Page 354 - DENG504_LINGUISTICS
P. 354

Linguistics



                  Notes          The explicitness and the predictive nature of such a grammar can be summed up in one word, the
                                 word generative. But the term generative should not automatically lead to conclude that all generative
                                 grammars are transformational. Any type of grammar which is explicit and predictive is generative.
                                 But not all generative grammars are transformational.
                                 The Goals of Linguistic Theory and the Concept of Grammar
                                 Between 1933 and 1957 the structuralists were largely concerned with the problem of how to
                                 DISCOVER the phonemes, morphemes, etc. of the language. They also believed that in the interest of
                                 ‘empirical’, ‘scientific’ linguistics we must begin with the observed data and work ‘upwards’ from
                                 the sound system to the grammatical system, keeping the two in their right place in the sequence and
                                 completely apart. The sound system had to come first because it was felt that only the phonetic/
                                 phonemic aspects of language provided a basis for scientific statement, since meaning was outside
                                 the possibility of serious investigation. So what linguistic theory had to provide, was only a set of
                                 ‘discovery procedures’—procedures for discovering the structure of the language.
                                 So at that time, a grammar could be defined as a perfect, objective description of a language. And the
                                 ultimate goal of linguistics was to find rules which led to such grammars. Chomsky suggested, on
                                 the other hand, that this aim was both far too ambitious and far too limited in scope. It was too
                                 ambitious in that it was unrealistic to expect a perfect grammar from a mass of data. It was too
                                 limited because such grammars had no predictive power. They catalogued what had happened, but
                                 did not predict what would happen.
                                 Chomsky suggested that a grammar should be regarded, instead, as a theory or hypothesis about
                                 how a language worked. In the same way as a biochemist might formulate a hypothesis against
                                 actual living cells, so a grammar should be a hypothesis about language formulated and tested in the
                                 same way. If correctly formulated, such a grammar will be ‘a device that generates all of the
                                 grammatical sequences (of a language) and none of the ungrammatical ones.’
                                 The task of linguistics was to formulate such grammars and to work out principles and procedures by
                                 which they could be evaluated so that if a linguist was presented with two rival grammars, he could
                                 automatically identify the better of the two and reject me other. (So by ‘evaluation procedures’ is meant
                                 procedures for evaluating all the possible descriptions saying why one is better than the others).
                                 The tasks before the Transformational, Generative grammarian, therefore, were to formulate a linguistic
                                 theory which will account for all the linguistic levels of the language, inter-relate sound and meaning,
                                 to provide a linguistic methodology for structural description and evaluation procedures for evaluating
                                 the best of all the available models of grammar and rejecting the other ones, and to have explanatory
                                 and descriptive adequacy.
                                 By explanatory adequacy is meant the grammar’s capacity to offer an A.D., that is, a language
                                 acquisitions device
                                                Primary data — A.D. —G
                                 In other words, it should be adequate to establish linguistic universals, to develop a general evaluation
                                 procedure, to distinguish between the grammatical and the ungrammatical, between the more
                                 grammatical and less grammatical. That is, the theory must provide a practical and mechanical method
                                 for actually constructing the grammar, given a corpus of utterances. One way of testing the adequacy
                                 would be to determine whether or not the sequences it generates, are actually grammatical, acceptable
                                 to a native speaker, whether or not it mirrors the behaviour of the speaker.
                                 Descriptive adequacy is an external condition. That is, a grammar must produce all and only
                                 grammatical sentences of that language, should offer a description of all the sentences, must indicate
                                 all the grammatical functions such as Subject, Object, Adverb, Adjective, etc., must indicate and specify
                                 ambiguity, if any, and must indicate grammatical relations, and processes such as paraphrasing,
                                 embedding, conjoining, deletion, permutation, etc. Furthermore, it should determine selectional
                                 restrictions or constraints too.



        348                              LOVELY PROFESSIONAL UNIVERSITY
   349   350   351   352   353   354   355   356   357   358   359