14. Paul de Man¹

[Historical note: This chapter went to press before de Man’s secret past as a Nazi collaborator writing anti-Semitic tracts in Holland, and as a de facto bigamist in the US who had abandoned his Dutch family, came to light. Some writers have since speculated on the significance for his thematic invocations of “ambivalence” and ”ambiguity” (as when he says “the category of the self” is “double-faced  "(BI 105) , but I really don’t see the relevance of anyone’s private life to their professional writing, including my life, which has been such a primrose fountain of gossip and rumours of dalliance at conference cocktail parties.].

[1. The key to de Man citations is: ALC: Allegories of Reading (1979a); BI: Blindness and Insight (1983); RT: "The Rhetoric of Temporality” (1969); and SD: “Shelley Disfigured" (1979b).]

         Until his recent untimely demise, Paul de Man commanded a considerable following who "acknowledged" his “pre-eminence in the field of literary theory" (Godzich, preface to BI, xv). He was hailed by Hartman as a "boa-deconstructor" and by Culler as one of the “best practitioners of deconstruction” (DC ix; PS 16). Like many critics in his “generation," de Man was impelled by "local difficulties of interpretation” to "shift from historical definition to the problematics of reading” (ALG ix). The further probed, the more he became convinced that the "dialogue between work and interpreter is endless” (BI 32). The “act of reading," for which "criticism is a metaphor," is "inexhaustible” (BI 107).

          His vision might appear to be a dualistic or ironic version of the utopia sketched in Chapters 1 and 2.² [2. But De Man refers to “utopianism" as a “trap of impatient “pastoral” thought" (BI 241). For him, a utopia in my sense (Ch. 1) might be on the order of speaking with “a single voice that, by the rigour of its negativity, finally coincides with what it asserts" (ALC 172). Compare Note 15 on "rigour.”] He forever undertook to read texts and master their meanings while categorically asserting the elusiveness or impossibility of the procedures for doing so. With this tactic, he pursued the implications of deconstruction in a more rigorous and consequential way than the other critics I survey, although he tended in this direction, even before he read the books of Derrida, and the two writers proceed rather differently. ³ [3. De Man's convictions about the "infinite plurality of significations" and “the deep division of being” (BI 236f) were declared in a piece written even "before the advent of Structuralism on the literary scene" (BI 229).]

        For De Man, “all true criticism occurs in the mode of crisis," because it “puts the act of writing into question by relating it to its specific intent" (BI 8; cf. BI 62). “The act of criticism" is “defined" by “scrutinizing itself to the point of reflecting upon its own origin” and “necessity" (BI 8). “Recent developments" reveal additional “outward symptoms” of “crisis." One is “the incredible swiftness with which often conflicting tendencies succeed each other, condemning to immediate obsolescence what might have appeared as the extreme point of avant-gardisme briefly before” (BI 3f). “Crisis” is also diagnosed “when a “separation” takes place, by Self-reflection,⁴ [4. De Man's world abounds in entities doing things to themselves, whence the many “self-" compoundings (e.g. BI 18, 113, 134, 208, 212, 215, 220; ALG 131; SD 55).] between what, in literature, is in conformity with the original intent and what has irrevocably fallen away from this source” (BI 8). The “hesitations” of the formalists already indicated a "conception of poetic consciousness" as "essentially divided, sorrowful, and tragic” (BI 241).

        One resolution for the crisis could be that the “dizziness of a mind caught in an infinite regression prompts a return to a more rational methodology” (BI 10). The "attempts” to "inaugurate a more scientific study of literature have played an important part in the development of contemporary criticism" (BI 107).⁵ [Frye is included here too, along with Jakobson and Barthes (Note 21), but all three remained "between the two camps” of “science” and “immanence" (BI 107).] I.A. Richards based his "scientific claims" on "the promise” of a "convergence between logical positivism and literary criticism" (BI 241). “Certain structuralist tendencies" are on a "borderline”: they “try to apply extrinsic methods to material that remains defined intrinsically and selectively as literary language” (BI 107). “Since it is assumedly scientific, the language of a structuralist poetics would itself be definitely “outside” literature, extrinsic to its object, but it would prescribe (in deliberate opposition to describe) a generalized and ideal model of a discourse that defines itself without having to refer to anything beyond its own boundaries. " “Scientific” “methods" could apply only "if literature rested at case within its own self-definition" (BI 164).

         But “this stability” "assumed” by “the structuralist goal of a science of literary forms" "systematically bypasses the necessary component of literature" and “a constitutive part of its language": "the fluctuating movement of aborted self-definition. " “Literature can be represented as a movement and is, in essence, the fictional narration of this movement” (BI 159). “Moving from an actual, particular text to an ideal one” is a tactic to “avoid" “the logical difficulties inherent in the act of interpretation" (BI 107). In de Man's view, “the semantics of interpretation have no epistemological consistency and can therefore not be scientific" (BI 109). “Interpretation" is "like scientific laws" in being a “generalization that expands the range of applicability of a statement to a wider area" (BI 29). But whereas “the natural sciences" pursue "predictability," “measurement,” and "determination,” "interpretation" "claims” to "understand” the “phenomenon” and its "intent.”

         “In order to protect the rationality” of "science," “structural anthropologists” envisioned “a myth without an author,” and “linguists” “conceived of a meta-language without a speaker” (BI 11f). De Man warns that a “fundamental discrepancy always prevents the observer from coinciding fully with the consciousness he is observing; the same discrepancy exists in everyday language, in the impossibility of making the actual expression coincide with what has to be expressed, of making the actual sign correspond with what it signifies" (BI 11).

         An alternate resolution for the crisis could be to "propose a radical relativism” on every “level" from “specific” to "general” (BI 10). Here, “there are no longer any standpoints that can a priori be considered privileged, no structure that functions validly for other structures, no postulate of ontological hierarchy that can serve as an organizing principle from which particular structures derive.” “All structures are equally fallacious and are therefore called myths.” But the “myth” still depends on the "arbitrary act of interpretation that defines it." So again, we are brought back “from anthropology to the field of language, and finally, of literature" (BI 11). And if “conceptual language, the foundation of civil society," is “a lie superimposed upon an error," we "can hardly expect the epistemology of the sciences of man to be straightforward” (ALG 155).⁶ [6. This thesis ostensibly follows Rousseau, but de Man, who often speaks in the name of others, seems to endorse it.]

        This referring of fundamental issues to language and suspending or dissolving them there is among de Man's most thematic moves and helps make his work both intriguing and unsettling to the language-oriented critical profession. Even “in the most physical of modes,” "the abstraction and generality of a linguistic figure" can “manifest itself” (ALC 182). De Man's referral is totally unlike that of the structuralists, who evidently sought reassurance in the clarity of grammatical and logical forms. For de Man, language is precisely the archetypal space of unreliability and opaqueness, rent with complications, fissures, and ambiguities of every sort. We have to take de Man's word for it, because no empirical demonstrations are attempted: "reading" is “an act of understanding that can never be observed, or in any way prescribed or verified” (BI 107).

         De Man prefers to view "reading" as an "allegory" that "narrates the impossibility of reading" (ALG 77, 205). “The allegorical representation of reading" is "the irreducible component of any text” (ALG 77). "The assumption of readability" is “found to be aberrant”; “all readings are in error because they assume their own readability" (ALC 202). In de Man's terminology, a text is “unreadable in that it leads to a set of assertions that radically exclude each other," or to a confrontation of incompatible meanings” (ALG 245, 76). “If one of the readings is declared true, it will always be possible to undo it by means of the other; if it is decreed false, it will always be possible to demonstrate that it states the truth of its aberration" (ALG 76). "Language” is "necessarily misleading" and “just as necessarily conveys the promise of its own truth" (ALC 277).

        This view is supported with the deconstructive thesis, also articulated by Culler (OD 201), of "the aporia between performative and constative language” (ALG 131) (the latter alternately called “cognitive"). "A text is defined by the necessity of considering a statement, at the same time, as performative and constative" (ALG 270). Yet “performative rhetoric and cognitive rhetoric” “fail to converge" (ALG 300). In the “process of reading," “rhetoric is a disruptive intertwining of trope and persuasion or -- which is not quite the same thing -- of cognitive and performative language" (ALC ix). “Considered as persuasion, rhetoric is performative but when considered as a system of tropes, it deconstructs its own performance" (ALG 131). Again, the issue is dissolved into language: “rhetoric is a text in that it allows for two incompatible, mutually self-destructive points of view, and therefore puts an insurmountable argument in the way of any reading or understanding.” "The predicament is linguistic rather than ontological or hermeneutic” (ALG 300).

