Derrida analyzes Saussure’s
Cours de linguistique générale and the linguistics of the first half
of the present century in the chapter of this book entitled “Linguistics and Grammatology,”
and the argument about Saus-sure is best presented there. We might simply say that Saussure
was not a grammatologist because, having launched the binary sign, he did not proceed to put
it under erasure. The binary opposition within the Saussurian sign is in a sense paradigmatic
of the structure of structuralist methodology. “We must doubtless resort to pairings like those
of signifier/signified and synchronic/diachronic in order to approach what distinguishes struc-
turalism from other modes of thought.”66
In the passage where Lévi-Strauss acknowledges his debt to Troubetzkoy, for example, we
notice the reference to a study of the unconscious infra-structure. In Derrida, via Freud, there
would be a difficulty in setting up the opposition between the conscious and the unconscious
within the sub-
((lix))
ject as the founding principle of a systematic study. The unconscious is undecidable, either the
always already other, out of reach of psychic descriptions, or else it is thoroughly and
constitutively implicated in so-called conscious activity. Further, as I have pointed out, the
opposition of the subject and the object, upon which the possibility of objective descriptions
rests, is also questioned by the grammatological approach. The description of the object is as
contaminated by the patterns of the subject’s desire as is the subject constituted by that never-
fulfilled desire. We can go yet further and repeat that the structure of binary oppositions in
general is questioned by grammatology. Differance invites us to undo the need for balanced
equations, to see if each term in an opposition is not after all an accomplice of the other: “At
the point where the concept of differance intervenes . . . all the conceptual oppositions of
metaphysics, to the extent that they have for ultimate reference the presence of a present, . . .
(signifier/signified; sensible/intelligible; writing/speech; speech [parole]/language [langue];
diachrony/synchrony; space/time; passivity/activity etc.) become non-pertinent.” (Pos 41)
It is therefore not too extravagant to say that “writing” or “differance” is the structure that
would deconstruct structuralism—as indeed it would deconstruct all texts, being, as we shall
see, the always already differentiated structure of deconstruction.
It should by now be clear where the structuralists have stopped short, or what they did not
begin with. Thev have not thought the “sous rature.”67 It is as if they have grasped only
Nietzsche’s “knowledge,” showing us the interpretive power working through human society,
so that all its studies become “genealogical,” an unending decipherment of sign-chains. How
close to that aspect of Nietzsche this passage from Roland Barthes sounds! : “... structural
man . . . too listens for the natural in culture, and constantly perceives in it not so much stable,
finite, ‘true’ meanings as the shudder of an enormous machine which is humanity tirelessly
undertaking to create meaning, without which it would no longer be human.”68 But it is also
as if the gravest lesson of that knowledge, its need for abdication, has not been imagined by
the structuralists. Nongrammatological structuralism cannot afford to cultivate the will to
ignorance: “Homo significans: such would be the new man of structural inquiry!” 69
The solution is not merely to say “I shall not objectify.” It is rather to recognize at once that
there is no other language but that of “objectification” and that any distinction between
“subjectification” and “objectification” is as provisional as the use of any set of hierarchized
oppositions. Derrida sets this forth most energetically in two early essays where he deals with
two structuralist critics who take elaborate precautions against objectification. I have already
referred to one—“Structure, Sign, and Play in the Sciences of Man”—where Derrida
interprets Lévi-Strauss’s attempt
((lx))
at a mythomorphic criticism of myth. The other is « Cogito et l’histoire de la folie »—a
critique of Michel Foucault’s Histoire de la folie à l’âge classique.70
Foucault writes, writes Derrida, as if he “knew what madness meant” (ED 66). Foucault
speaks thus for Reason, madness’s other—if his own binary opposition is to be trusted. Yet he
wishes to speak for “madness itself” (ED 56), write “the archaeology of [its] silence.”71 But
how can this be more than merely rhetorical? For an archaeolgy is perpetrated through
discourse, imposing reason’s syntax upon folly’s silence. Indeed Foucault recognizes the
problem and on occasion articulates it.
But, “to say the difficulty, to say the difficulty of saying, is not yet to surmount it” (ED 61).
Foucault sidesteps precisely this issue, says Der-rida, by misreading Descartes.
Foucault sees in Descartes one of the exemplary separators of reason and madness. Derrida’s
reading of Descartes on folly is an elegant bit of deconstruction; he spots the moment of the
forgetting of the trace in Descartes’ text. Descartes, he argues, gives the name “folly” to the
fire-reflexive cogito—before the “I think” can be reflected upon and pronounced. In the
prereflexive cogito “folly” and “I think” are inter-changeable, intersubstitutable. There the
distinction between reason and folly does not appear. There the “cogito” cannot be
communicated, made to appear to another self like my own. But when Descartes begins to
speak and reflect upon the cogito, he gives it a temporal dimension, and distinguishes it from
madness. The relationship between the prereflexive cogito (which is also madness) and the
temporal cogito (which is distinct from madness) is thus analogous to that between the
precomprehended question of Being and the propositional concept of Being. The possibility
of discourse is lodged in the interminably repeated movement from the one to the other—
from “excess” to a “closed structure.” (ED 94) Foucault, not recognizing this, still remains
confined within the structuralist science of investigation through oppositions.
This is a dated Foucault, the Foucault of the sixties. Even then he was violently unwilling to
be called a structuralist, and he gets into this section of my preface because he diagnoses an
age in terms of its epistémé, the self-defined structure of its knowing. This particular
characteristic of Foucault’s work has not disappeared. To diagnose the epistemic structure, he
has had, with repeated protestations to the contrary, to step out of epistemic structures in
general, assuming that were possible. To write his “archaeologies,” he has had to analyze
metaphors privileged by a particular age in what Derrida would call “meta-metaphorics.” By
describing grammatology as “a history of the possibility of history that would no longer be an
archaeology,” (43, 28), Derrida seems to declare an advance over Foucault. And by deny-ing
the status of a positive science to grammatology, he “erases” the ad-
((lxi))
vance. Perhaps there is an attempt to rewrite the Foucauldian method in “The White
Mythology,” Derrida’s extended essay on metaphor: