Instead of adding directly (the Nietzsche titles on my bookshelf glare at me sternly–don’t be reactive!) to some of the heated recent blogging, tweeting, and other exhausting backs-and-forths over the power of words and the question of self-policing of language in feminist, queer, and critical race fields of knowledge production, I feel moved to think out loud about what “discourse” is and how it can be usefully analyzed. Some of the tense debate of recent months, not to mention days, seem to me to be a repetition of questions raised (but undoubtedly not well answered) by the linguistic turn: do words hurt? How? Why? And what is the function of discourse in an era of endemic injury-identity?
In 1976, Foucault authored a short and remarkable text, “Le discours ne doît pas être pris comme…,” an accompaniment to the film project La Voix de son maître, the documentary debut of directors Gérard Mordillat and Nicolas Phillibert. (The text is included in the third volume of Dits et écrits, but unless I am mistaken it has not been translated into English.) Mordillat and Phillibert’s film, which profiles managers at twelve large companies, has come to be remembered as a groundbreaking diagnostic of the transition in the French economy from a dying industrial Fordism to a financialized capitalism. By not interviewing workers at the companies and juxtaposing their voice dialectically to that of the bosses, but instead inviting the bosses to set the agenda and create the setting for their interviews so that their strategic logic could be laid bare, Mordillat and Phillibert manage to pull off escaping, in Foucault’s estimation, a simplistic oppositional understanding of what a discourse is (i.e. managerial discourse v. worker discourse). For this reason, we are given an incredible opportunity to critically analyze the emergence of managerial discourse in the transition to post-Fordism.
Instead of treating discourse merely representationally, in the case of La Voix de son maître, “It is a matter of revealing discourse as a strategic field, where the components, tactics, and weapons never cease to circulate from one camp to the other, to be exchanged between adversaries, and to be used against those who use them” (my translation). If there is an ongoing struggle here, rather than a dialectical opposition that would end in clear victory (such victories always being surface distractions for an exercise of power relations), it is because the discourse is the medium common to the two struggling forces at hand. A discourse is not a mere accumulation of things said or sayable about a certain object, a set of structural rules for its intelligibility, nor a brute epistemological grid; all these are second-order derivatives of discourse as a field of strategic confrontation of force, where discourse is first of all “a weapon of power, of control, of subjection, of qualification and disqualification.”
Foucault’s short piece ends thusly: “Discourse is, with respect to the relation of forces, not merely a surface of inscription, but the technician” [un opérateur]. In the introduction to Society Must Be Defended, Arnold Davidson translates these final few sentences, ending them as “not merely a surface of inscription, but something that brings about effects” (xx). The capaciousness of that translation gives me pause; I see in Foucault’s deployment of opérateur an interesting evocation of the technical operation of machinery, something that draws our attention closer to the ever-reticent word dispositif that continues to elude useful translation. Discourse is not first and foremost a representational field but a dispositif that corresponds to an exercise of power; one consequence of that exercise is representation, but we must not mistake one level for another.
Davidson helpfully places “Le discours ne doît pas être pris comme…” alongside Society Must Be Defended, for Foucault takes up in those lectures perhaps most explicitly the project of “the logic of strategics” (xix), which Davidson defines as an attempt at “provid[ing] an alternative system of representation of the nondiscursive social field,” that is to say, a mode of representation that does not derive from juridical sovereignty. Part of this project entails re-approaching discourse, no longer so much in the mode of archeology under which went Foucault’s early works, but through what I think could be fairly termed a materialist treatment of discourse that itself begins at that “nondiscursive social field.” Discourse is here a specific weapon configured by a certain exercise of power, often a struggle for the truth effects of knowledge. As Davidson summarizes, then, discourse operates on two levels in the lectures Foucault gave in 1975-76: a “tactical productivity” and a “strategic integration” (xx) wherein truth is a weapon in a battle (xxi). In the second lecture of Society Must Be Defended, Foucault touches on this question by way of the enigmatic phrase “an ascending analysis of power” (30), which gives context to the two levels described by Davidson. Using the more general (if no less specific in its deployment) term “power,” Foucault explains:
it is important not to, so to speak, deduce power by beginning at the center and trying to see how far down it goes, or two what extent it is reproduced or renewed in the most atomistic elements of society. I think that, on the contrary…we should make an ascending analysis of power, or in other words begin with its infinitesimal mechanisms, which have their own history, their own trajectory, their own techniques and tactics, and then look at how these mechanisms of power, which have their solidity and in a sense, their own technology, have been and are invested, colonized, used, inflected, transformed, displaced, extended, and so on by increasingly general mechanism and forms of overall domination (30).
This passage is well read through the narrower lens of the strategic analysis of discourse as correlate of a certain exercise of power. That is to say, an ideological critique of discourse on representational terms is to already begin too late, to fall into the trap of a discourse on truth, to capitulate to its victory by refusing to see its tactical mode and strategic organization. Instead, a discourse should not be taken tautologically, which would be to accept its globality and self-referentiality; one must begin at its capillaries and seize upon its diffuse, dis-integrated contexts that are only through historical territorializations given an integration. In this mode of analysis we too become, merely by reading, tactical players in the strategic field.
Of course Foucault is continuing a line of analysis here that, surely at the time, took elliptical aim at Derrida (in La société punitive, the lectures from 1972-73, he takes time to directly critique “il n’y a pas hors texte,” continuing the spat inaugurated over Histoire de la folie). More than that, though, Foucault renders the content of discourse a relatively useless field of intervention. To quarrel over content will have no effect on the tactical organization of a discourse. As he says in reference to the founding of the authority of the university for the disciplines of knowledge in the early nineteenth century, what is at stake here is “a control that applies not to the content of statements themselves, to their conformity or nonconformity to a certain truth, but to the regularity of enunciations” (184).
Keeping that in mind, it strikes me as telling that, to take a radioactive example, the angered reaction at a certain recent academic blog post about “trigger warnings” and censorship have focused on the content of the discourse put forth (it is not true, it is false, etc.), not the tactical strategy it was (whether or not it succeeded being another matter) attempting to put into play, and whether or not that strategy might be one worth dialoguing about (even if the blog post in question presented itself as ultra-sovereign and truthful, hardly inviting dialogue). Investing our analytico-political energy in a struggle over the content of a discourse is, I think, following Foucault, to begin at a derivative level, much to our loss. It is also extremely exhausting (an aesthetic-affective-political problem also worthy of its own discussion) because it does not alter the exercise of disciplinary knowledge carried out under even the most anti-disciplinary names, such as feminist, queer, or critical race studies.