         “The critique of metaphysics” can thus be “structured as rhetoric” (ALG 131). De Man is not thinking of the “naively pejorative sense in which the term” “rhetoric” "is commonly used" for “a tool" in "the manipulation of the self and of others"⁷ [7. Yet he keeps referring to the rhetoricity of texts as “seduction,” performed by Rilke, Proust, Rousseau, Blake, or Blanchot (ALC 21, 69, 93, 181, 184, 190, 200; BI 62), as well as by literature and figural language in general (ALG 115). Compare the notion of “metaphor” “as a language of desire and as a means to recover what is absent” (ALC 47), reminiscent of Bleich's (far more literal) explanation of the rise of language in the infant.], or a "fraudulent grammar used in oratory" -- as "opposed to a literal use of language” (ALC 173, 130). Nor does he put much stock in "establishing a taxonomy of tropes” (ALC 63). He views "tropes" not as “grids,” but as “transformational systems” that produce "substitutive reversals" (ALG 63, 113). ⁸ [8. De Man finds it “notoriously difficult, logically as well as historically, to keep the various tropes and figures rigorously apart” (BI 284). “Numerical and geometrical models, assuming the specificity of each particular trope, though unavoidable, are in the long run untenable" (ALC 66). As if to prove his point, he likes to play off the tropes against each other. Although “metaphor” based on "analogy” and implying “identity and totality” is made to oppose "metonymy" based on "contiguity" and implying a “purely relational contact,” the "superiority of metaphor” has to be asserted by “using metonymic structures” (ALG 14f, 70). “Chiasmus" is seen less as a criss-cross scheme for word placement (de Man likes it, as in “text of desire” and “desire for text,” ALC 289), than as a “crossing that reverses the attributes of words and things" (ALC 38).] He wants to probe the "structures and relays" in which "properties are substituted and exchanged," and “resemblance” is “used” "as a way to disguise differences” (ALG 62, 16).

        To point up the "figural dimension of language" in general, de Man adduces "the divergence between grammar and referential meaning” (ALG 270). Following generative linguistics, he defines “grammar" as “the system of relationships that generates the text and that functions independently of its referential meaning"; indeed, "grammatical logic can function only if its referential consequences are disregarded" (ALG 268f). “We call text any entity that can be considered from such a double perspective: as a generative, open-ended, non-referential grammatical system and as a figural system closed off by a transcendental signification that subverts the grammatical code to which the text owes its existence” (ALG 270). “The logical tension between figure and grammar is repeated in the impossibility of distinguishing between two linguistic functions,” "performative and constative," which (as we were just told) “are not necessarily compatible.” This contention disrupts the “literary semiology” of “Barthes, Genette, Todorov, Greimas, and their disciples" that "lets grammar and rhetoric function in perfect continuity," "without apparent awareness of a possible discrepancy between them” (ALC 6). In certain passages, though, "grammar" can be “rhetoricized," and "rhetoric" can be “grammatized” (ALC 15).

         Like Derrida, de Man is impressed by Nietzsche's “critique of the main concepts underlying Western metaphysics” (ALC 119). “The critique” indicates that “in these innocent-looking didactic exercises, we are in fact playing for very sizeable stakes" --the more so as “the key" is “literature," that is, “the rhetorical model of the trope” (ALG 15). Nietzsche remarked that "logic'" is "'an imperative, not to know the true, but to posit and arrange a world that should be true for us'" (ALG 120). “The convincing power of the identity principle" essential to logic and to its demands for "noncontradiction" “is due to an analogical, inetaphorical substitution of the sensation of things for the knowledge of entities” (ALG 122, 119).⁹ [9. Compare Einstein (1956: 68): “concepts can never be regarded as logical derivatives of sense impression. But didactic and heuristic objectives make such a notion inevitable. Moral: it is impossible to get anywhere without sinning against reason.”] Nietzsche's contentions raise “the possibility of unwarranted substitutions leading to ontological claims based on misinterpreted systems of relationship," such as “substituting identity for signification” (ALG 123). The postulate of logical adequacy" "might well be based on a similar aberration” that cannot be proven right or wrong"; but “our ontological confidence has forever been shaken.”

        Here again, language is made the agent and model. “The unwarranted substitution of knowledge for mere sensation becomes paradigmatic for a wide set of aberrations all linked to the positional power of language in general, and allowing for the radical possibility that all being, as the ground for entities, may be linguistically” posited, "a correlative of speech acts" (ALC 123). “The linguistic model as speech act” is “established” to be “universal” and yet “voided” of “epistemological authority” (ALG 129). "The possibility for language to perform is just as fictional as the possibility for language to assert." "The critique of metaphysics" "deconstructs” “the illusion that the language of truth" “could be replaced by a language of persuasion” (ALC 130). "Thought as action" is an "illusion" resulting from an "illegitimate totalization from part to whole” (ALC 129f).¹⁰ [10. De Man routinely doubts the propriety of “totalizations” (cf. ALG 237, 249, 256; BI 32, 34), though he acknowledges “the desire for totality” to be “an inherent need of the human mind” (BI 54) (cf. Note 12). And, as I remark at the end, he replaces the totalizing of the literary work with a totalizing from the work to all literature (p. 277).]

        As we saw, de Man would "account for” “the allegorical mode" by describing all language as figural" (BI 136). A “text” may “go beyond this, however” and make its “statement" in "an indirect, figural way that knows it will be misunderstood by being taken literally.” It can “tell the story, the allegory of its misunderstanding: the necessary degradation" of "metaphor into literal meaning." “The. only literal statement that says what it means to say is the assertion that there can be no literal statements” (BI 133). Only “a figure of the unreadability of figures” might "no longer" be "a figure" (ALG 61). “As the report of the contradictory interference of truth and error in the process of understanding,” "the allegory of reading" "would no longer be subject to the destructive power of this complication” (ALG 72). "The statement of the enigma that gives language its necessarily referential complexity might itself be no longer a representation but a single voice that, by the rigor of its negativity, finally coincides with what it asserts” (ALC 172). These prospects would seem to outline de Man's own utopia (though he wouldn't call it that), in which reconciliation is paradoxically attained in the moments of most radical negation." ¹¹ [11. Officially, "reconciliation” too is opposed, especially one “of the self with the world by means of art,” as proposed by Starobinski (BI 218) and Iser (AR 135, 154), and foreshadowed by some Marxists (cf. Jameson in Ch. 17).]

         Not too surprisingly, the closest approximation of this utopia turns out to be literature (cf. Ch. 2). Thanks to its "necessarily ambivalent nature," "literary language" “can only tell this story" – “the allegory of its misunderstanding" – “as a fiction, knowing full well that the fiction will be taken for fact and the fact for fiction” (BI 136). "Literature does not fulfil a plenitude”; its “imagination takes its flight only after the void" “that separates intent from reality," “the inauthenticity of the existential project has been revealed" (BI 34f).¹² [12. If the world could "become a more complete, more totalized reality than that of everyday experience," “art would be the expression of a completed reality, a kind of over-perception” (BI 34) (cf. Doubrovsky, 1966). De Man rejected this prospect on the authority of Merleau-Ponty, whose work led lser to envision a reader being “made to ideate a totality" (AR 141).] Through “irony and allegory" (concepts we will engage further on), "the relationship between sign and meaning is discontinuous"; “the sign points to something that differs from its literal meaning and has for its function the thematization of this difference” (BI 209).

