Sie sind auf Seite 1von 278


Cahiers suisses de litterature generale et comparee

Schweizer Hefte fr allgemeine
und vergleichende Literaturwissenschaft
Quaderni svizzeri di letteratura generale e comparata

Paul- de Man Kolloquium

8. - 10. Juni 1989
Universitt Zrich

Berne Frankfurt am Main New York Paris

Illustration de la couverture tiree de Thomas Rowlandson: The Second Tour

of Dr. Syntax. London 1822.

Gedruckt mit freundlicher Untersttzung der Universitt Zrich.

ISSN 0179-3780

Verlag Peter Lang AG, Bern 1991

Nachfolger des Verlages der

Herbert Lang & Cie AG, Bern
Alle Rechte vorbehalten.
Nachdruck oder Vervielfltigung, auch auszugsweise, in allen Formen
wie Mikrofilm, Xerographie, Mikrofiche, Mikrocard, Offset verboten.
Druck: Weihert-Druck GmbH, Darmstadt


Barbara Johnson: Poison or Remedy?

Paul de Man as Pharmakon ........................... .. . .... ... .......... .......

Peter Hughes: The Usual Terror, the Unusual Suspects ..................

Zusammenfassung ....................................................................


Ortwin de Graef: A Stereotype of Aesthetic ldeology:

Paul de Man, Ernst Jnger .. .. ... .. .... .. ... .. ... .. .. ... .. .. .. .. .... .. .. ... .. ... ..
Zusammenfassung ....... .......... . ...... ............ .. .. .. .............. ............


Hans-Jost Frey: Die Verrcktheit der Wrter .. ..... .. .. .. .... .... .... .. .. .. ...


Heiner Weidmann: Verrcktheit?

Korreferat zu Hans-Jost Frey: "Die Verrcktheit der Wrter" . ... 103
Christiaan L. Hart Nibbrig: La Musique de la theorie ou:
Que veut dire "representer"?
A propos de la "deconstruction" de Paul de Man ...................... 109
Zusammenfassung .... .... ... .. .............. .. ................ ................. ...... 124
Norbert Gabriel: Hlderlin, Heidegger und Paul de Man .......... ...... 125
Abstract .. ................... ............. .. .... .......... .................... . ........... .. 13 8
Marc W. Redfield: De Man, Schiller,
and the Politics of Receptiori .................................................... 139
Zusammenfassung ....... ...... ........ .. ... .. .... .. ..... ............................. 167
Laura Quinney: Skepticism and Grimness in Shelley ... .. .. . .... .... ... .. . 169
Zusammenfassung ........................................................ :........... 182

Claude Reichler: Sens propre et nom propre

dans le Second Discours de Rousseau. .... .. ..... .. .. . ..... .................. 183
Abstract ............................................................................ ........ 192
Peter Grotzer: Paul de Man lecteur de Georges Poulet .................... 193
Zusammenfassung .................. .. .. ...... ..... .. .... ............... .... .. ....... .. 204
Neil Hertz: More Lurid Figures .................................................. .... 205
Zusammenfassung ..... ...... .. .... ............ .... .. .. ... .... .. ......... ............. 240
William Flesch: GlieDErMANn: Defacement as Autobiography .... 241
Resume .................................................................................... 257
Jonathan Culler: Tue Future of Paul de Man .. ..... .. .......... ..... .. ........ 259
Resume .................................................................................... 274
Mitarbeiter ....................................................................................... 275

Barbara Johnson


Merciless and Consequent

This pharma.kon, ,this "medicine", this philter, which acts as both remedy and
poison, already introduces itself into the body of discourse with all its ambivalence. This charm, this spellbinding virtue, this power of fascination, can be altemately or simultaneously - beneficent or maleficent. If the pharma.kon is
"ambivalent", it is because it constitutes the medium in which opposites are op-

posed, the movement and the play that links them among themselves, reverses
them or makes one side cross over into the other.
[...]lt is [...] the prior medium in which differentiation in general is produced.
[...] Writing is no more valuable, says Plato, as a remedy than as a poison.
There is no such thing as a harmless remedy. The pharmakon can never be simply beneficial.
Jacques Derrida, "Plato's Pharmacy''

lt is hardly smprising that the discovery ~f Paul de Man's collaborationist writings should have polarized critics into postures of attack and
defense. Despite frequent dismissals of the affair as a "teapot tempest",
something important seems to be at stake in the "glee" of denunciation
or the tortured and tortuous rhetoric of extenuation. Tue very asymmetry
between the ease of attack and the discomfort of defense deserves comment. Deconstructors have often characterized humanist resistance to deconstruction as an intolerance for paradox, ambiguity or undecidability.
Yet in our current inability either to excuse or to take leave of de Man,
we are now getting a taste of our own pharmakon.
Polarization around de Man can hardly be said to have originated
with the current scandal, however. More than any other literary theorist,
I think, he has always provoked a vehemently split response. For deconstructors, he was a model of rigor, lucidity and integrity; for humanists,
he was a radical nihilist; for materialist critics, a closet conservative. All
three evaluations center on his privileging of language. For many, de

Man's work emitted a highly demanding imperative not to shirk the responsibilities of reading or, as it is sometimes put, not to take the impossibility of reading too lightly. For others, his focus on the unreliability
and randomness oflanguage undennined the foundations ofWestern values. And for still others, his characterization of wars and revolutions as
byproducts of linguistic predicaments was a denial of history and a refusal of politics.
In some ways, all of the above are accurate. If one does not question
the nature of Western values or the definitions of history and politics,
then one would have to assign de Man to the "poison" position in each,
not in his early writings, in which he was himself an upholder of Western values welcoming a revolutionary New Order, but in his later writings. But what if the poison in this case were precisely not the opposite
of the remedy, but an attempt to get at the poison-remedy split at its
The journalists and polemicists are not wrong in locating the specificity of de Man's theory in his focus on language. Their mistake, however, lies in reassigning the certainties they say he takes away. lf language is no longer guaranteed to be reliable or truthful, then it must
"always" be unreliable, false, or biased. lf not necessary, then arbitrary;
if not meaningful, then indeterminate; if not true, then false. But de
Man's analyses do not perform such certainty-reassignments. Rather,
they question the very structure and functioning of such either/or logic.
To question certainty is not the same as to affirm uncertainty:
In a genuine semiology as weil as in other linguistically oriented theories, the
referential function of language is not being denied - f ar from it; what is in
question is its authority as a model for natural or phenomenal cognition. Literature is fiction not because it somehow refuses to acknowledge "reality", but
because it is not a priori certain that language functions according to principles
which are those, or which are like those, of the phenomenal world 1.
lt is by no means an established fact that aesthetic values and linguistic values
are incompatible. What is established is that their compatibility, or lack of it,
has to remain an open question2.

What complicates the picture even further is the fact that, while we
might be able to teil the difference between linguistic and purely phe1 Paul de Man, "The Resistance to Theory", in The Resistance to Theory, Minneapolis, The University of Minnesota Press, 1986, p. 11 (=Rn. (Emphasis mine).
2 Paul de Man, "Tue Return to Philology," inRT, p. 25. (Emphasis mine).

nomenal or aesthetic structures ("no one in his right mind will try to
grow grapes by the luminosity of the word 'day'"), the distinction is not
at all clear in the case of ideology or politics, because "wh~t we call ideology is precisely the confusion of linguistic with natural reality, or reference with phenomenalism". From this de Man goes on to assert:
lt follows that, more than any other mode of inquiry, including economics, the
linguistics of literariness ["literature as the place where this negative knowledge
about the reliability of linguistic utterance is made available"] is a powerful and
indispensable tool in the unmasking of ideological aberrations, as well as a determining factor in accounting for their occurrence. (Kl', p. 11)

In the years just prior to his death, de Man seems indeed to have been

moving toward establishing a more explicit link between his own theoretical stance and a critique of the ideological foundation of N azism.
Christopher Norris has pointed to that link by entitling his study of de
Man Deconstruction and the Critique of Ae.sthetic Ideology. As Walter

Benjamin was one of the first to point out, fascism can be understood as
an aestheticization of politics. In several late essays, de Man locates a
crucial articulation in the construction of a protofascist "aesthetic ideology" in Schiller's misreading of Kant's Critique of Judgment. Schiller's
misreading of the aesthetic in Kant involves a denial of (its own) violence. Schiller's vision of "the ideal of a beautiful society" as "a weil executed English dance" has exerted a seductive appeal upon subsequent
political visions .. In his essay entitled "Aesthetic Formalization"3, de
Man juxtaposes to this notion from Schiller a short text by Kleist, ber
das Marionettentheater, in which the grace of such a dance is shown to
be produced by substituting the mechanical(a puppet or a prosthesis) for
the human body. Schiller's "aesthetic state" is thus an ideal that can only
be produced by mutilation and mechanization. The dance-like harmony
of a state can only arise through the repression of differences within. In
one of the last lectures de Man delivered before his death, he makes the
political ramifications of this aesthetic state even clearer:
As such, the aesthetic belongs to the masses [ ... ] and it justifies the state, as in
the following quotation, which is not by Schiller:

"Art is an expression of feelings. Tue artist is distinguished from the nonartist by the fact that he has the power to give expression to what he feels.
3 "Aesthetic Formalization: Kleist's ber das Marionettentheater", in The Rhetoric of
Romanticism, New York, Columbia University Press, 1984 (= RR).

In some form or another: the one in images, a second in clay, a third in

words, a fourth in marble - or even in historical forms. The statesman is an
artist too. The leader and the led ('Fhrer und Masse') presents no more
problem than, say, painter and colour. Politics are the plastic art of colour.
This is why politics without the people, or even against the people, is sheer
nonsense. To shape a People out of the masses, and a State out of the People, this has always been the deepest intention of politics in the true
lt is not entirely irrelevant, not entirely indifferent, that the author of this passage is from a novel of Joseph Goebbels. Mary Wilkinson, who quotes the passage, is certainly right in pointing out that it is a grievous misreading of
Schiller's aesthetic state. But the principle of this misreading does not essentially differ from the misreading which Schiller inflicted on his own predecessor, namely Kant4.

Micha.el. Ein deutsches Schicksal in Tagebuchblttern (1929)

De Man's insistence on violence - disfiguration, death, mutilation - is

not a personal predilection for horror, but rather a deep suspicion of false
images of harmony and enlightenment. Hidden within the aesthetic appeals of the political images by which he himself was once seduced were
forms of violence unprecedented in human history. lt seems undeniable
that if "the linguistics of literariness is a powerful and indispensable tool
in the unmasking of ideological aberrations, as weil as a determining
factor in accounting for their occurrence", the ideological aberrations he
is unmasking were once his own.
lt could be objected that his relation to such "aberrations" remains
purely cognitive, that "accounting for" occurrences may not be the only
possible response to history, and that the ideology de Man "unmasks"
remains, in fact, masked. The political implications of his cognition remain at odds with the political implications of his performance. His refusal to teil his own story, which can be seen both as self-protection and
as self-renunciation, was also a silencing of the question of the origins or
consequences of his acts of cognition in the world. His unmasking of
aberrant ideologies maintains a metaphorical, rather than a metonymical,
relation to history. Yet those acts of cognition, however insufficient they
may seem now, are not to be discarded because of this refusal to go further. In the absence of any guarantee as to Paul de Man's moral character
or political vision, his writings remain indispensable in their insistence


Paul de Man, "Kant and Schiller", unpublished manuscript.

that the too-easy leap from linguistic to aesthetic, ethical, or political

structures has been made before, with catastrophic results.
Yet that insistence is never made unironically, in such a way as to
imply that the errors are ultimately avoidable. A typical concluding sentence takes the form of a double negative:
With the critical cat now so far out of the bag that one can no longer ignore its
existence, those who refuse the crime of theoretical ruthlessness can no longer
hope to gain a good conscience. Neither, of course, can the theorists - but, then,
they never laid claim to it in the first place5.

Tue fact that the theorists never laid claim to a good conscience is by no
means reassuring, especially in light of recent developments. Nor does it
erase the impression of moral self-satisfaction this sentence conveys. In
this typical pharmakon-like ending, de Man makes it difficult to teil
which of the possible remedies is more poisonous. All the more so since
in the original publication of this essay in the Times Literary Supplement, the final sentence had read "neither, of course, can the terrorists".

The Inhuman and the Impersonal

Things happen in the world which CaIUlot be accounted for in terms of the human conception of language. [ ...] Understand by nihilism a certain kind of critical awareness which will not allow you to make certain affirmative statements
when those affirmative statements go against the way things are.
Paul de Man, "Walter Benjamin's The Task of the Translator"

Tue pleasure with which de Man manipulates terms like "ruthlessness" is

unsettling, however much one wishes to believe he is only "waming
against unwarranted hopeful solutions". Something of what is at stake
may be gleaned from an exchange published in The Resistance to Theory
between de Man and Meyer Abrams. The exchange occurred during the
discussion of de Man's lecture on Walter Benjamin's "Tue Task of the
Translator". Elaborating on the Statement that "Benjamin says, from the
beginning, that it is not at all certain that language is in any sense human", de Man explains:
5 ''The Return to Philology", in RT, p. 26.



de Man:

de Man:

Tue "inhwnan," however, is not some kind of mystery, or some kind

of secret; the inhuman is: linguistic structures, the play of linguistic
tensions, linguistic events that occur, possibilities which are inherent
in language - independently of any intent or any drive or any wish or
any desire we might have.
I want to go back to the question [...] about language being somehow
opposed to the human. I want [...] to provide a different perspective,
just so we can settle the matter in a different way. And that perspective won't surprise you because you've heard it before and expect it
That's very hwnan.
Suppose I should say, as many people have said before me, that instead of being the nonhuman, language is the most human of all the
things we find in the world, in that language is entirely the product of
human beings. [...] Now, suppose that, alternatively to looking at the
play of grammar, syntax, trope, somehow opposed to meaning, I
~hould say - and l'm not alone in saying this - that language, through
all these aspects, doesn't get between itself and the meaning, but instead that language, when used by people, makes its meaning. [ ...]
What can be more human than the language which distinguishes human beings from all other living things? [ ... ] All I want to do is present the hwnanistic perspective, as an alternative, an optional alternative, which appeals to me. Instinctively, it appeals to me.
Well, it appeals to me, also, greatly; and there is no question of its appeal, and its desirability. Tue hwnanistic perspective is obviously
there [...] [But] a certain kind of critical examination [...] has to take
place, it has to take place not out of some perversity, not out of some
hubris of critical thought or anything of the sort, it has to take place
because it addresses the question of what actually happens. Things
happen in the world which cannot be accounted for in terms of the
human conception of language. And they always happen in linguistic
terms. [...] And good or bad things, not only catastrophes, but felicities
also. [...] One could say, with all kinds of precautions, andin the right
company, and with all kinds of reservations - and I think it's a very
small company - that Benjamin's concept of history is nihilistic.
Which would have to be understood as a very positive statement about
it. [...] Understand by nihilism a certain kind of critical awareness
which will not allow you to make certain affirmative statements when
those affirmative statements go against the way things are6.

Earlier in the discussion, de Man situated the crux of the difference between himself and Abrams as follows: "If one speaks of the inhuman, the
fundamental non-human character of language, one also speaks of the
fundamental non-definition of the human as such". Tue problem is not
6 "Conclusions: Walter Benjamin's Tue Task of the Translator, "'in RT, pp. 96-104.


one. of deciding whether language is or is not human, but rather of

knowing exactly what the word "human" means. Language becomes the
pharmakon within which it is both impossible and "desirable" - indeed,
urgent - to separate the human from the inhuman. But from what standpoint ~s such a statement being said?
Tue question of the humanness or inhumanness of language is very
much tied to the question of a lieu d'enonciation. In an early and fundamental essay on Mallanne, entitled "Poetic Nothingness", de Man quotes
a famous letter written by Mallanne at a tuming point in his poetic career:
This last year has been a temifying one. My thought has worked through to a
Divine Conception. [...] I write to inform you that I am impersonal now, and no
longer the Stephane you once knew - but an aptitude the spiritual universe has
for seeing itself and for developing, through whatonce was me7.

As de Man demonstrates in his discussion of Benjamin, from the Divine

to the inhuman il ny a qu'un pas. De Man follows very much in this tradition of impersonality, which was to lead Mallanne to the theory of the
"elocutionary disappearance of the poet, who leaves the initiative to
words". As Hillis Miller points out, "the first person pronoun is used
rarely and sparingly by de Man. [... ] This goes along with an austere
rigor that makes his essays sometimes sound as if they were written by
some impersonal intelligence, or by language itself '8. This eclipse of the
seif by language is both the content and the rhetorical mode of de Man's
writing. I have analyzed elsewhere both the grammatical errors and the
personifications that mark this apparent eclipse9. Here I would like to
retum to de Man's gloss on Benjamin's suggestion that language might
not be in any sense human. Tue example discussed is strange. Inhumanness seems to inhere in the lack of correspondence between the Gennan
word "Brot" and the French word "pain".

7 Quoted in Paul de Man, "Poetic Nothingness", Critical Writings 1953-1978, Minneapolis, The University of Minnesota Press, 1989, p. 25.
8 J. Hillis Miller, "'R~ading' Part of a Paragraph in Allegories of Reading", in Reading
de Man Reading, ed. Lindsay Waters and Wlad Godzich, Minneapolis, The University of Minnesota Press, 1989, p. 165.
9 See my essay entitled "Rigorous Unreliability", in A World of Difference, Baltimore,
The Johns Hopkins University Press, 1987.


Tue translation will reveal a fundamental discrepancy between the intent to

name Brot and the word Brot itself in its materiality, as a device of meaning. If
you hear Brot in this context of Hlderlin, who is so often mentioned in this
text, I hear Brot und Wein necessarily, which is the great Hlderlin text that is
very much present in this - which in French becomes Pain et vin. "Pain et vin"
is what you get for free in a restaurant, in a cheap restaurant where it is still included, so pain et vin has very different connotations from Brot urul. Wein. lt
brings to mind the pain franfais, baguette, ficelle, bOlard, all those things - I
now hear in Brot "bastard". This upsets the stability of the quotidian. I was very
happy with the word Brot, which I hear as a native because my native language
is Flemish and you say brood, just like in German, but if I have to think that
Brod [brood] and pain are the same thing, I get very upset. lt is all right in English because "bread" is close enough to Brot [brood], despite the idiom
"bread" for money, which has its problems. But the stability of the quotidian, of
my daily bread, the reassuring quotidian aspects of the word "bread", daily
bread, is upset by the French word pain. What I mean is upset by the way in
which I ~ean - the way in which it is pain, the phoneme, the term pain, which
has its set of connotations which you in a completely different directionlO.

Later, in his response to Abrams, de Man equates the inhuman with loss
of control. That "there is a nonhuman aspect of language is a perennial
awareness from which we cannot escape, because language does things
which are so radically out of our control that they cannot be assimilated
to the human at all, against which one fights constantly" (RT, p. 101). At
another point, however, when questioned by Neil Hertz about how a
word like "inhuman" can be derived from the connotations of Brot, de
Man responds by treating the example itself as a loss of control:
Weil, you're quite right. I was indulging myself, you know, it was long and I
was very aware of potential boredom, felt the need for an anecdote, for some
relief, and Benjamin gives the example of pain and Brot, and perhaps shouldn't
... whenever you give an example you, as you know, lose what you want to say;
and Benjamin, by giving the example of pain and Brot - which comes from him
- and which I've banalized, for the sake of a cheap laugh .... (RT, p. 95-96)
(ellipses in original)

In its ellipses, its anacoluthons, its denials, this passage comes very close
to total incoherence. De Man dismisses the example as a cheap laugh,
but his nervousness suggests that he sees it as a slip. What has the slip let
show, and what does it have to do with the inhuman? lt seems to me that
what the slip reveals is not the inhuman but rather the human as a loss of
control - the de Man that suddenly says "I get very upset", "my native
10 'The Task of the Translator", in RT, p. 87.


language is Hemish", later feels this small outbreak of exhibitionism as

the intrusion of something over which he has lost control. What de Man's
categories of human and inhuman seem to lack is a concept of the unconscious. Though he may have reasons to feel it with particular acuity, he
is not alone in experiencing the approach to the mother tongue - or perhaps to the mother as such - as the threat of a loss of control. Indeed,
even without assigning an unconscious meaning to the example (seeing
in it de Man's relation to his mother, or to Belgium), one would have to
say that de Man's riff on pain and Brot sounds suspiciously like a process of free association, a rather exuberant glide along what Lacan would
call the signifying chain.
De Man's desire to replace the unconscious with the randomness of
language is made explicit in his analysis of Rousseau's slip - Rousseau's
blurting out of the name "Marion" as the stealer of a ribbon. De Man
Because Rousseau desires Marion, she haunts his mind and her name is pronounced almost unconsciously, as if it were a slip, a segment of the discourse of
the other. But the use of a vocabulary of contingeny ("le premier objet qui
s'offrit") within an argument of causality is arresting and disruptive, for the sentence is phrased in such a way as to allow for a complete disjunction between
Rousseau's desires and interests and the selection of this particular name. Marion just happened to be the first thing that came to mind; any other name, any
other word, any other sound or noise could have done just as weil and Marion's
entry into the discourse is a mere effect of chance 11.

In a way, the entire debate between Lacan and Derrida conceming the
purloined letter is summarized here in this. discussion of a purloined ribbon. That such was de Man's own understanding of it can be seen not
only in the Lacanian phrase "the discourse of the other" but also in the
fact that the original title of "Excuses" was "The Purloined Ribbon".
De Man's refusal of psychoanalysis, bis desire for chance rather than
desire to have the last word, is of a piece with his adoption of a stance of
impersonality. But just as his desire to erase the self leaves residues in
the grammar of his essays, so too does his self-effacement leave
psychoanalytically readable traces in the domain of pedagogy. For the
remainder of this paper, I would like to analyze de Man as a pedagogical,
rather than merely as an intellectual, figure. I think that it is impossible

Paul de Man, "Excuses", Allegories of Reading, New Haven, Yale University Press,
1979, p. 288.


to understand the intensity of the current debates about de. Man without
taking the pedagogical arena into consideration. De Man himself characterized his work as "more pedagogical than philosophical; it has hlways started from the pedagogical or the didactic assignment of reading
specific texts rather than, as is the case in Derrida, from the pressure of
general philosophical issues"12. And the editors of the special issue of
Y ale French Studies entitled 'Tue Lesson of Paul de Man" go so far as to
say "He was never not teaching"13.

What ls a Teacher?
A giving which gives only its gift, but in the giving holds itself back and withdraws, such a giving we call sending.
Martin Heidegger, On Time and Being
(quoted by Alan Bass in his introduction to
Derrida's The Post Card; Bass's italics)
What haunts are not the dead, but the gaps left within us by the secrets of others.
Nicolas Abraham, "Notes on the Phantom"

Tue title of this section is an allusion to Michel Foucault's influential essay, "What is an author?" In that essay, Foucault is at pains to distinguish between a biographical "author" and a textual "author function". In
the same way, Paul de Man seemed to want to distinguish between the
person of the teacher and the intellectual process in which he or she is
Overfacile opinion notwithstanding, teaching is not primarily an intersubjective
relationship between people but a cognitive process in which self and other are
only tangentially and contiguously involved. Tue only teaching worthy of the
name is scholarly, not personal; analogies between teaching and various aspects
of show business or guidance counseling are more often than not excuses for
having abdicated the task14.

In her memorial tribute to de Man, Ellen Burt describes eloquently the

extent to which de Man succeeded in existentializing this view of the
12 "An Interview," in RT, p. 117.
13 The Lesson of Paul de Man, Yale French Studies #69, ed. Peter Brooks, Shoshana
Felman, and J. Hillis Miller, New Haven, Yale University Press, 1985, "Forcword".
14 ''The Resistance to Theory", inRT, p. 4.


teacher, by conveying "as complete a detachment from the claims of

subjectivity or individual personality as was possible"15. A different kind
of example of his pedagogical self-effacement occurs in his introduction
to a special issue of Studies in Romanticism which presents the work of
six of his students. While eventually speaking about "my generation"
and "their generation", he nevertheless totally avoids the first person
pronoun in the long opening paragraph in which he refers to their participation in "a [rather than "my"] year-long seminar sponsored by the
National Endowment for the Humanities" or "various graduate seminars". "lt would be an injustice to see in them only the product of a single 'school' or orthodoxy, thus reducing their challenge to mere anecdote". Rather, he asserts, "the essays collected in this volume come as
close as one can come, in this country, to the fonnat of what is referred
to, in Gennany, as an Arbeitsgruppe, an ongoing seminar oriented towards open research rather than directed by a single, authoritative
voice"16. Yet no one is fooled by these denials of authority. They serve
only to increase the impression of authority he conveyed. Again and
again both admirers and critics testify to his authority as paradoxical.
Tue memorial tributes are eloquent on the subject:
I want to speak not about de Man's power as a teacher and as a writer, but about
the extraordina:ry intellectual authority he exerted on his friends and colleagues,
at least on me. Paul de Man much disliked words like "power", "force", or
"authority", especially when applied to the academic world. He would have
smiled ironically again and more than a little scornfully at the idea that he had
what those words name, though obviously he did.
(J. Hillis Miller)
His jokes would always be in some sophisticated manner joking at his own expense, pedagogically disclaiming, in a Nietzschean manner, his own authority.
[...] Paul disclaimed his own authority, yet none had more authority than him.
(Shoshana Fehnan)
Tue last thing he probably would have wanted tobe was a moral and pedagogical - rather than merely intellectual - example for generations of students and
colleagues, yet it was precisely his way of not seeking those roles that made
him so irreplaceably an exception, and such an inspiration.
(Barbara Johnson)

15 "In Memoriam," in The Lesson of Paul de Man, p. 11.

16 Studies in Romanticism, 18, Winter 1979, p. 495.


lt has recently been remarked that the unearthing of de Man's early writings has induced deconstructors to abandon their anti-biographical stance
and to search for intentions, mitigating circumstances, and contextual elements that might help to understand the man. What this reveals, however, is not that the person of the author has suddenly been brought back
from exile, but that that person was always already there, idealized as
impersonal. In other words, it was not despite but rather because of his
self-effacement that students and colleagues were led to substantialize
and idealize him, as if the teacher as person could simply be deduced
from the teacher-function. They took him as a metaphor, not a
metonymy, of his persona. Just as in a psychoanalytic situation the analyst's silence allows the analysand to construe him or her as a subjectpresumed-to-know, de Man's silence about himself (including - but not
restricted to - his past) created a blank on which admirers could project
their idealizations. This fascination clearly had its dangers - but where
did we get the idea that powerful teaching could ever be purely beneficent? Western philosophy indeed originated with a text that saw the
teacher as pharmakon, perched between enlightenment and corruption.
De Man's blank was in fact itself already constructed out of transferential idealization. As de Man told Stefano Rosso, "I have a tendency to
put upon texts an inherent authority. [... ] I assume, as a working hypothesis, (as a working hypothesis because I know better than that), that the
text knows in an absolute way what it's doing"17. In a passage in which
de Man describes his discovery through Reuben Brower of pedagogy as
a subversive activity, he explains teaching as an induction into this type
of heuristic transference onto a text as a site of (perhaps inhuman)
Students, as they began to write on the writings of others, were not to say anything that was not derived from the text they were considering. They were not
to make statements that they could not support by a specific use of language
that actually occurred in the text. They were asked, in other words, to begin by
reading texts closely as texts and not to move at once into the general context of
human experience or history. Much more humbly or modestly, they were to
start out from the bafflement that such singular tums of tone, phrase, and figure
were bound to produce in readers attentive enough to notice them and honest
enough not to hide their non-understanding behind the screen of received ideas
that often passes, in literary instruction, for humanistic knowledge. [...] Mere
reading, it turns out, prior to any theory, is able to transform critical discourse
17 "An Interview", p. 118.


in a manner that would appear deeply subversive to those who think of the
teaching of literature as a substitute for the teaching of theology, ethics, psychology, or intellectual history. Close reading accomplishes this often in spite
of itself because it cannot fail to respond to structures of language which it is
the more or less secret aim of literary teaching to keep hidden 18.

If this type of teaching is subversive - and it is certainly subversive in

the ways de Man describes - it is because it is materialist _, it takes language not only on the level of meaning but on the level of meaning-production and -disruption. But in another way, this teaching is also deeply
conservative, not in its content but in the frame it draws around that
content. Tue instructions to the students are phrased in the grammar of
an absolute but hidden authority: "students were not to ... they were to
... ". Listen to the description Richard Ohmann gives of a not-so-different
set of instructions to the student taking the Advanced Placement English
Another thing a student is supposed to be is objective. The Acom Book says
that his Advanced Placement English course will teach him how to read andrespond to works of literature, but if the descriptive material and the examinations
are any indication, the Advanced Placement Program actually teaches the student not to respond to literature, not with his feelings. His concem must be with
"organization of the elements of the poem", with "particular uses of language"
that express a contrast, with the function of minor characters, with the way
structure, imagery, and sound contribute to the whole meaning of a poem "your feeling about the poem is important'', he is implicitly told, "only as the
outcome of careful reading". His role is that of the neutral instrument, recording
and correlating the facts and drawing conclusions. If any need or interest other
than the formalistic drove him to read the work, or indeed, if something within
turns him against the work, he will quickly learn to suppress these unwelcome
responses. They are not among the competencies that will move him a step up
the ladder. To his reading of a poem he is supposed to bring the techniques he
has mastered, and only those. He is, in other words, alienated in very nearly the
Marxian sense. and, of course, the ideal student is of the middle class. Docility,
care, tidiness, professional ambition, the wish for objectivity, these are all qualities valued particularly by the middle class and encouraged in its young19.

By thus drawing a frame around the text as a sujet suppose savoir, the
teacher can teach the student not to ask certain questions about the literary canon or about the teacher. lt is no accident that few students ever
18 ''The Return to Philology'', RT, pp. 23, 24.
19 Richard Ohmann, English inAmerica, New York. Oxford University Press, 1976, p.


asked de Man what he had done during the war. De Man's subversive
teaching certainly unsettled many of the assumptions that have accompanied the humanist understanding of the canon, but he did nothing to unseat the traditional white male author from his hiding place behind the
impersonality of the universal subject, the subject supposed to be without
gender, race, or history. He created a slightly idiosyncratic canon of his
own (in part through throw-away lines like "in the profession you are nobody unless you have said something about this text"), but he did not
suggest tliat there were multiple literary histories, or readers with completely different senses of what was urgent. Perhaps it was not his place
to do so. His pedagogy was a pedagogy of self-difference and self-resistance within a traditional understanding of canonical texts and questions.
lt is up to us to open the subversiveness of teaching further - without
losing the materialist conception of language that remains de Man's truly
radical contribution.


Peter Hughes


[...] dans l'imagination, tout perd ses formes naturelles, et tout s'alrere, et l'on y
cree des liberres comme les yeux creent des figures dans les nuages.
Saint Just

Those who have been engaged in writing conflicting epitaphs on Paul de

Man's life and work have at least one thing in common: blindness to an
ideology that underlies both his early wartime articles and much that is
surprising and strange about his later work. Even before his death in
1983, de Man had been praised as an unmasker of ideology, an unveiler
of rhetoric. And yet he was at the same time also attacked as a negative
mystic who conferred on texts motives and powers he would not grant to
their authors. After his first death there was another, the bringing to light
in 1987 of the reviews and articles he had written in wartime Belgium,
most of them in 1941 and 1942, most of them for the then collaborationist newspaper Le Soir. In the months since then these articles, most
of which are more dreary than dreadful, have been read as signs or manifestos of an "aesthetic ideology" that stretches back to Kant or even of
what Benjamin described as "the aesthetizing of politics" and as the sinister attraction of fascism 1. There can be little doubt that the thread of
Paul de Man's life and career was cut or broken by the personal and political calamities of those years: the ten years of silence that followed is
evidence enough. But to suggest, as some friends and enemies have, that
the rest of his life and work were devoted to undoing such an aesthetic
ideology, or to furtively propagating fascist ideas, strikes me as questionable on the one hand and slanderous on the other. Both views seem
blind to the possibility that the conflict in his writing and thinking may
be inward and imaginative rather than linear and extemal. And that I
1 See especially Christopher Norris, Paul de Man: Deconstruction and the Critique of
Aesthetic ldeology, New York & London, Routledge, 1988, pp: 1-27, 177-97;
Jacques Derrida, "Like the Sound of the Sea Deep within a Shell: Paul de Man's
War", Critical Inquiry, Spring 1988, 14, 3, pp. 590-652.


shall argue is the possibility raised by those wartime articles. lf we read

them as they might have been read by a contemporary, or as de Man
himself would read them, they cast a different light on his literary
politics and poetic theory. Far from implying or anticipating an aesthetic
ideology of romantic poets and Gennan philosophers, these early articles
are pervaded by the spirit of what Jean Paulhan had just described as "La
Terreur dans les lettres" - to cite the subtitle of his Les Fleurs de Tarbes,
which appeared in book fonn in the late summer of 19412.
The pervasive spirit of literary terrorism, whose spoor Paulhan traced
all around him in the critical writing of his time, from academic discourse to f euilleton pieces like Paul de Man's, is marked on the one hand
by a suspicion or even fear of the power of words, by an ascetic aversion
to rhetoric and style. But it is a divided spirit, deeply ambivalent. Its
denigration of tropes as commonplaces goes hand in hand with an exaltation of literature to the status of myth or the truth-claim of philosophy.
Conviction that the idea is worth more than the word and that language is
dangerous to thought converge in Paulhan's definition: "La definition la
plus simple qu'on puisse donner du Terroriste, c'est qu'il est misologue",
(p. 64 n.). In a note to his own definition, Paulhan points out how this
misologism dovetails with the deconstructive analysis of "grands mots"
and high sounding tenns. Such analysis of politics, or even of dreams,
tends to the conclusion that the man who speaks of liberty and equality,
"or even talks of flight or a ball - the dream being here only another sort
of language - is not at all thinking what he seems to be thinking" (p. 64
n.). The earlier version that appeared in his Nouvelle Revue Fran~aise in
1936 opens with the Blakean advice, taken from Baudelaire, that persisting in folly, pushing excess to its limit, will undo it or produce its opposite; just so here, "we have pushed a special kind of Terror to its limits, and have discovered Rhetoric" (p. 231 ). This was Paulhan's original
plan, which called for a second part that would reassert the right and
claims of rhetoric. Although that part was never written, not at least as a
continuation of Les Fleurs de Tarbes, the whole of Paulhan's brilliant essay is haunted by the revelations to come: above all by the growing
2 An earlier and briefer version had appeared in the Nouvelle Revue Fran{aise in
1936, and the idea of the book goes back as far as a letter to Francis Ponge in 1925.
lt was meant to be followed at once by a second part that would restore the rhetoric
called in question by Les Fleurs de Tarbes. References here are to the Idees I Gallimard edition (Paris, 1973), which includes both versions: subsequent page references will be given in the text.


awareness that the denial of rhetoric is its assertion, the recognition of its
power. Georges Perec's novel La. Disparition, to take a more recent example, eliminates the letter "e", not by erasure but by its "disappearance"
from the text. Perec refuses to write any word including the letter into his
narrative. Hence the letter "e", because of that refusal - which involves
writing the novel around the black hole of that missing letter - becomes
the most important letter in the book. Tue refusal to speak or write a
commonplace or cliche, the avoidance of a rhetorical figure, recognizes
and asserts that commonplace, cliche, or figure.
This is a central paradox of Les Fleurs de Tarbes, and part of its
lasting or later fascination for Paul de Man. We do not know how early
he. read Paulhan's essay, but he often told friends.of his interest in Paulhan3, and his last interrupted seminar at Yale was concemed with Les
Fleurs de Tarbes4. As I read de Man's wartime articles these questions
slip away into the background, tobe replaced by the sense that these articles are being read by Paulhan. Many of Paulhan's examples in the
1941 version of Les Fleurs de Tarbes, many of his terrorist suspects,
come in fact from polemical and political joumalism very similar to the
pieces published during that same year in Le Soir. This eerie impression
deepens with the realization that three of the strangest of the early articles suddenly make sense - a sense very different from the irony of double-talk that has been found in them - if we recognize in them the divided spirit of terrorism.
Two of these pieces are strange because. they turn accusingly on two
of the most prized of collaborationist authors in occupied and Vichy
France, Henri de Montherlant and Robert Brasillach. In his perceptive
discussion of de Man's wartime writing, Jacques Derrida notices the
strangeness of these attacks of Montherlant's Solstice de juin and Brasillach's Notre avant-guerre, but he misreads crucial passages out of an er~
ror that masks the link between these early suiprises and the strangeness
of de Man's mature writingss. What we discover about de Man's role.and
coded style in the literary terrorism of collaborationist Europe points
ahead to aspects of his later work: his oscillation between sceptical wit
3 As in a letter to Hans-Jost Frey, where de Man prefers him to Adorno andin several
conversations with J. Hillis Miller (both private communications).
4 I am indebted to Cynthia Chase for her notes on the meetings of this seminar and on
its discussion of de Man's later reading of Les Fleurs de Tarbes. This later reading
forms part of my study in progress on rhetoric in Paulhan and de Man.
5 Critical /nquiry (Spring 1988), pp. 612-614.


and sudden prophecy, an unpredictable alternation between lucidity and

the cloud of unknowing. To overlook this link, to see de Man as an unmasker of ideology, is at once true and deceptive, because it at once
makes it possible to read him and impossible to account for the oddity of
the experience. As Jonathan Culler has put this dilemma, "One can only
make sense of his writings if one already has a sense of what they must
be saying and can allow for the slippage of concepts, working to get over
or around the puzzling valuations, the startling assertions, the apparently
incompatible claims"6 . Part of the puzzle, as we shall see, lies in the
"apparently incompatible claims" of literary terror itself; at once austere
and frivolous, misological and as it were, pathological: words bring only
more words to mind, and vice versa. Tue Terror presented by Paulhan, as
Maurice Blanchot observed, finally becomes literature itself, condemned
either to silence or the saving grace of unending illusion7.


est indispensible d'ecrire tres lisiblement pour faciliter le contrle des

autorites allemandes.
(lnstruction on French wartime correspondence card sent by Jean Paulhan to
Francis Ponge in August, 1941).

To trace the links between terrorism and deconstruction we need to turn

back to de Man's two articles and then to a third, which reviews Ernst
Jnger's personal account of the invasion of France and two books on the
"new order" arising in Europe. I cannot hide from myself, and should not
from the reader, what a disturbing experience the actual reading of these
pages has been for me. Unlike the pieces from H et Vlaamsche Land,
given to me translated and transcribed, the articles from Le Soir are on
photocopies of the newspaper pages themselvess. They appear in the
midst or on the edge of a mosaic of headlines, editorials, articles, decrees
and photos. If we read Paul de Man's contributions as they would have
been read by a contemporary, we see them surrounded and framed by
6 ''The lesson of Paul de Man", Yale French Studies, no. 69, p. 106.
7 "Comment la liuerature est-elle possible," in Faux Pas, Paris, Gallimard, 1943, p.

8 I owe these copies, on which my readings are based, to William Flesch and Ortwin
de Graef.


images of collaboration and betrayal, by Aesopian texts that are hard to

read but impossible to forget. None of the commentary I have so far seen
on these articles has called any attention to the historical and political
context they imply. This context is also a key to the code de Man's articles were written in, and to neglect or overlook that context is to miss
their tone and even their point. The missing contexts, the surrounding
columns of propaganda that have been cut away, created an eerie double
impression that has now been dispelled or hidden. For while many of the
articles seem inoffensive out of context, they are made horrible by the
company they keep - as in Le Soir's antisemitic issue of 4 March 1941.
Any one who tries to break the code of these articles will be thwarted
at every turn by the way they have been published: the thin border of
context on the photocopies circulated in the autumn of 1987 has been cut
away, and nothing explanatory has replaced it9. The articles have been
estranged as facsimiles (some of them barely readable) of their originals,
and nothing has been done to bring them closer through notes or commentary. The editors claim that "This collection should also contribute to
the study of journalism, specifically political and cultural joumalism in
an occupied country" (p. vii), but nothing in their brief preface tells the
reader why and how that study is made hard and tantalizing by the
workings of censorship, the growth of fahles and deniable allegories, the
absence of an explanation that is itself an explanation - by, in a word,
the rhetoric of ideology that Paul de Man was then and later so quick to
see and subvert. The blank refusal to read through his articles to the
palimpsest of propaganda and historical interpretation on which they are
inscribed, or even to relate them to the context of Le Soir or Het Vlaamsche Land in which they. appeared, robs them of interest and even of
Any study of writing in an occupied country, for example, must take
into account the relations between authorized and clandestine publication
or broadcast, between the authorized Le Soir and the fugitive and manifold versions of La Libre Belgique or even more fleeting and forbidden
tracts, mimeographed sheets found in mail-boxes, slogans and orders
painted overnight on walls. Because, as Ernst Gombrich has observed of

9 Paul de Man, Wartime Journalism, 1939-1943, ed. Werner Hamacher, Neil Hertz,
and Thomas Keenan, Lincoln, Nebraska and London, University of Nebraska Press,
1988. Tue anti-semitic supplement, copied and published in full, is the exception to
this rule of deletion.


the relations of BBC and Nazi radio propaganda during the second world
war, the authorized (Nazi) source, the Gennan Horne Service, often had
to answer or deny reports broadcast by the BBC without admitting the
source or even the existence of such reportslO. To make such an admission would be to admit that many Gennans, despite the risks involved,
listened to the forbidden broadcasts from London, believed them, and
passed on what they had heard. In a similar way, Le Soir, which was
throughout the period of Paul de Man's employment the most important
authorized (that is, collaborationist) francophone newspaper in Belgium,
both masks and reveals its intertextual relations with clandestine reports
and inadmissible events. And these relations, passed over by the editors
in a silence that oddly repeats Le Soir's own attempt at concealment,
pervade his articles from the first.
To take just one example among many that show the need to read
Paul de Man the way he read others, as an allegorist, we might look
more closely at the way he reads accounts of Belgian and French defeats
and Gennan victories in the Blitzkrieg of 1940. There are several such
accounts reviewed in Le Soir, and the subject from the start troubles him,
evoking accusations of false interpretations and defences of the valor and
humanity of the Germans. Confronted by (or having chosen to review)
two opposed accounts of the campaign, de Man wannly approves one,
made up of articles that had already appeared in the Fascist paper Le
Pays reez11, that praises the courage and skill of the Belgian army, concluding that not they, but rather their French and British allies, had been
defeated, thus forcing the Belgians to capitulate. Against this de Man
sets and pillories the personal record (Ma deuxieme guerre) of a commandant Rousseaux, who fought from the statt of the brief campaign
until he was captured on the z5th of May. De Man writes: "Selon lui,
l'armee belge a manque a ses devoirs les plus elementaires; eile ne s'est
meme pas serieusement battue et a toujours re~u !'ordre de fuir avant
d'avoir vu l'ennemi de pres. De telles affinnations doivent etre combattues avec energie, si on ne veut pas laisser se divulguer de fausses inter10 Myth and Reality in German War-Time Broadcasts, London, Tue Athlone Press,
1970, pp. 10-13.
11 As the official organ of Uon Degrelle's Rexist party, Le Pays reel had been financed by both Mussolini (as early as 1936) and Hitler (since 1939), see J. GerardLibois and Jose Gotovitch. L'An 40: La Belgique occupee, Brussels, CRISP, N.D.,
pp. 33-36, hereafter cited as L'an 40. Tue article discussed here appeared in Le Soir
for 25 February 1941.


pretations des evenements". Setting aside for the moment the impertinence of this, in which the young annchair strategist refutes the old soldier's bitter experience, we should notice the imperatives and prohibitions of de Man's conclusion. Behind them lies, among other motives, an
attempt to refute the charge that was widely made against the anny, the
King, and the King's closest advisers, chief among them Henri de Man,
that they caved in and capitulated much too easily, againstthe conviction
of the government and the greater part of the population, that they might
still have stopped the Gennan advance as they had in 1914. Tue earlier
war was by the winter 1940/41 both a reproach and a call to resistance
for many Belgians. Even commandant Rousseaux's title, Ma deuxieme
guerre, called attention to this, as did the widespread charge that defeatists and Fascists had urged or ordered flight rather than the finn defence of 1914. De Man explicitly rejects Rousseaux's critical comparison
with "la bravoure de 1914-18", and there may be an even more unmentionable reason for bis rejection and uneasiness than those already given.
Evoking events and documents of the time may enable us to reconstruct
tliat untellable allegory.
Shortly before the 11 th of November 1940, despite strict Gennan
prohibitions against public assembly and any unauthorized spreading of
news or attempts to influence popular opinion, fliers and graffiti began to
appear all over Brussels, urging celebration of this memorial date, which
feil on a Monday, of the earlier war against Gennany12. Employers were
urged to give their workers the day off, shopkeepers to close, children
and students to stay away from classes, believers to go. to church, and
every one who could to pay respects to the monument to the Unknown
Soldier and the Monument aux Anglais. All of this was accompanied by
slogans ("Vive l'Angleterre! Vive la R.A.Ff') that called attention to the
fact that the war was by no means over and was in fact then going badly
for the. Axis powers. Thousands took part in what marked the beginning
of middle-class and urban opposition to the Gennan occupation and to
their Belgian collaborators. Parallel demonstrations took place in
Antwerp, Liege, and Verviers, but not in Flanders or in general among
fanners or workers. Taken by surprise, the Gennan military and police
struck back through arrests and threats made secretly to their Belgian
underlings. But once again the propagandist code translated this alann
into soothing and belittling reports to Gennany, in which von Falken12 I am indebted here. to the documented account in L 11.n 40, pp. 368-7 6.


hausen, the Wehnnacht commander, insisted that the demonstrations

were limited to the liberal bourgeoisie and bad not involved the masses.
lt was translated again by their Belgian subordinates into the terrorist
austerity of a proclamation that echoes the tone of de Man's article: "La
population comprendra que des incidents aussi regrettables que ceux qui
ont ete provoques le 11 novembre par quelques elements irresponsables
de la population sont de nature adesservir les interets les plux sacres du
pays"13. Notice that this sacred severity of tone imparts a message that
remains vague; first because to say what happened would be to contradict the proclamation itself, and second because official silence was
counterbalanced by widespread public witness and clandestine reports.
The allegorical text, both here and in de Man's articles, refuses to name
its subject.
To grasp what lies (in more than one sense) behind this refusal of
reference we have to bear in mind that propaganda, in this like irony or
satire, can only be read and understood in relation to a world of events
and discourse from which the text or proclamation deliberately diverges
or distances itself. Such different ways of "saying the thing that is not"
are also ways of not saying the thing that is, or is feared. Hence a third
translation of the events and clandestine accounts of the 11 th November,
so abusive that both events and their openly secret interpretation are
buried under insults. The author of this attack, which ap~ared in the
collaborationist pages of Le Nouveau journal on the 15 November,
1940, was its publisher Paul Colin: "Quelques bandes de braillards et de
snobinettes, applaudis par deux quarterons d'ecerveles, ont organise dans
les rues de Bruxelles des bousculades d'autant plus indecentes qu'elles
pretendaient s'inscrire sous le signe du patriotisme"14. In trying to read
this and similar propaganda we have to remind ourselves that even here a
code is at work. The dismissive "snobinettes", for example, may be read
as a shocked reaction to the role played in the demonstrations by lycee
and university students, the well-brought up daughters (and sons) of the
Brussels bourgeoisie 1s.
13 L'An 40, p. 375.
14 L'An40, p. 375.
15 Tue tone and choice of insults is reminiscent of the more recent attack by Louis de
Pauwels of Le Figaro on the lycee and university students who demonstrated in
Paris and throughout France against the policies of the Chirac government in late
1986, in which he notoriously suggested that they were "atteints d'un SIDA mental".
That insult in turn alludes to a right-wing code about disease, morality, and politics,


We will see how central this attack on the bourgeoisie is to the terrorist rhetoric of de Man's articles, but we might note that Colin seems
disturbed by appeals to patriotism. I detect this disturbance in the shift of
tone toward the end of his attack, for example, when abuse gives way to
more 9ecent language ("sous le signe du patriotisme"). Some of the historical reasons for this disturbance should now be clear to us, but there
were others that were more cultural and personal. Paul de Man was
writing in Le Soir for a Walloon (i.e. French-speaking Belgian) audience,
among whom his articles try to evoke sympathy and interest for both
Flemish and literature and culture. And yet that same Walloon
audience had already had several brutal reminders that it was thought to
be "ungennanic", pro-French, bourgeois, and even, apart from the Rexists, anti-Fascist. Flemish- and French-speaking soldiers, for example,
were not segregated within the Belgian army, but they were separated
the moment they surrendered to the Germ.ans. As soon as the capitulation
took place, Flemish soldiers were released and allowed to retum to their
homes. The Walloons were kept in prisoner-of-war camps for
years, many thousands of them until they were liberated by the Allies in
1944 or 1945. This partly racial discrimination was one of the two chief
causes for the early chilling of Belgian attitudes toward the Germans the other being the starvation of the civilian population - and Hitler
made no bones about this policy. Although the Fhrer had not by midJuly 1940 made up his mind about the future of the Belgian state, he ordered every possible advancement for the Flemings, but no favors whatever for the Walloons: "Den Wallonen sind keinerlei Vergnstigungen
zu gewhren''16. And yet Paul de Man, Aemish by origin and granted
further special privileges and rations by the Germ.ans, writes in French to
assure his Walloon audience that they are clearly better and better-off
than the decadent French, and are indeed living in the best of all possible
worlds. Even his attempts to mediate between cultures ran counter to policy. As soon as the Militrverwaltung took over in the summer of 1940, the distribution facilities of the Agence Dechenne (for
which de Man later worked) were used to divide the two cultural groups
by stopping all circulation of French-language publications in Aemish
but it backfired because the bourgeoisie of Paris, unlike that of wartime Brussels,
did not have to swallow such an insult, which was all the more offensive to the
many who, not knowing the code, took (or refused) it quite literally.
16 Hitler's wishes were expressed in a letter to the Army High Command from General
Keitel of 14 July 1940: it is reproduced inLl1n40, facing p. 201.


areas and to seal off Walloon areas by keeping out newspapers and magazines from France. All of this oppression and exclusion of his audience
had to be glossed or passed over in silence.
There is even the embarrassment of language itself. De Man's written
command of his native language, as Ortwin de Graef obseived in translating his articles from H et Vlaamsche Land, was rather shaky; often
clumsy and ungrammatical. To praise the language in which the praise is
written is meant tobe self-confirming. But if the praise is badly written
the result is self-subverting. So much of his zeal to promote and translate
the vlkisch qualities of Flemish/culture is undermined by both the manner and matter of the attempt. As in the voice that asserts in The Waste
Land, "Bin gar keine Russin, stamm' aus Litauen, echt deutsch", both
rhetoric and reference cast doubt on the assertion. Though not on the
voice itself, on its strange predicament.
This self-denying voice resonates through de Man's later work, at
once warning against alienation and aporia and yet arguing that it is inescapable. His plight was from the start a poignant if at times uncertain
echo of Michelet's lament in Le Peuple: he feit himself at one with the
people, but every attempt to express that unity confirrned his estrangement from a people whose language he could not speak. Tue way out of
that predicament was the transforrning power of the imagination as it
was conceived by the romantics. Tue Lyrical Ballads, read as both evidence and disproof of that power, remind the reader through Wordsworth's preface of this quandary of language, to which Coleridge later
retumed in his Biographia Literaria. The quandary remains, and one of
the reasons for Paul de Man's continuing significance is that he translates
it out of the primitive code of "La Terreur dans les lettres" into the
subtlety of philosophical concerns with language that reflects the undermining achievement of Nietzsche and Heidegger. His translation (or
transformation) of the terrorist code into a decoding process that undermines the referential force of literary texts - in the name of a deeper
grasp of their rhetorical power - transforrns the imagination into a sibyl
that promises to liberate us from the bafflement created by its riddles ...
by telling us even more riddles.
lt is striking that from the very beginning de Man wams against what
Paulhan describes as "le pouvoir des mots", never against their impotence: against a madness of words provoked rather than resolved by the
imagination. This seems even more striking when we re:tlect on his silence conceming this other role of the imagination, this very different

view of language. That other view emerges in Geoffrey Hill's sense of

the imagination as redemption or atonement for the guilt created by language, "an anxiety aboutfau.x pa.s, the perpetuation of 'howlers', grammatical solecisms, misstatements of fact, misquotations, improper attributions"17. Tue literary imagination accordingly atones for specific and
recognized barbarisms and mistakes, which may range from a slip of the
tongue to the Big Lie of propaganda. Hili even draws a parallel with a
severe judgment of Simone Weil that was itself based on her condemnation of wartime propaganda: "lt seems to me one of the indubitable signs
of Simone Weil's greatness as an ethical writer that she associates the act
of writing not with a generalized awareness of sin but with specific
crime, and proposes a system whereby 'anybody, no matter who, discovering an avoidable error in printed text or radio broadcast, would be entitled to bring an action before (special) courts' empowered to condemn a
convicted offender to prison or hard labour". Such punishments, and
worse,. were inflicted on wartime collabOrators ana propaganaists for the
crime of "avoidable error" in the use and abuse of language. But what
redemption is possible, what sentence must be served, if the error of language is unavoidable?

Our chains rattle, even while we are complaining of them.


The wartime articles in French, where the error of language can only be
countered by the asceticism of terror, attack literary style as a class
weapon and obstacle to a revolutionary Ordre Nouveau. In discussing an
article by Drieu la Rochelle in the Nouvelle Revue Fran~aise he had
usurped from Jean Paulhan, de Man quotes approvingly Drieu's conclusion that France had aged into abstraction and decadence because "La
civilisation fran~aise a cesse d'etre fondee sur le sens du corps"18. In this
the French are a contrast to the Gennans of the new Nazi Kultur praised
by Bertrand de Jouvenel in his book Apres Ia defaite, which. "foumit une

17 "Poetry as 'Menace' and 'Atonement'", in The Lords of Limit, New York, Oxford
University Press, 1984, p. 7.
18 Le Soir, 18 March 1941.


analyse tres pertinente de l'evolution morale qui mena les jeunes allemands a devenir les plus achames adversaires de la bourgeoisie
democratique triomphatrice de 1914-1918. Jusqu'a present, on avait trop
peu pris au serieux cette revolution interieure. Le fascisme etait considere comme une espece de folie passagere, tandis qu'il est au contraire
une reaction extremement nonnale et durable devant des circonstances
creees par la politique mondiale [... ]". lt was these revolutionary
pretensions of Nazism that de Man shared and that accounted in his mind
for the military triumph of Germany. And not only in his mind, because
although the "revolutionary" N azism of Rhm and the SA, which
Heidegger had also supported, had been crushed within the Reich, it was
still being touted at home and exported for possible belief abroad.
Gennan radio propaganda, as Gombrich has pointed out, observed a
discreet silence about the more loony intricacies of Nazi ideology and
concentrated instead on this simple opposition between youth and
senility, revolution and reaction: "There was no comparison possible, as
Goebbels wrote, between the first world war and the second, for now the
Gennan anny carried the spell of invincibility, being preceded by the
magic of a glorious revolution [... ] the slogan of the young nations was
sufficiently vague tobe flexible and sufficiently emotional tobe rousing,
and so the Gennan armies were made to march into France to the strains
of the Frankreich Lied 'We come and smash their old and corrupt world
to bits'"19.
The strangeness of de Man's task in celebrating Germany's victory
over France deepens when we realize that he is writing in French. lt is in
French that he condescendingly notes that "les Fran~ais ne se sont pas
encore habitues a l'idee que la creation de l'organisation mondiale nouvelle ne depend plus d'eux"20. And there is at least one moment when the
collision between French prose style and German conquest strikes him
forcibly enough to produce comment. Writing admiringly about Ernst
Jnger's war diary account of the invasion of France, he closes with
some bemused remarks about the translation, which he finds "too
perfect". The explanation of this paradoxical criticism is that the translation ends by giving the impression that the book had actually been written in French, "ce qui, surtout lorsqu'il raconte l'histoire d'un Allemand

19 Gombrich, Myth and Reality, p. 7.

20 Le Soir, 18 March 1941.


envahissant la France, a quelque chose d'etonnamment choquant"21. His

response to Jnger yields a further and final surprise. In its mythic vision, in its formal perfection, he finds what the literary terrorist seeks
beyond all the fiddle of style and subjectivity - which he here invidiously finds in Gide - and that is the text exalted to the status of universal
truth. As in other texts of this sort that come to mind, the truth on offer
may only manifest itself to those who have eyes to see and ears to hear.
In one of the articles mentioned already, on Brasillach's Notre avantguerre, de Man mentions Brasillach's fearful inability to grasp, at a Nazi
Nuremberg rally, "cette importance soudaine du politique dans la vie
d'un peuple" (12 August 1941). Just below this review is a brief and
cryptic announcement: "Radio-Bruxelles consacre une emission au depart de la Legion 'Wallonie"', which could be taken by contrast to show
that many Belgians grasped something Brasillach had missed. For the
"Legion Wallonie" was a volunteer corps of French-speaking Belgians,
Iater in 1943 incorporated into the Waffen-SS, who were being honored
as they set off on a road to the East that ended for so many in shame and
death. Their commander was to be the Rexist leader Leon Degrelle,
Hitler's favorite among the Belgian Fascists, and their last stand was to
be in Pomerania during the Spring of 1945. Of eight thousand men, only
seven hundred survived, many to face Belgian courts22. Their enthusiasm, a contrast to Brasillach's, is also a contrast between the Nazi
"nouvelle ere", the "new era" named and welcomed by de Man, and the
nostalgia for the old order (Notre avant-guerre) implied by Brasillach's
book and by his inability to enter into the spirit of Nazi spectacle.
Jacques Derrida reads this remark as possibly "overdetenninable" but
concludes that it criticizes the Nazis and distances de Man from Brasillach's Nazi sympathies. And yet the force of the judgment, clarified by
the texts that appear around it, is surely that de Man distances himself
precisely from Brasillach's aestheticism, his Alexandrian incomprehension of the "the life of a people"; in a word, Brasillach does not understand the cause for which he was shot.after the Liberation of France: he
was not Nazi enough. De Man's judgment seems to Derrida to hint at
irony, but it might better be read as a quite unironic declaration of the
21 Le Soir, 23 June 1942. The book reviewed was Jardins et routes, a translation by
Maurice Betz of Grten und Strassen.
22 See Hans Werner Neulen, Eurofaschismus und der Zweite Weltkrieg, Munich, Universitas Verlag, 1980, pp. 69-78, 154.


terrorist principle that stylish phrases and Byzantine refinement mask the
vlkisch truth of things.
Irony after all is a lie that subverts its own power to deceive. The paper Paul de Man wrote for was devoted to the opposite endeavor: the attempt to subvert the evidence of its readers' senses. lt had at all costs to
present an image or mosaic of a calm contented society united behind
humane Gennan authority. The problem, however, was that virtually every photo or paragraph that sought to convey that message collided with
a contradiction, often in the picture or prose itself. The children of veterans are shown above the Brasillach article being sent off by train to pass
"de delicieuses vacances" at Limbourg. And why? Because more than a
. year after the defeat of Belgium's anny many thousands of their fathers
are still in Gennan prisoner-of-war camps, leaving their families destitute. Although Belgium, like Holland and Norway, had been defeated,
her govemment went into exile in England and continued in a state of
war with the Third Reich. King Leopold III, advised by Henri de Man,
had ordered the surrender of the Belgian Anny, but there was no Vichy
govemment, no arguably legitimate collaborationist regime. There was
instead a military occupation by the Gennan army, a Militrverwaltung
backed by police forces and economic despotism that deported forced
labor to Germany, ransacked the country for money, machines and food,
starved the civilian population, and - through its large and powerful
Propaganda-Abteilung - took over newspapers such as Le Soir that then
put the best possible. face or mask on an ugly situation. lt should be remembered, as Belgian sources have pointed out, that the entire editorial
board of Le Soir, without exception, had refused to work for Nazi propaganda after the capitulation in 1940. All of them were fired, to be replaced by a more amenable group that included Paul de Man. From the
start, including his own opinion that henceforth collaboration was the
task of every reasonable person, Le Soir had somehow to balance fantasies against surreal actuality. lt had to balance reports of Petain's
speeches to boy scouts against police raids (razzias) to seize grain and
secretly slaughtered pigs; jokes about BBC broadcasts against the news
that French Jews were forbidden to have or listen to radios - all of these
hallucinatory contrasts can be found juxtaposed on the cropped and random photocopies of Le Soir.
The ugly face of military occupation required a language of euphemism and blandness. "La litterature


devant les evenements"

was how de Man avoided mentioning by name the shocks of defeat, be34

trayal, and resistance coverd by "the events" in this heading for one of
his articles. His language was definitely Aesopian, as a number of readers of the Le Soir essays have noticed. But the fable is less comforting
than it may seem. Through several articles and reviews on and about
books conceming the war, especially about the invasion of Belgium and
the fall of France, de Man downplays their interest and even regrets their
way of dwelling on the past. This softpedalling of battles and defeats was
part of German propaganda's concem that they might give rise to a desire for revenge - as indeed they did - but it was also I think a more personal concem of Paul de Man. He did not want to dwell on the "avantguerre" or the war itself because he was already looking ahead to the
"apres-guerre" that was meant to follow an early German victory. Any
event or any book that called attention to the fact that Belgium was still
formally, and to some extent actually, at war with the Thir~ Reich, and
that the hope of early victory, after which the social revolution could begin, had faded into an everwidening war that by the end of 1941
stretched around the world and grouped Russia, Britain, and the United
States against Germany and Italy, any such reminder increasingly dismayed him. Like his uncle Henri, whose protege he was, Paul de Man
had come to fascism from the left, and the puritanical tone he adopted
toward Brasillach and Montherlant was part of his rejection of the social
and cultural hierarchy they were part of and, in his eyes, tainted by. A
little mouming, as Hamlet reminds us, goes a long way, and a little too
much nostalgia is a dangerous thing. Hence his genuine and otherwise
surprising refusal of both the shining lacquer of style and the commonplaces of traditional politics, especially when they seem to be embedded
in an old order or a dangerous kind of recollection23.

