"At the Far End of this Ongoing Enterprise..."

The master is destined, then, not to smooth out the field of relations but to upset it, not to facilitate the paths of knowledge,
but above all to render them not only more difficult, but truly impracticable.
—Maurice Blanchot, The Infinite Conversation

1. Paul de Man's brief introduction to the 1979 issue of Studies in Romanticism devoted to the "Rhetoric of Romanticism" might be understood as his most explicit treatment of the question of legacy. The introduction is a strange and often contradictory text in which de Man provides an historico-fictional account of his own "generation"—understood synchronically and diachronically, both as a group of individuals and as an act of genesis. At the same time, by editing a volume of work by his students, de Man here introduces a "generation," one that already in 1979 is understood to be his issue.

2. De Man initially intended this special issue on "The Rhetoric of Romanticism"—which shares its title (in advance) with his own posthumous collection of essays on Wordsworth, Kleist, Shelley, Hölderlin, and Yeats—to be an outlet for work emerging from his 1977-78 NEH seminar. Yet, the final table of contents includes only two essays—Stephen J. Spector's "Thomas de Quincey: Self-Effacing Autobiographer" and William Ray's "Suspended in the Mirror: Language and the Self in Kleist's 'Über das Marionettentheater'"—that originated in the NEH seminar.¹ The remaining essays, by Cynthia Chase, Barbara Johnson, Timothy Bahti, and E.S. Burt, developed from de Man's regular graduate courses at Yale. In some obvious sense, then, de Man's introduction responds to the question of the relation between the work introduced and his instruction of its authors. In other words, the essay responds to the implicit question—phrased in the idiom of Kleist's "Marionettentheater," the occasion of de Man's most sustained discussion of pedagogy—of whether or not, in pulling a few strings to gather these essays and get them published, he also "pulls their strings."² At the same time, the introduction raises other questions—about the relation between scholarship and pedagogy more generally, about literary history, inheritance, and freedom. Do de Man's voice and authority alone give these essays their motion and their force? Are these essays, like an automaton, moved by an external generator, rather than by their own spontaneous or automotive energy? And, if so, is this external generation also the source of their extraordinary elegance? Is it the authoritative teacher himself who allows for their "light touch," that is, for the very grace that he will claim distinguishes their work from his and exposes the "awkward" and "lopsided" efforts of his generation, the very grace that he goes so far as to suggest also means his death?³ What is de Man's relation to his students' virtuosic displays of rigorous reading, that is, what is his relation to this putative legacy? What is his relation to it as legacy, and what, if anything, does de Man's introduction have to do with this legacy's production—or for that matter, its foreclosure?

3. From the outset, David Wagenknecht, the editor of Studies in Romanticism, questioned the work's capacity to stand on its own. On 10 April 1978, upon having read Timothy Bahti's "Figures of Interpretation, The Interpretation of Figures: A Reading of Wordsworth's 'Dream of the Arab,'" Wagenknecht wrote to de Man to say that the essay could appear in the journal only if de Man took responsibility for it and for the issue as a whole:

   Was it Huckleberry Finn who said of Pilgrim's Progress the statement was interesting but tough? In my case I have to confess the toughness finally overcame my interest: my erasures in the margins reflect an original annoyance which I found more difficult to overcome than to erase. The style, which simultaneously maximizes fussiness and vagueness—verbal operatives tend to be so abstract that almost anything can be related to, equilibrated with, adequated to anything else—caused me at last to mourn for Wordsworth. But I made up my mind to be more manly, to overcome prejudice, and to press on. Having done so, my attitude remained fundamentally the same. (Paul de Man Papers MS-C 4)⁴

   In a confessional (which here is also a fictional) mode, Wagenknecht complains that despite his best efforts at engagement, Bahti's essay left his energies "overcome." Yet beyond his distress with Bahti's style, this difficulty seems due, at least in part, to belief in reading as a dialectical practice. Paralyzed by the essay—and its relentless verbal chain—Wagenknecht calls upon his "manliness" to help him overcome his paralysis. But to no avail. The failed labor of attempted negation (the negation of prejudice, and weakness more generally) leads to further frustration, and thus to the recovery of dialectical practice beyond reading in mourning. Yet, rather than mourn a loss of virility or the apparent failure of dialectics, Wagenknecht laments the loss of the subject of Bahti's essay: Wordsworth himself.⁵ For unlike Wordsworth's poetry, which puts the reader "in the company of flesh and blood," Bahti's essay threatens the reader's manliness and power of mind. Rather than fulfill the romantic project through its contemporary criticism, Bahti's essay, in Wagenknecht's account, leaves its subject dead.⁶

