
L

9so
o

ts40
o
30
&"
C
920

l0
o

lower middle upper

working working working
class
class
class

lower
middle

middle
middle

class

class

Social class

Figure 43a Social class and the

use

af (ing) in casual

speech in

Norwich
r00
90
,o

6
'd

80

70
60

o
o
o

a
C)

tro

o
o
o
o-

50
40
30
20

l0
0
casual
speech

formal
style

reading
a

passage

reading
a word list

Figure 43b Speech style and the use of (ing) among upper working
class residents

of Norwich

268

Lexical variation and change

100

90

Eso

.E

hla
c

;,60
o
c50
o

3+o
o
Sio
E

8zo

l0

10-19 20-29 30-39 40--49 50*59 6G{9

7O+

Age in years

Figure 44 Changes in the use of (ing) in Norwich across the

generations

Historically, however, it must be considered as a morphological variable since

[r4] and [ln] come from two different Old English morphemes and still retain
signs of their former grammatical roles within present day variation. The relevant changes between Old English and the English of about 1400 are set out
in table 32:

Changes since Old English

reduction of final
loss of final

/e/ lo /a/

/a/

reduction of -nd- cluster

English of about l4oo

Old English -inde

(verbal suffix)

Old English -ingel-ynge

(verbal noun suffix)

-inde

-tDe

-ind

-irJ

'tn
-in

-i4

-io

By the mid fifteenth century , -ing lrad encroached on -ind's territory as a

verbal suffix in the south of England, but retained its more restricted role in
the north and in parts of East Anglia. In Modern English times, we can see
that this geographical variation (-rn in the north and -ing in the south) has

Variation and change in

morphology

269

evolved nationwide into social and stylistic variation. The former roles of
-inde and -inge are, however, still reflected in present-day variation. Research
has demonstrated that [rn] is much more likely to be found in progressives
(Madonna is singing again) and verbal complements (I don't mind listening to
Madonna) than in nominal -rrg lorms (I don't like Madonna's singing).Tlne
-inl-ing alternation, therefore, retains morphological importance, as well as
being a salient marker of social and stylistic information around the Englishspeaking world (exercise 7).
trinally in this section we shall consider the role of social contact on
morphological variation- In section 4, we saw how the strength ol social networks in the speech community has a considerable effect on the maintenance
of local dialect forms and susceptibility to language change- In a study carried
out on the speech of the African American and White populations ol
Philadelphia, Sharon Ash and John Myhill have found that there is a strong
link between ethnicity, social network ties and the use of certain non-standard
morphological features. We have already noted that one prominent characteristic of African American Vernacular English (AAVE) is the absence of the
suffix -s as a marker of third person singular agreement (see the examples in
(214) above). Additionally, possessive -s is not used in this dialect and we find
examples such as those in (217):

(zt7)

a. I met his brother wife

His cat name is Peanut

Ash and Myhill's research has revealed that there is a strong relationship
between the use ol these AAVE features and the levels of social contact
between whites and African Americans in Philadelphia- Those blacks who
have very little contact with whites use the AAVE leatures most, while those
with more contact with the white population use them less. Similarly those
whites who have little contact with the African American community rarely if
ever use the AAVE features, while those who have more contact do use these
features, albeit rarely (see figure 45). Network links with other ethnic groups
have led, in this case, to a weakening in the use of the ethnic variety and the
adoption of linguistic features from'outside'.
In this part of the book, we have introduced a range of concepts which are
necessary for coming to terms with the rich variety of processes on which
different languages rely lor forming complex words, focusing on English in
section l0 and taking account of aspects of other languages in section I l. Just
as progress in understanding sound systems requires a way lor describing
sounds accurately (the IPA of section 2), so discussion of word formation is
dependent on classification oi words into certain types, and we took the first

t'

Lexical variation and change

270

@ Blacks with little contact with Whites

90

Blacks with considerable contact with Whites

n whites with considerable contact with Blacks

80

-E
d

7o

P60
o

940
S30
o

920
o

A.

l0
possessive -s deletion

third person present

s deletion

Morphological variable(figures for the fourth category of Whites with little contact with Blacks are too
small to show on a graph of this nature)

Figure 45 Ethnicity, levels of interethnic contact and the use

A

AV E morp

ho lo g ica

of

fea tn'e s

steps in this direction in section 9 (see also section I B). A parallel aim throughout sections 9 to I I has been to sketch a view of the lexicol representations
which are an integral part ol a grammar (see the introduction, p. 4), constituting, as they do, the lexicon. Such representations, as well as having phonological and syntactic aspects (see ( I I a)), also encode lhe meanings ol lexical items,
and section 12 has examined how such meanings might be described and also
raised the issue of how the overall structure of the mental lexicon might be

understood in terms of meaning relations; that is, as well as coming to terms

with the internal structure of a lexical representation, we proposed that
meaning relations such as hyponymy and meronymy are useful in determining
the ways in which lexical representations are related to each other.
With basic concepts in place, the next three sections of this part have sought
to establish their usefulness in the study of the acquisition of words by small
children (13), the processing and storage (in a mental lexicon) of words by
adults (la) and the difficulties in perceiving and producing words which can
arise as a consequence

of brain damage (15).