        The whole line of argument leads straight to the point where de Man, like most of our critics, was heading all along. He expounds the dilemmas of language in such radical terms as to seem highly original and controversial. Then, he presents literature as both the essence and the counterpoint of language and founds his authority as the visionary mediator of indispensable, though disruptive insights. “The possibility of reading,” an act which is "prior to any generalization about literature," “can never be taken for granted” (BI 107). “The literary text is not a phenomenal event that can be granted any form of positive existence," but "merely solicits an understanding that has to remain immanent because it poses the problem of intelligibility in its own terms" (a reason why “all interpretation has to be immanent”) (BI 107f).¹³ [13. De Man is more doubtful about how far criticism can be “intrinsic" in the sense of Wellek and Warren. Whereas “form" was previously “considered to be the external trappings of literary meaning or content,” “intrinsic, formalist criticism" has made “outer form” “the intrinsic structure" and “referential meaning” the “extrinsic" (ALG 4). “This reversal" shows that “the polarities of inside and outside" situate “formalism” in "the prison house of language." In “the recurrent debate opposing intrinsic to extrinsic criticism" “the inside/outside metaphor” is “never being seriously questioned” (ALC 5). De Man, however, uses “the couple grammar/rhetoric” to "disrupt and confuse” the “neat antithesis”; and finds a case where “the inside is always already an outside” (ALC 12, 199).] Reading is thereby dramatized to a degree that makes de Man's achievements all the more striking -- a tactic we reencounter, greatly megaphoned, in Bloom's work (Ch. 15).¹⁴ [14. As if to tip us off, de Man opens his volume by comparing his announcement of "crisis" with Mallarmé's “rhetoric of crisis” whose “mock sensationalism" and “ironic slant” “baffled his foreign audience” (BI 3ff).]

        In this project, de Man predictably “equates the rhetorical, figural potential of language with literature itself” (ALG 10; cf. ALC 15). In his view, "literary” “language” “takes for granted” that “sign and meaning can never coincide”;. The “essence”  of “literature" is the “mirror-effect” of “asserting" its “separation from empirical  reality, its divergence, as a sign, from a meaning that depends for its existence  on the constitutive activity of this sign” (BI 17). A “definition” of “literary" is thus derived: “any text that implicitly or explicitly signifies its own rhetorical mode and prefigures its own misunderstanding as the correlative of its rhetorical nature; that is, of its “rhetoricity" (BI 136). “The text” “accounts for” this "rhetoricity” by "postulating the necessity of its own misreading.”

      Yet precisely because the “literary mode” is “a form of language that knows itself to be mere repetition, mere fiction and allegory," it is "the only form of language free from the fallacy of unmediated expression” (BI 161, 17). The "paradox inherent in all literature” is that it “gains a maximum of convincing power at the very moment that it abdicates any claim to truth" (ALG 50). So "literature” is “condemned (or privileged) to be forever the most rigorous.¹⁵ [15. “Rigor" seems to be a mental awareness of the twists and evasions of language (cf. ALC 19, 131, 172, 207) -- just what de Man has. It doesn't guarantee "the epistemological authority of the ensuing results” (BI 289).] and consequently, the most unreliable language in terms of which man names and transforms himself” (ALG 19).

        These contentions restore the special ontological status of literature, though not very reassuringly. The “structuralist literary critics” had thought it "imperative to show that literature constitutes no exception, that its language is in no sense privileged in terms of unity and truth over everyday forms of language" (BI 12).¹⁶ [16. I.A. Richards is also said to "refuse to grant aesthetic experience any difference from other human experiences” (BI 234).] They wanted to “show that the discrepancy between sign and meaning (signifiant and signifié) prevails in literature in the same manner as in everyday language." We thus find "the trend in Continental criticism, whether it” draws on "sociology, psychoanalysis, ethnology, linguistics,” or “philosophy," of launching "a methodologically motivated attack on the notion that a literary or poetic consciousness is in any way a privileged consciousness” (BI 9).

         This trend was closely allied to the denial of the "privileged subject” (BI 12). "In structuralism the loss of the intentional factor” "is due to the suppression of the constitutive subject” (BI 32f). Similarly, “language" is for Blanchot (1949) a “consciousness without a subject," and for Poulet (1965) “the medium" for "a radical questioning of the actual, given self, extending to the point of annihilation” (BI 69, 98). Blanchot (1955: 202) wanted "to take the work for what it is and thus to rid it of the presence of the author” (BI 64). “Language” “speaks and writes by itself”; and “the work has an undeniable ontological priority over the reader" (BI 69, 64). In Poulet's vision, “the critic” also “relinquishes his own self in his encounter with the work” (BI 97). “The subject that speaks" in his “criticism” “is a vulnerable and fragile subject whose voice can never become established as a presence” -- "the very voice of literature" (BI 101).

        These estimations indicate that de Man too is inclined to deny the privileged subject and disperse it into language. We might, he says, “conceive" the “categories" of "subject" and “object” “as standing in the service of the language that has produced them" (ALG 37).¹⁷ [17. A Rilke poem is the occasion for this view, which sounds just like de Man to me. The following quote claims to speak for Baudelaire (BI 213).] "Language" “divides the subject into an empirical self, immersed in the world, and a self that becomes like a sign in its attempt at differentiation and self-definition” (BI 213). “The reflective disjunction not only occurs by means of language as a privileged category, but it transfers the self out of the empirical world into a world constituted out of, and in, language.

         ”Literary "fiction, far from filling the void" “the human self has experienced” within itself," “asserts itself as pure nothingness, our nothingness stated and restated by a subject that is the agent of its own instability" (BI 19). Through “the literary work,” the “subjectivities” of “author" and "reader cooperate in making each other forget their distinctive identity and destroy each other as subjects" (BI 64). Correspondingly, the “loss of the representational function in poetry" “goes parallel with a loss of a sense of selfhood” (BI 172). ¹⁸ [18. These views are aired in regard to Lévi-Strauss, Blanchot, Yeats, and Hugo Friedrich (BI 19, 64, 172). But compare the remark on Mallarmé: "poetry does not give up” “its dependence on the fiction of a self that easily" (BI 182).] For such reasons, De Man feels "literature can be shown to accomplish in its terms a deconstruction that parallels the psychological deconstruction of selfhood in Freud" -- a motive for “the intensity of the interplay between literary and psychoanalytical criticism" (ALC 174).¹⁹ [19. Freud's concept of “defense" seems close to de Man's diagnosis that “the blindness of the subject to its own duplicity has psychological roots since the unwillingness to see the mechanism of self-deception is protective” (BI 113). Freudian imagery is certainly prevalent in de Man's vision of language and the self (Note 31).]

        Nietzsche's “critique of metaphysics" concurs by “showing” "that the idea” of “the human subject as a privileged viewpoint is a mere metaphor by means of which man protects himself from his insignificance by forcing his own interpretation of the world upon the entire universe, substituting a human-centred set of meanings that is reassuring to his vanity for a set of meanings that reduces him to being a mere transitory accident in the cosmic order" (ALG 111). This "metaphoric substitution is aberrant but no human self could come into being without this error.” “Making the language that denies the self into a centre rescues the self linguistically at the same time that it asserts its insignificance" (ALG 111f). “The usual scheme which derives truth from the convergence of self and other” (as invoked by Hirsch and Bleich, for instance) is "reversed" “by showing that the fiction of such a convergence is used to allow for the illusion of selfhood to originate."

        Still, de Man continues to accord literature a special place -- as discourse that, while it doesn't “escape” the “duplicity" in “everyday language” (BI 9), is at least conscious enough of it to draw characteristic consequences. "The claim of literary language to truth and generality" is “based on a duplicity within a self that wilfully creates confusion between literal and symbolic action in order to achieve self-transcendence as well as self-preservation" (BI 113). Literature “invents fictional subjects to create the illusion of the reality of others. But the fiction is not myth, for it knows and names itself as fiction. It is not a demystification, it is demystified from the start" (BI 18). "Literary history could in fact be paradigmatic for history in general, since man himself, like literature, can be defined as an entity capable of putting his own mode of being into question" (BI 165).

        De Man's portrayal of literature expediently aligns it with deconstruction, a move Culler diagnosed in Miller as well). In his magisterial fashion, de Man draws this equation for all language, though adducing a literary text each time, often by Nietzsche or Rousseau. One such “text” “establishes that deconstruction is not something we can decide to do or not to do at will; it is coextensive with any use of language, and this use is compulsive" (ALG 125). "The paradigm for all texts consists of a figure (or a system of figures) and its deconstruction" (ALC 205). "Allegories" of "unreadability" are also “allegories of the deconstruction and the reintroduction of metaphorical models" (ALG 257). “Semantic dissonance” is the “residue of meaning that remains beyond the text's own logic and compels the reader into an apparently endless process of deconstruction" (ALG 99). For de Man, “poetic writing is the most advanced and refined mode of deconstruction" (ALG 17).

