CHAPTER 3

WORD MEANING

Group 5 (4b) :
Arief maulana hasan
Isti siti patimah
Rian riswandi
Yanriski stiasih
 INTRODUCTION

In this chapter we turn to the study of word meaning or lexical

semantics the traditional descriptive aims of lexical semantics
have been:

(a) to represent the meaning of each word in the language and

(b) to show how meanings the meanings of word in language are
interrelated,
The meaning of a word is defined in part by its relation with
other word in this language.

For example:
I saw my mother just now

we know without any further information, that the speaker saw

a woman as we will see, there are a couple of ways of viewing
this: one is to say that this knowledge follows from the
relationship between the uttered word mother and the related
but unspoken word woman.
Another Example:

a. my bank manager has just been murdered

b. my bank manager is dead

c. rob has failed his statistics exam

d. rob hasn’t passed his statistics exam

e. this bicycle belong to Sinead

f. Sinead owns this bicycle
There are many different types of relationship such as murder
and dead, fail and pass, belong and own. That word usually
used by poets, philosophers and writers of law.

The study of word meanings, especially the changes that seems

take place over time are also the concern of philology, and of
lexicology.
As a consequence of these different interests in word meaning
there has evolved a large number of terms describing differences
and similarities of word meaning.
 WORDS AND GRAMMATICAL
CATEGORIES

It is clear that grammatical categories like noun, preposition etc.,

though defined in modern linguistics at the level of syntax and
morphology, do reflect semantic differences: different categories
of words must be given different semantic descriptions.

example :
a. Names e.g. Fred Flinstone
b. Common nouns e.g. dog, banana, tarantula
c. Pronouns e.g. I, you, we, them
d. Logical words e.g. not, and, or, all, any
Looking at the types of words, we can say that they operate
in different ways: some types may be used to refer (e.g.
names), others may not (e.g. logical words); some can only
be interpreted in particular contexts (e.g. pronouns), others
are very consistent in meaning across a whole range of
contexts (e.g. logical words); and so on. It seems too that
semantic links will tend to hold between members of the
same group rather than across groups. So that semantic
relations between common nouns like man, woman, animal
etc. are clearer than between any noun and words like and,
or, not, and vive-versa.
Note too that this is only a selection of categories: we will have
to account for others like verbs, adjectives adverbs,
prepositions, etc. Having said this, we deal mainly with nouns
and verbs in this chapter; the reader should bear in mind that
this is not the whole story.
 Words and Lexical Items

Words can be identified at the level of writing, where we are

familiar with then being separated by the white space, where we
can call them orthographic words. They can also identified at the
levels of phonology, where the same semantic word can be
represented by several grammatically distinct variants.
 Example: walks, walking, walked below are three different
grammatical words:

a. He walks like a duck.

b. He’s walking like a duck.
c. He walked like a duck.

For semantics, we will want to say these are instances of the

same lexeme, the verb walk.
 What is words?

Our first impulse: the word has been defined as the symbolic,
linguistic counterpart of a single concept.

Sapir (1949a:32) : the word is merely a form, a definitely material

molded entity that takes in as much or as little of the conceptual
material of the whole thought as the genius of the language cares to
allow.

Bloomfield (1984:178) : word is a free form which does not

consists entirely of (two or more) lesser free forms; in brief, a word
is a minimum free form.
There have been other suggestions for how to define words
grammatically :

Lyons (1968) for example, discusses another distributional

definition, this time based on the extent to which morphemes
stick together. This is shown by numbering the morphemes.
Internal cohesions (Lyons 1968: 202-4)
the1 +boy2 + s3 + walk4 + ed5 + slow6 + ly7 + up8 + the9 +
hill10
a. slow6 + ly7 + the1 + boy2 + s3 + walk4 + ed5 + up8 + the9 +
hill10
b. up8+ the9 + hill10 + slow6 + ly7 + walk4 + ed5 + the1 +
boy2 + s3
c.*s3 + boy2 + the1
d.*ed5 + walk4

This works well for distinguishing between the words walked an

slowly, but as we can see also leaves the as a problem case. It
behaves more like a bound morpheme than an independent
word : we can no more say

*boys the than we can say just the in isolation.

 We can see an example of three grammatical words
representing lexemes by looking at the word foot in the
following sentences:
He scored with his left foot.
They made camp at the foot of the mountain.
I ate a foot long hot-dog.

