51562-Article Text-77341-1-10-20100222
51562-Article Text-77341-1-10-20100222
51562-Article Text-77341-1-10-20100222
Abstract: This article outlines the challenges relating to the lemmatization of the lexical items
which are either polysemous or homonymous, as experienced during the compilation of the Sesotho
sa Leboa–English Bilingual Dictionary. These problems can be ascribed to a lack of objectivity result-
ing from an inadequate knowledge regarding the etymological relationships of the meanings of
some lexical items which are subsequently misclassified as either homonyms or polysemous words.
This often causes improper lexical entries in dictionaries, i.e. polysemous words may be lemma-
tized as homonyms and vice versa.
To unambiguously distinguish between lexical items which are either homonyms or polyse-
mous words, lexicographers should in this regard consider the criteria suggested by scholars:
— In accordance with the relatedness/unrelatedness criterion, the lexicographer will need to
determine the extent to which the lexical items are related before entering them in the dic-
tionary.
— The etymological criterion will help the lexicographer to determine the relatedness of the
lexical items for inclusion in the dictionary according to their historical connection.
Using these criteria to lemmatize a lexical item properly in the dictionary, the lexicographer will be
required to determine whether a lexical item can be regarded as one word with more than one
meaning (a polysemous word), or as two different lexical items which have the same spelling
(homonyms). To make this distinction, the lexicographer will need to research the historical back-
ground of the lexical item. Even though these criteria are important aids for distinguishing
polysemy and homonymy, the problem of subjectivity is not altogether removed since the criteria
do not specify the level or degree of relatedness/unrelatedness at which the lexical item(s) can be
said to be polysemous or homonymous.
* This article was presented as a paper at the Eleventh International Conference of the African
Association for Lexicography, organized by the Tshivenda National Lexicography Unit,
University of Venda for Science and Technology, Thohoyandou, Republic of South Africa,
5–7 July 2006.
Deur hierdie kriteria te gebruik om 'n leksikale item korrek in die woordeboek te lemmatiseer, sal
van die leksikograaf vereis word om te bepaal of 'n leksikale item beskou kan word as een woord
met meer as een betekenis ('n poliseme woord), of as twee leksikale items wat dieselfde spelling het
(homonieme). Om hierdie onderskeid te tref, sal die leksikograaf die historiese agtergrond van die
leksikale item moet navors. Selfs al is hierdie kriteria belangrike hulpmiddels vir die onderskeiding
van polisemie en homonimie word die probleem van subjektiwiteit nie heeltemal oorkom nie,
omdat die kriteria nie die vlak of graad van verwantskap/nieverwantskap aandui waarby die lek-
sikale item(s) as poliseem of homoniem beskou kan word nie.
1. Introduction
ymy accurately in academic and scientific researches. Stern (1931), Ullman (1962),
Lyons (1977), Leech (1981), Hurford and Heasley (1983), Taylor (1989), and
Louwrens (1994) are among those scholars who researched these sense rela-
tions. According to Leech (1981: 228), homonyms are 'roughly two or more
words having the same pronunciation and/or spelling, but different in mean-
ing'. This definition of homonymy is in line with that of Macdonald (1977: 625)
who defines a homonym as 'a word having the same sound and perhaps the
same spelling as another, but a different meaning and origin'.
From these definitions it can be deduced that homonymy covers both
written and spoken forms, but it is possible to have partial homonymy (or het-
eronymy), where the identity is within a single medium as in homography and
homophony (Crystal 1991: 167). The Encarta Concise English Dictionary (2001)
defines a homograph as 'a word that is spelt in the same way as one or more
other words but is different in meaning' and a homophone as 'a word that is
pronounced in the same way as one or more other words but is different in
meaning and sometimes spelling'.
In Sesotho sa Leboa there are, however, also tonal changes which need to
be considered when describing homonyms. If two or more words are spelt the
same, the tonal patterns must also be taken into account: heterotonal hom-
onyms have different tonal patterns (Louwrens 1994: 75), while homotonal
homonyms have the same tonal pattern (Louwrens 1994: 76).
The following are examples of two groups of words spelt the same in
Sesotho sa Leboa:
fola cool down
fola line up (to queue)
fola smoke (a pipe, cigarette, etc.)
fola samp, crushed maize
mogolo salary
mogolo throat
The words in the first group of four, however, do not have the same pronun-
ciation: the first two are pronounced with the mid-low vowel /O/, therefore
/fOla/, while the last two are pronounced with the mid-high vowel /o/, there-
fore /fola/. But when the tonal patterns of these two pairs of words are consid-
ered, the following variations emerge:
[fOla]: fólà 'cool down' (high-low tone) and fòlà 'line up' (low-low tone),
and
[fola]: fòlà 'smoke' (low-low tone) and fólà 'samp, crushed maize' (high-low
tone).
The words in this group are therefore heterotonal homonyms.
When the pronunciation and tonal patterns in the second group of two are
investigated, it appears that they are the same in each case. Both words have
the pronunciation /moVOlO/ and the tonal pattern mògòlò (low-low-low). The
436 V.M. Mojela
Hurford and Heasley (1983: 123) say that a case of polysemy is one where 'a
word has several very closely related senses'. And Mojela (1991: 31) also
explains it as a case where 'one word may have a set of more than one different
but related meanings'. According to Taylor (1989: 99), polysemy is the associa-
tion of two or more related senses with a single linguistic form. Illustrating it
with an example, he says: 'The word "bird" can refer to many different kinds of
creature — robins, penguins, ostriches, etc. These different kinds of creature are
members of the category in virtue of similarity to a single prototype represen-
tation.'
