| |
 |
 |
 |
 |
 |
| 421.1.3.1.2.1: SUBSTANTIVE UNIVERSALS |
|
| By substantive
universals we mean those fixed set of primitive elements
from which all natural languages must draw. For example,
at the level of phonology we may consider the sound segments
as analyzable into a fixed set of phonetic features such
as voicing, nasality, aspiration etc. these are called
distinctive features. Similarly sentences of any language
can be represented in terms of a fixed set of syntactic
categories like noun, verb, adjective etc. in the area
of semantics also we may want to say that lexical items
of any language will refer to specific types of objects,
states, actions etc. In short, substantive universals
define the total range of elements in a given area a subset
of which alone will be selected by any given language. |
|
| 421.1.3.1.2.2: FORMAL UNIVERSALS |
|
| As the name
indicates a formal universal has to do with the form of
rules of a grammar. Formal universals for instance, can
specify the type of rules that we may require to form
sentences in any language. Similarly formal universals
can be taken as the conditions or principles which may
govern the operation of phonological, morphological or
syntactic rules in a language. In generative grammars,
for instance, we recognize a type of rules called transformational
rules whose basic function is to move elements from one
position to other or delete elements in specific syntactic
structures. For example, in the English interrogative
sentence whom did you see in the library?, the presence
of the direct object 'whom' in the beginning of the sentence
as against its expected position, that is , after the
verb, may be formalized in terms of a movement transformation.
This transformation movers a Wh-question word to the clause
initial position in English. |
|
| 421.1.3.1.2.3: LANGUAGE VARIATION |
|
| Having considered
how we can approach some of the details of what UG may
consist of, we can now look into how the question of language
variation is answered in the generative approach, more
approrpriately, in the theory which is now called Government
and Binding theory or GB, for shot. From the discussion
so far it should be clear that by UG we mean those substantive
and formal properties which underline human language in
gerneral. But this does not mean that each and every one
of these features or principles will apply to each language.
To illustrate this let us take an example from phonetic
universals. You have already noted that sound units of
all natural languages can be analysed into a limited set
of phonetic features called distinctive features. Now
it is not necessary that all languages must select all
the features from this universal inventory. You may notice
that Hindi and Bengali utilize the feature of UG, does
not mean that the child's learning of a language is automatic.
First and foremost the child's innate linguistic faculty
has to be triggered by proper linguistic input. A child
who does not hear any language spoken around him or her
will never learn a language. Secondly, the child will
have to find out which particular subsets of the principles
and categories are applicable in the case of his language.
What is the value of a given parameter in a language to
which she is exposed? For instance whether or not the
language in question is a pro-drop language. Besides these
the child will have to sort out the details of the lexicon,
that is, the words, their phonetic forms and meanings.
A child will also have to learn the morphological details
of his language. |
|
| 421.1.3.2: DIFFERENCES BETWEEN THE TWO
APPROACHES |
|
| In the foregoing
sections we have considered the two major approaches to
linguistic universals. Now we can note down some of the
differences involved in the two approaches. |
|
| (1) |
In Greenberg's Approach
it is necessary that we draw our data from a wide
range of languages. In Chomsky's approach, on the
other hand, a detailed abstract study of a given
language can as well lead us to the postulation
of significant universals. Since, according to Chomsky,
language universals are innate properties of human
mind they can be as well available in one language
as in another. |
| (2) |
A second point of difference is
that in Greenberg's framework linguistic universals
are statable in terms of data available at a more
overt or concrete level of linguistic analysis.
As against this the principle and conditions that
have been suggested as universals in the generative
approach have been arrived at by assuming a high
degree of abstractness in linguistic analysis. It
is not that Greenberg's universals do not involve
any abstractness, but the degree of abstractness
required in formulating the universals varies in
the two methods. |
| (3) |
The two approaches also differ with
regard to the possible explanations from language
universals. We have already noticed that in the
generative approach language universals are innate,
that is we are born with them. Greenberg and his
followers do not make any claim to innateness. |
| In the foregoing sections
we have concentrated mainly on language universals
pertaining to syntax. We will now discuss briefly
a few examples of language universals from other
levels of language structure. |
|
|
|
|
|
|
 |
 |
 |
 |
| |
|
|
|