        The same argument cannily justifies deconstructive “critical writing," from which the "poetic" “may differ" in the economy of its articulation, but not in kind” (ALG 17). If (as Schlegel avowed), “all “modern” literature" is "characterized" "by the ineluctable presence of a critical dimension," “critics can be granted the full authority of literary authorship” (BI 80). "Poetry is the foreknowledge of criticism," which “merely discloses poetry for what it is” (BI 31). Indeed, "the critical and the poetic components are so closely intertwined that is it impossible to touch the one without coming into contact with the other” (BI 80).

        Since (as de Man assured us), "they are not scientific, critical texts have to be read with the same awareness of ambivalence that is brought to the study of non-critical literary texts" (BI 110). Because “the criterion of literary specificity” depends on “the degree of consistent 'rhetoricity' of the language,” a “critical” “text” that “signifies its own rhetorical mode” is “not more or less literary than a poetic text,” albeit criticism proceeds "by means of statements” and poetry “avoids direct statement" (BI 136f).
         De Man's version of the Yale-school power play, whereby critics demand the rank of authors and poets, is restrained and uneasy compared to the arrant versions of Bloom and Hartman. For one thing, de Man's view of literature is not calculated to exalt (despite Culler's idea that it “celebrates" “great writings of the past,” OD 276). For instance, “literary" and “critical texts" share the "discrepancy between meaning and assertion" that “is a constitutive part of their logic" (BI 110). The two also share the “immanence” of “posing the problem” of "intelligibility in its own terms" (BI 107). Just as literature “becomes authentic when it discovers that the exalted status it claimed for its language was a myth," the critic can follow the same “intent at demystification that is more or less consciously present in the mind of the author” (BI 14). “Texts engender texts,” including criticism no doubt, “as a result of their necessarily aberrant semantic structure" (ALC 162). Such parallels with literature hardly endow criticism with the glory or monumentality Bloom and Hartman sometimes conjure.

         Moreover, deconstructive criticism is doomed to remain trapped in the same quandaries of language it purports to explicate (cf. Ch. 12). “Deconstructive readings can point out the unwarranted identifications achieved by substitution, but they are powerless to prevent their recurrence even in their own discourse, and to uncross, so to speak, the aberrant exchanges that have taken place”; “they leave a margin of error, a residue of logical tension that prevents the closure of the deconstructive discourse” (ALC 242). In fact, “deconstructions of figural texts engender lucid narrative which produce, in their turn” a “darkness more redoubtable than the error they dispel" (ALG 217).

          In consequence, we face “the discouraging prospect of an infinity of similar future confusions, all of them potentially catastrophic in their consequences” (ALC 10). “Nothing therefore prevents the deconstructive labour” “from starting all over again" (ALG 245). At best, we can proceed with a critical “rigour" which is “fully aware of the misleading power of tropes," and which “pursues its labours regardless of the consequences, the most rigorous gesture of all being that by which the writer severs himself from the intelligibility of his own text” (ALC 131, 207).

        De Man's model of the uneasy symbiosis between literature and criticism, or between text and commentary, bears as its label a programmatic oxymoron: "blindness and insight." This concept of “blindness,” “implying no value judgmerit," is “the necessary correlative of the rhetorical nature of literary language" (BI 141). “A penetrating but difficult insight” into that “nature” can “be gained only because the critics" are "in the grip" of their own "peculiar blindness," which may be “inextricably tied up with the act of writing itself” (BI 106). Possibly, some "characteristic of literary language causes blindness in those who come in close contact with it." Indeed, the “most enlightened" "writers" elicit a “particularly rich aberrant tradition" that forms “the basis, in fact, of literary history” (BI 141). "Since interpretation is nothing but the possibility of error, by claiming that a certain degree of blindness is part of the specificity of all literature we also reaffirm the dependency of the interpretation on the text and of the text on the interpretation"  -- a welcome affirmation for criticism.

        De Man is not implying that “what the critic says has no immanent connection with the work, that it is an arbitrary addition or subtraction, or that the gap between his statement and his meaning can be dismissed as mere error. The work can he used repeatedly to show where and how the critic diverged from it, but in the process of showing this our understanding of the work is modified and the faulty vision shown to be productive” (BI 109). “The discrepancy between the original and the critical text” is "given immanent exegetic power as the main source of understanding” (BI 110). The critic's “statement" shows "blindness", whereas his or her “meaning" provides “insight”; such is the "constitutive discrepancy in critical discourse.” The “interaction between critical blindness and critical insight" is “a necessity dictated and controlled by the very nature of all critical language" (BI 111). A “double movement of revelation and recoil will always” pervade “a genuine critical discourse" (BI 289). The critic is continually “forced to resort to paradoxical formulations, such as defining the modernity of a literary period as the manner in which it discovers the impossibility of being modern” (BI 144).

        Still, de Man allows for several variants of literary blindness according to one's point of view. If the author is “not-blinded" because he or she “accounts at all moments” for the text's “rhetorical mode," then the “blindness is transferred from the writer to his first readers"; the task of the subsequent "critical reader” is to "reverse the tradition and momentarily take us closer to the original insight" (BI 139, 141). But if “the literary text itself has areas of blindness,” then “reader and critic coincide in their attempt to make the unseen visible" (BI 141). Or, if the "literary text” is both "critical” and "blinded," then the critic's “reading” “tries to deconstruct this blindness" by "undoing, with some violence," what Derrida calls “the “orbit of significant misinterpretation” (BI 141, 116). These various arrangements give de Man substantial flexibility for manoeuvring his own function in relation to an author or critic. They also further blur the margin between literature and criticism by consigning both to a ward where visual diseases are abruptly contracted or cured.

        "Writing critically about critics thus becomes a way to reflect on the paradoxical effectiveness of a blinded vision that has to be rectified by means of the insights it unwittingly provides" (BI 106). In several of well-known essays, de Man “reflects” in just this fashion. He develops special techniques to preserve the positive achievements of criticism not in spite of, but because of his rejection of its declared principles and results. His negativity affirms while it denies, and rescues underlying truths from behind published errors. This procedure parallels the deconstruction that conserves its undermined counterpositions (cf. Ch. 13).

        De Man examines a range of critics whose "insight seems” “to have been gained from a negative movement that animates the critic's thought, an unstated principle that leads his language away from his asserted stand, perverting and dissolving his stated commitment to the point where it becomes emptied of substance, as if the very possibility of assertion had been put into question" (BI 103). “The critics’ "language could grope toward a certain degree of insight only because their method remained oblivious to the perception of this insight” (BI 106). If the critic is "by definition incompetent to ask" “the question of his own blindness,” then "the insight exists only for a reader in the privileged position of being able to observe the blindness as a phenomenon in its own right,” and “so being able to distinguish between statement and meaning.” This “privilege” is de Man's own claim to authority, now that he has argued away the right to determine a correct or original reading for a text.²⁰  [20. However, he still uses the term “correct” for what matches his own beliefs (ALC 135, 166, 202, 229; BI 181, 288). Views he doesn't hold are “wrong" (BI 112) or in “error" (BI 219, 226). “A mistake” is not the same as the "blind spot” every writer must have (BI 139).]

       A case in point is his treatment of the New Critics, whom he proposed, "before the advent of Structuralism on the literary scene,” "to introduce" “to French readers” while warning against "the dead-end of formalist criticism" (BI 229).²¹ [21. The piece is 'also an attack on the early Barthes, whose views are said to match the New Critics’ (BI 231f, 234, 240f) and whose proclamations of scientific method must have galled de Man.] The New Critics practised "reassuring criticism" by "suggesting” a "balanced and stable moral climate," and "refused to grant aesthetic experience any difference from other human experiences” (BI 234f). But despite their idea of a “perfect continuity between the sign and the thing signified," they collided with the "fundamental ambiguity” “constitutive of all poetry” and with an “infinity of valid readings" -- the ultimate cause being “the deep division of Being itself” (BI 232, 236f). The New Critics clung to "a reified notion of a literary text as an objective thing" and succeeded in "describing" “literary language as a language of irony and ambiguity” only by making “the concept of form" “function in a radically ambivalent manner, both as a creator and undoer of organic totalities" (BI 104). Thus, their "final insight" "annihilated the premises that led up to it."

