The ideas that Althusser and Foucault articulate seem very different, but may actually address different aspects of the same phenomenon. Althusser argues that the primary function of ideology is to construct a subject—in both senses of the word (i. e. as self-presence and as subordination to an other—note, not just "another," but "an other"): in effect, for Althusser one has a sense of self because one is subordinated in ideology. The ideology interpellates or "hails" the individual as it constructs the individual's sense of her or his subjectivity. Although Foucault himself says he wishes not to identify discursive practice (or "discursive formation") with ideology (38), nonetheless one can affiliate discursive practice and ideology as Althusser is defining it as one's imaginary relationship to real conditions. Foucault complicates the issue by considering the way in which possible subject positions proliferate under the impetus of discursive practices. So, where Althusser addresses himself to the paradigm of an a-historical (hence synchronous) ideology, Foucault addresses himself to the syntax of ideologies, which he calls discursive practices, and how different syntaxes arise historically and produce an ever-shifting range of subjects, all of them an "imaginary" that arises from the "real conditions." In Foucault as in Althusser, that "imaginary" sense of self is absolutely naturalized because, after all, one can be nothing other than what one always already is. Both Althusser and Foucault make imperative the idea that no one or no group is responsible for the rise and development of ideology or discursive practices, bur rather that each discourse writes itself. Althusser associates that notion with the ideological state apparatus that serves as the set of social realities that produce the customary behaviors that define one's subject position; Foucault calls the phenomenon the discursive practices that function in exactly the same way, i. e. as the habitual actions and statements that define a subject position. I'd like to pursue the idea that each discourse writes itself in order to suggest just how important it is in a postmodernist perspective.
One of Foucault's major works is the analysis of sexuality, in a series of volumes published under the running title, History of Sexuality. Please note, by the way, that the title tells you right away that Foucault is engaged in understanding a discursively constructed "object"—sexuality—rather than the "thing" that underlies that "object"—sex (see Archaeology 47): he wants to know what one does "with" and "to" sex, how one talks about it, to whom one talks about it, what aspects of it are significant to talk about, etc. As he engages his subject in the first place, Foucault is very careful to demonstrate that the discursive practice at any given point in history has a genealogy. We in the 20th century are the heirs of 19th century discursive practices; they were the heirs of 18th century discursive practices; the 18th century inherited the discursive practice of 16th and 17th century religious commands; 16th and 17th century religious habits were the heir of 12th century confessional practices—and so on back to the discursive practice of the ancient Greeks. The point is that no one is responsible for the discourse at any given period: the discourse arises more or less in the same way as a new species arises in the natural world, in response to new "environments," in this case social and cultural environments that are themselves discursive practices. Moreover, just as intermediate stages between two species are difficult to define, so the question of historical periodicity is also difficult to define. Foucault substitutes "genealogy" for historical periods, and examines the organization of a given genealogical stratum as an archaeology.
Foucault's choice of sexuality as the focus for his last major work is not accidental: as he says in the first volume, sexuality is one of the most important discourses in the formation of identities, of subjects who are subjects because they've been constructed by the self-written discourse of sexuality. The connection within a given archaeological level of what Foucault says and what Althusser says is pretty clear. What Foucault says about homosexuality is particularly enlightening in this context: up to the 19th century sodomy was an act; suddenly, in the discourses of the 19th century, homosexuality became a way of life, in which sodomy was an act of some importance but which entailed much more than simply that act. The homosexual was constituted as a "subject," as an identity that defined a self, by being made a "subject" of study.
The choice of sexuality as focus traces one supremely important discourse and shows how it opens the way to an infinite series of fragmentations of the human body according to nuances of sexual habits: each fragment is the locus for "subjects" to be formed. But the richness of sexuality—or of the discourses Foucault refers to in Archaeology, medicine, grammar, etc.—as a locus for discourse also permits some degree of obscurity in illustrating the role and function of discourse. What I'd like to do, then, is to bring up another focus for discourse, less central and so perhaps clearer as an illustration of how discourse functions. Once again, notice that the discourse writes itself—no one controls it.