4. In aligning the possibility of reading Bahti's essay with the necessary admission of Wordsworth's death, Wagenknecht implies that Bahti is responsible for this death—or at least for its actualization. Moreover, Bahti is held responsible not only for sacrificing Wordsworth (a dead man), but for a sacrificial "assault" against "the man Harold Bloom has called 'the most defiantly Wordsworthian of modern critics,'" Geoffrey Hartman, Bahti's living teacher, who has most rigorously accounted for the dialectical, which is also to say, sacrificial, structure (Akedah) at work in Wordsworth's poetry (Bahti 607).⁷ Bahti's treatment of Hartman thus leads Wagenknecht to consider the essay in the context of a scene of instruction—and raises the question, which will recur in de Man's introduction, of the violence that constitutes that scene.

5. In his letter to de Man, Wagenknecht praises—and also dismisses—Bahti's piece as "a brilliant graduate student 'performance.'" He calls "honest" the fact that Bahti devotes the first nine pages of his essay to Hartman, but then admits that "the consequent impression that this is a very parochial, intra-Yale performance is very unpleasant" (Wagenknecht's emphasis).⁸ He explains that "it is sometimes difficult for the reader less obsessed with [Hartman's] texts than the author is even to derive a clear sense of what Hartman was talking about," and he proceeds to worry that an essay caught between "obsequiousness" and "assault" against one's teacher "begins to suggest unpleasant things about life in graduate school."⁹ Indeed, for Wagenknecht the "unpleasant" performance—this performance of unpleasantness—seems to be an Oedipal performance, a performance, he insists, that has less to do with the reading of texts than it does with the relations between students and their teachers. Yet, if it is the essay's violent "obsession" with Hartman that Wagenknecht finds suspect, and if it is Bahti's claim that mediation in Hartman is not a work of mind but of trope that he finds interesting, if not completely convincing, he responds not by holding Bahti finally accountable for the assault, but rather by inviting de Man to account for the work of his student and to become the issue's guest editor.¹⁰ Wagenknecht agrees to publish Bahti's paper in Studies in Romanticism only on the condition that de Man will accept responsibility for its violence. As he tells de Man: "I think that there should be clear indication that the editorial choices were finally yours, that the issue in the last analysis, has a guest editor. To me—should this be your inclination—this would all the more imply the necessity of an introductory preface by yourself discussing the seminar and explaining your attitude toward the various essays" (de Man MS-C 4). Wagenknecht sets de Man up as the teacher, and does so with the full acknowledgement that to be a teacher to these students is also to risk being their victim.

6. After several months of silence, de Man finally responds to Wagenknecht: he submits a table of contents for the volume, defends the publication record of the other essays' authors (Johnson, Chase, and Burt), and agrees "to write a brief introduction about the methodological assumptions that stand behind the choice and the treatment of the topic" (de Man 5 October 1978).¹¹ As an essay on the scene of instruction as well as a key moment in it, this introduction—like Wagenknecht's letter—bears witness to the essays' violence. It registers not only the carefully articulated assault against Hartman or the thoroughgoing disarticulation of Wordsworth, but also the violence disguised by loyalty and discipleship.

7. At first glance, the introduction seems a complex, even defensive, response to Wagenknecht's double request. In it, de Man offers both an argument about the event of these papers (one that resonates with his introduction to Carol Jacobs's The Dissimulating Harmony) and an exemplary autobiographical account of the study of romanticism from the 1950s to the 1970s. The introduction opens with an apparent rejoinder to Wagenknecht's dismissal of Bahti's essay as the acting out of an infantile drama performed in a local idiom. De Man begins:

