| |
 |
 |
 |
 |
 |
| 413.4.3.3: THE PROBLEMS |
| |
| Though
the collection of data from an unwritten language appears
easy and interesting, there are many problems in doing
so, which R.A. Singh (1982) has listed out as follows: |
| |
| (1) |
Problem of segmentation
and identification of words: In written languages,
the segmentation of morphemes and identification of words
used in a sentence is easy because of the presence of
traditional grammar, well-tested and accepted devices
and fixed criteria. There would be space between the words
for potential pause when we try to segment the oral speech.
Thus in unwritten languages there is no scope for such
help as they have no script or written form. The lexicographer
has to record the speech of the people, analyze them,
segment and identify the words. |
| (2) |
If the unwritten language belongs to
an isolating-agglutinating type of language, then the
determination of the lexicographic word is
even more of a problem. Let us see what we mean by these
terms. In such a language the whole grammatical process
involves prefixation and suffixation of morpheme. The
roots in such a language readily accept any number of
prefixes and suffixes. Any number of words can be derived
by a mere juxtaposition of various morphemes to the root
or the stem. For a native speaker it may not be a problem
because he can easily coin the words, as he likes. But
the problem is that he is unable to express himself systematically.
The lexicographer has to work hard to understand and analyse
the components. Otherwise he may consider a phrase as
a word. He has to identify them correctly in terms of
the main root, prefixes and suffixes and arrange them
alphabetically. |
| (3) |
The other problem is the collection
and selection of set combinations and compounds from
such languages. In written languages, the identification
of compounds may not be a problem, because there would
be some clues such as lack of word boundaries and/or hyphenated
spelling of compounds. The lexicographer gets chance to
collect data wherein he may come across many occurrences
of such units. That would be useful to him in the determination
of set expressions. For unwritten languages, the lexicographer
has to collect different contexts with varied collection
of data and then find out compounds and set combinations
in recorded stories, namations and conversations. |
| (4) |
In Jaintia language, the following words are used
with different shades of meaning. |
| |
a. P khan 'break(stick)'
b. Pya? 'break(bottle)'
c. tKue 'break(rope)' |
|
| |
The difference cannot be elicited by the
simple gloss 'break'. Hence the identification of words
from glosses there is always a possibility of not getting
the appropriate word. Extensive collection of materials
alone can solve this problem. |
| (5) |
A great difficulty in elicitation of
data for unwritten languages is presented by an isomorphism.
For a single object there may be two or more words in
the unwritten language whereas there is only one gloss.
For instance, there are two words in Jaintia, yuukhi
and rati? to denote razor in English. Suppose
the lexicographer asks for the equivalent word for a razor,
the informants will give him two terms. The lexicographer
must then be careful in such cases to collect all such
options. Similarly the words, baskets, spheres, knife
may have several equivalents in most of the unwritten
languages. Even if the lexicographer shows the object
to the informants, they may give related words and not
the exact word. |
| (6) |
Identification and determination
of meaning for the culture-bound words and flora
and fauna etc., are another problematic area. For cultural
words, one-word gloss may not be adequate. The lexicographer
must provide the whole cultural information related to
such words. That is why many of the dictionaries compiled
for the unwritten languages are more of an ethnographic
account with a considerable amount of encyclopaedic information
than being a conventional dictionary. So for such items
with cultural significance, detailed data should be collected
about the given culture. Similarly the words related to
fauna and flora too need some detailed information. |
|
| |
| If
the lexicographer wants to compile a good dictionary he must
be very careful about the above-mentioned problems. The lexicographer
has to collect not only a number of words, but also different
meanings in different contexts. For this purpose the compiler
should record all the linguistic and physical contexts. Such
data should be collected thoroughly and systematically. The
excerpts should be compared and similarities and dissimilarities
in usage of the words noted to mark the different meanings of
the word. The lexicographer must also ask the informants to
produce examples of the collocational possibilities of a particular
word. The informants may be asked to give synonyms and antonyms
of each word. There will also be useful in the determination
of meanings of the lexical units. |
| |
| The
lexicographer must give special treatment for the functional
words. He must collect extensive data, because such functional
words are so frequently used that there is more scope of having
varied and diversive meanings. To collect all the meanings and
usages in vogue the lexicographer should extract and scrutinize
a large number of contexts. |
| |
 Points
to Remember  |
| |
| QUESTIONS |
| |
| A. Write what you know about the
tasks: |
| |
(i) Identification of data
(ii) Determination of meaning
(iii) Encyclopaedia
(iv) Collocation
(v) Semantic categorization |
|
| |
| B. Write short notes on: |
| |
1. If you have to classify verbs
of your own language in writing dictionary, which
category labels would you like to use?
2. What is meant by 'Semantic domains"?
3.What are the special problems of material collection
from unwritten languages?
4. What are the defects of the word-list method
of data collection? |
|
|
| |
| |
|
|
|
 |
 |
 |
 |
| |
|
|
|
