Linguistic Structure and Sentence Production

Abstract

It is generally accepted that a careful distinction must be drawn between theories of a language user’s competence and theories of a language user’s performance. The former theories describe what the user knows about the sentence of his language, whereas the latter theories describe how this knowledge is put to use in the task of producing or understanding particular sentences. This paper is concerned with problems involved in the second type of theory, and in particular, with the task of providing a model of the control system involved in sentence production. The total sequence of operations involved in the production of a sentence is assumed to be analyzable into at least three components: (a) the class of operations involved in determining the content of the message to be encoded, (b) the class of operations involved in converting the message to a sequence of morphemes, and (c) the class of operations involved in converting this morpheme sequence into the appropriate commands to the articulatory system. Only the second of these three components will be considered in this paper.

This paper was prepared while the author was a Research Fellow at the University of Melbourne, Victoria, Australia. The research reported was made possible by the generous technical assistance of the following people, to whom the author wishes to express his gratitude: R. B. Lees, Belan Togrol, D. Cuceloglu, Toshio Iritani. Miho Steinberg, P. Tanaka, J. Oki, E. E. Davis, N. Viernstein, M. Clyne and M. Garrett.