GRAMMAR AND LEXICON

Oleh Lartini (2040107114)
Grammar, however, is not restricted to the study of form of function words. It is concerned, more widely, with categories such a tense, gender, number and with syntactic functions such as subject and object. Some of them may be marked in a language by form words, but they may equally be marked by morphemes or even by form words. While there is a problem of establishing what are the relevant grammatical categories in any language, it is irrelevant for semantic whether a grammatical category is indicated by a form word, a morpheme or the order of the words.

For example, we find that English marks past tense with the past tense morpheme (usually indicated as –ed ). But there is no similar morpheme to indicated the future, this is marked by the verbs shall and will or by be going to (it may also be indicated by other verbal forms with the appropriate adverbs as in I’m flying to Cairo tomorrow and I fly to Cairo tomorrow). Other languages may use inflection where English and most familiar language use form words. Thus the English conjunctions after, when, while, if are translated Bihn (a Cushitic language of Ethiophia) by endings of verb. Nearer home, Finnish has many complex ‘case’ systems, containing not only ‘nominative’, ‘accusative’, ‘ablative’ etc, all of Which are familiar from Latin, but also ‘elative’, ‘illative’,’ adhesive’ and others. These last ones would translate characterize it English out of, into, on, as.

In modern linguistics the problem of the distinction between the grammar and the lexicon is often posed in terms of the distinction between sentences that are unacceptable or deviant for grammatical reasons, and those that are excluded on lexical grounds. There is no apparent difficulty about recognizing grammatically deviant sentences. An example would be *the boy is in the garden. This breaks only one grammatical rules, but we can easily invent sentences that seem to conform to no rules at all as *been a when I tomato. In contrast we shall rule out on different grounds *the water is fragile, *the flower walked away. With these the issue is on be of collocation. Which determines the possible co occurrence of water with fragile and flower with walk.

There have, however, been opposing views on the question whether these two kinds of restriction, one grammatical, the other lexical, are, in principle, different. One argument to sustain the difference is that a sentence can be grammatically correct, yet at the same time totally defiant in lexical terms. If a sentence can thus conform to grammar, but be completely deviant lexically, it would seem that grammar and lexicon are distinct. Earlier, incidentally, Carnap had made the same point by inventing a sentence that does not contain any English word at all yet seems to be quite grammatical in terms of English – Pirots karulize clatically(1937: 2).

Some linguists believe that just as a grammar could be wholly formal, and that we need not concern ourselves with the meaning of any of our grammatical categories, so, too, a total statement of all collocation possibilities of a word would be sufficient to characterize it linguistically. Indeed, some went so far as to believe that the set of collocation possibilities of a word was essentially the meaning of that word for the linguist.

What is relevant to semantics is that he was concerned with restrictions on the co occurrence of items within a sentence, so that we shall no permit *the idea cut the tree,* I drank the bread,*he frightened that he was coming,* he elapsed the man. in all these examples it is clear that we have chosen item that, in some way, do not fit the verbs. The last examples are clearly a matter of grammar in that frighten does no take a that-clause, while elapse is an intransitive verb that does not take any object at all. With the other two examples it is a matter, however, of the incompatibility of lexical items, of certain nouns (as subjects or objects) with certain verbs. While nothing the difference between these two types, Chomsky proposed to deal with them in similar ways. In both cases he stated, as part of the specification of the verb, the environment in which it may occur. Thus elapse was shown as not occurring with an object noun phrase, and frighten not occurring with the following that-clause ( or rather it was not shown that they can so occur , since the specification would state what is possible, not what is possible). Similarly cut would be shown to need a concrete subject, and drink a liquid object. This was achieved in terms of components (concrete) and (liquid). These are selection restriction. Any sentence which did not comply to them was ruled out and the grammar would not generate it.

The lexical restrictions, it has been suggested (Haas 1973:147-8), are not a matter of rules but of tendencies, not of Yes/No, but More/Less, when judge in terms of deviance, unfortunately this leads us to the problem ‘When is a rule a rule?’, for there is no clear line between grammatical or lexical deviance. Some sentences are clearly ungrammatical and are simply to be ruled out or corrected, while others are odd only in a lexical way and can, with some imagination, be contextualized. But there are others that are half-way, and we are not really sure whether their deviance is lexical or grammatical.

Consider, for example, *the dog scattered. This is not simply a matter of the collocation of dog with scatter, for the verb scatter is normally used only with plural nouns (the dogs scattered), or with collective nouns (the herd scattered). It would seem, therefore, that a grammatical rule is being broken and that we should amend to the dogs scattered (or the dog was scattered). But cannot we imagine a dog with magical powers whose way of avoiding its enemies was to break into many peaces and ‘scatter’ over a wide area? Indeed we can, and so we have found a possible, if far fetched, contextualization for the dog scattered. The deviance would seem, there for to be lexical rather than grammatical. But I am not really sure. Can we say, the dog scattered even ………….. Context? Or would the dog scattered itself be more appropriate. My indication here shows that we are on the borderline of grammar and lexicon.