Each of these uses has a different meaning, and we can reflect

this by identifying three lexemes. We have three senses of the
word foot. We could represent this by numbering the senses:
foot1: part of the leg below the ankle;
foot2: base or bottom of something;
foot3: unit of length, one third of a yard.
 Once we have established our lexemes, the lexicon will be a
listing of them with a representation of:
The lexeme’s pronunciation;

1. Its grammatical status;

2. Its meaning;
3. Its meaning relations with other lexemes.
 Problems with Pinning Down Word
Meaning

Different native speakers might feel they know the meaning of a

word, but come up with somewhat different definitions. Some of
this difficulty arises from the influence of context on word
meaning, as discussed by Firth (1957), Halliday (1966) and Lyons
(1963). These contextual effects seem to pull word meanings in
two opposite directions.
The first, restricting influence is the tendency for
words to occur together repeatedly, called
collocation. Halliday (1966), for example, compares
the collocation patterns of two adjectives strong and
powerful, which might seem to have similar
meanings. Though we can use both for some items,
e.g. strong arguments and powerful arguments,
elsewhere there are collocation effects. For example
we talk of strong tea rather than powerful tea; but a
powerful car rather than a strong car. These
collocations can undergo a fossilization process until
they become fixed expressions. We talk of hot and
cold running water rather than cold and hot running
water, and say they’re husband and wife, rather than
wife and husband. A similar type of fossilization
results in the creation of idioms.
Contextual effects can also pull word meanings in the other
direction, toward creativity and semantics shift. In different
contexts, for example, a noun like run can have somewhat
different meanings
a. I go for a run every morning.
b. The tail-end batsmen added a single run before lunch.
c. The ball-player hit a home run.
d. We took the new car for a run.
e. He built a new run for his chickens.
f. There’s been a run on the dollar.
g. The bears are here for the salmon run.
 Some writers have described this distinction in terms of
ambiguity and vagueness. The basic idea is that in examples
of vagueness the context can add information that is not
specified in the sense to be selected. The problem, of course,
is to decide, for any given example, whether one is dealing
with ambiguity or vagueness. The main reason for this is
context. Ambiguity is usually more potential than real since
in any given context one of readings is likely to fit the
context and be automatically selected by the participants;
they may not even be aware of readings that they would
naturally prefer in other contexts. This means that they
involve inventing a sentence and a context where both
readings could be available. One test proposed by Kempson
(1977) relies on the use of abbreviatory forms like do so, do
so too, so do. These are short forms used to avoid repeating a
verb phrase, e.g:
a. Charlie hates mayonnaise and so does Mary.
b. He took a form and Sean did too.
 Such expressions are understandable because there is a
convention of identity between them and preceding verb
phrase; thus we know that in example a Mary hates
mayonnaise and in example b Sean took a form. Kempton’s
sense relies on this identity; if the preceding verb phrase has
more than one sense, then whichever sense is selected in this
full verb phrase must be kept the same in the following do so
clause.
a. Duffy discovered a mole.
b. Duffy discovered a small burrowing mammal.
c. Duffy discovered a long dormant spy.
This relies on the two meanings of mole, and it therefore a case of
lexical ambiguity. If we add a do so as in following example:
d. Dufy discovered a mole, and so did Clark.
 Whichever sense is selected in the first clause has to be
repeated in the second, i.e. it is not possible for the first clause
to have mammal interpretation and the second the spy
interpretation, or vice versa. By contrast where a word is
vague, the unspecified aspects of meaning are invisible to this
do so identity. Basically, they are not part of the meaning and
therefore are not available for the identity check. We can
compare this with the word publicist which can be used to
mean either a male or female, as below show:

a. He is our publicist.
b. She is our publicist.

And in the example below:

c. They hired a publicist and so did we.

It is quite possible for the publicist in the first clause to be male
and in the second is female. Thus this test seems to show that
publicist is unspecified, or ‘vague’ for gender. We can see that the
vagueness allows different specifications in do so clauses, but the
different senses of an ambiguous word cannot be chosen.
 This do so identity seems to work, but as mentioned earlier, its
use relies on being able to construct examples where the same
sentence has two meanings. Other tests for ambiguity rely on one
sense being in a network of relations with certain other lexemes
and another sense being in a different network. So for example
the run above, might be in relation of near synonymy to another
noun like jog, while run in ex e might be in similar relation to
nouns like pen, enclosure, etc. Thus while the b sentences below
are fine, the c versions are bizarre:

a. I go for a run every morning.

b. I go for a jug every morning.
c. ?I go for an enclosure every morning.

a. He built a new run for his chickens.

b. He built a new enclosure for his chickens.
c. ?He built a new jog for his chickens.
This sense relations test suggest that run is ambiguous between the
example a and example e in readings.

There are a number of other tests for ambiguity, many of which are
difficult to apply and few of which are controversially successful;
see Cruse (1986: 49-83) for a discussion of the tests. It seems likely
that whatever intuitions and arguments we come up with to
distinguish between contextual colouring and different sense, the
process will not be an exact one.