As with most things nowadays, what helped me to conceive of what I'm about to say was TV. On Friday night an ad for one of those incredibly numerous talk shows came on. The man told me that his show would treat of "overweight people who have found others that think like them and are now happy with their overweight partners." It took me a few minutes to recognize how bizarrely typical of discourse this statement is. I was especially struck by the implication that "overweight" people think in a particular way, that they have a peculiar identity. I wondered what that way might be and what might have led to that particular way of thinking. Then a series of loci (or "surfaces"—Archaeology 41) for the discourse of obesity became more and more obvious to me.
First of all, obesity is a bit like sex: one talks around it, but rarely comes right out and names it. Paradoxically, both to talk and not to talk about it mean that one is actually granting to the phenomenon a great deal of importance—in other words, one is treating obesity as a thing about which one can talk or about which it is better if one is silent. Obesity then is not just simply something that is somehow attached to a person. Compare, for instance, the way one talks about an arm to the way one talks about obesity. In the first case, the subject is unproblematic and so not subject to silences (except in the case of the discourse of baseball, for example, where the rotator cuffs of pitchers become important; or of industry where one loses an arm; or of genetics, where mutations produce armless children—and here too, the same delicacy of naming the thing comes into play); in the second case the mere fact of obesity is seen as a problem: one can argue, in fact, that the general silence about obesity is a measure of its problematics. But in fact, the silence does not extend itself to all registers. The truth of the matter is that a whole mechanism for the examination of obesity has come into being. There are psychologists who specialize in obesity. There is Overeaters Anonymous. There are experts who warn about the dangers of heart disease. There are angry tax-payers who argue that the obese should not be covered by health insurance because their diseases are their own fault. There are moralists who descant on the imperative of taking responsibility for oneself, etc., etc. There is a First Lady, Michelle Obama, who takes up the cause of children and obesity. Knowledge about obesity is elaborated in such registers—and note that "register" means each of the spaces in which conversation about obesity is licensed—the doctor's office (a medical register), the counselor's couch (a psychological register), and so on.
By this point the nature of the discourse becomes clear: because being overweight is a "problem," as suggested by silence and by what is said about it, it follows that the condition of being overweight is contested. The medical and psychological discourse on the subject makes it clear that one should not be overweight. A secondary market of "cures" for the problem arises—spas for the obese, diet drinks (Pepsi, Coke [decaffeinated{classic or not} as well as caffeinated {classic or not}], Seven Up, etc., etc.), diet meals (also low sodium or not), diet pills, weight loss clinics, etc., etc.
Finally, then, overweight people are made subject to/of these discourses. They can no longer simply be overweight. They are now centrally aware of the characteristic that distinguishes them—i. e. the signifying difference that defines them. They become not just overweight, but the overweight, the obese—the problem of obesity, in short, becomes a locus for the crystallization of a subjective sense: they are subjects (the obese) because they've been made subjects of/in the discourse of obesity. Here, by the way, I think that one can see pretty clearly where Foucault differs from Althusser, in ways that make Foucault a much more sophisticated analyst.
But the discourse does not stop here. As Foucault says elsewhere, discursive practices represent lines of power and the contest between lines of power. Having made the obese subjects of their own girth, the discourse now empowers them to re-visualize their material circumstances. Next thing you know, there are overweight activist groups demanding that turnstiles in the New York subway system be made wider. A radical group in San Francisco blows up a revolving door that was not wide enough to accommodate the obese. Demands for wider seats on buses, trains, airliners proliferate. Political Action Committees form and lobby Congress and the President. Newspaper pundits take sides—George Will, surprisingly, comes in on the side of the obese. In short, the political system is engaged in the discourse of obesity and soon federal law mandates that vehicles participating in interstate commerce of any sort have wider seats. Transportation industries grumble at the requirement, and form the Industrial Standards Organization, which finances an ad campaign to encourage self-loathing on the part of the obese. The Obese Liberation Movement is formed as a response, and takes a case all the way to the Supreme Court, which decides that freedom of speech comes before the emotional tranquility of the obese. But the political power of the Obese Liberation Movement forces a retooling of all of the factories concerned with seats in transportation. That entails entirely new industrial models for die and tool manufacturing. Before long even seating arrangements in private homes begin to change simply because furniture manufacturers use the same tools to build domestic as public seating arrangements. So all chairs, sofas, benches, etc. are now made wider. Here again a secondary market in the parts business flourishes.