3. Lexicographic challenges
The major challenge facing lexicographers is the making of a clear and objective
distinction between polysemous and homonymous lexical items which are
entered as lemmata in their dictionaries. Such a distinction is not always possi-
ble with all lexical items in a language. Lexicographers with little knowledge of
the etymology of the lexical items in their languages will have problems in
entering polysemous and homonymous lexical items in their dictionaries, since
they will not know the extent to which the lexical items are related. In this
regard, Lyons (1977: 550) says the following: 'The difference between homon-
ymy and polysemy is easier to explain in general terms than it is in terms of
objective and operationally satisfactory criteria.'
Lyons' statement shows that the distinction between polysemy and hom-
onymy is more complex than is generally perceived. He gives two criteria to
simplify the complexities of identifying polysemy from homonymy, i.e.
— the etymological criterion, and
— the relatedness/unrelatedness criterion.
The etymological analysis of the lexical item tshela 'six' will show that the
word originated from tshela 'jump'. Counting in the Sotho tradition is done
with the help of the fingers starting with the fingers of the one hand, and when
the fifth finger is reached, 'jumping' to the other hand to find the sixth finger, in
this way giving the number tshela 'six' its name. Obviously, the lexicographer
with little (or no) knowledge of the etymology of this lexical item will enter
tshela twice as two unrelated homonymous lexical items, i.e.
tshela jump
tshela six
The etymological investigation of the lexical items tšhela 'pour (liquid, sand,
etc.)' and tšhela 'pay tax' reveals that the second lexical item originated from
the first one. The pouring of money into the tax-collector's coffer is metaphori-
cally compared to the pouring of liquid, sand, etc. into a container. In this way,
the two meanings become related, so that tšhela should be regarded as a
polysemous lexical item.
The etymological relationship between the meanings of šupa 'seven' and
šupa 'to point' can be explained by the fact that the number seven got its name
from the seventh finger, which is used for pointing. The seventh finger or the
'pointing finger' is also referred to as tšhupa baloyi, which means the finger
used to point out witches/wizards, i.e. to indicate who performed witch-
craft/wizardry. Therefore, this relationship between šupa 'seven' and šupa 'to
point' gives the lexicographer a reason to enter šupa as a polysemous word
instead of two homonymous words.
438 V.M. Mojela
to refer also to 'graveyards' and 'game reserves'. The meaning shift is so drastic
that only the researcher with a knowledge of the etymology of this lexical item
would be able to trace the relationship between the three meanings of dirapa.
6. Conclusion
The main objectives of this article can be summarized by the answers to the
following questions:
— Are lexicographers aware of the complications pertaining to the identi-
fication and the ultimate distinction between polysemy and homonymy?
— Do lexicographers realize that it is very difficult, if not impossible, to be
objective in distinguishing between homonymy and polysemy?
— Are lexicographers aware of the importance of having sufficient knowl-
edge of the etymological history of lexical items before distinguishing
between polysemous and homonymous lexical items?
— Do lexicographers know that relatedness of meaning is only measured in
degrees, and that the degree to which lexical items are related depends
on the individual researcher's subjective conclusion?
A knowledge of the suggested criteria for identifying polysemy and homon-
ymy is important in making a distinction between these two sense relations,
but this does not always solve the question of subjectivity in distinguishing
between polysemy and homonymy, since the criteria are always associated
with the semantic relationship of the lexical items. Furthermore, it is obvious
that the lexicographer will be faced with numerous lexical items for which it
will not always be possible to trace the etymological history.
References
Crystal, D. 1991. A Dictionary of Linguistics and Phonetics. Third Edition. Oxford/Cambridge, Mass.:
Blackwell Publishers.
Hurford, J.R. and B. Heasley. 1983. Semantics: A Coursebook. Cambridge: Cambridge University
Press.
Leech, G. 1981. Semantics: The Study of Meaning. Harmondsworth: Penguin Books.
Louwrens, L.J. 1994. Dictionary of Northern Sotho Grammatical Terms. Pretoria: Via Afrika.
Lyons, J. 1977. Semantics, Volumes 1 and 2. Cambridge: Cambridge University Press.
Macdonald, A.M. 1977. Chambers Twentieth Century Dictionary. Bath: Pitman Press.
Mojela, V.M. 1991. Semantic Changes Accompanying Loan-words in the Northern Sotho Lexicon. Un-
published M.A. Dissertation. Pretoria: Vista University.
Rooney, K. (Ed.-in-chief). 2001. Encarta Concise English Dictionary. London: Bloomsbury.
Stern, G. 1931. Meaning and Change of Meaning. Gothenburg: Goteborgs Hogskolas Arrskrift.
Taylor, J.R. 1989. Linguistic Categorization: Prototypes in Linguistic Theory. Oxford: Clarendon Press.
Ullman, S. 1962. Semantics: An Introduction to the Science of Meaning. Oxford: Basil Blackwell.