        In his critique of critiques, de Man seems to defer to the lord high executioner of western metaphysics, Jacques Derrida. The latter is saluted for applying “the rigor and intellectual integrity of a philosopher whose main concern is not with literary texts to restore the complexities of reading to the dignity of a philosophical question" (BI 110). This accomplishment goes beyond that of “critics like Blanchot and Poulet, who make use of the categories of philosophical reflection" but "erase the moment of actual interpretive reading, as if the outcome of this reading could be taken for granted in any literate audience." In Derrida's work, the “interaction between critical blindness and critical insight” appears "no longer in the guise of a semiconscious duplicity, but as a necessity dictated and controlled by the very nature of all critical language" (BI 111). “The discrepancy implicitly present in the other critics here becomes the explicit centre of the reflection." Hence, “Derrida's work is one of the places where the future possibility of literary criticism is being decided, although he is not a literary critic in the professional sense.”

        Yet even Derrida gets diagnosed as blind regarding Rousseau, who plays a central role in De la Grammatologie (1967). De Man “claims Rousseau to be" a “non-blinded author" that Derrida, in "blindness," “misreads" as a blinded one "for the sake of his own exposition and rhetoric" (BI 141, 139). The “misreading" was “deliberate": “insights that could have been gained from the “real” Rousseau" are deployed to “deconstruct a pseudo-Rousseau” fabricated by "the established tradition of Rousseau interpretation" (BI 139f). Derrida does not find in Rousseau an author who "fell back into confusion, had faith, or withdrawal"²² [22. Even the “sympathetic and penetrating" critic Jean Starobinski (1961) tended to "reduce Rousseau from the status of a philosopher to that of an interesting psychological case” (BI 113).], but places him in a “tradition that defines Western thought in its entirety: the conception of all negativity (non-being) as absence and hence the possibility” of a “reappropriation of being (in the form of truth, of authenticity, of nature, etc.) as presence" (BI 112, 114).

        De Man retorts that in Rousseau, “a vocabulary of substance and of presence is no longer being used declaratively but rhetorically”; “Derrida misconstrues as blindness what is instead a transposition from the literal to the figural level of discourse" (BI 138f). “Rousseau escapes from the logocentric fallacy" of “favoling voice over writing" “to the extent that his language is literary," because “literature” “demystifies" “the priority of oral language over written" (BI 137f). “Derrida" is “unwilling or unable to read Rousseau as literature” and “reproaches" him “for doing exactly what he legitimately does himself” -- the “established philosophical procedure" of using “a philosophical terminology with the avowed purpose of discrediting this very terminology” (BI 137ft).

        The disputed source-work is Rousseau's disquisition on the origin of language, whose thesis that figural usage precedes the literal is shared by de Man, as we have seen (see also ALC 149-54). Derrida, however, states that Rousseau's “theory of metaphor is founded on the priority of the literal over the metaphorical meaning," so that the text reveals “a moment of blindness in which Rousseau says the opposite of what he means to say” (BI 133). “Rousseau” "relinquishes" "literal meaning” for "the designation of objects” but retains it for an “interiorized" "object,” namely “a state of consciousness, a feeling or a passion" (cf. Derrida 1967: 389) (BI 133). This interpretation empowers Derrida to place Rousseau within “Western thought" at “the moment when the postulate of presence is taken out of the external world and transposed within the self-reflexive inwardness of a consciousness" (BI 134). De Man, in contrast, views Rousseau as already deconstructive, provided we read the philosophical works with the same rhetorical awareness as the literary works (cf. also ALC 159, 220, 247).

        Derrida expends “a considerable and original interpretative effort that has to move well beyond and even against the face-value of Rousseau's own statement" in order to "demonstrate the strict orthodoxy of Rousseau's position with regard to the traditional ontology of Western thought" (BI 122f). “Representation is conceived as imitation in the classical sense of eighteenth-century aesthetic theory” and thus "confirms rather than undermines the plenitude of the represented entity” (BI 123). De Man disagrees: “Rousseau's theory of representation is not directed toward meaning as presence and plenitude but toward meaning as void"; “the sign is devoid of substance, not because it has to be a transparent indicator that should not mask a plenitude of meaning, but because the meaning itself is empty" (BI 127).

         The whole dispute may not seem terribly vital, since both critics agree on the priority of figurality, whether or not Rousseau is counted an ally. Moreover, de Man is not passing a negative judgment on Derrida, but merely gathering fresh evidence that “blindness" is “the necessary correlative of the rhetorical nature of literary language" (BI 141). Still, the debate intriguingly reveals how the authority of one's own statement must be founded by dismantling that of someone else's (an idea Bloom carries further). Having abandoned, at least in principle, the search for the real meaning of texts, de Man claims instead the right to decide what constitutes "blindness" or “insight," and thus the right to support an argument with a source that either disagrees or agrees with him. Derrida prepares his counterposition by making Rousseau's text say the opposite of what it seems to; de Man counters Derrida's text and restores the statement of Rousseau's. We might have saved a deal of trouble by taking Rousseau's word for it from the start. But the deconstructionist mind needs “a sparring partner,” “an antithetical mask or shadow" to “gain its momentum" (cf. BI 140) –[ even if the partner is just another such mind.

        In a similar blend of assertion and negation, de Man envisions a “rhetoric of reading reaching beyond the canonical principles of literary history" but still using them “as the starting point of their own displacement” (ALC ix). De Man heretically vows that “considerations of the actual and historical existence of writers are a waste of time from a critical viewpoint” (BI 35). “What we usually call literary history has little or nothing to do with literature"; “literary interpretation” “is in fact literary history” (BI 165).

        Like that of the subject, the author's role is marginalized in de Man's view of “literature," which “originates in the void that separates intent from reality” (BI 34). Rousseau,” for instance, attained “efficacy" for his “text” by making it "a version” of a “rhetorical model" of “language over which he had no control” (ALC 277). Or, Proust tried to celebrate the self-willed and autonomous inventiveness of a subject" and yet produced “images relying" on “semi-automatic grammatical patterns" (ALC 16).

        Within such an outlook, the question of whether “the themes" of a work are “the expression” of the author's “own lived experience" may seem “irrelevant" (ALC 50). But de Man does introduce authorial history when it supports a reading, as in: “the situation of the scene" “corresponds to the actual predicament of Rousseau at that time"; or the “extension of meaning” is “consistent with the thematic concerns of Mallarmé's other works of the same period" (ALG 175; BI 179).

         Predictably though, De Man is inconsistent about the consistency of the author. On the one hand, the fact that “Rilke says just about exactly the opposite in a prose text" "hardly invalidates our reading” of a poem (ALG 42). On the other hand, “the unpublished fragments, contemporaneous with the main text" by Nietzsche “reduce” it to “an extended rhetorical fiction devoid of authority" (ALG 101). We see here more illustrations of my reservation about Hirsch's validation procedure: the critic subjectively decides what to do with authorial evidence.

        De Man's thinking rejects the traditional "historiography” based on “an organically determined view of literary history" (ALC 80). “The critical “deconstruction” of the organic model'²³ [23. In Nietzsche's “terminology,” the “organic” approach is called “monumental” (ALC 81).] "creates radical discontinuities and disrupts the linearity of the temporal process to such extent that no sequence of actual events" ”could ever acquire, by itself, full historical meaning” (ALG 81). As we'd expect, de Man avers that “society and government derive from a tension between man and his language" (ALG 156). The “political nature” of “all forms of human language" is to be construed “not in the representational, psychological, or ethical sense," "but rather in terms of the relationship, within the rhetorical model, between the referential and the figural semantic fields" (ALC 156f).

        We might surmise that for de Man, history is essentially the history of rhetoric. In “the classical sense of eighteenth-century aesthetic theory," “representation is conceived as imitation” and “functions as a memnotechnic sign that brings back something that happened not to be there at the moment, but whose existence in another place, at another time, or in a different mode of consciousness is not challenged" (BI 123). Accordingly, “classical” “theories" "repeatedly state as the main function of art" “the possibility of making the invisible visible, of giving presence to what can only be imagined,” and thereby lending “the ontological stability of perceived objects" to “what lies beyond the senses" (BI 124). “The mimetic imagination is able to convert non-sensory, “inward” patterns of experience (feelings, emotions, passions) into objects of perception and can therefore represent as actual concrete presences, experiences of consciousness devoid of objective existence” (BI 123f).

         In “romantic” “texts," on the other hand, “consciousness does not result from the absence of something, but consists of the presence of a nothingness; poetic language names this void with ever-renewed understanding” (BI 18) By asserting that "this persistent naming is what we call literature," de Man binds literature much closer to romanticism than to classicism.