   The essays collected in this issue come as close as one can come, in this country, to the format of what is referred to, in Germany, as an Arbeitsgruppe, an ongoing seminar oriented towards open research rather than directed by a single authoritative voice. Some of the papers originated in a year-long seminar sponsored by the National Endowment for the Humanities conducted at Yale during the year of 1977-78. It was entitled "The Rhetoric of Romanticism," and the title seemed suitable enough to be retained in this expanded version of the initial group. The additional papers were often written in connection with various graduate seminars, but it would be an injustice to see in them only the products of a single "school" or orthodoxy, thus reducing their challenge to mere anecdote. (495)

   The essay initially describes not the success of these works, but their limitation: "They come as close as one can come" to an Arbeitsgruppe—to this truly democratic possibility of free inquiry and "open research," a possibility that remains for de Man paradoxically unrealizable within an American (rather than a German) institution—and in a seminar organized through and funded by the National Endowment for the Humanities.¹² Yet whether this lingering authoritarianism is attributed to national or personal identity, it indicates that the essays, whose freedom and innovation he will recognize, also fail to constitute a truly open seminar. Rather, gathered under the title of the NEH seminar that de Man organized, they appear to be indistinguishable, at least in name, from the work of Paul de Man.¹³ However, as soon as he states these essays' only limited freedom, he also claims that it would be an "injustice" to understand them as "products of a single 'school' or orthodoxy" or as the echo of "a single authoritative voice." It would be an injustice, because this attribution would coincide with the avoidance of reading—the substitution of a history of the works' formation for the difficulty of their reading, and thus the assumption that one knows in advance what these essays say and do. Indeed the substitution not only would be fallacious and limited, but it also would rely upon an inaccurate and ideological version of history. More than merely claim this, de Man claims that the papers demonstrate it—which is precisely the reason why they must be read. While they may have emerged in connection with a seminar that did not fully eschew the authoritative voice of its teacher, these essays—despite their ambitions—ultimately break with, rather than sustain, his lesson. They inherit not a methodology (despite de Man's acknowledgement of the remarkable closeness of their reading), but a textual object (the romantic canon).¹⁴

8. Nevertheless, the event of these essays proves to be indissociable from the question of de Man's legacy, and this becomes especially clear when de Man registers the opposition between their event and an historical (or psychoanalytic) account of their emergence:

   Both in what these papers have in common and in what sets them apart from each other, something is happening that is by no means confined to the idiosyncracies of a particular configuration of individuals. It has to do with a larger question which can be considered as a generational process—although this perspective, too, is misleading; one can as little wish away the innovative and subversive impact of these essays by attributing it to Oedipal struggles as to academic provincialism. The validity of a genetic or generation model for literary history is one of the received ideas that the papers leave behind. It is a matter of chance that the contributors turn out to belong by and large to the same generation, thus providing at least a conveniently fallacious point of view from which to attempt a collective characterization of their achievement.¹⁵

   To insist that "something is happening" here is to acknowledge this work as an event. In happening and so in differing (indeed in differing from itself)—despite all aim and intention—this work involves a rupture, one that de Man cautions against domesticating through a fiction of generation (whether construed as a narrative of cultural mythology or intellectual history).

9. De Man sets out by looking beyond the obvious "idiosyncrasies of a particular configuration of individuals" (e.g., the generation of students who studied literature with de Man at Yale in the 70s) in order to recognize their evocation of "a larger question which can be considered as a generational process." If "generational process" names the event of this work—generation understood as temporalization (and temporalized) rather than as a stable configuration of individuals—it does so as a question, the question of generation itself, and the mark of generation's constitutive incompletion ("Is something happening?" And further: "If something is happening, what is happening?" "Is nothing happening?" "Can nothing happen?").¹⁶ Generation—as a generational process—comes to signal a truly temporal predicament: it signals the impossibility of a birth just the same as a death.¹⁷