23 Both of these refusals are part of his cultural terrorism, but they may also directly
echo some of the political slogans that were part of his daily life. Several articles
extol "European" values and ,,Tue defence of the Occident", phrases that were in
thernselves part of the liturgy of right-wing politics. During the weeks and months of
late 1941 and early 1942, during the very period of their recurrence in his articles,
these phrases had a further and pointed significance: they were meant to counter the
declaration by Churchill and Roosevelt, in the Atlantic Charter of August 1941, of
Allied .war aims and principles. This inspired Hitler to circulate the slogan of
"European solidarity" and summon a meeting of his own allies to proclaim a
"European New Order". lt was during these weeks that we find as well-placed a
source as Ciano, Mussolini's Foreign Minister and son-in-law, writing from Hitler's
headquarters, "Now the fashionable slogan is that of 'European solidarity'. Europe


This brings us to the second of de Man's surprising attacks, a review

of Montherlant's Solstice de juin - like Brasillach's Notre avant-guerre a
book that touched all the neives we have just laid bare. lt is exactly the
kind of text that Denida once identified in Positions as "embarrassing",
and therefore one most likely to open itself to an interesting reading.
Derrida here notices that there is something incidentally odd about de
Man's definition of language, quite apart from his even stranger rejection
of Montherlant's style. Language is defined along with manners and
customs, as "material and direct", a token of the terrorist reduction of
language to instrumentality. Tue heart of his attack on Montherlant is
really an attack on literary language itself and on writers in general as
empty stylists: "parce que les litterateurs sont capables d'exprimer des
lieux communs avec elegance, on en fait des oracles et on ecoute leur parole comme un message providentiel [... ] On est etonne de la naivete et
de la nullite de certaines de leurs sentences lorsqu'on les depouille du
vemis brillant qu'une forme soignee leur confre" (11 Nov. 1941). lt
would be hard to find a more unvamished example of Paulhan's terrorism, which he characterizes by precisely this" rupture avec le lieu commun".
lt is also a crude prevision of the subtle insistence in de Man's later
work on the separability of figurative language and cognitive value. This
insistence, as a later title and argument like "the epistemology of
metaphor" serves to show, is anchored in the notion that meaning is the
standard of language. And meaning, even here, is unveiled and shown to
be empty. Tue political roots of this terrorism, its technological shoving
of style aside as a kind of decadence, are visible in his contempt for writers who have not either gone to the people or gotten down to the hard
work of finding solutions and remedies for their "veritables carences":
"C'est f l'oeuvre des specialistes qualifies et non pas de dilettantes
touche-a-tout qui n'ont pour eux que leurs facilites oratoires ou la qualite
de leur style". Tue philistinism of this could hardly be closer to the terrorist revulsion from style, hardly further removed from any sharing in
an aesthetic ideology. De Man prefaces all of this by tuming against
Montherlant a poigant image that the author of Solstice de juin applied to
all those who bad written about the war and its consequences: "Aux
ecrivains qui ont trop donne, depuis quelques mois, a l'actualite, je
- the Fhrer said - besides being a geographical expression is a cultural and moral
concept". See Ciano's Diplomatie Papers (London, 1948), p. 459.


predis, pour cette partie de leur oeuvre, l'oubli le plus total. Les journaux,
les revues d'aujourd'hui, quand je les ouvre, j'entends rouler sur eux
l'indifference de l'avenir, comme on entend le bruit de la mer quand on
porte a l'oreille certains coquillages". De Man hails this as a "just and
severe sentence", a phrase that echoes the austerity of as Saint-Just, and
then inflicts it on Solstice de juin itself. Such a consignment to oblivion
evokes Nietzsche, just as the figure of the seashell's rumor of the sea
evokes Wordsworth's story of the shell and the stone, and difference
between the two, the sentence and the simile, shows us some of the difference between the writer and his reviewer.
This contrast seems all the sharper if we notice in closing how it all
seemed to yet another contemporary, one who reviewed both Montherlant's book and Les Fleurs des Tarbes. This was Maurice Blanchot, who
relished in Solstice de juin all of the romantic irony that de Man detested.
And in his extended essay on Paulhan he writes the missing second part,
the missing rhetoric that even when wrtnen is still missing, because if
taken to its conclusion, Paulhan's vision of terror becomes indistinguishable from literature itself. Even in his lighthearted essay on Montherlant,
whose title "De l'insolence consideree comme un des Beaux-Arts" plays
on De Quincey's "Murder Considered as One of the Fine Arts", Blanchot
catches what remains an ominous paradox of style - "On prend une attitude, mais on prend l'attitude de ce qu'on est reellement"24. When he
tums to Paulhan, he traces over the book that is offered as an anatomy of
terror another book that anatomizes and dissects literature, calling even
its existence in doubt. After reading them all - Paulhan, de Man, Blanchot - and allowing for all the differences, we are left in closing with the
recognition that all three are caught, late or soon, by the laws of literary
terror. And Paul de Man, who in his early writings tried to dispossess
language, went on tobe haunted by it. Flight and pursuit become undecidable, escape hardly possible. All writers share in acts of terror, there
are only unusual suspects.

24 Both essays, the second of which is entitled "Comment la litterature est-elle possible", are found in Faux Pas, Paris, Gallimard, 1943.



Paul de Mans Kriegsjournalismus hat zu einer groen Debatte gefhrt, die aber nur selten
im Zusammenhang mit seiner ambivalenten Haltung gegenber der Rhetorik untersucht
wurde. Diese Ambivalenz geht zurck auf seine stetige Bewunderung fr Jean Paulhan,
dem Herausgeber der Nouvelle revuefrtm:faise, und auf dessen Ablehnung der Rhetorik,
wie sie klar und deutlich in Les Fleurs de:Tarbes, ou la terreur dans les lettres (1936, 2.
erweiterte Auflage 1941) formuliert wurden. Nach Jean Paulhan drckt sich der literarische Terror - ein Ausdruck, der auf die soziale Bedeutung der Terreur zur Zeit der
Franzsischen Revolution verweist - durch zwei offenbar entgegengesetzte Haltungen
gegenber der literarischen Sprache und ijrrer Rhetorik aus. Rhetorik wird einerseits als
eine billige Sammlung von lieux communS verachtet und soll vom Autor vermieden werden; dieser soll ber das Sublime, das er in der Literatur entdeckt, berichten. Beide Haltungen lassen sich bis auf de Mans frhe! Schriften zurckverfolgen. Daraus ergibt sich
die oft schwer zu verstehende Doppelhaltung: seine Angriffe auf das, was de Man den
Stil des Mandarin nennt (wie ihn Henri de Montherlant und Robert Brasillach anwenden), aber auch seine ernstgemeinte Aufforderung zum Opfer im Sinne der Revolution
und sein gleichzeitiges Lob fr den be~schwenglichen Stil von Ernst Jnger. All das
geistert in de Mans spterem paradoxen Apsatz zur Definition der Rhetorik mit, wird von
seiner aesopischen und verneinenden Spr~che in bezug auf Kollaborationsschriften sogar
noch unterstrichen. Beim nheren Hinschauen auf das, was man "Umtext/Umfeld" der
Zeitung Le Soir nennen mchte, aber auch seine eigenen Artikel in diesem Blatt, machen
deutlich, wie aktiv er in die Rolle eines '"Jierroristen" geschlpft ist. Trotzdem zeigt sich,
da, was Maurice Blanchot als Paradoxon von Paulhan bezeichnet hat, auch auf Paul de
Man anwendbar ist: vor der Rhetorik zu fliehen, heit von ihr verfolgt zu werden, und
die Aufhebung der Sprache bedeutet die A~fuebung der Literatur schlechthin.


Ortwin de Graef



Contrary to common belief, literature is not the place where the unstable epistemology of metaphor is suspended by aesthetic pleasure, although this attempt
is a constitutive moment of its system.
Paul de Man, "Tue Epistemology of Metaphor"
En dat voert ons tot een der diepste problemen van de kennis in het algemeen,
probleem dat wij hier niet kunnen stellen.
Paul de Man, ..Literatuur en sociologie.. l

Paul de Man and Ernst Jnger: what of this "and"? If it may still seem
slightly eccentric to center an essay on de Man on a juxtaposition of his
name and that of a writer whose name is never mentioned in what we
might still call, all too hastily, but not therefore illegitimately, his critical
writings proper, the reflection on that possible eccentricity is not, nor
does it have to be. What of this "and"., then? Is it the "and" of, for instance, "Madame de Stael et Jean-Jacques Rousseau", "Wordsworth und
Hlderlin", "Keats and Hlderlin" - that is to say, an "and", "und", or
"et" suggesting an exercise in comparative literature particularly tuned at least initially - to the identification and interpretation of correspondences where the very project of comparative literature (or comparative
ideology) has already ordered us to expect them? Or is it the perhaps
rather more wilful "negative" version of the generic conjunction of comparative reading, announcing an investigation of differences rather than
convergences, at least to the informed reader - in an essay, say, on "Kant
and Schiller"? Or is this "and" perhaps really the apparently more enigmatic "with", the "with" of "Kant avec Sade" (or the reverse), for instance? And is it at all possible to uphold such intuitive differentiations
between various modes of conjunction?

"And this leads to one of the deepest problems of knowledge in general, a problem
we cannot pose here" (Het Vlaamsche Land, 27-28 September 1942).


In the conception of this essay, these heavily programmed questions

have figured - and still figure - prominently. The juxta-position suggested itself, of course, in the course of a study of Paul de Man's wartime
joumalism; more precisely, in response to the unmistakable emphasis on
the epochal exemplariness of Jnger's work that marks this joumalism2.

From March 1942 through March 1943, 14 texts by de Man mention or review
Jnger's work: 2 reviews of average length of (translations of) Auf den Marmorklippen (M) and Grten und Strassen (GS); one short essay on Jnger's work in general;
2 more one-paragraph reviews each of M and GS, plus another one-paragraph review of Das Abenteuerliche Herz (AH); and finally, six mentions of Jnger in different contexts (usually as contrasting example of how literature should be written). All
these texts are available in Paul de Man, Wartime Journalism, 1939-1943, ed. by
Werner Hamacher, Neil Hertz and Thomas Keenan, Lincoln, University of Nebraska
Press, 1988. For convenient reference, I will here list these 14 pieces in their order
of appearance in the above volume (i.e. according to source and date) - the texts will
be identified in the body of this paper by means of the abbreviation WJ, plus page

Le Soir
"Chronique litteraire. 'Sur les Falaises de Marbre', par Ernst Jnger. Deux ouvrages d'actualite", 31March1942 (WJ, pp. 216-17).
"Les possibilites du recit de guerre: 'Le Chemin des Errants' par Louis Fonsny",
30 April 1942 (WJ, pp. 228-229).
"Chronique litteraire. 'Jardins et Routes', par Ernst Jnger", 23 June 1942 (WJ,
pp. 244-45).
"Chronique litteraire, Aspects de la pensee allemande: 'Le Livre du souvenir',
de R. Kassner", 15 September 1942 (WJ, p. 269).
"Chronique litteraire. Apropos d'un concours liueraire", 22 Septembre 1942
(WJ, p. 270).
"Chronique litteraire. Technique du roman: 'Cadavre exquis', par Louis Carette.
- 'L'Herbe qui tremble', par Paul Willems", 24October1942 (WJ, pp. 275-76).
H et Vlaamsche Land
"Duitsche Letteren. Een groot schrijver: Ernst Jnger", 26-27 July 1942 (WJ,
pp. 319-20; trans. pp. 321-22).
"Menschen en Boeken. Blik op de huidige Duitsche roman-literatuur", 20 August 1942 (WJ, pp. 323-24; trans. pp. 325-26).
"Literatuur en sociologie", 27-28 September 1942 (WJ, p. 331; trans. p. 332).
Bibliographie Dechenn.e
"Jnger, Ernst: Sur les Falaises de Marbre", May 1942 (WJ, p. 349).
"Jnger, Ernst: Jardins et Routes (Pages de Journal 1939-1940)", July 1942
(WJ, p. 361).
"Jnger, Ernst: Op de Marmerklippen", January 1943 (WJ, p. 376; trans. p.
"Jnger, Ernst: Le Coeur aventureux", January 1943 (WJ, p. 376).
"Jnger, Ernst: Grten und Strassen", March 1943 (WJ, p. 384).


But once this suggestion was made, it proved tobe far from clear exactly
how one should go about pursuing it, even while it became more and
more evident that such a pursuit was necessary. The promise the conjunction contained could not be seriously dismissed, but neither could it
be easily circumscribed - and this, presumably, not exclusively due to
the bad infinity of comparability which everyone engaged in the practice
of comparative reading is more than familiar with.
In the present essay, the systematic articulation of the possibilities of
the conjunctional promise will only receive a very preliminary treatment.
The best part of our attention will be devoted to an attempt to trace the
contours, in Jnger's work, of a particularly powerful version of what de
Man in his later work explicitly and critically began to address as "the
aesthetic ideology". Needless to say that this selective reading cannot
claim to adequately render a profile of the entirety of Jnger's massive
(and massively repetitive) work - but the profile that will be traced
should invite renewed recourse to that work in order to question (it is
hoped) the validity of my insistent implicit claim that the monumental
"wrongness" of Jnger's aesthetics has an obvious, important, complex
and negative rapport with de Man's rhetorical reading. Only towards the
end of this reading shall I explicitly suggest some contrapuntal resonances of this aesthetic ideology with de Man's critical undertaking.
These suggestions, it should perhaps be added, will hardly come as a
surprise-which is precisely why they should be prepared for.


After the completion of a first version of the present essay, the companion volume to
Wartime Journalism was published: Werner Hamacher, Neil Hertz and Thomas
Keenan (eds.), Responses: On Paul de Man's Wartime Journalism, Lincoln, University of Nebraska Press, 1989. A number of the essays in this volume contain remarks
or reflections on Jnger that may be useful here - e.g. Ian Balfour, "'Difficult Reading', De Man's Itineraries" (pp. 6-20; p. 9); Cynthia Chase, "Trappings of an Education" (pp. 44-79; p. 50); S. Heidi Krueger, "Opting to Know" (pp. 298-313; passim);
and Allan Stoekl, "De Man and Guilt" (pp. 375-85; pp. 376-77, p. 384). See also, in
the same volume, Rodolphe Gasche's perhaps slightly over-confident but nonetheless arguable observation that Jnger, of all the German writers de Man reviewed,
was "the only one he admired" ("Edges of Understanding", pp. 208-20; p. 220).


"Mach kei Loch in de Natur" -

Georg Bchner

One of de Man's first references to Jnger, in a review of the French

translation of the first of his Second World War diaries, Grten und
Strassen, offers a convenient passage through which we can enter our
matter. In this review, de Man casts Jnger as a universal nature, whose
principal goal it is to merge (se confondre) with the world surrounding
him, to find equilibrium and harmony in a complete participation in nature. To illustrate this point, he then refers to a passage in which Jnger
records what de Man calls his extase in the face of a shell:
Objet pur, sorti de la nuit des temps, d'une fmesse de structure parfaite - rien ne
pouvait parler plus directement ason etre profond. [... ] c'est toute la magie de la
creation que Jnger y retrouve, dans tout ce qu'elle a d'ephemere et de
merveilleux. (WJ, p. 244)

Tue passage in question - dated June 19th 1940- is densely suggestive.

Walking through the badly damaged park of the Castle of Montmirail,
where he and his staff have taken temporary residence at that stage of the
French campaign, Jnger, in search of a souvenir (Andenken), finds the
petrified shell of a Wendelschnecke (a wentletrap), "[ ... ] der durch die
Explosion der Tiefe des Kreidegrundes entrissen war." His diary then
records, as usual, his reading of the object, of this "persnliches
Geschenk" - a la recherche du f ond pierreux:
Sogleich begriff ich, den die Nhe der Vernichtung in diesen schnen Tagen
doch gleich einem Schatten begleitet hatte, die Belehrung: der Mont Mirail war
einst eine Klippe im Kreidemeer, und unzerstrbar ruht das Wunderbare in ihm,
von dem auch dieses Schloss mit seinen Grten nur eine Bildung, ein flchtiges
Gleichnis wie das Gehuse dieser Muschel ist3.

Notwithstanding its obvious and specific stereotypicality (and we shall

have the occasion to judge this qualification more accurately later on),


Ernst Jnger, Grten und Strassen (first published 1942), in: id., Werke. Band 2:
Tagebcher II, Strahlungen, Erster Teil, Stuttgart, Klett, s.d. [1963], pp. 25-230; p.
187. All quotes from Jnger in the present essay are from this edition of the Werke.
Henceforth, references to Grten and Strassen will be given in the body of the text
(GS, plus page number).

this lesson, this Belehrung, deseives to be spelled out attentively. Tue

maivellous or wondrous (das Wunderbare) is that which is unzerstrbar,
that which, in the proximity of annihilation, challenges and foils that annihilation - but the lesson itself is only arrived at in die Nhe der Vernichtung, as a result of an explosion that has violently ripped out the lesson's object from the inner depth of the cliff. lt is important to note that
the lesson does not teach tliat the shell is unzerstrbar, or the cliff, but
only that das Wunderbare, which reposes in the cliff and which is intimated to the obseiver through the shell, is indestructible. In addition to
this, the Belehrung Jnger so instantly grasps also identifies its own generative principle - alluding, perhaps, to the Chorus Mysticus in Faust as a process of moulding; the shell is "nur eine Bildung ein flchtiges
Gleichnis" of the indestructible, as are, importantly, the castle and its
gardens, which, in this respect, resemble (wie) the "house" of the mollusc. Tue didactic purport of the passage, then, is to drive home a concatenation that runs from the woundrous over the natural object to the
cultural object via the fonnative process of Bildung as it is contained
(begriffen) by the subject.
We have underscored that this insight into a universal figuration this analogia (or genealogia) entis4 - is arrived at the expense of consid4

"Insofern es zu den Aufgaben des Verstandes gehrt, die Dinge nach ihrer Verwandtschaft zu ordnen, zeigt sich der kombinatorische Schluss dadurch berlegen,
dass er die Genealogie der Dinge beherrscht und ihre hnlichkeit in der Tiefe
aufzuspren weiss. Der einfache Schluss dagegen sieht sich auf die Feststellung der
Oberflchenhnlichkeit beschrnkt und plagt sich damit ab, am Stammbaum der
Dinge die Bltter zu messen, deren Grundmass jedoch im Keimpunkt der Wurzel
verborgen liegt." Jnger, Das Abenteuerliche Herz. Zweite Fassung: Figuren und
Capriccios (first published 1938), in: Werke. Band 7: Essays III, Das Abenteuerliche
Herz, pp. 177-338; pp. 195-96. Henceforth, this work will be referred to in the body
of the text (AH2, plus page number). Cf. also, of course, and running ahead a little:
"Ainsi, toute la Philosophie est comme un arbre, dont les racines sont la Metaphysique, le tronc est la Physique, et les branches qui sortent de ce tronc sont toutes
les autres sciences [ ...]", Martin Heidegger, "Einleitung zu 'Was ist Metaphysik?':
Der Rckgang in den Grund der Metaphysik", in: id., Wegmarken, Frankfurt am
Main, Klostermann, 1978, pp. 361-77, pp. 36LFor a succinct commentary on the
opposition between the kombinatorische and the einfache Schluss in Jnger, more
particularly as these relate to the relation of science to metaphysics, see Christian
Graf von Krockow, Die Entscheidung: Eine Untersuchung ber Ernst Jnger, Carl
Schmitt, Martin Heidegger, Stuttgart, Enke, 1958, pp. 110-12. Jnger's own perception of this relation in the short essay "Der kombinatorische Schluss" which we have
just quoted from, deserves tobe mentioned here: "Wo es aber dem Genius auch immer beliebt, das Feld der Wissenschaften zu betreten, da liefert er den Leuten vom


erable violence. Still, if this violence were "only" (and this adverbial
slight is of crucial significance) the empirical violence of an explosion,
of a real bomb which, as a matter of historical, accidental fact, did conjure up the wentletrap from the interior of the Mount Mirail, it could be
argued that the Belehrung of das Wunderbareis in essence independent
from this violence, that the insight into the tropological unity of the universe - for that is what Jnger adumbrates here, in the shadow of annihilation- is only accidentally, only in this particular and contingent case,
due to the exercise of violence. A closer examination of Jnger's aesthetics, however, can reveal that the link between violence and understanding (Begriff) as it operates in the most confidently articulated parts
of that aesthetics is by no means as tenuous as such an interpretation
would suggest. (But prior to embarking upon such an examination, we
should perhaps already point out that the easily discernible fact that this
examination will remain implicated in the very tropological processes it
seeks to uncover, while forcing us to remain alert, cannot in itself invalidate this examination's opposition to that tropology. What concerns us is
the tropological thrust of Jnger's aesthetics, and it is one of the principal
features of that thrust that it does not exhaust itself explicitly in the
recorded actualisation of its potential, which is also why it appears to
succeed in getting away with just about everything (in fact, that is exactly what it is intended to do). To read this tropology, therefore, is to
participate, up to a point, in its impetus, to be gathered in its momentum,
and it is only in this complicity with its programme that the import of the
mechanics of these aesthetics can be measured with a margin of adequacy. This measurement, as I will try to demonstrate, is what will allow
us to gauge some of the quasi-detenninative processes connecting the
aesthetic ideology with its critique5.)


Fach ein kurzes, entscheidendes Gefecht, indem er sie, die gewissermassen in der
geraden Linie gegen ihn anrcken, mit Leichtigkeit zu berflgeln und aus den
Flanken zu erschttern vermag. Am schnsten und schnellsten tritt seine berlegenheit in der Kriegskunst hervor." (AH2, p. 195) This realisation of truth's metaphorical army of tropes into the "real" army of the (military) Genius obeys a principle
which we shall try to demonstrate as fundamental to Jnger's thought.
One of the crucial problems here is the fragmentary nature of Jnger's work, which
for the most part consists of diaries and short essays. While fragmentariness is in itself by no means incompatible with systematicity (see, for instance, Rodolphe
Gasehe, The Tain of the Mirror: Derrida and the Philosophy of Rejlection, Cambridge, Harvard University Press, 1986, e.g. pp. 178ff.), Jnger's fragmentary exposition does contain incompatibilities which may not be "merely" blind moments of

Es lag in diesem Funde auch ein Zug von Alchimie - vom Stein der Weisen,
der die Dinge verwandelt durch ein Wunder, das sich in unserm Innern vollzieht. So wre mir, auch wenn man mich aufgefordert htte, mir unter den alten
Bildern, Bchern oder Schtzen des Schlosses ein Erinnerungsstck zu whlen,
doch nichts begehrenswerter erschienen als dieses Schneckenhaus. Wir mssen
einen Stand erreichen, wie er dem berfluss der Erde angemessen ist, in dem
sich in Gold verwandelt, was unsere Hand berhrt. (GS, p. 187)

At first sight, we appear to be in a very familiar plea for modesty and renunciation here6: those who have grasped the lesson have no need for literal wealth, as the lesson teaches them, precisely, that such literal wealth
is only the ephemeral figure of the real wealth resting in everything their
hands can touch (and not, it should be recalled, although this particular
passage does not emphasize this, as yet identical to everything their
hands can touch, for what is touched is itself, too, but a Bildung of the
wealth)7. Yet, the sad tale of King Midas the passage also alludes to already traces a rather less benign reading of this same doctrine of transformation (Verwandlung: the Wendelschnecke is a suspiciously appropriate object of this demonstrative didactics; as appropriate, in fact, as
Jnger weil realises, as the setting of the lesson - Mount Marvel), and
the continuation of the diary entry responds to that trace with implacable,
be it unacknowledged, rigour.

the potential system itself, but may point to a different mode of thinking that prefigures the critique of the system. Such incompatibilities, however, and this is the fundamental but strategic flaw of the following exposition, will not be rigourously investigated here.
Similar passages occur throughout Jnger's work: thus, for instance, the narrator of
Auf den Marmorklippen: "Wenn wir zufrieden sind, gengen unseren Sinnen auch
die kargsten Spenden dieser Welt. Von jeher hatte ich das Pflanzenreich verehrt und
seinen Wundem in vielen Wanderjahren nachgesprt. Und wohl war mir der Augenblick vertraut, in dem der Herzschlag stockt, wenn wir in der Entfaltung der
Geheimnisse erahnen, die jedes Samenkorn in sich verbirgt." Jnger, Au/den Marmorklippen (first published 1939), in: Werke. Band 9: Erzhlende Schriften, pp. 187298; p. 199.
A survey of the etymology of the word "weal(th)", the literal meaning of which
"now" is a metonymy for its "original" meaning ("the state of being well") could
bear this lesson out; a critical reading of such a survey could then question the metaphysics of the trope/lesson (see, as always, Jacques Derrida, "La mythologie
blanche: la metaphore dans le texte philosophique", in: id., Marges de la philosophie, Paris, Minuit, 1972, pp. 247-324.


Tue reading of the shell concluded with the reflection we have just
quoted, and the diary then goes on to register another event that occurred
later on the same day. At sunset, Jnger and his men set out to go
hunting in the valley. They chance upon a big and prosperous-looking
farm, whose owners appear to have fled, having left behind "a dark
gnome" to tend to the cattle. Tue encounter with this singular (and totally
inebriated) character soon modulates from the comical into the eerie:
Dann schien ein Freudenrausch sich seiner zu bemchtigen, er fasste den Herd,
die Sthle und Tische, ja selbst die Wnde an: "Das gehrt mir, das auch, das
auch'', und endlich - ich fand die Steigerung nicht bel - zeigte er noch eine
zerschlissene Mtze vor: "Das gehrt auch mir, alles gehrt mir." Dann wieder,
ein langes Messer wetzend und sich scheu umsehend: "Die Patronin ist nicht
da, wartet, ich werde euch eine Gans abschlachten, nicht das Weibchen, weil es
jetzt Junge hat, sondern das Mnnchen, das schmeckt sehr gut. Alles gehrt
jetzt mir." Zuletzt wurde uns grausig zumute, und wir verliessen den Ort. Der
Gnom verfolgte uns noch eine Weile um uns zurckzurufen; wir sahen, wie er
Hhner ergriff und in die Luft schleuderte, so dass sie kreischend davonflogen.
(GS, p. 188)

This time there appears to be no Belehrung to be gained - at least none

that the text cares to explicitate. Tue juxtaposition of this little narrative
with that of the wentletrap does, however, prompt some remarkable (but
not therefore extra-ordinary) propositions. Thus, it could be argued that
what is grausig in this scene is not so much the grotesque quality of the
scene it refers to but rather the fact that it allows for a peculiarly consistent specular identification of classical Unheimlichkeit. As a rule,
Jnger is anything but slow in elaborating such (specular) potential, and
the fact that he does not do so here is perhaps not insignificant: the possible parable here is indeed very different from the reassuring little Bildungsgeschichte the shell had in store. Even on a first simple referential
level, it could already be remarked that the analogy suggesting that the
violence of war has procured Hauptmann Jnger possession of Montmirail in much the same way as it has procured the dark gnome possession
of the farm is considerably less flattering to this former than the wentletrap's praise for his noble renunciation of literal wealth. A more interesting correspondence, which elaborates on this contrast, is yielded in
the gnome's somewhat absurd pride in his worn cap, in the Steigerung
which Jnger tellingly judges tobe "nicht bel": for just like the wentletrap is transformed into gold by the touch of the initiate's hand, so the

cap the gnome imperiously holds up for inspection has become an object

of proud possession - the pattem that leads from the paintings, the
books, and the treasures of Montmirail to the petrified shell in the first
narrative is accurately mirrored in the Steigerung from hearth, chairs, tables and walls to "eine zerschlissene Mtze". The affective difference in
Angemessenheit between the gnome and the Hauptmann, between the
inner Wunder and the overpowering Freudenrausch, between berhren
and anfassen, or begreifen and ergreifen, cannot suppress the truly
gnomic and by far more fundamental Belehrung of the later passage:
"Das gehrt auch mir, alles gehrt mir" - or, and more crucially, "Das
gehorcht mir, alles gehorcht mir": everything listens to me, obeys my
voice, my touch, my gaze, my Geist8. Angemessenheit, it should be recalled, can work either way; more accurately: it works one way only, according to the measure of obedience enforced by the alchemy of possession and property. Let us repeat this by tuming to another trope.

* * *
("La pensee reve des choses et d'elle-meme dans un miroir Oll les images la
representent teile qu'elle est dans son froid sommeil. Elle se donne un regard
qui juxtapose a la chose sa signification et a cette signification une nouvelle
realite qui en est comme le prolongement dans le neant. L'au-def de l'esprit est
figure par ce monde qui nous est ouvert." Maurice Blanchot, "Une oeuvre d'Emst Jnger")

In a fragment of Das Abenteuerliche Herz, a diary of ideas which de

Man also reviewed, Jnger introduces an image that is interestingly similar to that of the wentletrap: "In den Sammlungen des Leipziger mineralogischen Instituts sah ich einen fusshohen Bergkristall, der bei der
Tunnelbohrung aus dem innersten Stock des Sankt Gotthard gebrochen
war- einen sehr einsamen und exklusiven Traum der Materie." (AH2, p.
189)9. To all appearances, the image functions as the objective correla8

And, perhaps, proleptically, my (Jnger's) name: cf. the section in the bibliography
of Jnger's work that is devoted to "Ernst Jnger zubenannte Insekten, Schmetterlinge, Molllusken und ein Moor" (Hans Peter des Coudres & Horst Mhleisen, Bibliographie der Werke Ernst Jngers, Stuttgart, Cotta, 1985, pp. 135-36). Particularly
interesting here is the (modest) wentletrap-like Chilostoma (Cingulifera) cingulatum
juengeri which Jnger discovered/described/claimed in 1983.
Tue translation de Man reviewd is presumably based on the second version of the
book (see note 3) which Jnger published in 1938. Tue first version was published
in 1929 under a different subtitle (Aufzeichnungen bei Tag und Nacht); it is reprinted


tive of an immediately preceding thesis which is aimed, with some little

scorn, at Swedenborg's condemnation of those who jealously hoard their
dreams and cognitions ("Trume und Erkenntnisse"):
Wie aber ist es mit der Verachtung des Geistes davor, sich auszumnzen und in
Kurs zu bringen- mit seiner aristokratischen Abgeschlossenheit in den Zauberschlssern Ariosts? Das Unaussprechliche entwrdigt sich, indem es sich
ausspricht und mitteilsam macht; es gleicht dem Golde, das man mit Kupfer
versetzen muss, wenn man es kursfhig machen will. (AH2, p. 189)

lt will be noticed that the appropriateness and the power of the image of
the crystal (its private, exclusive answer to Jnger's question: "So ist es
mit der Verachtung des Geistes") hinges on its hiddenness, which is also
its purity; but it is not difficult to bring to light the paradox this entails.
For the image, because it functions as an image, has already been
"contamined", in that it has been made kursfhig, in that it has been extracted from the desired seclusion of the ineffable and has been uttered
as image - as an image, moreover, of "something" which had already
suffered an equally incongruous circulation before, as (unalloyed) gold.
Tue point being that if the ineffable were to be allowed for in its ineffability, it would not be relayed through a tropological chain which, moreover, pretends to denounce just such a tropological subsumption. Tue
crystal, through the alchemy of Jnger's discourse, has been turned into
gold and is thus already versetzt with copper, gesetzt as image, and
entsetzt of its purity, Ch71d this because, as an image, it obeys, and is possessed by, the pre-existent universalising discursive intent of the antidiscursive principle it serves to imagine. Jnger's reluctance to establish
the crystal explicitly as an image of his thesis may perhaps signal an
awareness of the paradox, but it by no means resolves it. In fact, the
paradox is even more deeply embedded in the logic of the passage than

alongside the second version in the same volume of the Werke (pp. 25-176). As a
rule, we shall refer to the second version only - otherwise the reference will be
identified as AHJ. For an admirable appraisal of the numerous and substantial differences between both versions (which exceeds our present scope), see Karl Heinz
Bohrer, Die sthetik des Schreckens: Die pessimistische Romantik und Ernst
Jngers Frhwerk, Mnchen, Hanser, 1978 (succinctly, for instance, on pp. 264-66,
269 and 424ff.).


we have suggested: indeed, the text does not rest satisfied in its impossible troping of the unfigurable, it also tropes the "carrier" of this untenable image beyond its limits, thus admirably crystallising its own inconsistency.
In the first version of this fragment, Jnger posed a question which
he deleted from the second version, and which goes to the core of the
problem the passage slips into: "Welche Sprache ist frei vom Arbeitsgeruche des Gefhlstransports?" (AHJ, p. 33). Tue answer can certainly
not be the language in which the crystal is cast: the image, in effect, is
only this Gefhlstransport. Tue wishful projection of the allegedly ineffable products of the sensitive and aristocratic Geist into the crystal occurs by means of the classical crossing of properties which endows matter with consciousness: matter ''dreams" the crystal in loneliness and exclusivity, just like mind produces its "Trume und Erkenntnisse" in
aristocratic seclusion. In one and the same exchange, the "products" of
the mind mimick the solidity of material objects, while these latter receive the properties of dreams and cognitions - or at least, that is what
the chiasm attempts to installlO. Under closer scrutiny, however, the pat10 This chiasmic reversal is one of the key tropes in most (of Jnger's) writings. A
striking example occurs in another piece in Das Abenteuerliche Herz, "Fortuna's
Unkraut". Jnger has just described a particular "argument" that visited him like a
flash in the course of a dream about agame of cards: "Durch solche Bilder leuchtet
uns zuweilen ein, dass es eine. besondere Art, vielleicht eine Kurzschrift, des
Denkens gibt, die das Element der hnlichkeiten und Anklnge im Grunde erfasst
und spielend beherrscht. Da gengt uns der Klang eines Wortes, eine unbekannte
Sprache zu verstehen. In die harmonische Ordnung einbezogen, wandelt sich der
erstbeste Gegenstand, den wir erblicken, zum Universalschlssel um. I Dies und
nichts anders begrndet auch den eigentlichen Reiz aller Glcksspiele. Die rote
Serie gibt dem Spieler mehr als Geld; sie schenkt ihm jenen Glauben, dessen wir im
Innersten bedrfen - nmlich mit der Welt verschworen, mit ihr im Einverstndnis
zu sein. Wenn die Kugel fr uns rollt, das Blatt sich fr uns wendet, kosten wir
einen erlesenen Genuss - den Genuss einer geheimsten, materiellen Intelligenz. In
der Tat ist das Glck nichts anderes als die Elementarform der Intelligenz - im
Glck denken die Dinge, denkt die Welt f.r uns mit." (AH2, pp. 241-42; emphasis
added). The apparent oscillation in this passage, between a (possibly mistaken)
belief (Glauben) and a certainty of a present or future harmony (harmonische
Ordnung, "In der Tat ... "), is highly typical of Jnger's musings and can usually be
shown to be indeed only apparent - in the final analysis (which is what Jnger
avowedly strives for, afinal analysis), the element of certainty typically prevails. On the chiasmic reversal of properties of matter and consciousness, see de Man,
passim. (An early instance which is particularly interesting here, as it involves a
shell, is de Man, "Image and Emblem in Yeats" (excerpt from 1960 Ph.D.


tem cannot but break down, for in its very production the Geist that
claims it as its product cannot but render null and void its altemate claim
to splendid ineffability, which was the rationale for its assumption of the
solidity of matter in the first place. lt is only as a consequence of the
mind's refusal to allow for the material that matter can be said to dream
at all, and it is in this utterance that the mind misses the purpose of its
trope by doing exactly that which it pretends to contemn: "Was aber ist
es mit der Verachtung des Geistes davor, sich auszumnzen und in Kurs
zu bringen?" Rather than being the subjective genitive it appeared to be,
the Traum der Materie tums out tobe minted as an objective one: the
image of the crystal is the mind's dream of matter, and it only succeeds
in recuperating and transfonning matter into mind, the ineffable into the
discursive, even while and because it intends to do exactly the opposite.
Again we encounter the figural and universalist ontology we saw operative in the previous passage: the object is made to undergo a Verwandlung at the discretion of the subject and consequently disappears,
or, more accurately, appears only insofar as it has shed the alterity of
matter that escapes the phenomenal reduction to objectivity. As this
definition suggests, in its tracing of the transfonnation from one object
into another, the difficulty of thinking against this transfonnation is not
tobe lightly dismissed. In order to think (about) it at all, it is necessary to
recognize the epistemological thrust of the transformation: the obedience
of the object to the image imposed on it is indeed an opening up of the
object to cognition, but the crucial element in this process is that this
cognition is presented as a cognition of matter measured as the source of
the object. We will retum to this difficulty by reading yet another fragment in a similar vein, but before we do so we may want to touch briefly
upon Paul de Man's awareness, in his 1943 review of Das Abenteuerliche Herz, of the epistemological involved - as witness the following praise:
On y retrouve en effet cette vision etonnemment cosmique qui caracterise
Jnger, capable de decouvrir l'universel dans le plus minime objet. 11 se penche
sur une pierre, sur un insecte avec le regard de celui qui peryoit les mysteres les

dissertation), in: id., The Rhetoric of RomanJicism, New York, Columbia University
Press, 1984, pp. 145-238; pp. 152-60, pp. 192-93, p. 197. On this shell, see also
Andrzej Warminski's Epigraphs", preceding his "Terrible Reading", in: Responses,
cit., pp. 386-96; p. 397.)


plus caches de la matiere. Et ce n'est pas f une fausse profondeur, qui ne se

manifeste que par un verbiage delicat donnant une superficielle impression de
poesie. Bien au contraire, on atteint par cette voie a une reelle connaissance,
non pas celle qu'apporte la raison, mais celle plus rare, qui nait d'un contact
avec les sources memes des etres et des choses. (WJ, p. 376)

Let us turn to another trope in order to repeat this.

('"We have a machine down there,' the Sergeant continued,. 'that splits up any
smell into its sub- and inter-smells the way you can split up a beam of light with
a glass instrument. lt is very interesting and edifying, you would not believe the
dirty smells-that are inside the perfume of a lovely lily-of-the-mountain.'" Flann O'Brian, The Third Policeman)

In the short essay on "Der Stereoskopische Genuss" in Das Abenteuerliche Herz, Jnger expounds his own special theory of aesthetic perception, which centers on the concept of the split sense: "Stereoskopisch
wahrnehmen heisst, ein und demselben Tone gleichzeitig zwei Sinnesqualitten abgewinnen, und zwar durch ein einziges Sinnesorgan"
(AH2, p. 198). If the coercive mode of this kind of perception is already
present in the verb abgewinnen, it comes fully to the fore in a further explanation of "stereoscopy": "[ ... ] ihre Wirkung liegt darin, dass man die
Dinge mit der inneren Zange fasst. Dass dies durch nur einen Sinn, der
sich gleichsam spaltet, geschieht, macht die Feinheit des Zugriffes gross"
(AH2, p. 200). After this very graphic description of sense-splitting and
inner pliers (or forceps), the following lines are somewhat surprising:
Die wahre Sprache, die Sprache des Dichters, zeichnet sich durch Worte und
Bilder aus, die so ergriffen sind, Worte, die, obwohl uns seit langem bekannt,
sich wie Blten entfalten und denen ein unberhrter Glanz, eine farbige Musik
zu entstrmen scheint. Es ist die verborgene Harmonie der Dinge, die hier zum
Klingen kommt[ ...] (AH2, p. 200)

Tue true language, then, is a language of words and images that (not unlike the chickens snatched up by the drunken dark gnome) have been ergriffen by der inneren Zange and that are said to subsequently blossom
forth like flowers. Notwithstanding its apparent sophistication and its
(slightly deceitful) appropriation of Hlderlin's notoriously enigmatic


versell, the structure of the aesthetics and epistemology of the passage

remains the same as before: mind extracts, more or less violently, something from reality which tums out to be, or into, a word or image which
then, in turn, obediently intimates the reelle connaissance that measures
the music of "die verborgene Hannonie der Dinge". In this particular
case, however, a slight swerve from this basic pattem lights up an important feature that in other cases is easily glossed over; for here it is said
not that an object is ergriffen but that words and images are extracted
from language in the same way as objects of stereoscopic perception are
abducted from reality, and that these extracts themselves become objects
11 Tue nearly obsessive concem with flowers is one of the most characteristic surface
features of Jnger's aesthetics - a particularly instructive example is the fragment
"In den Treibhusern" (the German word for the hothouse is very velicitous here),
which forms part of' what the author proudly refers to as his Kritik der Orchideen
(Kant is never far away here - or perhaps rather the contrary, as I shall suggest later
on): "Am Nachmittag tat ich den gewohnten Rundgang durch die Treibhuser, um
meine Kritik der Orchideen zu bereichern, der ich die Spielregel zugrunde gelegt
habe, dass diese Blumen als Schauspielerinnen zu besprechen sind. Meine bung
besteht darin, sie lange und mit gedankenloser Starre zu betrachten, bis sich gleichsam durch Urzeugung das Wort einstellt, das ihnen angemessen ist." (AH2, p. 219).
The duplicity of Jnger's aethetics is quite conspicuous here, notably in the opposition between his Spielregel, which is blatantly anthropomorphic, and the pretense to
Urzeugung of the adequate work. Again, we may remark that Angemessenheit works
both ways: one way only, and it is not that which Jnger's floral hypnosis pretends to
follow. - Fora reading of "Worte, wie Blumen", see, of course, de Man, "Structure
intentionnelle de l'image romantique", in: Revue Internationale de philosophie, 14,
1960, pp. 68-84 (a revised translation made by de Man in 1970 is reprinted as
Chapter 1 of The Rhetoric of Roma.nticism). Fora further reading of de Man's reading, especially interesting here as it specifically engages an opposition between
flowers and stones (which would appear not to hold for Jnger), see Dirk De Schutter, "Words like Stones" (forthcoming in the proceedings of the June 24-25 1988
Antwerp conference on de Man). - "Worte, die, obwohl uns seit langem bekannt,
sich wie Blten entfalten" - this implies that the words were already part of the economic system of circulation of coinage, that they were already kursfhig. A sustained reading of this trope will have to take into account the Satz from Goethe on
which Heidegger somewhat enigmatically concludes his letter to Jnger, and which
revolves around the conception of words and expressions as heilige Zeugnisse and
wahres quivalent or as Scheidemnze oder Papiergeld (see Heidegger, "Zur Seinsfrage", in: Wegmarken, cit., pp. 379-419; p. 419), especially as the same letter also
relays the same Hlderlin verse alluded to here (p. 417). No such reading can be undertaken here, but its future possibility deserves to be pointed out as proof of the
dense and determined (but how?) grafting our initial conjunction invites (cf. note
28). See also note 14 for "Zur Seinsfrage" and note 12 for a further excursion on


(flowers) rather than the other way about. Now, one could say that this
hardly matters, as the net effect of the operation - an insight into the hidden unity of everything that is - remains the same, but that is precisely
the point: the net effect is indeed always the same, for the simple reason
that it is not an effect but a precept which the typical tropology Jnger
employs (and borrows from a powerful tradition) transforms into a
quasi-effect by circulating it in a figural system whose potential is never
fully actualised "on the page", so to spealc. lt matters little to this tropology whether one takes one's cue from a natural object or from a linguistic construct: in the uncritical reading of Hlderlin's enigmatic verse in
which the system is founded, there simply is no real difference between,
say, words and flowers, and to circulate the one is always already to ac-
tivate the other (which is also to say that to circulate the One is always
already to obliterate the Other), but what exactly is the process of this
stereoscopic language?
Stereoscopic language is language that sees - and seizes, with der
inneren Zange - stereotes, solidity, and imprints itself, as cognition, in
this solidity, reading its own imprint as the truth of Being. This stereotyped truth of Being is rarely called "the truth of Being"; it figures as das
Wunderbare, das Ein und das Alles, die Harmonie der Dinge which is
purportedly read or perceived (the difference between them beingimmaterial, if not delusive, in this vision) in everything that is - the apparent modesty of refraining from calling this percept the truth of Being is
part of the efficacy of the system, as is the temporality of perception and
understanding it relies upon. This temporality - the path from perception
to the apperception of the Belehrung of that perception - should not be
taken to be a true temporal unfolding: the temporality of the stereotype is
part of its self-justification, of its concealment of its own fundamental
metaleptic neglect. In the aesthetics of the stereotype, which is also the
epistemology of obedience, matter is but that in which a preconceived
type is stamped, and the type is precisely that which purports to be
formed by matter in the process of stamping, a process which Jnger also
calls ablesen, and which is very different from reading as de Man,
among others, has taught us to think (of) it. Indeed, ablesen is what one
does when one reads a dial: it is a "reading" that only reads what the dial
reads, and the dial only reads what itis read to read. Das Ablesen, so to
speak, lest ab: it empties out, it reaps, it abstracts and extracts from that
which it pretends to read, which is ever only the readable as what has
already been read, its stereognosis seizes only what it has already typed


out, or carved out - in this sense, Ablesen recalls Heidegger's reading of

das Rechnen: "Das Rechnen lsst anderes als das Zhlbare nicht
Thus, what remains categorically unthought and unallowed for
throughout is the un-typed stereotes, the materiality that resists all typing
because it "is" (but what is this "is"?) materiality (or, if you will, "the
real thing", a "type" "qu'il faut mettre a la porte"). What remains also
unthought - but it is the same oblivion - is the "true" object of the type's
stereophronetic confinnation (Jnger speaks repeatedly of a Besttigung)
12 Heidegger, "Nachwort zu 'Was ist Metaphysik?"', in: Wegmarken, cit., pp. 301-310;
p. 306. Ablesen: In a later text, first published in 1963, Jnger has considerably refined the stereotypology structuring his theories in the thirties and forties, without,
however, succeeding in thinking back to the fundamental obfuscation inherent in
that aesthetics. As we are not here concerned with a balanced and encompassing assessment of Jngers aesthetics ("as much"), we may, for the time being, suffice by
quoting a few lines that exemplify the persistence of the universal tropo-typology of
everything that is in this later essay (Jnger, ''Typus, Name, Gestalt", in: Werke.
Band 8: Essays W, Fassungen, pp. 383-473. "Die Schwierigkeit liegt darin, dass ein
Geformtes leichter zu beschreiben ist als eine Form. Einen Bolzen, eine Schraube,
eine Feder vermgen wir zu schildern, whrend sie vor uns auf dem Tisch oder auf
unserer Hand liegen. Das gilt nicht fr Typen und Gestalten: Der Typus kommt
nicht in der Natur und die Gestalt nicht im Universum vor. Wir mssen beide, wie
eine Kraft an ihrer Wirkung oder einen Text an seinen Zeichen, an den Erscheinungen ablesen" (p. 386; emphasis added). Tue fragment immediately following this
one reworks an observation already made in the 1947 text "Sprache und Krperbau"
(published in the same volume of the Werke, pp. 49-103). I first quote the 1964 version, then the 1947 one: "dass wir bei diesem Ablesen von einer sichtbaren auf eine
unsichtbare Harmonie schliessen, deutet sich bereits in der Sprache an. 'Begreifen'
bezeichnet eine reale und eine geistige Antastung. 'Form' ist sowohl das geformte
Objekt als auch die Mater, in der es gegossen wird. 'Mnze' ist sowohl das Geldstck, das von Hand zu Hand geht, als auch die Anstalt, in der es zu Tausenden
geprgt wird: der Ort seiner Individuation. I Das betrifft unser Thema: Wir sehen die
Prgung, aber nicht den Prgstock; wir sehen die Mnzen, aber wir sehen die Mnze
nicht. Ob berhaupt eine solche Mnze bestehe und wo sie zu vermuten sei: das war
von jeher der schrfste Prfstein der Urteilskraft. Das Thema stellt nicht nur Fragen,
sondern es verndert den Menschen, der sie beantwortet." (p. 389). - "Die Sprache
verfhrt hier wie die Natur mit den Versteinerungen: sie gibt entweder Abdruck oder
Kern, die beide sich oft zum Verwechseln hnlich sind. Osmotisch dringen usseres
und Inneres ineinander ein. Die Sprache wird nie hinreichend Aufklrung darber
geben, ob das Gesagte Kern oder Abdruck ist, ob es dem Reich der Dinge oder der
Vorstellung entstammt. Ein Doppelsinn wohnt selbst einfachen Worten inne, wie
Eindruck und Form. Ein Eindruck wird empfangen und geussert; Form ist sowohl
Type wie Mater, Geprgtes wie Prgendes." (p. 86; emphasis added, except for
Form and Eindruck).


of the truth of Beingin qualifications of hannony, unity and wonderfulness which bracket their object (Being) by installing themselves as
cognitions (Erkenntnisse) of Being which are provided by Being and are
as such fundamentally independent from the language in which they are
postulated. Stereotypical language, in short, participating as it does in the
(stereo )scopic-gnostic drive for immediacy, denies itself as language by
rhetoricallyclaiming its ground at the far side of rhetoricity, that is: in its
claim to perceive the coincident originary coinage of Geist, Materie, and
Being - in the postulate of a perception that would challenge Wordsworth's famous founding question: "Oh! Why hath not the mind I Some
element to stamp her image on I In nature somew.hat nearer to her own?"
(Prelude, Book V, vv. 44-46):
Der Sprachgeist ruht nicht in den Worten und Bildern; er ist in die Atome
eingebettet, die ein unbekannter Strom belebt und zu magnetischen Figuren
zwingt. So allein vermag er die Einheit der Welt zu erfassen, jenseits von Tag
und Nacht, von Traum und Wirklichkeit, von Breitengraden und Zeitrumen,
von Freund und Feind-in allen Zustnden des Geistes und der Materie. (AH2,
p. 316)13

Now, the obvious objectionableness of such oracular judgements

notwithstanding, the "experience" of the stereotype on which it relies
should not be dismissed too hastily. We should be wary of conceiving of
stereotypical language as "something" that can be replaced at will by
some "other" language which would be the product of a higher
awareness. The forgetfulness of this language - its oblivion to
materiality and. to that which it confinns to have read in its own impress
"in" a solidity which is nothing but this impress, hence also its
forgetfulness to itself - is not, to use the famous image of Heidegger's
letter to Jnger, the forgetfulness of the philosopher who forgot his

13 Cf. also: "Aber immer ist vom Autor zu verlangen, dass ihm die Dinge nicht vereinzelt erscheinen, nicht treibend und zufllig - ihm ist das Wort verliehen, damit es an
das Ein und das Alles gerichtet wird." (AH2, p. 183); or, more intricately, "Am Stoff
ist also kein Mangel, doch soll ihm die Sprache noch etwas hinzufgen. Sie hat das
Wasser wieder herbeizuzaubem, das mit und ber diesen Gebilden spielt - ein
Wasser, das zugleich bewegt und durchsichtig ist." (AH2, p. 182). (The system is
completed with the statement, from "Sprache und Krperbau", that "Wasser ist Erdvergeistigung" (p. 81) - this pattem of belated completion/actualisation is, as we
suggested, part and parcel of Jnger's strategy.)


umbrella1 4 lt is not something that can be remedied at the discretion of

the will, by returning to the forgotten in order to retrieve it, for instance.
But it can be kept at bay (by not retrieving "it", precisely - by reading it,
perhaps), which is not what happens in the passages from Jnger the
predecing schematics was abstracted from. For what structures these
passages is not what they explicitly state in their intentness on a reality
which they qualify as hannonious, wondrous, and one; to the contrary,
rather, the hannonious, the wondrous, and the one are the rhetorical
measures by which that reality is forced to shine forth (called into being)
as, indeed, the truth of Being, irrespective of the text's refraining from
naming this truth of Being by calling it such. Or, more accurately, as a
result of that very refrain. A next and final trope will allow us to put this
pattem into a somewhat broader perspective.

In one of the "Strandstcke" in Das Abenteuerliche Herz, Jnger describes a spectacle (Schauspiel) he observed while standing on a cliff
overlooking the sea: the comings and goings of a colony of birds nesting
in the face of the cliff draw a pattem of "zauberhafte Regelmssigkeit"
over the "blanke Scheibe" of the sea. Immediately, and as usual, Jnger
grasps the lesson Figuren of this type provide: "In ihrer tellurischen
Mathematik bieten sie eines der mchtigsten Schauspiele dar, in dem
sich hllenloser als sonst Gewalt und Ordnung dieser Erde offenbart"
(AH2, p. 263)15. In these figures, moreover, "[ ... ] fhlen wir, wie in einer
14 Cf. Heidegger, "Zur Seinsfrage", p. 409. As is well-known, this text was first published in 1955 in a Festschrift for Jnger's sixtieth birthday; its original title, "ber
'Die Linie"', critically mirrored that of the essay Jnger wrote for Heidegger's sixtieth birthday in 1950 (Jnger, "ber die Linie", in: Werke. Band 5: Essays/, Betrachtungen zur Z-eit, pp. 245-289). For a reading of this exchange, see Philippe Lacoue-Labarthe, "Typographie", in: Sylviane Agacinski et al., Mimesis des articulations, Paris, Flarnmarion, 1975, pp. 165-270; esp. pp. 180-84. See also LacoueLabarthe's further remarks on and rehearsals of the points made in this seminal text
in "La transcendance finie/1 dans la politique" (p. 168), "Poetique et politique" (p.
186), and "Oedipe comme figure" (pp. 203-05)- all three texts are collected in: Lacoue-Labarthe, L'Imitation des modernes (Typographies 2), Paris, Galilee, 1986.
15 For the import of the notion of mathematics, cf. also "Der kombinatorische
Schluss": "Das kombinatorische Vermgen unterscheidet sich vom nur logischen insofern, als es sich stets in Fhlung mit dem Ganzen bewegt und nie im Vereinzelten


Urmelodie, Verwandtes [... ] anklingen -", to wit, "das khne Doppelspiel des Geistes": "Auf die eine Seite strebt dieses Spiel der hchsten,
metallischen Ausprgung des Bewusstseins zu, auf der anderen verliert
es sich in die Zonen der elementaren Gewalt."
With this intimation of this doubleness, of this oscillation between the
impress of consciousness and the realm of elementary violence, it might
seem as if this passage, contrary to the schematics we sketched before,
does acknowledge an otherness that escapes the typology (the type being
that which impresses, or imprints the (metallic) Ausprgung). This,
however, would neglect the fact that the oscillation itself is asserted as
circumscribed or delimited by "das khne Doppelspiel des Geistes" as it
is mirrored in the mathematics of the figure which is in return said to
represent it in its revelation of the "Gewalt und Ordnung dieser Erde"
that is verwandt to the Geist. The continuation of the passage may help
to clarify what is involved here; in the Geist's two "opposed" inclinations
(Neigungen) is concealed the "Einheit und Mannigfaltigkeit unser so rtselhaften Welt'', and this unser should be read. very restrictively: "Nichts
ist so sehr bezeichnend fr uns wie dieses Nebeneinander von furchtbar
entfesselter Kraft und der unbewegten Khnheit der Anschauung - das
ist unser Stil, ein Stil von vulkanischer Przision, dessen Eigenart man
vielleicht erst nach uns erkennt." (AH2, p. 264; emphasis added). Rather
than being a stage beyond, or a retreat from, stereotypology, this passage
confirms it in its fundamental thrust: it is the ontology of the type as it is
enforced by a powerful "us", and thus bespeaks what Lacoue-Labarthe
has rightly called, at several occasions and with curious insistence on his

verliert. Wo es das Einzelne berhrt, gleicht es einem Zirkel aus zweierlei Metall,
dessen goldene Spitze im Zentrum fusst. Dabei ist es in weit geringerem Masse auf
Daten angewiesen; es beherrscht eine berlegene Mathematik, die zu multiplizieren
und zu potenzieren versteht, wo die gewhnliche Rechenkunst sich mit einfachen
Additionen behilft." (AH2, p. 195; see also note 4). - Indispensable readings of such
mathematical totalisation are de Man, "Anthropomorphism and Trope in the Lyric"
and "Aesthetic Formalization in Kleist's ber das Marionettentheater" (resp.
chapter 9 and 10 in The Rhetoric of Roma.nticism), as weil as "Pascal's Allegory of
Persuasion", in: Stepen Greenblatt (ed.), Allegory and Representation, Baltimore,
Johns Hopkins University Press, 1981, pp. 1-25. For instance, in the first text, "If
number can only be conquered by another number, if identity becomes enumeration,
then there is no conquest at all, since the stated purpose of the passage was, like in
Pascal, to restore the one, to escape the tyranny of number by dint of infinite
multiplication." (p. 250).


baptismal perfonnance. "[ ... ] en toute rigueur, une onto-typo-logy"l6.