        Like Hayden White and the early Culler, de Man explores literary history in terms of the evolution of dominant rhetorical tropes.²⁴ [24. De Man vacillates between the terms “modes” and “tropes” in the early paper (e.g. RT, 207, 192). White’s scheme of tropes is summarized in Note 14 to Ch. 13. Culler has since repudiated his own scheme (Ch. 13).] To propose his own version, he naturally has to devise “a historical scheme that differs entirely from the customary picture” and "demystifies" the established “terms” (BI 208, 211). He accuses prior critics of “associating” "rhetorical terms with value judgments that blur distinctions and hide the real structures" (BI 188). He now calls for “a more systematic treatment of an intentional rhetoric.”

           “The entire historical and philosophic pattern changes a great deal" if one assumes that “the dialectic between subject and object does not designate the main romantic experience, but only one passing moment in a dialectic, and a negative moment at that" (BI 204f cf. BI 208) — in defiance of a long tradition ²⁵ [25 See also BI 208. Failure to achieve a synthesis of “subject and object” may be an advantage, linked to “the most original and profound moments in the works, when an authentic voice becomes audible” (BI 204f). Compare de Man's high esteem for Blanchot (ranked above the entire “nouveau roman") in this regard (BI 64). On the utopian projection of the synthesis in Hegelian and Marxist thought, see Jameson (MF 38, 44, 141, 146) and Iser's more hopeful projection (Note 11). On the unity of "mind and nature” compare Wasserman (1964), Abrams (1965), and Dieckmann (1966) (RT 181f, 184).] De Man mistrusts “conceptions of metaphor" as "a dialectic between object and subject, in which the experience of the object takes on the form of a perception or a sensation”; and of the "symbol as a unit of language in which the subject-object synthesis can take place” (BI 193, 199).

         De Man's view of language being irremediably divided, he devalues the “symbol," whose “superiority" "over allegory” was “asserted” by Coleridge and Goethe (BI 191, 189). “In the world of the symbol, life and form are identical" for Coleridge; "its structure is that of the synecdoche, for the symbol is always a part of the totality that it represents" (BI 191). “Consequently, in the symbolic imagination, no disjunction of the constitutive faculties takes place, since the material perception and the symbolical imagination are continuous, as the part is continuous with the whole." Eventually, “the supremacy of the symbol, conceived as an expression of unity between the representative and the semantic function of language, becomes a commonplace that underlies literary taste," “criticism,” and “history" (BI 189).

        De Man traces this outlook to his elder contemporary, Hans-Georg Gadamer. In Wahrheit und Methode, “Gadamer makes the valorization of symbol" “coincide with the growth of an aesthetics that refuses to distinguish between experience and the representation of this experience” (BI 188). “The poetic language of genius is capable of transcending this distinction and can thus transform all individual experience directly into general truth." “The subjectivity of experience is preserved when it is translated into language; the world is then no longer seen as a configuration of entities that designate a plurality of distinct and isolated meanings, but as a configuration of symbols ultimately leading to a total, single, and universal meaning." Hence, "the symbol is founded on an intimate unity between the image that rises up before the senses, and the supersensory totality that the image suggests” (BI 189).

        All these conceptions don't suit de Man's temperament. He brushes aside “the asserted superiority of the symbol" as a "tenacious self-mystification” (BI 208). He swears that the "symbolical style will never be allowed to exist in serenity" or "to gain an entirely good poetic conscience." At best, the symbol is "a special case" that "can lay no claim to historical or philosophical priority over other figures" (BI 191). At worst, it is a “mystified form of language” and a “pseudo-synchronic structure” that “misleads one into believing in a stability of meaning that does not exist" (BI 132f).

         De Man elects to champion “allegory," a mode at whose expense the symbol had been elevated (cf. BI 188ff, 208). For Coleridge, “the allegorical form appears purely mechanical, an abstraction whose original meaning is even more devoid of substance than its “phantom proxy,” the allegorical representative" (BI 191f). For Gadamer (1960: 70), "allegory” is like "non-art, in that'" it "'has run its full course” "as soon as its meaning is reached" (BI 188f). Allegory was reduced to “a sign that refers to one specific meaning and thus exhausts its suggestive potentialities once it has been deciphered" (BI 188). It “appears as the product of the age of Enlightenment" and is “reproached" for being "dryly rational and dogmatic in its reference to a meaning that it does not itself constitute" (BI 189). Even in "recent French and English studies of the romantic and post-romantic eras," "allegory is frequently considered an anachronism and dismissed as non-poetic” (RT 175).

          De Man's counterargument invokes "the rhetoric of temporality." He "locates” “the dialectical relationship between subject and object” "entirely in the temporal relationships that exist within a system of allegorical signs. It becomes a conflict between a conception of the self seen in its authentically temporal predicament and a defensive strategy that tries to hide from this negative self-knowledge” (BI 208). Whereas "nature” can reveal “endurance within a pattern of change” or “a rneta-temporal stationary state beyond the apparent decay of a mutability" that “leaves the core intact," the “self” is “caught up entirely within mutability” (BI 196f). "The temptation exists, then, for the self to borrow" “the temporal stability that it lacks from nature, and to devise strategies by means of which nature is brought down to a human level" (BI 197). This “borrowing" could maintain the “illusionary priority of a subject" (BI 200).

         The “allegory" offers a better resolution for the predicament, because in its world," "time is the originary constitutive category" (BI 207). "The prevalence of allegory always corresponds to the unveiling of an authentically temporal destiny. This unveiling takes place in a subject that has sought refuge against the impact of time in a natural world to which, in truth, it bears no resemblance" (BI 206). The "relationship between signs necessarily contains a constitutive temporal element": “the allegorical sign” must "refer to another sign that precedes it" (BI 207). The "essence of this previous sign" is to be "pure anteriority," apparently because the sign was nowhere inscribed. “The meaning constituted by the allegorical sign can then consist only in the repetition.” Accordingly, “allegory designates primarily a distance in relation to its own origin, and renouncing the nostalgia and the desire to coincide, establishes its language in the void of this temporal difference.” "Allegory” is thereby able to “suggest a disjunction between the way in which the world appears in reality and the way it appears in language” (BI 191).

         This account is symptomatic for the whole rhetorical movement focusing on intertextuality, the way any text refers to others. This phenomenon could be described as “temporal" in the sense that within the entire language system, the use of any sign other than a utter neologism presupposes its prior uses (in Derridean diction, it bears their "traces”). On the other hand, these uses are also believed to coincide at least enough to guarantee that people know what the sign should mean; and the exact timing of those uses is often neither known nor relevant. Allegory is apparently portrayed by de Man as a text which coincides with some uninscribed prior text it declines to declare, so that -- in contrast to the traditional view de Man rebukes -- its meaning cannot be exhausted.

        On close inspection, criticism also might become an allegory by dissolving the prior text to which it refers. An “allegory of reading" might have “universal significance” if it could be a “report of the contradictory interference of truth and error in the process of understanding” (ALC 72). His own writings might be considered an endless attempt to provide just such a report. He insistently reflects on the “allegorical” discourse that “repeats" a “potential confusion between figural and referential statement” or raises the prospect of "the impossibility of reading” (ALC 116, 205). The whole "blindness and insight” dynamics also rests on the irredeemable fusion of error and truth. The danger is that de Man's general critique of language might allow the allegory to gradually absorb all discourse, or at least all literature. Some such tendency might indeed be detected in his development between the “temporality" essay (1969) and his Allegories of Reading (1979), whose title is instructive in this regard.

        Temporality also undergoes a related expansion that threatens to absorb all discourse. Due to “the structural characteristics of language," “the visual perception which creates a false illusion of presence has to be replaced by a succession of discontinuous moments that create the fiction of a repetitive temporality” (BI 131f). "All sequential language is a dramatic, narrative language,” because “language is a diachronic system of relationships." Correspondingly, “literature is an entity that exists not as a single moment of self-denial, but as a plurality of moments that can, if one wishes, be represented -- but this is a mere representation -- as a succession of moments or a duration" (BI 159). "Literature" "is, in essence, the fictional narration of this movement." Hence, “chronology, is the structural correlative of the necessarily figural nature of literary language" (BI 133).