Exercises

2'7t

Finally, in section l6 we have examined variation with respect to two of the

principle components of the lexical entries, the semantic representation and
the morphological shape of word forms serving particular grammatical functions. We have seen cases where each of these may be subject to variation
within the speech of an individual, across social groups, between dialects and
at different stages in the historical development of a language or dialect. For
a full description ol a speaker's behavioul then, the simple representations we
have presented in section l0 are not fully adequate, however, we can be confident that they constitute the basic core or nucleus over which variation can be
defined.
Our last major theme is the sentence, to which we now turn in the final part
of the book.

i,,.9ie1t"1t9

,.sauna....
:.|sffa|6:,,

:i

t'

256

Lexical variation and change

(210)

television: Greek tele ('far')

Latin

vrsro

('sight, thing seen')

microscope: Greek mikros ('small') + Greek skopein ('observe closely')

photograph: Greek photo ('light') + Greek graphos ('written')

Borrowings, then, are words which originated in one language (or dialect), but
which have come to be used in another, even by people who don't speak the
'lending'language- These borrowings are very olten assimilated to the phonological and morphological structure of the new host language (exercises I, 2 and.3).

Register: words for brain surgeons and soccer players,

hairdressers and lifesavers

A register is the specialised vocabulary common to a particular trade,
occupation, topic or activity. Hairdressers, soccer players, brain surgeons and
undertakers all have specialised words or uses of words which refer to concepts
particularly common or specific to their activity or profession. As a soccer
player you might nutmeg your opponent (kick the ball between their legs) or
play a one-two; you might ask a hairdresser for afat top or abob, or need a
surf lilesaver to rescue you from a rip (a dangerous backcurrent on a surf
beach), but you are unlikely to ask a brain surgeon for a lobotomy.It is of some
interest that occupations, interests, etc. can have some impact on the important idea of a basic level of categorisation introduced in section 13. Thus,
whereas for many of us, dog corresponds to a basic level category, for those of
us preoccupied with dogs, the basic level shifts down to that of particular
breeds. Similarly, while all of us are familiar with such words as beech, ash and
elm,many of us are not in a position to distinguish these different types of tree,
For those of us who are arboreally challenged in this way, it is plausible to
suggest thal tree appears at the basic level in our categorisation systems.
llowever, this is not the case lor botanists and tree surgeons. For them, the
basic level of categorisation will be that ol beech, ash and elm or, indeed, the
more specific level of copper beech, mountain ash, etcOften, people consider that the registers of doctors and lawyers (and even

linguists) hinder communication and understanding. The term jargon' is

sometimes used to reler to the confusing registers ol particular occupations.
In some senses, registers are 'in-group' varieties, which lead to accurate and
speedy communication of information among those that know and use them,
but confuse those who don't. It is obviously important that the doctor tells the
nurse that you have had a coronary infarcture or a stress fracture of your left
tibia, but what you want to know is that you've had a heart attack or broken

Biscuit or

cookie?

257

your left leg. In this medical example, the use of a special register is clearly a
necessity - the leg, for example, has several major bones and it is vitally important for the nurse to know which one you've broken. Some registers, howeveq
are deliberately confusing so as to hinder understanding by outsiders. This
may be because the group speaking the particular register wants to maintain
a sharply contrasting identity, or maybe has something to hide (exercise 4).

Biscuit or cookie? Variation and change in word choice

Consider (21l):

(2ll)

ConcePt

Word to refer to concept

Britain:
. U.S.A.:
Australia:

biscuit
cookie
biscuit/cookie

The thin flat, often round, usually sweet, hard but crumbly thing we eat during
our coffee break is called different things in different English-speaking areas.
In Britain it is usually relerred to as a biscuit and in the USA it is a cookie,
Australia is experiencing the initial stages of language change with the word
biscait gradually losing out to the word cookieSuch geographical differences in word choice are well known. Most people
are familiar with the US-UK contrasts between sidewalk and pavement, gas
and petrol, pants and trousers, elevator and lift, vacation and holiday. Just as
borrowing is frequent in situations of language contact, as we saw earlier, it is
also very common when dialect contact arises- In the past century, within the

English-speaking world, most interdialect borrowings have come lrom

American English, with the newly borrowed words pushing out or beginning
to push out older words, usually of British English origin. Thus, we find examples such as those in (212), where in each case, the American English lorm is
replacing or has replaced the British English equivalent:

(212)

British English
housey

bakery
minerals
pictures

American English
bingo
baker's shop
soft drinks
movies

lorry

truck

chips

fries
potato chips

crisps