What must interest us now, in our limited perspective, is that the type,
the Prgung, the Gestalt, is unclerstood, in Jnger, as Lacoue-Labarthe,
taking recourse to Heidegger, also explains, as "donation de sens",
which is why it ultimately has to be "la figure d'une humanite" - the
figure, in this case, of a - of unser Still?. What is this type, or face, of (a)
humanity? Has it already arrived or is it still in the process of becoming?
Is it a totality or is it (still) (only) a fragmentation? The efficacy of
Jnger's text, here as elsewhere, resides in its apparent ability to affirm
both of these alternatives (and thus only one). For what does it say?
First, there is the bold (khn) double game of the Geist between the
highest metallic impressions of consciousness on the one hand, and the
swerve into the realm of elementary violence on the other. This bold
double game is then asserted to be a juxtaposition (Nebeneinander) of
immobile bold (Khnheit, again) Anschauung and terrible unleashed
force, and this juxtaposition is called a style, "unser Stil". The hold game
of the Geist is staged between consciousness and violence into a style
that juxtaposes bold Anschauung and unleashed force. This pattem is
further complicated by the next statement on the state of affairs:
Wir fahren durch diese Welt wie durch eine titanische Stadt, die hier der Schein
von schrecklichen Brnden erleuchtet, whrend dort die Werkleute an den
Grundrissen gewaltiger Bauten beschftigt sind. Es wechseln sich in schneller
Folge Bilder eines tiefen und dumpfen Leidens, das sich wie in Trumen vollzieht, mit der dmonischen Unverletzbarkeit des Geistes, der das Chaos dem
Bann seiner Lichter und Blitze und seiner kristallischen Figuren unterwirft.
(AH2, p. 264) 18
16 Lacoue-Labarthe, "Typographie", cit., p. 181; see also the rehearsals referred to in
note 14.
17 "Mais qui, nous?", Derrida, "Les fins de l'homme", in: Marges de la philosophie,
cit., pp. 129-64; p. 164. We have amalgamated here, as does Lacoue-Labarthe, the
notions of Typus and Gestalt (and Prgung) - as we have already pointed out (note
12), Jnger has later developed a distinction between Typus and Gestalt, to which
we shall retum later on (note 21).
18 This last sentence, it will be noticed, is syntactically ambiguous: "mit" can either be
read as qualifying the altemation (wechseln) itself, or it can be seen as governing the
second element in the altemation. In the fst case, there would only be an alternation between Bilder, in the second the plurality of these Bilder would itself altemate
with a unity of invulnerable Geist. For reasons that will become clear, the present
reading opts for the second alternative, but the curious suggestion of the first (an interminable succession of images of suffering presided over, in some only vaguely
articulate fashion, by the Geist) ought not to be ruled out.


What this passage, oddly establishes is that Geist has now taken in the
subordinate position of Bewusstsein and Anschauung in the previous assertions: Geist is that which subjects a chaos to its figures in a larger altemation (wechseln) with something else (deep and mute suffering),
which. in fact pertains to the other pole of the Nebeneinander that was
earlier asserted to be the style of Geist. Rather than being an element in
the game of Geist, in which Geist tends to lose itself but which is still
part of its style, the realm of elementary violence is now something
which has yet tobe subjected but which escapes a total subjection in an
itinerary of altemation. The conclusion vf the passage, however, redistributes the components of the pattem yet again by rehearsing the initial
specular/spectacular figure of Geist in order to project a resolution into
the future of an Aufgabe in which this initial figure would become, in
fact, the figure of Geist (which indeed it already was):
Aber wie sich hier das Bild der Meeresflche mit den scharfsinnigen Bewegun-

gen der insektenhaften Vgel vereint, so sind auch Orte zu ahnen, an denen
diese beiden grossen Motive sich nhern und ineinander einschmelzen, und es
ist mglich, dass sich in dieser Deckung der metaphysische Teil unserer Aufgabe verbirgt. (AH2, p. 264).

The seductiveness of this passage stems from its prophetic projection

into a possible future, in which the Doppelspiel of Geist between consciousness and violence, or the Nebeneinander of unleashed force and
bold Anschauung, would no longer be the wechseln between Geist and
chaos (which it never was), but the Deckung of both, of what is called
somewhat earlier in the text "die Elementar- und die Ordnungsseite unserer Macht". To put this differently: the future Deckung would unify
what is also unified (vereint) in the Schauspiel of the birds and the sea,
but the important point is that this spectacle was already a revelation of
the "Gewalt und Ordnung dieser Erde", itself verwandt to the bold Doppelspiel of the Geist. The Aufgabe would thus indeed be to accomplish a
Deckung of "die Elementar- und die Ordnungsseite unserer Macht", of
what is already proper to "us". This, then, is quite rightly the metaphysical part of the ontotypological project: it is the repetition of what the
project had already accomplished, as stereotypology, as aesthetics of
obedience, and as universal tropology. lt is the stereotype of the stereotype of the Geist, which is both identical to Bewusstsein as part of everything that is ("diese Erde") and, through a process of synecdochal to-


talisation, identical to this everything ("unsere Welt")19. As such, "der

metaphysische Teil unserer Aufgabe" is "our" already accomplished
Aufgabe itself: it is the paroxysm of the constitutive and thoroughly
metaphysical oblivion ofthe stereo-ontotypology.
What then, again, is the "'type of (a) humanity" in this metaphysics?
Heidegger, speaking of Jnger's Arbeiter, gives an elucidating answer:
Die Gestalt ruht im Wesensgefge eines Menschentums, das als Subiectum
allem Seienden zugrundeliegt. Nicht die Ichheit eines vereinzelten Menschen,
das Subjektive der Egoitt, sondern die vorgeformte gestalthafte Prsenz eines
Menschenschlages (Typus) bildet die usserste Subjektivitt, die in der Vollendung der neuzeitlichen Metaphysik hervorkommt und durch deren Denken
dargestellt wird20.

What this confinns most crucially to our preliminary purpose here - and
thus abstracting from the "substance", but not the "fonn", of the question
of technology and the Arbeiter - is that the metaphysical "part" of
Jnger's self-assigned task is indeed that task itself. Insofar as it appropriates the transcendence of Being into the "Reszendenz" (or "Rcktritt") of the preconceived type of a humanity as "Quelle der Sinngebung" - which, as we have tried to demonstrate, is its fundamental prin19 The instability of crucial terms such as Geist and Bewusstsein in Jnger's aesthetics
is, as will have become clear, an indispensable principle in the construction of that
aesthetics: it is the linguistic indeterminacy that always accounts for the erasure of
the indeterminate linguisticity of the system. lt is, in other words, the principle that
makes it possible to transform incompatibilities into the only apparent doubleness of
allegedly supra-rational poet-ontologies (like Jnger's) that propose to speak "das
Ein und das Alles". - Synechdochal totalisation (such as that operation here between
Bewusstsein as part of the earth and Geist as principle of the world) is another typical pattern in the type of system Jnger expounds - it is most strikingly elaborated in
his "Historia in Nuce: Die Ergnzung": "Wenn wir eine bestimmte Farbe einige Zeit
betrachten, bringt unsere Netzhaut die Ergnzung hervor. Wie jede sinnliche Erscheinung, so hat auch diese ihren geistigen Bezug; wir drfen aus ihr schliessen,
dass uns ein Verhltnis zur Welt als zu einem Ganzen gegeben ist. Wenn irgendeiner
ihrer Teile unsere Aufmerksamkeit bermssig in Anspruch nimmt, so ruft der Geist
wie ein Heilmittel das Fehlende herbei. [... ] Und so wohnt uns auf allen Gebieten ein
Hang zur Ergnzung inne, der heilende Wirkung besitzt. I Besonders schn tritt das
in der Erscheinung des grossen Historikers hervor: unsere Geschichte, die eine
Geschichte der Parteiungen ist, wird durch ein gttliches Auge ergnzt. Architektonisch gesprochen, zeichnet der Historiker in den babylonischen Plan unserer
Anstrengungen die Bgen ein, deren W ahmehmung sich den handelnden Mchten,
die den tragenden Pfeilern gleichen, notwendig entzieht." (AH2, pp. 250-51).
20 Heidegger, "Zur Seinsfrage", cit., p. 390.


ciple throughout -, it marks the prefiguration of the completion of metaphysics in and of itself'21. That this is an all-important and dangerous descent we shall presently, however briefly, try to illustrate.

("Et insensiblement, une personnalire-type s'est trouvee decoupee." Paul de Man, Le Soir, 28 octobre 1941)

Tue Verwandlung of Transzendenz into Reszendenz indeed implies a

danger: the danger of the Verwandlung itself. That is to say: while the

For "Reszendenz" and "Quelle der Sinngebung", see ibid., p. 392. "Rcktritt" (not
bo be confused with Heidegger's "Rckgang") is a terrn employed by Jnger in an
interesting entry (March 21st 1943) in Das Zweite Pariser Tagebuch. The passage is
remarkable for its expression of doubt as to the "rightness" of the totalizing thrust of
the ontotypology and, more typically, for its subsequent demolition of that doubt.
Jnger recounts a conversation he had with (again, typically) a one-eyed Jesuit of
the jenseits-persuasion, and reflects: "Die Unterhaltung gemahnte mich an einen
meiner frhen Zweifel: ob nicht beim Rcktritt in die Einheit uns ein Genuss verlorengehe, den nur die Zeit und nur die Mannigfaltigkeit gewhren kann, und ob
sich nicht gerade darin der Grund zu unserer Existenz verberge, dass Gott der Individuation bedrftig sei. Ich hatte das Gefhl so oft beim Anblick der Insekten und
Meerestiere und all der unerhrten Wunder der Lebensflut. Der Schmerz ist tief bei
dem Gedanken, dass es eines Tages von all dem Abschied zu nehmen gilt. [Note the
preparation of the refutation, already inscribed in the argument: the individuated
(insects etc.) are part of an all, of a Lebensflut, a gathering continuity echoed in the
gathering moment of Schmerz (algos: see Heidegger's "Zur Seinsfrage" on Jnger's
"ber den Schmerz")-OdG.] I Demgegenber ist zu sagen, dass wir beim Rcktritt
Organe gewinnen werden, die wir nicht kennen, obwohl. sie in uns angelegt und
vorgebildet sind, so wie etwa die Lungen im Kind, das die Mutter im Schosse trgt.
Die krperlichen Augen werden gleich unserer Nabelschnur verdorren; wir werden
mit einer neuen Sehkraft ausgestattet sein. Und wie wir hier die Farben im
Aufgeteilten sehen, so dort mit hherem Genuss ihr Wesen im ungeteilten Licht."
(Jnger, Das Zweite Pariser Tagebuch (first published 1949), in: Werke. Band 3:
Tagebcher III, Strahlungen, Zweiter Teil, pp. 9-304; p. 27.). - In a previous note
we referred to Jnger's later sophistication of his typology in ''Typus, Name,
Gestalt". The following lines, which form .the onset of that essay reveal both
Jnger's pretense to a non-metaphysical part of his project and the inevitable contradiction of that pretense in its immediate relegation into the realm of human Erfassung and Konzeption: "Gestalt und Typus sind Formen der hheren Anschauung.
Die Konzeption von Gestalten verleiht metaphysische, die Erfassung von Typen
sichert geistige Macht. Beide beschftigen daher zu allen Zeiten als grosse Themen
das Denken und Nachdenken." (p. 385).


imposition of the type already entails the metaphysical oblivion of the

ontological question, it also immediately furnishes a legitimation of the
translation of this linposition (for there are others) onto the very mundane reality of, perhaps not merely for instance, war. This is not to sug. gest that the ontotypology is only ever coincident with the crudeness of
propaganda - it is to suggest, though, that it all too easily can come to
participate in such propaganda, even if it styles itself against the orthodoxy of that propaganda in a specific historical frame of reference. Tue
possible oppositional mode of a particular elaboration of the ontotypological project is not tobe discounted by any means - more concretely,
in the present case, Jnger's complicated resistances to developments in
Gennany before and during the Second World War has tobe credited to
the full, as it appears in his stereotypology itself -, but the dangerous
proximity of the fonnative processes, the Bildung of the ontotypology to
the justificatory myths of totalitarian politics cannot be glossed over too
lightly either.
A detailed treatment of this problem can clearly not be offered here,
but at the same time sketchiness cannot be condoned in this particular
matter, even if only because what is under consideration is precisely the
sketch provided by the Geist in the ontotypology and its further transfonnation into a scheme for reality. Which is why we shall suffice here
by pointing to Paul de Man's sketchy journalistic reading of Grten und
Strassen, a reading which admirably illustrates the Verwandlung of the
allegedly inconceivable but heavily visible (stereotyped) primitive forces
of Being into the "ordinariness" of what might perhaps be called the ontical:
Les livres [ ... ]de Jnger planent dans des spheres ou seules les forces primitives
agissent, creant la seremte la plus pure vu le plus noir chaos. On y voit des
hornmes lances par des puissances qui les depassent infiniment et qu'ils ne
pourraient ni connaitre ni gouvemer; on y goiite des flicites et des extases qui
sont d'essence divine, des tounnents d'une brutalite bestiale. Dans "Jardins et
Routes" ces elements apparaissent parfois, mais sous une forme plus moderee,
domestiquee en quelque sorte par les brides de la realite: les hornmes dechaines
sont des soldats, les monstres sanguinaires des tanks et des canons - choses
sornme toute assez ordinaires. (WJ, p. 244)

And again, around the same time (middle of 1942), in another brief characterisation of Grten und Strassen that succeeds in juggling universality
and particularity in exemplary fashion - according to the example set by
the "conception du monde" of the metaphysics of obedience precisely:


En effet, c'est moins l'image d'une grandiose operation militaire que l'aspect inrerieur d'une amed'artiste qui vit dans ces pages. Rarement Jnger s'eleve audessus de sa destinee personnelle pour englober d'un vaste coup d'oeil la portee
universelle de l'experience unique alaquelle il a pris part; tel n'est pas le but de
ce joumal intime. Mais ce n'est pas 13 en diminuer l'interet, car les reactions, les
reflexions et les meditations d'un etre d'elite constituent un spectacle aussi riche
en enseignements que le plus synthetique des tableaux historiques. Et c'est bien
f ce qui fait l'attrait principal de ce livre: l'occasion de penetrer dans la vie personnelle d'un des plus grands ecrivains du moment et de puiser dans cette connaissance des enseignements precieux sur une conception du monde qui, meme
au contact des plus cruelles realites, parvient a conserver une sererure souriante
mais d'une admirable elevation. (WJ, p. 361)22

Such is the sketch of smiling serenity - Jnger himself would probably

speak of desinvolture23 - with which the ontotypology succeeds in conceiving the world at war. And it is here that a responsible articulation of
22 Tue oscillations of this passage, arguably overdetermined by those in Jnger's own
writings, are characteristic of de Man's readings of Jnger in general: it allows him,
for instance, to state both that Jnger's work offers a "reelle connaissance" (WJ, p.
376) and that "Jamais [Jnger] ne nous apprendra une chose concrete et il n'enrichira
pas nos connaissances" (WJ, p. 216). Similarly, Jnger's thought is (indirectly) characterised as "[ ... ] une pensee qui ne raisonne pas mais qui 'voit' avec une puissance
et une penetration incomparable" (WJ, p. 269). (Needless to add that this in-determinacy is precisely what is at stake in the primal scene (and Critique) of the aesthetic ideology.) Tue same vagueness also determines the concept of universal particularity central to the complementary nationalism in a European frame of which
Jnger is, appropriately, de Man's favourite incarnation (alongside Rilke).
23 "Zur Desinvolture" (AH2, pp. 264-66): Desinvolture, a word which Jnger is particularly fond of, is variously defined as "d[ie] gttergleiche berlegenheit", "die Unschuld der Macht", "Wuchs und freie Gabe und als solche dem Glck oder der Zauberei weit eher als dem Willen verwandt", "die unwiderstehliche Anmut der Macht",
and "eine besondere Form der Heiterkeit" which is itself one of man's "gewaltige
Waffen", which he carries as "gttliche Rstung, in der er selbst die Schrecken der
Vernichtung zu bestehen vermag". See also, in Grten und Strassen, "Zur
Desinvolture. Hier liesse sich noch erwhnen das Wort 'gracious', zu dem uns
gleichfalls die Entsprechung fehlt. Die Paarung von Macht und Anmut ist bei uns zu
selten, um eigene W orte hervorzubringen, und diese Sprdigkeit hat uns im Grunde
im Lauf der Weltgeschichte den guten Anspruch oft verscherzt." (GS, p. 102). -Tue
serenite souriante de Man ascribes to Jnger in the passage just quoted "echoes"
some of the phrases Maurice Blanchot uses to characterise this author ("Une oeuvre
d'Ernst Jnger", in: id., Faux Pas; Paris, Gallimard, 1943, pp. 296-301) - e.g.
"sereine angoisse" (p. 300), "une hautaine et tranquille tristesse" (p. 300) and
"dignite cruelle" (p. 301). - For the Anmut of power (desinvolture), see, once more,
de Man, "Aesthetic Formalization in Kleist", p. 270; and the brilliant reading of this
essay in Cynthia Chase, ''Trappings of an Education" (in Responses, cit.).


the aesthetics of obedience with its attendant ideologies could seriously

commence. (Perhaps. For "we" . keep saying this, embarrassed by the
facile narrative this suggestion grafts onto (not only) de Man's "critical
career", and, at least up to now, largely unable to dispell this unease by
effecting just this "responsible articulation" in a more than cursory
fashion.) The present frame, however, urges us to circle back, and ahead,
to the juxtaposition we set out from - the question we thus so symptomatically abandon here remains to be thought.

What, then, is the import of this most rudimentary impression of Jnger's
ontotypology in our conjunction? How does this sketch, which is itself
evidently tributary to de Man's reading24 (that, I will add, is the one crucial point not to be forgotten), resonate with, or perhaps even in, that
reading? A very general remark supplementing this sketch will have to
suffice to suggest some possible entries into this question (or, rather, to
confirm some of the entries the sketch has already programmed). The
immediate danger in tracing such entries is, as will also have been
understood, that they tend to become all too easily implicated in what
Derrida has so forbiddingly but indispensably called "the policeman's
petty game"25 -we can only be cautious.
The, to my knowledge, only reference to Jnger de Man made in
print after the war occurs in a remarkable short essay "The Inward Generation", which he published in 1955, while a student at Harvard26. In
24 lt is, perhaps, slightly disturbing that de Man's "master concepts" have the ironical
habit of leading their own predictable life in criticism following his work - in this,
they all too often figure as literary criticism's answer to Holmes and W atson.
25 Derrida, "Like the Sound of the Sea Deep Within a Shell: Paul de Man's War", in:
Critical Inquiry, 14, 3, Spring 1988, pp. 590-652; p. 642.
26 De Man, 'The Inward Generation" - now reprinted in: id., Critical Writings, 195378, edited and introduced by Lindsay Waters, Minneapolis, University of Minnesota
Press, 1989, pp. 12-17 (henceforth CW). Tue essay begins with the following, excessively suggestive paragraph: "There always is a strange fascination about the bad
verse which great poets write in their youth. They often seem more receptive than
any to the mannerisms and cliches of their age, particularly to those which their later
work will reject most forcibly. This early work, therefore, is often a very good place
to discover the conventions of a certain period and to meet its problems from the inside, as they appeared to these writers themselves." (CW, p. 12).


this essay, de Man sketches the paths along which a "modern conservatism" tends to justify its defensive and passive retreat into a certain
permanence: notably, by pointing to the failure of the "blend of revolutionary spirit and aesthetic refinement" (CW, p. 14) that was typical of
the twenties and thirties - and, more in particular, of "those near-great
writers whose ability to catch the mood of the times always put them in
the center of events". Characteristic of these writers, among them Jnger,
is that they were "forcefully committed politically", but ended up "[ ... ]
writing off this part oftheir lives completely, as a momentary abberation,
a step toward finding themselves", and that they were staunch defenders
of certain aesthetic values inherited from their symbolist predecessors
which in their own work, however, tended to whither into "banality and
imitation". De Man then identifies the underlying principle of this failure, which, as he puts it, the catastrophe of the war as weil as a
"mounting mechanization and automaticism" seived to expose:
What happened is that the political as weil as the aesthetic were being used, not
for what they represented in thernselves, but as a protection which shielded
them from their real problerns. Political systems of the left and the right, and
literary experimentations which had originated before them, provided an organized framework within which they could fit and act, without really returning to
the questions out of which these systerns and experiments had arisen. (CW, p.

De Man does not explicitate just how this judgement would apply to
Jnger, but that is, as we have repeatedly tried to underscore, not our
main concem here27. What matters to us now is the use to which de Man
puts this judgement in his larger argument.
Jnger's failure, it is argued, was due to the fact that he did not retum
to.the questions out of which the tradition in which he had installed himself had arisen, but rather used that tradition to fend off its originary
problems. These questions and problems, it then transpires, are in fact
those that became prominent in the "ontological crisis" (CW, p. 15) de
Man aligns with the advent of romanticism: they are, in fact, appurtenant
to the ontological question itself. Importantly, though, this does not mean
that the "modern conseivatism" which would decry the failure of com27 Reference should once again be made to Bohrer, op. cit., which is (among other
things) the indispensable starting point for any analysis of Jnger's evolution. A very
different study of this evolution, the project of which is particularly interesting here
is Von Krockow, op. cit..


mitments like Jnger's is any more attentive to the causes of this failure:
first "[ ... ] it should be proven [... ]", de Man ordains, that "today's conservatism" is "[ ... ] not just doing the same thing, in a more vicious and
destructive way". Which is, of course, the opposite of what will be
"proven": indeed, not only is "modern conservatism" - which de Man
suspects of being "just another fonn of nihilism" (CW, p. 13)28 - guilty
of forgetting the ontological question, but it does so in a way which is
much more insiduous than that of the unquestioning assumption of traditions arising from this question which writers like Jnger chose:
"modern conservatism" in fact represses the question by giving in to a
"desire for serenity" (CW, p. 15), to the temptation of permanence29,
which is the denial of nothing less than history. The essay's subsequent
reference to Heidegger as the philosopher who can offer an alternative
for this repression was only tobe expected.
What this last explicit appraisal of Jnger by de Man teaches us on
the preliminary level of this sketch is fairly obvious: in the larger question of the assessment of de Man's wartime journalism, it can provide a
number of difficulties the investigation of which may contribute to what
Jacques Derrida has outlined as the "articulation" of de Man's '"early
writings"' with "the work to come"30. While trying not to fall prey ("if
possible") to either of the "two more or less symmetrical errors'' such an
investigation is highly vulnerable to, this articulation could be conducted
along the following (well-worn) lines: a sketch, yet again, crudely.
28 The question of nihilism is, of course, the question of Being in "ber 'Die Linie"'
andin "ber die Linie".
29 "Serenity" - Jnger, too, possessed a "sererute souriante". For the issue of permanence, see the roughly contemporaneous de Man, ''Tentation de la permanence", in:
Monde Nouveau, 93, Octobre 1955, pp. 49-61 (trans. CW, pp. 30-40). "Tue Inward
Generation" uses the same example for the author who has fallen prey to this temptation as ''Tentation" does: Malraux and his eponymous Noyers de l'Altenbourg.
Malraux is, moreover, listed alongside Jnger (and Pound and Hemingway) as one
of the "near-great writers" who failed in their political activism and aesthetic formalism and subsequently embraced a "nihilistic conservatism" (CW, p. 16). We
might underscore here that Jnger's aesthetics as we have read it in the present paper
is equally susceptible to the allegations de Man here levels at Malraux. For the sake
of completeness, we may add that in another text from the same period ("Le devenir,
la poesie", in: Monde Nouveau, 105, Novembre 1956, pp. 110-24; trans. CW, pp. 6475), de Man also launches an attack on the "ambiance de sererute dont on peut demander si elle a vraiment ere meritee" (p. 111) that characterises the thought of
those tempted by permanence.
30 Derrida, "Like the Sound of the Sea ... ", cit., pp. 640-41.


Tue ontotypology our reading of Jnger has drawn out could arguably be identified as one of the principal targets of de Man's critical
enterprise. If so, however, it would have tobe categorically distinguished
from the justificatory target it forms for the "modern conservatist"
(nihilist) denunciation: on the contrary, de Man's critique is precisely
aimed at an analysis of the forgetfulness that is constitutive of, for instance, Jnger's thought and consequently also envisages the repression
of the question that aesthetics "only" forgets. In other words, de Man's
reading could be said to be an attempt to critically think again the questions the ontotypology forgot - which are the originary questions of what
he calls romanticism31 - by approaching the systematicity of that oblivion as it was already inscribed in the romantic predicament from its very
inception; that is to say: in the rhetoricity of, in the categorical crossings
engendered by, that predicament.
Thus, far from being a wholesale and uncritical rejection/repetition
of the aesthetic ideology he was implicated in - through Jnger, among
others -, de Man's subsequent inquiry could be demonstrated to take seriously the problematics whose oblivion resulted in that ideology's constitution. His critical itinerary - which is an iteration, not an obsessed
repetition, a "Schritt zurck"32, not a "pacifying and diplomatic transla-

31 Tue persistence of "romanticism" as a horizon for de Man's thought can be traced

without too much difficulty throughout his work - some examples (which, evidently, stand in need of attentive comment): "We know all this; the characteristics of
romanticism are now apart of literary history. But we do not generally realize that
we are still living under the impact of exactly the same ontological crisis. Never
have the truly great minds of romanticism, such as Rousseau, Hlderlin, or Hegel
been more familiar and more directly concerned with our own situation" (CW, p.
15). - "Die Hauptpunkte, um welche die heutigen methodologischen und ideologischen Auseinandersetzungen kreisen, knnen fast immer direkt auf das romantische
Erbe zurckgefhrt werden" ("Wordsworth und Hlderlin", in: Schweizer Monatshefte, 45, 12, Mrz 1966, pp. 1141-55; p. 1141), - "[ ...] the question of Romanticism
can no longer be asked in the manner to which we are still accustomed and [...] the
genetic and monumental patterns that are commonly associated with Romanticism
have lost much of their authority." ("lntroduction", in: Studies in Romanticism, 18,
4, Winter 1979, pp. 495-99; p. 499).
32 For another, and much more detailed, survey of de Man's work, from which this
term is borrowed, see Werner Hamacher, "Unlesbarkeit", in: de Man, Allegorien des
Lesens, Aus dem Amerikanischen von Werner Hamacher und Peter Krumm, Mit
einer Einleitung von Werner Hamacher, Frankfurt am Main, Suhrkamp (Neue Folge,
Band 357), 1988, pp. 7-26; p. 8.


tion"33 of themes already solidly in place from the outset - would then
have to be followed attentively in its increasingly intricate encounter
with the aesthetic ideolgy (ideologies), in a critique which, interestingly, gains in acumen to the extent that the rhetoric of Being (and Heidegger) gets to be less explicitly pronounced and a profound new encounter with the difficult narrative of metaphysical and transcendental
principles in Kant is engaged in, all this "up until" the inevitable affirmation of the impossibility for thought to come to tenns with itself, to
tenninate (with) itself in the system it can neither uphold nor resist. The
affinnation, that is, of "the bottom line", of "the prosaic materiality of
the letter" which "[ ... ] no degree of obfuscation or ideology can transform [... ] into the phenomenal cognition of aesthetic judgement"34.
33 Derrida, art. cit., p. 641.
34 De Man, "Phenomenality and Materiality in Kant", in: Gary Shapiro and Alan Sica
(eds.), Hermeneutics: Questions and Prospects, Amherst, University of Massachussets Press, 1984, pp. 121-44; p. 144. See also, in the same text, the challenging reflections on the interplay of metaphysical (ideological) and transcendental (critical)
principles: "Tue possibility of maintaining the causal link between them is the controlling principle of rigorous philosophical discourse: philosophies that succumb to
ideology lose their epistemological sense, whereas philosophies that try to by-pass
or repress ideology lose all critical thrust and risk being repossessed by what they
foreclose" (p. 123). - Our last instance and comparability: Kant, in the "Allgemeine
Anmerkung" following 29 of the Third Critique (which de Man here discusses):
"[ ...]man muss den Ozean bloss, wie die Dichter es tun, nach dem, was der Augenschein zeigt, etwa, wenn er in Ruhe betrachtet wird, als einen klaren Wasserspiegel,
der bloss vom Himmel begrenzt ist, aber ist er unruhig, wie einen alles zu verschlingen drohenden Abgrund, dennoch erhaben finden knnen." - Jnger, in the third
"Strandstck": "Das Meer nahm das Aussehen einer blanken Scheibe an, von deren
Umkreis das gefiederte Leben strahlenfrmig zu einem geheimen Mittelpunkt
zusammenschoss, um sich dann in der gleichen Ordnung wieder zu zerstreuen. Es
schien den einschlfernden Glanz dieses Spiegels noch zu erhhen, dass sich das
feine Netz der Flugbahnen wie eine strenge Gradeinteilung auf ihm ausbreitete"
(AH2, p. 263). -De Man, commenting on Kant's Wasserspiegel: ''The 'mirror' of the
sea surface is a mirror without depth, least of all the mirror in which the constellation would be reflected. In this mode of seeing, the eye is its own agent and not the
specular echo of the sun. Tue sea is called a mirror, not because it is supposed to reflect anything, but to stress a flatness devoid of any suggestion of depth. In the same
way and to the same extent that this vision is purely material, devoid of any reflexive or intellectual complication, it is also purely formal, devoid of any semantic
depth and reducible to the formal mathematization or geometrization of pure optics"
(p. 136). Devoid of any reflexive or intellectual complication: that is, "free" from
ontotypological appropriations of, among other things, mathematics and optics as
reservoirs of master tropes.


Tue thinking of this bottom line, which Rodolphe Gasehe has so lucidly read as the "incontoumable" impossibility "of' thinking35, cannot,
as this crude blueprint might have suggested, be grasped as an effect of
an initial aberration into aesthetic ideology - it can, precisely and. rigorously, not be grasped at all, which is ultimately the reason why it will
nevertheless always be preseived in the narrative of the aberration that
forgot it in the first place. The only thing the narrative can do, perhaps,
for the time being, is, as was suggested before, to keep at bay the violence - the ontotypological, metaphysical, political violence which is always demanding the justification the aberration provides. Ultimately - if
we can still say this - this means that the narrative of our conjunction, of
the Nebeneinander with which we began, should not be allowed to turn
into a Deckung: that is the critical part of our "Aufgabe", which assuredly has not yet been faced, has only, perhaps, been faced up to, in an
imitation (and thus a betrayal) of "true mouming" such as the preceding.

This paper fonns part of a Ph.D. project I am.engaged in as research assistant of the National Fund for Scientific Research (Belgium) at the
Katholieke Universiteit Leuven. Thanks to Dominic Christiaans for
technical support.

35 Gasehe, "Apathetie Criticism" (lecture delivered at the conference on de Man referred to above, note 11) - a modified version of this lecture forms part of Rodolphe
Gasehe, "In-Differenee to Philosophy: de Man on Kant, Hegel and Nietzsche", in:
Lindsay Waters and Wlad Godzieh (eds.), Reading de Man Reading, Minneapolis,
University of Massachussets Press, 1989, pp. 259-294.



Cette communication prend son point de depart dans la constatation de l'importance singuliere accordee al'oeuvre d'Ernst Jnger dans les ecrits joumalistiques qu'a publies Paul
de Man pendant la guerre. La lecture de quelques passages plutt programmatiques dans
les Oeuvres de Jnger qui ont ete commenrees par de Man sert a suggerer les contours
d'une solidification specifique de l'"ideologie esthetique" - qu'on pourrait appeler la
"stereo-onto-typology" - qu'elabore la pensee de Jnger. Ensuite, la reception de cette
ideologie dans les chroniques litreraires de de Man est confronree, de fa~n preliminaire,
avec ses pensees ulrerieures concernant l'ideologie esthetique.
En bref, le propos principal de ce texte est d'offrir un essai d'articulation entre, d'une
part, le travail joumalistique pendant la guerre de de Man et, d'autre part, la pensee
critique qu'il a developpee des les annees 50, et ceci en lisant effectivement la litrerature
propagee dans ses chroniques a travers les l~ons de lecture entamees dans son "oeuvre
proprement dite". Le fait que ni la structure, ni les "conclusions" de cet essai ne meritent
la qualification de sensationnel pourrait, peut-etre, donner apenser.


Hans-Jost Frey


In einem spten Vortrag ber Benjamins Aufsatz Die Aufgabe des bersetzers wirft de Man im Zusammenhang mit einer englischen berset-

zung von Benjamins Text die Frage auf, ob die Sprache des Menschen
eine menschliche Sprache sei. Benjamins Unterscheidung zwischen dem
Meinen und der Art des Meinens aufgreifend, gibt er zunchst zu, dass
das Meinen intentional sei. Ich weiss, was ich sagen will, und was ich
sagen will hngt von mir ab. Bei der Art des Meinens ist das nicht mehr
so sicher, denn die sprachlichen Mittel, die zur Verfgung stehen, um
das zu Sagende zu sagen, stehen nicht im Belieben des Redenden, sondern sind ihm von anderswoher gegeben. "Tue way in which I can try to
mean is dependent upon linguistic properties that are not only [not] made
by me, because I depend on the language as it exists for the devices
which I will be using, it is as such not made by us as historical beings, it
is perhaps not even made by humans at all. Benjamin says, from the be..,
ginning, that it is not at all certain that language is in any sense human"l.
Fr de Man besteht das Nichtmenschliche der Sprache darin, dass wir sie
nicht hervorbringen, sondern bernehmen, dass wir ihr also mindestens
so sehr ausgeliefert sind wie wir sie beherrschen. "If language is not necessarily human - if we obey the law, if we function within language, and
purely in terms of language - there can be no intent; there may be an intent of meaning, but there is no intent in the purely formal way in which
we will use language independently of the sense or the meaning" (ib.).
Die Sprache ist insofern nicht menschlich, als sie der Kontrolle durch
den sprechenden Menschen entgeht. Es finden in ihr Dinge statt, die der
Mensch sich nicht zuschreiben kann.

Paul de Man, The Resistance to Theory (RT), Manchester University Press, 1986,
s. 87.


Dieser nichtmenschliche Aspekt der Sprache ist das Hauptthema der

im Anhang abgedruckten Diskussion. Er wird unter anderem durch den
Hinweis auf die Frage nach dem gttlichen oder menschlichen Ursprung
der Sprache verdeutlicht, eine Frage, die nicht entschieden werden kann,
die aber durch ihr blosses Auftreten die Unmglichkeit veranschaulicht,
die Sprache als ein menschliches Produkt zu erklren. ''That it is divine
or not makes little difference, and the more you take the sacred out of
this picture, the better. But it indicates a constant problem about the nature of language as being either human or nonhuman. That there is a
nonhuman aspect of language is a perennial awareness from which we
cannot escape, because language does things which are so radically out
of control that they cannot be assimilated to the human at all, against
which one fights constantly" (RT, S. 101). Wieder ist das Indiz fr das
Nichtmenschliche an der Sprache die Unkontrollierbarkeit dessen, wa_s in
ihr geschieht. Diese Bestimmung des Nichtmenschlichen scheint eine
solche des Menschlichen einzuschliessen. Der Mensch wre durch die
Grenzen seiner Herrschaft definiert. Das Menschliche wre das Kontrollierbare. Und diese Bestimmung des Menschlichen wre selber gerade
dadurch eine menschliche, dass sie als Definition das Menschliche unter
Kontrolle bringt. Nichts ist weiter von de Mans Auffassung entfernt als
eine solche Ableitung, obwohl seine eigene Formulierung sie ennglicht
und zu verlangen scheint. Wenn von der Unkontrollierbarkeit der Sprache gesprochen wird, so ist damit gerade jenes definitorische Reden in
Frage gestellt, das die Herrschaft ber das Definierte sicherzustellen
glaubt. Ein Satz wie: "Das Nichtmenschliche der Sprache ist ihre Unkontrollierbarkeit" ist formal eine Definition, aber diese untergrbt sich
selbst und verunsichert sich radikal. Der Versuch, das Nichtmenschliche
der Sprache zu bestimmen, zielt darauf, es unter Kontrolle zu bringen.
Was aber kontrolliert werden soll, ist das, was dadurch bestimmt ist, dass
es sich der Kontrolle entzieht. In dem Masse aber, als es sich der Kontrolle entzieht, ist es nicht bestimmbar, auch nicht als das Unkontrollierbare. Die Bestimmung der Sprache als unkontrollierbar ist keine Bestimmung, sondern deren Unmglichkeit. Sie vernichtet sich selbst. Was
heisst das anderes, als dass sie radikal ausser Kontrolle gert? In ihr geschieht, was sie vergeblich unter Kontrolle zu bringen sucht: dass die
Sprache der Kontrolle entgleitet. Was sich auf diese Weise zersetzt, ist
die Definition des Nichtmenschlichen der Sprache, mit ihr aber auch jede
daraus ableitbare Bestimmung des Menschlichen. Indem die Definition
des Nichtmenschlichen sich selbst auflst, entfllt die Mglichkeit, dar-


aus irgend ein Wissen ber das Menschliche als das Gegenteil des
Nichtmenschlichen zu gewinnen. Davon her wird eine eher unvermittelte
usserung de Mans zugnglich:"[ ... ] there is, in a very radical sense, no
such thing as the human. If one speaks of the inhuman, the fundamental
non-human character of language, one also speaks of the fundamental
non-definition of the human as such, since the word human doesn't correspond to anything like that" (RT, S. 96). Wenn man also vom Nichtmenschlichen der Sprache spricht, so wird die Sprache nicht am Menschlichen als an etwas Bekanntem und Verfgbarem gemessen, denn das
Menschliche wre, um verfgbar zu werden, auf die Sprache angewiesen, die nicht menschlich und nicht verfgbar ist. Die Unterscheidung
zwischen dem Menschlichen und dem Nichtmenschlichen wird dadurch
weder hinfllig noch entbehrlich, aber es erweist sich als unmglich, sie
zu machen. Die Rede von der Nichtmenschlichkeit der Sprache entsteht
aus der Unmglichkeit, die Sprache auf den Menschen zurckzufhren
und von ihm aus zu verstehen. Aber ebensowenig gelingt es, den Menschen durch die Sprache zu erklren. Zu sagen, dass der Mensch dadurch
Mensch sei, dass er Sprache hat (vgl. RT, S. 100), wrde bedeuten, dass
das Wesen des Menschen in etwas Nichtmenschlichem bestnde. Weder
erklrt sich die Sprache durch den Menschen, noch der Mensch durch
die Sprache. Die Sprache entgleitet dem Menschen, und der Mensch entgleitet der Sprache.
Der erste Teil dieses Satzes - die Sprache entgleitet dem Menschen ist das, worauf die zitierten Stellen vor allem eingehen, und was aufgrund des Gesagten einigermassen einsichtig geworden sein drfte. Es
bleibt aber noch dunkel, was es heissen kann, dass der Mensch der Sprache entgleitet. De Mans Satz, wonach die Rede vom Nichtmenschlichen
der Sprache die Undefinierbarkeit des Menschlichen einschliesse, "since
the word human doesn't correspond to anything like that", gibt einen
Hinweis. Dass das Wort menschlich in keiner Weise bestimmt, was das
Menschliche sei, bedeutet, dass es keinen Zugang dazu erffnet, sondern
leer bleibt. Das Menschliche ist nur als Wort gegeben. Es ist eine
sprachliche Fiktion, von der kein sicherer Weg zu etwas fhrt, was aussersprachlich als Mensch gegeben wre. Dass Mensch eine sprachliche
Fiktion ist, heisst nicht, dass es keine Menschen gibt, sondern nur, dass
wir nicht wissen, was wir so nennen, oder was die Sprache so nennt. Der
Mensch, der sich selber in der Sprache verloren geht, geht auch der
Sprache dadurch verloren, dass keine Definition ihn fasst. Oder anders:
gerade dadurch, dass er definiert wird, geht er verloren. Den Allgemein-


begriff Mensch gibt es nur in der Sprache. Den Menschen gibt es nur als
Wort. Das Menschliche des Menschen, das, was allen Menschen gemeinsam ist und macht, dass sie Menschen heissen knnen, ist nur durch
die Sprache gegeben und beruht auf der Unterdrckung der Unterschiede, die zwischen den einzelnen Menschen bestehen. Es ntzt nichts,
einzuwenden, gewisse Zge seien allen Menschen gemeinsam und andere nicht, und die gemeinsamen ergben die Bestimmung des Menschlichen. Nicht nur gelingt es nicht, die Grenze, die es geben muss, zu ziehen, sondern das Gemeinsame - z.B. die Sprache - ist solcher Art, dass
es sich gerade nicht als das Menschliche fassen lsst. Dieser unverlssliche Status des Wortes Mensch lsst sich mit Hilfe des Kapitels
M etaphor in Allegories of Reading verdeutlichen: Es handelt sich hier
um eine Art ergnzenden Gegentext zum Benjamin-Vortrag. Whrend
dort von der Nichtmenschlichkeit der Sprache gesprochen wird, geht es

hier um die Sprachlichkeit des Menschen.

Der Text, an dem das gezeigt wird, ist die Stelle in Rousseaus Essai
sur l'origine des Iangues, in der es darum geht, dass die figrliche Sprache der wrtlichen vorausgeht, dass also der eigentliche Sinn erst aus der
Metapher gewonnen wird. Rousseaus Beispiel ist der Mensch, der zum
erstenmal einem anderen Menschen begegnet und diesen einen Riesen
nennt, weil er sich bedroht fhlt. Die Metapher Riese hat die eigentliche
Bedeutung ich habe Angst, wobei die Angst ein Zustand der Ungewissheit ist, in dem man nicht weiss, ob tatschlich eine Bedrohung besteht. Die Unertrglichkeit der unentscheidbaren Spannung der Angst
fhrt zu dem gewaltsamen Entscheid, dem andern die Grsse und Strke
zuzuschreiben, welche der Angst als realer Grund dient. "This is done by
means of a metaphor (calling the other man a giant), a substitutive figure
of speech ( "he is a giant" substituting for "/am afraid") that changes a
referential situation suspended between fiction and fact (the hypothesis
of fear) into a literal fact"2. Die Konfrontation des Menschen mit dem
Menschen fhrt zunchst in eine radikale Ungewissheit. Es ist keineswegs das verbindende Gemeinsame als das Menschliche des Menschen,
was zuerst aufscheint, sondern die Fremdheit als bedrohliche. Die Metapher Riese betont nicht die Gleichheit, sondern den Unterschied. Allerdings bleibt es nicht dabei, denn das Wort Riese wird durch das Wort
Mensch ersetzt. Das Wort Mensch ist fr Rousseau "le mot propre" und


Paul de Man, Allegories of Reading (AR), New Haven and London, Yale University
Press, 1979, S. 151.

macht das Wort Riese zur Metapher, indem es dem in der lliusion der
Angst Befangenen seinen Irrtum bewusst macht. Das Wort Riese wird
fortan als Metapher zum Ausdruck der Leidenschaft verwendet, die es
hervorgebracht hat.
At?er das Wort Mensch, das die Metapher ersetzt, ist weniger eigentlich als es scheint. Die Angst wird dadurch berwunden, dass man feststellt, dass der andere nicht grsser ist, und daraus den Schluss zieht,
dass er auch nicht gefhrlich sei. Dieser Schluss, der, wie alltglich zu
erfahren ist, nicht zwingend ist, begrndet den Begriff Mensch und die
Gleichheit der Menschen. Dass der andere, obwohl er gleich gross ist,
mir wohl oder bel gesinnt sein kann und die quantitative Gleichheit somit nicht gengt, um die Ungewissheit, die zur Angst gefhrt hat, zu
beseitigen, wird unterdrckt. Im Wort Mensch ist deshalb der Irrtum
ebenso wirksam wie im Wort Riese. Beide sind Metaphern, aber whrend Riese die Verschiedenheit betont, legt sich Mensch auf die Gleichheit fest. Die musion der Angst ist die Verschiedenheit des Gleichen, die
Illusion der Begrifflichkeit ist die Gleichheit des Verschiedenen. Die
numerische Gleichheit, die wrtlich genommen werden muss, damit der
Begriff Mensch mglich wird, ist eine Metapher, weil die Entsprechung
von Aussen und Innen keineswegs feststeht. Erst dadurch, dass das Verschiedene vergessen oder unterdrckt und die Metapher wrtlich genommen wird, kann die Illusion entstehen, Mensch sei "le mot propre".
Der Begriff Mensch ist aber lediglich die berwindung der mglicherweise berechtigten Angst durch die lliusion der Gleichheit. Diese illusionre Gleichheit ist die Ermglichung der Gesellschaft (AR, S. 155).
Diese ist, wie der Mensch, eine sprachliche Fiktion. Nicht dass es keine
Menschen und keine Gesellschaft gbe, aber es gibt keine Gewissheit
darber, was sie sind. Der Schein einer solchen Gewissheit wird immer
durch Festlegungen erreicht, die fragwrdig sind, weil sie dort entscheiden, wo nicht entschieden werden kann, wo aber entschieden werden
muss, weil die Spannung der Unentscheidbarkeit nicht auszuhalten ist.
Der Entscheid und die Festlegung .sind deshalb ebenso unvermeidlich
wie unhaltbar. Die Gefahr der Festlegung liegt darin, dass ihre letztliche
Unbegrndbarkeit vergessen und an ihre Richtigkeit geglaubt wird. Aber
dieses Vergessen ist seinerseits unvermeidlich, weil der Entscheid immer
die Sprachlichkeit vergisst oder unterdrckt, um die mit der Sprache
notwendig verbundene referentielle Ungewissheit zu berwinden. Nicht
zu wissen, was das Menschliche des Menschen ist, und im Zweifel dar-

ber sein, ob das, was das Menschliche genannt wird, mehr als ein Wort

sei, ist schwieriger als zu behaupten, der Mensch sei dieses oder jenes.
Die Definition und der Begriff sind weder zu umgehen, noch ist es
wnschbar, dass man ohne sie auskme. Aber das hindert nicht, dass Begriffe Metaphern sind und keinen Anspruch etheben knnen, Aussersprachliches zu treffen. Daran zu erinnern wird de Mans Text nicht
Die Stelle aus Allegories of Reading ber das Wort Mensch bei
Rousseau mndet in den Satz, was Rousseau Wahrheit nenne, bezeichne
"the suspicion that human specificity may be rooted in linguistic deceit"
(AR, S. 156). Dieser Satz ist zusammenzunehmen mit dem anderen, wonach die Sprache nicht menschlich ist. Um zu verifizieren, ob das, was
das Menschliche heisst, eine sprachliche Tuschung ist, msste das
Menschliche verfgbar sein. Der Begriff Mensch bleibt aber immer eine
Hypothese, die nur dadurch brauchbar wird, dass man ihren hypothetischen Charakter vergisst. Was so vergessen wird, ist die Sprachlichkeit
des Begriffs. In de Mans Satz ist die Sprachlichkeit erinnert und der unsichere Status des Begriffs wird wiederhergestellt. Gewiss ist jetzt nur,dass es das Menschliche in der Sprache gibt (als das, was mglicherweise, aber nicht sicher, eine sprachliche Tuschung ist). Diese Gewissheit besteht lediglich darin, dass es keine Gewissheit ber das
Menschliche gibt, auch nicht die, dass es nur sprachlich ist. Eine solche
Gewissheit wre der Sprache berlegen und wrde sie beherrschen, was
bedeuten wrde, dass die Sprache menschlich wre und der Mensch sie
zur Verfgung htte. Es ist das Nichtmenschliche der Sprache, das die
Gewissheit ber die Sprachlichkeit des Menschlichen verbietet, und es
ist der Verdacht der Sprachlichkeit des Menschlichen, der die Menschlichkeit der Sprache in Frage stellt.
Die Unbestimmbarkeit des Menschlichen "[ ... ] deprive[s] man forever of a central identity" (AR, S. 140). Diese Dezentrierung des Menschen hat mit der Sprache zu tun, die einerseits die Mglichkeit ist, von
"Menschen" zu sprechen, anderseits aber keinen Zugang zum Menschlichen gibt und dessen Status in der Schwebe lsst. Der Satz, wonach der
Mensch keine zentrale Identitt hat, drfte das Wort Mensch nicht verwenden, da mit ihm diese Identitt gerade gesetzt wird. Aber nur ihre
Setzung erlaubt es, ihr Fehlen zu sagen und die Setzung zu zersetzen.
Beides geschieht in der Sprache. Dass das Gesagte entsagt wird, heisst
nicht, dass es das Menschliche nicht gibt, sondern dass es nicht gehabt
wird, weil es nur sprachlich gegeben ist und sein Status daher ungewiss

bleibt: weil es, wenn man so will, versagt bleibt.


Ein Satz wie "Die Sprache ist nicht menschlich" - oder irgend ein anderer - bezeugt die Notwendigkeit, der man redend untersteht, die Wrter
so zu :verwenden, als wsste man, was sie bedeuten, und als trfen sie
Wirkliches. Der unsichere Status des Wortes Mensch hat aber zur Folge,
dass dieser referentielle Bezug in Frage gestellt ist. Man braucht das
Wort Mensch, ohne zu wissen, was der Mensch ist, aber so, als wsste
man es. Dieses Wissen um die referentielle Zuverlssigkeit der Sprache,
das die Redenden einander zusprechen, ist gemss de Mans Analyse von
Rousseaus Text nicht aus der Sprache zu gewinnen, die es gerade verunsichert, aber es ist die Bedingung jedes Verstehens und damit die Bedingung der Lesbarkeit von Texten. "If to read is to understand writing [... ]
then it presupposes a possible knowledge of the rhetorical status of what
has been written. To understand primarily means to detennine the referential mode of a text and we tend to take for granted that tbis can be
done" (AR, S. 201). Um ber den referentiellen Status einer Rede Klarheit zu gewinnen, muss man das Figrliche vom Wrtlichen, das Gesagte vom Gemeinten unterscheiden knnen. Ihre Referentialitt ist in
dem Masse unsicher, als ihre Rhetorizitt es ist. Diese Beziehung lsst
sich noch einmal an Rousseaus Beispiel Mensch klren. Die Bezeichnung Mensch kommt dadurch zustande, dass von der usseren auf die
innere Gleichheit geschlossen wird. Dieser Schluss ist fragwrdig, weil
die Annahme der bereinstimmung von Aussen und Innen nicht zwingend ist. Er kann zutreffen oder nicht. Wenn die Erfahrung die Annahme
besttigt, so ist Mensch wrtlich zu nehmen als ein mot propre, das eine
Wirklichkeit trifft. Wird die Annahme durch die Erfahrung widerlegt, so
ist Mensch eine Metapher, weil dann als gleich angenommen wird, was
verschieden ist. Insofern als beide Mglichkeiten bestehen, ist Mensch
als Allgemeinbegriff eine Metapher, denn keine Einzelerfahrung mit einem andern lsst sich verallgemeinern. Es ist jederzeit mglich, dass die
Begegnung mit ihm das nchste Mal anders verluft. So bleibt der referentielle Status von Mensch ungewiss. Es ist nie sicher, ob das Wort eine
Wirklichkeit trifft oder ob es eine Fiktion und damit eine Metapher ist,
welche die Verschiedenheit des hnlichen vernachlssigt, und der deshalb keine Realitt entspricht. Diese Unsicherheit ber den rhetorischen
Status des Wortes verhindert, dass es als Allgemeinbegriff Gltigkeit beanspruchen kann. Als ein solcher kann es nur funktionieren, wenn die
Unentscheidbarkeit seines Status durch einen Gewaltakt beseitigt wird,


und wenn mehrere sich darauf einigen, dass es keine Metapher sei, sondern referentiell gelten soll. Diese Konvention besteht darin, dass entschieden wird, wofr das Wort steht. Es wird definiert und innerhalb dieser Abgrenzung als gltig anerkannt. Damit ist ein Mass gegeben, an
dem ablesbar ist, wann das Wort eine Metapher ist und wann nicht. Zu
dieser Konvention heisst es bei de Man: ''Tue innumerable writings that
dominate our lives are made intelligible by a preordained agreement as
to their referential authority; this agreement however is merely contractual, never constitutive. lt can be broken at all times and every piece of
writing can be questioned as to its rhetorical mode [... ]"(AR, S. 204).
Die referentielle Autoritt der Wrter und Texte beruht immer auf
einem Beschluss, der widerrufbar ist. Dadurch wird alles verunsichert.
Ohne das Ernstnehmen von Metaphern gbe es keine Gesellschaft. Die
Autoritt der Gesetze liegt nicht in ihnen selbst, sondern ist dekretiert
und kann wieder aufgehoben werden. Die Gesetze gelten nicht, weil sie
richtig sind, sondern weil sie erlassen werden, und die Moral, auf die sie
sich absttzen, ist wiederum ein Text, dessen referentielle Autoritt auf
blosser bereinkunft beruht. Die menschliche Gesellschaft funktioniert
nur aufgrund der Blindheit gegenber der Rhetorizitt der Texte, auf
denen sie beruht. Jede Aufdeckung der unsicheren Referentialitt von
Texten ist deshalb eine Gefhrdung der Sicherheit geltender Ordnungen.
Die Infragestellung der referentiellen Autoritt des Textes ist die Infragestellung seiner Lesbarkeit Wenn Lesen als Verstehen Klarheit ber
den rhetorischen Status des Geschriebenen voraussetzt, so ist der Text,
bei dem der Entscheid zwischen Figrlichkeit und Wrtlichkeit nicht
getroffen werden kann, nicht lnger lesbar. Indem de Man zeigt, dass die
Texte, die er liest, diese Unterscheidung verunsichern, liest er sie auf
ihre Unlesbarkeit hin, das heisst darauf hin, dass sie nicht im Verstandenwerden aufgehen, sondern opak bleiben und durch den radikalen Widerstand, den sie dem Verstehen entgegensetzen, dieses selbst zeniitten.
Der zeniittete Verstand, ist der Verstand, den man verliert. Deshalb erscheint in de Mans Text im Zusammenhang mit der Unlesbarkeit die
Veniicktheit des Lesers. Wenn man vom Zusammenhang von Unlesbar..;
keit und Veniicktheit spricht, wie es jetzt gerade geschieht, so ist das bereits wieder der Versuch, den Verlust des Verstandes zu verstehen und
die Verrcktheit einzuordnen, whrend die Erfahrung der Unmglichkeit
zu lesen gerade diejenige des Zerfallens von Ordnung ist. Aber auch der
zerfall von Ordnung und sogar ihr Z-erfallensein ist noch nicht Verrcktheit, solange die Ordnung als zerfallene noch erinnert wird. Und das

Wort Verrcktheit selbst misst das Fehlen der Ordnung an der Ordnung
als dem in der Verrcktheit gestrten .Zurechtgeiiickten. Um von VerIiicktheit zu reden braucht man ein Mass, an dem sie ablesbar ist, aber es
ist gerade der Verlust dieses Masses - der referentiellen Autoritt, die
daiiiber entscheidet, was figrlich und was wrtlich ist-, der Unmglichkeit zu lesen erfahren wird.
Es lohnt sich, die Stellen, an denen de Man auf die Verrcktheit zu
sprechen kommt, nher anzuschauen. Dort, wo entwickelt wird, dass
Verstehen die Entscheidung zwischen dem Figrlichen und dem Wrtlichen verlange, und dass dies blicherweise keine Schwierigkeiten bereite, wird folgendes Beispiel eingefhrt: "We do not usually assume, for
example, that someone suffers from hallucinations merely because he
says that a table has four legs; the context of common usage separates
the figural meaning of the catachresis (which, in this case, leads to the
referent) from its literal denotation (which, in this case, is figural)." AR,
S. 201). Die Wahl der Katachrese als Beispiel erklrt sich wohl dadurch,
dass hier das Verhltnis zwischen der figrlichen und der eigentlichen
Bedeutung komplexer ist als in anderen Fllen. Die Katachrese ist nicht,
wie man das von einer gewhnlichen Metapher annimmt, eine Figur, die
an die Stelle eines Wortes tritt, das die Sache beim Namen nennt, sondern sie steht anstatt eines Wortes, das es gar nicht gibt. Wenn das, worauf der Tisch steht, Bein genannt wird, so ist das zwar eine Metapher,
aber diese lsst sich nicht durch ein nicht figrliches Wort ersetzen, weil
ein solches nicht existiert. Die Figur fhrt hier direkt zur Sache und muss
nicht erst auf die eigentliche Bedeutung hin bersetzt werden. Die richtige Interpretation der Figur wird durch die Konvention ("the context of
common usage") gewhrleistet, die festlegt, dass in diesem Fall die
figrliche Bedeutung zur Sache (dem Tisch) fhrt, whrend die wrtliche Bezeichnung Bein hier figrlich wird. Wenn nun die Unterscheidung
so problemlos gelingt, so .fragt es sich, weshalb der Gedanke an die
Halluzination berhaupt aufkommt. Dies wird einsichtig, wenn man bedenkt, dass die Katachrese so funktioniert, dass das gewohnte Verhltnis
zwischen dem Figrlichen und dem Wrtlichen umgekehrt wird, indem
man nicht mehr dadurch zur Sache kommt, dass man von der Figur auf
das eigentliche Wort und von diesem auf die Sache zuiiickgeht, sondern
so, dass man die Figrlichkeit der Figur ignoriert und von ihr aus direkt
zur Sache kommt. Bei der Katachrese ist die figrliche Bedeutung referentiell und die wrtliche Bedeutung figrlich zu lesen. Diese Umkehrung ist potentiell verwirrend, denn sie strt das vertraute Verhltnis von


Figrlichkeit, Wrtlichkeit und Referentialitt. Wenn bei der Katachrese

im Gegensatz zur gewohnten Ordnung nicht das Wrtliche, sondern der
figrliche Sinn referentiell zu nehmen ist, besteht die Gefahr, dass der
figrliche Sinn auch wrtlich genommen wird. Das fhrt aber bald einmal zur Halluzination, in der der Tisch zum vierbeinigen Ungeheuer entstellt wird. Die Entstellung ist eine Disfiguration auch im rhetorischen
Sinn, denn sie besteht darin, dass das Wort Bein nicht mehr als Figur,
sondern wrtlich verstanden wird.
Die halluzinatorische Dimension der Katachrese entfaltet de Man in
dem spteren Aufsatz "Hypogram and Inscription" (RT, S. 27f.). Hier
geht es um die Prosopopoiie als die Figur, die dem eine Gestalt, ein Gesicht gibt, was nicht sinnlich wahrnehmbar ist. Sie bringt zur Erscheinung, was nicht erscheinen kann. Dadurch geht die Prosopopoiie einen
Schritt weiter als die Katachrese. Diese bezeichnet figrlich, was keinen
eigentlichen Namen hat, dadurch aber in seiner sinnlichen Prsenz nicht
in Frage gestellt ist. Jene macht sichtbar, was unsichtbar ist. Die Katachrese ist eine unreduzierbare Figur, die referentiell zu lesen ist, whrend
die Prosopopoiie eine Fiktion ist, der keine wahrnehmbare Wirklichkeit
entspricht. Darin liegt ihr gespenstischer Charakter. "lt is the visual
shape of something that has no sensory existence: a hallucination" (RT,
S. 49). In der Halluzination ist die Referentialitt der Wahrnehmung
selbst in Frage gestellt. Was in der Halluzination wahrgenommen wird,
existiert nicht ausserhalb dieses Wahrgenommenwerdens, das eine Tuschung ist. Aber es als Tuschung zu erkennen ist deshalb schwierig,
weil der in der Halluzination Befangene immer etwas Wirkliches wahrzunehmen glaubt. Wenn aber die Halluzination mit der Gewissheit der
Wahrnehmung verbunden ist, so muss mit der Mglichkeit gerechnet
werden, dass die Gewissheit der Wahrnehmung halluzinatorisch sein
knnte. Die blosse Mglichkeit, dass es Halluzinationen gibt, verunsichert die Unterscheidung zwischen Wahrnehmung und Halluzination.
Dazu sagt de Man: "This means, in linguistic tenns, that it is impossible
to say whether prosopopeia is plausible because of the empirical existence of dreams and hallucinations or whether one believes that such a
thing as dreams and hallucinations exists because language pennits the
figure of prosopopeia. The question 'was it a vision or a waking dream?'
is destined to remain unanswered. Prosopopeia undoes the distinction
between reference and signification on which all semiotic systems [... ]
depend" (RT, 49f.). Die Prosopopoiie selbst als Figur ist von dieser Ver-

unsicherung betroffen. Sie lsst sich als Figur nur bestimmen, wenn es

mglich bleibt, das, was in ihr erscheint, als etwas nicht Wahmehmbares
und damit als Fiktion zu bestimmen. In dem Masse, als die Grenze zwischen W ahmehmung und Halluzination verschwimmt, wird es auch unsicher, ob die Prosopopoiie referentiell zu lesen sei oder nicht, das heisst,
ob sie eine Figur ist oder nicht.

