This paper has tried to outline the nature of the phenomena to be incorporated in a theory of communicative competence. It began with a re-assessment of what linguists consider to be the "intrinsic competence of the ideal hearer-speaker" and tried to show that such competence is much broader than the restricted aspects of it that linguistic theory, as currently understood is able to handle. There is merit in the argument that linguistic theory ought to restrict itself to phenomena of language competence that are capable of being incorporated in the strict formalization of transformational generative grammars since such a restriction enhances the likelihood of cumulative additions to our knowledge about the linguistic phenomena. In pursuing such a strategy, one should be very careful not to claim for linguistic theory a power it does not have. The definition of linguistics as the account of "an ideal speaker-hearer's intrinsic competence" is infelicitous in this respect, as are the various claims that linguists have made al)out "semantic interpretation and meaning.

The proposal for a theory of communicative competence is motivated by the recognition of how far short of the mark linguistics has come to account for speakers' ability to understand utterances in a communicative context. It is by no means certain that an account of communicative competence will ever be capable of achieving the kind of strict formalization that now exists for the description of syntactic phenomena. On the other hand, it seems quite plausible that we can go far beyond the formalization that has been achieved so far in the fields of literary criticism and descriptive ethnographies of communication. This remains a hope for the future. At the present time, much more descriptive work is needed, and is going on, in psychology sociology, and anthropology. Ultimately, the theory of communicative competence will be at least as general as any psychological theory of the individual or any sociological theory of interacting men.