Lexical Ambiguity Revisited On Homonymy PDF
Lexical Ambiguity Revisited On Homonymy PDF
Lexical Ambiguity Revisited On Homonymy PDF
Abstract: The aim of this paper is to assess the two descriptively controversial cases of lexical
ambiguity known as polysemy and homonymy in an attempt to illustrate a practical solution
to the lexicographic treatmentă oПă tСeseă pСenomenКă ЛКseНă onă PТerreă όrКtС’să reПerentТКlă
КpproКМСă КnНă onă InРrТНă δossТusă όКlФum’să prКРmКtТМă КpproКМСέă ψotСă КММountsă restă onă tСeă
mind-world relations that characterise the representation of polysemy and ultimately point to
a combination between conceptual atomism and pragmatic-focused analysis as the framework
most likely to solve the paradox of polysemy.
Lexical ambiguity is considered the most common form that this phenomenon
characteristic of natural languages tends to take, due to the numerous occurrences of
polвsОЦousă аorНsă КЧНă toătСОă vКrТousă МКsОsă oПă СoЦoЧвЦвă ПОКturТЧРă ТЧă ЦКЧвă oПă tСОă аorlН’să
languages. However, it is also the most controversial, both theoretically and practically, since
speakers of a language cope with polysemy with little, if any, effort, whereas scholars find it
difficult to define and account for this linguistic phenomenon, which has caused some
lТЧРuТstsă toă tКlФă КЛoută Кă “polвsОЦвă pКrКНoб”ă (Ravin and Leacock, as well as Taylor). This
paradoxical situation results from the problems linguists have encountered in their attempts to
clearly distinguish between polysemy and homonymy proper, on the one hand, and to offer a
transparent but clear-cut description of lexical ambiguity on the other.
TСОă аorНă “polвsОЦous”ă Тsă ПКră ПroЦă ЛОТЧРă polвsОЦousμă Тtă СКsă ЛООЧă РТvОЧă oЧlвă oЧОă
НОПТЧТtТoЧ,ăЧКЦОlвă “СКvТЧРă ЦultТplОă ЦОКЧТЧРsăorăТЧtОrprОtКtТoЧs”,ăКsăopposОНătoăСКvТЧРăoЧlвă
one meaning or interpretation and, thus,ă ЛОТЧРă “ЦoЧosОЦous”ă – e.g. the word walk is
polysemous, as proven by the following examples:
The first known use of the word polysemous dates back to 1884, but the notion was
made popular by Michel ψrцКlăТЧă1κλιέ Merriam-Webster Dictionary traces the origin of the
term, to the Late Latin polysemus, from the Greek polвsēmos (-poly- many + -sēmК- sign).
Thus, polysemy is a characteristic displayed by some words and phrases that may enjoy
multiple yet somehow related interpretations.
1047
SECTION: LANGUAGE AND DISCOURSE LDMD I
TСОă аorНă “СoЦoЧвЦous”,ă ПТrstă rОМorНОНă ТЧă 1θβ1,ă МКЦОă vТКă tСОă δКtТЧă homonymus,
from the Greek Сomōnвmos, meaning “oПătСОăsКЦО ЧКЦО”,ă-homo- same + -ōnвmos- named).
In its strictest sense, a homonym is a word that not only sounds the same, but is also spelled
the same as another word which has a different meaning: fluke – a parasitic flatworm of the
class Trematoda, fluke – pКrtăoПăКăаСКlО’sătКТlăorătСОăЛКrЛОНăСОКНăoПăoЧăКЧăКrroаăorăКăСКrpooЧă
or an anchor, and fluke – a chance occurrence, a fortunate accident, a stroke of good luck
(The Free Dictionary). There is, however, a looser sense of tСТsătОrЦμă“КăаorНătСКtăsouЧНsăor
is spelled the same as another word but has a different meaning, technically called a
homophone (same sound: e.g. die and dye) or a homograph (same spelling: e.g. the metal lead
and the present tense of the verb lead)”ă(Wiktionary). Thus, homonymy is a feature displayed
by a word that sounds the same and/or is spelled the same as another word, but has a
different, unrelated meaning. This linguistic relation may either be complete (full
homonymy) or partial (homophony or homography).