Schlimmer als verrckt

Was ist nun der Status dieser referentiellen Unsicherheit? In Allegories

of Reading ist sie das, was als die Unmglichkeit, das Figrliche und das
Wrtliche auseinanderzuhalten, die Unlesbarkeit ausmacht, und im Zusammenhang mit der Unlesbarkeit erscheint in de Mans Text die Vertcktheit. Dabei ist die Stelle ber den potentiell halluzinatorischen Charakter der Katachrese erst eine Andeutung, die zudem in einem Zusammenhang erfolgt, in dem es um die gewhnlich bestehenden Mglichkeiten geht, das Figrliche vom Wrtlichen zu trennen Erst wo das
nicht mehr mglich ist, erscheint das Wort madness, und zwar auf eine
Weise, die man genau nehmen soll. De Man untersucht hier die im dialogisierten Vorwort zu Rousseaus Julie durchgehaltene Weigerung
Rousseaus, seinem Gesprchspartner zu sagen, ob er der Autor der
Briefe oder nur deren Herausgeber ist. De Man deutet den Autor hier
nicht als ein Subjekt, sondern als die Metapher fr die Lesbarkeit des
Textes. Der Autor ist die Instanz, die gebraucht wird, um den referentiellen Status des Textes zu bestimmen und so dessen Lesbarkeit sicherzustellen. Der folgende Text beschreibt die Verfassung des Lesers, fr den
die Referetialitt des Textes unentscheidbar bleibt:
Asked whether he can respond to the pathos of the text, he replies:"I can conceive of this effect with regard to you. If you are the author, the impact is easy
to understand. If you are not, I can still conceive of it ..." (2: 18). What he could
not tolerate, however, is the impossibility of distinguishing between the alternatives. This would leave him dangling in an intolerable semantic irresolution.
lt would be worse than madness: the mere confusion of fiction with reality, as
in the case of Don Quijote, is mild and curable compared to this radical dyslexia. 27
27. "Voulant etre ce qu'on n'est pas, on parvient a se croire autre chose que ce
qu'on est, et voila comment on devient fou. En montrant sans cesse a ceux qui
les lisent les pretendus charmes d'un etat qui n'est pas le leur, [les romans] les


seduisent, ils leur font prendre leur etat en dedain, et en faire un echange irnaginaire contre celui qu'on leur fait airner." (2:21). This is a simple figural
exchange in which the two specular poles, "leur etat" and "un etat qui n'est pas
le leur," are clearly to be distinguished. The madness can be considered the
madness of another which does not threaten the sanity of the reader. (AR, S.

Dieser Text handelt von der Verfassung des Lesers, der sich der Unlesbarkeit des Textes ausgesetzt sieht. Der Zustand, in dem er sich befindet,
ist nicht etwa der der Veniicktheit, sondern er ist schlimmer als Verrcktheit ("worse than madness"). Es scheint, dass die Veniicktheit hier
nicht mehr als Abweichung von einer Nonn auftritt, sondern selber zu .
einer solchen wird, woran etwas gemessen wird, was noch schlimmer ist.
Das Verstndnis der Abweichung setzt die Bestimmung des Masses voraus, an dem sie abzulesen ist. Was als Verrcktheit zu gelten hat, geht
aus der Anspielung auf Don Quijote und aus der Anmerkung hervor.
Don Quijote kann Wirklichkeit und Fiktion nicht auseinanderhalten. Er
nimmt die Romane, die er liest, ernst. Ebenso werden in dem in der Anmerkung zitierten Text die Romanleser durch die Vorspiegelung einer
unwirklichen Welt dazu verfhrt, ihre wirklichen Verhltnisse zu vergessen und sich fr etwas anderes zu halten, als sie sind. Verrckt ist
gemss dieser Bestimmung, wer das Unwirkliche fr wirklich hlt, oder,
sprachorientiert ausgedrckt, wer das nur sprachlich Gegebene referentiell liest. Um diesen Irrtum erkennen zu knnen, muss man wissen, was
wirklich ist, eine Anforderung, die nach de Mans berlegungen zur unsicheren Referentialitt der Sprache nicht leicht erfllbar zu sein scheint,
die aber dem Leser des zitierten Rousseautextes keine Schwierigkeiten
bereitet, da dieser Text selbst deutlich auseinanderhlt, was man ist, und
was man nicht ist. "The madness can be considered the madness of
another which does not threaten the sanity of the reader."
Im Rahmen von de Mans Argumentation erfllt dieser Text durchaus die
ihm zugedachte Aufgabe, eine Definition der Verrcktheit zu liefern.
Liest man ihn aber in dem Zusammenhang, in dem er in Rousseaus
Vorwort zu Julie steht, so ergibt sich eine Verunsicherung, weniger der
Definition als solcher, als ihrer Anwendbarkeit, das heisst ihrer Referentialitt. Dass die Bercksichtigung des Kontextes auch von de Mans Text
her nicht abwegig ist, zeigt die Erwhnung Don Quijotes, auf den Rousseaus Text ebenfalls Bezug nimmt. Das Zitat in der Anmerkung stammt
aus einer lngeren Entwicklung, die damit einsetzt, dass der Gesprchspartner N. zu bedenken gibt, man knne den Briefwechsel so nicht ver82

ffentlichen, weil das Publikum Menschen, die so schreiben, nicht ernst

nehmen wrde. Die gnstigste Reaktion wre allenfalls noch, "qu'on s'en
amust comme d'autant de vrais fous. Mais les longues folies n'amusent
gueres: il faut ecrire comme Ceivantes, pour faire lire six volumes de
visions"3. Darauf folgt eine Argumentation, die darauf zielt, die Verrcktheit der Romanfiguren als Vernunft und die Vernunft der Leser als
Verrcktheit erscheinen zu lassen. Rousseau unterscheidet zwei Leserschichten: eine stdtische (gens du monde) und eine lndliche
(campagnards). Die stdtischen Leser werden die Menschen in einem
Buch, das ein einfaches Leben und dessen einfache Werte beschreibt,
verrckt finden. Sie wollen Romane, die eine knstliche Welt aufbauen,
wie es die ihre ist. Gerade das ist aber das Schdliche des Romans: dass
er den Leser in die lliusion lockt und ihm ein fiktives Glck vonnacht.
Der einfache Leser gert in eine verrckte Welt. Das ist die Verrcktheit
Don Quijotes, nur sind es jetzt nicht mehr die Figuren von Rousseaus
Roman, die verrckt sind, sondern die Leser der ..schlechten" Romane,
die sich verfhren lassen. Vom Standpunkt des stdtischen Publikums
aus sind Rousseaus Figuren Verrckte, aber nach der Wertordnung, die
dieser Roman vertritt, ist die stdtische Gesellschaft verrckt, die eine
illusionre Welt fr wirklich hlt, so wie naive Romanleser es tun. Alles
mndet in den Satz: "Voila, Monsieur, comment la folie du monde est
sagesse" (OC, II, S. 22), der wohl so zu verstehen ist, dass das, was die
Gesellschaft (le monde, les gens du monde) fr Verrcktheit hlt, Weisheit ist. In Rousseaus Text sind Vernunft und Verrcktheit letztlich vertauschbar. Was fr das eine oder das andere gehalten wird, hngt davon
ab, was man fr wirklich und was man fr fiktiv hlt. Nun scheint allerdings die Argumentation eine klare Tendenz erkennen zu lassen, das
Einfache und Natrliche, das die. Welt von Julie vertritt, gegenber der
Verstiegenheit und Knstlichkeit der stdtischen Gesellschaft zu privilegieren. Das wrde bedeuten, dass es diese Gesellschaft ist, welche in der
lliusion befangen ist und die Fiktion fr Wirklichkeit hlt, whrend die
Landleute sich an das Wirkliche halten. Aber diese Auflsung lsst sich
nicht halten, denn die Hauptfrage des Vorworts ist gerade die nicht zu
beantwortende Frage nach dem referentiellen Status der Briefe. Wenn
diese fingiert sind, so ist auch die Wirklichkeit der einfachen Leute fiktiv, und es wird wieder unentscheidbar, wer verrckt und wer vernnftig

Jean-Jacques Rousseau, OEuvres completes, Paris, Bibliotheque de la Pleiade, 1964,

Bd. II, S. 18.


ist. Dass bald die Verrckten vernnftig, bald die Vernnftigen verrckt
erscheinen, beeintrchtigt die Definition der Verrcktheit nicht, sondern
weist lediglich darauf hin, dass sie schwer anwendbar ist, weil jede Anwendung voraussetzt, dass feststeht, was wirklich und was unwirklich
ist. Rousseaus Text gibt also nicht nur eine Bestimmung der Verrcktheit, sondern er hebt vor allem die Fragwrdigkeit jeder Festlegung dessen heivor, was als Wirklichkeit gelten soll, auf welche die Verirrung der
Irren zu beziehen wre.
Fr die Zwecke von de Mans Argumentation gengt es, wenn Verrcktheit als ein Zustand bestimmt ist, der sich von einem anderen, vernnftigen, abhebt. Diese Bestimmung liefert der erste Satz der zitierten
Rousseaustelle: "Voulant etre ce qu'on n'est pas, on paivient a se croire
autre chose que ce qu'on est, et voif comment on devient fou." Merkwrdigerweise zitiert de Man falsch. Der Wortlaut stimmt zwar, aber bei
Rousseau steht dieser Satz nicht am Anfang, sondern am Schluss der
Stelle, die de Man bernimmt. Die Umstellung ist ohne Zweifel eines der
zahlreichen Versehen, durch die Allegories of Reading sich auszeichnet,
das aber seiner Art nach weniger dem Setzer als dem Autor anzulasten
ist, der es mindestens bersehen hat. Unabhngig davon, was dazu gefhrt hat, erscheint Rousseaus Text bei de Man entstellt, und nichts verbietet, diese Entstellung als mit konstitutiv fr de Mans Text zu betrachten. Zunchst fllt auf, dass die Umstellung keine ernsthafte Sinnstrung
zur Folge hat. Die einzige Schwierigkeit entsteht dadurch, dass das Pronomen ils ("ils les seduisent [... ]"),das bei de Man durch "[les romans]"
ersetzt ist, unverstndlich wird, weil der vorausgehende Satz nicht von
den Romanen handelt, wie es bei Rousseau der Fall ist (''L'on se plaint
que les Romans troublent les tetes: je le crois bien. En montrant sans
cesse [... ]").Im brigen ist die Vertauschbarkeit der Stze dadurch gegeben, dass sie beide ungefhr das Gleiche zu sagen scheinen. Der Unterschied liegt darin, dass der eine den Zustand der Verrcktheit allgemein
beschreibt ("voila comment on devient fou"), whrend der andere strker
auf das Beispiel des Romanlesers ausgerichtet ist. Bei Rousseau wird die
allgemeine Bestimmung aus dem besonderen Fall abgeleitet, whrend in
de Mans Umstellung das Beispiel die allgemeine Formulierung illustriert. Die Mglichkeit der Bestimmung von Verrcktheit ist das, was
bei de Man im Vordergrund steht und deshalb vorweggenommen wird.
Die Anwendung ist sekundr. Sie ist das, was bei Rousseau die Definition jedesmal dadurch ausser Kraft setzt, dass sie die Referentialitt des

festen Bezugspunktes als Mass ("ce qu'on est", "leur etat") in ihrer


Fragwrdigkeit erkennbar macht. An dieser Infragestellung ist de Man

hier nicht interessiert, weil es ihm vielmehr darum geht, Verrcktheit als
etwas Bestimmbares dem gegenberzustellen, was schlimmer als Verrcktheit ist.
Das entstellte Zitat gibt noch zu einer weiteren berlegung Anlass.
Rousseaus Text, wie er bei de Man erscheint, ist verrckt. Er ist es in
dem wrtlichen Sinne, dass seine Stze verschoben sind. Er ist es aber
auch im Sinne der Bestimmung von Verrcktheit, die er selbst gibt, denn
die Umstellung der Stze enthllt sich im Vergleich mit der "Wirklichkeit'' von Rousseaus Text. Unterbleibt dieser Vergleich, was meistens
der Fall sein drfte, so wird der bei de Man erscheinende Text als Rousseaus Text, das heisst als etwas genommen, was er nicht ist, wofr er
sich aber selbst hlt. Diese durchaus halluzinatorische Situation lsst sich
aber nicht ohne weiteres bereinigen, indem man Rousseaus Text neben
de Mans entstelltes Zitat hlt. Denn selbst wenn der "richtige" Text bekannt ist, ist der Status des Zitats noch nicht geklrt. Dieses ist weder
Rousseaus Text, denn dieser ist ja verndert worden, noch de Mans Text,
denn alles in ihm ausser die Abfolge der Stze stammt von Rousseau.
Das entstellte Zitat lsst sich nicht mehr auf einen Autor zurckfhren.
De Man kann es nicht beanspruchen, und Rousseau kann es nicht verantworten. Der Text kann so, wie er in Allegories of Reading steht, keinerinstanz zugeschrieben werden, die dafr einsteht. Sein referentieller
Status ist unentscheidbar. Der Zustand, in den diese Einsicht versetzt,
kann weder als normal noch als verrckt bezeichnet werden, denn weder
herrscht in ihm die filusion, in der man das Unwirkliche fr wirklich
hlt, noch gibt es in ihm eine feste Norm, an der gemessen werden
knnte, was wirklich ist und was nicht.
Das fhrt zu dem, was in de Mans Text als das erscheint, was
schlimmer als Verrcktheit ist. Es ist die Verfassung des Lesers, der ber
den referentiellen Status des Textes im Ungewissen bleibt. Der Verrckte und der Vernnftige sind vergleichsweise in einer einfachen Situation. Der Verrckte verwechselt zwar das Nichtwirkliche mit dem
Wirklichen, weil ihm ein Kriterium fehlt, beides zu unterscheiden, aber
das braucht ihn nicht zu stren, solange er an seine filusion glaubt, das
heisst, solange er in der filusion lebt, er verfge ber das Mass der
Wirklichkeit. Der Vernnftige v~rfgt ber dieses Mass, was im Kontext
von de Mans berlegungen allerdings nicht mehr heisst, als dass er sich
daran hlt, jene Metaphern wrtlich zu nehmen, bei denen dies nach
geltender Konvention zulssig oder wnschbar ist, aber dieses Mass gibt

ihm die Mglichkeit, die Abweichung zu bestimmen und Verrcktheit zu

diagnostizieren. Von Verrcktheit kann man berhaupt nur reden, wenn
es eine Norm gibt, an der sie gemessen werden kann. Der Zustand des
Lesers, den de Man im Auge hat, ist nun aber weder der des Verrckten,
noch der des Vernnftigen. Es ist ein Zustand, in dem man weder der
Tuschung verfllt und das Unwirkliche fr wirklich hlt, noch ber eine
Mglichkeit verfgt, wirklich und unwirklich zu unterscheiden. Man ist
in der Ungewissheit darber suspendiert, ob man sich tuscht oder nicht,
ohne dass eine Mglichkeit offen bleibt, es zu verifizieren. In dieser Situation funktioniert der Gegensatz vernnftig/verrckt nicht mehr. Er ist
dadurch ausser Kraft gesetzt dass man weder verrckt ist und glaubt vernnftig zu sein, noch vernnftig ist und sich dadurch vor Verrcktheit
schtzen kann, sondern dass man auf keine Weise mehr wissen kann, ob
man verrckt ist oder nicht. Dieser Zustand, der schlimmer als verrckt
genannt wird, ist der Zustand der unentscheidbaren Referentialitt der
Sprache. Es ist die Erfahrung der Unmglichkeit, aus dem Bannkreis der
Sprache auszubrechen, obwohl die Sprache unablssig ber sich hinaus
verweist. In Rousseaus Vorwort zu Julie ist der Autor als der aussertextliche feste Bezugspunkt, der ber die Referentialitt des Textes zu entscheiden htte, das Mass, das dem Leser entzogen ist und ihm nicht mehr
erlaubt, Fiktion und Wirklichkeit zu trennen.
Der Leser, der in der Ungewissheit ber die Referentialitt des Textes suspendiert bleibt, ist der Leser, der nicht referentiell liest, den Text
nicht auf ein Aussersprachliches hin zu verlassen versucht, sondern sich
der Sprachlichkeit alles Gesagten bewusst bleibt. Aber es wre naiv, diesem Leser, dessen Zustand im brigen schlimmer als der des Verrckten
ist, einen hheren Erkenntnisgrad zuschreiben zu wollen und seine Beziehung zum Text als ein erstrebenswertes Ziel misszuverstehen. Auch
wenn die Einsicht in die ungewisse Referentialitt des Sprachlichen gegenber der filusion des Verrckten, der sich fr vernnftig hlt, wie
auch des Vernnftigen, der sich fr vernnftig hlt, eine Ent-tuschung
bedeutet und deshalb der Wahrheit nher kommt, so besteht doch diese
Annherung an die Wahrheit darin, dass man den Glauben daran verliert,
ihr nahe zu sein. Die Einsicht in die unsichere Referentialitt des
Sprachlichen ist sogleich ihrem eigenen Befund unterworfen. Sie ist keineswegs ein Ausweg aus der Tuschung, sondern erliegt ihr selber in
dem Masse, als sie an ihre eigene Wahrheit glaubt. Sie besttigt sich erst
darin, dass sie ihrerseits wieder verunsichert wird. Deshalb ist der unent-

scheidbare referentielle Status der Texte nicht eine Erkenntnis, in der


man sich einrichten kann, sondern eine Erkenntnis, die sich allenfalls
dann bewhrt, wenn man ihr zuwiderhandelt. Dashalb ist der Zustand der
unentscheidbaren Referentialitt der Sprache unhaltbar und wird, sobald
er sich herstellt, durch einen gewaltsamen Entscheid beseitigt, aber auch
besttigt. Die Besttigung liegt darin, dass der Entscheid sich nicht fundieren lsst, die Beseitigung geschieht dadurch, dass man so tut, als wre
der Entscheid fundiert. Dadurch, dass man das, wovon man nicht weiss,
ob es referentiell gelesen werden kann, dennoch referentiell liest, unterdrckt man das Wissen um den mglicherweise halluzinatorischen Charakter dessen, was man als wirklich setzt, und flchtet in eine Vernunft,
die vielleicht verrckt ist, oder in eine Verrcktheit, die vielleicht vernnftig ist, aber auf jeden Fall in einen Zustand, in dem man sich verhalten kann, und in dem der unhaltbare Moment, in dem Verrcktheit
und Vernunft nicht mehr zu unterscheiden sind, berwunden ist, anderseits aber durch das Gewaltsame des Entscheids, den keine Erkenntnis
fundiert, gerade besttigt wird.
Gerade weil der Zustand jenseits der Verrcktheit unhaltbar zu sein
scheint und immer schon beseitigt. ist, wenn er in den Blick tritt, kann
man versucht sein, ihn redend zu umkreisen und als die Unmitte zu
beleuchten, in der er sich jedesmal verflchtigt. Der Ausdruck
"schlimmer als verrckt" bezeichnet nicht nur etwas, was anders als das
Verrcktsein ist, sondern etwas, was mit dem Verrcktsein so zusammenhngt, dass es eine Steigerung und Verschlimmerung der Verfassung
ist, die verrckt genannt wird. "Schlimmer als verrckt" ist nicht ein Gegensatz zu "verrckt", sondern ein Zustand, der im Sinne des Verrcktseins ber dieses .hinausgeht. Anderseits wird dieser Zusand aber nicht
dadurch erreicht, dass die Verrckung als die Abweichung von der Norm
und vom Normalen strker wird, denn eine solche quantitative Vernderung wrde lediglich zu einer schlimmeren Verrcktheit, nicht aber zu
etwas fhren, was .schlimmer als diese ist. Obwohl das, was schlimmer
als Verrcktsein ist, eine Art Steigerung des Verrcktseins ist, ist es doch
auch etwas qualitativ anderes als dieses. So stark der Verrckte auch
immer von dem abweichen mag, was als normal gilt, seine Verrcktheit
bleibt eine Position, nicht nur fr den Vernnftigen, der den Grad ihrer
Abweichung von der Norm misst und sie als verrckt diagnostiziert,
sondern auch fr den Verrckten selbst, der an seine Welt glaubt und sie
vertritt. Deshalb kann der qualitative Sprung, der vom Verrcktsein zu
dem fhrt, was schlimmer ist als verrckt zu sein, nur darin bestehen,
dass nicht mehr nur der Grad der Verrckung sich steigert und damit die


Position des Verrckten sich verschiebt, sondern dass die Position berhaupt verlorengeht.


Der Verlust der Position ist keine Verrcktheit mehr. Alle Verrckung
ist immer nur Verrckung der Position. Deren Verlust kann man Entrkkung nennen. Die Entrckung ist nicht wie die Verrckung eine Verschiebung von einem Ort zum andern, von einer Stelle zur andern, sondern die Entstellung in die Ortlosigkeit. Der Verrckte ist anders als die
andern, weil er eine andere Position hat. Der Entrckte, der ohne Position ist, ist nicht mehr nur anders, sondern er ist niemand mehr. In der
Entrcktheit gibt es kein Subjekt mehr. Entrcktheit ist keine Situation,
worin jemand sich befindet, denn das Situiertsein in der Situation ist gerade das, was in ihr verlorengeht.
Gemss de Mans Deutung von Rousseaus Voiwort zu J ulie ist Entrckung - was schlimmer als Verrcktsein ist - das, was dem Leser
zustsst. Entrcktheit ist der Zustand, in den der Text den Leser durch
seine Unentscheidbarkeit bringt. Es liesse sich zeigen, dass dies nicht nur
fr diesen besonderen, sondern fr jeden Text gilt, der in de Mans Sinn
Text ist. So heisst es im Zusammenhang mit Nietzsche: "Rhetoric is a
text in that it allows for two incompatible, mutually self-destructive
points of view, and therefore puts an insunnountable obstacle in the way
of any reading or understanding" (AR, S. 131; vgl. AR, S. 126, 270).
Dass der Text unlesbar ist, bedeutet nicht einfach, dass er fremd und undurchdringlich gegenber bleibt. Vielmehr wird der Leser durch die Unentscheidbarkeit des Textes lesend in die Positionslosigkeit entrckt.
Aber was entrckt ist, ist nicht mehr der Leser, denn die Unlesbarkeit ist
nicht eine andere Situation, in die der Leser versetzt wird, sondern die
Vernichtung des Lesers als Leser. Diese Gefhrdung des Lesers macht
die Unertrglichkeit der Unentscheidbarkeit des Textes aus und verhindert, dass sie ausgehalten wird. Der Leser setzt der Gefahr der Entrkkung sogleich den sachlich nicht begrndbaren und daher abrupt gewaltsamen Entscheid entgegen. Diese Entgegensetzung ist die Setzung des
Lesers als Position, seine Rckkehr in die nonnale Verrcktheit der Vernnftigen oder Verrckten. Der Leser konstituiert sich als Leser, indem
er dort Stellung nimmt, wo es keine Stellungnahme gibt. Die unmgliche


Position ermglicht die Lektre, in welcher in dem Masse, als sie scheitert, weil sie den Text als unlesbaren nicht einholt, die Positionslosigkeit
der Entrcktheit erscheint. Die J.,eserposition - der Leser als Position, als
Setzung - ist eine Figur, die das, wofr sie steht, dadurch zugnglich
macht, dass sie sich selbst zersetzt und untergrbt Der Leser ist die
Prosopopoiie der Entrucktheit, aber die Entrcktheit erscheint in ihm in
dem Masse, als seine Position im Lesen der Entsetzung verfllt und das
fiktive Gesicht der Entruckung entstellt wird. Die Entstellung ist die Disfiguration der Prosopopoiie, deren de-facement. Die Entstellung des Lesers ist die Auflsung seiner Position, seine Entrckung in die Positionslosigkeit und seine Vernichtung in der Unlesbarkeit des Textes, die
nicht anders erscheinen kann als in dessen Gelesenwerden. Die Unentscheidbarkeit des Textes muss immer entschieden werden, damit er gelesen werden kann, aber das Lesen fhrt immer zu der Unlesbarkeit zuruck, der es durch den gewaltsamen Entscheid zu entgehen hoffte.
Die Stelle ber die Entrckumg in Allegories of Reading lsst sich
von den berlegungen zur Position her noch einmal betrachten. Die Entruckung des Lesers in die Unentscheidbarkeit des Textes wird durch die
Wendung "worse than madness" umschrieben. In ihr ist Verrucktheit
eine Position, die aber sogleich in das aufgelst wird, was schlimmer ist
als sie, wobei nur die Setzung der Verrucktheit deren Entsetzung in die
Entruckung ennglicht. Die Setzung der Verrucktheit als Position wird
in de Mans Text dadurch begrundet, dass in der Anmerkung eine Rousseau-Stelle ber die Verrucktheit zitiert wird. Aber das Zitat ist als entstelltes gestellt und durch die Versetzung seiner Stze entautorisiert und
ohne Autoritt. Nun knnte man allerdings einwenden, dass die Autoritt
des Textes sich wiederherstellen lasse, da man den richtigen Text kenne.
Aber der unentstellte Text steht, wie sich gezeigt hat, in einem Kontext,
der die Unterscheidung zwischen dem Wirklichen und dem Unwirklichen, auf der die Verrucktheit als Position beruht, verunsichert, so dass
er selber die Bestimmung dieser Position verhindert und seine eigene
Autoritt untergrbt. Auch wenn die Entstellung des Zitats ein zuflliges
Versehen ist, bleibt sie die Metapher fr die davon unabhngig nachweisbare Reihe der Ent-stellungen (Auflsungen von Positionen), durch
die jeder Setzung immer wieder der Boden entzogen wird. Die Formulierung "schlimmer als verrockt" fr den Zustand der Unentscheidbarkeit
des referentiellen Status des Textes setzt Verrucktheit als Position voraus. Diese setzt ihrerseits voraus, dass wirklich und unwirklich sich unterscheiden lassen, und diese Unterscheidung setzt voraus, dass man


ber den referentiellen Status Bescheid weiss. Was aber referentiell zu

lesen sei und was nicht, ist wiederum eine als gltig gesetzte Konvention, deren Fragwrdigwerden in die Unentscheidbarkeit entrckt, in die
der in Rousseaus und de Mans Text besprochene Leser gert. Die Beziehung zwischen Setzen und Voraussetzen errtert de Man an einer anderen Stelle in Allegories of Reading (AR, S. 124). Die Setzung verleugnet
ihren hypothetischen, referentiell ungewissen Status, indem sie voraussetzt, was nicht vorausgesetzt werden kann, weil es seinerseits eine
hypothetische Setzung ist. Durch jede Setzung entfernt sich de Mans
Text von der Positionslosigkeit, die er zu sagen versucht. Indem er aber
jede dieser Setzungen wieder zersetzt, nhert er sich der Positionslosigkeit, aus der er immer wieder eine Position machen muss, um darber
reden zu knnen.
Die Unausweichlichkeit dieser Spannung zeigt sich in de Mans Text
dort am deutlichsten, wo die Positionslosigkeit als der Zustand, der
schlimmer ist, als verrckt zu sein, als radikale Dyslexie bezeichnet
wird. Radikal ist das, was zur Wurzel gehrt, was zu dem zurckreicht,
aus dem alles hervorwchst. Die Dyslexie hingegen steht hier fr die
Entwurzelung der Sprache. Radikale Dyslexie ist ein Text, in dem das
Wort Dyslexie die im Wort radikal gesetzte Wurzel entwurzelt, der aber
gleichzeitig die Dyslexie - die Entwurzelung - dadurch, dass sie radikal
genannt wird, als Wurzel setzt, was sie nicht sein kann.
Es gibt eine weitere Stelle in de Mans Werk, an der das Wort madness erscheint. Gegen Ende des Aufsatzes "Shelley Disfigured" findet
sich der Satz: "No degree of knowledge can ever stop this madness, for it
is the madness of words"4. Noch bevor man sich fragt, wie Verrcktheit
hier vom Kontext her bestimmt ist, gibt der Satz selbst zu bedenken, dass
die Verrcktheit eine solche der Wrter ist. Nicht jemand ist hier verrckt, sondern die Sprache. Whrend in Allegories of Reading der Zustand des Lesers beschrieben wird, ist hier kein Subjekt mehr erhalten,
dem die Verrcktheit zugeschrieben werden kann. Aber die beiden Texte
sind einander vielleicht nher, als sie es vorerst zu sein scheinen, denn
die Unlesbarkeit des Textes, der sich der Leser angesichts der unentscheidbaren Referentialitt ausgesetzt sieht, ist ja als der Zustand, der
schlimmer ist, als verrckt zu sein, die Zersetzung des Lesers als Position und seine Entrckung in die Positionslosigkeit, also seine Vemich4


Paul de Man, The Rhetoric of Romanticism (RR), New York, Col1llllbia University
Press, 1984, S. 122.

tung als Subjekt. Es knnte daher sein, dass die Verrcktheit der Sprache
nicht so weit weg von dem ist, was schlimmer als Veniicktheit ist.
Der Zerfall des Subjekts ist in dem zitierten Satz daran ablesbar, dass
das, was in ihm the madness of words genannt wird, unkontrollierbar ist
Es gi~t keine Instanz, welche das, was als Sprache und in der Sprache
geschieht, im Griff hat. Was im Entgleiten der Herrschaft ber die Sprache aufgeht, ist das Nichtmenschliche der Sprache, das, was dem Einfluss dessen, der Sprache verwendet, entrckt ist, und in das. die Sprache
ihn entrckt. Was als the madness ofwords abluft, untersteht also keinem Wollen und Knnen und ist nicht die Manifestation einer Freiheit,
sondern scheint einer blinden Notwendigkeit zu gehorchen. Die
Notwendigkeit des Ablaufs ist seine Unvenneidlichkeit Die Einsicht in
was geschieht ennglicht auf keine Weise, etwas daran zu ndern:"No
degree of knowledge can ever stop this madness." Der unausweichliche
Vorgang, der the madness of words heisst, wird in de Mans Text an der
Lektre von Shelleys Gedicht The Triumph of Life vorgefhrt. Da es
aber mehrere Hinweise darauf gibt, dass dieser Text als reprsentativ fr
alle Texte betrachtet wird, kann man eine vom Beispiel losgelste Darstellung versuchen.
Die Sprache bedeutet Dadurch ist sie ein Gefge von Beziehungen.
Was bedeutet, ist auf sein Bedeutetes bezogen und alles, was bezogen
ist, bedeutet. Sinn ist Zusammenhang. Alle Ordnung ist Herstellung von
Beziehung. Als die Mglichkeit von Bedeutung bestimmt die Sprache
das Menschenmgliche. Aber damit einer sprechend die8e Mglichkeit
wahrnehmen kann, muss ihm die Sprache schon gegeben sein. Was einer
sagt, ist bis zu einem gewissen Grad, was er sagen will, dass er es aber
sagen kann, hngt nicht von ihm ab, sondern davon, dass es Sprache
gibt Die Sprache ist die Voraussetzung dafr, dass einer etwas sagen
kann. Indem das Sprechen des Sprechenden die Sprache voraussetzt, ist
es auf diese Voraussetzung bezogen und bedeutet sie. Jedesmal, wenn
gesprochen wird, ist dadurch auf die Sprache als die Ermglichung ihres
Stattfindens verwiesen. Dadurch zeichnet sich im Sprechakt eine Dimension ab, die es nicht mehr erlaubt, ihn nur unter dem Gesichtspunkt des
Sprechenden zu sehen, der sich der Sprache bedient. Jedesmal, wenn gesprochen wird, ist das nicht nur die Setzung von Bedeutung durch denjenigen, der spricht, sondern es ist zugleich die Setzung der Sprache als
Voraussetzung diese Setzung. Die Setzung der Sprache ist aber nicht der
Akt dessen, der spricht, sondern sie ist vielmehr die Voraussetzung dafr, dass er berhaupt spricht. Niemand erfindet die Sprache, und jede


Instanz, der die Setzung der Sprache zugesprochen werden knnte, ist ihrerseits immer nur eine sprachliche Setzung. Die Sprache als die Voraussetzung ihres Stattfindens ist voraussetzungslos. Sie ist nicht durch ein
Subjekt gesetzt, das als das sie Setzende ihr bergeordnet und ihre Voraussetzung wre. Die Setzung der Sprache ist auf nichts zurckbeziehbar. Sie ist ein abrupter, beziehungsloser Akt, der in seiner Unbezogenheit ohne Bedeutung und Sinn ist. Dass es Sprache gibt, ist nicht verstehbar. Jeder Sprechakt hat insofern, als in ihm die Sprache selbst sich
setzt, an dieser Abruptheit teil. Jedes Sinngefge verdankt sich letztinstanzlich einem Setzungsakt, der ohne Sinn ist und das Gefge dadurch
aufreisst, dass er sich auf keine Weise in dieses integrieren lsst, denn
jeder Versuch, dem abrupten Akt Sinn zu geben, diskreditiert sich schon
dadurch, dass er unternommen wird, besteht er doch darin, das Unbeziehbare in die Beziehung zu bringen.
Dennoch ist dieser Versuch unvenneidlich. Wenn die Setzung der
Sprache als die Voraussetzung des Sinngefges uneinholbar ist, so hngt
dieses Sinngefge von einem Akt ab, der sich nicht in es einfgen lsst
und als abrupte Setzung unverstndlich und sinnlos bleibt. Aber dieser
Akt, der ohne Voraussetzung ist, ist nicht ohne Folgen. Insofern als er
das sprachliche Sinngefge ennglicht hat, ist er auf dieses bezogen und
bekommt nachtrglich Sinn. Aber diese Sinnverleihung verfehlt notwendig gerade die Unbezogenheit und Abruptheit der Setzung von Sprache.
Die Bedeutung, die dem abrupten Akt zugesprochen wird, ist immer
falsch, weil sie fr das gilt, was keine Bedeutung hat. Aber gleichzeitig
ist die falsche Bedeutung unumgnglich, weil der abrupte Setzungsakt
im Rckblick als die Voraussetzung des Sinngefges, das er ermglicht
hat, auf dieses bezogen ist. Dort, wo de Man davon spricht, wie der Setzungsakt durch die Verleihung von Bedeutung in das Sinngefge integriert wird, heisst es: "But this is radically inconsistent: language posits
and language means (since it articulates) but language cannot posit
meaning; it can only reiterate (or reflect) it in its reconfirmed falsehood"
(RR, S. 117f.).
An jedem Sprechakt ist die abrupte Setzung von Sprache mitbeteiligt, die der Sprechende sich nicht zusprechen kann, die er aber dennoch
immer wieder dadurch unter Kontrolle zu bringen sucht, dass er ihr einen
Sinn gibt. Weil dies nur redenderweise geschehen kann, ist der Versuch,
die Herrschaft ber das Unbeherrschbare zu gewinnen, diesem wiederum
selbst ausgeliefert. Was auf solche Weise sich meldet, ist das Nichtmenschliche der Sprache, das, was sich auf keine Weise auf den Men-


sehen zurckfhren und in die Abhngigkeit von dem, der spricht, bringen lsst. Aber weil die Mglichkeit des Sinns von diesem Unzugnglichen abhngt und das heisst: mit ihm zusammenhngt, ist der unmgliche und notwendig falsche Sinn immer schon gefordert und nicht zu unterdrcken, selbst wenn er in seiner Falschheit durchschaut wird. Die
Einsicht in diese Falschheit kann allerdings verhindern, dass man der
Mystifikation verfllt, verstanden zu haben, aber das durch diese Klarsicht wiederhergestellte Nichtverstehen ist nicht eine Verfassung, in der
es sich zur Ruhe kommen lsst, sondern es fordert das Verstehen als die
Knpfung neuer Beziehungen, obwohl zum voraus klar ist, dass sie
falsch und unhaltbar sind.
Im Rckblick auf die frher besprochene Stelle aus Allegories of
Reading kann man jetzt in dem Bedrfnis des Lesers in Rousseaus Vorwort, die Rede auf einen Redenden, den Text auf einen Autor zurckzufhren, den Versuch sehen, die Setzung der Sprache dem Sprechenden
zuzusprechen, denn nichts anderes geschieht, wenn der Autor als eine
aussersprachliche Instanz angenommen wird, die fr den Text einsteht.
Aber in dem Masse, als es gerade nicht der Sprechende ist, der die Sprache setzt, und als diese umgekehrt die Voraussetzung dafr ist, dass
Sprache berhaupt stattfindet, ist der Autor die Figur fr eine Setzung,
die ihm gnzlich entzogen und unzugnglich ist, und der er, indem er sie
redend mitvollzieht, ausgesetzt ist, ohne darber verfgen zu knnen.
Der Autor als die nachtrglich dem Sinngefge vorausgestellte Voraus- \
setzung ist die Prosopopoiie derVoraussetzungslosigkeit, das heisst der \
abrupten Setzung von Sprache. Der Autor ist zwar einerseits die Figur
fr die Autoritt ber den Text und damit, wie es in Allegories of Reading heisst, fr dessen Lesbarkeit, aber er ist anderseits und gleichzeitig
die Figur fr das Fehlen dieser Autoritt und damit deren Verunsicherung. Der Sinn des Sinngefges insgesamt wird dadurch fraglich. Nicht
der Sinn, der sich in ihm als kohrentes Netz von Beziehungen herstellt,
aber der Sinn dieses Sinns. Insofern als es unverstndlich bleibt, dass es
Sprache gibt, ist die sprachliche Ordnung nirgends zu verankern und
schwebt in einem ungewissen Raum an einem ungewissen Ort. So reich
die Beziehungen in ihr sein mgen, so bleibt sie doch als ganze unbeziehbar, von fraglichem Sinn und fraglicher Referentialitt. Die Unmglichkeit, den sprachlichen Setzungsakt unter Kontrolle zu bringen, entwurzelt das Sinngefge als ganzes, stellt alles, was in ihm geschieht, in
seinem Sinn in Frage und ffnet damit die Frage nach dem Sinn, die
immer in dem Masse, als sie beantwortet wird, neu zu stellen ist, weil die

Antwort dem Sinn gibt, dessen Sinn, wenn es einen hat, unerreichbar ist.
"No degree of lmowledge can ever stop this madness, for it is the madness of words." Sobald Sprache geschieht, gibt es Bedeutung und Sinn.
Aber ob der Sinn nun sprachlich bleibt oder dadurch getragen ist, dass
die Sprache selbst Sinn hat, bleibt offen. Die Verrcktheit der Sprache
ist, dass sie stndig Sinn gibt, ohne dass man je weiss, ob dies Sinn hat.
Sie gibt Sinn, als gbe es ein Mass, woran der Sinn dieses Vorgangs der
Sinngebung messbar wre. Sobald Sprache geschieht, widerspricht sie
der Ungewissheit, ob es ein solches Mass gibt, und dem Wissen, dass es
unerreichbar ist. Die fortwhrende Konstituierung von Sinn, der vielleicht keinen hat, ohne dass dieser Zweifel dem Vorgang Einhalt gebieten wrde, weil er selber sich als sinnvoller konstituiert, ist ein Kreisen
der Sprache in sich selbst, das delirierenden Charakter hat. Weil es kein
anderes Mass a}.s die Sprache selbst gibt, fehlt das Mass, an dem sie zu
messen wre. Die Sprache ist nicht situierbar, sondern in die Ortlosigkeit
entstellt, verrckt nicht im Sinne der Abweichung von einer Nonn, sondern im Sinne des gnzlichen Fehlens einer solchen.

Die Verrcktheit lsst sich noch auf andere Art annhern und beschreiben. Wenn das Sinngefge auf der uneinholbar entzogenen Voraussetzung beruht, dass es Sprache gibt, und wenn dem Stattfinden von Sprache immer nur nachtrglich Sinn verliehen werden kann, so ist der Sinn
des Sinns unsicher, weil das Sinngefge nicht auf etwas bezogen werden
kann, worin es fundiert wre. Sobald aber Sprache geschieht und Sinn
sich konstituiert, wird diese Unsicherheit verleugnet und eine Position
angenommen und eingenommen, die als fester Bezugspunkt fungiert.
Diese Position ist die berwindung der Ortlosigkeit durch die Situierung. Aber der gewonnene Ort bleibt eine nicht begrndbare Fiktion und
die Position die Prosopopoiie der Positionslosigkeit. Wird dies durch;,.
schaut, so erweist sich die Position als das, was man eine Verrckung
bezglich der Positionslosigkeit bezeichnen knnte, wrde man diese
dadurch nicht in eine Position - die Position, keine zu haben - verflschen. Die als Fiktion erkannte Position wird unhaltbar und muss abgebaut werden. Da aber ihre Auflsung wiederum nur redend geschehen
kann, ist sie ihrerseits positionsbildend und der gleichen Erosion ausge-


setzt wie diejenige, die sie zersetzt. Der Prozess von Setzung und Entsetzung, der auf diese Weise in Gang kommt, ist endlos, weil es weder eine
Position, die haltbar wre, noch ein Reden gibt, das nicht das Einnehmen
einer Position wre, so dass das Reden unaufhrlich zwischen der Position und ihrer Unhaltbarkeit fluktuiert. Dieser endlose Prozess der Verrckung von jeglicher Position, ohne dass es doch ein endgltiges Loskommen von der Position gibt, ist die Ironie.
Die Ironie ist ein ausgezeichneter Zug aller Texte de Mans, aber sie
ist nur einmal, in dem 1969 erschienenen Aufsatz "The Rhetoric of
Temporality", ausfhrlich behandelt worden. Im Zusammenhang mit
Baudelaires Essay De l'essence du rire, wo anlsslich der Beschreibung
einer Pantomime von Schwindel die Rede ist, verbindet de Man die Ironie mit der Vencktheit: "Irony is unrelieved vertige, dizziness to the
point of madness. Sanity can exist only because we are willing to function within the conventions of duplicity and dissimulation, just as social
language dissimulates the inherent violence of the actual relationships
between human beings. Once this mask is shown to be a mask, the authentic being underneath appears necessarily as on the verge of madness"5. Dieser Text unterhlt etliche Beziehungen zu der Stelle ber Verrcktheit in Allegories of Reading. Der Zustand der Gesundheit oder
Vernnftigkeit wird durch die Bereitschaft bestimmt, gewisse Konventionen als gltig anzuerkennen und ernst zu nehmen. Dem entspricht im
andern Text der Entscheid darber, was Wirklichkeit und was Fiktion ist.
Aber so wie dort diese Unterscheidung verunsichert wird, so ist hier die
Konvention als eine Ordnung verstanden, deren Gltigkeit durch nichts
als die bereinkunft gesichert ist, und deren sachliche Fundierung ausbleibt. Die Konvention ist die Maskierung eines verleugneten - hier als
"authentisch" bezeichneten - Zustandes, der Verrcktheit heisst. Was
damit gemeint ist, lsst sich aus dem Kontext erschliessen. Die Ordnung,
in der der Vernnftige funktioniert, ist eine Fiktion, die einen verleugneten Zustand maskiert und verbirgt. Von der Demaskierung wre zu
erwarten, dass sie die Wahrheit an den Tag bringt. Stattdessen erscheint
die Vencktheit, als ein Zustand, in dem die Wahrheit ferner als je ist.
Wenn die konventionelle Ordnung die Mglichkeit der Position, des
Standpunktes ist, so entzieht die Enthllung der Ordnung als Fiktion der
Position den Boden. Der Verlust der Position ist der Schwindel (vertige).

Paul de Man, Blindness and Insight. Essays in the Rhetoric ofContemporary Criticism. Second Edition (BI), London, Methuen, 1983, S. 251f.


Er erfasst denjenigen, dessen Standfestigkeit durch den Blick in die umgebende Leere bedroht ist. Dass Verrcktheit hier in Verbindung mit
Schwindel erscheint, ist ein Hinweis darauf, dass Verrcktheit nicht die
Verschiebung der Position bezglich einer nach konventioneller Nonn
als nonnal geltenden meint, sondern die Verunsicherung der Position
berhaupt und ihren Verlust in der Entrckung.Verrcktheit ist hier das,
was in Allegories of Reading als schlimmer als verrckt beurteilt wird.
Wenn Nonnalitt, Gesundheit, Vernnftigkeit als Fiktion und Maske
durchschaut werden, so enthllt doch die Entlarvung kein Gesicht hinter
der Maske, sondern nur dessen Fehlen. Die Einsicht in die Fiktivitt geltender Ordnungen fhrt nicht zur Findung einer Wahrheit, auf der sich
eine gltige Ordnung errichten liesse, sondern ist gleichzeitig die Einsicht in die Unmglichkeit einer fundierten Ordnung. An dieser Einsicht
des Ironikers ist nicht nur wichtig, worauf sie sich erstreckt, sondern dass
sie eine Einsicht ist. Denn die Entrckung als Verlust der Position, die in
ihr aufgeht, ist der Verlust der Mglichkeit von Einsicht. In dem Masse,
als eine Einsicht den Verlust der Position begleitet, ist die Positionslosigkeit nicht erreicht, und es bleibt in der Entruckung selbst eine Stellung
nehmende und daher ihr gegenberbleibende Instanz. Dieser unmgliche
Abstand kennzeichnet die Struktur der Ironie.
[... ] absolute irony is a consciousness of a non-consciousness, a reflection on
madness from the inside of madness itself. But this reflection is made possible
only by the double structure of ironic language: the ironist invents a form of
himself that is "mad" but that does not know its own rnadness; he then proceeds
to reflect on his rnadness thus objectified. (BI, S. 216)

Die Formulierungen im ersten Teil dieses Zitats stellen das Problem der
unmglichen Gleichzeitigkeit des Bewusstseins mit dessen Verlust. Der
Widerspruch, den der unvermittelte Aufeinanderprall des Unvereinbaren
herstellt, wirft die Frage auf, wie Ironie berhaupt mglich sei. Sie ist es
dadurch, dass Bewusstsein und Unbewusstsein auseinandergenommen
und einander gegenbergesetzt werden. Das Miteinander wird in ein
Nacheinander, Simultaneitt in Sukzession umgewandelt. Man kann das
in de Mans Sinn eine Allegorisierung der Ironie nennen. Wichtig daran
ist, dass die Positionslosigkeit (madness, non-consciousness) wieder zu
einer Position gemacht, "objektiviert" werden muss, damit es mglich
wird, daruber zu reden. Das Bewusstsein, das den Verlust der Position
begleitet, und die Rede, die ihn sagt, sind gezwungen, ihn wieder zu einer Position zu verfestigen, um ihn zu fassen. Das heisst, dass der Ironi-


ker, der die Fiktion durchschaut und sie als solche blossstellt, die Auflsung der Fiktion wieder als Fiktion aufbauen muss, und dass diese Fiktion ihrerseits der Ironisierung ausgesetzt werden muss, soll der Zustand
der Mystifikation durch die emstgenommene Fiktion nicht wieder hergestellt werden. Der so in Gang kommende Prozess ist unabschliessbar,
weil die Sprache nicht anders kann als Ordnungen zu errichten und sie
als gltig zu postulieren, obwohl das Wissen, dass sie es nicht sind, nie
aussetzt. Die Konventionen der Gesellschaft sind Fiktionen, aber die Positionslosigkeit, in die der Abbau der Fiktion fhrt, ist es nicht minder,
und sie ist der Ironie ebenso ausgesetzt. Ironie kann deshalb niemals zu
einer Position im Sinne einer vertretbaren Haltung werden. In dem
Masse, als sie es werden muss, um mglich zu werden, muss sie sich
sich selber ausliefern und wird notwendig zu der Ironie der Ironie, als die
Friedrich Schlegel sie dargestellt hat. "No degree of knowledge can ever
stop this madness, for it is the madness of words." Dem Ironiker, der
weniger mystifizien als andere Redende zu sein scheint, da er die Fiktivitt aller Ordnungen erkennt, bleibt es in keinem Moment erspart, dem,
was er entlarvt, .dadurch zu verfallen, dass er es entlarvt. Es bleibt ihm
nur die endlos sich wiederholende Auflsung der Fiktionen, die er nicht
glaubt, und auf die er dennoch angewiesen bleibt. Dieses Angewiesensein erinnert noch einmal an den Schwindel. Dieser befllt nicht den
Fallenden, der keinen Ort mehr hat, sondern den, der noch steht, ohne es
zu knnen. Schwindel grndet in der Position, die er gefhrdet.
Die Ironie ist der Ausdruck der Unmglichkeit der Fundierung von
Sinngefgen. Sie ist die unabschliessbare Abgrndung des Sprachlichen,
die sich selber als sprachliche Ordnung konstituieren muss, weil Rede
auch dann noch Gltigkeit postuliert, wenn sie sich als nicht fundiert
ausgibt. Ironie ist die Figur der Disfiguration. Sie durchschaut alle Ordnungen als Masken ohne Gesicht, als Prosopopoiien ihres Fehlens. Aber
die ironische Disfiguration fhrt nicht zu einem nichtfigurativen, eigentlichen Sinn, denn das der Figur Zugrundeliegende ist das radikal Entzogene, die entzogene Wurzel. Die Ironie verhindert das Ernstnehmen der
Figur, ohne dass sie ihm etwas anderes Ernstzunehmendes entgegenzusetzen htte als das Nichtemstnehmen der Figur. Weder kann man das
Figrliche noch ernstnehmen, noch ist ohne es auszukommen. Ironie. ist
Suspendierung zwischen der Diskreditierung der Figur und der Unmglichkeit, auf sie zu verzichten. Sie geschieht als das fortwhrende Verlassen der Fiktion, ohne dass sie bei einer Wirklichkeit ankommt.


Dass die sprachliche Ordnung sich nicht fundieren lsst und dennoch unvermeidlich ist, bedeutet, dass sie fortfhrt, eine ihr usserliche Gewhrleistung zu postulieren, ohne dass diese je sicherzustellen wre. Das
ist die Unentscheidbarkeit der Referentialitt der Texte, von der de Mans
berlegungen zur Veniicktheit ausgingen. Die Metaphorik der Verrcktheit steht in de Mans Texten fr die Entwu_rzelung der Sprache.
Alles, was innersprachlich geschieht, bleibt von ungewisser Referentialitt, und die Erkenntnis, dass es so ist, vermag nichts daran zu ndern.
Die Verrcktheit der Wrter ist die aus der Unmglichkeit der Fundierung von Sprache sich ergebende Unbestimmbarkeit ihres Status. Das
gilt auch fr das Wort Verrcktheit, das jetzt immer die Verfassung der
Wrter charakterisiert hat. Von der Veniicktheit der Wrter zu sprechen
ist eine Metapher, wenn Verrcktheit ein gestrter Geisteszustand ist, der
Wrtern nur zugeschrieben werden kann, wenn sie vermenschlicht werden. Aber die Verrcktheit der Wrter ist insofern eine besondere Metapher, als sie nicht nur die Wrter anthropomorphisiert, sondern auch eine
exakte Beschreibung der Metapher liefert. Nicht nur ist die Verrcktheit
der Wrter eine Metapher, sondern die Metapher ist die Verrcktheit der
Wrter, sofern man das Wort Verrcktheit aus seiner erstarrten Metaphorizitt befreit und ihm seine wrtliche Bedeutung "V~rschiebung" zurckgibt. Die metaphorische Verschiebung ist - in jedem Sinne - die
Veniicktheit der Wrter. Es ist nicht zu entscheiden, ob die Wendung
Verrcktheit der Wrter eine metaphorische Aussage ber die Wrter
oder eine wrtliche Aussage ber die Metapher ist. Diese Unentscheidbarkeit ist aber gerade die Verrcktheit oder - in dem Masse, als die
Veniicktheit sich wieder als Position zu fixieren droht - d~e Ironisierung
der Verrcktheit zu dem, was schlimmer als Veniicktheit ist. Die Verrcktheit der Wrter ist die unreduzierbare Rhetorizitt der Sprache.
An diesem Satz bleibt die Vermenschlichung der Sprache in der
Metapher von der Verrcktheit der Wrter auffllig. Darin scheint noch
einmal, verschoben, die Frage nach der Menschlichkeit der Sprache und
der Sprachlichkeit des Menschen auf. Verrcktheit ist die Verletzung
von Normen, die selber nur ernstgenommene Metaphern sind, die unbersetzbar bleiben und nicht fundiert werden knnen. Wenn die Norm
als nicht fundierbare Fiktion demaskiert wird, hrt Verrcktheit auf,
messbar zu sein, was sich auch so sagen lsst, dass sie alles, auch das

Mass, woran sie zu messen wre, erfasst. Damit zerfllt die Mglichkeit,

Nonnale und Verrckte auseinanderzuhalten, und damit letztlich die

Mglichkeit der Bestimmung des Menschlichen. Die Unmglichkeit,
ber die Sprache hinauszukommen, bedeutet, dass keine solche Bestimmung darber hinauskommt, die verrckte Fiktion von etwas zu sein,
dessen Fehlen sie maskiert. Das Menschliche ist eine sprachliche Fiktion, aber die Sprache ist nicht menschlich, sondern sie ist bezglich aller
mglichen Bestimmungen des Menschlichen verrckt. Sie ist es, weil sie
den fiktiven Ordnungen, die der Mensch in ihr errichtet, nicht unterstellt
ist, sondern sie erst ennglicht. Die Verrcktheit der Sprache ist die des
Menschen, der in ihr in dem Masse, als er zu sich zu kommen meint,
ausser sich ist. Ist der Mensch, der in der Sprache ausser sich ist, dadurch
bestimmt? Als das Wesen, das dadurch bestimmt ist, dass es sich entzogen ist? Eine solche Bestimmung wrde fr alles gelten, was gemeinhin
als nichtmenschlich gilt. Die darin sich ausdrckende Auslieferung an
die Sprache besteht darin, dass der Mensch sich durch sie bestimmt, die
ihm radikal unverstndlich bleibt, und sich deshalb verliert, indem er
sich versteht.
Die doppelte Auslegung des Ausdrucks Verrcktheit der Wrter hinterlsst eine Schwierigkeit, der man nicht ausweichen sollte. Die Mglichkeit, die Wendung einmal wrtlich, einmal metaphorisch zu nehmen,
ist an das deutsche Wort Verrcktheit gebunden, dessen Metaphorik
durchsichtig geblieben und jedem, der das Wort gebraucht, zugnglich
ist. Die Metaphorik des englischen madness ist undurchschaubar. Das
Wort bezeichnet einen Geisteszustand und kann in der Wendung the
madness of words nicht anders als metaphorisch verstanden werden. Die
vom deutschen Audruck ausgehenden berlegungen scheinen deshalb
nicht fr den englischen Text zu gelten, und es stellt sich die Frage, wie
sie auf ihn zu beziehen sind, und ob es nicht unzulssig ist, von einer
bersetzung her ber einen Text zu reden. Was ist der Status der doppelten Auslegung der Verrcktheit der Wrter?
Die Auslegung beruht auf einer bersetzung. bersetzung ist Verrckung. Was hier von der einen in die andere Sprache verrckt wird, ist
die Verrcktheit selbst. Die Verrcktheit der Wrter ist also nicht nur
eine metaphorische Aussage ber die Wrter, und auch nicht nur eine
wrtliche Aussage ber die Metapher, sondern sie kann auch auf die
bersetzung von madness durch Verrcktheit bezogen werden. Der
Ausdruck Verrcktheit der Wrter thematisiert, was mit dem englischen
Ausdruck the madness of words im bersetztwerden geschieht. Davon
ausgehend wre zu fragen, was Veniickung im Zusammenhang mit

bersetzung meint. Man kann madness auch anders bersetzen, zum

Beispiel mit Wahnsinn. Die Mglichkeit der doppelten Auslegung, die
das Wort Verrcktheit bietet, wrde dann entfallen. Dass man das eine
oder andere der beiden Wrter whlt, ist ein freier, das heisst beherrschter, Entscheid oder kann es mindestens sein. Dass aber das Wort Verrcktheit als mgliche bersetzung von madness berhaupt zur Verfgung steht, ist unkontrollierbar und eine nicht berechenbare Gunst der
deutschen Sprache, die, indem sie dieses Wort zur Verfgung stellt, an
der Auslegung des englischen Ausdrucks the madness of words mitarbeitet. Man kann diese Mitarbeit der Sprache zufllig nennen und sie
ablehnen, weil sie zufllig ist. Der Zufall ist das Abrupte, das auf nichts
zurckgefhrt werden kann und daher unverstndlich bleibt. Es scheint,
dass berlegungen, die auf dem zufllig zur Verfgung stehenden Wort
Verrcktheit aufbauen, nichts zum Verstndnis eines englischen Textes
beitragen, in welchem das Wort madness vorkommt. Wenn man aber
berlegungen, die vom deutschen Wort ausgehen, fr das englische
Wort nicht gelten lsst, so bedeutet das in letzter Konsequenz, dass Verrcktheit nicht als bersetzung von madness anerkannt werden kann.
Wenn beide Wrter dasselbe meinen, so muss die Metaphorik des deutschen Wortes auf das englische Wort bertragbar sein, auch wenn sie in
ihm nicht enthalten ist. Die gemeinte geistige Verfassung bleibt die gleiche, ob man nun Deutsch oder Englisch spricht, sie ist nicht an das Wort
gebunden und bleibt ber die Grenzen der Einzelsprachen hinweg unverndert. Wenn aber madness fr den Englischsprachigen etwas anderes ist
als die deutsche Verrcktheit, so ist die Metaphorik des deutschen Wortes unbertragbar, und was beide Wrter je meinen, bleibt an sie gebunden und diskreditiert das Wort der anderen Sprache als bersetzung.
Das unlsbare Problem der Ablsbarkeit des Gedankens von der
Sprache, dem kein Versuch zu verstehen, was beim bersetzen geschieht, entgeht, braucht hier nicht einmal gestellt zu werden. Es dient
lediglich der Situierung der doppelten Auslegung des Ausdrucks Verrcktheit der Wrter. Diese Auslegung ist offensichtlich an das deutsche
Wort Verrcktheit gebunden, und das englische Wort knnte dazu keinen Anlass geben. Dennoch beansprucht die berlegung zum deutschen
Ausdruck, zum Verstndnis des englischen Textes beizutragen, was nur
der Fall sein kann, wenn die berlegungen, die das Wort Verrcktheit
auslst, auch fr madness gelten und die beiden Wrter ein Selbes meinen, das. von ihnen ablsbar ist. Die Bindung der Sache an das Wort und
ihre Ablsbarkeit von ihm werden demnach gleichzeitig vertreten, ob100

schon sie einander ausschliessen. Damit ist der Entscheid ber das Verhltnis von Wort und Sache ausgesetzt. Dass dieser Entscheid nicht
mglich ist, versuchen de Mans Texte unermdlich vorzufhren, Madness und Verrcktheit knnen nicht gleichzeitig dasselbe und Verschiedenes sagen. Jede der beiden Versionen ist, gemessen an der andern, verrckt, aber es gibt kein Mass, woran sie beide gemessen werden knnten
und aufgrund dessen zwischen ihnen zu entscheiden wre. Die Anwendung der frher vorgelegten doppelten Auslegung des Ausdrucks Verrcktheit der Wrter auf den englischen Text lsst sich deshalb weder
legitimieren noch einfach abweisen. Die Beziehung zwischen madness
und Verrcktheit ist ein weiteres Beispiel fr jene Unentscheidbarkeit,
die schlimmer als Verrcktsein ist.