10. And so, it is in calling for a reading of this work rather than for an account of its emergence that Paul de Man's own life—his history, his survival, his legacy—comes into question. While de Man claims that the pieces themselves invalidate an historical account that would position them as his legacy or as a Yale production, it is not only their abandonment of "a genetic or generational model for literary history" that challenges the charge of "academic provincialism." Rather, what de Man calls their "innovative and subversive impact" already indicates the displacement of and difference from their teacher that their mimesis enacts. While refusing to read the essays merely as following in his own footsteps, and refusing to reduce their violence and their impact to an "Oedipal struggle," de Man nevertheless frames their event as a betrayal. In stating the radicality of the work collected here, de Man disavows responsibility for it (a gesture that is indistinguishably generous and dismissive), and goes so far as to call the work that would appear to be the legacy of his seminar the instrument of his own death.¹⁸ To say (as de Man does) that "something is happening," and that what is happening "is by no means confined to a configuration of individuals" (495), is also to say that what occurs is a generation that exceeds generation. There is a generational process beyond generational models, a generation that withholds grasping and maintains itself in the form of a question of generation, a generation as the event of generation, that is also, and indistinguishably, a generation without generation. Yet, the claim to this work's importance and its event beyond narrative, the claim to its generation beyond generation, remains a devastating accusation.

11. In order to elaborate his accusation, de Man offers a lengthy (and surprising) parable of his generation and its emergence. The narrative of his generation—a generation whose attachment to the study of romanticism he attributes to guilt, anxiety, and failure—calls into question the enduring status of genetic models of literary history supposedly eschewed by this generation.¹⁹ Like the narratives that would account for the emergence of the work introduced here, this narrative only can be a literary example. However, it begins in an uncomfortably avuncular tone and a clichéd or mechanical idiom in de Man's suggestion that when "people of my generation" began to write, they suffered. They actually had to confront Romanticism as a question of history; they actually had to face up to betraying their elders:

    People of my generation (now roughly speaking in their fifties) interested in Romanticism began to write in the shadow of historical works that considerably refined on preconceived notions of periodization but without losing the sense of historical order to which these works owed their learning and aesthetic discrimination. The answer to such questions as: What is (or was) Romanticism and did such a thing ever occur? became increasingly difficult to formulate but the question itself continued to make sense. (495-96)

    De Man's description of his generation suggests the extent to which he is its product, not despite, but because of, his abandonment of history for theory. The fact that, as he explains in its preface, The Rhetoric of Romanticism—his only book on romanticism—is a fragmentary collection of essays "establish[ed]" by his editor, rather than the historical reflection that he intended Allegories of Reading to be, indicates not the abandonment of history for theory, but the resistance of history and the so-called "historical outlook."²⁰ This claim, on the one hand, relies upon the fiction of generation that domesticates the difficulty of reading and situates de Man's failure to effect an historical study as an historical or generational effect. But it also recognizes de Man's enduring attachment to romanticism (despite his "more theoretical inquiries into the problems of figural language" [RR viii]) as sustaining "the sense of historical order" that dominated the work of his elders.²¹ De Man explains that although his teachers' work led many of the more faithful students of his generation to "start out with the ambition to write their own syntheses or summae of Romanticism" (on the model of their teachers), no volumes of this sort ever were produced. "For all I know," de Man continues, "some may still be about to succeed, yet the fact remains, looking back over the production of the last twenty years, that no general works on Romanticism were produced comparable in scope and serenity to those of the previous decades. More important perhaps, the reasons for this apparent failure became themselves part of the problem" (496).²² Allegories of Reading, which appeared in the same year as the Studies in Romanticism issue, is one of these failures. In its preface, de Man explicitly aligns his work—his desire for and failure to execute an historical project, the interruption of narrative by reading, and the remainder of history as a canon—with that of his generation:

    Allegories of Reading started out as a historical study and ended up as a theory of reading. I began to read Rousseau seriously in preparation for a historical reflection on Romanticism and found myself unable to progress beyond local difficulties of interpretation. In trying to cope with this, I had to shift from historical definition to the problematics of reading. This shift, which is typical of my generation, is of more interest in its results than in its causes. It could, in principle, lead to a rhetoric of reading reaching beyond the canonical principles of literary history which still serve, in this book, as the starting point of their own displacement. (AR ix; my emphasis)

    If Allegories of Reading is not an "historical reflection on Romanticism," if it is the failure or the interruption of that work, this may not be exactly the same as saying—as de Man will say—that it is "in no way a book about romanticism or its heritage" (RR viii). At stake in this description is the very interruption of cognitive and performative language that de Man explains will emerge in the book itself—i.e., the interruption of the difference between a book about romanticism (cognitive, constative) and a book of romanticism and its disruption (performative).²³ In each of these cases, the introductory apparatus raises the question of history and generation—the question of romanticism and the trace of its enduring order on the work of Paul de Man. And, in the introduction to the volume of Studies in Romanticism (as, in the forward to the second edition of Blindness and Insight), the constellation of historical and generational questions emerges as inextricable from guilt and betrayal.²⁴