According to the above-mentioned definitions, if a word displays multiple rather
similar meanings as part of a large semantic field, it is a case of polysemy:
get can mean - procure (I’llăget the money.)
-fetch (I’llăget the kids.)
-become (ώО’ll get angry.)
- understand (I get it.)
fleet can mean - КllătСeăsСТpsăoПăКăМountrв’sănКvв
(The Spanish fleet appeared from behind the dark clouds.)
- a number of ships, aircraft or road vehicles managed as a unit
(εвăunМle’sătrКnsportăМompКnвăСКsăКălКrРe fleet of service vehicles.)
Conversely, if two or more distinct concepts share the same name, it is a case of homonymy:
To sow the seeds of revolt is all that he wanted. (the verb – to plant) versus Have you found
the sow? (the noun – female pig).
Not all cases are clear-cut, however, since relatedness is clearly a rather vague notion
and, as such, does not constitute an infallible test for polysemy. Leech (227) shows that this
concept is two-fold: two or more meanings are historically related if they originated from the
same source, but they may also be psychologically related if speakers of a language
intuitively feel that there is a link between those meanings. Moreover, historical relatedness
does not imply psychological relatedness, and vice-versa1284. όТllЦorОăКЧНăχtФТЧs’săНОПТЧТtТoЧă
of polysemy (100) is three-fold: 1. the various senses of a polysemous word have a central
orТРТЧνăβέătСОălТЧФsăЛОtаООЧătСОsОăsОЧsОsăПorЦăКăЧОtаorФ,ăКЧНăγέăuЧНОrstКЧНТЧРătСОă‘ТЧЧОr’ăoЧОă
contributes to the understanding of tСОă ‘outОr’ă oЧОέă SТЧМОă etymology may help, but is not a
definitive test, lexicographers also rely on ЧКtТvОă spОКФОrs’ă ТЧtuТtТoЧă КЧНă УuНРОЦОЧt when
1284
δООМСă(1λλίμăββκ)ăquotОsăStОpСОЧăUllЦКЧЧ’săОбКЦplОμă“ОКr”ăοă"organ of hearing" Old English ēКre, from
Proto-Germanic *auzon (cf. Old Norse eyra, Danish øre, Old Frisian are, Old Saxon ore, Middle Dutch ore,
Dutch oor, Old High German ora, German Ohr, Gothic auso), from PIE *ous- with a sense of "perception" (cf.
Greek aus, Latin auris, Lithuanian ausis, Old Church Slavonic ucho, Old Irish au "ear," Avestan usi "the two
ОКrs")ăvОrsusă“ОКr”ăοă"grain part of corn" from Old English ēКr, (West Saxon), æСer (Northumbrian) "spike, ear
of grain," from Proto-Germanic *akhaz (genitive *akhizaz; cf. Dutch aar, Old High German ehir, German иСre,
Old Norse ax, Gothic ahs "ear of corn"), from PIE root *ak- "sharp, pointed" (cf. Latin acus "husk of corn,"
Greek akoste "barley;" see acrid). See the Online Etymology Dictionary available at:
http://www.etymonline.com/index.php?term=ear Although historically they are homonyms, present-day
speakers view them as polysemous words.
1048
SECTION: LANGUAGE AND DISCOURSE LDMD I
deciding whether to assign different entries or only one to words whose status is ambiguous.