Heiner Weidmann

Korreferat zu Hans-Jost Frey, "Die Verrcktheit der Wrter"

"No degree of knowledge can ever stop this madness, for it is the madness of words"l. In diesem Satz aus "Shelley disfigured" klingt alles mit,
was man heute mit dem Namen Paul de Man verknpfen mag: die Vorstellung von bestimmten Einsichten - irgendwelcher "degrees of knowledge" - im Bereich des Lesens. (Und auch wenn diese darauf hinauslaufen, dass es weder einen eingrenzbaren Bereich des Lesens noch darin
irgendwelche bleibenden Einsichten geben kann, so kann man auf eine
solche Vorstellung doch schwerlich verzichten.) Danach besteht die
'Lehre' de Mans etwa in der Erkenntnis der grenzenlosen Figuralitt der
Sprache: Es gibt nur Metaphern, und das eigentlich und wrtlich zu
Nehmende ist nichts als eine durch Metaphern produzierte illusion; oder
in der Erkenntnis der endgltigen Nichtreferentialitt des Sprachlichen:
Von der Sprache aus gelangt man niemals zu einer Wirklichkeit, es sei
denn zu einer sprachlich hergestellten.
In Hans-Jost Freys Essay "Die Verrcktheit der Wrter" erscheint
der titelgebende Satz erst spt, um dann bis zum Schluss stndig prsent
zu bleiben. Vorerst wird ein Begriff von "Verrcktheit" bei de Man in
einzelnen Momenten exponiert, und dem gehe ich in meinem Korreferat
noch einmal nach. - Diese Verrcktheit tritt erstens auf als "Halluzination", als ein Zustand, der noch kaum Verrcktheit genannt zu werden
verdient, und erscheint zweitens als eine Verfassung, die schon
"schlimmer als Verrcktheit" heisst.
1. De Man lese die Texte, die er liest, auf ihre Unlesbarkeit hin. Die
Lesbarkeit eines Textes wre dadurch gegeben, dass man betreffend seines referentiellen Status sicher sein kann. Das kann natrlich nicht heissen, dass jedes Wort fr Wirkliches steht und wrtlich zu nehmen ist,
vielmehr knnte jedes auch eine rhetorische Figur sein, wie der Leser

Paul de Man, The Rhetoric of Romanticism, New York, Columbia University Press,
1984, s. 122.


weiss; nur msste er erkennen knnen, welcher Fall jeweils gerade vorliegt. Aber eben dafr fehlt das Kriterium, und diese Lesbarkeitsbedingung bleibt unerfllt. Damit knnte das Lesen jederzeit halluzinatorisch
sein. Der Halluzinierende nimmt nmlich das figural zu Verstehende literal, er fasst das, was nicht wirklich ist, wie etwas Wirkliches auf, und
zwar mit der Gewissheit, dass es wirklich sei. Aber diese Gewissheit hat
auch der vernnftig Wahrnehmende, und allein sie ist es, die ihm sagt,
dass er nicht halluziniert. In der Absicht, zu zeigen, dass das Lesen wegen der letztlichen Unbestimmbarkeit der Referentialitt halluzinatorischen Charakter hat, geht de Man - wie Frey einsichtig macht - besonders solchen rhetorischen Figuren nach, die in ihrer ganz korrekten und
regelrechten Anwendung das Verhltnis von Literalem und Figuralem
verkehren und untergraben: der Katachrese als einer gar nicht figrlich,
sondern direkt referentiell zu nehmenden, der Prosopopoiie als e~ner
berhaupt referenzlosen Figur.
2. Dieselbe Unbestimmbarkeit, die die milde Verrcktheit der niemals auszuschliessenden Halluzination bewirkt, kann auch der Grund fr
eine weit bedenklichere Verfassung sein. In dieser befindet sich zum
Beispiel jener Leser, der im Vorwort von Rousseaus Nouvelle Heloise
auftritt und darauf drngt, zu wissen, ob er es mit dem Autor der folgenden Briefe oder nur mit ihrem Herausgeber zu tun habe. Dringend muss
er es wissen nicht bloss aus Neugierde, sondern um den referentiellen
Modus des Textes bestimmen, um ihn lesen zu knnen. Beides knnte er
hinnehmen, dass die Briefe authentisch wren oder fiktiv, aber: "What
he could not tolerate, however, is the impossibility of distinguishing
between the alternatives. This would leave him in an intolerable semantic irresolution. lt would be worse than madness: the mere confusion of
fiction with reality, as in the case of Don Quijote, is mild and curable
compared to this radical dyslexia"2. Wenn also vernnftig derjenige ist,
der das Wirkliche bestimmen kann, der das Figrliche figrlich und das
Wrtliche wrtlich nimmt, dann gilt derjenige als verrckt, und zwar nur
harmlos und heilbar verrckt, der beides verwechselt, aber schlimmer als
Verrcktheit wird der Zustand dessen genannt, von dem gar nicht mehr
gesagt werden kann, ob er im einfachen Sinn verrckt sei oder nicht; das
ist "der Zustand der unentscheidbaren Referentialitt der Sprache"3.


id., Allegories of Reading. Figural Language in Rousseau, Nietzsche, Rilke, and

Proust, New Haven/London, Yale University Press, 1979, S. 202 (AR).
Hans-Jost Frey, "Die Verrcktheit der Wrter", oben S. 86.

Jene "Verrcktheit" im Satz "No degree of madness can ever stop

this madness, for it is the madness of words" scheint nun eher jener Zustand zu sein, welcher "worse than madness" heisst, als die milde und
kaum so zu nennende "madness" der Halluzination, der Verwechslung
von Fiktion mit Realitt. Doch das knnte tuschen. An der einfachen
Verrcktheit leide der Leser, der trotz der Unlesbarkeit liest, a~r seine
Situation sei noch beneidenswert verglichen mit der Situation dessen, der
die .Einsicht in die Unlesbarkeit gewonnen habe: jedoch nur lesend ist
diese Einsicht zu gewinnen, und also nur so zu gewinnen, dass man sie
auch schon verliert. Hans-Jost Frey hlt das deutlich fest: "Der Leser, der
in der Ungewissheit ber die Referentialitt des Textes suspendiert
bleibt, ist der Leser, der nicht referentiell liest, den Text nicht auf ein
Aussersprachliches hin zu verlassen sucht, sondern sich der Sprachlichkeit alles Gesagten bewusst bleibt. Aber es wre naiv, diesem Leser, dessen Zustand im brigen schlimmer als der des Verrckten ist, einen hheren Erkenntnisgrad zuschreiben zu wollen und seine Beziehung zum
Text als ein erstrebenswertes Ziel misszuverstehen"; denn "die Einsicht
in die unsichere Referentialitt des Sprachlichen ist sogleich ihrem eigenen Befund unterworfen." Wenn darum "der Zustand der unentscheidbaren Referentialitt der Sprache. unhaltbar" ist4, bedeutet das, dass jede
Erkenntnis von Verrcktheit selbst nur ein Beispiel dieser Verrcktheit
abgeben kann, dass der Zustand des 'Schlimmer-als-verrckt' jederzeit in
ein blosses, einfaches 'Verrckt' zurckgefallen sein muss. Die Unhaltbarkeit der berlegenheit eines Lesers ber einen Leser - und wenn es
die berlegenheit in der Verrcktheit sein sollte -- weist Frey an eben je'."
ner Stelle bei de Man nach, wo dieser in der Nouvelle Heloise die milde
Verrcktheit eines Don Quichote dargestellt findet: Einern (Roman-)
Leser kann es passieren, dass er verfhrt wird, etwas anderes fr wirklich
zu halten, als wirklich ist, - "et voila comment on devient fou". Diese
"madness", eine simple Vertauschung, "can be considered the madness
of another which does not threaten the sanity of the reader"5. Diese allzu
bequeme berlegenheit des Lesers scheint nun aber gerade an dieser
Stelle de Man selbst vorzuwerfen. Denn er liest hier, wie Frey zeigt,
Rousseau reduktiv in eben dem Sinn, dass er ber Referentialitt entscheidet und Lesbarkeit herstellt. Frey findet nmlich, wenn er bei Rousseau nachliest, weit kompliziertere Verhltnisse vor:

loc. cit.,id., S. 86-87.

Paul de Man, AR, S. 202.


Die stdtischen Menschen (gens du monde) werden die Romanfiguren, Landleute (campagnards) in einer einfachen Welt, nur verrckt finden. Sie erwarten von Romanen die Darstellung einer knstlichen, komplizierten Welt, wie es ihre eigene ist, um sie fr wirklich halten zu knnen. Das ist aber ihre Verrcktheit, - falls nmlich die dargestellte lndliche Welt wirklich ist. Nun scheint zwar Rousseau diese lndliche Romanwelt gegenber der stdtischen Leserwelt eindeutig zu privilegieren:
und dennoch falle die letzte Entscheidung darber, was von beidem lliusion und was Wirklichkeit sei, bei ihm nicht, weil eben die Frage, ob die
Briefe fiktiv oder authentisch seien, offen bleibe. Demnach liefert Rousseaus Text also keine Bestimmung der Verrcktheit, wie de Man ihn
liest, sondern stellt die Mglichkeit dieser Bestimmung in Frage und erreicht damit den Zustand, der schlimmer als Verrcktheit sein soll. - Die
Frage drngt sich auf: Warum simplifiziert hier de Man, - was man freilich nur bemerkt, wenn man de Man mit de Man wiederliest; und warum
unterschlgt er hier ausgerechnet das, was er doch schliesslich zu zeigen
unternimmt? "An dieser Infragestellung ist de Man hier nicht interessiert", oder: "Fr die Zwecke von de Mans Argumentation gengt" diese
Vereinfachung6, befriedigt als Antwort nicht. Ich versuche eine Antwort
zu geben, indem ich Frey mit Frey wiederlese.
De Man unterschlgt in seiner Rousseau-Lektre bezeichnenderweise eine Variante der milden Verrcktheit. Wenn diese nmlich in der
simplen Verwechslung von Figuralem und Literalem besteht, dann ist
derart verrckt der Leser, der Fiktives fr wirklich, aber auch der, der
Wirkliches fr fiktiv hlt und das wrtlich zu Nehmende figrlich auffasst. Dass von diesem nicht gesprochen wird, ist wohl mehr als eine
kleine unbedeutende Nachlssigkeit. Genau das knnte nmlich der Fall
von de Man selbst sein. Denn ebenso unvenneidlich, wie sich bei Rousseau die Tendenz durchsetzt, eine von zwei Mglichkeiten der Verrcktheit zu privilegieren, geschieht dies auch bei ihm. Immer wieder drngt
es sich auf, ihm die Einsicht in die unbegrenzte Figuralitt und Nichtreferentialitt der Sprache zuzuschreiben. Aber dass dies nicht die Einsicht
de Mans gewesen sein kann, dass sie als Einsicht auch ganz einfach
falsch wre, das hat Hans-Jost Frey in einem frheren Aufsatz ber de


Hans-Jost Frey, loc. cit., S. 85 und 84.

Man, "Undecidability", klar gemacht?. Die Unentscheidbarkeit der Referentialitt ist etwas gnzlich anderes als die Entscheidung fr Nichtreferentialitt. Und doch handelt es sich nicht nur um ein Missverstndnis
von unaufmerksamen Lesern de Mans, wenn ihnen immer wieder die Infragestellung der Sicherheit zur Behauptung der Unsicherheit gert.
Auch bei de Man schon passiert es wie von selbst, dass die in der Unsicherheit erreichte Positionslosigkeit umschlgt in eine - immerhin avancierte - Position. Man knnte sagen, dass die Unertrglichkeit des Zustandes, der schlimmer als Verrcktheit sein soll, nicht nur stets behauptet, sondern so bezeugt wird, dass er sofort in einen ertrglichen
umgewandelt erscheints. Zu diesem Zustand gehrt nmlich die Ungewissheit, dass man zwar vielleicht verrckt, vielleicht aber hchst vernnftig ist und nur insofern etwas absonderlich, als man sich einbildet,
verrckt zu sein.
Es scheint, als habe Hans-Jost Frey die Pointe von "Undecidability",
dass de Mans 'Lehre' nmlich unhaltbar wre, in diesem Text zurckgenommen; und man kann auch erraten, warum. Denn dass nicht nur jedes
Wort eine Metapher, sondern jede Metapher auch keine sein knnte, das
ist keine Einsicht fr die, die es schon lngst gewusst haben und auch
weiterhin davon ausgehen werden. Vielmehr vermag gerade der Leser,
welcher sich der unhintergehbaren Sprachlichkeit des Sprachlichen bewusst geworden ist, auch nicht einmal mehr zu entscheiden, ob das
Sprachliche nur-sprachlich sei oder nicht, und die Referenz entgeht ihm
so grndlich, dass er auch nicht ausschliessen kann, dass sie ihm vielleicht berhaupt nicht entgeht, - aber zum Bewusstsein bringt er sich
dies als die durchaus ertrgliche Einsicht in die unmgliche Referentialitt alles Sprachlichen.
"No degree of knowledge can ever stop this madness, for it is the
madness of words": das heisst nicht, dass diese Verrcktheit auch ohne
erkennendes Bewusstsein weitergeht - obwohl es allerdings nicht die
Verrcktheit in einem Bewusstsein, in dem des Lesers sein kann, wel7

id., "Undecidability", in The Lesson of Paul de Man, ed. Peter Brooks, Soshana
Felman. J. Hillis Miller, (Yale French Studies; 69) New Haven, Yale University
Press, 1985, S. 124-133.
Dabei ist aber nicht zu vergessen, dass es eine wirklich unertrgliche Verrcktheit
geben kann, die doch unbeschreiblich ertragen werden muss und angesichts derer
der Gebrauch des Begriffs "Verrcktheit" bei de Man nur fahrlssig oder metaphorisch wre, wenn nicht genau genommen die Verrcktheit eben ausgespart wrde:
die Rede ist nur von einem Weniger- und einem Mehr-als-Verrckt.


ches vielmehr durch das Lesen erst hergestellt wird -, sondern es heisst,
dass sie auch dann weitergeht, wenn das Bewusstsein ganz zum Einsatz
kommt, und mehr noch: dadurch weitergeht, dass es ganz zum Einsatz
kommt und sie aufzuhalten versucht. Unlesbarkeit wre ziemlich ertrglich, wenn man ganz einfach nicht lesen knnte. Aber zu erfahren ist sie
nur so, das man trotz allem und vielleicht auch ganz einfach und ohne
weiteres - liest.


Christiaan L. Hart Nibbrig



A propos de la deconstruction de Paul de Man.

Etrange, tout d'abord: comme nous parvient de fa~on silencieuse, insensible, plate, ce que Paul de Man nous donne a lire. A lire, comme si la
lecture deconstructive n'etait qu'accomplissement et execution de l'autodeconstruction du texte lu. La lecture n'est pas 'notre' lecture dans 1a
mesure ou eile se sert exclusivement d'elements de langage presentes par
le texte lui-meme [... ] La deconstruction n'est pas ce que nous aurions
ajoute au texte, mais c'est eile au contraire qui le constitue en tout premier lieul. Selon de Man, c'est le texte lui-meme qui fait valoir que
la deconstruction n'est pas quelque chose dont l'accomplissement ou non
dependrait de notre decision et de notre volonte. Elle est coextensive
avec l'utilisation dulangage en general [ ... ]2. On peut se poser la question - toute hermeneutique et depassee qu'elle puisse paraitre - de savoir
qui donc fait usage de quel langage, quand et de quelle maniere; et cette
question, apparemment reglee par l'impossibilite analytiquement arrangee d'y repondre, se pose de f~on inquietante plus que jamais comme
la question du lieu et du statut d'une teile undecidability sans cesse conjuree3. 'How can we know', ainsi pourrait-on transfonner le dernier vers,
eher a de Man, du poeme de Yeats Among School Children, 'how can
we know the reader from the text?' 4. Et cela justement a l'endroit d'une
1 Semiology and Rhetoric, in: Allegories of Reading, Figural Languagein Rousseau,Nietzsche, Rilke, andProust, Yale University Press, 1979, p. 17 (=AR).
2 Rhetoric of Persuasion, AR, p. 125.
3 cf. Hans-Jost Frey, Undecidability, in The Lesson of Paul de Man, Yale French
Studies, Number 69, 1985, p. 124ff. - La question reste ouverte de savoir si
l"indetermination' est un constat textuel ou une experience intellectuelle induite par
un travail conceptuel.

4 Sur la base de l'essai nietzscheen ber Wahrheit und Lge im aussermoralischen

Sinne, de Man deconstruit la categorie du 'soi' dont ne peut se passer ni
l'henneneutique, ni la theorie des performatifs et de la communication, en tant que
concept fonctionnel intratextuel: Le langage, en se pla~ant au centre, devalorisant


lecture cerebrale a l'extreme, qui pousse le lecteur par une debauche de

moyens a s'empetrer dans le texte a mesure qu'elle le met hors jeu, eva. cuant son point de vue de sa responsabilite, et lui arrachant le texte des
mains. Ainsi, pourrait-on penser, le texte prend-il la forme d'une machine
productrice de frustration pour la soif de signification, de meme que sa
lecture ascetique prend la forme du chef-d'oeuvre d'un artiste affame,
dont la minceur rejoint finalement ceile du papier, support de l'ecriture.
Celui qui lit ainsi, mefiant envers toute forme d'ancrage - que ce soit
dans le sujet, dans l'objet ou dans une signification mettant en relation
les deux terrnes, signification toujours engendree par un acte de representation anterieur a la relation et incontoumable -, de teile maniere qu'il
consente et souscrive a l'immanence radicale d'un jeu de representation
intenninable dont les signes se substituent a l'infini, celui-Ia, confiant
dans le manque total de fiabilite du langage dans sa constitution textuelle, ne cherche qu'a se faire surprende de maniere critique, lors de son
activite de lecture, en flagrant delit de mouvement de fuite; d'autant plus
qu'une teile deconstruction, veritable travail de for~at qui se presente
comme s'il n'existait rien de plus facile, assimilant le lecteur a Xerox ou
le soi, sauve celui-ci a l'instant ou elle lui attribue l'insignifiance et la vacuite d'une
simple figure de langage. Le soi ne peut subsister comme soi que sous la condition
de se deplacer dans le texte qui le nie. Le soi, qui fut d'abord le centre du langage
comme son referent empirique, devient maintenant le langage du centre fictif, metaphore du soi. [...]La deconstruction du soi comme une metaphore n'aboutit pas dans
la separation rigoureuse de categories (le soi et la figure ), mais dans un echange de
proprietes qui rend possible la subsistance des deux, bien entendu au prix de leur
verite litterale (Rhetoric of tropes in: AR, p. 111). De Man deplace encore davantage dans la structure du texte le 'moi' dit et le 'moi' lu au moyen du vingtieme
paragraphe de I'Eneyclopedie de Hegel: [ ...] quand je dis: 'je', je veux dire moi en
tant que celui-ci, a l'exclusion de tous les autres; mais ce que je dis, 'je', c'est en fait
chacun: In this sentence, the othemess of 'jeder' does not designate in any way a
specular subject, the mirror image of the I, but precisely that which cannot have a
thing in common with myself; it should be translated, in French, not as 'autrui', or
even as 'chacun', but as 'n'importe qui' or even 'n'importe quoi'. De Man conclut
d'apres l'exclusivite reciproque de 'vouloir dire' et 'dire', 'discemer' et 'poser', que le
je in its freedom from sensory determination, is originally similar to the sign
(Sign as Symbol in Hegel's 'Aesthetics', Critical lnquiry, 8, 1982, p. 769). Le lecteur du signe <<je devient lui-meme un signe, comme pour donner raison a Hlderlin: Ein Zeichen sind wir, deutungslos ... (Nous sommes des signes, depourvus de
signification... ) (Mnemosyne, seconde Version). - Cf. egalement Christiaan L. Hart
Nibbrig, Spiegelschrift, Spekulationen ber Malerei und Literatur, Francfort, 1987,
p. 31sq., p. 304. Manfred Frank, Die Unhintergehbarla!it von Individualitt, Francfort, 1986.


Laser, ne veut pas laisser discemer dans ses fruits, sous la fmme de metastatements synthetisants, detachables de chaque exemple concret: la
nature.morte, allegorie de leur caractere inconsommable. Voila pourquoi
il est discutable de sortir de leur contexte des passages comme le suivant:
A literary text is not a phenomenal event that can oo granted any form of positive existence, whether as a fact of nature or as an act of the mind. lt leads to no
transcendental perception, intuition, or knowledge but merely solicits an understanding that has to remain immanent because it poses the problem of its intelligibility in its own terms. This area of immanence is necessarily part of all critical discourse. Criticism is a metaphor for the act of reading, and this act is
itself inexhaustible5.

La lecture est insondable sous la condition de sa propre structure metaphorique, structurellement identique au texte, qui devient ainsi le toboggan de la lecture - metaphore au genitif, qui caracterise la relation de
maniere indiscemable, subjectivement et objectivement a la fois, comme
une traduction metaphorique, c'est-a-dire comme une meta-metaphore,
dans laquelle se dissout simultanement sa consistance. Impossible de
determiner lequel des deux ples se meut par rapport a l'autre. C'est un
galop intellectuel sur place. Une sorte d'antiderapant a l'intention d'une
lecture ala fois mue et mouvante peut ala rigueur ~tre foumi par cette
attention egalement flottante que Freud conseillait contre les dangers
du transfert et du contre-transfert, qui sont transposes ici du texte sur la
scene de sa lecture. 11 s'agit d'ecouter avec la troisieme oreille, comme
l'enseignait Lacan, en prenant garde aux coupures, dont la plus importante est celle entre le signifiant et le signifie. Virus intellectuel, engendrant une veritable manie: il faut encore et toujours s'en approcher et y
penetrer, si l'on veut parvenir a s'en degager.
Si l'un est pris pour l'autre, comme c'est le cas de maniere consequente dans le procede de Paul de Man, plus rien n'est donne. Cela conduit, a travers une perspicacite croissante pour la figuralite de ce qui parait etre a proprement parler, et qui est ainsi jete bas teile une presence
mena~ante de par sa fermete ontique, a l'assourdissement methodiquement mis en oeuvre envers toute audibilite phenomenale du discours
textuel, qui en reste sans voix. Une voix qui de toute fa~on ne retentit
que comme le resultat d'un transfert a partir de l'ecriture, c'est-a-dire
qu'elle ne retentit pas, mai~ devient lisible comme une simple metaphore
5 Tue Rhetoric of Blindness, in: Bliruiness and /nsight, Essays in the Rhetoric of
Conlemporary Criticism, University of Minnesota Press, 1983, p. 107 (=BI).


de subjectivite, laquelle est dechiffree de son cte non pas comme

l'origine de l'expression mais comme la metaphore de la perte de

l'origine, en tant qu'echo produit par sa vaine representation. La demonstration deconstructive de la figuralite omise, negligee, de ce qui est en
soi (des Eigentlichen), qui des le debut veut toujours dire autre chose que
ce qui est dit litteralement, et qui par consequent est deja soi-meme une
signification transposee, ne correspond pas ala production hermeneutique d'une signification remplissante, mais a une fonctionnalisation progressive et a une evacuation formalisante de tout contenu, ce qui en un
mot est nouvellement considere comme susceptible d'appartenir a la figuralite. 'Disfiguration' comme le resultat d'une mise en scene analytique
avec ses propres moyens de persuasion rhetorique, qui engendre une impossibilite de contrle au moyen d'un controle conceptuel extreme: une
surdetermination. Non pas que le texte se mette a retentir. En lieu et
place: un bruissement blanc6. verdetennining is a symptom of despair
(desespoir) as weil as of control7 L'inverse du travail de Penelope, qui
consiste a defaire de jour ce qui fut noue de nuit, prend sur le dos, avec
ses retablissements abstrayants, ce qu'il deblaie couche apres couche.
Cependant l'investigation systematique de ce que Paul de Man appelle
avec pertinence le caractere photographique de la representations
comporte un angle mort pour la theorie de la representation. Le discours
sous forme de metalangage de la pretendue 'lecteur' escamote
l'intermediaire, en pretendant expressement ne pas etre un meta-discours
du langage, ne pas etre une representation de la representation, cet intermediaire, dans lequel le mouvement de lecture inverse est sollicite avec
son accumulation respective d'actes de traduction, cet intennediaire, qui
nous montre de prime abord que la representation inverse mise en oeuvre
entre en conflit avec lui, et de quelle fa~on il devient audible que le langage se trompe9 (sich verspricht) et que le texte parle le langage de
6 Afin de preciser mes quesions, il serait utile d'effectuer une comparaison avec la
maniere clont Theodor W. Adorno lit des textes musicalement>>, en se basant sur
l'experience qu'une philosophie qui s'eieve ne sonne pas autrement que 1a grande
musique. - Cf. p. ex. l'essai de Hegel Skoteinos oder Wie zu lesen sei, qui est place
sous le signe de cette phrase de Borchardt: Ich habe nichts als Rauschen (Drei
Studien zu Hegel, Francfort, 1963).
7 Hypogram and Inscription, in: The Resistance to Theory, University of Minnesota
Press, 1986, p. 43 (=
8 Anthropomorphismus und Trope in der Lyrik, in: Allegorien des Lesens, Francfort, 1988, p. 198.
9 AR,p. 277.



l'auto-resistance10. A quoi il faut ajouter qu'avec une teile reflexivite

structurelle attribuee au langage lui-mfme, comme c'est le cas ailleurs,
on ne peut ecarter sans autre l'hypothese d'un retour a l'ontologie avec
l'affinnation du caractere grammatical de la consciencell, dans la mesure
ou les circonstances qui en sont la condition dans un champs de representation donne sont passees sous silence. The error is not within the
reader. Language itself dissociates the cognition from the act12. L'affirmation de 1a verite peut alors tre racontee comme l'histoire de son
egarement, non pas chez le lecteur, et non dans le seul texte, mais dans le
piege relationnel dans lequel s'est fourre le lecteur a travers la lecture;
c'est en sa qualite de piege relationnel (dans lequel la representation de la
duperie de la representation se trouve elle-meme prise au piege comme
une duperie) que l'impossibilite de resoudre l'aporie entre le caractere figuratif de la constatation et la figuralite perfonnative, entre la figuralite
en tant que connaissance et en tant que representation devient lisible
comme l'allegorie de son illisibilite (en tant que piege relationnel, la
lecture se rend elle-mme lisible comme piege relationnel). Index sui et
Jalsi. Cela est egalement valable pour le test d'auto-application d'une deconstruction deconstruite, laquelle, dans la mobilite vivace de son autocontradiction, fait tomber sur un os celui qui tente l'experience. Ce qui
est a demontrer le sera d'autant mieux que le resultat du test sera mauvais.
Paul de Man, qui 1it Rousseau, qui lit Derrida, croit tre en mesure de
demontrer que ce qui arrive chez Rousseau, est exactement ce qui arrive
chez Derrida:
a vocabulary of substance and presence is no longer used declaratively but rhetorically, for the very reasons that are being (metaphorically) stated. Rousseau's
text has no blind spots: it accounts at all moments for its rhetorical mode; Derrida misconstrues as blindness what is instead a transposition from the literal to
the figural level of discourse13.

10 RT, p. 28.
11 Cf. egalement Hegel on the Sublime, in: Displacement, Derrida and after, edited
with an Introduction by Mark Krupnick, Indiana University Press, 1983, p. 146: <<lf
we say that language speaks, that the grammatical subject of a proposition is language rather than a self, we are not fallaciously anthropomorphizing language but
rigorously grammatizing the self. Tue self is deprived of any locutionary power; to
all intents and purposes it may as weil be mute.
12 Prornises (social contract), AR, p. 277.
13 The Rhetoric of Blindness, BI, p. 138s.


Mais c'est lui-meme qui produit cet effet de par sa propre lecture, et il
rejoint les deux autres, bouclant ainsi la boucle, car il n'en va pas autrement pour lui que pour eux. Ainsi se realise ce que de Man reconnait
ailleurs de fa~n generale: la deconstruction etablit la conclusion erronee de la refrence d'une maniere necessairement referentielle. On ne
peut y echapper ... 1 4. Pas meme lorsqu'il s'agit de la representation nonrefrentielle de la musique comme dans l'Essai sur l'origine des langues
de Rousseau que de Man dirige contre la deconstruction effectuee par
Derrida de la conception rousseauiste de la representation. Certes, les
theories esthetiques de la representation qui ont cours au xvrne siede et
qui desesperent de quelque chose d'analogue a une presence originelle de
la meme fa~on qu'elles propagent la croyance en le retablissement de
cette presence perdue, sont soumises au paradigme d'une representation
picturale. Sauf que cet etat de chose doit etre court-circuite de maniere
circulaire avec l'autre, a savoir que la peinture - ce qui voile la poetique
de l'ut pictura poesis - se subordonne a la representation sous fonne de
langage aussi longtemps qu'elle peut compter sur le caractere traduisible
des histoires qu'elle raconte. Rousseau, cite par de Man contre Derrida
au moyen des passages des treizieme et seizieme chapitres, consacres a
la musique, de son essai, retoume la hierarchie classique des arts:
music is called superior to painting despite and even because of its lack of substance [...] Music becomes a mere structure because it is hollow at the core, because it 'means' the negation of all presence. lt follows that the musical strucUlre obeys an entirely different principle from that of strucrure resting on a 'fll'
sign, regardless of whether the sign refers to sensation or to a state of consciousness 15.

11 est frappant que ces metaphores de l'excavation et du remplissage, qui

signalent souvent chez de Man le passage a une lecture transposee et
transposante, reaffinnent, de meme que les guillemets, le schematisme
binaire entre l'inteme et l'exteme, le propre et le figure, qu'elles sont censees critiquert6. 11 n'y a pas que cet usage de plus en plus frequent de
14 Rhetoric of Persuasion, AR, p. 125.
15 BI, p. 127s.
16 L'espace inaugure par la lecrure transposee et transposante de de Man correspond a
cet etat de magma agite et agitant precooant toute liberte du langage et toute separation entre la matiere et l'esprit, l'objet et le sujet, que Platon decrit dans le Timee
sous le nom de chora. Avant de faire cela, il le met en scene en langage, comme
s'il etait l'eieve de Paul de Man: [ ... ] incapable de dire la verite a l'etat de veille, a
savoir - puisque celle qui l'a fait naitre ne lui appartient pas meme en propre, et


quotation-marks pour marquer la figuralite du. discours textuel qui soit

frappant chez de Man, et qui retire tout fondement Stahle a la figuralite
de son propre discours, que ce soit en tant que susceptible d'etre refrentialisee qu'en tant qu'apte a refrentialiser, mais on remarque aussi le fait
que quotation converge finalement avec signification chez le de Man
tardif, laquelle, coupee definitivement de la structure intentionnelle du
sens, n'est reconnaissable que dans les citations repetees d'un modele depourvu en lui-meme de senset de signification:
the principle of signification [...] is no longer a sign-producing function (which
is how Hegel valorized the sign in the Encyclopedia) but the quotation or repetition of a previously established semiosis; it is reduced to the preordained motion of its own position. (Hegel on the Sublime, p. 150)

Ainsi, la signification du mot 'signification' n'est plus praticable que

comme un 'mot', et eile se retrouve coupee d'elle-meme dans ce qu'elle
sigrune. De tous temps, ene nest aeJa quune cuauon. comme Ies signes
qui 'signifient' une signification empruntee - guillemets, mots soulignes,
italiques -, et que nous ne remarquons plus, parce qu'eux-memes sont
prives de signification, en dehors de leur fonction de signe indiquant une
citation. Nous les lisons - voila ou l'on veut en venir a travers l'effort
d'une lecture teile que la pratique Paul de Man - avec la plus grande facilite, sans meme nous en rendre compte, de maniere fnctionnelle.
Par ailleurs, on peut avant tout remarquer que dans ladite critique de
de Man, la structure de la representation musicale, dans sa description
theorique, prend la place de son audibilite, en mettant l'accent sur le
Statut semiotique et non-sensible du signe musical. De la musique se
met aretentir effectivement Ia ou eile ne retentit pas: dans les intervalles,
melodieuse a !'horizontale, et harmonieuse a la verticale, dans les interstices de ce systeme de relations dynamiques et differentielles qui est sa

qu'elle est toujours emportee comme une apparence appartenant a une autre - qu'il
incombe pour cette raison al'image de se former dans une autre, s'agrippant en quelque sorte a l'essence, ou eile n'est rien du tout; mais que vient en aide a l'etant
veritable le discours vrai de par sa precision, qu'aussi longtemps que quelque chose
est ceci comme une autre chose, mais que. cela l'est a nouveau comme une autre,
l'une des deux, devenue une dans l'autre a un moment ou a un autre, ne sera pas simultanement la meme chose et deux choses separees (Timte, 52). Cf. a ce sujet
Julia Kristeva et sa justification du langage poetique a l'intersection entre thetique
et chora>> in La revolution du langage poetique, Paris, 1974, p. 22sq.


nature propre17. De Man met en scene tout cela avec veive, revenant ala
charge de maniere repetee, prenant son elan avant de foncer a coups de
definitions (music is ... music becomes ... music can never ... etc.) qui
peuvent ~tre lues comme metaphore de sa conception du langage, bien
avant qu'il n'impute a Rousseau, dans l'esprit d'une transposition en arriere, le fait de ne parler metaphoriquement que de langage, Ia ou il parle
de musique. Ainsi de Man se voit-il rejoint et depasse par le caractere
traduisible de son propre discours:
Musical signs are unable to coincide: their dynamics are always oriented toward
the future of their repetition, never to the consonance of their simultaneity.

A cela correspond la fin de l'essai de Proust, ou l'on peut lire, concernant

le langage, que Ce que nous appelons le temps reside justement dans
l'incapacite de la verite de concorder avec elle-m~me1s. La musique, en
devenant langage, en extirpe la musicalite.
C'est le privilege de la musique par rapport ala litterature, de la voix
par rapport al'ecriture, du propre par rapport au figure, de !'immediat par
rapport au mediat, qui est enjeu19 dans la lecture de laNaissance de
la tragedie apartir de l'esprit de la musique de Nietzsche. La voix du
texte est mise en avant contre le texte qui repose sur l'autorite d'une
voix humaine (AR, p. 95). L'autorite de la voix du narrateur, comme
on peut le montrer, doit legitimer un acte dans lequel l'aporie d'une
representation immediate, elle-m~me une absurdite logique, doit ~tre
depassee (AR, p.96). De Man a tf fait de venir a bout de ce passage ou
Nietzsche prend en consideration, en regard de la partie finale de Tristan
et Iseut, l'effet mortel de la griserie provoquee par l'experience musicale
sans le concours du verbe et de l'image, et cette fois sans l'etalage habituel de minutie discursive, comme si l'affinnation defensive apollinienne que l'on devine ici sous le masque dionysiaque concordait avec
l'affinnation sous-jacente que l'experience n'est pas possible immediatement, sans l'intennediaire du langage, en finde compte n'est pas possible tout court sans payer un tribut mortel, et comme si le mythe d'une
17 Cf. Christiaan L. Hart Nibbrig, sthetik der letzten Dinge. Francfort, 1989, demier
18 S'il est vrai que la phrase est censee etre juste a travers l'auto-diffrentiation potentielle de la verire, alors le diapason a la forme d'une replique logique, lien entre les
enonces. qui est fausse si, et seulement si, le premier enonce est vrai et le second
19 AR, p. 90.


teile fatalite n'etait pas tout-a-fait infonde, mythe qui ne peut resister au
caractere ridicule d'une description litterale. Le plaisir lui-meme,
peut-on lire de maniere analogue chez Derrida, apropos d'un passage du
huitieme livre des Confessions de Rousseau, depourvu de symbole et
d'artifice, qui nous procurerait la pure presence eile-meme (et nous
mettrait en harmonie avec eile), pour autant qu'une teile chose soit possible, serait en fin de compte un synonyme de la mort20.
Chaque lecteur de la Naissance de la tragedie sait, selon de Man,
qui use implicitement d'une rhetorique d'exclusion, par quel subterfuge le
pouvoir destructif d'une verite immediate peut etre neutralise: au lieu
d'etre directement eprouvee, eile est representee (AR, p. 131). La representation fait office d'experience de substitution, d'ersatz, d'experience.
Tout astucieux que ce soit, c'est une strategie d'immunisation contenu
dans sa propre critique.
Afin d'extiiper par la critique le schema genetique comme le seul
permettant la representation dans la revalorisation logocentrique que
pratique Nietzsche - Dionysos, pere de tous les arts, qui donne finalement une voix a la sphere apollinienne, qui n'attendait que cela -, de
Man met en parallele la description de cette fin d'opera dans laqueile le
drame apollinien commence aparler avec une sagesse dionysiaque, et
ou eile se nie eile-meme ainsi que son apparence apollinienne, avec un
autre passage qui n'a pas ete retenu dans le texte definitif et qui est consacre a la Sainte Cecile de Raphael - une visualisation de la musique:
- aucun sonne s'echappe de ce monde apparemment abandonne a la
musique, qui s'eteint a l'instant ou eile commence a retentir aux oreilles
de l'observateur, oubliant que tout cela n'arrive que de maniere imagee.
Quelque chose d'analogue se produit a la fin de l'essai de de Man, a contre-poil de sa lecture, comme si le schema genetique s'imposait contre sa
deconstruction, en depit ou a cause d'une sensible sympathie avec la dimension apollinienne, qui en soi n'est ni vraie ni fausse, VU que son
horizon se recouvre avec la conscience de sa propre nature illusoir (AR,
p. 98). Avons-nous bien entendu: nature? La deconstruction du texte
de Nietzsche, poursuit-il, n'a pas lieu entre des affinnations, comme
dans une reponse logique, ou dans un mouvement dialectique, mais
s'accomplit entre les affirmations d'un metalangage portant sur la nature
rhetorique du langage d'une, part, qui met en doute ces affirmations. Le
20 Jacques Derrida. Grammatologie, traduit par Hans-Jrg Rheinberger et Hanns
Zischler, Francfort, 1974, p. 286.


resultat de ce jeu d'interactions n'est pas une simple negation (l.c.).

Mais il engendre de son cte un reste de signification, qui reste hors
de portee de la logique du texte et force le lecteur apenetrer dans un processus manifestement illimite de deconstruction, afin de le rendre analytiquement audible en tant que dissonance semantique (AR, p. 99).
Sauf qu'il resonne a proprement parler aussi peu que la musique dans le
tableau de Raphael, mais seulement de fa~on imagee, figuree. Ce qui reste de signification, poursuit-il, peut etre traduit en un enonce, bien
que l'autorite de cette seconde enonciation ne puisse etre comparable a
celle de la voix dans le texte, s'il est lu narvement. L'enonce secondaire,
non-autoritaire, qui resulte de la lecture, devait etre un enonce au sujet de
la limitation de l'autorite textuelle (AR, p. 98). Avec la perte de la
narvete, la lecture retire sa 'voix' au texte, en divisant celle-ci, teile une
sorte de diapason, en un axe constatatif et un axe performatif; eile devient une metaphore de l'inaudible pour la voix analytique - eile semble
parler a la fin au nom du style socratique, qui est passe sous silence chez
Nietzsche au profit de la filiation genetique entre la sphere dionysiaque
et la sphere apollinienne -, qui entend de plus en plus. La nature
rhetorique du langage est sa nature metaphorique - rien de moins que
sa nature. La metaphore n'est pas 'vraiment' l'essence qu'elle signifie,
mais eile peut etre comprise comme un renvoi a quelque chose ou la signification et l'essence cofncident. [... ] La metaphore ne signifie pas ce
qu'elle dit, mais en fin de compte eile dit tout de meme ce que signifie
son enonciation [... ](AR, p. 90s.). Elle possede donc une voix qu'on lui
conteste pourtant. To the extent that language is figure (or metaphor, or
prosopopeia), peut-on lire dans le prolongement de Wordsworth, it is
indeed not the thing itself, but the representation, the picture of the thing
and as such is silent, mute as pictures are mute21. Et nous tous, que
nous lisions ou que nous ecrivions, sommes sourds et muets, pas seulement silencieux, Ce qui englobe la manifestation possible et volontaire
du son, mais au contraire silencieux comme des images, eternally deprived of voice and condemned to muteness (ibid). La voix de la metaphore n'est que la metaphore d'une voix. C'est egalement valable pour
'Dionysos', auquel Nietzsche communique la voix, et qu'il fait a son tour
communiquer la voix. 11 doit, en tant que musique ou langage, ainsi de
Man tente-t-il de contrer la pretention totalisante du schema de repre21 Autobiography As De-Facement>>, in The Rhetoric of Romanticism, Columbia University Press, 1984, p. 80 (= RR).


sentation genetique dans sa structure topologique, soit appartenir a la

sphere teleologique du texte, mais il n'est alors qu'erreur et duperie, soit
appartenir a la 'nature', et etre alors separe radicalement et a jamais de
toute fonne d'art, etant donne qu'aucun pont ne pourra jamais relier le
domaine nature! des essences au domaine textuel des fonnes et des valeurs (AR, p. 100). La refrence negative ala nature, qui est une fois
de plus ornee significativement de guillemets, est elle-meme aussi impropre que ceile a proprement parler dont on affinne l'inaccessibilite et
l'inaptitude a etre representee. Le pont que cette metaphore est en vain
cense jeter, est lui-meme une metaphore. Qui ne retentit pas. Et qui
donne en lieu et place un visage a ce qui est sans voix, comme le fait le
'master trope' pour la 'prosopopeia'; ce visage est la figure de la voix,
eile-meme figure, comme si eile venait d'au-def de la mort qui trace
cette ligne de. demarcation entre le texte et la vie, comme la voix d'une
epitaphe22. Un changement de nom implicite se produit chez de Man, a
savoir que le signe mythologique 'Dionysos' devient 'Hennes', en tant
que vehicule de la trans-position. La naissance de La tragedie dramatise une multitude de genres, avec l'aide desquels la distinction entre
l'etre et l'apparence peut etre surmontee; ce que nous avons appete le
schema genetique est exactement la possibilite de ce depassement, representee dans un recit metaphorique, grce auquel Dionysos peut penetrer
dans un monde d'apparences et, d'une certaine fa~on, demeurer tout de
meme Dionysos (AR, p. 101). Un tel changement du nom paternel,
Dionysos, s'effectue au nom du fils, Apoilon, celui du fils se produit finalement au nom de l'esprit de la deconstruction. Bien entendu, la voix
n'est pas leguee, seulement le nom, qui recouvre, de fa~on figuree, au
nom de qui on peut continuer aparler en epelant a rebours, de maniere
deconstructive, le schema genetique, avec une autorisation etrangere.
Ainsi se deroule sous nos yeux ce que de Man met egalement en action
lors de sa lecture de Proust: une metonymisation de la metaphore, et le
sapement de son autorite. En se revelant etre une meta-metaphore, de
par son incapacite a remplir sa fonction de pont, eile entralne egalement
le 'schema genetique', qu'elle semblait servir, dans le tourbillon de sa

22 Cf. Hypogram and Inscription, ibid. et Cynthia Chase, Giving a face to a Name,
de Man's Figures, in: Decomposing Figures, Rhetorical Readings in the Romantic
Tradition, Tue Johns Hopkins University Press, 1986, p. 82sq.


mise-en-abyme, incapable de totalisation23. Ainsi, l'un ne remplace finalement pas l'autre dans le sens d'un rapport de substitution metaphorique, pretendu, illusoire, mais dans un rapport de simple voisinage, aveuglement et de maniere fortuite:
Car si les modeles genetiques ne sont qu'un exemple de mystification rhetorique
parmi d'autres, et si le rapport entre le sens figure et le sens propre d'une metaphore, comme c'est le cas dans ce texte, est interprete en termes genetiques,
alors la metaphore devient une metonymie aveugle, et tous les jugements de
valeur qui se dessinent de maniere si precise dans IaNaissance de la tragedie, une theorie melocentrique du langage, la conscience pan-tragique de soi et la
conception genetique de l'histoire -, ont l'air creux [are made to appear hollow ], lorsqu'ils sont exJXlses a la clarte d'une nouvelle lumiere ironique. (AR,
p. 102)

Cette note finale aveuglante indique le prix a payer pour l'acuite audi~ive
analytique: la perte de la vue. De Man produit la cecite qu'il critique.
Immunite irritante: le lui reprocher n'aurait pu que lui convenir. Significatifs, f aussi: les recipients metaphoriques. L'espace creux, vide aussi
bien que rempli par la lumiere de l'ironie, conditionne la possibilite de la
resonance et de la voix. A sa place, de Man offre la sienne ala lumiere.
De Man considere le poeme Am Rande der Nacht tire du Buch der
Bilder de Rilke comme un exemple pour la plus classique des metaphores, asavoir l'idee du transfert d'un espace interieur aun espace exterieur
[ou vice versa] avec l'aide d'une representation analogique24.
L'excavation du monde des objets - des corps de violons, avec leur
caisse de resonance, le signe integral de leurs oui'es - evacue plus que jamais le sujet parlant de tout gue, dans sa pretention a etre une corde vibrante, et agit a l'image de was in den Dingen irrt, et qui veut quitter
les tenebres pour la lumiere. Toujours est-il que cette errance se voit
attribuer une voix:

23 Au sujet du probleme de la mise-en-abyme de la meta-metaphoricitie chez Derrida,

cf. la tentative de clarification de Rcxlolphe Gache, The Tain of the Mirror, Derrida
and the Philosophy of Rejlection, Harvard University Press, 1986, p. 233sq.
24 Tropes (Rilke), AR, p. 35.


Ich soll
silbern erzittern: dann wird
Alles unter mir leben,
und was in den Dingen irrt
wird nach dem Lichte streben,
das von meinem tanzenden Tone,
um welchen der Himmel wellt,
durch schmale, schmachtende Spalten
indie alten
Abgnde ohne
Ende fllt ...

Il serait errone, selon de Man, d'interpreter cette lumiere comme la

clarte de la connaissance de soi .... La lumiere est la transformation d'un
etat de confusion et d'inconscience (reve, sommeil, errance) dans la version sonore de ce meme etat. La figure est la metaphore d'une sonorisation, non d'une prise de conscience (AR, p. 36). Une teile sonorisation
devient cependant visible, Sous nos yeux, comme on peut le lire a plusieurs reprises, sous nos yeux de lecteurs. Et pourtant pas a proprement
parler. L'instrument de musique ne represente pas la subjectivite d'une
conscience, mais un potentiel inherent au langage; c'est la metaphore
d'une metaphore. Son elaboration comme un processus acoustique
devient transposable point par point sur 1a description rendue visible de
l'objet (AR, p. 37), image trompeuse d'une description, dans laquelle la
structure de l'objet est celle d'un potentiel figure de langage. Donc invisible egalement: l'objet est au sens propre un instrument de musique.
L'embolt:ement total des figures fait resonner le langage comme un violon (ibid). Pour ainsi dire, metaphoriquement, dans urie inversion en
chiasme du son en une chose visible, bien.que les yeux du lecteur, au
sens figure, ne peuvent percevoir la figuralite tant inaudible qu'invisible
dans le poeme de Rilke qu'en considerant le visible comme la voix metaphorique de l'audible. De Man, conscient de cette duperie, s'eclipsant
comme une sirene, en queue de poisson, al'apogee de sa demystification
magique, se fait finalement rejoindre par les vers fran~ais de Rilke:
Masque? non. Tu es plus plein,
mensonge, tu as des yeux sonores.

Les images sonores, lues de cette maniere, demasquent toute expression

comme l'effet d'une representation et montrent que le sujet, dans la confusion du langage et de la musique, n'est pas ou l'on croit entendre sa


voix. Sans origine, sans gtte, coupes de leur source qui n'est que leur
echo. Bien entendu, c'est ainsi que parlent des robots. The principle of
intelligibility, in lyric poetry dcpends on the phenomenalization of the
poetic voice25. Une autre fois, se contredisant de maniere comprehensible: The lyric depends entirely for its existence on the denial of phenomenality as the surest means to recover what it denies26. 11 s'agit Ia de
debusquer l' ideologie qui dissimule exactement dans la Substitution
entre la realite du langage et la realite naturelle, de la reference phenomenale27. La solution que propose de Man pour ne pas se laisser pieger
par la trompeuse sensualite de la representation consiste a la demasquer
comme une figure, et cela signifie aussi, paradoxalement, qu'il faut toujours la referencer.
Un exemple eloquent pour la fa~on dont de Man coupe le son en lisant est sa critique devote au point d'en apparattre cynique de ce qu~ dit
Jauss apropos de la fin du poeme de Baudelaire Spleen II:
Desonnais tu n'es plus, matiere vivante!
Qu'un granit entoure d'une vague epouvante,
Assoupi dans le fond d'un Sahara brumeux;
Un vieux sphinx ignore du monde insoucieux,
Oublie sur la carte, et dont l'humeur f arouche
Ne chante qu'aux rayons du soleil qui se couche.

Ici, dit Jauss, l'autre monde irreel de la Beaute s'avance en plein jour,
transfigure et transfigurant, et transforme finalement le paysage melancolique du Spleen en beaute bizarre et detaches d'un son pur qui ne retentit pour personne; un poeme donc anticipant Mallarme, qui fait disparattre dans sa representation l'objet lui-meme, afin de trouver sa veritable matiere dans l'activite poetique donnant naissance au poeme2s.
Une teile reabsorption du represente par la representation sous le signe
de la pure resonance serait la realisation poetique du paradoxe de Gdel.
Pour de Man, se referant au concept hegelien du signe comme un embleme de l'oubli29, le sphinx chantant n'est que le sujet grammatical - en

Lyrical Voice in Contemporary Theory, in: Lyric Poetry: Beyond New Criticism,
ed. P. Parker/C. Hosek, Cornell University Press, 1985, p. 55.
Anthropomorphism and Trope, AR, p. 259.
RT, p. 10.
Hans Robert Jauss, sthetische Erfahrung und literarische HermefU!utik, Munich,
1977. p. 332s.
Cf. Hypogram and /nscription, in: RT, p. 43: L'ecriture est l'elimination determinee
de la determination.


verite, il n'y a que son humeur farouche qui chante -, Cut off from its
consciousness: what he 'sings' can never be the poem entitled 'Spleen';
his song is not the sublimation but the forgetting, by inscription, of terror, the dismembennent of the aesthetic whole into the unpredictable
play of the literary letter3o. La signification phenomenale d'un chant comme la Certitude sensible de Hegel - se voit ainsi evacuee par la
raison, videe de sa substance, materialisee, au profit de la signification
litterale. Le resultat, ce sont ces choses inoures que de Man mentionne
ailleurs en rapport avec la theorie des anagrammes de Saussure31 .
Tue 'choses inou'ies' would precisely be that the phonic, sensory and phenomenal ground of poetic diction has been unsettled, for the laws for the dispersal of
the key word in the text, be it as ana-, para- or hypogram, are not phenomenally
nor even mathematically perceivable. Since the key word is the proper name in
all its originary integrity its subdivision into discrete parts and groups resembles, on the level of meaning, the worst phantasms of dismemberment to be
found in D.P. Schreber's Denkwrdigkeiten eines Nervenkranken. We would

then have witnessed, in effect, the undoing of the phenomenality of language

which always entails (since the phenomenal and the noumenal are binary poles
within the same system) the undoing of cognition and its replacement by. the
uncontrollable power of the letter as inscription32.

Matiere vivante - comme chez Baudelaire a la lumiere de la lecture de

de Man. Et pourtant, le sphinx, tombeau du souvenir, le poeme dans son
ensemble, espace creux de la memoire, seraient tout-a-fait en mesure,
comme chez Rilke, de garder une voix a l'interieur, laquelle, une fois
toute Contenance eliminee, emet alors un bruissement comme d'un COquillage vide. Derrida doit penser de maniere analogue quand il ecrit a la
fin des Memoires ... :
Il devait penser que les oreilles fines savaient l'entendre, et qu'il n'avait meme
pas besoin de faire des confidences sur la guerre ace sujet. En fait, il ne parlait
que de cela. n n'ecrivait que de cela. n m'arrive de me dire: il supposait peutetre que je le savais ne serait-ce qu'a le lire, tout ce clont il ne m'avait jamais
parle. Et peut-etre, en effet, le savais-je obscurement. Je l'entendais sourdement.
'Comme le bruit de la mer .. .'33.

30 Reading and History, RT, p. 70.

31 Jean Starobinski, Les mots sous les mots: Les anagrammes de Ferdinand de Saussure, Paris, 1971.
32 Hypogram and /nscription, RT, p. 37.
33 Jacques Derrida, Memoires, pour Paul de Man, Paris, 1988, p. 229.


Ce n'est pas sans raison que Paul de Man lui-meme a choisi d'introduire

son essai Semiologie et rhetorique avec cette devise empruntee aPascal,

qui doit inviter nos yeux a ecouter en silence 1orsque nous lisons:
Quand on 1it trop vite ou trop doucement on n'entend rien. Tant que de
Man, contre Jauss, dit ce qui n'est certainement pas chante dans la lumiere declinante du soleil couchant, c'est-a-dire le poeme comme sujet et
contenu du chant, en tant que sonorisation de ce qui est lu, il donne toujours dans le phenomenalisme phonocentrique qu'il combat (What he
'sings' ... , et non his 'singing'). Cependant, des qu'il rend lisible allegoriquement le chant (das Singen) comme le fonctionnement du langage poetique, celui-ci devient en meme temps, s'il est lu de cette maniere, l'allegorie d'une telle lecture. Elle expulse ce qu'elle refoule, et une
sonorite devenant ecriture nous force en meme temps a decouvrir la similitude entre textes et partitions musicales. Celui qui lit de fa~on a demasquer ce qu'il 1it dans son fonctionnement comme porteur symbolique
de son et de sens, et par consequent, regardant au travers de la notation,
devient capable d'eteindre, celui-la commence a entendre, Ia ou la lecture
par ses propres moyens fait faillite, ce qui flamboie en s'eteignant, a entendre, non pas ce qui est ecrit - car ce qu'il lit n'est deja plus la -, mais
comment il lit, comment c'est ecrit. Et comment, mais comment?
Traduit de l'allemand par Marc S. Abbhl


Das Folgende gilt, versuchsweise, einer Selbstapplikationsprobe de Manscher Dekonstruktion und der Errterung des expliziten und impliziten Darstellungsbegriffs, der seine
theoretische Praxis leitet. Indern den Texten die Musikalitt und Phnornenalitt, zugunsten ihrer Figuralitt, letztlich ihrer nackten Materialitt ausgetrieben wird, ergibt sich,
gegen die Absicht solcher 'Lektre', da sie gelesen werden wie musikalische Partituren.


