The aim of this paper is to reflect upon Michel Foucault’s (1926–1984) contributions to discourse studies in general, and to pragmatics in particular. In order to accomplish this aim, this study examines his central tenets in the understanding of discourse, and how his work has been received and incorporated in those fields, both in the development of current concepts and theoretical proposals, and in the kinds of analyses which have been suggested. Given that, to a certain extent, the employed approach is also Foucaultian, rather than presenting a mere inventory, this paper focuses on those aspects which enabled, at a given time and place, the development of specific objects of study, theoretical models, and techniques of analysis, and it also raises several theoretical problems and questions in this field. In the first place, it deals with those Foucaultian concepts that are considered to have shaped a new understanding of discourse, in a process through which discourse has emerged as the object of a field of knowledge (Sections 2 & 3). In addition to this, the paper discusses how this new understanding has introduced changes in the analytical practice, allowing the adoption of particular perspectives, such as a ‘critical’ perspective (understood in the widest sense) and the emergence of particular aims, such as the ‘interventional aim’ (taking part in the present discursive and social order). Both can be seen as consequences of analysts’ increasing awareness of and concern about the social and political consequences of discourse, and of discourse analysis itself (Section 4). (See Mchoul & Grace 1993, for a similar attempt; for a reviews of Foucault’s work as a whole see, among others, Dreyfus & Rabinow (1982). One of the difficulties encountered is the plurality of readings of Foucault’s work. This generally acknowledged multiplicity has been associated with different traditions, and cultural and geographical areas. Rorty, for example, assigns to the European continent what he considers a Nietzschean reading, while he attributes a liberal (reformist) one to the Americas (in fact North America). But these multiple readings can also be derived from Foucault’s texts themselves, the choice depending on which topics attract readers, in a process of reading which constitutes a philosophical journey (Enrici 1999). As will become clear in the paper, these multiple options seem to have their roots in the Protean nature of some of the main Foucaultian concepts (i.e. discourse itself; see, 2.1), which is also linked to Foucault’s procedure of ‘problematization’ (that is, questioning what is generally assumed). And, finally, this plurality also explains how the assessment of his legacy causes frequent controversies, and why different authors, from different and even opposed approaches, refer to his work, and claim to have been influenced by him. This is attested not only in the more radical discursive approaches, in which the relativist-constructionist perspective is stronger, but also in ‘constructivist structuralist’ approaches, concerned with the constraining role of social structures, as well as with the active process of the production of social practices which can transform social structures.