The solution that Leech (228) advances to the polysemy - homonymy problem depends on
acknowledging the psychological reality of lexical rules: if people identify a metaphorical
МoЧЧОМtТoЧă ЛОtаООЧă ЦОКЧТЧРs,ă “аСКtă ПroЦă Кă СТstorТМКlă poТЧtă oПă vТОаă Тsă Кă СoЦoЧвЦв,ă
resulting from an accidental convergence of forms, becomes reinterpreted in the context of
present-НКвăEЧРlТsСăКsăКăМКsОăoПăpolвsОЦв”έăTСОăМlКssТПТМКtТoЧăoПăаorНsăКsăpolвsОЦousărКtСОră
than homonymous based on the intuition criterion is bound to be subjective. We might be
dealing with polysemy, even though the various meanings are relatively far apart, or with
homonymy, despite the overlap between different meanings. As a result, some speakers may
treat the respective item as one word, whilst others might think that there are two or more
different words. Thus, even though the different meanings of polysemous words should be
treated under a single dictionary lemma, whereas homonyms should feature as separate
lemmata, this is not always the case: in some dictionaries even polysemes are listed as
different entries, whereas in others homonyms feature together, under a single entry. This
situation is, explains Frath (50-η1),ă ТЧă НТrОă ЧООНă oПă “an exhaustive semiotic and cultural
analysis”ă– basically, the confusion springs from our tendency to view language and reality as
ТsoЦorpСТМ,ăаСТМСăТs,ăoЛvТouslв,ă“КăПКllКМв”μă
Mankind has kept trying, at least since Aristotle, to organise and comprehend the
real world, essentially by matching it with language. Language is a repository of
kЧoаlОНРОă РКТЧОНă Лвă ЧКЦТЧРă КЧНă rОlКtТЧРă oЛУОМtsέă […]ă When words refer to
separate objects, we quite naturally believe that the objects and the words must
share some features. Yet it fails with homonyms, because of some real world
discrepancies or because we have forgotten why our forebears named these
objects so. And we wonder: why are these two very different things referred to by
the same word? We name the mystery homonymy, and add it to the repository.
Homonymy is disappointed polysemy; we resort to it reluctantly.
To the human mind, even historically unrelated items can appear to be somewhat
related. A connection can be found, with a stretch of imagination, by invoking some
suЛsuЦТЧРă МoРЧТtТvОă ОlОЦОЧtsă lТФОlвă toă МoЧstruОăКЧă “uЧНОЧТКЛlО”ă lТЧФă ЛОtаООЧă tаo or more
meanings. Thus coerced into resemblance, historical homonyms are turned into polysemous
words, due to our overwhelming preference for polysemy and distaste for homonymy.
The boundaries between polysemy and homonymy seem to be characterised by flux,
rather than fix, as a result of semantic shift, referential dispersion, semantic closeness, human
subjectivity and, first and foremost, a lack of firm ground-rules on the basis of which the
distinction between the two is to be drawn.
As Frath (44) shows, the lexicographic treatment of polysemy and homonymy is
bound to be problematic, since a dictionary is not meant to explain language, but rather to
provТНОă“a snapshot of usage at one particular moment”έăTСОăsОЦКЧtТМătrОКtЦОЧtăoПăpolвsОЦв,ă
however, should not normally be hindered to the same extent, as it may offer an accurate and
detailed description of language from a theoretical point of view. Two different approaches
have been identified: one that operates with literal versus figurative meanings, with
metonymical versus metaphorical connections, and another that operates with core versus
specific meanings triggered by context or by generative rules. Frath (47-48) argues that the
1049
SECTION: LANGUAGE AND DISCOURSE LDMD I
former, known as the linear theory, fails to work because, on the one hand, there is no clear-
Мută ЦОtСoНă oПă ТНОЧtТПвТЧРă аСТМСă tСОă lТtОrКlă ЦОКЧТЧРă Тsă КЧН,ă oЧă tСОă otСОră СКЧН,ă “tСОă lТЧФă
between literal and derived meanings cannot always be specified with certainty, even when it
is established.”ăTСОălКttОr,ăФЧoаЧăКsăthe subsuming theory, also fails on two counts: firstly,
it is unable to explain exactly how speakers use the context to carry out the selection process
КЧН,ăsОМoЧНlв,ăТtăМКЧăЧОТtСОră“НОПТЧОătСОăorТРТЧКlă(lТtОrКlăorăsuЛsuЦТЧР)ăsОЦКЧtТМăОЧtТtв”,ăЧoră
“ОбplКТЧătСОălТЧФsăЛОtаООЧătСОăorТРТЧКlăКЧНă(НОrТvОНăorăsuЛsuЦОН)”ăЦОКЧТЧРsăoПăКăаorНέ
SСoаТЧРă tСКtă “polвsОЦвă Тsă Чotă Кă sТЧРlОă СoЦoРОЧОousă pСОЧoЦОЧoЧ”,ă όrКtСă (ηη-56)
attempts to explain it in terms of reference. Thus, if a word can be used to refer to various
aspects of the same object, the case is that of referential polysemy; if a word is employed to
refer to a number of similar objects, the case is that of lexical polysemy; and, finally, if words
identical in some respect are used to refer to unconnected objects, the case is that of
homonymy:
τЧОăoПătСОăЦostăПrОquОЧtlвăquotОНăКЧНăНТsМussОНăМКsОsăТsătСКtăoПătСОăаorНă“ЛКЧФ”έăTСОrОă
are several reasons for its popularity. Firstly, oЧО’să ТЧtuТtТoЧă ЦТРСtă НТМtКtОă tСКtă tСОă sКЦОă
combination of letters/sounds cannot possibly be treated both as one word with several related
meanings and as two different words with unrelated meanings (web York handout).
