Salahaddin University-Erbil

College of Languages

English Department

MA Studies/ 1st semester

Differentiating Homonymy from Polysemy

A Term Paper submitted to the English Department,

College of Languages as part of the requirement for MA

course in Semantics

By

Date

January 2021
Abstract

This paper deals with the relations between polysemy and homonymy, and how
they are distinguished. It aims to give the criteria that are used to differentiate
between the two terms. It also determines which criterion is better and what is the
weakness of the criteria. The paper tells as how the dictionaries usually treat
homonyms and polysemous words. It also tries to give some solutions to solve the
ambiguity between the two. Actually the problem is left without a clear-cut
solution. They cannot be distinguished easily.

Keywords: Homonymy, Polysemy, Etymology, Dictionary

Introduction

Homonymy and polysemy are two kinds of sense relations.

Homonyms are words which have the same (written or spoken) form but
different meanings. Homonyms could be absolute which means the
words are identical in all their forms and are from the same part of
speech, or they could be partial which means the words with different
meanings are identical in some of their forms but not in all. Polysemy is
the characteristic of one word with a number of different related
meanings. Two criteria are usually used to differentiate between
homonymy and polysemy which are: Etymology and relatedness of
meaning. When the two different meanings of a form have the same
origin, it is considered as one polysemous word; but when they are from
different sources, they are considered as homonyms.
Polysemy

The word polysemy is from Greek (“poly” many and “semy” meanings). It can be
defined as one form (lexeme, word) which has several different meanings that are
related to each other, (Yule, 2017).
The following examples illustrate this phenomenon.
For example the word “head” can refer to different senses:
1. Dr. Charles is the head of the department.
2. He kissed the top of her head.
3. The problem is inside her head.(mind)

In the examples above the word “head” has three different meanings.

In the first example it means the person who is in charge of the department,
in the second one it means the part of the body which is on top of our body, and in
the third one it means our mind.

The word “foot” can refer to foot of person, foot of bed, and foot of mountain; thus
it is an example of polysemy. The word “date” has these meanings: point in time, a
romantic meeting and an appointment.

4. What date is it today? (Point in time)

5. I’ve got a date with Jane. (Romantic meeting)

6. Let’s make a date to see each other. (An appointment)

Polysemous words are given one entry in the dictionaries, i.e. the distinct
senses of the word are listed under one major heading, (Croft and Cruse, 2017).
Etymologically, Polysemous words are from one source, which means the different
senses of one lexeme are from one common origin, (Palmer, 1976).
Al-Suliman (2016) states that “polysemy is a matter of single lexemes in single
languages”(p. 212).

What Al-Sulaiman wants to say here can be illustrated by words that have
both a “literal” meaning and one or more metaphorical (transferred) meanings as
the case of colour adjectives in English.

For instance, “green” can mean “unripe” based on a metaphor that most of the
fruits have green colour when they are not ripe. And from this the meaning of
“immature” taken on the basis of a metaphor that builds similarities between the
process of getting experience and the process of ripening in fruits. This multiplicity
of meaning of “green” here as it can be seen in English may not be seen in other
languages by the same sort of metaphorical motivation.

Moreover this cannot be observed su for other colour adjectives. For

example, although the majority of the fruits have red colour when they are ripe, but
red does not mean “ripe”, or “mature”. Similarly in German, “blau” can mean
“drunken” as well, but in English “blue” does not have such a meaning.

In short Polysemy is the property of one lexeme having a number of distinct

meanings which are semantically linked.

Homonymy

When a single form, either written or spoken, has more than one unconnected
meaning, i.e. totally different meanings, it is called homonymy, (Yule, 2017).
Palmer (1976) argues that homonymy is a number of words with an identical form.
Examples of Homonymy:

Bank (financial institution) _ bank (ground alongside a river)

Race (contest of speed) _ race (ethnic group)
Pen (writing instrument) _ pen (enclosed space)
Pupil (Student)_ Pupil (a part inside the eye)
Date (point in time) date (a kind of fruit)
Each pair of these examples above can be seen as two words which have
completely unrelated meanings, but having the same form. Therefore, they are
called homonyms. Alsulaiman (2016) classifies homonyms into two types:
Homophones and homographs. The former means that two words or more are
written differently but pronounced the same that have completely different
meanings such as:

Sight/ site; right/ rite/write; son/sun; sew/so.