Norbert Gabriel


ber das Verhltnis de Mans zu Hlderlin ist viel, ber das zu Heidegger
ist noch mehr gesagt worden. Weniger jedoch hat man bislang ber die
Bedeutung der besonderen Konstellation "Hlderlin-Heidegger" fr de
Man gesprochen, darber, wie sie seine Auffassung von der Dichtung,
dem Dichter geprgt h~t. Weder Einzelheiten der Heidegger-Exegese
noch Details der Hlderlin-Philologie werden _hier also im Vordergrund
stehen, sondern die Frage, wie de Man mit Hlderlin die Kritik an Heidegger entwickelt, wie er mit Heidegger seine Deutung Hlderlins in
dang gebracht und wie er schlielich in der Verbindung von Hlderlin
und Heidegger seine Theorie des Poetischen entfaltet hat. Besonders
deutlich lt sich dies an zwei Aufstzen de Mans verfolgen: seinem ersten Hlderlin-Essay, der 1955 in der franzsischen Zeitschrift Critique
erschien und seiner letzten Hlderlin-Arbeit, einem Artikel, der 1970 in
der New York Review of Books publiziert wurde. Doch auch in den anderen drei dazwischenliegenden Hlderlin-Aufstzen ("Keats and Hlderlin", 1956; "L'image de Rousseau dans 1a poesie de Hlderlin", 1965;
"Wordsworth und Hlderlin", 1966) wird die Auseinandersetzung mit
Heidegger, wird die grundstzliche Orientierung an ihm deutlich. Doch
dort sind de Mans Hlderlin-Interpretationen in einen weiteren literaturgeschichtlichen Rahmen gestellt. So mu hier aus Raumgrnden auf eine
Behandlung dieser Aufstze verzichtet werden. Es kann aber auf diese
Aufstze auch verzichtet werden, weil sie fr unseren Zusammenhang
grundstzlich nichts Neues bringen und hier wesentlichen Positionen in
de Mans erstem und letztem Hlderlin'.Essay prgnanter formuliert
1 De Mans Hlderlin-Essays in chronologischer Folge: a) "Les exegeses de Hlderlin
par Martin Heidegger'', Critique, XIl (1955), S. 800-819. Ins Englische bersetzt
von Wlad Godzich unter dem Titel "Heidegger's Exegeses of Hlderlin'', in: P.d.M.:
Bliruiness arui /nsight. Essays in the Rhetoric of Contemporary Critiscism, Minneapolis, University of Minnesota Press, 1983, S. 246-266. - b) "Keats and Hlderlin",


Die erste Abhandlung ber "Heideggers Erluterungen zu Hlderlin"

ist als Rezension zweier Hlderlin-Bcher erschienen, die in ihrer Anlage, der Fragestellung und Gedankenfhrung entscheidend von Heideggers Philosophie und seinen Hlderlin-Erluterungen bestimmt sind. Das
eine Buch stammt von Else Buddeberg, das andere von Beda Allemann2.
Doch de Mans Essay ist kaum eine Rezension zu nennen. Er ist vielmehr
eine breit angelegte Auseinandersetzung mit Heideggers Erluterungen.
De Man reflektiert hier vor allem sein eigenes Verhltnis zu Heidegger.
Man kann seinen Aufsatz geradezu als Versuch einer Loslsung aus dem
Bannkreis der Heideggerschen Philosophie lesen. Zweifellos ist de Mans
Aufsatz, wie man nicht mde wurde festzustellen, eine deutliche Kritik
an Heidegger3. Doch ist dies nur dessen eine Seite. Denn andererseits
bedeutet die Kritik an Heidegger keinen radikalen Bruch mit ihm, ja
nicht einmal, wie sich zeigen wird, eine grundstzliche Ablehnung seiner
Erluterungen - eher das Gegenteil. Es ist der Versuch, mit Heidegger
gegen Heidegger zu denken, in der Kritik im Besonderen, dennoch
grundstzliche Positionen zu bernehmen.
Comparative Literature, VIII (1956), S. 28-45. - c) "L'image de Rousseaus dans la
poesie de Hlderlin", Deutsche Beitrge zur geistigen berlieferung, V (1965), S.
157-183. Ins Deutsche bersetzt von Renate Bschenstein unter dem Titel
"Hlderlins Rousseaubild", Hlderlin-Jahrbuch, XV (1957-68), S. 180-208. Ins
Englische bersetzt von A. W arminski unter dem Titel "The Image of Rousseau in
Hlderlin's Poetry", in: P.d.M., The Rhetoric of Romanticism, New York, Columbia
University Press, 1984, S. 19-45. - d) "Wordsworth und Hlderlin", Schweizer
Monatshefte, XLV (1966), S. 1141-1155. Ins Englische bersetzt von Timothy Bahti
unter dem Titel "Wordsworth and Hlderlin", in: P.d.M., The Rhetoric of Romanticism, cit., S. 47-65. - e) ''The Riddle of Hlderlin", The New York Review of
Books, XV (19. November 1970), S. 47-52. - Die beiden hier behandelten Aufstze
werden nach dem Erstdruck zitiert. Die Seitenangabe erscheint fortan im Text.
2 Else Buddeberg, Heidegger und die Dichtung. Hlderlin. Rilke, Stuttgart, 1953. De
Man bezieht sich in seiner Quellenangabe auf den ein Jahr zuvor in der Deutschen
Vierteljahresschrift, XXVI, erschienen Aufsatz E. Buddebergs "Heidegger und die
Dichtung: Hlderlin", den er flschlicherweise unter dem des Buchtitels anfhrt. Beda Allemann, Hlderlin und Heidegger, Zrich-Freiburg, 1954. Allemanns Name
wird von de Man konsequent "Alleman" geschrieben, dies wird auch in dem ins
Englische bersetzten Neudruck des Aufsatzes von 1983 nicht verbessert.
3 So Andrzej Warminski, "Heidegger Reading Hlderlin", in A.W., Readings in Interpretation. Hlderlin, Hegel, Heidegger (Theory and History of Literature 26),
Minneapolis, University of Minnesota Press, 1987, S. 45-71; hier S. 69 mit Anm. 37,
S. 208; vgl. a. Anselm Haverkamp, "Verschwiegener Lorbeer: 'Im Hofe aber wchset ein Feigenbaum' (Hlderlin, 'Andenken')", Poetica, XX (1988), S. 218-233; hier


Zunchst fragt de Man kritisch nach den Grnden fr Heideggers

Hinwendung zu Hlderlin. Zeithistorische und ideologische Grnde erklrt de Man fr eher peripher. Fr wichtiger hlt er die Tatsache, da
Heidegger im Zuge seiner Metaphysikkritik auf Hlderlin zurckgreife.
Heideggers Pldoyer, die abendlndische Geschichte der "Seinsvergessenheit" zu berwinden und sein Appell, zum "Sein" zurckzufinden,
habe einen Zeugen bentigt, der die Mglichkeiten eines solchen
Sprunges aus der Seinsvergessenheit schon bewiesen habe. Bei Hlderlin, so beschreibt de Man Heideggers Standpunkt, sei das "Sein" Wort
geworden. Da es nun aber gerade Hlderlin ist, auf den Heidegger zurckgreift, erklrt de Man mit jenem inzwischen so oft zitierten, berhmten Satz: "c'est que Hlderlin dit exactement le contraire de ce que
lui fait dire Heidegger" (S. 809). Dieser Satz kann, aus philologischer
Perspektive, die Fehlinterpretation Hlderlins durch Heidegger erklren.
Aber das ist nur die eine Seite. De Man schreibt nmlich weiter: "A ce
niveau de pensee, il est difficile de ctistinguer entre une proposition et ce
qui constitue son contraire. Dire le contraire, c'est encore parler de la
meme chose quoiqu'en sens oppose, et c'est deja beaucoup lorsque, dans
un dialogue de cet ordre, les deux interlocuteurs parviennent a parler de
la meme chose. On peut dire, que, en effet, Heidegger et Hlderlin parlent de la meme chose" (S. 809). Dies relativiert nun den vorausgegangenen Satz in seiner scheinbar negativen. Eindeutigkeit. Zwei Punkte
sind hier von Bedeutung:
1. Heidegger, der sich dem kritischen Dialog mit Hlderlin versagt
habe, wird nun, im Minimalkonsens des gemeinsamen Gegenstandes,
doch zum "Gesprchspartner" Hlderlins. Eine wichtige Voraussetzung
dafr ist, da de Man Heideggers eigene Charakterisierung seiner Erluterungen akzeptiert. Er sagte von ihnen, da sie nicht mit den ~te
rien philologischer Interpretation beurteilt werden knnten. Zwischen
Hlderlin und Heidegger herrscht de Man ein besonderes
Verhltnis. Heidegger vermge, auch wenn sein Vorgehen zuweilen
philologisch desastrs sei, manches genauer zu erkennen als die philologischen Interpreten Hlderlins. Heidegger wird so von de Man fr die
Heidegger-Exegese gleichsam "gerettet".
2. Wenn de Man Hlderlin und Heidegger als "Gesprchspartner"
zusammenfhrt, so hlt er grundstzlich an einer zentralen Position Heideggers fest. Mit Heidegger nimmt er an, da es Hlderlin in seiner
Dichtung um das "Wesen des Poetischen", es ihm um das "Sein" gehe.
Nur aufgrund dieser Prmisse kann de Man das Verhltnis Hlderlin127

Heidegger als einen polaren und komplementren Gegensatz bestimmen.

Darin zeigen sich die zwei Mglichkeiten, das "Wesen des Poetischen"
zu definieren, dichterisch auf das "Sein" zu reagieren. Am Beispiel von
Hlderlins "Feiertags"-Hymne erlutert de Man dann die unterschiedlichen, gegenstzlichen Positionen Hlderlins und Heideggers.
Das Ergebnis des Vergleichs lt sich so zusammenfassen: Heidegger sage, da das Sein, das gleichgesetzt werde mit dem Heiligen, dem
Unaussprechlichen, im Gedicht Wort werde. Der berhmten Zeile aus
der Hymne "Das Heilige sei mein Wort" werde von Heidegger die konjunktivische Form genommen. Er lese es als Indikativ: "Das Heilige ist
mein Wort". Bei Hlderlin dagegen werde das Heilige, das Unaussprechliche, das Sein - und festzuhalten ist, da de Man grundstzlich
Heideggers Gleichsetzungen dieser Bestimmungsmomente nachvollzieht
- nie unmittelbar, sondern immer nur vermittelt Wort. Seine Dichtung
sei als Mglichkeit zu begreifen, sich dem Sein zu nhern, aber allerdings immer nur in einem offenen und nie abzuschlieenden Proze.
Hierbei sind zwei Punkte zu beachten: 1. De Man zeigt, wie bei Hlderlin die poetische Sprache, im Blick auf das Sein, das Heilige, das
Unmittelbare von einer Situation des Mangels gekennzeichnet ist. In
temporaler Perspektive ist die poetische Sprache gegenber dem Sein einerseits nachtrglich, zum anderen ist sie vorgngig. Einerseits ist das
Sein in seiner Unmittelbarkeit immer vor dem vermittelnden Wort und
von ihm nie zu erreichen, anderseits ist die poetische Sprache in derErwartung, in der Haltung der Sehnsucht auf das Sein hingespannt. Aus
diesem Bewutsein nimmt das Gedicht den Ton des Gebets und der Bitte
an. Das Sein selbst aber kann niemals im Gedicht Wort werden. - 2. Die
poetische Sprache aber hat damit nicht den Bezug zum Sein verloren, es
steht nicht nur in negativer Beziehung zu ihm. Sie reflektiert vielmehr
das "Sein" in der von de Man beschriebenen eigentmlichen Doppelstruktur. Die Unmittelbarkeit des Seins ist fr Heidegger nicht nur die
Ermglichung des Akts der Vermittlung, sondern auch der Akt selbst, sie
ist gleichzusetzen mit der poetischen Sprache als Vermittlung. Fr de
Man aber ist die Vermittlung nicht identisch mit einem der beiden Elemente - dem Unmittelbaren oder dem Mittelbaren - in der, wie er sagt,
"Gleichzeitigkeit ihrer gegenseitigen Ausschlielichkeit". Als eine dritte
Entitt enthalte sie vielmehr beide Elemente. Die poetische Sprache reflektiere in diesem Sinne die Doppelstruktur des Seins. Es erscheine stets
in zwei Aspekten, die in Hlderlins Dichtung in verschiedene Begriffs-

paare gefat wrden: Natur und Kunst, das Chaotische und das Gesetz128

hafte, das Gttliche und das Menschliche, Himmel und Erde. Die poetische Sprache vennge zwischen diesen beiden Aspekten zu vennitteln,
sie in einer Balance zu halten. Ihr gelinge es jedoch nie, sie zu vereinen,
in dem Sinne, da sie zusammenfielen. "Mais [le poete] n'est jamais revenu sur son savoir de sa structure necessairement antithetique - " (S.
Im Fortgang der Argumentation verblat aber dieses spannungsvolle
Moment der Vennittlung von Unmittelbarem und Mittelbarem in der Beschreibung von Hlderlins Position. Nun wird der negative Aspekt im
Bezug der poetischen Sprache zum "Sein" strker betont, die Tatsache,
da sie das "Sein" nie wird erreichen knnen, sie sich in steter Vermittlung wiederholen mu. Im Gegenzug zu Heidegger verlagert sich der
Akzent zudem vom Akt der Vennittlung auf das "Ergebnis" der Vermittlung, auf das, was sie leistet. Bei Heidegger ist das die Unmittelbarkeit, bei Hlderlin, fr de Man, das Vennittelte, die stete Mittelbarkeit.
Wo bei Heidegger das Gedicht "gelungen" ist, es sein Ziel - das Heilige
Wort werden zu lassen - erreicht hat, steht bei Hlderlin, fr de Man,
das notwendige Scheitern, das Bewutsein der Unmglichkeit, je das
Heilige unmittelbar zu erfassen4.
Von Hlderlin, wie de Man ihn versteht, aus betrachtet, erscheint die
Beschreibung von Heideggers Position als eine grundstzlich kritische.
Doch ist mit dieser diametralen Gegenberstellung de Mans Aufsatz
nicht beendet. In der Rckkehr zum Ausgangspunkt seiner berlegungen, wo Heidegger und Hlderlin als "Gesprchspartner" bezeichnet
wurden, werden sie hier wieder zusammengefhrt. De Man geht es um
die Bestimmung des "Wesens des Poetischen", um eine "gltige Poetik".
Hlderlin und Heidegger reprsentieren fr ihn die zwei Mglichkeiten,
das "Poetische" zu bestimmen und zwar in der Weise, da nun deren
beide Positionen in einem Proze der Vennittlung verbunden werden.
Die Position Heideggers ist die der Unmittelbarkeit, die Hlderlins die
des Vennittelten, der Mittelbarkeit. De Man greift also hier auf jene
scheinbar unvernderbare Gegenberstellung von Hlderlin und Heidegger zurck, die er, wie wir sahen, durch eine einseitige Akzentuierung
von Hlderlins eigentlicher Position - der Vennittlung von Unmittelbarem und Mittelbarem - erreichte. Dies ennglicht es ihm, nun gerade
sie zum Modell machen zu knnen fr seine eigene Bestimmung des
4 Vgl. hierzu auch Hans-Jost Frey, Studien ber das Reden der Dichter, Mnchen,
1986, S. 135ff.


Poetischen, die sich darstellt in der zu vermittelnden Spannung von Heidegger und Hlderlin, von Unmittelbarkeit und Mittelbarkeit des Poetischen - immer im Blick auf dessen Bezug zum Sein.
In der Auseinandersetzung, der Kritik, aber auch der grundstzlichen
Orientierung an Heidegger entwickelt also de Man zunchst seine Auffassung von Hlderlins poetischer Sprache als Vennittlung von Unmittelbarem und Mittelbarem. Die temporale N achtrglichkeit der poetischen Sprache gegenber dem Sein wird akzentuiert in der polaren Konfrontation von Heidegger und Hlderlin, in der Auffassung vom Poetischen als Spannung von Unmittelbarkeit und Mittelbarkeit. Diese Gegenstze fhrt de Man dann in seiner Theorie des Poetischen nach dem
von ihm entwickelten Modell von Hlderlins poetischer Sprache in der
Form der Vermittlung zusammen, in der dann auch Heideggers Position
ausdtiicklich nicht nur als mgliche, sondern auch als notwendige Form,
das "Wesen des Poetischen" zu bestimmen, erscheint.
De Man folgt in seiner Auseinandersetzung mit Heidegger den von
ihm gebahnten Wegen. Die Perspektive auf Hlderlins Gedicht ist dabei
sehr eingeengt. Der Blick gilt allein dem "Helligen", dem Wort, dem
Verhltnis von Unmittelbarem und Mittelbarem. Nun ist aber gerade
"Wie wenn am Feiertage ... " eines jener Gedichte Hlderlins, die sich
ausgeprgt an Ereignissen der aktuellen Geschichte entznden. Hier wird
die Gegenwart der Franzsischen Revolution mit einem mythischen Ereignis, einem anderen Blitz, dem der Dionysosgeburt, in Beziehung gesetzt. In Heideggers Interpretation wie in .de Mans Auseinandersetzung
mit ihr und seinem eigenen Bemhen, ein Verstndnis dieses Gedichtes
zu entwickeln, verliert Hlderlins Gedicht jedoch jegliche Verankerung
in Geschichte und Mythos. Fr Heidegger erscheinen diese zweitrangig
im Vergleich zum Heiligen, das er den Zeiten und Gttern gegenber als
das Ursprnglichere urid Kommende bezeichnet. Wenn de Man Hlderlins Position der Mittelbarkeit hervorhebt, dann folgt er mehr der Logik
von Heideggers Argumentation, als da er sie aus dem Gedicht selbst
entwickelte. Das erklrt sich daraus, da es de Man ebensowenig wie
Heidegger, wie de Man ja festhlt, um die Interpretation des HlderlinGedichtes geht, sondern um ein Gesprch ber das "Wesen des Poetischen". So aber wird der dichterische Text nur zum Ausgangspunkt fr
eine Theorie der Literatur, in der er zu verschwinden droht5.
5 Vgl. die kritische Analyse von Heideggers Deunmgsverfahren gerade auch am Beispiel von "Wie wenn am Feiertage ... " durch Bernhard Bschenstein, "Die Dichtung


Das an der "Feiertags"-Hymne entwickelte Strukturmodell des Poetischen kehrt in de Mans letztem Hlderlin-Essay von 1970 (''The Riddle
of Hlderlin") wieder, nun jedoch mit charakteristischen Vernderungen.
Die Fixierung auf den Begriff des Seins ist verschwunden, Heidegger
nicht mehr explizit prsent. Hlderlin allein steht nun im Vordergrund.
Darin wird ein Weg von Heidegger zu Hlderlin erkennbar, der zugleich
auch der einer deutlichen Herausarbeitung von de Mans eigener literaturtheoretischer Position ist. In Diktion und Darstellungsweise unterscheidet sich dieser Aufsatz grundlegend von dem fihen Essay. Dies
mag mit seinem Erscheinungsort zusammenhngen. Es ist keine Abhandlung fr ein wissenschaftliches Journal, sondern eine Rezension fr
eine Wochenzeitung, der New York Review of Books, in der de Man Michael Hamburgers Hlderlin-bersetzung bespricht6. De Man entwickelt
aus seiner Kritik an Hamburgers bersetzung eine hchst interessante
Deutung von Hlderlins Dichtung, die er bei aller Komplexitt der angesprochenen Probleme doch auerordentlich przis und eingngig vorzutragen vermag. Ich will im folgenden einige Punkte dieser HlderlinDeutung hervorheben.
1. De Man betont den historischen Zusammenhang, in dem Hlderlins Werk steht und zwar sowohl in zeithistorischer wie auch in geistesgeschichtlicher Perspektive. Er weist auf die Bedeutung der Franzsischen Revolution ebenso hin wie auf die verbreitete Griechenland-Sehnsucht.
2. De Man umreit knapp und przis die immens komplexen Probleme, die sich stellen, wenn man es unternimmt, die Funktion der Gtter, des Gttlichen in Hlderlins Dichtung zu bestimmen. Er zeigt, da
sie weder als anagorisch, symbolhaft, pantheistisch noch im gngigen
Sinne als 'religis' begriffen werden drfen. De Man stellt es nachdrcklich als offene Frage hin, wie diese mehr als menschliche Perspektive,
die durchgngig in Hlderlins Dichtung eingenommen werde, bestimmt
werden knne. So wichtig de Mans Fragestellung ist, so wenig darf man
dem Umkehrschlu verfallen und annehmen, Hlderlins Dichtung habe
nichts mit den traditionellen Formen der poetischen Darstellung des Re-

Hlderlins. Analyse ihrer Interpretation durch Martin Heidegger", 'Zeitwende,

XLVID (1977), S. 79-97.

6 Friedrich Hlderlin, Poems and Fragments, translations by Michael Hamburger,
Ann Arbor, University of Michigan Press, 1970.


ligisen mehr zu tun. Im Gegenteil: Hlderlins Dichtung ist in ihren

Denk- und Sprachformen eindeutig geprgt von jenen Traditionen, die
de Man hier in ihrer Bedeutung einschrnkt. Gerade die Rckbindung an
eine Botschaft bildet dann auch die Grenze gegen poetologische Positionen der Modeme, wo das Ich in der sich sprechenden Sprache verschwinden soll. De Man hat vllig recht, wenn er betont, da Hlderlins
Dichtung nicht mit den Inhalten jener religisen Tradition einfach
gleichgesetzt werden drfe. Um aber das Eigentmliche, das Neue an
Hlderlins Dichtung bestimmen zu knnen, mu man sie in ihren Traditionsbindungen beschreiben. Erst so wird der Blick frei auf jene Momente, in denen sie sich von ihnen lst.
Aus den bislang genannten zwei Punkten, dem historischen Kontext
von Hlderlins Dichtung und der problematischen Funktion der Gtter,
ergibt sich fr de Man ein dritter wesentlicher Punkt zur Charakterisierung von Hlderlins Dichtung. In ihm werden sie gleichsam aufgehoben,
aber auch in charakteristischer Weise relativiert. De Man erinnert
zunchst an Hlderlins Drei-Tne-Lehre, der Tatsache, da seine Gedichte sich aus abwechselnd vorgetragenen ''Tonarten", dem naiven,
dem heroischen und dem idealischen konstituieren. De Man konkretisiert
diese drei Tonarten nach inhaltlichen Kriterien als Beschreibungen, besonders von Landschaften, als abstrakte Maximen oder Gnomen, d.h.
also "philosophischen" Partien und als Passagen von dramatischer Intensitt, wobei das Adjektiv "dramatisch" jedoch sehr unglcklich gewhlt
ist, weil es im Zusammenhang von Hlderlins eigenen poetologischen
berlegungen in vllig anderen Bedeutungszusammenhngen steht. Mag
dies zunchst etwas vereinfacht wirken, zeigt de Man doch in einzelnen
knappen Stilanalysen, da die Drei-Tne-Lehre, sieht man von den spekulativen, philosophischen und poetologischen Begrndungsversuchen
ab, die diese Lehre bei Hlderlin und seinen Interpreten immer umranken, durchaus auf diese Weise eingngig beschrieben werden kann und
sich in ihrer Funktion und Bedeutung erklren lt.
De Man sucht nun die "unifying voice" (S. 49) dieser verschiedenen
Tne. Die Bestimmung, die er gibt, scheint auf den ersten Blick tautologisch zu sein. Als den 'Grundton' jener drei verschiedenen Tne in Hlderlins Poesie bezeichnet er die Poesie selbst oder genauer, die poetische
Sprache. De Man entwickelt seine Bestimmung an einem Satz aus Hlderlins Hymne "Der Rhein". Es ist Beginn der vierten Strophe: "Ein
Rthsel ist Reinentsprungenes". Durch eine ausfhrliche, genaue Interpretation dieser Stelle, vor allem aber des Kontextes und des Gangs des

Gedichts, kommt de Man zu einer berraschenden, spannungsvollen

Michael Hamburger bersetzt das Wort "Rthsel" mit "mystery". De
Man lehnt diese bersetzung ab und schlgt statt dessen "riddle" vor. "If
the 'pure origin' were a mystery, it would be vain to search for its precise
meaning anywhere: the poem would have tobe read as a prayer or incantation [... ]. But if 'pure origin' is a 'riddle', then the poem has a different function. A riddle is not, in itself, out of the reach of knowledge, but
is temporarily hidden from knowledge by a device of language that can,
in turn, be deciphered only by another operation of language" (S. 50). De
Man bestimmt nun das "Reinentsprungene" im Verlauf der Hymne vom
Beginn des Rheins, dort wo er noch der Gefahr unterliegt "to lose itself
in the mystery of origin" ber Rousseau bis hin zur Gestalt des Sokrates
als einen temporalen Pr~ze, vom Chaotischen zur Ordnung, vom Instinkt zum Bewutsein, bei dem schlielich im Bewutsein, in der Reflexion der frhere Impuls in der Ruhe und Stille aufgehoben ist. Das
fhrt ihn dazu, die Gnome "Ein Rthsel ist Reinentsprungenes" in ihrer
spannungsvollen Umkehrungsmglichkeit zu lesen: "the sentence means
not only that 'pure origin' is a riddle but that the riddle itself is one of the
entities which can lay claim to pure origin" (S. 51). Wird das Reinentsprungene als Rtsel aufgefat, so wird der Aspekt der temporalen Prozessualitt von der Sehnsucht zur Reflexion, vom Instinkt zum Bewutsein betont. Das Rtsel als Reinentsprungenes wiederum verweist auf jenen Aspekt, wo im hchsten Bewutsein der Impuls des Ursprnglichen
noch bewahrt ist. In der poetischen Sprache vollzieht sich also die Verwie auch die mgliche Entrtselung des "Reinen". Wenn in der Wendung "Reinentsprungenes", das "Reine", wie de Man schreibt, "designates a proper balance between desire and reflection, between instinct
and consciousness between action and interpretation" (S. 51 ), so kann
die poetische Sprache eben das Gleichgewicht zwischen diesen Spannungen erreichen, weil sie den temporalen Aspekt der Nachtrglichkeit
und zugleich die Gleizeitigkeit dieser sich gegenseitig ausschlieenden
Paare umschliet, weil sie auch vennittelt zwischen dem Ursprung, dem
Unmittelbaren und dem Mittelbaren.
Unschwer lt sich hier in der Bestimmung der poetischen Sprache
die Analogie zur Definition des "Wesens des Poetischen" erkennen, die
de Man in seinem Heidegger-Essay gegeben hat. Doch ist die ontologische Fundierung der Struktur der poetischen Sprache aufg.egeben. Jetzt
wird ein anderer Aspekt deutlich, nmlich deren allegorische Fonn.

Zunchst wurde die Komplementaritt der Pole, die Gleichzeitigkeit

zwischen System und Chaos, Reflexion und Gefhl etc. betont. Doch ist
die Mglichkeit der Temporalisierung miteinbeschlossen, wobei die gegenstzlichen Elemente aus ihrer Simultaneitt heraustreten und als Anfangs- und Endstadium einer narrativen Entwicklung fungieren. Genau
diese Permutation zeigt das geschichtsphilosophische Schema, das de
Man in der "Rhein"-Hymne verfolgt und das vom mysterienhaften Ursprung des Rheins zur rtsellsenden und rtselstellenden Figur des Sokrates gleichsam auch eine Durchdringung von Poesie und Philosophie
entfaltet. Komplementaritt und Temporalisierung, Gleichzeitigkeit und
Mittelbarkeit, die den Bruch zwischen den gegenstzlichen Elementen
nicht aufheben, wohl aber in einem spannungsvollen Gleichgewicht zum
Austrag bringen knnen, weisen als Strukturmomente der poetischen
Sprache auf das, was bei de Man spter im Bereich des Figuralen als
Allegorie bestimmt wird.
Doch diese Interpretation der "Rhein"-Hymne gelingt de Man nur,
weil er die letzte Strophe des Gedichtes unterschlgt. Es endet eben nicht
mit der ausgleichenden, der vermittelnden Figur des Sokrates, sondern in
der Anrufung des Freundes Sinclair mit einem aktuellen Zeitbezug und
einer auffallenden Betonung der disharmonischen Momente im Geschichtsverlauf?.
Dir mag auf heiem Pfade unter Tannen oder

Im Dunkel des Eichwalds gehllt

In Stahl, mein Sinklair! Gott erscheinen oder
In Wolken, du kennst ihn, da du kennest, jugendlich,
Des Guten Kraft, und nimmer ist dir
Verborgen das Lcheln des Herrschers
Bei Tage, wenn
Es fieberhaft und angekettet das
Lebendige scheinet oder auch
Bei Nacht, wenn alles gemischt

7 In frheren Fassungen wurde zunchst der vterliche Freund Wilhelm Heinse angerufen. Zu dieser Vernderung, zur Bedeutung Heinses fr die "Rhein"-Hymne und
der wahrscheinlichen Herkunft der Formulierung "In Stahl[ ... ] gehllt" von Heinses
Beschreibung des famesischen Herkules, den er als "Gott in Stahl" bezeichnete,
siehe den Vortrag von Bernhard Bschenstein ber die "Rhein"-Hymne, den er 1988
auf der 20. Jahresversammlung der Hlderlin-Gesellschaft in Kassel gehalten hat.
Der Vortrag erscheint im Hlderlin-Jahrbuch XXVI (1988-89).


Ist ordnungslos und wiederkehrt

Uralte Verwirrungs.

Man knnte hier noch auf die Mglichkeiten hinweisen, die sich aus de
Mans Ansatzpunkt fr die Hlderlin-Deutung ergeben. Renate Bschenstein hat aufgefhrt, welche Bedeutung der allegorischen Ausdrucksform
in Hlderlins spten Texten zukommt9.
Man knnte ferner auf das Strukturprinzip der "Offenheit" eingehen,
auch und insbesondere in seinen temporalen Bezgen. Ich mchte hier
jedoch zuletzt noch auf den Zusammenhang von Historie, Geschichtlichkeit und Dichtung zu sprechen kommen. Gerade dort ist zu erkennen,
wie Denk- und Strukturmodelle, die de Man in seinen berlegungen
zum Verhlnis Heideggers zu Hlderlin entwickelt hat, weiterhin benutzt
Fr de Man kommen in Heideggers Zugriff auf Hlderlin verallgemeinernde und a-historisierende Tendenzen zum Ausdruck. Indem Heidegger, in seiner metaphysikkritischen Intention, Hlderlin in die Geschichte des Seins oder vielmehr der Seinsvergessenheit einordnet, bersehe er geistesgeschichtliche zusammenhnge und zeithistorische Bezge. Die Geschichte, die Historie wird bei Heidegger als uerlich und
zufllig abgewertet gegenber dem "Wesentlichen", das in der Dichtung
zur Sprache kommt. Gleichzeitig betont Heidegger die grundstzliche
Bedeutung der "Geschichtlichkeit", die "wesentlicher" verstanden werden solle als die Historie. Im dichterischen Kunstwerk drcke sich diese
Geschichtlichkeit in der Vereinung der drei Ekstasen der Zeit aus. Die
Vergangenheit und die Zukunft. werden in der Gegenwart des Gedichts,
in der es sich rhythmisch ereignet, verbunden.
Ein Blick auf de Mans Auffassung der Historie, wie sie in unserem
Zusammenhang von Interesse ist, erffnet sich am Ende seines letzten
Hlderlin-Essays. Hier erlutert de Man das Verhltnis von Geschichte
und Dichtung. Er kritisiert zunchst jene Auffassung, die glaubt, Hlderlins prophetische, eschatologische Aussagen als ein Versprechen auf
eine bessere Zukunft verstehen zu knnen. In der Ablehnung dieses et8 Friedrich Hlderlin, Smtliche Werke (Groe Stuttgarter Ausgabe), hrsg. von Friedrich Beiner, 8 Bde, Stuttgart 1946-1985, Bd 2: Gedichte nach 1800, S. 148.
9 Renate Bschenstein, "Hlderlins allegorische Ausdrucksform, untersucht an .der
Hymne 'An die Madonna'", Jenseits des Idealismus. Hlderlins letzte Homburger
Jahre (1804-1806), hrsg. v. Christoph Jamme und Otto Pggeler, Bonn, 1988, S.


was einfltigen Standpunktes (ein dichterischer Text wird wrtlich genommen als Prophezeiung) ist de Man ebenso zu folgen, wie in seiner,
nun im Gegenzug, getroffenen Feststellung, da die Dichtung sich stets
negativ auf die Geschichte beziehe, und zwar in dem Sinne, da sie gegenber der unmittelbaren historischen Erfahrung als Reflexion, also
Vermitteltes, immer nachtrglich sei. Und schlielich kann man de Mans
Schlufolgerung nur zustimmen, wenn er sagt, da die Dichtung, da
Hlderlins Dichtung "in its effectiveness lies well below history but well
above it in its wisdom" (S. 52). Aber sind, so ist nun anderseits zu fragen
"Effektivitt", die "unmittelbare historische Erfahrung", schlielich die
Kritik an einem einfltigen Wrtlichnehmen dichterischer Aussagen
tatschlich die Kriterien, um die Dichtung in ihrer Geschichtlichkeit, und
das wird hier gemeint, ihrem Bezug zur Historie, ihrem Eingebundensein
in die Geschichte, in ihrer Komplexitt zu erfassen? Oder ist dies nicht
vielmehr eine etwas eindimensionale Auffassung des Verhltnisses von
Geschichte und Dichtung? Sicherlich ist die Dichtung nachtrglich gegenber der unmittelbaren Wahrnehmung. Doch ist sie zugleich immer
schon eine eigene Form geschichtlicher Erfahrung. Sie hebt in sich Geschichte auf und ist selbst eine Form der Geschichte. Und mu man sich
schlielich nicht, gerade auch bei Hlderlin, zumindest mit dem Anspruch der Dichter auseinandersetzen, groe Sinnentwrfe liefern, mit
und durch die Dichtung auch auf die Geschichte wirken zu knnen?
Festzustellen ist, da de Man in seiner Beschreibung des Verhltnisses von Geschichte und Dichtung, hnlich wie Heidegger, zu einer Abwertung der Historie neigt. Es scheint, als wolle er die Dichtung aus der
Historie herausnehmen. Entweder steht sie unterhalb oder weit ber der
Historie, nicht jedoch in ihr. Gleichzeitig aber betont de Man, wie Heidegger, die wesentliche "Geschichtlichkeit" der Dichtung. Sie ist, wie de
Man hier feststellt, gegenber der unmittelbaren historischen Erfahrung
in ihrer Mittelbarkeit notwendig nachtrglich. Diese temporale Struktur,
ein wesentliches Bestimmungsmoment der Allegorie, ist jedoch auch,
wie wir sahen, fr de Man ein Kennzeichen der poetischen Sprache. Die
"Geschichtlichkeit", die temporale Struktur der Dichtung besteht darin,
da sie in ihrer Gegenwart notwendigerweise nachtrglich ist gegenber
einem "Ursprung", etwas "Unmittelbarem" und dies zuknftig stets nur
wiederholen kann. Die Dichtung, der Dichter ist umgeben von einem,
fast mchte man sagen, existentialistischen Pathos der Vergeblichkeit,
der verpaten Erfllung, das zugleich noch erhht wird durch seine permanente Reflexion darauf. "When Hlderlin evokes the possibility of

future moments of historical splendor", stellt de Man fest, "such evocations are accompanied by the foreknowledge that people will be conscious
of the achievement of these periods when they have ceased to be and
have become in turn parts of the past" (S. 52). In ihrer "Geschichtlichkeit'' wird die Dichtung, wie Suzanne Gearhart schon feststelltelO, bei de
Man gegen die als krude erscheinende Geschichte abgedichtet. Insofern
werden auch jene beiden Hinweise de Mans auf die historische, literarund geistesgeschichtliche Einbindung von Hlderlins Dichtung relativiert. Sie sind nun nicht mehr als periphere Faktoren, die das
'Eigentliche' der Dichtung, ihre "Geschichtlichkeit" unberhrt lassen.
Am Ende seines Essays scheint de Man dann selbst einer, von ihm stets
kritisierten, verklrenden Stilisierung der Figur Hlderlins zu erliegen,
wenn er seine Sprache als "the autonomous movement of a mind that
establishes its own domain" (S. 52) bezeichnet. Die Dichtung scheint
wesentlicher zu sein als die Geschichte. Wo Heidegger sie zum Ausdruck des Seins erhob, ist sie bei de Man - in negativer Umkehr Vergeblichkeit und Scheitern. Eine Auffassung der Dichtung, die sie als
verpate Erfllung und immerwhrende Anstrengung begreift, mag in
Sympathie fr Hlderlins tragisches Schicksal, die Magie seiner Sprache
und die Anziehungskraft seiner Persnlichkeit ihr Bild formen - ein
vollstndiges Bild Hlderlins ist dies jedoch nicht. Es kommt vielmehr
darauf an, Hlderlin in seiner spannungsvollen Gegenstzlichkeit zu begreifen, in seiner Einbindung in die Geschichte und dem Willen, sie zu
berwinden, in seinem Streben nach Ordnung und Gestalt und den Tendenzen der Destruktion, dem Gelingen seiner Dichtung und ihrem Milingen.
De Man entwickelt in der Auseinandersetzung und in der Kritik an
Heidegger eine Auffassung des Poetischen, die in ihrer Grundintention,
dem Bezug auf das Sein, noch an Heidegger orientiert ist. Das Strukturmodell des Poetischen als Spannung von Unmittelbarkeit und Mittelbarem, bleibt dann, weitgehend befreit von seiner ontologischen Fundierung, weiterhin bestimmend fr de Mans Charakterisierung von Hlderlins Dichtung, wobei nun das temporale Moment der N achtrglichkeit,
wie sie fr seine Theorie der Allegorie wichtig wird, deutlicher betont
wird. Diese Auffassung des Poetischen und seiner temporalen Struktur
liegt dann auch de Mans Beschreibung des Verhltnisses von Geschichte
10 Suzanne Gearhart, "Philosophy Before Literature: Deconstruction, Historicity, and
the Work of Paul de Man", Diacritics, XIII (1983), S. 63-81.


und Dichtung zugrunde, in der sich die einstige Orientierung an Heidegger wieder zeigt. Motiviert freilich jetzt aus anderen zusammenhngen,
wird bei de Man die Historie abgewertet und gleichzeitig die "Geschichtlichkeit" der Dichtung, ihre Temporalitt heivorgehoben. Wie bei Heidegger fhrt dies zu einer stilisierenden Erhhung Hlderlins, bei Heidegger zum Propheten, Zeugen und Hter des Seins, bei de Man gleichsam zu einem existentialistischen Sisyphos, der, um die Vergeblichkeit
seines dichterischen Unternehmens wissend, es dennoch und immer
wieder von vorne beginnt.
Jacques Derrida zitierte einmal zustimmend die Bemerkung eines
Freundes von de Man, der ihn als "Hlderlin in Amerika" bezeichnetell.
Vielleicht ist deutlich geworden, was Hlderlin, gerade in der Kritik und
in der Orientierung an Heidegger, fr die Entwicklung de Mans als Literaturtheoretiker bedeutete und wie sehr Hlderlin fr ihn zum persnlichen Inbild seiner Auffassung der Dichtung und des Dichters geworden


In his Hlderlin essays de Man develops a notion of poetry and the poet that is essentially
indebted to Heidegger. To Heidegger the poem is the expression of the "Holy", of the ineffable; to Hlderlin, and to de Man, the poet only can say the mediated. But for Heidegger and de Man the tension between the ineffable and the mediated governs the temporal
structure of poetry and its relationship to history. In the context of his "Seinsgeschichte"
Heidegger ahistoricizes Hlderlin. He makes him the prophet and the witness of Being.
De Man emphasizes the historicity of poetry. But also, as his interpretations of "Wie
wenn am Feiertage ... " and the "Rhein"-hymn show, he takes Hlderlin out of history,
making him a second Sisyphos who, conscious of the necessary failure of his poetry,
nonetheless repeats the anempt.



Jacques Derrida. Memoiresfor Paul de Man, New York, 1986, S. 10.

Marc W. Redfield


Das der Furcht im Charakter der Bedeutsamkeit Begegnende ist etwas Abtrgliches, wie Aristoteles sagt, ein x:a.x:6v, malum, und zwar ist dieses Abtrgliche immer etwas Bestimmtes. Wir wrden, wenn wir den Begriff hier
schon htten, sagen, etwas Geschichtliches, etwas Bestimmtes, das in die vertraute Welt des besorgenden Umganges hereinbringt.

Over the last ten years, the work of Paul de Man has not become any
easier to assimilate. Frank Lentricchia could not have been more wide of
the mark when in 1983 he predicted that the "war between traditionalists
and deconstructors" would "draw to a close by the end of this decade",
with de Man "rediscovered as the most brilliant bero of traditionalism"2 .
For even if de Man's youthful contributions to Le Soir had remained hidden a few more years in the archive, it is clear that Lentricchia would
bave lost bis wager. The furor over de Man's wartime joumalism has at
least bad the virtue of making manifest the extraordinary violence with
which his mature work is resisted. Doubtless, a measure of institutional
success continues to attend "de Manian" criticism. lt would be astounding if this were not the case, given the visible rigor of the methodology,
the prestige and relative power that de Man himself was able to achieve,
the cultural force of certain notions of comparative literature, theory, European philosophy, and so on. As a rule, however, contemporary criticism quarantines and ignores de Manian theory by way of various hegemonic strategies of inclusive exclusion, supplemented by extravagant
gestures of anthropomorphization and rejection. One could with considerable justice invert Lentricchia's formulations and claim that the most

1 Martin Heidegger, Prolegomena zur Geschichte des Zeitbegriffs [1925], Gesamtausgabe, 20, Frankfurt, Vittorio Klostermann, 1979, p. 395.
2 Frank Letricchia, Criticism and Social Change, Chicago, University of Chicago
Press, 1983, p. 39.


significant realignments of institutional power in literary studies during

the 1980's amount to wholehearted approval of the rhetoric of Criticism
and Social Change. Nothing, it seems, is more obvious than the political
inadequacy of de Man's texts. The task of pursuing some form of
"historicism" has taken on the self-evident necessity of an ethical imperative. "lt is a facf', de Man wrote in 1972, "that this sort of thing happens, again and again, in literary studies"3. What happens perhaps a little
more rarely in literary studies is the event of an exemplary figure such as
de Man, capable of inspiring the most lurid gestures of monumentalization and ritual sacrifice.
The pages that follow seek to articulate de Man's theoretical text with
the politics of his reception and with the question of politics. I shall be
pursuing the notions of history and politics that infonn de Man's late
texts, mounting an argument for their credibility and political usefulness.
However, my purpose is also to account, by way of the same vocabulary,
for the resistance his writing inspires. This topic acquires interest when,
like de Man, we understand "resistance" as a necessary component of
any act of reading. Overt displays of "resistance to theory", in other
words, should be understood as spectacular versions of the subtler
problematic posed by theory's "resistance" to itself. Par from composing
a frivolous exercise in self-reflexivity, this problematic defines the difficult necessity of a political criticism. The complement of fear and repression is idealization and identification: both are predicated upon a monumentalizing gesture without which no response to de Man seems able to
come into being. The very act of commenting, favorably or unfavorably,
on his work draws one into a network of effects characterizable in both
institutional and libidinal tenns. The politics of criticism and the politics
of charisma intersect within the event of this fortuitously anthropomorphic proper name. One is thus led to pursue what might otherwise seem a
needless complication: the relation in de Man's text between history,
3 Paul de Man, Allegories of Reading: Figural Language in Rousseau, Nietzsche,
Rilke, and Proust, New Haven, Yale University Press, 1979, p. 4. Subsequent references are indicated parenthetically in the text by acronym AR and page number.
Quotations from de Man's other books are indicated by page number and acronym,
as follows: Blindness and /nsight: Essays in the Rhetoric of Contemporary Criticism, Second Edition, Minneapolis, University of Minnesota Press, 1983, BI; Critical Writings, 1953-1978, Minneapolis, University of Minnesota Press, 1989, CW;
The Rhetoric of Romanticism, New York, Columbia University Press, 1984, RR; The
Resistence to Theory, Minneapolis, University of Minnesota Press, 1986, RT.


politics, and pathos. Further reasons for privileging this cluster of issues
will unfold as we negotiate de Man's theoretical propositions. But we can
suggest the nature of this topic's interest, and open the question of de
Man's "own" resistance to ("de Manian") theory, by considering, in the
most naive and literalistic fashion possible, the affective career of the
word "history" in his writing over thirty years.
De Man's essays have tended to address the question of history in an
elevated tone. With surprising regularity they have sought closure in
dramatic, aphoristic invocations of the historical. Occasionally the mood
is neutral or upbeat, as when, at the end of his clearly essay on the theme
of Faust, de Man writes that a genuinely thematic reading must "pass
from myth to idea, and from idea to formal theme, before being able to
become history" (CW, p. 88)4. More often, the tone is closer to that of
the closing phrase of "Tue Dead-End of Formalist Criticism", as it invokes "the sorrowful time of patience, i.e., history" (BI, p. 245). Tue existential idiom of these early texts, their thematization of history in terms
of a non-naturalistic, death-directed temporality, clearly favors but does
not entirely explain the recurrence of such a tone in essays so frequently
marked by a refusal of pathetic language5. Tue question is of interest
because de Man's penchant for granting the word "history" rhetorical
charge does not disappear as his attention shifts to rhetoric. His most famous, or infamous, aphorism on history is memorable partly because it is
- and has the ring of - a closing sentence: "the bases for historical
knowledge are not empirical facts but written texts, even if these texts
masquerade in the guise of wars or revolutions" (BI, p. 165). An essay
devoted to themes of political action in Rousseau ends with the dramatic
4 Tue closing cadence is slightly less portentous in the original. French: "... avant de
pouvoir devenir une histoire". Paul de Man, "La critique thematique devant le theme
de Faust", Critique, 120, May 1957, p. 404.
5 Tue refusal. of pathetic language, of course, hardly impedes, and if anything encourages, the recurrence of a pathetic tone. I am drawing attention here to one thematic
regularity among several in a general. rhetoric of mourning that one encounters
throughout de Man's work. For a study of the rhetoric of sacrifice in de Man, see
Minae Mizumura, "Renunciation", Yale French Studies, 69, 1985, pp. 81-97. Fora
particularly rich thematization of the temporal. pathos of history, see de Man's 1967
lecture "Time and History in Wordsworth", which has recently been recovered for
lhe archive in a special issue of Diacritics, 17, 4, Winter 1987, pp. 4-17. Tue lecture
holds special interest for critics interested in de Man's shift from existential to
rhetorical. terminologies, since he gave the lecture again in 1972, modified in ways
that the Diacritics text records in footnotes.


dramatic proposition that "textual allegories on this level of rhetorical

complexity generate history" (AR, p. 277). And in the late essays that
principally concern us here, de Man's prose will often acquire
extraordinary intensity at the very moment when he is repudiating the
pathos made available by notions of historical time. In "Shelley
Disfigured", an essay that bears on the historicity of an aesthetic object
"that has been unearthed, edited, reconstructed, and much discussed"
(RR, p. 93), de Man's tone, grimly elegiac throughout, rises memorably
as he concludes the essay with a resurrection of Shelley's dead body and finally, with a reintroduction of the charged word "history":
"Reading as disfiguration, to the very extent that it resists historicism,
turns out to be historically more reliable than the products of historical
archeology" (RR, p. 123). But perhaps the most dramatic instance of
such a deliberately pathetic renunciation of pathos occurs in the last
sentence of "Anthropomorphism and Trope", where the work of "true
'mourning"' unrolls as a bleakly sublime list of deprivations: "Tue most
it can do is enumerate non-anthropomorphic, non-elegiac, noncelebratory, non-lyrical, non-poetic, that is to say, prosaic, or better,
historical modes of language power" (RR, p. 262, de Man's italics). Tue
text performs what it denies, going to some length, in fact, to deliver a
certain version of the elegiac satisfaction it is renouncing.
History is of course not by any means always, in de Man's work, the
object of sibylline utterance or the cynosure of a concluding sentence.
Essays such as "Literary History and Literary Modernity", which thematize history at length, are for that reason, in fact, more rather than less
representative of an oeuvre that could with some justice be described as
obsessed by the task of thinking Romanticism, and literature in general,
as historical events. But when the question of "distinguish[ing] rigorously between metaphorical and historical language", between a mystified and an authentic perception of the historical, appears with its fll
force (BI, p. 164), de Man writes more elliptically, and at a significantly
higher pitch, than is usually the case. Naive as it would be to imagine
that de Man is "repressing" some entity called history, we should also
not hasten to call such rhetorical perfonnances self-reflexive. Certainly
one of their - quite seductive - functions in the late essays is to exemplify the difficulty of rendering "true mourning"; but the persistence with
which the word "history" has attracted rhetorical energy in de Man's
writing over three decades suggests the pressure of a panem irreducible
to what we ordinarily call the self-consciousness of an author or text. To

interpret this disturbance in the de Manian text within the tenns of the de
Manian text - which is to say, within the logic of an interrogation that
disqualifies for closure the "withinness" of logic or self-consciousness is a compelling and perhaps impossible task.
For various pragmatic reasons, in much of what follows I shall be
centering attention on de Man's late essays on. Kant and Schiller. With
Kant, the aesthetic definitively enters the institution of philosophical discourse, and, according to de Man, the question of Kant's reception composes not just the philosophical possibility of aesthetic judgment, but the
political burden of critical thought. "For it is as a political force that the
aesthetic still concems us as one of the most powerful ideological drives
to act upon the reality of history" (RR, p. 264). Tue late essays on German pre-Romantic and Romantic authors- on Kant, Schiller, Kleist, and
Hegel - take as their target an understanding of Romanticism deriving
from Hegel, which situates Schiller's ber die sthetische Erziehung des
Menschen ("the wellspring of romantic criticism", as Rene Wellek
claims) on a path leading from subjective to objective idealism, from the
Kritik der Urteilskraft to the Vorlesungen ber die sthetik6. De Man refigures this teleological commonplace into an economy of demystification and regression in which the name "Schiller" operates as a personification of aesthetic ideology7 Produced by, and yet incommensurate with
the "historical" event figured in the Kritik der Urteilskraft, the
"reception" of Kant takes its coordinates from Schiller's treatise, which
in its turn figures the most disastrous of political possibilities: we are
told at the end of de Man's late lecture draft, "Kant and Schiller", that
Goebbels' misreading of Schiller in his 1929 novel Michael "does not
differ essentially" from Schiller's misreading of Kants. In less dramatic,
6 Rene Wellek, A History of Modern Criticism: 1750-1950, London, Jonathan Cape,
1955, vol. I, p. 255. For Hegel's famous claim tliat Schiller broke through "die Kantische Subjektivitt und Abstraktion des Denkens", see G.f;W. Hegel, Vorlesungen
ber die sthetik, I, Werkausgabe, 13, Frankfurt a.M., Suhrkamp, 1970, p. 89. Fora
critique of Schiller's contribution to nineteenth-century notions of "culture", see
David Lloyd, "Arnold, Ferguson, Schiller: Aesthetic Culture and the Politics of
Aesthetics", Cultural Critique, 2, Winter 1985-86, pp. 137-69.
7 Schiller's name appears with some regularity in de Man's work, usually signifying a
certain misreading of Rousseau (see RR, pp. 20-6 passim and AR, pp. 137, 176,
208). However, Schiller only becomes a figure of emblematic stature when de Man
begins to write explicitly on the reception of Kant's Kritik der Urteilskraft.
8 "Kant and Schiller" was delivered at de Man's penultimate Messenger lecture at
Comell University in March, 1983. This lecture and another unpublished talk to


but perhaps equally significant ways, "Kant and Schiller" and "Kant's
Materialism" also yield what are pretty much the only explicit reflections
in the de Manian corpus on gender politics.
In negotiating de Man's invocation of Schiller, therefore, we engage
the question of political criticism as a question of reception. That question returns upon itself as one of our reception of de Man, and of "de
Man's" reception of "himself'. Tue genial but genuine tone of accusation
de Man adopts in "Kant and Schiller" as he reiterates one of the more
venerable commonplaces of Schiller criticism - that Schiller lacks philosophical rigor, has misunderstood Kant, and so on - is not simply a pedagogical device designed to animate a semi-improvised lecture9. This
personification is substituting for the dense pathos of essays like those on
Kleist, Baudelaire, or Shelley, and is ironically rehearsing the closure of
reception: if Schiller anthropomorphizes the aesthetic, de Man anthrowhich I shall be referring, "Kant's Materialism", are scheduled for publication in The
Aesthetic ldeology, edited by Andrzej Warminski, forthcoming from the University
of Minnesota Press. Since page references cannot be had at this time, my practice in
what follows has been to restrict quotation as much as possible to relatively long,
easily locatable excerpts. Where de Man's oral delivery occasioned uninteresting
solecisms, I have edited them out.
9 Schiller, the vulgarizer of Kant, the overpragmatic dramatist or overidealistic poet
incapable of genuine philosophical cogitation, is a stock character in German literary
history from Schiller's own time onward. Schiller's patron, the Duke of Augustenburg, wrote apropos of an early version of the sthetische Erziehung: "Our good
Schiller is not cut out for a philosopher; he needs a translator to elaborate his fine
phrases with philosophic precision, and to transpose him from the poetic into the
philosophic mode". Hans Schulz, Schiller und der Herzog von Augustenburg in
Briefen, Jena, 1905, p. 153. Quoted by Elizabeth M. Wilkinson and L.A.
Willoughby in their introduction to their translation of Schiller, On the Aesthetic
Education of Man, Oxford, Oxford University Press, 1967, p. cxxxviii. Fora summary of the main lines of Schiller's twentieth-century reception, and a glowing defense of Schiller, see Wilkinson and Willoughby pp. xlii-lxvii. - Finding fault with
Schiller is also a gesture with a more specific history. Adorno's remarks on Schiller
in sthetische Theorie are reminiscent of de Man's, and might constitute a useful
point of entry for a study of the complex presence of Adorno in de Man's writings.
"Die bei Kant beginnende Fetischisierung des Geniebegriffs als der abgetrennten,
nach Hegels Sprache abstrakten Subjektivitt, hat schon in Schillers Votivtafeln
kra elitre Zge angenommen. Er wird potentiell zum Feind der Kunstwerke; mit
einem Seitenblick auf Goethe soll der Mensch hinter jenen wesentlicher sein als sie
selbst. Im Geniebegriff wird mit idealistischer Hybris die Idee des Schpfertums
vom transzendentalen Subjekt an das empirische, den produktiven Knstler zediert".
Theodor W. Adorno, sthetische Theorie, Gesammelte Schriften, 1, Frankfurt a.M.,
Suhrkamp, 1970, p. 255.


pomorphizes the source of its error. The seductive promises of a certain

monumental self-reflexivity are in place, as are those of more banal
scenarios of naming and blaming. One will have no trouble imagining de
Man exorcising his own Schillerian wartime joumalism; and readers
willing to repeat in fll the Schillerian gesture will find in that image of
human self-interrogation relief from other questions.