Nevertheless, if one employs the criterion of psychological relatedness with words like
“ЛКЧФ”,ă oЧОă ЧООНsă toă НТstТЧРuТsСă ЛОtаООЧă tСОă КppКrОЧtlвă НТППОrОЧtă ЦОКЧТЧРsă – financial
institution versus slope adjoining a river versus row of similar things – whilst acknowledging
that for each of these homonyms there are also polysemous uses, as the following entries
featured in Collins English Dictionary clearly show:
bank1
n
1. (Economics, Accounting & Finance / Banking & Finance) an institution offering
certain financial services, such as the safekeeping of money, conversion of domestic
into and from foreign currencies, lending of money at interest, and acceptance of bills of
exchange
2. (Economics, Accounting & Finance / Banking & Finance) the building used by such
an institution
3. a small container used at home for keeping money
4. (Group Games / Gambling, except Cards) the funds held by a gaming house or a
banker or dealer in some gambling games
5. (Group Games / Card Games) (in various games)
a. the stock, as of money, pieces, tokens, etc., on which players may draw
b. the player holding this stock
6. any supply, store, or reserve, for future use a data bank a blood bank
1050
SECTION: LANGUAGE AND DISCOURSE LDMD I
bank2
n
1. a long raised mass, esp. of earth; mound; ridge
2. (Earth Sciences / Physical Geography) a slope, as of a hill
3. (Earth Sciences / Physical Geography) the sloping side of any hollow in the ground,
esp. when bordering a river the left bank of a river is on a spectator's left looking
downstream
4. (Earth Sciences / Physical Geography)
a. an elevated section, rising to near the surface, of the bed of a sea, lake, or river
b. (in combination) sandbank, mudbank
5. (Mining & Quarrying)
a. the area around the mouth of the shaft of a mine
b. the face of a body of ore
6. (Engineering / Aeronautics) the lateral inclination of an aircraft about its longitudinal
axis during a turn
7. (Engineering / Civil Engineering)
Also called banking, camber, cant, superelevation: a bend on a road or on a railway,
athletics, cycling, or other track having the outside built higher than the inside in order
to reduce the effects of centrifugal force on vehicles, runners, etc., rounding it at speed
and in some cases to facilitate drainage
8. (Group Games / Billiards & Snooker) the cushion of a billiard table
bank3
n
1. an arrangement of objects, esp. similar objects, in a row or in tiers a bank of dials
2. (Transport / Nautical Terms)
a. a tier of oars in a galley
b. a bench for the rowers in a galley
3. (Communication Arts / Printing, Lithography & Bookbinding) a grade of lightweight
writing and printing paper used for airmail letters, etc
4. (Electronics & Computer Science / Telecommunications) Telephony (in automatic
switching) an assembly of fixed electrical contacts forming a rigid unit in a selector or
similar device
Secondly, according to the Online Etymology Dictionary, the first two entries are, indeed, to
be viewed as homonyms:
“ПТЧКЧМТКlă ТЧstТtutТoЧ”,ă lКtОă 1ηМέ,ă ПroЦă ОТtСОră τlНă ItКlТКЧ banca or Middle
French banque (ТtsОlПăПroЦătСОăItКlТКЧăаorН),ăЛotСăЦОКЧТЧРă“tКЛlО”ă(tСОăЧotТoЧăТsăoПătСОă
ЦoЧОвlОЧНОr’să ОбМСКЧРОă tКЛlО),ă ПroЦă Кă ύОrЦКЧТМă sourМОă (МПέă τlНă ώТРСă
German bank “ЛОЧМС”)έ
1051
SECTION: LANGUAGE AND DISCOURSE LDMD I
“ОКrtСОЧă ТЧМlТЧО,ă ОНРОă oПă Кă rТvОr”,ă Мέ1βίί,ă proЛКЛlвă ТЧă τlНă EЧРlТsСă Лută Чotă КttОstОНă ТЧă
surviving documents, from a Scandinavian source such as Old Norse banki, Old
Danish banke “sКЧНЛКЧФ”,ă ПroЦă Proto-Germanic bangkon “slopО”,ă МoРЧКtОă
with bankiz “sСОlП”έ
Nevertheless, more recent etymological information about the lexical item in question