          If de Man's reasoning tends to make allegory swallow everything, much the same happens with irony. He decides that “literary history" is "the dialectical play" between “allegory" and “irony," as well as “their common interplay with mystified forms of language (such as symbolic or mimetic representation), which it is not in their power to eradicate” (BI 226). "Irony and allegory" have "a shared structure" wherein “the relationship between sign and meaning is discontinuous, involving an extraneous principle," but this "aspect may well be a description of figural language in general" (BI 209). The two are further “linked in their common demystification of an organic world postulated in a symbolic mode of analogical correspondences or in a mimetic mode of representation in which fiction and reality could coincide" (BI 222). “The two modes" are "two faces of the same fundamental experience of time," that is, "determined by an authentic experience of temporality which, seen from the point of view of the self engaged in the world, is a negative one" (BI 226).

        Still, the two may differ somewhat in their temporality. Irony is more likely to “appear as an instantaneous process" occurring in "the instant at which the two selves, the empirical as well as the ironic, are simultaneously present" – “two irreconcilable and disjointed beings" (BI 225f).²⁶ [26. The “symbol,” which de Man devalues, is also given “simultaneity" (BI 207). But this exception from the pervasive temporality of literature is not clear to me.] Thus, "irony is a synchronic structure, while allegory” (whose “temporality" I outlined above) “appears as a successive mode capable of engendering duration as the illusion of a continuity that it knows to be illusionary" (BI 226). Irony is "the mode of the present" and "knows neither memory nor prefigurative duration, whereas allegory exists entirely in an ideal time that is never here and now but always a past or an endless future.” Its link to the "present" rather than to “an ideal time" makes “irony come closer to the pattern of factual experience and recapture some of the factitiousness of human experience as a succession of isolated moments lived by a divided self.”

        This referral back to the "divided self" we might well have foreseen. De Man's world is distinctly schizophrenic, with language being both cause and effect, plus model and expression. Hence, we are not surprised when de Man relates “irony" to “a problem that exists within the self” (BI 211; cf. BI 212, 214, 219, 226). The self undergoes “duplication” or “multiplication,” so that “irony" is “a relationship, within consciousness, between two selves" (BI 212).²⁷ [27. However, this division is not "intersubjective," because one self cannot be “superior” to the other, for instance, in wanting to exercise “power" or "violence" over the other, or to “educate and improve" it (BI 212; cf BI 64). But psychoanalysis, whether of Freud, Horney, and Maslow (and common sense, by the way) agrees that one part of the self is precisely seeking power over another.] "The writer” “constitutes by his language” an “ironic, two-fold self,” but “only at the expense of his empirical self, falling (or rising) from a stage of mystified adjustment into the knowledge of his mystification” (BI 214). Thus, the loss of experienced immediacy is accompanied by a gain in self-awareness. “The ironic language splits the subject into an empirical self that exists in a state of inauthenticity and a self that exists only in the form of a language that asserts the knowledge of this inauthenticity. " This account agrees with de Man's notion of literature as language that is at least aware of the “separation from empirical reality" (BI 17).

        Once stated in such fundamental terms, "irony possesses an inherent tendency to gather momentum and not to stop until it has run its full course; from the small and apparently innocuous exposure of a small self-deception it soon reaches the dimensions of the absolute” (BI 215). De Man's theorizing correspondingly progressed from a concept of irony as a “trope” (RT 192) to the point where “irony is no longer a trope but the undoing of the deconstructive allegory of all tropological conditions, the systematic undoing, in other words, of understanding" (ALC 301). “Irony” "dissolves in the narrowing spiral of a linguistic sign that becomes more and more remote from its meaning, and it can find no escape" (BI 222). "It can know” “inauthenticity, but can never overcome it; it can only restate and repeat it on an increasingly conscious level” without ever “making its knowledge applicable to the empirical world.” "All true irony" "has to engender” "irony to the second power or “irony of irony,” which “asserts and maintains its fictional character by stating the continued impossibility of reconciling the world of fiction with the actual world” (BI 218). “The author” “asserts" "the ironic necessity of not becoming the dupe of his own irony and discovers that there is no way back from his fictional self to his actual self” (BI 219).

        De Man thereby enters Schlegel's vision: “the dialectic of self-destruction and self-invention” which “characterizes the ironic mind is an endless process that leads to no synthesis” (BI 220). "Irony engenders a temporal sequence of acts of consciousness which is endless"; “irony is not temporary but repetitive, the recurrence of a self-escalating act of consciousness” (RT 202). The author “permanently” intrudes into the work in order to “disrupt the fictional illusion" and "prevent the all too readily mystified reader from confusing fact and fiction and from forgetting the essential negativity of the fiction” -- a strategy termed “parabasis" (BI 218f, 228; RT 200, 209; ALC 300f) (after the portion of Greek comedy sung to the audience as a direct message from the author). Besides literature, “all philosophical discourse" has “ironic allegory” for its "rhetorical mode” (ALC 116).

        Like other aspects of deconstruction, the proliferation of irony is by no means "harmless" or "reassuring" (BI 214f). True, a certain “freedom" is gained from “the unwillingness of the mind to accept any stage in its progression as definitive" (BI 220).²⁸ [28. Barthes (1970) “liberating theory of the significance in S/Z is said to “imply a complete drying up of thematic possibilities,” such as would occur in "pure poetry” wherein the “emblematic object is revealed to be a figure without the need of any discourse" (ALC 48). De Man seems to attack Barthes despite the similarities of their views on "free play” (cf. Note 21).] But we also approach a “dizziness to the point of madness"; “sanity can exist only because we are willing to function within the conventions of duplicity and dissimulation” (BI 215f). “Once this mask is shown to be a mask, the authentic being underneath appears necessarily as on the verge of madness” (BI 216) Hence, “absolute irony is a consciousness of madness, itself the end of all consciousness; it is a consciousness of a non-consciousness." Using “the double structure of ironic language,” "the ironist invents a form of himself that is “mad” but that does not know its own madness; he then proceeds to reflect on his madness thus objectified.” De Man's personal touch is his serene acceptance of this dizzying irony that Bloom and Hartman resolve to combat by restoring the power of the Word (Ch. 15-16).

          When he gives a critical reading of a literary work, de Man's rhetorical moves are rendered consistent and predictable by his theoretical program. “One has the feeling beforehand that one knows what de Man will do, yet one still is awed by the elegance, precision, and economy of his performance” (Godzich, preface to BI, xvf). Wherever he looks, de Man finds his own ideas once more, and duly extracts his conceptions of language, literature, and the self as the eventual consequence of every text. In this way, his “readings” are indeed “allegories" not of reading in general, as he likes to insist, but of a style of reading so specialized Godzich admits it could not be “duplicated” by anyone else.

        For example, de Man's reading of The Triumph of Life by Shelley keeps broadening out into "a more general interpretation" (SD 62). The poem “warns us that nothing, whether deed, word, thought, or text, ever happens in relation, positive or negative, to anything that precedes, follows, or exists elsewhere, but only as a random event whose power, like the power of death, is due to the randomness of its occurrence" (SD 69).

        As usual, the evasiveness of language gets top billing. The poem indicates that “the latent polarity implied in all classical theories of the sign allows for the relative independence of the signifier and for its free play in relation to its signifying function" (SD 60). “Particularly meaningful movements" can be “generated by random and superficial properties of the signifier rather than by the constraints of meaning. " A “figure" of language is “constituted” by “the alignment of a signification with any principle of linguistic articulation whatsoever” (SD 61). A similar process marks “the passage from tropological models such as metaphor, synecdoche, metalepsis, or prosopopoeia (in which a phenomenal element, spatial or temporal, is necessarily involved) to tropes such as grammar and syntax (which function on the level of the letter without the intervention of an iconic factor)” (SD 61f).

        The tactics are the same when the poet's use of the term “shape" is "identified as the model of figuration in general" and as “a figure for the figurality of all signification” (SD 61f). “The figure is not naturally given or produced," but “posited by an arbitrary act of language”; and “the positing power of language is both entirely arbitrary, in having a strength that cannot be reduced to necessity, and entirely inexorable in that there is no alternative to it” (SD 62f). "Language posits and language means (since it articulates) but language cannot posit meaning; it can only reiterate (or reflect) in its reconfirmed falsehood” (SD 64).

