Linguistic Competence: What Is It Really?

'Conoscere e distinguere' (B. Croce): To attain scientific knowledge is, first of all, to make distinctions. Hence, if I were asked what I consider to be my main contribution to the understanding of language and consequently to the foundation of linguistics or, to put it in other words, what constitutes my permanent frame of reference, the very often implicit fundamental principle underlying my treatment of the different, general, or particular linguistic problems, I would answer that it is a relatively simple distinction, one also made intuitively by the speakers of any language, which became entirely clear to me only around I955; in any case, it was only in 1955 that I formulated it explicitly and undertook to justify it.1 This distinction, which originated as a result of the discomfort with the distinction langue/parole, language/speech in the context of the post-Saussurean discussion of these notions, concerns in reality the levels of language, but applies first of all to what in the last decades has been called 'linguistic competence' and what I called at that time and still continue to call 'linguistic knowledge' (saber lingiiistico). And I consider this distinction to be important, as it enables us to assign a precise position to the different problem areas of linguistics and to its various questions with respect to the complex object language. It has been, for me at least, a helpful epistemological frame of reference for the interpretation not only of the various linguistic problems ranging from that of linguistic change to that of translation and of linguistic correctness, but also of the structure of the linguistic disciplines themselves and of recent developments in linguistics.2 I shall say more about this in the second half of this lecture. Let us first consider the distinction as such and its justification. Alan Gardiner, one of the most penetrating commentators of the distinction made by Saussure between langue and parole, interprets this distinction as a distinction between language as the activity of speaking and the underlying knowledge or technique of this activity (science, knowledge): 'Speech is ... a universally exerted activity.... In describing this activity, we shall discover that it consists in the application of a universally possessed science, namely the science which we call language. ... Language is a collective term, and embraces in its compass all those items of knowledge which enable a speaker to make effective use of word-signs.'3 For Gardiner, speech and language in this sense are not coextensive: inasmuch as it is realized, language is entirely contained in speech; speech on the other hand is not only the realization of language, since it exhibits, along with 'the facts of language', features by which it differs from language or rather by which it goes beyond language ('facts of speech'). Gardiner thinks that the distinction thus formulated