In academic and artistic discourses the new awareness of obesity engenders a re-view of critical judgments and practices. A new study of Shakespeare's Henry IV plays argues that the obese Falstaff is the real hero of the sequence. Post-obese artists refigure the ideals of beauty and soon the new image becomes evident in TV commercials where overweight models are in demand. The Industry Standards Organization withdraws all of its ads and capitulates entirely to the new requirements inscribed in law and practice.
But, of course, there are discourses that resist the hegemonic power of the discourse of obesity. The medical community objects against obesity on the essentialist ground that being overweight is bad for health. A radical wing of the Obese Liberation Front therefore bombs some clinics, and a splinter group of that wing spikes Diet Kool Aid with psychoactive drugs. But by this point the discourse of obesity has intersected so many other discursive practices—industrial, political, civil libertarian, artistic, etc., etc.—that even the AMA comes to a compromise solution of the problem. A new ideology of "safe weight" is formulated. Still, in furtive conversations that take place in locker rooms everywhere, "fat jokes" are whispered from skinny person to skinny person.
I stop here simply because by this point I hope that the nature of "discourse" is clear. I also hope that the reason why one really cannot define the term is clear: to define it would mean that one gives it a rational formulation—but the characteristic of discourse is that it evades rationality and invades material practice—or, as Foucault says, it works within "systems of dispersion" (Archaeology 37).
Finally let me also point out that a discourse of obesity is not far-fetched. Whether it achieves the sort of significance that the discourse of sexuality achieves is, of course, much in doubt simply because not everyone is obese, but everyone is sexed. This suggests that not all discourses engage the same degree of social energy. So, for instance, the discourse of cellulite, which once was pretty noticeable on TV, seems to have come to naught. But being overweight may be closely affiliated with the discourse of sexuality itself—the pleasures of eating are, psychoanalytically speaking, akin to the pleasures of sex, as Mr. Palomar discovered when he finds himself in the presence of a pound of goose fat. If the discourse of obesity is thus associated with the discourse of sexuality, then their coincidences and overlappings may energize the discourse of obesity and make it become culturally significant. In fact, in the silencing of obesity one might argue that it already is a significant discourse, so that anorexic and bulemic women—and it's not irrelevant to the coincidence of food and sex that for the most part it's a particular sex that suffers from those ailments—are embodying the silence of the discourse. In any case, the discourse of obesity, like the failed discourse of cellulite or like the very powerful discourse of smoking, is one of the discourses of the self that, Foucault suggests, are made possible in the first place by the discourse of sexuality itself. All these discourses enable the subject: without them, one cannot crystallize a self from out of the undifferentiated mass of human possibilities. Such enabling, Foucault suggests, leads to a rich creativity—the whole industry of wider seats, for instance. But the self thus made into a subject is also a mere fragment within the continuum of human practices. In that sense, the crystallization of a self according to discursive practices has the effect of isolating the individual within the limits of the discourse—that is to say, if we mix Althusser and Foucault, of subjecting him to the power of the self-written ideology. Such paradox, as well as such fragmentation, is the condition of postmodernism.
Note as well that the contestation of the idea of "obesity" makes possible a point of historical inflection, a borderline or liminal moment: before the contestation we're in one historical period; after the contestation we're in another one. For Foucault, then, historical periods are really an expression of the way that discourses sort themselves out, not simply a matter of one monarch or president ruling and then being followed by another monarch or president.