actually suggests that the two should feature under one lexical entry, since, as Frath (50-51)
explains, Bank-mound, bank-financial institution and river bank actually have a common
origin, the Gothic word benc, which originally referred to a small mound. Then it was also
used metaphorically for river bank, and also as a metaphor for a low table, especially those
where bankers used to exchange money, which gave Italian banca, French banque, and
English bank.
Lyons (28) also points out that the two senses of the word bank did coincide, at one
point in the evolution of the Italian language, as bankers used to sit on the riverbanks when
conducting business. Thus, going to the financial institution was tantamount to going to the
margin of the river, hence to the bank. Consequently, historical linguistics research revealed
an actual link between the two modern senses of bank. Yet most people are not aware of this
link and fail to see the connection, viewing the two as homonyms. Still, it seems that not all
the meanings of bank are related (the meanings in 1, 2, 3, 4, 5 and, respectively, 7, 8, 9 are
two somewhat distinct groups of examples linked to one another, whilst 6 could be seen as
remotely linked to either of the groups, but, as mentioned above, the connection between the
meanings of the words used in the two separate groups is, nowadays, rather blurry):
1. They demolished the bank yesterday. (noun – building)
2. δОt’săРoătoătСОăbank to get a loan. (noun – financial institution)
3. This morning the bank called about our mortgage. (noun – personnel)
4. Now you are the bank. (noun – when playing Monopoly)
5. δОt’săРoătoătСОăbank to donate some blood. (noun – storage facility)
6. The typical operator used to sit in front of a bank of switches. (noun – row of similar
things)
7. δОt’să РoătoătСОă bank to watch the fishermen. (noun – sloping land beside a body of
water)
8. I had just finished piling up the snow, when he ruined it all by reversing into the bank.
(noun – large pile of snow, sand, earth, etc.)
9. The divers hid behind a bank of about two feet high, hoping to get away from the
shark. (noun – a ridge, an undersea elevation) (web afv.gr homonymy and polysemy)
1052
SECTION: LANGUAGE AND DISCOURSE LDMD I
homonymy (the different usages of the word refer to distinct objects: some kind of storage
facility, row of similar items and pile of snow, sand or earth).
όrКtС’să КММouЧtă oПă polвsОЦвă КЧНă СoЦoЧвЦвă (ηζ-55) is based on two premises: that
“аorНsă СКvОă usă surЦТsОă tСКtă tСОă oЛУОМtsă tСОвă rОПОră toă КrОă soЦОСoаă rОlКtОН”ă КЧНă tСКtă “tСОă
subsuming entity is deduced from usage, it is not cause of usКРО”ă (ОЦpСКsТsă ЦТЧО)έă TСОă
former premise implies that if various objects are referred to by employing different words the
speakers do not surmise the existence of an obvious relation between those objects. The latter
implies that words are not receptacles in which conceptual sub-entities are lying in wait only
to later be actualised, rather words can be used to refer to one or more objects. If one word is
used to make reference to clearly related objects this is to be viewed as a case of referential
polysemy. When one word refers to objects that we perceive as related, though their senses
may not be directly linked, the polysemy is lexical. In those cases where no connection is
identified by the speakers between the two or more referents, yet they are conveyed by one
and the same word, the phenomenon is labelled homonymy, whether the objects happen to
share a signifier by mere accident, or the link existed once but is now lost due to the fact that
the object that had provided the link no longer exists, or owing to the fact that the connection
itself was forgotten and, hence, lost.
