Reply to van der Hoven

At end of his review of Ferdinand de Saussure Mr. van der Hoven claims that theorists of language should address themselves to substantive questions of such as what processes do we choose some formulations and reject others as inadequate or false. Since linguists have never defined their task as that of describing how we choose what to say about a particular situation, Mr. van der Hoven seems to be proposing a grand new vision of linguistics, based on dubious assumption that a grammar of a language should distinguish between true and false statements. This confusion emerges at very beginning of his critique. He quotes claim that a semiological analysis of conventions of astrological discourse not be fundamentally affected if future discoveries were to prove that everything astrologers said were true. The same rules would underlie astrological discourse whether predictions they yield are true or false. To this he objects that serious users of this largely obsolete language, truth (or at least supposed truth) of astrological statements is perhaps single most important factor in discourse. Doubtless it is, and thus concerns of a consumer of astrology are very different from concerns of semiologist who is analyzing astrology as a signifying system. Mr. van der Hoven implies that semiologist ought to change his purpose and investigate whether astrological predictions are true or false, since this is what worries serious users, but that would not be semiological investigation at all; it would be astronomy statistics, and perhaps sociology. The distinction between study of rules of a system and study of truth or falsity of particular statements or discourses produced by system is absolutely basic, but since Mr. van der Hoven has apparently failed to grasp it, let us take another example. If someone were to utter statement John is easy to please, his interlocutor might be particularly anxious to know whether this were really true or whether it were false, a lie designed to reassure. Truth or falsity of statement might be the single most important factor in discourse, but this does not mean that linguist should abandon his attempt to describe grammar of English and concentrate instead on truth or falsity of this sentence. The question of how speaker decided what to say about John and what led him to pick this information rather than another might be a fruitful topic of research, but it is fundamentally different from linguistic project of working out underlying system of rules of a language. Failure to distinguish two bespeaks confusion and can only result in confusion. Next is problem of arbitrary nature of sign, which is somewhat more complicated. Mr. van der Hoven cites as a critique of this principle fact that may be historical reasons for emergence or evolution of a form: there may still be appropriate reasons why meaning of 'cattle' evolved as Culler suggests it did. In fact, his formulation here is too weak: in principle every change has a cause. As it happens, attempts to explain emergence of a particular form, adoption of one

Lexicography and Language Studies · Linguistics and language evolution · Natural Language Processing Techniques · Philosophy