12. Scholars of his generation, de Man claims, discovered the impossibility of carrying out the work of their teachers. Their desire to write "syntheses or summae of Romanticism" left them—like the Jena Romantics in Blanchot's description—without work in the unproductive discovery "that the writing of literary history and the reading of literary texts are not easily compatible." They discovered that "distinctions become so diversified that no discussion of generations, movements, or specific experiences of consciousness is any longer conceivable" (497-98). Yet, de Man's account of this discovery, the discovery of his generation, emerges in a narrative of his generation, the very narrative that here is rendered impossible.

13. At the zero-point of his reflection—the point at which the inconceivability of a discussion "of generations, movements, or specific experiences of consciousness" (in the case of romanticism) occurs within a narrative "of generations, movements [and] specific experiences of consciousness," de Man not only describes, but also dramatizes, the double bind of his generation. Indeed, the account he provides is indissociable from the predicament whose discovery he reports in it. In other words, his narrative is an example of the trap that it presents. It is an allegory, or what de Man calls his "invention," the invention of a "generation," and at the same time it indicates invention's impossibility:

    Caught between historical norms inherited from their predecessors and their own reading practice, the 'generation' which, for the sake of convenience, I have chosen to invent (and to which no individual will exactly correspond), finds itself in an awkward double bind, reflected in writings that are lopsided in their emphasis on textual analysis as compared to the paucity of the historical results to which they continue to aspire. The tension produced frustrating books and teachers, skillful at best in the techniques of reading but inconclusive with regard to the problems which the readings discover. That so many remained selectively interested in Romanticism is clear evidence of a persistent commitment to the historical outlook that keeps haunting the textual analyses as their bad conscience. (498)

    In the first place, invention refers to the fictional status of the generation that de Man calls his own. It is not what invented or produced him, but rather what he "has chosen to invent." If this means, as he explains, that the genealogical narrative he provides apparently has no referent, his identification of its fictional—indeed fabular—status does not lead him to abandon this narrative. Rather, as soon as he dismisses its referential function and claims this narrative a fable or invention rather than an accurate report, he also produces the referent that he denies. He renders—in a manner that runs counter to expectation—an account of "the study of romanticism [that] would also necessarily be a reflection on our own historical predicament, our history."²⁵ De Man's fable refers at the very instant that it invents. His dismissal of the referential truth of his narrative occurs within a narrative that recovers his dismissal of this "generation" as an invention without referent. It generates the referent whose existence it denies. This specular (self-referential or autobiographical) structure neutralizes the interruptive power of de Man's claim to invention and allows the narrative of generation to proceed without incident. With his generation, de Man demonstrates what Jacques Derrida has called the "invention of the impossible," the impossible invention, as the only invention possible.²⁶

14. In fact the failure (and fable) of this invention — it is a failure because what he claims has no referent is reflected in the text that claims to invent it — can be understood to constitute the "double bind" in which, de Man goes on to explain, "this generation" finds itself. This "generation" emerges as an invention not because it lacks a referent (it is self-referential: the text enacts the predicament de Man reports), but because it is the (textual) emergence of the unprecedented (which may be another way of calculating fiction). Here, invention is a coming into existence rather than (as, and despite) the denial of existence. Invention (the invention of a generation, of this generation) cannot maintain the purity of its fictional or fabular status.²⁷

15. What matters here is that the double bind—the "tension"—that de Man renders and describes (i.e., that he invents) bears upon teaching.²⁸ For de Man, this double bind, far from being a reflection of the "authoritative voice" that he describes at the essay's opening, instead "produced frustrating books and teachers"—books and teachers that remain "inconclusive" rather than synthetic or summative, books and teachers trapped in the double imperative of close textual analysis and a conflicted yet "persistent commitment to the historical outlook."²⁹ While these scholars over and again discover their own duplicity, they also fail to resolve it. Moreover, their enduring attachment to "the historical outlook" becomes, in de Man's account, inextricable from an attachment to romanticism, and romanticism ("our contemporary") becomes inextricable from the double bind of reading and history. In this respect, the "methodological assumptions that stand behind the choice and the treatment of the topic," and which de Man promised his introduction would explain, are not deliberate or rational, but, at least in the case of de Man's (invented) generation, are the effect of an irrepressible—indeed, "haunting"—inheritance. Romanticism is the sign of their "bad conscience."³⁰

