On Discourse

Abstract

Etymologically discourse finds its origin in the Latin verb discurrere, to run about, probably by way of the French form discourir. This particular genealogy is more indicative than it might first appear because discourse, as the term is used in classical rhetoric, emphasises language as motion, as action. If the rhetorician was concerned with arresting language so that he could specify the various relations into which words could enter, to classify the figures and the topics, discourse constituted both the object and the aim of the study. Rhetoric started from and ended with the running together of the forms and the subjects in a continuous utterance—in, exactly, a discourse. It is within this perspective that we can consider discourse as indicating the articulation of language over units greater than the sentence. The major divisions of rhetoric accomplish just such supra-sentential divisions: exordium, narratio, argumentatio, refutatio, peroratio (Curtius 1953: 70). A particular set of articulations will produce a field of discursivity—the site of the possibility of proof and disproof (it can be recalled that the study of rhetoric found its early rationale in relation to forms of popular law (cf. Barthes 1970: 175)). It is to such a notion of discourse that Cahiers pour l’Analyse made reference when, in the avertissement to the first number, Jacques-Alain Miller defined the magazine’s task as the constitution of a theory of discourse, specifying that by discourse ‘we understand a process of language that truth constrains’ (1966: 5). And, in a reference to the content of that first number, Miller made clear that the constraint of truth has as its inevitable corollary the production of a subject, a subject divided by the very process of language that calls it into being.