One tends to speak easily of the essential or radical figurativeness of

language. The assumption often seems to be that this insight is easily
bome, or even fundamentally inconsequent. Having renounced all metaphysical and representational naivete, including, of course, the naivete of
believing that we could ever utterly renounce representational logic,
metaphors of grounding, notions of truth and lie, etc., we would, it
seems, be in a position to forsake linguistic for other, more practical or
obviously political topics. Versions of this pragmatic assurance surface
repeatedly in comtemporary criticism. And yet, if the radical figurativeness of language is granted, or suspected, all else in the de Manian narrative follows.
lt follows, first, that the paradigmatic condition of reading is a condition of suspense between a literal and a :figurative meaning. Since any
literal meaning is vulnerable to being read as a figure for another meaning, itself a figure, and so on, language as trope must be understood as a
process of circulation devoid of extemal support. Since, however, a
meaning, in order tobe read, must be taken in isolation from the possibility of tropological displacement, the condition of reading is structured
by a double possibility: that of figuration, and that of propriety of reference. This difference - the difference between the figural and the proper
- is itself that of figure; No extemal principle can regulate this difference
a priori, since no referent can definitively ground tropological displacement. This is why de Man writes at the beginning of Allegories of
Reading that "the grammatical model of the question becomes rhetorical
not when we have, on the one hand, a literal meaning, and on the other
hand a figural meaning, but when it is impossible to decide by grammatical or other linguistic devices which of the two meanings (that can
be entirely incompatible) prevails" (AR, p. 10). The figure that accounts

for and describes the possibility of the difference between literal and figurative meaning is the figure of this difference's undecidability. Radical
figuration implies the radical undecidability of figure. This undecidability defines, finally, the "text" (AR, p. 10), because there is no linguistic
vantage point extemal to it. Undecidability is what is given to us to read,
though by definition it cannot necessarily be read. What is given to us to
read is the possible impossibility of reading. This aporetic imperative
generates the plot of de Man's theoretical text.
One consequence of'rhetoric's radical suspension of meaning is that
language can no longer be understood primarily as an intentional structure. Tue popular idea that deconstruction "makes no difference" because prejudices are irreducible and one has to make decisions anyway,
etc., forwards the kind of complacency that might be underwritten by
substituting for rhetorical undecidability a phenomenological notion of
"suspension" (Aufhebung), in which the referent is bracketed through an
intentional actlO. But intention directs itself toward meaning; and if all
meaning is implicated in an undecidability of meaning arising from a
process of semantic substitution, then this process of substitution is possibly indifferent to meaning and intention. Language as figuration cannot
be reduced to a play of intentions, because language's formal principle of
articulation (or figuration) cannot be determinately motivated. We shall
retum to this problem in a moment, but consider first another implication
of radical figuration: the narrative or cognitive dimension of its error.
In order to be read, a figure must figure forth an aberrantly literal
meaning. Rousseau's primitive man, on bis way to language, sees another primitive man and experiences fear: out of fear he exaggerates the
other's size, and invents a primitive metaphor, "giant". Since this
metaphor has a proper meaning - fear - it is a proper metaphor, for all its
referential inaccuracy. But fear is not actually a proper meaning, being
"the result of a possible discrepancy between the outer and the inner
properties of entities" (AR, p. 150). Metaphor, in coming into legibility,
imposes meaning on undecidability (for "it remains an open question, for
whoever is neither a paranoiac nor a fool, whether one can trust one's

10 Edmund Husserl, Ideen zu einer reinen Phnomenologie und Phnomenologischen

Philosophie, Erstes Buch, ed. Walter Biemel, Husserliana., m, Haag, Martinus Nijhoff, 1950, p. 64 (para. 31).


fellow man"): the metaphor "giant" "freezes hypothesis, or fiction, into

fact and makes fear, itself a figural state of suspended meaning, into a
definite, proper meaning devoid of alternatives" (AR, p. 151). This dense
parable, which sets the stage for de Man's long and passionate engagement with Rousseau in Allegories of Reading, initiates figural narrative,
the allegory of the (im)possible of figure. An a priori condition of uncertainty has generated metaphor ("giant"), a reading seif (by virtue of
the internalized propriety of fear), and the possibility of referential denomination (the "giant" will be domesticated as a conceptual metaphor,
"man"). Figuration betrays itself, obliterating its own radical figurativeness. Put slightly differently, the consequence of referential indetermination is insistent referentiality. Language, de Man insists, must refer.
Like Marcel driven away from his books and out into the garden by his
grandmother, like the critics who at the beginning of Allegories of
Reading "cry out for the fresh air of referential meaning" (AR, p. 4), language turns away from its own figurativeness to produce literal meanings always marked in advance by the process of figuration that has produced them. Reference cannot be "avoided, bracketed, or reduced tobeing just one contingent property among others" (AR, p. 207). Werner
Hamacher has thus been led to organize a powerful account of the de
Manian system around the notion of an impossible and categorical referential imperative. "Language is imperative. lt is imperative because its
referential function gives the directions for possible reference, even if no
referential meaning answers to it and even though it corresponds to no
referent"l l. One could supplement the imperative "Reference must occur" with a variant characterization: "Intentionality must occur". And the
correlate of such imperatives is that "Reading must occur". The same
principle of error that produces these effects of reference and intentionality also marks them with the necessary possibility of being read as
mere figures. Referential indeterminacy "generates the illusion of a subject, a narrator, and a reader", and "the metaphor of temporality" (AR, p.
162). But since these illusions are figures of a figure, they bear within
them their own critique. In this sense they are self-deconstructive; but

11 Werner Hamacher, "Lectio: de Man's Imperative", in Lindsay W.aters and Wlad

Godzich, ed, Reading de Man Reading, Minneapolis, University of Minnesota
Press, 1989, p. 185.


since the deconstruction cannot halt or avoid repeating the error it reads,
"it engenders, in its turn, a supplementary figural superposition which
narrates the unreadability of the prior narration" (AR, p. 205). This second-degree narrative is what de Man calls allegory. Of such narratives
and their allegories "one should remember that they are the unfolding
and not the resolution of the chaotic uncertainty which Rousseau calls
fear" (AR, p. 162).
Consequently, it is possible to think of critical philosophy as the
thematic equivalent of allegory: of a critique of trope that is enabled by
the same spiral of error that produces referential illusion. The more rigorously the critique is pursued, the more surely it will reveal, unwittingly
and to no epistemological profit to itself, the tropological process that
enables it. And in the process, a certain limit to the notion of trope will
appear. As de Man recapitulates in "Kant and Schiller":
[T]he passage from trope to performative [...] occurs always, and can only occur, by way of an epistemological critique of trope. Tue trope, the epistemology
of tropes, allows for the 'critical discourse, transcendental critical discourse, to
emerge, which will push the notion of trope to an extreme, trying to saturate the
whole field of language, but then certain linguistic elements will remain which
the concept of trope cannot reach. [ ...]

Tue notion of the "performative" returns us to the topic we broached

earlier: the possible indifference of substitutive pattem to semantic detennination. Transcendental critical discourse is the critique of the possibility of trope: that is, of the figural structure that generates the epistemological field of truth and falsity as the task of judging literal and figural meaning. This narrative discovers undecidability as the condition of
its possibility. Semantic undecidability implies the potential irrelevance
of the principle of articulation to the meanings it articulates. Since
"figure" names the conjunction of signification with a principle of substitution, the notion of figure must now be revised to signify "the alignment of a signification with any principle of linguistic articulation whatsoever, sensory or not [... ] The iconic, sensory, or, if one wishes, the
aesthetic moment is not constitutive of figuration". Thus "the particular
seduction of the figure is not necessarily that it creates an illusion of sensory pleasure, but that it creates an illusion of meaning" (RR, p. 115).
Since the principle of articulation is possibly arbitrary, it becomes necessary to consider the role of a performative imposition of meaning on
random difference. A catachretic prosopopeia must "give face" to struc148

tural differences that can then be read as signs12. Figure must be figured.
Such a collusion between figuration and positional power is not cognitively masterable, for it is radically inconsistent: "language posits and
language means (since it articulates) but language cannot posit meaning;
it can only reiterate (or reflect) it in its reconfirmed falsehood. Nor does
the knowledge of this impossibility make it less impossible" (RR, pp.
117-18). Tue critique of trope finds its limit in its passage to a notion of
language as perfonnance. Twinned with that impossible perfonnance, as
we have seen, is the possible randomness of the articulative patterns that
will be yoked to meanings. This randomness of articulative pattern is
what de Man, in his late texts, calls "materiality".
De Man's most elaborately showcased parable of the materiality of
languagc is worth examining in some detail, since it organizes his readings of Kant and, indirectly, his reading of Schiller. lt is far beyond my
means here to reproduce the dense argument of de Man's reading of the
Kritik der Urteilskraft in "Phenomenality and Materiality in Kant" andin
the shorter, unpublished lecture "Kant's Materialism". For our purposes
it will suffice to note a few guiding themes; and "Kant's Materialism"
holds particular interest for us, since in this text de Man sets out to correct a misreading of the role of the "empirical" in Kant by reevaluating
the Kantian notion of affect. Kant does indeed attempt to resolve the divergence between form and content in the sublime by way of the affectivity of the subject. Thus, as Kant's rigorous transcendental critique of
trope forces the emergence of a language of power (in his text's abrupt
shift to a "dynamic" sublime), affective judgments take the place of rational judgments and we appear to reenter an empirical world of "assault,
battle, and fright" - for in the dynamic sublime, mental faculties must
struggle with nature, and an emotion such as admiration must do battle
with another emotion, such as fear. However, this strategy is not entirely
the "return of the empirical" it might seem. De Man claims that Kantian
typologies of affect tend to take their organizing principle from the
"dictionary" rather than from "experience", and that Kant is "often
guided by external resemblances between words rather than by the inner
resonances of emotion". Tue Third Critique's elaborate contrast between
12 See for a rigorous treatment of this predicament, Cynthia Chase's chapter on de
Man, "Giving a Face to a Name'', in her Decomposing Figures: Rhetorical Readings
in the RomanJic Tradition, Baltimore, Johns Hopkins University Press, 1986, pp. 82-



surprise (Verwunderung) and admiration (Bewunderung), for instance,

might weil be underwritten by no better organizing principle than the accidental similarities and differences of signifiers. The dynamic sublime's
concatenation of power and affect thus figures, as de Man reads it, language 's perfonnance of meaning.
The most sublime affect, Kant teils us, is in fact the absence of affect
(Affektlosigkeit), a noble a-pathy linked in turn to the grandeur of architecture. This conclu_sion surfaces in the midst of a set of dictionary-discriminations between sublime, active, male affects and beautiful, langorous, female ones; and de Man remarks:
the interpretation of the architectonic as a principle of masculine virility, as
pure macho of the variety (whatever the word may be), seems inevitable. But to quote Derrida: "When erection is at stake, one should never be
too much in a hurry - one should let things take their course (il faut laisser la
chose se faire)."[ ...] lf erection is indeed "la chose", then it is likely tobe anything but what one, or should I say men, think(s) it tobe.

Eventually I shall be reinvoking these comments, but consider for now

their ultimate object: an extraordinary paragraph in which Kant illustrates a general principle of aesthetic judgment: natural objects capable
of producing sublime effects must be considered from a radically nonteleological viewpoint. Kant provides as examples two landscapes and a
human body:
Wenn man also den Anblick des bestirnten Himmels erhaben nennt, so mu
man der Beurteilung desselben nicht Begriffe von Welten, von vernnftigen
Wesen bewohnt, und nun die hellen Pmikte, womit wir den Raum ber uns erfllt sehen, als ihre Sonnen; in sehr zweckmig fr sie gestellten Kreisen bewegt, zum Grunde legen, sondern blo, wie man ihn sieht, als ein weites
Gewlbe, was alles befat; und blo unter dieser Vorstellung mssen wir die
Erhabenheit setzen, die ein reines sthetisches Urteil diesem Gegenstande beilegt Eben so den Anblick des Ozeans nicht so, wie wir, mit allerlei Kenntnissen
(die aber nicht in der unmittelbaren Anschauung enthalten sind) bereichert ihn
denken [ ...] sondern man mu den Ozean blo, wie die Dichter es tun, nach
dem, was der Augenschein zeigt, etwa, wenn er in Ruhe betrachtet wird, als
einen klaren Wasserspiegel, der blo vom Himmel begrenzt ist, aber ist er unruhig, wie einen alles zu verschlingen drohenden Abgrund, dennoch erhaben
finden knnen. Eben das ist von dem Erhabenen und Schnen in der Menschengestalt zu sagen, wo wir nicht auf Begriffe der Zwecke, wozu alle seine
Gliedmaen da sind, als Bestimmungsgrnde des Urteils zurcksehen [...]13.
13 Immanuel Kant, Kritik der Urteilskraft, ed. Wilhelm Weischedel, Werkausgabe, X,
Frankfurt a.M., Suhrkamp, 1974, pp. 196-97. Quoted by de Man in "Kant's Materi-


If one takes this passage at its word, following its (impossible) injunction

to see non-teleologically, then "the only word that comesto mind" to describe it, de Man writes, "is that of a material vision" ("PMK", p. 135).
The passage resembles but differs decisively from Romantic pairings of
mind and nature: "No mind is involved in the Kantian vision of ocean
and heaven. To the extent that any mind, any judgment, intervenes, it is
in error - for it is not the case that heaven is a vault or that the horizon
bounds the ocean like the walls of a building". The eye sees only what
the eye sees, as the tautology of Augenschein indicates: this Schein is
thus neither illusory nor real, and consequently Kant's architectonic figures, read aesthetically, are not figures: "Heaven and ocean as building
are a priori, previous to any understanding, to any exchange or anthropomorphism. [... ] Kant's vision can therefore hardly be called literal,
which would imply its possible figuralization or symbolization by an act
of judgment" ("PMK", p. 135). "lt is in no way possible to think of this
stony gaze as an address or an apostrophe", de Man adds in "Kant's Materialism". "The dynamics of the sublime mark the moment when the infinite is frozen into the materiality of stone, when no pathos, anxiety or
sympathy is conceivable; it is indeed the moment of a-pathos or apathy,
as the complete loss of the symbolic."
Aligning this materiality with the scene's optical and architectonic
thematics, de Man thereby coordinates the "material" with the category it
traditionally opposes, the "formal". A nonteleological consideration of
the architectonic would not imply its total disintegration: "sea and
heaven, as the poets see them, are more than ever buildings." However,
"it is no longer certain that they are articulated (gegliedert)" ("PMK", p.
142).. What is lost is not all definition, as would be the case in a classical
postulate of matter without fonn, matter as pure potentiality; rather, what
is lost is the possibility of establishing an intemal necessity for the pattems of relations that allow signs to function as signs. The concatenation
of matter and form in "aesthetic vision" produces, within the context of
organic structure that infonns Kant's critical enterprise, a narrative of
dismembennent that ceases at minimal units of fonn: the vault of the
heavens; the limbs (Glieder) of the body; the letters of a word. "To the
alism" and in "Phenomena.Iity and Materiality in Kant", in Gary Shapiro and Alan
Sica, ed., Hermeneutics: Questions an4 Prospects, Amherst, University of Massachusetts Press, 1984, p. 133, 142. Subsequent page references to this latter essay
are given in parentheses, preceded by "PMK".


dismembennent of the body corresponds a dismembennent of language,

as meaning-producing tropes are replaced by the fragmentation of sentences and propositions into discrete words, or the fragmentation of
words into syllables or finally letters" ("PMK", p. 144). To view a
"letter" nonteleologically, of course, would not be to view it as part of an
alphabet, or as the instrument of a sign. Kant's eye thus sees at the heart
of the aesthetic "the absolute randomness of language, prior to any figuration or meaning" (AR, p. 299). If we graft on another of de Man's
tenns, we can say that this eye is seeing history.


No word in the de Manian lexicon retums to us more altered and charged

than the word "history". Earlier I had occasion to note the salience of the
tenn's rhetorical career in de Man's oeuvre; and at this point we can appreciate what is at stake when, in two of his very last essays, he offers to
align history with the "errancy oflanguage":
AB such, history is not human, because it pertains strictly to the order of language; it is not natural, for the same reason; it is not phenomenal, in the sense
that no cognition, no knowledge about man, can be derived from a history
which as such is purely a linguistic complication; and it is not really temporal
either, because the structure that animates it is not a temporal structure. Those
disjunctions in language do get expressed in temporal metaphors, but they are
only metaphors. (RT, p. 92)

In "Kant and Schiller" de Man defines the historical as the passage to

perfonnative and material notions of language that marked the exhaustion of figural, cognitive narration:
[H]istory ist not thought of [here] as a progression or a regression, but is
thought of as an event, as an occurrence. There is history from the moment that
words such as "power" and "battle" and so on emerge on the scene; at that moment things happen, there is occurrence, there is evenL History is therefore not
temporal, it has nothing to do with temporality, but [rather it has to do with] the
emergence of a language of power out of a language of cognition.

A far more exhaustive study of de Man's texts than I have been able to
furnish would be necessary before any definitive interpretation of de


Man's notion of history could be attempted. What I propose to offer here,

more modestly, is an argument for reading such definitions patiently.
Even in its most conventional sense, a "historical event" does not
possess or produce a "meaning" in the same way that a sign does. However, a historical event must nonetheless stand in a certain relation to an
ensemble of meaning-effects: i.e., a text. lt is a banal but easily forgotten
truth that death or pain or catastrophe "in themselves" do not possess the
slightest historicity. One way to characterize the peculiar quality of what
we call a "historical" event would be to say that it disrupts a text, in addition to helping constitute it. The event, as event, stands in a relation to
the text that the text itself cannot control; only retrospectively will it acquire fll status as a narrative event. Usually this historical characteristic
is rendered in referential language as the irruption or "resistance" of the
real. De Man employs terms that sound deceptively close to referential
language: in "Kant and Schiller" he speaks of the "occurrence" as that
which "has the materiality of something that actually occurs", that
"leaves a trace on the world, that does something to the world as such".
This notion of occurrence "is not in any sense opposed to the notion of
writing". A specifi~ally inscriptive violence is inseparable from historicity, as is a certain blank undeniablity: "by the fact that [the event] occurs
it has truth, truth value, it is true". There must be a sense in which historicity resists figuration, which is one' ground for de Man's notorious resistance to periodizations and genetic historicisms: "Such a narrative can
be only metaphorical, and history is not fiction" (BI, p. 163). To resist
figuration is to resist substitutive patterns of presence and absence that
articulate what we call the phenomenal - and human - world.
This ascetic, "material" notion of history may be difficult to acceptI4.
But there is a political thrust to de Man's thought that we are now ready
to negotiate. The "political" is defined in "Hegel on the Sublime" as the
prosaic "discourse of the slave", the "undoer of usurped authority"15 that is, as the enumeration of material, "historical modes of language
power". The political in this sense takes as its object aesthetic ideology.
Mystified, totalizing instantiations of aesthetic ideology make possible
the most damaging of political consequences. De Man's paradigmatic

14 For an infonned and careful account of de Man's notion of history, see Kevin Newmark, "Paul de Man's History", inReading de ManReading, pp. 121-35.
15 Paul de Man, "Hegel on the Sublime", in Mark Krupnick, ed., Displacement: Derrida and After, Bloomington, Indiana University Press, 1983, p. 153.


case is, as mentioned earlier, the "reception" of Kant by Schiller, whose

work "condenses the complex ideology of the aesthetic in a suggestive
concatenation of concepts" 16, and thereby reveals the aesthetic as what it
"primarily" is: "a social and political model" (RR, p. 264).
"Reception", in de Man's late texts, names the phenomenalization of
"history" by way of aesthetic syntheses. Schiller, rewriting the Kantian
sublime in his early essay "Vom Erhabenen", and subsequently elaborating the aesthetic into a full-fledged political system in ber die sthetische Erziehung des Menschen, domesticates and naturalizes the
Kantian critique by reproducing it as idealist empiricism17. I shall be
attempting a close reading of Schiller's text in the final section of this essay, and at this point wish only to recall the main lines of de Man's semiimprovised and relatively sketchy critique. The paradox that de Man addresses, and accounts for, is that an allegiance to the pragmatic or empirical makes possible the most thoroughgoing idealism. A pragmatic deflation of linguistic issues leads to ever more coercive linguistic structuration. Tue aesthetic renders language a property of the world; and in

16 De Man's phrase is actually describing the achievement of Elizabeth M. Wilkinson

and L.A. Willoughby's edition and translation of Schiller's sthetische Erziehung
(see note 9). This extraordinary bilingual edition, with its two hundred page introduction and extensive commentary, constitutes one of the most monumental - and
monumentalizing - gestures of canonization in recent scholarship. Quotes from the
sthetische Erziehung in what follows are from this edition, and are indicated by
letter and paragraph number: thus "1.1" for instance, means First Letter, first paragraph.
17 Readers who are not Gennanists, and who intend to work through de Man's essay in
greater detail, may benefit from a sense of the dates and occasions of the Schiller
texts. Schiller began to read Kant's Kritik der Urteilskraft intensely in the spring of
1791, and wrote the relatively obscure essay "Vom Erhabenen" in the spring of
1793. Also in 1793, Schiller composed for the Duke of Augustenburg the letters
that, massively transformed and elaborated, became the sthetische Erziehung of
1795. In 1801 Schiller discarded the first half of "Vom Erhabenen'', and republished its second half under the title "ber das Pathetische" in Kleinere prosaische
Schriften: the usual scholarly guess is that, eight years after its composition, Schiller
found the essay's first half too dependent on Kant. Schiller revised the sthetische
Erziehung for republication in Kleinere prosaische Schriften in 1801, but the
changes were relatively minor: the significant transfonnations in Schiller's aesthetic
theory had occurred between 1793 and 1795. De Man will speak of "Vom Erhabenen" as "early Schiller" for this reason. To avoid confusion, it should also be
noted that in 1795 Schiller published another essay on the sublime, "Uber das Erhabene", which de Man mentions but does not discuss.


doing so, it gives the world over to the indifferent cruelty of tropological
structures fundamentally alien to the universe of meaning they articulate.
Schiller's strategy, which is that of aesthetic ideology, is twofold. On
the one band, he grounds figural pattem in the phenomenal world by underst~ding chiasmic oppositions and transfers as the expression of
drives (Triebe). On the other band, he polarizes Kant's argument, recoding Kant's troubled passage from a "mathematical" to a "dynamic" sublime, for instance, as a binary opposition between a "theoretical" and a
"practical" sublime - an opposition that in Schiller's mature text, the
sthetische Erziehung, becomes an opposition between a Formtrieb, allied with reason, law, and other totalizing imperatives, and a sinnlicher
Trieb or Stofftrieb, which pursues the sensuous appeal of the moment.
Tue Formtrieb and the Stofftrieb find a peculiar mode of synthesis in
what Schiller calls Wechselwirkung, "reciprocal action": a chiasmus that,
given its purely formal nature, laclcs intemal necessity, and is forced to
derive its necessity from what Schiller takes to .be the incontrovertible
empirical facticity of the human. Language is thus grounded in the
"human" with exemplary force; and out of this synthesis Schiller derives
the most humanistic of drives, at once the sign, the cause, and the effect
of the human, the play-drive or Spieltrieb, directed at the appearance,
Schein 18
In short, Schiller's text produces and polices a representational concept of language. The phenomenal world of "reality" appears to direct
the mimetic exchange - even though binary oppositions such as that
between "language" and "reality'' are sheerly linguistic. This is to say
that mimesis is a trope, and that the formal pattems that permit the polarization and valorization of terms such as empirical and ideal, particular and general, etc., are not natural- that is, seif-evident and self-identical - but cognitive or tropological - that is, linguistic. A discourse that
18 De Man is not exaggerating, though as I hope to show later - and as de Man would
doubtless be the first to acknowledge - Schiller's text is more strained and complex
than de Man's comments might suggest. lt is certainly true that "the human" functions as a pragmatic, conceptually arbitrary principle of closure in the sthetische
Erziehung. When complications grow troublesome Schiller is given to saying things
lik:e, "But enough! Self-consciousness is there" (19.11); and at a crucial point in the
treatise, not far removed from the passages that concem us, we are told that Reason
must posit humanity.and beauty- i.e., the Wechselwirkung that defines the beautiful

and the human - be.cause Reason is Reason. "Wie aber eine Schnheit sein kann,
und wie eine Menschheit mglich ist, kann uns weder Vernunft noch Erfahrung
lehren" (15.4).


uncritically naturalizes linguistic structures will thus shuttle between opposites that imply each other. Tue initial privilege granted the phenomenal world can be - and is - revoked by chiasmic inversion: from a valorization of the empirical, one passes with ease to a celebration of the
spiritual. Language, initially domesticated as a reflection of empirical
drives or intentions, can now receive inverse valuation as a prefiguration
of the ideal. Tims the aesthetic is both domesticated and granted exemplarity - in Schiller's case to the point of making aesthetic harmony the
telos of individual and collective pedagogy, and a model for the State.
Tue synecdochi~ power of trope guarantees the passage from individual
to nation, artwork to culture, pedagogy to politics; and the logical end to
the system is the aesthetic state, the Staat des schnen Scheins, which is
for Schiller an ideal, realized only in a beautiful soul or within a circle of
friends (27.12), but which is in its turn vulnerable to tropological reinforcement and empiricization. lt is thus that de Man can claim that
Goebbels's vulgarization of Schiller repeats, however crudely, the essential gesture of Schiller's own text:
Tue statesman is an artist too. For him the people is neither more nor less than
what stone is for the sculptor. [...] Politics are the plastic art of the State, just as
painting is the plastic art of color. This is why politics without the people, or
even against the people, is sheer nonsense. To shape a People out of the masses,
and a State out of the People, this has always been the deepest intention of politics in the true sense19.

Tue continuity between statesman and artist, life and art, human being
and aesthetic object, so ferocious as to expunge any overt recognition of
violence (there can be no politics "against the people" in a structure of
such symmetry), violates every cautious, humanistic gesture tobe found
in the sthetische Erziehung. But it does not violate the treatise's deepest
logic. Tue "human" names an effacement of violence, not least, as de
Man remarks in closing "Kant and Schiller", when the "human" itself
discovers the necessity of deriving its closure from binary valorizations:
"Just as the sensory becomes without tension the metaphor for reason, in
19 Joseph Goebbels, Michael. Ein deutsches Schicksal in Tagebuchblttern, Mnchen
1933 (1929], p. 21. Cited (and translated) by Wilkinson and Willoughby, p. cxlii. I
have been unable to consult Goebbels's text in the original German. For an instructive account of Schiller's importance for the Nazi culture industry, see Georg Ruppelt, Schiller im nationalsozialistischen Deutschland: Der Versuch einer Gleichschaltung, Stuttgart, J.B. Metzlersche Verlagsbuchhandlung, 1979.


Schiller, woman becomes without oppression a metaphor for man". The

cost of aesthetic ideology in real violence and actual oppression can be
as enonnous or as modest, as literal or as symbolic, as any particular
context happens to pennit. The tropological patterns that make such distributions of meaning possible are essentially indifferent to the notion of
the human they enable. lt is the spectre of such indifference that humanism seeks to exorcise by appropriating and naturalizing linguistic structures: a gesture that sustains itself only in the mode of violent repetition,
since the principle of its success is also that of its disarticulation. A
threat is being taken as a solution, and the meaning and the perfonnance
of such a constitutive act of expulsion must thus ultimately be at odds20.


Tue affect proper to the irruption of "historical modes of language

power" ist more often than not, in the de Manian corpus, terror. Confronted with the possibility of the "uncontrollable power of the letter as
inscription", Saussure proceeds with a caution that "supports the assumption of a terror glimpsed" (RT, p. 37). The vision of sea and heavens is "a terrifying moment in a sense - terrifying for Kant, since the entire enterprise of philosophy is involved in it" - though de Man hastens
to discredit the idea of Kant "shuddering in bis mind" as he scribbled:
"Any literalism there would not be called for. lt is terrifying in a way we
don't know. [... ]" ("Kant and Schiller"). However, "literal" affect does
have its place in the de Manian allegory of reading: it derives, as we
have seen, from the effacement of undecidability that produces the possibility of literal meaning, which is to say the possibility of trope. A
rhetorical critique of language thematizes affect as a dimension of language's resistance to the random violence of its own inscription. Affect
resists history, insofar as it manifests itself as a dimension of a referential
imperative in flight from its own impossibility. Rousseau's parable of
20 This essay was on its way to publication before I had the chance to consult Cynthia
Chase's extraordinary essay, "Trappings of an Education", in Responses: On Paul de
Man's Wartime Journalism, ed. Werner Hamacher, Neil Hertz, and Thomas Keenan,
Lincoln, University of Nebraska Press, 1989, pp. 44-79. Chase's analysis of the historical and political dimensions of de Man's work constitutes to my knowledge the
most far-reaching study of these issues to date.


primitive man, experiencing fear in the face of language's event, is indeed a paradigmatic text for the de Manian narrative. Fear is a privileged
affect in a discourse about resistance. As an "empirical" affect, fear is an
illusory effect of metaphor's need for a proper meaning. And prior to becoming properly affective, we recall, fear was an impersonal epistemological suspension of semantic detennination (AR, p. 150-1): an allegorical personification of reading per se, and thus perhaps another figure for
the "true mouming" or non-empirical "terror'' that history occasions.
lt is thus perhaps also no accident that this figure of "fear'', so crucial
to the narrative of Allegories of Reading, should provide the axis for one
of the mostdensely intertextual negotiations in de Man's oeuvre21. A few
years earlier, in ''The Rhetoric of Blindness", de Man had taken issue
with Derrida's reading of the covert propriety of Rousseau's originary
"giant" metaphor - which professes non-referentiality, but actually denotes a proper and intemalized meaning, fear. "Rousseau's text has no
blind spots", de Man had claimed: Rousseau's text knows the truth of
radical figuration, and if the spontaneous metaphor "giant" finds its
proper meaning in fear, this is simply because Rousseau has made a
"mistake" in selecting fear to exemplify metaphor. "The choice of the
wrong example to illustrate metaphor (fear instead of pity) is a mistake,
not a blind spot" (BI, p. l 39n). When Allegories of Reading recodes fear
as the exemplary affect, structured like a trope that defaces its own figurativeness, de Man recodes the "mistake" as the undecidability between
"mistake" and "error": the metaphor must deface itself to compose itself
- though its instantiation may also be a random mistake. De Man's
reading of fear as mistake, by implication, was a mistaking of error
spurred by an error of mistaking. "If 'mistake' is random and contingent
21 Titls essay leaves aside, but wishes to recall and evoke, the Heideggerian subtext
constantly legible in de Man's work, and especially prominent in this chapter of Allegories of Reading. Though de Man's terms are dictated by the task of interpreting
Rousseau's text, it is not entirely coincidental that the operative, disputed tenn
should be the inauthentic (that is, im-proper: uneigentlich) affect "fear" rather than
the authentic Angst of a Dasein tuming away from its own potentiality. De Man's
intervention here should be read in tandem with bis gesture to replace the Heideggerian (and Kantian) "consciousness in itself' with "rhetoricity" (see AR, p. 175 and
note), and would ultimately have to be thought in relation to the occurrence or
Ereignis: to history as Geschichte, as that which occurs, as in the line of Hlderlin's
that encapsulates de Man's invocation and displacement of Heidegger: "Es ereignet
sich aber das Wahre." For Heidegger's classic discussion of Angst, see Sein und 'Zeit,
Tbingen, 1927, pp. 184-91.


[... ] and 'error' is systematic and compulsive [... ] then I have stated, in a
variety of terminologies, the impossibility of ever coming to rest on one
or the other side of this distinction", de Man remarks in a late text, recalling his "rash assertion" in ''The Rhetoric of Blindness" as an
example of mistaking error for mistake22. The (allegorical) nexus is fear,
or at least the wishful possibility of fear. And the stakes of mistake are,
of course, ethico-political as wen as epistemological.
Ideology, "the confusion of linguistic with natural reality, of reference with phenomenalism" (RT, p. 11), is the mistaken error built into
language: "lt is true that tropes are the producers of ideologies that are
not longer true" (RR, p. 242). "No degree of knowledge can ever stop
this madness, for it is the madness of words" (RR, p. 122). We have no
choice but to apostrophize the dead, monumentalize the text, phenomenalize the sign. And to the extent that the error is an error, its
equally inevitable: an epistemological critique of trope is "in no one's
power to evade" (RT, p. 69), even though the critique will discover
nothing more than the possible mistakenness of its error. Since we as
reading subjects are the products of this language machine, the exigent
contingency of its operation is replayed on the level of ethics. We cannot, de Man teils us, halt the madness of prosopopeia; however, we do
not have to delude ourselves into taking this process as a source of value:
such a belief "leads to a misreading that can and should be discarded,
unlike the coercive 'forgetting' that Shelley's poem analytically thematizes". But the discardable misreading then becomes difficult or impossible to discard: the "aesthetification of texts" describes also "their use, as
in this essay, for the assertion of methodological claims made all the
more pious by their denial of piety" (RR, p. 122). And so it goes: a spiral
of error that draws within it our ethical selves and the consciousness in
which we cannot help but believe; as the "product" of language's error,
we have no choice but to continue to choose. The ethical tonality of de
Man's writing reiterates the mistaken truth of error.
The rigor with which de Man stages this predicament is what makes
his work so difficult to read. It is easy to make the mistake of not reading
at all, as when Frank Lentricchia claims that de Man teaches political
22 Paul de Man, "A Letter from Paul de Man", Critical Jnquiry, 8, 3, Spring 1982, pp.
50'), 510. For a rigorous study of de Man's shift in position with regard to
Rousseau's "giant" metaphor, see Hans-Jost Frey, "Undecidability", Yale French
Studies, 69, 1985, pp. 124-133.


quietism by projecting "all those paralytic feelings of the literary onto

the terrain of society and history" 23 A more attentive reading discovers,

with J. Hillis Miller, that under the terms of de Manian thought, the
reader "must take responsibility for (the reading) and for its consequences in the personal, social, and political worlds"24. To adapt Kafka's
phrase: in the de Manian universe there is an infinity of "paralysis" - but
not for us. We cannot dwell within undecidability; reading must take
place, and to read is to judge: Miller is correct to extend the consequences of this model to the world of practical reason. Such is, for that
matter, the entire burden of aesthetic judgment. But if Lentricchia is
simply wrong, Miller is not simply right: he, too, domesticates de Manian theory by implying that language is an ethical entity25. We must
take responsibility, but responsibility is not ours to be taken. We must
act ethically, but we should not delude ourselves into thinking that such
action can be genuinely said to have value.
Tue intense, bleak pathos of de Man's work, particularly of his late
work, responds to the tenacity with which he pursues the impossible necessity of the ethical. In its full elaboration, the de Manian system - and
in its inevitable error, it is a system, teachable and generalizable, "the
universal theory of the impossibility of theory" (RT, p. 19) - is so thoroughly in control of the impossibility of ever being in control, that the
critique's, and the critic's, ethical imperative, recognized and named as
an impossible imperative of language, necessarily rewrites its intentionality in the mode of the pathetic. Tue system has accounted for this gesture long ago: such pathos repeats the illusory hypostatization of "the deconstructive passion of a subject" (AR, p. 199). And the subject whose
passion could animate such a system would be a "giant" indeed: "as far
beyond pleasure and pain as he is beyond good and evil, or, for that
matter, beyond strength and weakness. His consciousness is neither
happy nor unhappy, nor does he possess any power. He remains however
a center of authority to the extent that the very destructiveness of his ascetic reading testifies to the validity of his interpretation" (AR, pp. 1734). He would incamate the pathos of a-pathy, the sublime Affektlosigkeit

23 Lentricchia, op.cit., p. 50.

24 J. Hillis Miller, The Ethics of Reading: Kant, de Man, Eliot, Trollope, James, and
Benjamin, New York, Columbia University Press, 1987, p. 59.
25 I examine Miller's argument and develop my reservations in greater detail in
"Humanizing de Man", Diacritics 19, 2, Spring 1989, pp. 35-53.


of a subjectivity that recuperates phallic interiority in the mode of invulnerable impotence. He would derive castration out of disarticulation,
achieving thereby the funereal grandeur of an architectonic erection.
Thus to the pathos deriving from the power of de Man's thought corresponds the monumentalization of de Man as teacher, thinker and text.
From a certain perspective it makes little difference whether this monumentalization occurs in the mode of celebration or defiance: whether de
Man's text is fetishized and imitated, as in this essay, or whether it is
castigated and ritually sacrified. From a certain perspective it is also relatively indifferent whether one speaks of institutional effect or libidinal
investment: of the professionalization of de Manian theory, or the coercion of de Manian charisma. Both these modes of recuperation appear
united with exemplary force in the grotesque, funereally monumental issue of Yale French Studies dedicated to de Man, and an essay in that issue by Carol Jacobs provides an exemplary trope for the paradoxes that
control bis reception. "[De Man] may offer us a mirror of sons, but bis
writings [... ] are an aegis to which the head of the Medusa is affixed and
which we contemplate at our own risk"26. Jacobs is analyzing representations of the Medusa, and her remark is motivated and inspired by its
context; but as is often the case with figures, this figure of decapitation
cuts many ways. lt freezes de Man's visage into stone, evading, monumentalizing, and gendering his text at a stroke. A similar gesture can be
found in Schiller. Within the tenns of a de Manian problematic, the
Medusa's head is in essence a figure of reception27.

26 Carol Jacobs, "On Looking at Shelley's Medusa", Yale French Studies, 69, 1985, p.
27 In pursuing such a connection between de Manian and feminist concerns, we rejoin
the work of Cynthia Chase and Neil Hertz: see especially Chase's "The Witty
Butcher's Wife: Freud, Lacan, and the Conversion of Resistance to Theory", Modern
Language Notes, 102, no. 5, 1987, pp. 989-1013, and Hertz's chapters "Medusa's
Head" and "Afterword" in his The End of the Line: Essays on Psychoanalysis and
the Sublime, New York, Columbia University Press, 1985. Fora sustained reading
of what I have been calling de Man's "reception" of himself in terms of figurations
of gender, see Hertz's "Lurid Figures" inReading de ManReading, pp. 82-104.


De Man's interpretation of Schiller, as we have seen, centers on Schiller's
uncritical deployment of tropological structure as a defense against
trope. Imposing rigid polarities that stabilize and naturalize differences,
Schiller's text evades the perils of aesthetic Schein by relegating language to a mimetic role:
[Kant's Augenschein is] certainly not in opposition to rea.lity, but was precisely
what we see and as such rnore real than anything else, though it is reality which
exists on the level of vision. [...] And [in the case of both Kant and Hegel] there
is a road that goes frorn this notion of Schein to the notion of rnateriality. Such a
road cannot be found in Schiller, and that is why for Schiller the concept of art,
which at that rnornent is rnentioned and is stressed, will always and without
reservation be a concept of art as irnitation, as nacha.hmende Kunst.

That last claim, while quite correct, is vulnerable to the charge of not being sufficiently nuanced. Schiller's notion of Schein appears in the treatise's penultimate letter as the outward sign (Phnomen) of the psyche's
aesthetic mode. As such, the object of the play-drive, aesthetic Schein is
in one sense radically anti-mimetic: while Being (Dasein, Wesen) proceeds from nature, Schein proceeds from man. Any appearance that pretends to (natural) being or (referential) truth is not aesthetic Schein, or is
not being perceived aesthetically: in this sense, Schein is non-referential,
though in another sense it is the most referential of signs, since it refers
to the Human. Obeying the classic manoeuvres of what Jacques Derrida
has called "economimesis", Schiller's text thus recuperates mimesis by
way of an analogical chain leading from Schein to Man to what Schiller
sometimes calls "Nature" and sometimes "Absolute Being" or "the Godhead"28. This covert imitative chain incites the retum of the very language of mimesis that the text denies. The binary opposition between
Schein and Wesen, appearance and reality, is maintained with such enthusiasm in Schiller's text that the opposite of the real drifts implacably
into its classical role of being an image of the real, and thus with no apparent sense of contradiction Schiller can indeed write that the Spieltrieb
is followed by the "shaping spirit of imitation" (nachahmender Bildungstrieb) (26.7). De Man is not wrong, but the manoeuvres of



Jacques Derrida, .. Econornimesis", in S.

Paris, Flarnrnarion, 1975, pp. 55-93.


et al., Mime;yi;y dc;y artic"latianJ,

Schiller's idealist empiricism are more complex than Kant and Schiller"
allows for. If no road leads to the "material" in Schiller, what signs
mark, at least, the road's closure?
One way to pursue the track of Schein would be to examine the origins of the drive proper to it, the Spieltrieb, which makes its appearance
near the middle of Schiller's treatise under curious conditions. Schiller
has just identified the principle of chiasmus, Wechselwirkung, with the
principle of the human, and he is now moving from (what he, at least,
calls) a transcendental critique to more empirical considerations. A pure
Wechselwirkung between man's formal drive and bis sensory drive exists
only as an ideal, as "the Idea of Human Nature" (14.2); in the empirical
world we can only approach this ideal asymptotically, through time.
Schiller is then faced with the question of what enables the asymptotic
approach. And though at other points in the treatise originary questions
are dismissed as pre-critical distractions, here Schiller proposes a curiously empirical and ambiguous ontological fable. Its telling involves
Schiller in his most extended recourse, in this text, to the subjunctive
[ ...] so lange [der Mensch] nur empfindet, bleibt ihm seine Person oder seine
absolute Existenz, und, solange er nur denkt, bleibt ihm seine Existenz in der
Zeit oder sein Zustand Geheimnis. Gbe es aber Flle, wo er diese doppelte Erfahrung zugleich machte, wo er sich zugleich seiner Freiheit bewut wrde und
sein Dasein empfnde, wo er sich zugleich als Materie fhlte und als Geist kennen lernte, so htte er in diesen Fllen, und schlechterdings nur in diesen, eine
vollstndige Anschauung seiner Menschheit, und der Gegenstand, der diese
Anschauung ihm verschaffte, wrde ihm zu einem Symbol seiner ausgef.hrten
Bestimmung, folglich (weil diese nur in der Allheit der Zeit zu erreichen ist) zu
einer Darstellung des Unendlichen dienen.
Vorausgesetzt, da Flle dieser Art in der Erfahrung vorkommen knnen,
so wrden sie einen neuen Trieb in ihm aufwecken, der eben darum, weil die
beiden andern in ihm zusammenwirken, einem jeden derselben, einzeln betrachtet, entgegengesetzt sein und mit Recht fr einen neuen Trieb gelten
wrde. (14.2-3)

Tue subjunctive, I think, is taking up the strain felt by a passage that

does not entirely want tobe what it is - the account of a revelation. Of
Schiller's several, and often contradictory, accounts of the relation between ideal and real, this version, offered at a crucial point in the treatise, is coming close to proposing a Schein that precedes and incites the
drive proper to it. Before man can become man he must experience an
intuition of man, and such an intuition can only be had in the presence of

a schnem Schein that by the same token does not yet properly exist.
Schiller's essentially theocentric system would counter here with the
claim that the "human" exists always already in potentia, as a promise or
prefiguration (Anlage) at the "origin" of humanity (4.2); that is the sense
in which the "case" postulated in Letter 14 would merely "awaken"
rather than "create" the Spieltrieb. For though the transfonnation of a
mere Gegenstand into the specular, and spectacular, promise of a Symbol
suggests a dramatic positional act on man's part, here as elsewhere
Schiller's Fichtean gestures are actually being controlled by a more classic model of prefiguration and fulfillment. But the subjunctive mood of
the passage is responding to the .proximity of a threat. Either the prefiguration of the human is vulnerable, at the moment of its instantiation, to
chance, or the "object" is already Schein, a Schein before Schein that
would control the etiology of man's aesthetic education at the price of
imagining a Schein, an instance of beauty, stripped, at the outset, of the
"symbolic" character that defines it. Schiller's text is naturally dedicated
to closing off either possibility, but enough de Manian - or, according to
de Man, Kantian - burdens are bome by this "case" or "Fall" to require a
ceremony of exorcism, which takes place in the treatise's next letter.
For if it can only be postulated that "cases of this sort" occur in experience, Schiller's text can at least offer a certainty on the level of its own
engagement with the Fall of the aesthetic. In the wake of a long discussion of the beautiful, Schiller invokes the example of the Greeks, a people whose only error in the realm of the aesthetic was to "in den Olympus versetzen, was auf der Erde sollte ausgefhrt werden" (15.9). In
Olympus, at least, they gave face to beauty itself, and Schiller's letter
closes with a vision that operates rhetorically, and to some extent thematically, as a "symbol ofman's accomplished destiny":
Beseelt von diesem Geiste, lschten [die Griechen] aus den Gesichtszgen ihres
Ideals zugleich mit der Neigung auch alle Spuren des Willens aus, oder besser,
sie machten beide unkenntlich, weil sie beide in dem innigsten Bund zu
verknpfen wten. Es ist weder Anmut, noch ist es Wrde, was aus dem herrlichen Antlitz einer Juno Ludovisi zu uns spricht; es ist keines von beiden, weil
es beides zugleich ist. Indem der weibliche Gott unsre Anbetung heischt,
entzndet das gottgleiche Weib unsre Liebe; aber indem wir uns der himmlischen Holdseligkeit aufgelst hingeben, schreckt die himmlische Selbstgengsamkeit uns zurck. In sich selbst ruhet und wohnt die ganze Gestalt, eine
vllig geschlossene Schpfung, und als wenn sie jenseits des Raumes wre,
ohne Nachgeben, ohne Widerstand; da ist keine Kraft, die mit Krften kmpfte,
keine Ble, wo die Zeitlichkeit einbrechen knnte. Durch jenes unwider-


stehlich ergriffen und angezogen, durch dieses in der Feme gehalten, befmden
wir uns zugleich in dem Zustand der hchsten Ruhe und der hchsten Bewegung, und es entsteht jene wunderbare Rhrung, fr welche der Verstand
keinen Begriff und die Sprache keinen Namen hat. (15.9)

Schiller's treatise has never strayed further from Kant's dry, abstract
postulation of the "ideal of beauty" as a "human figure" capable of
summing up "the visible expression of moral ideas": "Die Richtigkeit
eines solchen Ideals der Schnheit", Kant continues, "beweist sich darin:
da es keinem Sinnenreiz sich in das Wohlgefallen an seinem Objekte zu
mischen erlaubt, und dennoch ein groes Interesse daran nehmen lt"29.
Tue Reiz of Schiller's Juno, meanwhile, is similar to that of the "human
figure" that Freud in his turn was to conjureup as an ideal of narcissism:
the woman whose "self-contented" aesthetic closure produces her "great
charm", which finds its "reverse side" in her "enigmatic being"30. Frozen
into monumental stone. schner Schein achieves its most radically formal figuration in the sthetische Erziehung, and could not be more
proximate to or distant from the "material" vision in Kant's Analytic of
the Sublime. Schiller's figure of affectless indifference substitutes its
gendered countenance for Kant's architectonic erection of sea and sky,
and its fetishistic rhythms of empirical "terror" and "ecstasy" for the terror of a disarticulation without meaning. Tue Medusa's head of aesthetic
ideology soothes through the fear and bliss of the possibility of experiencing fear and bliss: it marks the assertion of an act of identification
that would forget its figurativeness, and a dream of castration that would
discover its own deluded possibility by mouming the hypothetical
former existence of an erection that was not Kant's. At the considerable
political cost of grounding figuration in the symmetrical asymmetry of
gender difference, Schiller's text achieves the illusion of a desire forever
safe from language.

29 Kant, op.cit., p. 154.

30 Sigmund Freud. "Zur Einfhrung des Narzissmus", Freud-Studienausgabe, m,
Frankfurt a.M., S. Fischer Verlag, 1975, pp. 55-56. Fora relevant'reading of this
passage, see Sarah Kofman, L'enigme de la femme: La femme dans les textes de
Freud, Paris, Galilee, 1980.


For in naming the Juno Ludovisi, Schiller, miming and appropriating

Goethe's desire, domesticates a less naturalizable chain of substitutions
through a gesture of Oedipal rivalry31, Throughout the sthetische
Erziehung, classical statuary has bome a heavy figurative burden, representing the intersection of form and matter, meaning and medium, or,
most generally, reason and phenomenality, the articulation of which
composes the text's philosophical and political task. The fact that the nobility of the past can be preserved "in bedeutenden Steinen" (9.4),
"ein[ge]drck[t] in den verschwiegnen Stein" (9.6), means for Schiller
not only that atemporal Form, phenomenalized in art, can intersect the
temporal world, but that the aesthetic can underwrite and direct political
history. The aesthetic support (Sttze) which will ensure the endurance
(Fortdauer) of the political world as we know it while laboring
(etemally) at its transformation into the Aesthetic State (3.4), supports itself upon figures that evoke and evade the inscription: the random event
of meaning in "silent" stone. As a historical force the aesthetic may work
all too weil, but, at least in this text by .Schiller, not quite to the point of
effacing "the violence that makes- it possible" (RR, p. 289). The evasion
of aesthetic judgment, as de Man theorizes it, occurs as the imposition of
the coercive fascinations of a language of fear and desire, naturalized by
way of the binary polarizations of gender. Such scenarios reconfinn that
the tax levied by aesthetic ideology is not only thoroughly "empirical",
but that it is empirical because it is figurative, rather than vice-versa. To
identify such a moment as a defense against the event of language is thus
one way to begin to assess the violence characterizing the history of reception.



Goethe's fascination with the Juno Ludovisi dates from his Roman sojum; he installed a cast of the collossal bust ("my first sweetheart in Rome") in his rooms in
1787 and talked of taking it back with him to Weimar, but was eventually forced to
leave it behind In 1823, eighteen years after Schiller's death, Goethe obtained another replica, which dominates the "Juno-Zimmer" in what is now the Goethe Museum in Weimar. Schiller's invocation of the statue is tantamount to an explicit act
of homage, and would of course take its place in the narrative of adulation, desire,
insecurity and envy that constituted Schiller's side of what is perhaps the most ponderously canonized of literary friendships in Western literature.


Ein Hauptaspekt von de Mans Text zeigt seine Fhigkeit, die Problematik seiner eigenen
Rezeption zu theoretisieren auf. Dieses Voraussagevermgen konstituiert den epistemologischen und politischen Wert von de Mans Denken, verschrft aber zugleich die Versuchung, seinen Text als Autoritt zu "fetischisieren". De Mans Einblick in den
rhetorischen Charakter der Sprache kann nur gewonnen werden, wenn man denselben als
Einblick (d.h. Anschauung) falsch wahrnimmt. Die de Mansche Sprachkritik hat - und
voraussagt gleichzeitig - seine eigene Fehlanschauung als Darstellung des Erhabenen zur
Folge. Die spten Texte de Mans ber Kant und Schiller allegorisieren diese Problematik
durch ihre Darstellung der Schillerschen Rezeption von Kants "Analytik des Erhabenen"
in der Kritik der Urteilskraft. Wo die transzendentale Kritik Kants die echte Materialitt
der Sprache enthllt, kehrt die idealistische Empirik Schillers zur sthetischen
Naturalisierung tropologischer Strukturen zurck. Indem er diese Problematik in seinen
eigenen Texten wiederholt, lt de Man die Gewalt erkennen, mit der die sthetische
Ideologie ihre Illusion sttzt.


Laura Quinney


As against C.E. Pulos, who first argued that Shelley was a skeptic, of the
mild Socratic variety which would suspend confidence in our capacity to
make metaphysical determinationsl, it seems clear both that Shelley's
skepticism took a chillier and fiercer form, extending to suspicions of
life's malevolence, and that it was influenced as much by his literary aspirations as by his philosophical convictions. Any skeptical unveiling of
the world is inherently ambiguous, since it can mean the unveiling either
of .a numinous reality, or of the reality of nothingness. The word
"skepticism" accordingly ranges in meaning from doubt about the certainty or even the possibility of knowledge to annihilating dismissal of
what passes for mundane reality.
Skepticism begins, in any case, by claiming to furnish that wide survey that Wittgenstein calls "perspicuous representation". Wittgenstein
suggested that: "The concept of a perspicuous representation is of fundamental significance for us. lt earmarks the form of account we give,
the way we look at things." Thought promises us a "clear view of
things", and we therefore think it possible to command a full view of
whatever thought takes as its object. The desire for perspicuous representation may be in itself a "Weltanschauung", he added. In other words,
to esteem views ofthe world is already a kind of world-view2.
The flight over the earth and the heavens, one of Shelley's goveming
poetic ideas, realizes in physical form the claims of perspicuous representation. To take a relatively neutral example of this idea, there is the
ascension of Queen Mab's chariot, which rises over the oceans, past the
clouds, and beyond the sun:
1 See The Deep Truth: A Study of Shelley's Scepticism, Lincoln, University of Nebraska Press, 1962.
2 Philosophical lnvestigations, 3rd ed., trans. G.E.M. Anscombe, New York, Blackwell, 1958, 122.


The magic car moved on.

Earth's distant orb appeared
The smallest light that twinkles in the heaven;
Whilst round the chariot's way
Innumerable systems rolled.
And countless spheres diffused
An ever-varying glory. (I, vv. 249-55)3

Shelley was partial to this exhilarating fantasy of space travel. A somewhat lower, but similar flight is taken by the cloud, the skylark, the West
Wind, and the second of the two spirits, in the poems that bear their
names. In Hellas, it is Victorious Wrong who gains this birds'-eye view
on her descent toward earth, her prey:
I saw her, ghastly as a tyrant's dream,
Perch on the trembling pyramid of night
Beneath- which earth and all her realms pavilioned lay
In visions of the dawning undelight. (vv. 942-5)

This anthropomorphized image of flight readily combines with the purposes of Shelley's skepticism, since the identification of night as a
"pyramid" or "cone" and of Heaven as a "canopy" or "pavilion" does the
work of suggesting their limitedness, and the unknown magnitude of
what lies beyond.
Shelley's Ode to Heaven follows the skeptical logic inherent in these
metaphors. Apostrophizing Heaven as the "Palace-roof of cloudless
nights, /Paradise of golden lights", an anonymous celebrant flies out to
the limit of the Heavens, and looks back to see the planets, moons and
stars in their dance:
Glorious shapes have life in thee Earth and all Earth's company
Living globes which ever throng
Thy deep chasms and wildernesses
And green worlds that glide along [...] (vv. 10-14)

To this atheistic naturalist, human beings are the byword of evanescence:

"their unremaining Gods and they/Like a river roll away - Thou remainest such alway" (vv. 25-7). But "A Remoter Voice" intervenes, to
3 Unless otherwise noted, all quotations from Shelley are from Shelley's Selected Poetry and Prose, ed. Donald H. Reiman and Sharon B. Powers, New York, Norton,
1977. This edition provides the best texts for the poems that it contains.


take the idealist position that the Heavens are the evanescent apparition,
doomed to fade in the bigher reality of the ultimate, beyond "the portal
of the grave" (v. 32). A paradoxically "Louder and Still Remoter Voice"
has the last word, arriving at the most distant perspective, in wbich both
peopl~ and the Heavens are the liglltest motes of a radically expanded
What is Heaven? a globe of dew
Filling in the moming new
Some eyed flower whose leaves waken
On an unimagined world
Constellated suns W1Shaken
Orbits measureless, are furled
In that frail and fading sphere
With ten million gathered there
To tremble, gleam and disappear! (vv. 46-54)

This perspective, which by its position in the poem promises omniscience, dismisses to oblivion all the realities to wbich the other voices
gave authority. With this ascendence towards ever bigher overviews of
reality, the Ode to Heaven brings out how much skepticism depends on
the ambitions of perspicuous representation. At th~ same time, the
poem 's exhilarating widening of perspective makes manifest an intemal
logic of skepticism - its tenacity, its contagion, and its strange demand
for self-transumption. In professing its final authority, each voice teases
into being a more boldly annibilating position.
"Death is the veil wbich those who live call life" (III, iii, v. 112),
says the ambiguous Earth of Prometheus Unbound. Shelley's demand for
perspicuous representation realizes itself in figures of the veil and its
rending, and yet those figures are delicate and provisional to ihe point of
obscurity. There are layers, perhaps endless layers of veils, each of
whose rendings invokes the drama of perspicuous representation. In
Shelley, as elsewhere, unveiling is contagious. The dramatic gesture of
unveiling always makes itself vulnerable to its own unveiling as a veil.
Each unveiling must profess to be ultimate, at the same time that, by the
imperative of ultimacy, it calls forth its own supercession. Nevertheless,
in spite of bis insight into this self-defeating logic, Shelley did not abandon the attractions of perspicuous representation, and in fact bis last poems pursue it with the greatest energy. The rending of the veil underlies
the concluding lines of Adonais in wbich the elegist drifts towards suicide, as he beholds that ''The massy earth and sphered skies are riven!"

lt seems that perspicuous representation bends by some inner weight

towards grim revelation. Skepticism exposes the putative grounds of reality as a veil or screen of illusion. But once it begins exposing chimeras,
it seems condemned to go on exposing more chimeras, tearing away the
veil of still more fundamental grounds. The summoning up of a supramundane vision thus tends by an inner necessity to accelerating skepticism and dismissal. No one is surprised to find that the exhilarating
flights of Ode to H eaven should end in the bleak otherworldliness of

Heaven's light forever shines, Earth's shadows fly
Life, like a dome of many-coloured glass,
Stains the white radiance of Eternity. (vv. 461-3)

Skepticism is bound to generalizations about the nature of reality, and by

its logic of exposure and contagious self-transuming, it .seems bound as
weil to the premise that the truth it is constrained to reveal will be cruel.
Conversely, the paradigm ofthe truth's cruelty has no life apart from perspicuous representation.
Generalization itself is a means and mode of such a representation,
but one whose rhetorical form exerts its own attraction toward severity.
A sovereign finality is entailed in the requirement for condensation,
definitiveness and sharp persuasive force. Shelley's lyrics exemplify the
chann of generalization and its consequences. Wherever his lyrics interact with the anonymous fonns of wisdom literature - with proverbs and
aphorisms - they join in the sweeping designs of perspicuous representation. The first two stanzas of the 1821 lyric which Mary Shelley entitled Mutability consist entirely of succinct and violent apothegms:
Tue flower that smiles today
Tomorrow dies;
All that we wish to stay
Tempts and then flies;
What is this world's delight?
Lightning that mocks the night,
Brief even as bright.
Virtue, how frail it is!
Friendship, how rare!
Love, how it sells poor bliss
For proud despair!
But these though soon they fall
Survive their joy, and all
Which ours we call.


This poem makes manifest the tugging within generalization to darken

itself, as it works toward a conclusion whose bleakness supercedes that
of the laments leading up to it. The sources of joy die, but only after
having outlived joy itself. This escalation of despairing summaries
echoe~ the technique of self-transumption that energizes skepticism. But
it is more than an echo: the conjunction brings out the strange
interdependence of these particular rhetorical and conceptual forms:
poetic demands for sublimity and closure find a partner in critical
intelligence, doubt and disappointment.
The power of self-transumption finds another concise example in a
strong late lyric, ostensibly personal rather than forrnulaic, whose darkness lies in catastrophic generalization, the disappearance of the individual into the shadow of the aphorism and the law:
When passion's trance is overpast
If tenderness and truth could last
Or live, whilst all wild feelings keep

Some mortal slumber, dark and deep I should not weep, I should not weep!
lt were enough to feel, to see
Thy soft eyes gazing tenderly
And dream the rest - and bum and be
Tue secret food of fires unseen
Could thou but be what thou hast been!
After the slumber of the year
Tue woodland violets reappear;
All things revive in field or grove
And sky and sea, but two, which move
And form all others - life and love. -

The rhythm of the death of love is folded into the inexorable rhythm of
the earth. Yet the finality and impersonality of this reduction has a darkness not quite identical with the theme of the death of love. The severity
of the fiercely sweeping word, "all" (all things, all others), exerts an au:tonomous rhetorical magnetism. The charrns of generalization have here
exerted themselves through a strange logic, since it is in the absence of
life that things "revive". Totalized entities confront one another in a
mutually exclusive relation. "All things" revive but life and love- which
are all, in another sense. Tue structure of this ironic reduction has an authority independent of its subject. Whatever the terms in which they
filled it out, Shelley's poems pursued this reduction again and again.