And homographs are words which are similar in writing but pronounced
differently such as:

Lead /li:d/ (the first position in a race)

Lead /led/ (a heavy metal)

Kinds of homonymy

Absolute homonymy

Palmer (1976) believes that words are called absolute homonyms when they
meet three criteria: first, the meaning of the words should be totally different, i.e.
there is no relation between the meaning of the words. Secondly, they will be in
the same shape in all of their forms. And the last, they should be grammatically
equal.

There are many examples of absolute homonymous words:

Bark (outer part of a tree) bark (the sound a dog)
Spring (a season) spring (coiled metal)
Bank (financial institution) bank (side of a river)
In the last example, it is noticed that ‘bank₁’ has a totally different meaning
which has no link with the meaning of ‘bank₂’, but they are grammatically
equivalent, i.e. both are nouns and all their forms are identical.
In brief, for homonyms to be called absolute, they must meet the three
criteria mentioned above.

Partial homonymy

Lyons (1995) explains that partial homonymy is a case where there is a

shared form between two words with different meanings but not all their forms are
alike and/or they are not equal grammatically. In other words, one or two of the
criteria of absolute homonymy is missing.

For instance, “bear₁” (n) which is (an animal) and “bear₂” (v) which means (to
stand, to deal with) are examples of partial homonymy because they are not
grammatically equivalent.

As Lyons (1995) has exemplified, sometimes partial homonymy causes

great ambiguity. For example both verbs “find” and “found” have one identical
form which is found, but they have also other different forms from each other;
finds, finding, or founds, founding. Moreover, found as a form of ‘found’ is not
grammatically equal to found as a form of ‘found’. Here, as usual in English both
the last two criteria are missing. In the case of this example, the identical form
‘found’ of both verbs find and found is a transitive verb. Therefore, there are many
cases that found can be understood as form of ‘found’ and ‘find’.

(7) We found some buildings.

This sentence can be understood as present which contains the present form
of the verb ‘found’ or, as a past tense sentence which contains the past form of the
verb ‘find’. This sentence is lexically and grammatically ambiguous.

We can solve the ambiguity of (7) by putting have before found as in:

(8) We have found some buildings.

(9) We have founded some buildings.
In (8) found can be just understood as the past participle of the verb ‘find’; while in
(9) founded is the past participle of the verb ‘found’. The ambiguity of (7) will also
disappear if we change we to he or she as in (10).

(10) He/ she found some houses.

While in English singular and plural forms are identical in the past tense
verb-forms, there is a difference between third-person singular and plural forms in
the present simple tense. Therefore, in (10) found is recognised as the past form of
‘find’.

Distinguishing homonymy from polysemy

Polysemy is a case where a single word has a number of different senses

Palmer (1976). This idea of different meanings of a word has some problems.
 First, it is not so easy to decide whether two meanings are similar or distinct,
and for that reason identifying the number of meanings that a word has is
ambiguous. For example, the verb eat. In the dictionary three meanings are
distinguished for eat, the literal meaning which is ‘taking food’ and two derived
senses which are ‘use up’ and ‘corrode’. But ‘eating soup’ and ‘eating meat’ can
give different meanings in that what we use in eating each, in the former a spoon is
used and in the latter a knife and fork. Other than this, we can also differentiate
between ‘drinking soup’ and ‘eating it’. So here eat represents drink. The problem
is that whether to decide that they are two different meanings or not. It is clear that
there are different kinds of food which we eat in different ways. So shall we decide
that the word eat has a different meaning with each specific food? The right
decision is that we shall not consider all the possible different meanings, but we
should look for similarities of meanings as far as possible (Palmer, 1976).

