Foucault and Derrida have been disseminated and transcribed in this country for several years now, most often as the common discourse of an analytic addressed against humanism, against the subject, and therefore against the very privilege of "literature" and "humanism." We are still averse to a criticism that opens up literature to other discourses, that entangles it in quasi-scientific methods alien to its self-referential, holistic, or "emotive" space, that "reduces" it, as it were, to philosophy, history, politics, economics (or as Derrida might say, simply to écriture), to the discursive. Yet Foucault and, especially, Derrida signify more than a threat of literature's other, the threat of the discursive to the imaginary. They question the hierarchical division itself, the division on which the very ground of literature's privilege has been erected and maintained. What is called "deconstruction," a term identified with Derrida if not with Foucault, challenges the "whole shebang" (Stevens' metaphor) of hermeneutics, and not simply literary criticism, by not only inverting the priority of literature to analysis, or performative to cognitive language, but by radically shuffling the whole house of cards, by refusing the ontological margin between the two, or, better, by situating both in the same abysmal margin. By reducing their difference to différance. (p. 238)

Foucault's contribution to literary criticism consists largely of rewriting the place of a marginal, disruptive discourse in the history of discourses, or marking and re-marking its significant madness. He would appropriate literature as the excluded or repressed voice, that which cannot be accounted for by history. It is little wonder that so much energy these days is being spent on defenses (or reaction-formations) against French nihilism, though Foucault is the very antithesis of a nihilist; and Derrida is misread as a nihilist if one considers his thought as a via negativa or a negative theology. But neither can be considered within the frame of that orderly and humble discourse we define as literary criticism. (p. 241)

Derrida's strategies have been usefully graftable onto a discourse addressed exclusively to "literature" … [but] they have been vulnerable to angry distortion, hortatory dismissal, and, especially, savage reductionism because they have not been able to offer an alternative method, a recentering.

If Foucault has (for largely personal reasons) attacked Derrida for the sterile pedagogy of a criticism which reduces the world to a text and reading to a series of classroom exercises, he has addressed a problem that is taken up by Anglo-American critics of Derrida in a somewhat less theoretical manner. For Derrida does textualize the world, or sees it reduced everywhere to a text (in the sense that language is the inescapable reduction and that the concept "world," like "life," "experience," "perception," etc., is originarily metaphorical, or metaphysical, and hence textual). He therefore poses a question that is intolerable to humanism. That question concerns the "proper" relationship of literary "representation" to "reality," to "truth" or "being" or "presence," and calls into question the way a humanist culture sees literature as at once the revealer and the regulator of "truth." Derrida questions the priority of the author more radically than does Foucault, for whom discourse rules without the singularity of a subject. Derrida, then, alternately destroys the privilege of literature and gives it back the status of a "reality" as the irreducible simulacrum of being. Literature for him is the "scene" where we may find rehearsed the fabrications of centering and decentering that we call by various names, most often dialectics. But as Derrida reads it, the dialectical play, this self-reflexivity of literature, exposes its own fractiveness, unstrings dialectic, reveals its own artifices, just as it exposes the artifices of a metaphysics of being, that metaphysics itself is the "supreme fiction." (pp. 242-43)

Derrida prompts us to a new strategy of reading as "re-reading," to a reading that is always implicated with writing, by which he means either deconstruction or desedimentation. Literature, for him, has always been engaged in this reading/writing or writing/reading, an alogical strategy made egregiously necessary by the turn of modern literature, from Mallarmé and Nietzsche to Ezra Pound's theory of ideogrammic poetry … because of its revolt against representation and the renewal of emphasis on graphic strategies which "decenter" phonetic writing and put in question the "founding categories" of philosophy and science, in particular the "dominant category of the epistémè: being."… On the other hand, as Derrida recognizes, it has been literature or, better, literature as defined by a certain blind reading, that has been employed throughout the history of the West to authenticate and protect this category of "being," to represent it or mythologize it—a literature which philosophy and science (or their hermeneutical strategies) have in turn protected against radical analysis or deconstruction. Thus history interfaces with epic, the notion of the subject with the notion of the hero—but a Poundian re-writing of the "epic," which dislocates the hero and the authorial voice, which submits the text to a graphic marking of the voice, dissolves the narrative glue that held together the epic and the notion of historical repetition as continuity or a repetition of the same, and makes the epistémè of the epic tremble if not collapse. What Derrida insinuates into the idealistic discourses of the human sciences is a (question) mark that exposes the desire for a totalized reading or a totalizing writing, with its closures and unity, its proper representation, and its ideal of an adequate interpretation.