16. De Man's admission of romanticism as his generation's bad conscience sounds the moral of his tale, and it shapes the lesson that he offers to (and performs upon) his students. Whereas "Romanticism" signals the unexamined residue of "genetic or monumental patterns of history" (499), which is to say, whereas the enduring study of Romanticism in the work of his intellectual contemporaries indicates the enduring (in a "common" understanding of it) romanticism of their work and thus the bad conscience of close reading, de Man explains, in conclusion, that what is striking about the papers that he presents here is their apparent freedom from the disarticulations of neurotic attachment. He explains:

    The papers in this issue of Studies in Romanticism are remarkably free from this feeling of guilt. They perform their parricide with such a light touch that the target may not even realize what has hit him. The scope is certainly not wider, far from it. . . The selective corpus grows smaller and smaller and gets stuck, at times on a sentence, a title, or a word. But far from causing anxiety, the authors wrest their best findings from these obsessive interrogations. Techniques of rhetorical, as opposed to thematic, analysis are used with remarkable ease, with none of the nervousness which, speaking for myself, makes me feel as if someone were looking over my shoulder whenever thematic assertions can be shown to be subservient to rhetorical overdeterminations. (498)

    Here de Man accounts for two forms of guilt: on the one hand, the guilt suffered by those who remain attached to "the historical outlook," the guilt indicated by a continued attachment to the romantic canon, despite the insistence of rhetorical reading, and, on the other hand, the guilt that attends a rhetorical reading in which "thematic assertions can be shown to be subservient to rhetorical overdeterminations." These two sources of bad conscience are directly at odds: one is ordered by attachment; the other, by disavowal. It becomes clear that the competing tensions they produce would amount to frustration and paralysis. Yet, de Man attributes this work's graceful, rather than awkward, paralysis to an absence of guilt, a clear conscience that would follow from faithfully executing the de Manian lesson; and he attributes its ease and elegance, but above all its "best findings," to this capacity to remain stuck without any anxiety whatsoever. While his anxiety follows from breaking at once too much and too little from his teachers, the essays here, unlike the "frustrating books" of his generation, emerge as the work of good students, burdened neither by the inheritance of history nor the pressure on their inheritance that their readings produce. Nevertheless, the paralysis that marks the papers in this issue—coupled with the serenity of their limited scope—signals, as de Man implies, not the fulfillment, but the betrayal of a legacy.³¹ The assumption of the fulfillment of a legacy masks the extent of its betrayal, the betrayal that de Man uses his introduction to identify. Indeed, the confidence with which these papers carry out de Man's project and pedagogy relies upon genealogy rather than close reading, the reading that de Man solicits in his essay's final sentences.

17. De Man criticizes this work in words almost indistinguishable from praise (and bordering on resentment)—"The selective corpus grows smaller and smaller and gets stuck, at times, on a sentence, a title or a word. But far from causing anxiety, the authors wrest their best findings from these obsessive interrogations," or "Tropes are taken apart with such casual elegance that the exegeses can traverse the entire field of tropological reversals and displacements with a virtuosity that borders on parody" (498)—and he implies that close reading in these essays remains a moment within a dialectic of understanding, rather than its disruption. Thus, de Man points out that the "clarity," calmness and "casual elegance" of these papers takes the place of the "awkwardness," "lopsidedness," and "inconclusiveness" that marked the work of his generation. He implies that, rather than being merely an improvement upon the work of their predecessors in the resolution of their difficulties, these papers continue to discover difficulties, but because the authors recognize difficulty as the ambition of reading, their work continues on, nonplussed. It is in this sense—and in this absence of struggling awkwardness, even in their seeming awkwardness—that the papers fail to emulate the models "held up to them." However, their failure is an effect of their responsiveness to teaching. Close reading does not leave these papers struggling, but, to the contrary, as de Man says of Carol Jacobs's book, in them "the demonstration of . . . necessary incoherence becomes a remarkably sound narrative" (CW 223). This absence of anxiety should lead to a second anxiety, the anxiety that de Man admits to feeling "whenever thematic assertions can be show to be subservient to rhetorical determinations" (498). The absence of this anxiety—freedom from the fear that this work could be an assault against one's teachers, rather than a respectful elaboration of their work—is evidence of disruption rather than continuity. However, this is a disruption that seems for the most part never to interfere with the essays' grace or narrative coherence.

