This commentary is part of The Atlas Society's 1999 online "CyberSeminar" entitled " The Continental Origins of Postmodernism . "
Foucault (as Derrida interprets him) says that: “Descartes is not interested in madness, he does not welcome it, he does not consider it. He excludes it by decree” (47). Will Thomas then adds: “At some length, Derrida offers an alternative reading of Descartes, one that is ‘banal’ by his own admission.” This alternative reading amounts to saying: “Descartes does (implicitly) consider madness and does claim that it offers no exception to the certainty provided by the Cogito.” The passage Derrida quotes is one from the Discourse on Method, which declares that the Cogito is “so certain and so assured that all the most extravagant suppositions brought forward by the sceptics were incapable of shaking it.”
I find this debate about the history of philosophy more interesting than what Derrida has to say about his own philosophy. But having no fund of scholarship regarding Descartes, I must proceed to analyze the debate largely on the basis the facts presented above.
As Objectivists, we reject the skeptics’ game of “It is possible that...” But Descartes did not. He apparently believed that he could be certain of his knowledge only if he could exclude all objections based on the assertion of possibility, and he apparently believed that he had some way of knowing whether an assertion of possibility stood as an objection or had been overcome, even when the assertion had been uttered without evidence.
Now, early in the First Meditation, Descartes considers whether he is certain of knowledge that is based on sensory observation and that regards large objects near at hand. Is his certainty undermined, he wonders, by the fact that “certain persons, devoid of sense, whose cerebella are so troubled and clouded by the violent vapours of black biles that they constantly assure us that...they are clothed in purple when they are really without covering....” Descartes answers: “They are mad, and I should not be any less insane were I to follow examples so extravagant.” He then goes on to pursue the possibility that he is dreaming.
This is the passage that, Foucault says, shows Descartes excluding madness by decree. And this is the objection based on the possibility of madness that Derrida says Descartes later dealt with. Without going into the specific answer that Derrida believes Descartes made, let us look at the question schematically.
Descartes: I am certain of propositional knowledge based on the sensory awareness of large-scale objects near-at-hand.
Interlocutor: But it is possible that you are mad.
The problem is that, according to Descartes, the very possibility of making a counter-argument depends upon the assumption that he is not mad. As David Kelley put it in Evidence and Justification, Descartes held that “the acceptance of a proposition p is justified in accordance with some epistemological rule R only if the subject has determined that accepting p does comply with R” (7). But this determination can be made only if the subject is rational. To be specific: Descartes’ acceptance of the proposition “I am sane” is justified in accordance with some epistemological rule R only if Descartes has [rationally] determined that accepting “I am sane” complies with R.
On this basis, I think Foucault may have the better of the argument with Derrida, and perhaps this is why Descartes writes, in reply to the fourth set of Objections, “the power of thinking is asleep...in maniacs.” If the power of thinking itself is asleep in madmen, the question of arriving at certain knowledge does not arise for them. They could never get to the point of saying, “I may be mad, but I have grasped (by reason) that ‘I think, therefore I am.’” Therefore, Descartes does not welcome their presence in his argument.
Beyond the question of madness in Descartes’ philosophy, however, there lies the more interesting question of madness (and rationality) in Objectivism . (Not having any acquaintance with the seriously mad, I here rely upon the long quotation from a paranoid cited in David Kelley’s The Art of Reasoning [3rd edition, 578-79].)
Start with Kelley’s Evidence and Justification: “To know a fact inferentially is to know it by means of its relationship with other facts. Those other facts are the evidence for the conclusion....These facts, and the relationship between them, exist regardless of whether I know them or not. The concept of evidence pertains to what is ‘out there.’...[Nevertheless,] what justifies my acceptance of the conclusion is therefore not the evidence per se, but my awareness of the evidence....For a subject to be justified in accepting a given proposition, he must have some grasp of the evidential relationship on which it is based....If he knew the premises to be true but saw no relation between them and the conclusion, then his acceptance of the conclusion would be arbitrary” (12-14).
A question Kelley does not discuss explicitly is what we should make of a person (like the paranoid) who sees evidentiary connections everywhere. From a third-person perspective, we may say, in a particular case, “Yes. The facts he cites provide evidence for the conclusion he mentions.” But must we then go on to say, hypothetically, “If he actually grasps the evidentiary connection here, then he knows the conclusion”?
How exactly do I know that I am sane or rational?
I wonder if the matter is that cut-and-dried? If the paranoid is constitutionally incapable of distinguishing a grasp of an evidentiary relationship from a completely arbitrary relationship, can we ever say that he has grasped an evidentiary connection? If a judge rules in whatever way he feels like, but sometimes he feels like ruling in accordance with the evidence, is he rendering objective judgment in those cases? Perhaps that is a bad analogy, but it is the best I can think of.