Of Grammatology offers no prescription for literary criticism. In a most un-Eliotic sense, it may teach us to care and not to care, to approach literature not as the valorized text of the humanist tradition but as the ground-breaking texts that put that tradition and its tyrannous grammars in question. Derrida's inaugural chapter is tantalizingly called "The End of the Book and the Beginning of Writing," and while it provides a resume and critique of modern linguistics and grammatology (particularly since the early eighteenth century), its strategy is to undermine the very notion of this history. If we are situated in a time in which the "book" or the possibility of global systematic thought is precluded, we are not necessarily in the time of some new, immaculate, originary beginning for which writing is an honorific metaphor. If Hegel is Derrida's name for the last author of the book, he is no less the name of the thinker who opens philosophy to deconstruction: "the last philosopher of the book and the first thinker of writing."… Nietzsche and Husserl, Heidegger and Freud signify the diacritical unfolding of Hegel's extraordinary text, which in Derrida's reading is represented by something like Hegel's "pyramid." But it is perhaps modern literature, by which one might mean Flaubert, but more likely Mallarmé, and in particular a certain nonphenomenological reading of Mallarmé, that signifies for Derrida the dis-closure of the "book." And this "event," which is never to be located in a single text or a single philosophical argument, inaugurates itself in a "writing" that is, like literary criticism, an attempt at some totalization or recuperation of meaning but which doubles back upon itself in every gesture, undoing its own metaphors as it reworks the metaphors of the text it is addressing. The margin between literature as the "book" and criticism as "writing" is erased, and with it our habitual way of thinking of criticism as always posed before the blank face of the always literary revelation.

Derrida's text, then, is an exemplary questioning of the very notion of the exemplary. It cannot be a model for criticism because it poses itself against the notion of the model and of systematic methodology. The most abusive appropriation of Derrida is to cite him as the prescription of a method, since in his texts the movement of self-reflection, or the generalizations which he intermittantly makes on the progress of his own analytic discourse, are in themselves interventions in or questionings of the practice. The practice is always breaking itself up into several styles. And this might well serve as a lesson about the so-called self-reflexivity of modern literature, its tendency to comment on itself. This self-mirroring is never a proper mirroring, but like a "reading," it breaks up the "line," the development or unfolding, of the poem without completing or clarifying the "line." It refuses the poem's closure, and signifies the very thing most hostile to modernist poetics: that the literary text, like the critical text, can never be present to itself, but always plays between at least two dissonant texts, the one irreducible to the other. The model of criticism addressing literature has, according to Derrida, already been rehearsed in what we call literature, or the single literary text, just as it has been in the venerable books of philosophy. All texts are intertextual layerings, and to desediment a text, to grope down to its lower layers, is only to arrive at another text, a representation at the bottom of representation, as both Melville and Nietzsche revealed of literature and philosophy.

So if I conclude this "re-view" by citing a general argument for the reading Derrida proposes, a way of reading that may very well inaugurate new critical discourses, I must do so by denying that his proposals can ever direct or restrain the criticisms they initiate. In the midst of his reading of Rousseau, Derrida interposes a subsection that is a discourse on his method. In keeping with the tendency of his argument, it is an alternation of direction, a deviation, an asymptotic reprise. He calls the section "The Exorbitant. / Question of Method," and every term is a double, since exorbitance or excess is always that which cannot be accessible to method, and method is never accountable for the fragmentation of argument. Hence exorbitance marks the curious economy of method, even as it does Rousseau's notion of Nature. What is at issue in this section is Rousseau's "reading" of Nature, which Derrida construes to be the very anomaly of his system, since in Rousseau Nature is nothing if not a text (écriture) and therefore marked everywhere by the exorbitance of the notion of the supplément. This directs Derrida to the question not only of how to read Rousseau, but of Rousseau's reading, and therefore to the order that provisionally governs a text that presumes to represent something beyond or outside of language, and in turn grounds the language that the text employs. Moreover, there is the question of a certain author's deployment of metaphors or substitutions, his own freedom to use the received language, and the constraints of a language he can never totally dominate, that escapes his emphases, that means beyond his means…. (pp. 245-48)