18. Indeed, the "ease, lightness, and grace" that de Man recognizes in these papers makes them appear like the "unfortunates" who dance with prosthetic limbs in Kleist's "On the Marionette Theater."³² It is perhaps no accident that these papers seem to move with the fabricated legs that replace the awkward and unsightly immobility of amputation with "a virtuosity that borders on parody" (498). Yet de Man reminds us that this grace, as if the effect of prosthesis, signals only more broken legs — and in the French of Jacques Derrida rather than the English of Kleist's artists or of these young authors—broken legs that are a broken legacy. The essays' incomparable grace, as in the substitution of prosthesis for mutilation, hides the break of which it also is the evidence.

19. The violence hidden and demonstrated by elegance is what de Man calls a parricide, and this death (a broken leg, broken legacy) seems de Man's own. Blind to the violence that they wield against their elders, unaware that their essays are not the fulfillment of an inheritance but rather its disarticulation, and unburdened by the guilt or anxiety that their unconscious violence should command, these authors, de Man admits, keep their victim—a victim of their grace—oblivious as well. Like all death, this parricide remains an event foreign to consciousness and without proper witness. It remains unknown and unfelt by those good students whose imagined loyalty hides, but does not protect them from, this violence against their teacher; it also remains unknown, indeed unfelt, by the accusatory dead man who "may not even realize what has hit him."

20. Whereas de Man invented a generation in order to account for his relation to his teachers, he invents a parricide in order to account for his legacy. De Man's definition of these essays' violence as a parricide (not just murder, but murder of the father) admits the filiation (through violence and its prosthetic remedy) that he reads them to shatter. Moreover, while de Man establishes that the parricide will have taken place and recognizes that it is the actual, deadly outcome of this issue, he also acknowledges that it cannot take place, and remains what cannot be known to have taken place. As soon as he accuses these essays of violence, he also re-establishes (in a negative mode) their inheritance, their legacy as a broken legacy, and their generation. Thus, the accusation and recognition of a parricide becomes inseparable from the recovery of filiation, the production of guilt, and the insistence of an inheritance. By bearing witness to his own death, to a murder of which only he—and hence no one—is aware, de Man replaces the death he invents with the impossibility of dying.

21. It is in this sense that de Man appears like the Wordsworth he so famously describes. Proleptically speaking from beyond the grave and thus failing to recognize or know death in the moment when he claims to have suffered it, de Man presents the work of his students by accounting not for his influence on or generation of it (his pulling the strings), but for his own death through it. And yet, if in not killing the father, one kills him; if in killing the father one only fails to kill him—and leaves him living on beyond the death that never will have "hit" him—this is also because the murderous progeny (like Oedipus, blind in enterprise, and like the dancing amputees, graceful beyond measure) remain just as oblivious as he.

22. This recovery of filiation in parricide has its parallel in the more general return of history and ethics that de Man recognizes in these essays (and claims is their "most interesting occurrence of all") (498). As he describes:

    At their strongest moments, the shape of another critical discourse begins to emerge, and the critical analysis of the figuration gropes for its own context. This is often accomplished by ways of psychoanalytical schemes of understanding that are no longer ego-centered or by performative modes disencumbered of ethical considerations. The most interesting occurrence of all is that, at the far end of this ongoing enterprise, the question of history and of ethics can be seen to reemerge though in an entirely different manner, no longer predicated, as it was for us, on identifiable evasions of complexities. It would be preposterous to try to state succinctly, in paraphrase, how this reemergence of history at the far side of rhetoric can be said to take place, as if one could spare oneself the labor of reading accomplished in these papers. They deserve at least to have some of their own rigor applied to themselves. Such a reading would reveal that the question of Romanticism can no longer be asked in the manner to which we are still accustomed and that, by extension, the genetic and monumental patterns that are commonly associated with Romanticism have lost much of their authority. The new problems that appear as a consequence are not less redoubtable, but it is exhilarating to capture the moment at which the emancipation is taking place. (498-99)