όКlФuЦ’săprКРЦКtТМăКpproКМСăpoТЧtsătoăКăМoЦЛТЧКtТoЧăЛОtаООЧătСОăКtoЦТstăКММouЧtăoПă
lexical semantics and a pragmatic-focused analysis as the framework most likely to solve the
paradox of polysemy. In order to bypass the fallacies characterising the existing
decompositional theories of word meaning, Falkum (3) shows how a conceptual atomist
approach combined with the treatment of polysemy as a basically communicative occurrence
yields an approach able to offer a descriptively valid account of this phenomenon. Working
аТtСТЧă tСОă МoЧtОбtă oПă SpОrЛОră КЧНă WТlsoЧ’să rОlОvКЧМОă tСОorв,ă όКlФuЦă (βθη)ă ОбplКТЧsă that,
sТЧМОă spОКФОrsă possОssă “КЧă ТЧНОpОЧНОЧtlвă ЦotТvКtОНă prКРЦКtТМă ТЧtОrprОtКtТoЧă sвstОЦ,ă
automatically activated by verbal utterances, which is capable of rapidly generating new
sОЧsОsăТЧăМoЧtОбts”,ăКăМoРЧТtТvОălТЧРuТstТМăКpproКМСăТsăЛОttОrăsuТtОНătСКЧăКăpurely semantic one
when it comes to the interpretation of polysemous words. Thus, when the senses of a word are
conventionalised, and there exist more semantic encodings, they should be rendered as
separate lexical entries, being homonyms, whilst in the other cases, where there exists a single
encoded meaning that, in distinct contexts, can be interpreted slightly differently, the
interpretations being derived or constructed through pragmatic adjustment during the on-line
processing of the encoded meaning, the various senses should feature under one and the same
lexical entry. The representation of the word bank’să sОЧsОsă аТll,ă ТПă όКlФuЦ’să tСОorвă Тsă
КpplТОН,ă looФă tСОă sКЦОă Кsă tСОă oЧОă НОvТsОНă КММorНТЧРă toă όrКtС’să КpproКМСμă ЧКЦОlвă tСrООă
separate entries for the homonyms meaning some kind of storage facility, row of similar
items and pile of snow, sand or earth, and for the first and last entry there will feature, as
polysemes, 1, 2, 3, 4, 5 and, respectively 7, 8 and 9.
The two theories outlined above yield similar typologies that, on the one hand, might
effectively be employed by lexicographers when drawing the distinction between polysemous
lexical items and homonyms, and, on the other hand, can offer a more accurate description of
how polysemous words and, respectively, homonyms may be represented in the mental
lexicon. Both approaches have thus succeeded in steering clear of the problematic issues
traditionally associated with decompositional theories of word meaning by treating polysemy
1053
SECTION: LANGUAGE AND DISCOURSE LDMD I
cases as aspects of language meaning in use. Falkum views word meanings as unstructured
atomic concepts and emphasises the essential contribution of pragmatics in the derivation of
word senses, whilst Frath pertinently differentiates between referential and lexical polysemy,
a МruМТКlă НТstТЧМtТoЧă tСКtă ПОlТМТtouslвă МoЦplОtОsă όКlФuЦ’să КММouЧtέă IЧă МoЧМlusТoЧ,ă sТЧМОă ЛotСă
όrКtС’sărОПОrОЧМО-МОЧtrОНăКpproКМСăКЧНăόКlФuЦ’săprКРЦКtТМ-focused theory of polysemy rest
on the mind-world relations that characterise the representation of polysemy, a combination
of these two accounts could well provide an apposite framework for the interpretation of this
controversial linguistic phenomenon.
Bibliography
1054