Derrida refuses us the ideal of the "autotelic" text and, therefore, the ontological distinction between a creative and a critical discourse so formidably advocated by those like T. S. Eliot who, while recognizing the interplay of the creative and the critical in poetry, subordinated the critical to the creative and denied the critical text anything but a supportive function. It is precisely this attack on the autotelic text and the possibility of a recuperative criticism, however, that provokes some rather vulgar misreadings of "deconstruction"—those which appeal to a "post-modern" privilege, celebrating any new art that appears to break up the page or new criticism which would similarly imitate graphic anarchy. If Mallarmé and Pound represent a certain kind of experiment that for Derrida signifies the return of writing, and if his own involuted style (or styles) and his own occasional experiment in radical textuality (as in the bi-columnar Glas) seem to suggest that "deconstruction" is only a kind of gimmickry, that is the result of reading the "gesture" for the abyss it imitates. Postmodern playing with the page becomes another kind of imitative style and another form of closed, cognitive "reading." Derrida no more justifies a criticism written in any (subjective?) way than he permits a critical reading that can say everything. The critic must always contend with the metaphorical play of the text he re-writes, and over which he has no control.

It is a certain strategy of deferral, a spurring or fraying of rhetoric (and Derrida's terms here, borrowed from Freud, do not translate well), a certain kind of self-interfering structure (Hugh Kenner's term) or an elliptical interplay between the "imaginative" and "discursive" (imaginary differences) elements of the text, between its spatiality and its temporality, if you will, that identifies Derrida's murder of hermeneutics—so like Freud's notion of the murder of a text in Moses and Monotheism. If he can offer us a theory of "deconstruction," then, it is with pen and not tongue in cheek, with a stylus in his mouth, which must always produce (what Freud calls) Moses' stutter. Criticism can never imitate a literary text, and the literary text can never itself imitate anything but its own departure from itself. What Derrida calls the ideal of criticism, "doubling the commentary," has always been an uninterrogated and unrealizable desire, so that when he talks of its opposite, of going beyond this doubling or transparent "reading," he already marks this "doubling" as impossible. If literature has never been anything but a "deconstructive" reading, a reading of itself, whether one follows the theory of Paul de Man or of Harold Bloom, neither has criticism. This is all the more evident when criticism tries to "double" the text, to imitate it or represent it, and yet to feign its own humility, to efface itself so that the literature can speak directly. Derridean "deconstruction," which advocates another "reading," an opening of the text, therefore only describes what the illusory doubling or imitation does when it forgets. But when he recommends to us a method of "opening," of a radical re-reading, he can only remind us to remember what is always already forgotten. One cannot, as he says, "produce" a theory or task of "reading" except by a negative…. (pp. 248-49)

Criticism has never done what is has claimed: it has neither doubled the text nor left it uncontaminated. But it has never departed from the text, become its own creative moment. For the "creative," like "life," "experience," and "truth," is a metaphysical notion, which appeals to a "referent" beyond the text for which the text stands as substitute. Derrida, who poses as the an-archist against the tradition, has persisted in showing us the limited future of an illusion, of the ideal of a scientific criticism which might extract this referent beyond from that which in the text represents it, the text's privileged signs. But he has repeatedly warned against thinking of the careless freedom of its opposite, saying "almost anything," citing a meaning where literature offers only silence. He remains within "deconstruction," and thus within the language of metaphysics, a conservator of criticism in the best sense. (p. 250)