The discourse of Michel Foucault: A sociological encounter

Michel Foucault is a major source for the idea in critical accounting and organizational studies that identities (selves, subjectivities) are discursively constituted. This return to the text is intended as a clarification of what Foucault actually says on this matter and an assessment of how far it can be regarded as authoritative. The major conclusions are as follows.The subject matter of Foucault's ‘discursive’ phase is not discourse in its generality but islands of organization (‘discursive formations’) within it. To all intents and purposes these are bodies of knowledge and Foucault's focus is on those which he calls ‘human sciences’. His concern is to show that these can be understood as a rule-governed systems of discursive events. The alternative of an action-theoretic account is ruled out by Foucault's declared intention of avoiding recourse to a concept of human agency. Thus Foucault does not theorize discourse as an expression of human subjectivity. Rather he theorizes the subject as an image of the human being which is produced by, and presumed in, self-organizing systems of knowledge.In Foucault's work up to and including the Archaeology of Knowledge, therefore, the discursively constructed subject is not a flesh-and-blood human being at all. It is a thought-object constructed by, and within, the human sciences. Because there are a number of human sciences there are a corresponding number of constituted subjects, each of which, in the first instance, has currency only within its parent knowledge. In Foucault's earlier Order of Things, however, a unitary ‘contemporary subject’ is theorized as a composite of these constructs. Since the constituting discourses are depicted as evolving autonomously, Foucault is thus able to produce a history of ‘the different modes by which … human beings are made subjects’.All this means that any support from Foucault for the idea that subjectivities are discursively constituted in actuality must rest on Foucault's genealogical phase. In Discipline and Punish, the human sciences are depicted, not as self-organizing fields of knowledge, but as the theoretical arms of various regimes of behavioural correction. Foucault is convincing in his claim that this ‘power–knowledge’ has diffused outwards from the total institutions in which it was prototyped, thence to become the characteristically modern modality of power. He is much less convincing on the question of its effects. Despite Foucault's talk of ‘shaping the soul’, in fact, it is not clear that he has anything at all to say about this. The problem is that all of his descriptions of the various disciplinary orders are ‘top down’ accounts, relying either on the programmes of legal theorists and institutional reformers or on observation of institutional routines by official inspectors. The voice of the inmate is absent entirely, as is any evidence that disciplinary regimes achieve anything more than a calculative conformity to their behavioural dictates.This is not to deny that disciplinary power may impact on subjectivities. The point here is that such an effect needs to be evidenced rather than simply assumed on the basis of (what has been taken to be) Foucault's say-so. In critical accounting, unfortunately, the tendency has been to treat accounting as a discursive system or regime of power–knowledge and then cite Foucault as if this were sufficient to establish that it works through the production of subjectivities. The paper concludes with a discussion of two recent examples, one of which appeals to a concept of discursive constitution and one to the concept of power–knowledge.