      “Figuration” gets revenge by “performing the erasure of the positing power of language” and thus becomes "disfiguration,” being “the repetitive erasures by which language performs the erasure of its own position" (SD 64f). The “shape" is then “the figure of thought, but also a figure" of "the element in thought that destroys thought in an attempt to forget its duplicity" (SC 65). "The entire scene of the shape's apparition and subsequent wandering is structured as a near-miraculous suspension between these two different forces whose interaction gives to the figure the hovering motion which may well be the mode of being of all figures” (SD 55). The poem's own shape as a fragment (Shelley never finished The Triumph of Life) points to another “disfiguration" “the reader” has to "reinscribe"; yet this “challenge," this “fracture" too is deemed "present in all texts" (SD 67). “To read is to understand, to question, to know, to forget, to erase, to deface, to repeat" (SD 68).

        Shelley's recurrent imagery of light is construed to "create conditions of optical confusion that resemble nothing so much as the experience of trying to read The Triumph of Life, as its meaning glimmers, hovers, and wavers, but refuses to yield the clarity it keeps announcing” (SD 52f). "Light” is made to stand for “metaphor”: it is “the bearer of light which carries over the light of the senses and of cognition from events and entities to their meaning," but it "irrevocably loses the contour of its own face or shape” (SD 66). The “rainbow" represents the very possibility of cognition, even for processes of articulation so elementary that it would be impossible to conceive of any principle of organization, however primitive, that would not be entirely dependent on its power” (SD 57). “The figure of the rainbow is the figure of the unity of perception and cognition, undisturbed by the possibly disruptive mediation of its own figuration” (SD 58).

        Shelley's use of mirror imagery is also interpreted in tune with de Man's standard theses. "The scene is self-reflexive: the closure of the shape's contours is brought about by self-duplication” (SD 55), “The light generates its own shape by means of a mirror, a surface that articulates it without setting up a clear separation that differentiates inside from outside.” Thus, "the specular structure of the scene" is "merely an illustration of a plural structure that involves natural entities only as principles of articulation among others" (SD 62). Comparably, “the underlying assumption of a paraphrastic reading" is “of specular understanding in which the text serves as a mirror of our own knowledge and our knowledge mirrors in its turn the text's signification” (SD 58).

        These interpretations illustrate the critical moves whereby de Man allegorises the particulars of the literary work into his own theoretical precepts. In effect, the poem provides a commentary on de Man's interpretive procedures, rather than the other way around -- a tactic that hints why he might be anxious to equate criticism with literature. The openness of literature is paradoxically narrowed by de Man's obsession with a small group of consistent, interlocking conceptions about that openness. His entrenched blindness to unifying or stabilizing forces generates an unintended insight into the power of those forces. He looks for “randomness," “free play," and “disfiguration," but finds that “Shelley's imagery, often assumed to be incoherent and erratic, is instead extraordinarily systematic" (SD 69, 60, 57).²⁹ [29. Examples might be the “consistent system of sun imagery” or the “system of relationships" into which the “river” “enters” (SD 51, 53). Blanchot is described as having an “almost obsessive preoccupation with a few fundamental concerns” (BI 61) -- not a bad comment on de Man too.]

        We encounter the same self-deconstruction on the more general level of theory. De Man consistently denounces “totalizations from part to whole" as "illegitimate," "aberrant,” “fallacious," and the like (ALC 256, 129, 237, 249; BI 32, 34). Yet he can scarcely make a statement about a text without generalizing across the whole of language or literature, witness his insistent use of “all," “always,” or “any”.³⁰ [30 Thus, note the uses of “all" in ALC 50, 65, 111, 113, 160, 192, 202, 239, 245, 264, 290; “always” in ALC 152, 199, 205, 265;, and “any” in LC 76, 125f, 141, 182, 187, 226.] He revels in epigrammatic or apodictic pronouncements about “all language," “all readings,” “all writing,” "all metaphors," "all literature,” and so on (ALG Ill, 202, 290, 65, 50). These are too broad to be subject to proof, and de Man rarely offers any. He merely proves that he -- and perhaps he alone -- could make the same point with any text he pleased. Even Culler, who makes de Man the star practitioner for deconstruction, is troubled by the annoying” “strategy of omitting crucial demonstrations" (PS 16; OD 229).

        If deconstruction is something we cannot escape because it is built into language and literature, we might wonder why authors and critics didn't resolve to perform it centuries ago. De Man might say they were doing it without knowing it, blinded by a certain metaphysics; the modern practice of deconstruction then appears as an apocalyptic fulfillment of the history of ideas. We might also wonder why de Man, like Derrida, Miller, Hartman, and Bloom, is so selective about the works he reads, ones that seem ready-made for his project. (Rousseau, for example, seems to have come back into fashion in exactly this manner.) Whatever his selective criteria -- style, technique, authorial intention, or choice of topics and images -- they limit de Man's absolute avowals of inescapable figurality. We return to the problem latent in Holland's work: the skilled critic can read the way he or she says everybody does, but the ingenious and surprising results signal the exceptional nature of the performance.

        Both de Man and the recent Culler declare that everything is pretty well controlled by language. Among the subjects who dissolve beneath this control are the author and the reader. Deconstructionists did not intend merely to reanimate the New Critical polemic against the “intentional" and “affective” fallacies (cf. BI 24f). De Man (and Bloom even more) simply takes over from both author and reader all decisions about what should be done with the text. He assures us his proceedings are “irnmanent” or “intrinsic” (cf. BI 107f, 109f; ALC 4f). But the closeness of his criticism to the literary text is rather like that of antimatter to matter: precisely reflected, but violently explosive upon contact. “The further the critical text penetrates in its understanding, the more violent the conflict becomes" with “the original text" (BI 109). ³¹ [31. In Rousseau, “all examples for the ‘natural’ language of man are acts of violence” (ALC 140). In Shelley, an image is found for a “violent act of power achieved by the positional power of language” (SD 62; cf. SD 64). Compare the violence of de Man's own imagery with its Freudian overtones: “writing always includes the moment of dispossession in favour of the arbitrary power play of the signifier and from the point of view of the subject, this can only be experienced as a dismemberment, a beheading or a castration”; or "the representation of copulation or murder are the most effective emblems for the moment of literal significance that is part of any system of tropes” (ALC 296, 182). Bloom and Hartman share a similar tendency (cf. Note 27 to Ch. 14; Note 24 to Ch. 15; Note 21 to Ch. 19) – as if in a violent gang of “Yale’s Angels”.]

       Still, if de Man sees the self as divided, he has a right to be an example. Why shouldn't he exemplify his idea that “the category of the self” is so “double-faced that it compels the critic who uses it to retract implicitly what he affirms and to end up by offering the mystery of this paradoxical movement as his main insight" (BI 105)? And if he can diagnose Lukács, Poulet, and Blanchot to be writers whose "critical stance" “is defeated by their own critical results” (BI 106), might we not make a like diagnosis of him? Yet he designed the fail-safe mechanism of building this dualism directly into his theory, so that in the very moment we detect him deconstructing himself we find his theory confirmed. He cannot be wrong on one level without being right on another.

        Even after he denied that language can convey truth in the usual sense, de Man apparently could not overcome his nostalgia for some form of truth. His compromise was evidently as uneasy as it was complex: to seek new insights by deconstructing rich sources. Only in this circuitous way could he reconcile his own understanding with his theses about the universality of misunderstanding. The prospect that this procedure might imply his exclusion from the universe may have driven him steadily further into labyrinths of allegory and irony where the illusion of his own selfhood was, if not secured, at least preserved in the way deconstruction reweaves what it unravels. “Curiously,” “only in describing a mode of language which does not mean what it says" can one "actually say what one means” (RT 194).

        De Man certainly lived up to his own declaration that “literary criticism, in our century, has contributed to establishing” the “crucial difference between an empirical and an ontological self, in that respect, it participates in some of the most audacious and advanced forms of contemporary thought” (BI 50). His mistrust of metaphysics left him among those whose “ontological confidence has forever been shaken” (cf. ALG 123). If, in Nietzsche's words, “we have to cease to think if we refuse to do it in the prisonhouse of language” (PL i), de Man elected to dwell there steadfastly, not planning an escape, hardly even looking out the windows, but exploring and refining the inner layout as a space for constructing and dismantling a microcosm of the world outside. His tools, so far most rigorously employed (like Derrida's) in “the unmaking of a construct” (BI 140), may be turned to new uses by those who come after him. “However negative it may sound, deconstruction implies the possibility of rebuilding.”

Click here to go to Critical Discourse Main Page