The second problem as Palmer (1976) has mentioned is that we cannot make
any generalization about “differences of meaning”. There are words which have
both a literal meaning and one or more derived meanings by metaphor. The most
common examples are words of the body parts, foot, hand, eye, tongue, leg, etc.
Therefore, we use these expressions, hands of a clock, leg of a chair, eye of a
needle, foot of a mountain, etc. Metaphor is something almost random. We may
say that foot is suitable for mountains based on a metaphor that foot is the lowest
part of the body and eye for needles, but if we look at other languages, we see that
it is not the case. For instance in French, there is not such an expression.

The third problem of one linguistic form with a number of different

meanings as Palmer (1976) has pointed out is that whether we consider this form a
polysemous word ( one word which has more than one different related meaning)
or as an instance of homonymy (different words which are alike in shape). For
example, in the dictionary, flight is given one entry and therefore it is treated as
one word, but bank is given two entries for (financial institution) and (side of a
river) and is considered as two different words.
 These problems mentioned, to a great extent, cannot be solved easily and
make our understanding of meaning of words more difficult.

Criteria of identifying homonymy and polysemy

What is the difference between homonymy and polysemy? How can we

distinguish between homonymy and polysemy? These questions are not so easy to
answer as we said before. The difference between the two is not clear-cut.
However, there are some possible criteria to identify a form which has several
meanings as a polysemous word or as two homonymous words.

As mentioned in previous paragraphs, dictionaries give a single entry for

polysemous words, but separate entries for homonyms. What do the dictionaries
base their decision on, then?

There are usually two criteria that dictionaries depend on: etymology and
relatedness of meaning (Lyons, 1995). Etymology means the historical origin of
words, the diachronic study of words. When the same forms are from different
origins, they are considered as homonyms and the dictionary give them different
entries; but when they are known to be from the same etymological origin, they are
considered a single word and therefore the different meanings are listed under one
entry (Palmer, 1976). Croft and Cruse (2004,p.111) argue that “homonymous units
are derived from distinct lexical sources, and their orthographical/phonological
identity is due to either the loss of an original distinction due to language change,
or to borrowing, whereas polysemic units are derived from the same lexical source,
being the result of processes of extension such as metaphor and metonymy”.
For example, native speakers by intuition consider ‘bat₁’ (a small animal with
wings usually flies at night) and ‘bat₂’ (a wooden stick used in some sports to hit
the ball) different words and they are really from distinct origins historically. The
first one is derived from Middle English ‘bakke’ and the second one is from Old
English ‘batt’ (Lyons, 1995).
 Semanticists as Lyons (1995) and Palmer (1976) argue that the etymological
criterion is not satisfactory to differentiate polysemy from homonymy, since in
many cases, the historical source of language does not necessarily represent its
present state. There are words which are etymologically from the same sources, but
are treated as different words in present language. For instance, the word ‘pupil₁’
meaning (student) and ‘pupil₂’ meaning (the black area in the middle of the eye)
are considered as different words, i.e. homonyms today, but historically, they are
from one common source. So if we base our decision on etymology, they should be
examples of polysemy.

On the other hand, there are examples of words which are historically from
different sources, i.e. examples of homonymy, but they are treated as one word
with different meanings in the language by native speakers, i.e. they are examples
of polysemy Lyons (2005). For instance, the word ‘shock₁’ as in ‘shock of hair’
and ‘shock₂’ in ‘shock of corn’ are treated as one word, however, they are
historically from distinct sources.

Other examples are the words of parts of the body which are used
metaphorically for other things as we speak of hands of a clock, foot of a mountain,
eye of a needle, etc. and these are all recognised as examples of polysemy. And
similarly the word ear is used for the ear of corn and is regarded as one word
which has different meanings and so, an example of polysemy. Yet, historically
speaking the etymologists say that the ear of corn is not related to the ear of the
body in no way, then they are homonyms. But most native speakers of English
nowadays regard them as the same word with distinct meanings i.e. an example of
polysemy,(Palmer, 1976).

Moreover, In English there are words with different spelling, but

pronounced the same which are called homophones. There are pairs of
homophones that are from the same origin historically, however they have a
difference in spelling. Examples are metal and mettle, flour and flower. So if we
only depend on history, these pairs should be regarded as one word and an
example of polysemy. But this will be strange. Can words that have a difference in
spelling be regarded as one polysemous word? The dictionaries usually treat these
words as different words, for they list words in an alphabetical order, (Palmer,
1976).