Suppose, then, that we conclude the madman can never properly be said to grasp an evidentiary relationship. We are led to the conclusion that none of the madman’s judgments are knowledge, and thus we return in a new way to Descartes’ problem.
When I am awake, I know I am awake. When I am dreaming, I do not know I am dreaming. But the second fact does not undermine the first. To say it does, is the mistake of the “It’s possible” school. By parallel: If I am sane (rational), I can know that I am sane, even if a mad person cannot know he is mad. But how exactly do I know that I am sane/rational? It cannot be inferred knowledge, because all inferred knowledge depends on the fact that I am sane/rational (if my argument above is correct). Is it, then, an axiom or a corollary of an axiom?
We might well argue that it is one way of looking at the axiom of free will. “The integrative processes of conceptual thought must be initiated by conscious effort, and they must be consciously directed in such a way as to avoid error and exclude subjective whims, biases, and preconceptions. If we could not control our integrative processes volitionally, we could have no confidence in the validity of their products. In this respect, volition is a fundamental feature of conceptual thought, and the choice to think might also be described as the choice to be objective, to be governed by facts rather than whims” ( The Logical Structure of Objectivism , Beta Version, 28-29).
So perhaps Foucault was onto something. Descartes, whether he fully realized it or not, had no way (within the context of his philosophy) of presenting an argument against the possibility that he was mad. And, having no grasp of the Objectivist axioms, he could rule out the possibility that he was mad only “by decree.”
Will Wilkinson wrote:
I’ve been strapped for time, and have not read the Derrida essays in any detail. I began the first and then quickly petered out, finding Derrida’s opaque style annoying, and having been already exhausted by Saul Kripke’s and David Lewis’s respective brands of stupefying clarity. Anyway, here is what I thought of the first few pages of Derrida just in case I don’t find the time later to say anything more.
I found what I took Derrida’s first point in “Cogito and the History of Madness” to be maybe vacuous and probably false. The point: someone who is really mad can’t tell us what it is like being mad, because he is mad, and someone who isn’t mad can’t tell us what it is like being mad, because he isn’t.
If the point is that the experience of madness is entirely ineffable, that it is the sort of thing we have no vocabulary to describe, then fine. But I would think this point in need of an argument. Why is our usual complement of concepts insufficient for describing what madness is like? Of course, no description of madness will make the reader herself experience it. But we don’t expect an apt description of anger to make us experience anger. Why should madness be any different? Derrida seems to assume that an account of madness would have to draw upon a language with very special or impossible properties. He seems to think a *real* account of madness would require a mad language. But does an account of anger require an angry language, or an account of cats a catty language? I don’t suppose it does.
Now, I’m almost certain that this is a trivialization of Derrida’s position. But since I cannot otherwise determine what that position is, I’ll leave it at that.
One last thing. Derrida suffers terribly from logorrhea. He just can’t shut up. He goes on and on belaboring his point, delighting in his own noises. Others (no one here) have told me they find Derrida *playful*. I find him tiresome and overindulgent.
I apologize for the shallow and flippant nature of these comments. I hope to find time to read more, and thus say something more substantive later in the week.
David L. Potts wrote:
Will Wilkinson writes: “Others (no one here) have told me they find Derrida *playful*.”
This sounds to me like somebody parroting Derrida’s own self-promotional shtick. Play is one of Derrida’s key concepts, play being all we’ll have once objective science has been exposed as incompetent and discarded in the coming “future epoch of differance” (Of Grammatology 93). I think we’re supposed to “get” that Derrida’s puns and word games are a part of this “play,” this liberation from “linear” language. (Derrida attacks objectivity as requiring a “linear model” of language at 85-87. It is now common to use “linear” as an epithet of reason, and I wonder whether this is the source.)
Derrida has a book I saw in a bookstore--I think it was Glas--which is an oversized volume in which separate chunks of text are printed in all manner of different font sizes and faces, jumbled together on the same page. This also, I presume, is “play.”
Play is “the absence of the transcendental signified” (Of Grammatology 50). Or, in a somewhat less clear but semantically equivalent metaphor, “the disruption of presence” (“Structure, Sign and Play” 292). What this means is that symbols, including words and sentences, never have any extra-symbolic referents. They refer only to other symbols in an endless chain anchored to nothing.
If you take this conception seriously, it implies two things. One is, that since reference never really happens in language--that is, since there is never any reference to anything outside language itself--language really does become a game in a rather literal sense. It’s like chess: the pieces may be called “King,” “Knight,” etc., but they actually refer to nothing beyond the rules that comprise the game itself. Language therefore is play; it’s hard to see what else Derrida could call it. The other is, that more than just language is included in this play. If symbols don’t really refer, they aren’t significantly different from any other rule-governed system. Thus everything becomes “writing.” And writing is governed by nothing but play.
> Return to the parent page for this 1999 online CyberSeminar, "The Continental Origins of Postmodernism."