    For all the grace of these essays, their power coincides with their lost footing, and emerges when the analysis no longer "traverse[s] the entire field of tropological reversals and displacements," but, stumbling, "gropes for its own context." What the papers demonstrate, then—indeed what a reading of them would "reveal"—is not "what a poetics of literature and a theory of reading could be on the far side of literary history" (Bahti, Allegories 293), but, rather, that the "reemergence of history at the far side of rhetoric" is the disruption of grace and the recovery of de Man's legacy. It is the return of history as interruption. And while rhetorical reading may abandon "the question of romanticism" (and not just the answers to this question, as de Man claims was the case for his generation), and while it may displace the "authority" of "genetic and monumental patterns"—of genealogy and generation, and of romanticism as their sign—this abandonment, de Man argues, also recovers the question of history and ethics. It returns the question with which this issue is concerned: the question of legacy and of the possibility of the de Manian legacy.

23. The return of the question of history and of ethics "at the far side of rhetoric" emerges as both the failure and the invention of this generation. In guiltlessly performing their disarticulation of texts, these essays display the negativity and the violence of romanticism. However, this negativity—and its difficulty—conceals the negativity and the difficulty that remain unmanageable. This residual negativity—the blind underside of elegance—coincides with the return of the question of history and ethics beyond rhetorical reading, and, in this respect, proves indissociable from the question of the relation of these works (and these authors) to the work and the teaching of Paul de Man. Rather than having overcome the anxiety that attends simultaneously violating and suffering an inheritance, rather than leaving behind, once and for all, the question of history and ethics, as the trauma and the sign of an inheritance (and of the unavoidable violence of pedagogy), these essays are numb both to the violence they endure and the violence they enact. Yet they are not, de Man suggests, as free as they seem. As they apparently follow his lesson, they murder the paternal figure. In this sense, the question of the legacy remains. It remains as the anxiety that this introduction cannot fail to produce and the filiation that it does not fail to recover. "Legacies of Paul de Man" coincide with the irremissibility of ethics and history, which, in de Man's own account, means his death and the impossibility of his dying. Yet the consciousness of this impossibility also remains another form of blindness, which is to say that this legacy bears the structure of an event.

    * * *

24. It would have been tempting here, and at many points in this reading, to have considered de Man's essay—its association of history and guilt, its account of his generation's relation to the previous generation, and its apparent understanding of a certain approach to literary study—as being symptomatic of de Man's personal bad conscience, and to read this introduction as a rare but undeniable instance in which de Man, in accounting for his guilty attachment to romanticism and his guilty demonstrations of aesthetic ideology, might also be accounting for his own historical and political guilt. Read in this manner, the introduction then would seem to provide the evidence—often sought—of the bearing of de Man's wartime journalism upon his later literary criticism, and, by extension, on the work of his students. Depending upon the reader's desire or stance, this text would allow either for the indictment of literary theory and rhetorical reading (ahistorical, violent, etc.) or the discovery of a covert admission of guilt, one that would somehow allow us to understand de Man, not as having deceived his students, but as having forced them—already in 1979—to confront their teacher's bad conscience. Both of these scenarios would rely upon the same evidence and upon equally fallacious, which is to say, ideological, perspectives. It is tempting to read de Man's allegory of invention (which, as Derrida suggests, indicates an allegory of allegory and which we could extend to incorporate, to reflect and invent, the entire de Manian corpus) as a discreet allegory of his history. However, the non-possibility of appropriating this allegory to redeem or to kill (once again) Paul de Man also indicates the other side of this legacy: the endurance of the question of history and ethics beyond guilt and confession and their possible allegories, and beyond death—including the impossible murder that de Man here witnesses, the death by cancer that he suffers four years later, and the public outrage, confusion, and awkwardness that accompanies the discovery of his wartime journalism in 1984 (the same year that The Rhetoric of Romanticism posthumously appeared).