In short, the etymological criterion, the history of words, is not reliable all the time.

Relatedness of meaning is another criterion to decide between homonymy

and polysemy. When the different meanings of a word are related as in ‘neck ₁’
meaning ‘part of body’, ‘neck₂’ meaning ‘part of shirt’, ‘neck₃’ meaning ‘ part of
bottle’ and ‘neck₄’ meaning ‘ narrow strip of land’, they are all regarded as one
polysemous word with related meanings. But when the meanings of a lexeme are
completely different and unrelated, even if they have the same historical origin,
they are treated as homonyms. For example, as we mentioned before ‘pupil ₁’
which means (a child studies at school) and ‘pupil₂’ (the black part of the eyeball)
are from the same origin, but synchronically, looking at language at the present
day, because the meanings are unrelated, they are considered as different words,
i.e. homonyms, (Lyons, 1981).

Another way of distinguishing homonymy from polysemy as Palmer (1976)

argues is the number of antonyms that a word has for each different meaning. For
instance the word fair is used with a number of words like: (1) hair, (2) skin, (3)
weather, (4) sky, (5) judgement, and (6) tackle. The antonyms will be dark for (1)
and (2), foul for number (3) and (6), cloudy for (4) and unfair for (5). Here, we see
that the word fair with hair and with skin has the same antonym (dark) and fair
with weather and tackle as well (foul). It can be considered as polysemy when
there is the same antonym for different meanings and homonymy when the
antonyms are different. But this implies that fair with weather is more similar to
fair with tackle than with sky. However, it seems clear that weather is more related
to sky and tackle to judgement, but the antonyms does not give us any evidence for
that.
 So, all the criteria used for distinguishing homonymy from polysemy have
some issues and cannot give clear-cut distinction between the two.

Solutions for Deciding between Homonymy and Polysemy

Some studies of semantics at the present day suggest that we should cut the
“Gordian knot” and assume homonymy in all the words with different meanings
and neglect polysemy. However, this assumption seems attractive, but it does not
solve the issues and it misses the “theoretical point”. The number of different
meanings of words is not exactly determined. With passing time, words may give
other meanings that were not seen before. Meanings are ‘indefinitely extensible’
(Lyons, 1981).

Lyons (1981) also suggests another solution for distinguishing between

polysemy and homonymy which is neglecting the “semantic criteria” when we
define lexemes and depending on only “syntactic and morphological criteria”. This
criterion makes the words like ‘bank₁’ (financial institution) and ‘bank₂’
(alongside the river) which were regarded as two homonymous words into one
polysemous word with different meanings because there is no difference in their
syntactic and morphological structure. However, many scholars do not like this
solution, yet it is much supported than the other one.

Thus that the problem of differentiating between homonymy and polysemy

is not soluble.

Conclusion

The number of different meanings of a word is not clearly determined.

A form that has several different meanings is not easily identified as one word with
different meanings, i.e. a polysemous word or two different words with the same
shape, i.e. homonyms. Homonyms can be absolute or partial.
The etymological criteria that dictionaries usually base their decisions to
differentiate between homonymy and polysemy is problematic and cannot
solve the problem properly. The best criteria to differentiate homonymy from
polysemy is relatedness of meaning. The two cannot be easily distinguished
from each other. The two solutions which have been suggested for solving
the problem between polysemy and homonymy which are: considering
homonymy in all the cases or depending on syntactic criteria rather than
semantic criteria are refused by many scholars. Actually the problem is left
without a clear-cut solution.

References

Al-Sulaiman, M. (2016) Semantics and Pragmatics. Haval Art Printing Press

Lyons, J. (1981) Language and Linguistics. Cambridge, Cambridge University
Press
Lyons, J. (1995) Linguistic Semantics. United states of America, Cambridge
University Press
Palmer, F. R. (1976) Semantics, A New Outline. Great Britain, Cambridge
University Press
Yule, G. (2017) The Study of Language. 6th ed.,United Kingdom, Cambridge
University Press.