From Alastor, through Stanzas written in Dejection, Julian and Maddalo, The Cenci, Adonais, and The Triumph of Life, it is as if dark generalization had to go in search of its content. Thus these two lyrics enact in
small the self-transuming, and self-consuming itineraries, not only of
many longer Shelley poems, but of his career as a whole.
A self-transuming fierceness dominates Shelley's poetry, no matter
how "Utopian", celebratory or lyrical its appearance. Shelley sought to
re-invent the sublime in an erotic, but still de-naturalizing and passionate, mode. This is the work of the third, and particularly, the fourth acts
of Prometheus Unbound, acts which would otherwise seem puzzlingly
prolonged and redundant, however beautiful. Harold Bloom has helpfully written that "In Act IV the imagination of Shelley breaks away
from the poet's apparent intention, and visualizes a world in which the
veil of phenomenal reality has been rent, a world like that of the Revelation of St. John [... ]"4 Prometheus Unbound is structured as an unveiling
of progressively higher realities. The scope of the play radically widens,
escaping the limitations of time and human perspective. This dramatic
expansion of view - the rending of veils - obeys the same skeptical imperative that generates the structure of the Ode to Heaven: Act IV presents still louder and remoter voices than Act III. At the heart of Act IV
is the hymeneal dance of the Earth and the Moon, anthropomorphized so
as tobe enfolded in the Ode's "frail and fading sphere". But their supernatural passion is unquestionably uncanny, as will be remembered from
the Earth's notorious exclamation, "lt interpenetrates my granite mass",
or the Moon's description of sexual awakening:
Gazing on thee I feel, I know,
Green stalks hurst forth, and bright flowers grow
And living shapes upon my bosom move

This revelation of a post-apocalyptic paradise provides the occasion for

the exuberance of a liberated figurative language. The "thrilling life" of
poetic figures stands in for the awakening of "unimagined worlds". But
these science-fiction imaginings already begin again to touch on strange
and chilling effects. Act IV does deviate from the play's humanist
agenda, and from Shelley's intention, but it is by way of sustained obedience to the aspirations of his elevated poetics. Though Prometheus U n4 Introduction to Percy Bysshe Shelley: Modern Critical Views, ed. Harold Bloom,
New York, Chelsea House, 1985, p. 10.


bound has been allegorized as the flowering of Shelley's "hope", its hieratic style does not distinguish it from the fierce rhetoric of his putatively
"disillusioned" poems.
Shelley's poetry is characterized by this peculiarly ambivalent combination of passion and severity. For to the skeptical theme of the world's
blankness and poverty, Shelley joined a strong current of eroticism - or
rather, he heightened a latent one. In the course of traducing him, F .R.
Leavis called attention to Shelley's vivid and telling revisions of The
Recollection5. In the original version, Shelley had described an
ephemeral sensuous contentment in language that his revisions rendered
A spirit interfused around,
A thinking silent life;
To momentary peace it bound
Our mortal nature's strife; And still, it seemed, the centre of
The magic circle there,
Was one whose being filled with love
The breathless atmosphere.
("The Pine Forest of the Caseine Near Pisa")6

Leavis shrewdly identified the vocabulary of these lines as Wordsworthian, by contrast with the "characteristically Shelleyean attitude"7 that
took their place:
A spirit interfused around,
A thrilling silent life, To momentary peace it bound
Our mortal nature's strife;
And still I felt the centre of
The magic circle there
Was one fair form that filled with love
The lifeless atmosphere. (vv. 45-52)

Don~d Davie, Purity of Diction in English Verse, London, Routledge and

Kegan Paul, 1967, pp. 144-45.
6 Quoted from Poetical Works, ed. Thomas Hutchinson, new edition corrected by
G.M. Matthews, Oxford, Oxford University Press, 1970.
7 Davies' phrase.

5 See


Shelley's changes ("thrilling" for "thinking", "fonn" for "being" and

"lifeless" for "breathless") have, as Leavis said, the effect of supercharging these lines with eroticism. What makes them "characteristically
Shelleyean" is not just eroticism but, as Leavis did not remark, the particular, ambivalent fonn of it - the eroticism of one lovely thing shining
in darkness. This fonn of the erotic is inseparable from nihilism. The
change from "breathless" to '_'lifeless" produces catastrophe. Whereas
"breathless" may pass for a description of still air, "lifeless" is unmistakably fateful; it introduces the sense of phenomenality's emptiness, the
death-in-life of life, which is familiar from Shelley's other late poems.
All the revisions of The Recollection work to intensify the poem's
particular pathos of tenuous fantasy. In this way they are all eroticizing,
though not in the restricted, interpersonal fonn that Leavis had in mind.
The most interesting are the revisions of the fourth stanza, in which the
reflecting pool, not mimetically exact, but fruitful of beautiful mutations,
makes its appearance. The draft describes the reflection as "A purple
finnament of light I Which in the dark earth lay, I More boundless than
the depth of night, I And clearer than the day", while the final version
not only reads "purer than the day", but more subtly changes "A purple
finnament of light" to "A finnament of purple light". Pools of water on a
dark ground reflect a dark sky; the phrase "purple light" carries this image toward "the supematural. The natural world is transformed out of natural possibility by the properties of the reflecting medium.
Shelley constantly evokes the bleakness of this eroticism - the extent
to which the strange wonder of the erotic object throws the world's
poverty into relief. But it manifests itself nowhere more clearly than in
the ghastly hyperboles of Epipsychidion:
Seraph of Heaven! too gentle to be human,
Veiling beneath that radiant form of Woman
All that is insupportable in thee
Of light, and love, and irnmortality!
Sweet benediction in the etemal Curse!
Veiled Glory of this lampless Universe!
Thou Moon beyond the clouds! Thou living Form
Among the Dead! Thou Star above the Storm! (vv. 21-28)

Tue eroticism of such a moment is a fonn of despair, since the celebration of Emilia expresses itself in contempt for all other available reality.
Many quotable passages, which are generally discussed under the rubric


of Shelley's "search for the Ideal", might be cited here, but the conclu~
sion to be drawn is evident. lt is of very little importance. whether or not
the poetry celebrates a temporary incarnation of the Ideal; here the distinction between nihilism and affirmation is not profound.
The spectacle of the world's emptiness, which the panorama of perspicuous representation tends to produce, has special rhetorical attractions. Shelley's poetry often evokes its peculiar intensity. In this Shelley
was following, not only Humean skepticim, but a long line of wisdom
literature beginning with Ecclesiastes. "Lift Not the Painted Veil" both
dramatizes the doomed search for what Shelley calls "love", source of "a
secret correspondence with our heart"8, and treats the seeker himself as
our object of love, though unregarded and lost:
Lift not the painted veil which those who live
Call Life; though unreal shapes be pictured there
And it but mimic all we would believe
With colours idly spread, - behind lurk Fear
And Hope, twin Destinies, who ever weave
Their shadows o'er the chasm, sightless and drear.
I knew one who had lifted it .. he sought,
For his lost heart was tender, things to love
But found them not, alas; nor was there aught
Tue world contains, the which he could approve.
Through the unheeding many he did move,
A splendour among shadows - a bright blot
Upon this gloomy scene - a Spirit that strove
For truth, and like the Preacher, found it not.

This scenario of disappointment is familiar from. other Shelley poems.

But as framed by the directive, not to lift the painted veil, it joins the tradition of wisdom literature and participates in the paradoxes of didacticism within that tradition. The poem pictures what it says we ought tobe
blind to, and teaches what it begins by claiming we ought not to learn.
The wisdom it offers is as openly useless as that in the aphorism from
Ecclesiastes that appears as the epigraph to Shelley's early poem 0 n
Death: "There is nor work nor device, nor knowledge, nr wisdom, in
the grave, whither thou goest". Yet this uselessness in no way vitiates the
rhetorical energy of the proposition. On the contrary, it becomes for this

8 See Shelley's "Essay on Love", in Shelley's Prose, or The Trumpet of a Prophecy,

ed. David Lee Clark, Albuquerque, University of New Mexico Press, 1954, p. 170.


reason impressive in the highest degree. This story of knowledge has at

its heart the strength of severe and daimonic generalization.
From the prematurely defeatist Alastor through The Triumph of Life,
Shelley chose perpetually to re-enact the skeptical discovery of the
poverty of the world, recalling the trauma of this discovery to restless
life. Because skeptical disappointment is vivid only as an individual experience, he kept dramatizing it in tenns of tragic awakening. Leavis' influential canard, that Shelley's "characteristic pathos" is "self-regarding"9 emerges as the unfortunate consequence of Shelley's readiness to
enlist the first-person, if not himself, in this wintry drama.
Shelley regularly regretted the world's incapacity to saturate love or
imagination. lt is odd to think of this regret as itself erotic, but of course
it is. By the same doubling that structures "Lift Not the Painted Veil",
the Alastor poet leaves behind a world whose deadness has only been revealed by the withdrawal of bis own ephemeral, illuminating presence.
Tue final lines of the poem read:
Art and eloquence,
And all the shews of the world are frail and vain
To weep a loss that tums their lights to shade.
lt is a woe "too deep for tears", when all
Is reft at once, when some surpassing Spirit,
Whose light adomed the world around it, leaves
Those who remain behind, not sobs or groans,
Tue passionate tumult of a clinging hope;
But pale despair and cold tranquillity [ ...] (vv. 710-18)

This turn on the skeptical theme - when the disappearance of the unsaturated lover is treated as our erotic loss - helps to bring out how it is that
Shelley's gloomy poems are invested with erotic pressure. There is here
an eroticism-through-grimness which can be found to suffuse many
Shelley poems. Tue identification of an erotic object - lost or yet to have
- is not in itself so necessary; eroticism inheres in the freshness of skeptical disappointment.
Disappointment displaces eroticism without eradicating it. A diffused cathexis of the world, and the threat of its defeat, sustain eroticism,
which can thus be bom out of the generalizations of skeptical severity.
What Leavis calls "erotic" in the substitution of "thrilling" for "thinking"
9 Revaluation: Tradition and Development in English Poetry, London, Chatto and
Windus, 1969, p. 220.


and "lifeless" for "breathless" is just as erotic in Shelley's compellingly

dark generalizations and the strong, suspenseful rhythms in which they
are expressed, as in these lines from Adonais: "He hath awakened from
the dream of life - 'Tis we, who lost in stormy visions, keep/With phantoms an unprofitable strife". The distinctive ambivalence of Shelley's
poetry arises from the rhetorically contradictory character of its generalizations. This strange eroticism govems the daimonic splendor of Shelley's late poems.
Shelley knew the attractions of tragic representation weil, having
early in his youth written poems like On Death and the 1814-15 version
of M utability: "We are as clouds that veil the midnight moon;/How restlessl y they speed, and gleam, and quiver,/Streaking the darkness radiantly! - yet soon/Night closes round, and they are lost forever." These
first experiments in gloom appear to seek confonnity with some invisible
standard of literary strength. How is it that Shelley, who in his adolescence was addicted to morbid cliches, came to write persuasive and
original versions of the truth's austerity? His poetry does witness profound development of a technical order. The magnetism of a severe and
daimonic style brought Shelley to the ironizing precision of Adonais and
The Triumph of Life. Yet Shelley is romantically imagined to have followed over the course of his life an itinerary in which each phase of his
thought exposed the comparative nai'vete of the previous phase, and accelerated by one order of magnitude in the sublimity of its darkness. But,
as we have seen, the self-transumptions of skepticism already shape the
rhetorical structure of his individual poems.
Add to this that Shelley's poetry repeatedly dramatizes the debilitating incorporation of the truth - from Alastor to Julian and Maddalo to
The Triumph of Life. His interest in this theme has even given rise to odd
superstitions about the prophetic nature of his poetry. lt is not uncommon for critics to describe Alastor or "Lift Not the Painted Veil" as the
prophecy of Shelley's own destiny. But the very fact that he began in a
preoccupation with the itinerary of tragic discovery makes it difficult to
persist in describing the course of bis career as a blind enactment of it.
As the peculiar rhetorical structure of "Lift Not the Painted Veil" suggests, the changes in Shelley's poetry itself do not represent the movement from blessed ignorance to tragic knowledge. A more subtle movement within knowledge was in fact the subject of a curious fragment
from 1820:


Alas! this is not what I thought life was.

I knew that there were crirnes and evil men,
Misery and hate; nor did I hope to pass
Untouched by suffering, through the rugged glen,
In mine own heart I saw as in a glass
The hearts of others And when
I went among my kind, with triple brass
Of calm endurance my weak breast I armed,
To bear scom, fear, and hate, a woful mass!
(Oxford Shelley, pp. 633-34)

Tue fragment breaks off just at the point of distinguishing the speaker's
pre-experiential knowledge from his experiential discoveries, though it
takes work to imagine what dark truth he had yet to leam. Tue distinction may have been difficult to tease out as a constative one, a difference
in the content of knowledge. As long as dark generalization goes in
search of its content, then its increasing success cannot properly be ascribed to the result of deepening knowledge. Tue temptation is to portray
the difference between early and late Shelley as a difference in ideas, but
the difference - sometimes in ideas, sometimes in style, intensity, or expertise - actually reflects a constant progress along a route determined
by an autonomous literary imperative.
Tue odd sense of circularity of tautology in Shelley"s career - the
way in which bis poetry seems to have predicted its own fate, as weil as
Shelley's fate, and to have accomplished a trajectory toward tragedy with
which it was always pregnant - reflects its obedience to a necessity
within literature. Those who write about this force rationalize it by representing its bending arc as the emanation of Shelley's gradual and unforeseen disillusionment. Thus, while Judith Chernaik may write that
"these last lyrics confirm the darkening of [Shelley's] spirit", his
"pessimism", and the "withdrawal" of bis hopesi 0, or while Bloom may
praise "the wisdom of disillusion" that Shelley achieved after Epipsychidion, and the "realistic sorrow and wisdom" to which he ascended in
The Triumph of Lifel 1, they are only giving naturalizing and spiritualizing names to a literary necessity to which they respond at the same time
in finding that Shelley's "disillusionment" and "darkening" brought
10 The Lyrics of Shelley, Cleveland, Case Western Reserve University Press, 1972, p.
11 Visionary Company: A Reading of English Romantic Poetry, second edition, Ithaca,
Comell University Press, 1971, p. 335.


about the fulfillment of the dream of his poems. The themes of

demystification and dawning undelight themselves take shape within the
paradigm of the grimness of the truth. Shelley's critics follow this
paradigm in perceiving bis mind to have darkened, and in this way they
echo his skeptical vocation.
De Man's "Shelley Disfigured" obviously does not involve itself in
this picture of Shelley's deepening knowledge. But it may choose other
forms in which to participate in the paradigm of the grimness of the
truth. lt is an odd feature of the essay that it attributes to the poem such
an exact and accurate apprehension of the "linguistic predicament" that
de Man himself studied to uncover. lts "negative knowledge" does not of
course prevent the poem from suffering the random textual articulation it
describes, since it is fractured by "the actual death and subsequent
disfigurement of Shelley's body"12. But de Man does not argue, as he
does in roughly contemporaneous essays on Wordsworth and Rousseau,
that the text itself regresses from its knowledge. The Triumph of Life perfected an austerity which de Man recapitulates, changing the vocabulary
. but not the resonance of the poem's skeptical fierceness. In The Triumph
of Life, de Man's writing found a comparable text to enlist in its own
enterprise of consolidating philosopbical precision and rhetorical severity.
Shelley and his friends imagined that he bad fulfilled the destiny of
literature at his own expense. In her note on the poems of 1822, Mary
Shelley ascribes to bis poetry a strange collusion with his death (an ascription that many others make, including contemporary students of
[... ] though dreams and hues of poetry cannot blunt grief, it [poetry] invests his
fate with a sublime fitness, which those less nearly allied may regard with complacency. A year before he had poure.d into verse all such ideas about death as
give it a glory of its own. He had, as it now seems, almost anticipated his own
destiny; and when the mind figures his skiff wrapped from sight by the thunderstorm, as it was last seen upon the purple sea, and then, as the cloud of the tempest passed away, no sign remained of where it had been - who but will regard
as a prophecy the last stanza of Adonais? (Oxford Shelley, p. 679)

Despite the frisson of uncanniness lingering about this quotation, it is not

the case that literature conspired in Shelley's death, but that the spectre
of this uncanny possibility arose when his life conformed .to the fateful
12 The Rhetoric ofRomanticism, New York, Columbia University Press, 1984, p. 120.


arc pursued by bis poetry. The "sublime fitness" of which Mary Shelley
wrote, the conformation of Shelley's life to the designs of bis poetry, is a
variant of the "sublime fitness" occasioned by the successful self-transumption of tragic skepticism. Shelley's poetry recreated the grimness of
the truth, and his life illustrates its meaning.


Im allgemeinen sieht man Shelleys Entwicklung als einen Weg, der von jugendlichem
Idealismus zu einem desillusionierten Ende fhrte. Sein ausgeprgter Skeptizismus trieb
ihn aber von Anfang an dazu, Ideen und Stilelemente zu suchen, die das Paradigma der
finsteren Wahrheit erfllen wrden. Dieses Paradigma wird noch durch seine Bevorzugung der rhetorischen Strenge untersttzt. und in dieser literarischen Finsternis wurde
Shelley Meister. Wenn man das berhandnehmen des Finsteren als ein Zeichen der
wachsenden Verzweiflung interpretiert, so akzeptiert man die grausame Aufdeckung
durch die Wahrheit. De Man verineidet zwar diese psychologisierende Auffassung der
finsteren Sachzwnge in seinem "Shelley Disfigured", bemht aber trotzdem das
Paradigma der finsteren Wahrheit. Die kontrollierte Strenge seines eigenen Essays
verbindet sich harmonisch mit Shelleys Skeptizismus.


Claude Reichler



On se saurait trop insister sur l'importance de Rousseau dans l'oeuvre de

Paul de Man. Au temoignage de Derrida, en 1966 deja de Man etait sur
la voie d'une inteipretation nouvelle de Rousseaul. C'est d'ailleurs cet
ecrivain qui noue le dialogue entre Derrida et de Man a la fin des annees
soixante2 , dialogue essentiel pour l'essor de la deconstruction aux EtatsUnis. Toute la decennie 70 est marquee par une succession d'etudes sur
Rousseau, et se termine avec la parution de Allegories of Reading3. Dans
ce livre, Rousseau occupe les six demiers chapitres, toute la seconde
partie de l'ouvrage. C'est l'ensemble le plus long et le plus developpe
consacre aun auteur par Paul de Man.
On ne peut manquer d'y etre frappe par le caractere systematique de
l'inteipretation. Le commentaire, appuye sur de breves sequences, joue,
d'un chapitre al'autre, sur des anticipations et des retrospections, aux fins
de tenir ensemble, dans une texture serree et repetitive, le texte lu et les
concepts inteipretatifs. Dans un ordre qui suit a peu pres sa chronologie,
toute l'oeuvre de Rousseau est examinee, aussi bien les grands textes (Le
Discours sur l'origine de l'inegalite, La Nouvelle Heloise, l'Emile, le
Contrat social, les Confessions) que quelques autres, mineurs, auxquels
de Man corrtere une fonction strategique (ainsi Pygmalion ou Narcisse
pour l'analyse du moi). Tous les lieux communs de la critiqm~, ce qu'on
pourrait appeler la pensee Rousseau, sont passes en revue: la theorie
politique, le moi et l'autre, la passion, la foi, la loi, la definition de
1 V. Jacques Derrida, Memoires pour Paul de Man, Paris, Galilee, 1988.
2 En temoigne, dans l' aire fran~aise, la parution dans un des premiers numeros de la
revue Poetique, alors tres en vue, de la traduction de The Rhetoric of Blindness:
Rhetorique de la cecire. Jacques Derrida lecteur de Rousseau, Poetique, 4, 1970.
Le texte de de Man, revu, est paru dans Blindness and Jnsight, Minneapolis, Univ. of
Minnesota Press, 2e ed., 1983.
3 Allegories of Reading. Figural Language in Rousseau, Nietzsche, Rilke and Proust,
New Haven and London, Yale Univ. Press, 1979 (=AR).


l'homme, la musique, la denonciation du paraftre, la culpabilite,

l'exhibition ... Tous ces objets de pensee sont mis dans une lumiere nouvelle, parfois etincelante. Un champ problematique est dessine, complexe et riebe, grce aune reflexion qui cherche toujours ase depasser et
ase radicaliser.
Cet ensemble est ouvert par l'analyse du Second Discours
(Discours sur l'origine et les fondements de l'inegalite parmi les hommes, 1755), ou sont mis en place les concepts et les instruments qui vont
operer dans les chapitres suivants et dans de nombreux autres essais. La
confrontation avec de grandes interpretations recentes de Rousseau y
joue son rle egalement (Levi-Strauss, Starobinski, Derrida), surtout
quant a la nature du langage. La these centrale met clairement en evidence le renversement anti-humaniste en jeu dans les theories structuralistes et post-structuralistes: L'homme est une propriete du langage.
C'est cette these qu'il convient donc d'examiner, non dans ses applications generales mais en discutant les fondements qu'elle se donne, voire
la valeur de la caution qu'elle cherche dans le texte de Rousseau, puisque
les theories de maniennes se presentent d'une maniere anti-deductive,
dans une elaboration ameme le texte, une pratique de la lecture.

La doctrine

11 est sans doute utile, pour commencer, de rappeler le mouvement du

chapitre 7 de Allegories of Reading, premier de la serie de 6 chapitres
consacres a Rousseau, et intitule Metaphor. De Man part de la constatation d'une division entre les interpretes de Rousseau, certains se
consacrant a la theorie politique, d'autres a l'aspect litteraire de l'oeuvre,
comme s'il y avait deux domaines de l'interpretation, l'un litteral,
l'autre figure. Cette repartition pose la question de la relation entre le
texte et son refrent, question qui reflete d'ailleurs, pense de Man, la situation du Second Discours lui-meme, dans lequel une fiction narrative
(l'etat de nature) est proposee pour rendre compte de l'etat actuel de la
societe politique. Est aborde ainsi le probleme, central dans la reflexion
de de Man, de la valeur epistemologique de la litterature. Si on parvient a
degager cette valeur, on montre du meme coup l'unite de Rousseau, on
legitime le fait quune fiction puisse expliquer une realite; mais surtout,
pense de Man, on montre qu'en fait il n'y a pas de division dans le lan184

gage, qu'il n'y a qu'une sorte de langage, et que les categories admises
(politique, morale, litterature ... ) sont secondaires.
Le centre de l'etude de de Man porte donc sur la theorie du langage,
que Rousseau aborde comme accessoirement dans le Second Discours, et
dont il traite aussi dans l'Essai sur l'origine des langues4. Mais, pour
bien souligner l'importance de ces questions dans un texte qui semble ne
les aborder qu'obliquement, de Man conduit d'abord a une aporie la reflexion sur la liberte et la perfectibilite, qui constitue un centre explicite
de la premiere partie du Discours. Rousseau, argumente de Man, en faisant de la liberte l'essence de l'homme, definit ce demier comme un objet
impossible a connaitre, puisqu'il transgresse toujours le cadre qu'on lui
assigne. La liberte et la perfectibilite sont a la racine du changement historique (aussi bien technique que social et politique), mais aussi de la
negativite presente en l'homme, que chaque progres eloigne de son origine. Pour comprendre veritablement cette condition douloureuse, il faut
analyser, pense de Man, la description de la formation du langage, que
Rousseau aurait liee structurellement a la formation de la socit~te, et que
les critiques laissent toujours de cte, comme si elle constituait une
menace. (On sait combien cette idee du point aveugle compte pour de
Man, et toute l'influence qu'elle a eue. Elle joue un rle comparable a
celui que jouent ailleurs l'inconscient ou le signifiant, en permettant
au critique de decider lui-meme qu'il y ade l'insu chez un auteur, et que
c'est cela le plus important.)
Cette analyse, qu'on discutera plus loin, donne ade Man l'occasion
d'une meditation sur la metaphore et sur sa valeur cognitive, dans des
termes auxquels il reviendra tres souvent: la metaphore est aveugle, eile
est une erreur, eile defigure. Elle donne lieu aussi a un expose de
doctrine en philosophie du langage, ou sont presentes d'une fa~on tres
dense les principes qui deviendront ceux de la deconstructions. La for4 De Man cite le Secorul Discours d'apres l'oo. des Oeuvres completes, t. III, Pleiade,
1964 (la pref. et les notes du Secorul Discours sont dues a Jean Starobinski). 11 cite
!'Essai d'apres l'ed. A. Belin de 1817, qu'il a trouvee reproduite dans le supplement
au NO 8 des Cahiers pour l'analyse. Une edition separee est parue depuis lors, chez
Aubier Montaigne, 1974, introduction et notes par A. Kremer-Marietti. On peut consulter egalement la savante ooition de Paola Bora, Turin, Einaudi, 1989; sur les rapports de I'Essai et du Discours, v. les notes pp. 19-24. V. surtout l'edition toute recente de J. Starobinski, Folio/essais, 1990, avec preface, notes et annexes.
5 L'etude est parue d'abord sous forme d'article, en 1973: Theory of Metaphor in
Rousseau's Secorul Discourse, in Studies in Romanticism, 12, 2.


mulation la plus serree en est la suivante: All language is language

about denomination, that is, a conceptuel, figural, metaphorical metalanguage. 6 Aucun enonce ne peut porter sur des choses, mais seulement
sur des proces de denomination, eux-mfmes radicalement incertains
parce qu'ils reposent sur des deplacements de sens, que l'analyse retrouve
toujours derriere l'illusion de connaissance. Ainsi l'homme, le concept
d'homme, serait chez Rousseau une metaphore conceptuelle. Si la societe civile existe bel et bien, eile n'est pourtant pas fondee sur une volonte politique des hommes, mais sur une Structure linguistique que de
Man decrit comme a lie superimposed upon a error7. Tel serait le rapport entre la fable initiale (l'etat de nature) et la description socio-politique finale dans le Second Discours. Tel serait le veritable message de
Rousseau, que ses exegetes auraient jusqu'alors obstinement refuse
d'entendre: The political destiny of man is structured like and derived
from a linguistic model that exists independently of the subject.s La
nature, l'ethique, le theologique, le politique sont des effets du
langage; l'unique acces au modele qui les commande est la litterature,
comprise ici comme un langage conscient de son mode d'enonciation
rhetorique, de son erreur constitutive.
On reconnaft les articles doctrinaux de la pan-textualite, dont
s'autorisent les disciples de Paul de Man pour deconstruire tous les savoirs, traites comme des enonces metaphoriques, donc errones. Cependant ces speculations devenues dogmes sont loin d'etre demontrees, et
particulierement dans le commentaire qu'on vient de resumer, ou elles
prennent essor sur une faute de base dans l'interpretation du texte de
Rousseau, et creent pour se soutenir des monstres linguistiques et
philosophiques. Je n'entends pas dans ce tenne le jeu de mots connu sur
la monstration, mais plutt une allusion a son usage dans la langue du
dix-huitieme siede, ou il designe des creatures hybrides, du mulet a la
sirene, dont on pensait - ou dont on souhaitait - qu'elles etaient incapables de se reproduire ...

6 AR, pp. 152-3.

7 AR, p. 155.
8 AR,p.156.


Genealogie d'un monstre

11 faut en venir au detail de l'analyse que propose de Man sur les pages
consacrees par Rousseau ala formation du langage. 11 s'agit d'abord d'un
assez pref passage situe dans la premiere partie du Second Discours, pp.
146-151 de l'edition Pleiade. Qu'il me soit pennis de considerer un instant les embarras de l'origine des langues, dit Rousseau. 11 montre les
embuches logiques et chronologiques auxquelles on se heurte lorsqu'on
tente de savoir ce qui a commence, du langage ou de la societe, ou encore du langage ou de la pensee. 11 esquisse une serie d'hypotheses genealogiques, marquant, par paliers et en relation avec son propre systeme
philosophique, le debut du langage expressif dans le cri, puis le debut du
langage cognitif dans la pure designation, avant d'en arriver a
l'acquisition des idees generales et des termes generiques. Les substantifs ne furent d'abord qu'autant de noms propres, ecrit-il; et plus bas, de
maniere coneordante: Chaque objet re~ut d'abord un nom partieulier
(p. 149). Tel ebene reneontre dans la nature recevait le nom de A, explique-t-il; et tel autre, a ete, etait appele B: le monde etait une nomenclature infinie. Ce n'est que lentement que les observations aeeumulees
permirent aux hommes de decouvrir les earacteres communs et les proprietes eommunes a plusieurs individus, et de designer les ehoses par
des ctenominations communes et generiques (p. 149). Opposant donc
le nom propre au nom commun eomme le particulier au generique ou
eomme la pure deixis au maniement de rapports abstraits, Rousseau
decrit la genese de la capacite eoneeptuelle en mettant al'origine, selon
son systeme, le eoneret et l'immectiat contre le reflexif et le mediat.
Paul de Man suit pas a pas le raisonnement de Rousseau, et
l'interprete en des termes qui ne sont pas indifferents. Pour l'exemple des
deux ebenes, il parle de la denomination litterale du nom propre (AR,
p. 146; je souligne). Et il explique le mouvement de la conceptualisation
decrit par Rousseau comme un echange ou une substitution de proprietes sur la base de la ressemblanee, qui correspond exaetement a ladefinition classique de la metaphore teile qu'elle apparatt dans les theories
rhetoriques d'Aristote a Roman Jakobson. (Ibid., je souligne). De Man
cteplace les oppositions utilisees par Rousseau: nom propre I nom commun (ou particulier I general) devient litteral I figure. 11 confond la mise
en evidence de proprietes communes entre un nombre n d'unites, qui permet de designer ces unites par le meme tenne generique, et la perception
d'analogies partielles entre deux objets compares, qui conduit transfrer


a l'un le nom de l'autre. Aucun philosophe ni aucun ecrivain fonne a la

rhetorique classique ne s'aviserait de confondre ces deux. mouvements, et
de superposer la fonnation du concept a l'invention de la metaphore, Rousseau pas plus qu'aucun autre. De Man a besoin de decouvrir cette
pretendue Superposition pour introduire dans Sa reflexion Uil passage de
l'Essai sur l'origine des langues (le chap. III) intitule Que le premier
langage dut etre figure, dont il pense que les arguments sont entierement paralleles aceux du Second Discours9.
Dans ce chapitre, celebre chez les deconstructionnistes et rituellement rappele, Rousseau soutient l'idee paradoxale que le sens propre fut
trouve le demier. 11 l'appuie sur l'exemple d'un homme sauvage qui,
effraye par l'aspect d'autres hommes rencontres, leur aura donne le nom
de geants. Par la suite, voyant combien ces hommes lui ressemblent, il
invente un autre nom commun a eux et a lui, tel, par exemple, que le
nom d'homme (ibid., je souligne commun a). Rousseau oppose deux series: le sens figure, qui obeit al'affect, a la passion, a l'illusion, et le sens
propre, auquel sont associees la raison et la verite. A nouveau, mais dans
un autre domaine du langage, on reconnait le systeme de Jean-Jacques,
l'anteriorite de l'expressif, de l'emotif sur l'eclaire et le calculelD. De
Man pousse tres loin les remarques proposees par Derrida, et voit dans
l'erreur sur l'objet, et surtout sur l'autre homme, l'essence de toute metaphore. Suivant egalement Derrida, il superpose etroitement le passage du
Second Discours traitant de la formation du concept (l'exemple des deux
ebenes) et la question du langage figure. La notion de metaphore conceptuelle ou de concept metaphorique est fondee sur cette superposition. Pour lui, le nom propre et le sens propre se recouvrent, tout
comme, de l'autre part, le nom commun et le sensfigure.
C'est Ia un parfait contresens, une faute de lecture permise par
l'emploi du tenne propre dans deux acceptions qu'on ne doit pas confondre. Dans l'une, la propriete est celle du nom, du representant pre9 Ce parallelisme sera presente par la suite comme une antecedence logique de
l'Essai sur le Discours. Dans son commentaire, Derrida accordait egalement une
grande importance a l'etroite relation entre les deux ouvrages, faisant de l'Essai une
longue note du Discours (v. Jacques Derrida, De la Grammatologie, Paris,
Minuit, 1967, 2eme partie, chap. III). Pourtant, fvidence ne semble pas acquise, et
les recherches recentes inscrivent !'Essai dans la constellation des travaux de
Rousseau sur la musique, notamment du Dictionnaire de musique.
10 V.A. Wyss, Jean-Jacques Rousseau. L'accent de l'ecriture, Neuchatel, La Baconniere, coll. Langages, 1988.


linguistique d'un individu hors code; dans l'autre, eile est celle du sens,
du signifie bu du represente designe confonnement au code. Symetriquement, l'emploi fait par Rousseau de l'expression denominations
communes (dans le Second Discours, p. 149) pour designer les tennes
generiques dans la creation du concept, est confondu avec cet autre
nom commun a eux et a lui (Essai) qui represente, dans le processus de
correction de la metaphore passionnelle, le mot propre (ibid.). 11 est
interessant de noter que le contresens se trouve deja chez Derrida, dans
un entrelacement de notions tel que l'effort de clarification poursuivi par
Rousseau se perd totalement: On ne peut substituer le nom commun
adequat (homme) au nom de geant qu'apres l'apaisement de la frayeur et
la reconnaissance de l'erreur. Avec ce travail s'accroissent le nombre et
l'extension des noms communs. Par la, l'Essai communique etroitement
avec le second Discours: les premiers substantifs n'ont pas ete des noms
communs mais des noms propres. Le propre absolu est bien a l'origine:
un signe par chose, un represente par passion 11.
11 n'en va pas ici de cheveux coupes en quatre. Si le propre absolu
de Derrida ne repond qu'a un goiit de la fonnule, et ne met pas en cause a
lui seul sa theorie du supplement, Paul de Man, lui, echafaude bel et bien
toute sa philosophie du langage sur la confusion du trope et du concept,
indument extraite de la lecture de Rousseau. Que cette lecture fonne un
contresens, on peut.le montrer d'une autre maniere, en tenant compte de
la chronologie supposee par Rousseau dans l'acquisition du langage.
Dans le Second Discours, Rousseau decrit l'acquisition de la capacite
cognitive et du langage categoriel:c'est le passage, dans la longue duree,
d'un stade prelinguistique au stade linguistique. Dans !'Essai, il prend
pour exemple un homme sauvage deja en possession du langage, deja
confronte a d' autres hommes, dans un etat qui n'est plus celui de la nature, et decrit l'acquisition individuelle d'une capacite de rationalisation
et d'adequation du discours au vrai12. L'element mis enjeu a chaque fois,
11 J. Derrida, Op. eil., p. 395. 11 semble que Derrida ne fasse pas le meme contresens
que de Man. C'est qu'il est soucieux de faire apparaitre partout le logocentrisme
de Rousseau, sa conception ontologique du signe. Pour lui, chez Rousseau, au
commencement etait le propre. Pour de Man, c'est le contraire: au commencement et
partout, tout est figure. Dans la longue note de son edition italienne (v. ci-dessus, n.
4 ), Paola Bora s'efforce de tirer parti de cet embrouillamini pour rendre justice a
12 Rousseau decrit deux geneses separees, non confondables, et qui n'impliquent pas
les memes points de depart. Meme si l'on pense, comme Paul de Man, que


c'est la comparaison, con~ue comme premiere maitrise sur la nature, par

l'observation multipliee, puis comme premiere maitrise sur la passion et
l'illusion. La comparaison, concept ambivalent (Rousseau en fait dependre notamment sa critique du regard social et de la vanite), met l'homme
de plain pied avec le langage et avec le monde social; eile l'eloigne aussi
bien du pur concret que de l'immediatete.

Felix culpa
La notion de metaphore conceptuelle ou de concept metaphorique
est essentielle chez de Man. Elle apparait dans chaque chapitre de Allegories of Reading, avec sa reference au Second Discours comme texte
fondateur. Elle est presente dans de nombreux autres essais, notamment
dans Epistemology of Metaphort3, ou toute l'analyse en depend, mais
ou eile n'est pas mieux fondee. Locke, Condillac et Kant sont censes
tomber dans la trappe qu'elle ouvre sous leurs pas. On renonce a suivre
ici l'analyse de de Man, pensant que l'examen critique pratique dans un
texte ou la notion de metaphore conceptuelle est donnee pour un fondement theorique, a ete suffisamment convaincant.
Cet examen aura montre en tout cas que, contrairement a ce que laissent croire les procedures affichees par de Man, la theorie chez lui precede la pratique. La conviction de de Man: tout enonce est une figure et
dit autre chose que sa visee litterale, cette conviction conduit
l'interpretation. De proche en proche, eile dissout le sujet des enonces
dans la mecanique rhetorique des tropes, et evacue la recherche de la
verite au profit des strategies de persuasion. S'il importe aujourd'hui de
refuser cette theorie et de remettre Rousseau sur ses pieds, il importe tout
autant de conserver l'oeuvre de Paul de Man dans la maison des lettres.
l'affabulation narrative est une projection dans la duree d'une situation con:flictuelle
vecue dans un present dechire, on ne doit pas supprimer les ples de tension. La tension essentielle separe ici un langage de type ana.lytique, qui manie des denominations generiques en abstraction des objets concrets, et un langage de type
symbolique, qui rend campte a la fois de l'objet et du rapport que le sujet entretient
avec lui. V. La litterature comme interpretation syrnbolique, in L'lnterpretation
des textes, sous la dir. de C. Reichler, Paris, Minuit, 1989.
13 The Epistemology of Metaphor, Critical lnquiry, 5, l, 1978; repris dans Aesthetic


Refuser la theorie pour garder une pratique, c'est renoncer a la pantextualite qui regne a l'avant-scene doctrinale, mais c'est rester attentif a
deux elements essentiels qui seuls peuvent preserver les etudes litteraires
de la regression qui les menace. Quoique de maniere contoumee et equivoque, Paul de Man reconnaft dans ses lectures une valeur cognitive specifique au langage litteraire. Cette valeur, il la nomme rhetoricite et lui
accorde la capacite de signaler sa propre erreur, sa propre illusion.
Heureuse erreur, par laquelle persiste la conscience d'un dispositif mental specifique, irreductible aux visees de mai't:rise fonctionnelle et techniciste. Ce dispositif n'est autre que celui que vise Rousseau dans son
exemple de l'attribution du nom geants. Il est indissociable d'un sujet et
d'une relation immediate entre ce sujet et le monde. Le langage peut dire
adequatement ce lien, meme s'il le dit de maniere symbolique. Cette pensee qui se dit par figure, Paul de Man la fait travailler, il reconnatt sa
force et son caractere inarretable, contre les conceptions positivistes de
l'inteipretation litteraire. De Man a voulu voir dans ce mouvement de
l'inteipretation un enchainement d'aberrations, a la suite de la critique
nietzscheenne de la philologie. Mais il n'y a pas d'erreurs sans une
chance de verite: la encore, heureuses erreurs, qui emportent dans leurs
scories les eclats de justes paroles.
Il y a dans cette persuasion de l'erreur native, si ambigue .chez de
Man, un fennent de remise en cause inepuisable. Il me semble que, abandonnant les monstres philosophiques a leur destin solitaire, c'est ce
message-f qu'il faut retenir. Si on sait l'entendre, on est amene d'abord a
concevoir l'histoire litteraire, et l'histoire intellectuelle en general, d'une
tout autre maniere que celle qui a prevalu jusqu'ici. Le schema chronologique des causes et de la consequence, de la source et du derive, se complique singulierement. Le texte litteraire constitue une reinteipretation
des materiaux au moyen desquels i1 a ete elabore, c'est-a-dire, pour
une part, de textes anterieurs, qui sont ainsi reemployes et eclaires de
maniere particuliere (et, al'occasion, aberrante). L'histoire de la litterature devient alors l'histoire des inteipretations successives par lesquels
les textes prennent sens. L'inscription d'une oeuvre dans le temps est effectuee par les oeuvres qui la suivent, et la dechiffrent ou la meconnaissent en la reactualisant. On est amene aussi aune sorte de deontologie de
l'inteipretation, qui reste relativiste (et c'est sa limite), mais s'ouvre sur
une ethique de la recherche. On ne cherche jamais seul: on cherche contre ou avec d'autres, dans une relation la fois fratemelle et polemique,
puisqu'on part de ce qui leur etait reste cache pour produire notre tour,


al'adresse d'autrui, un champ de questionnement. Cette relation, qui n'est

pas de progres cumulatif mais constitue une sorte de passage du temoin, dans une exigence de comprehension toujours renouvelee, fait de
la recherche en litterature un processus de deplacements. Elle decrit sans
doute de maniere essentielle le lien entre le maitre et l'eleve, par le fait
qu'elle interdit al'eleve de rester un epigone, et de se tenir assure d'un


Tue reflection on Rousseau that forms the centre of the second part of Allegories of Reading, contributes fundamental elements to de Man's work. One of them will be analyzed
here: the term of 'conceptual metaphor' which is proposed in the interpretation of the
'Second Discours.' According to de Man, every statement is supposed tobe a [...] conceptual, figurative, metaphorical metalanguage. We analyze this theory in the text from
which de Man claims to derive it: the passages on language in the Second Discours and
in the Essai sur l'origine des langues. De Man's interpretation of those passages seems
wrong to us: Rousseau never confuseci the formation of concept and that of metaphor,
which involve different mental and semiotic structures. But if it seems necessary to refute
de Man's theory of language, one has nevertheless to approve his literary practice in so
far as it clarifies two points: the recognition of the specifically cognitive value of
literature and the capacity of permanently calling into question any given interpretation.


Peter Grotzer


Sein, das verstanden werden kann, ist Sprache.

(Hans-Georg Gadamer, Wahrheit und Methode, S. 450)

Parler de Paul de Man et de Georges Poulet dans le contexte du Colloque

de Zurich, c'est evoquer tout d'abord leur voisinage pendant les annees
1964 a 1969, quand ils etaient collegues ala Faculte des Lettres de cette
Universite, mais c'est surtout reflechir sur leurs conceptions de 1a

litterature et de la critique litteraire d'alors, telles qu'on peut les decouvrir

dans l'essai Verite et methode dans l'oeuvre de Georges Poulet, publie
dans le numero de juillet 1969 de la revue Critique, texte repris sous
forme completee dans le volume Blindness and lnsightl.
Avant d'entrer dans le vif du debat, retenons le fait suivant: l'oeuvre
impressionnante de Georges Poulet2 est actuellement quelque peu
oubliee, alors que les mises en question de Paul de Man, aussi
suggestives que discutables, ont provoque une maree de reactions et
trouvent leurs admirateurs et leurs detracteurs. Grand joueur qu'il etait,
Paul de Man serait le premier a s'en moquer, car il ne croyait pas trop a

1 Paul de Man, Blindness and Insight. Essays in the Rhetoric of Conternporary

Criticism. New York, Oxford University Press, 197L L'essai sur Poulet y porte le
titre The Literary Seif as Origin: The Work of Georges Poulet; l'auteur y
remarque que la version parue dans la revue Critique etait slightly shortened. Pour
ma part, je pense que le texte anglais est posrerieur au texte fran~ais; je m'y re:fere
s'il donne des eclaircissements supplementaires. Sauf indication contraire, je suis le
texte fran~ais.
2 Pour une liste des publications jusqu'en 1980 voir Marcel Raymond - Georges
Poulet, Correspondance 1950-1977, Choix et presentations par Pierre Grotzer,
Avant-propos par Henri Gouhier, Paris, Jose Corti, 1981, pp. 327-345. On y
ajoutera: La Pensee indeterminee. L De la Renaissance au Romantisme. II. Du
Romantisme au XXieme siecle, Paris, Presses universitaires de France, 1985 et 1987
(Collection Ecriture). Le tome m est SOUS presse.


ce qu'il disait: c'est que presque rien, et lui-m~me le moins, n'echappait a

son sens de l'ironie.

Il y a vingt ans, quand l'essai en question fut ecrit, Poulet etait au
centre du debat sur l'ancienne et la nouvelle critique3. Le texte Verite et
methode dans l'oeuvre de Georges Poulet est donc tout d'abord un
hommage a un collegue celebre, mais aussi une sorte d'adieu dans
plusieurs sens du terme. Comme pretexte, la revue Critique mentionne
la parution de Mesure de l'instant4 et de Benjamin Constant par luimeme5. Cependant, de Man se refere surtout aux deux essais de Poulet
sur Proust (dans Etudes sur le temps humain I et dans M esure de
l'instant), et il s'interroge sur le rapport entre deux notions qui jouent un
rle capital dans la demarche de ce critique: le point de depart et le
Le texte se developpe entre l'eloge et la mise en question; i1 met en
relief l'interpenetration entre la matiere critique et la matiere qu'on
pourrait appeler proprement poetique dans le sens de la modemite
teile que l'avait deja postulee Friedrich Schlegel. D'une part, de Man distingue chez Poulet une reflexion [... ] ala fois remarquablement stable et
homogene, aimantee vers une totalisation constamment en voie de
s'accomplir, et d'autre part, cette oeuvre passe pour ~tre intensement
mobile, pleine de rebondissements, de retours en arriere, de nouveaux
departs (p. 609).
D'un cfe la methode, et avec eile une certaine assurance, la saisie
graduelle des mouvements de la conscience en vue d'une vision globale
du monde interieur d'une oeuvre, de l'autre une verite, et avec eile la
complexite, la densite, l'interrogation jlermanente.
Que de Man se concentre sur ce second aspect, appele la face
privee d'une oeuvre qualifiee par lui de litteraire, qu'il pose des
questions sur le rapport entre les deux notions-clef, n'etonnera personne.
En lisant 1a phrase suivante, l'on remarquera tout de suite quels sont les
deux ples qui permettront a son texte de se developper: La fermete de
3 Cf. Les Chemins actuels de la critique, Centre culturel international de Cerisy-laSalle, 2-12 septembre 1966, Sous la direction de Georges Poulet. Textes revus et
publies par les soins de Jean Ricardou, Paris, Pion, 1967. Paul de Man y presenta
Ludwig Binswanger et le probleme du moi poetique, Repris egalement dans
Blindness and /nsight.
4 Etudes sur le temps humain W, Paris, Pion, 1968.
5 Paris, Editions du Seuil, 1968 (Les Ecrivains de toujours). De Manne s'y rerere que
brievement dans la version anglaise, B lindness and /nsight, pp. 88sv.


la methode ne se comprend que par egard a la delicatesse avec laquelle

se trace ici la figure d'une verite fuyante et rebelle (p. 610).
Ce qui interesse, est donc une figure, terme qui pour de Man
jouera un rle important dans les ecrits posterieurs a l'annee 1969, a la
Suite de sa lecture de Fontanier, Genette, Derrida. Sa reflexion s'attaque
aux principes de la methode, et il attire l'attention du lecteur sur leur
incompatibilite: ces principes impliquent une tension qui semble tre le
gage de la litterarite de la demarche critique. A cela s'ajoute wie tension
decouverte a l'interieur des differentes prises de conscience, con~ues
dans la plupart des cas comme la constatation lucide d'un manque d'etre,
d'une Situation de faiblesse OU meme de nullite.
Le procede employe a de quoi surprendre: au lieu de s'en tenir au
langage qui exprime cette experience premiere dans les etudes critiques
de Poulet, l'essai de de Man nous renvoie d'abord a un texte de 1924 Oll
l'auteur s'occupe des problemes du recit dans une perspective bergsonienne6. En ce temps, Georges Poulet venait de consacrer sa these de
doctorat a l'oeuvre de Balzac, et il etait sur le point d'ecrire le roman La
Poule aux oeufs d'or, qui a paru en 1927 chez Emile-Paul freres (sous le
pseudonyme Georges Thialet).
Ce renvoi a un texte qui precede l'oeuvre critique proprement dite de
Georges Poulet permet d'evoquer une attitude in statu nascendi qu'on
retrouvera dans bien des essais critiques Oll la rigueur methodique de
Poulet ira de pair avec sa conviction profonde qu'un oubli de soi-meme
est une condition indispensable pour qui veut se faire porter par l'oeuvre,
aussi bien dans l'elaboration d'un roman dans un cadre etabli que dans
l'activite critique.
L'optimisme exprime dans le texte sur le Bergsonisme, nous apprend
de Man, sera de courte duree, et il se retere avec raison a une citation
bien caracteristique de Nicole qui figure dans l'introduction au premier
tome des Etudes sur le temps humain: Nous sommes comme des
oiseaux qui sont en l'air mais qui n'y peuvent demeurer sans mouvement
ni presque en meme lieu, parce que leur appui n'est pas solide, et que
d'ailleurs ils n'ont pas assez de force et de vigueur pour resister a ce qui
les porte en bas.
6 A propos du 'Bergsonisme'. (Trente Ans de Pensee /rQnfaise, vol. ill, Le
Bergsonisme, par Albert Thibaudet.) Selection, Bruxelles, n 6, avril 1924, pp. 6575. Signe Georges Thialet. Pour les premiers essais et articles de Georges Poulet,
presque toujours signes Georges Thialet ou G.T., on consultera notre bibliographie
(cf. la note 2).


Dans un texte que de Man ne pouvait pas encore connaftre en

1968/69, Poulet evoquera la naissance de la conscience critique sur laquelle repose son oeuvre. Tres sensible ala presence vivante des textes qui nourrissent sa pensee en fonnation, il remarque: Retrouver cette
notation parfaitement fluide, qui, quel que ft l'objet quotidien qu'elle se
donnait, s'exprimait toujours avec la meme uniformite, tels devinrent le
but invariable de mes lectures et l'operation par laquelle je reduisais
toutes celles-ci a une substance homogene unique et comme liquide.
Derriere les formes, derriere les structures, derriere le flot continu des
mots ne demeurait plus qu'une seule chose: une pensee sans forme, toujours differente d'elle-meme dans ses manifestations successives mais
toujours inalterablement fidele a elle-meme dans son fond.7 Dans la
suite, Poulet montre comment il a reussi a se debarrasser du defile chaotique des pensees d'autrui qui pendant longtemps finissaient par
l'accabler et qui le privaient de la possibilite de proceder aune demarche
hermeneutique proprement dite. 11 s'agit de la decouverte qu'ecrire, c'est
se constituer le sujet de ces penseess. A partir de cette decouverte
(J'assiste aux phenomenes qui se deroulent en moi ), il con~oit toute
parole organisee comme naissant d'une prise de conscience initiale pour
tendre vers les points subsequents que tour atour eile allait toucher9.
Mais revenons au texte de de Man que nous nous sommes propose
de lire. Il circonscrit avec beaucoup de lucidite le moment premier de
cette experience: Le cogito semble corncider, chez Poulet, avec l'eveil
de ce sentiment de fragilite fonciere, qui n'est autre que l'experience
subjective du temps (p. 614).
11 n'est peut-etre pas inutile de se rappeler que le propos initial de
Georges Poulet n'avait pas ete de decrire la conscience du temps, mais
l'etemiteto. Pour ce lecteur, il y a un parallelisme etonnant entre l'experience du Janseniste et la situation initiale de l'homme dans bien des
textes litteraires, qu'il s'agisse d'un rapport de similarite ou d'un rapport
7 Conscience de soi et conscience d'autrui chez le critique, Allocution de Monsieur
le Professeur Dr. h.c. Georges Poulet, Nice, Fascicule Montaigne Preis 1970 de la
Stiftung F.V.S. zu HambUFg, 1970. Le texte est repris dans G.P., La Conscience
critique, Paris, Corti, 1971. La citation se trouve a la page 303. Dans ses essais
recueillis dans les trois volumes de La Pensee indeterminee (op. cit.), Poulet s'est
mis a la recherche de cette pensee sans forme.
8 La Conscience critique, op. cit., p. 305.
9 La Conscience critique, p. 306.
10 Conversation avec l'auteur de cet essai.


d'opposition 11. Le fait est qu'a partir d'une teile experience fondamentale
(par exemple l'eveil proustien dans une chambre d'htel oti la prise de
conscience pascalienne de la misere humaine), toute demarche intellectuelle est une tentative de trouver mieux, c'est-a-dire d'inventer, de creer,
d'atteindre une duree qui, dans le domaine spirituel, il est vrai, ne peut
etre obtenue sans la grce. D'ou la necessite de contrebalancer la fragilite
de cette situation de depart par une construction mentale et litteraire de
l'oeuvre qui se presente sous fonne de parcours lineaire, ce qui a l' air de
gener quelque peu Paul de Man, car il remarque: L'assurance methodologique provient de la possibilite d'etablir et de justifier ainsi un
scheme narratif qui pennet de parcourir la totalite d'une oeuvre dans
l'espace d'un recit critique coherent, partant d'une Situation initiale et
s'achemmant, par une suite de peripeties et de decouvertes, vers une
conclusion satisfaisante parce que prefiguree (p. 615).
Ce qui semble etre un eloge est au fond une mise en question
radicale: un point ne peut ftgurer en meme temps comme point de depart
et comme centre. Un principe structurel et organisateur fait intervenir
une anteriorite et une prospectivite, ce qui nous interdit de parler
d'un point de depart. Ce qui devalorise considerablement ce recit
critique coherent, c'est que la conclusion est censee etre. prefiguree;
voila ce qui risque d'enlever a ces recits Tun des traits significatifs de
tout texte litteraire: le caractere polysemique qui se prete a l'analyse
rhetorique et hermeneutique.
L'experience de la discontinuite, bien connue a Georges Poulet
depuis sa lecture des Jansenistes, de Pascal et des Refonnateurs12, il la
retrouve entre autres dans La Recherche du temps perdu, et cela lui
permettra de mieux voir ce qui separe Proust et le roman moderne de la
conception bergsonienne de la duree13. (Cela explique aussi pourquoi
Poulet se voyait dans l'impossibilite d'entrer dans l'oeuvre de Charles
11 fai esquisse le rapport etroit entre la vision du monde de Beguin. Raymond et Poulet
et leur activite critique dans Literary Criticism and Religious Consciousness: Marginalia on Albert Beguin, Georges Poulet, and 'Marcel Raymond. Art/LiteratureiReligion: Life on the Borders, Ed. by Robert Detweiler, Journal of the American Academy of Religion Studies, XLIX, 2, Scholars Press, Chico, Califomia, 1983.
12 Cf. aussi l'essai Continuire et Discontinuite, La Revue Reactionnaire, 15
november 1933, pp. 285-291.
13 Cf. a ce propos Bergson et le theme de la vision panoramique des mourants,
Revue de Theologie et de Philosophie, t. X., 1960, pp. 23-41. Repris comme
appendice a G.P., L'Espace proustien, Paris, Gallimard, 1963, SOUS le titre Bergson,
le theme de la vision panoramique des mourants et de lajuxtaposition.


Peguy qui, selon Beguin, marque la fin du Jansenismet4.) Ce qui est

produit par la juxtaposition des moments d'extase dans la Recherche, ne
cree dans la conscience pas une duree, mais le sentiment de
l'intemporalite du sujet.
Comment faut-il lire le passage suivant dans lequel de Man
circonscrit la methode de Poulet: Le point de depart, d'abord angoisse
parce qu'il se sentait separe d'un passe qui ne le soutenait plus, se libere
maintenant du faux poids qu'il croyait peser sur lui et devient le moment
createur par excellence. 11 est devenu 1a source de l'imagination poetique
proustienne ainsi que le centre du recit par lequel le critique nous fait
participer al'aventure de cette invention (p. 616).
Ceci est dit a propos de la lecture de Proust presentee par Poulet,
mais en meme temps - toutes proportions gardees -, ces deux phrases
me semblent decrire l'experience critique teile que la con~oit de Man; la
version anglaise est plus explicite acet egard, car apres la phrase Poulet
makes us share in the adventure of bis creation, nous lisons dans le
texte anglais: This critical narrative turns around the central affinnation: 'time recovered is time transcended'. Ne devrions-nous pas dire
dans ce contexte: Poulet recovered is Poulet transcended?
La lecture parallele du texte fran~ais et du texte anglais est par
endroits assez interessante. Prenons par exemple le passage suivant:
They [the two essays of Poulet on
Proust] do not set up scales of value or
make normative statements about an
assumed superiority of future over past.
Their interest stems instead from the
movement generated by their dialectical interplay. (Blindness and Insight, p.

[ ... ] c'est que leur signification ne reside

pas dans des conclusions qui etabliraient des valeurs ou des certitudes,
mais dans le mouvement, alternatif ou
pivotant, des polarites qu'elles contiennent (p. 617, c'est moi qui souligne).

Les lecteurs de de Man voient tout de suite que c'est ce mouvement, ce

dialectical interplay qui le fascine en premier lieu. Cet interet lui
pennet de mettre en relief l'originalite et l'efficacite de la demarche de
Poulet qui est moins de l'ordre de l'essence que de celle du devenir: ce
14 Cf. Albert Beguin, Peguy ou la fin du janserusme, Une Semaine dans le Monde,
22 mars 1947. Ce n'est pas seulement l'idee de la continuite entre le monde chamel
et le monde spirituel qui gene Poulet, c'est surtout l'Incamation, la conjonction
paradoxale de deux mondes dont l'un transcende radicalement l'autre. La notion
d'incamation lui est aussi suspecte quand eile est appliquee au texte litteraire.


qui importe, ce sont les moments de passage. Ainsi l'eternelle aventure

litteraire qui semblait ~tre menacee, est sauvee: le recit et la prefiguration repondent a cette preoccupation du devenir.
Des qu'on con~oit l'activite critique sous l'aspect de l'identification
du critique avec la conscience creatrice qui se verbalise dans l'oeuvre, on
aurait avantage a la rapprocher de l'experience decrite par Karl Jaspers,
qui con~oit la communication existentielle profonde comme une
transcendance du moi vers ce qu'il appelle das Selbst. Il n'y a pas de
doute que pour le lecteur, c'est l'oeuvre qui est toujours une source, une
origine. Paul de Man l'a tres bien senti, et il prend au serieux la qufe
intersubjective de Poulet Vecue dans l'intensite d'un authentique besoin
spirituel (p. 619)~ n va meme plus loin, car la verite 'du fait litteraire, la
vraie subjectivite se montre selon lui dans le mouvement, la tension, et
celle-ci, comme la conscience, n'est accessible que dans la generation
permanente de significations par et dans le langage. Ce que Poulet
appelle la conscience, prend donc chez de Man le nom de langage,
et il se demande pourquoi Poulet en parle si peu de fa~on explicite.

Poulet est-il un critique de l'ecriture comme le pretendent Gerard Genette

et Paul de Man? Commen~ons par exclure la premiere explication que de
Man donne pour cette mefiance a l'egard du langage: eile tiendrait a des
raisons tactiques et pour eviter des malentendus et a la volonte de
Poulet de separer sa methode d'autres methodes qui mettent egalement
un accent determinant sur le langage (p. 622).
Dans l'alinea suivant - c'est le dernier de son essai -, de. Man nous
renvoie ades causes qui ramenent asa problematique profonde: Il est
clair que, pour lui, le langage n'importe que dans la mesure ou il ouvre
l'acces au moi profond de la subjectivite pure, par opposition au moi
quotidien eparpille dans les choses (p. 622).
Arretons-nous quelques instants sur cette phrase. Il est question de la
Subjectivite pure, et dans un hommage a Poulet, qui date de 1982, de
Man ajoute: what counts for him is the experience of an unmediated
presence of the mind to itself, in a barely articulated or actualized act of
inner participation. Que faut-il penser de la remarque suivante que nous
lisons un peu plus tard: Poulet belongs to the highest tradition of
literary criticism in which the criticism as its name literally indicates is
primarily the criticism, the devalorization, and destruction of literature

itself1 5 Pour les inities, cette phrase n'a rien de choquant; il existe une
categorie de lecteurs et d'interpretes qui reussissent assez bien adetruire
la litterature - remarquez bien que je me garde de dire a la de-construire.
Mais que faut-il entendre dans ce contexte par litterature? Je crois que
sur ce point, les positions de Poulet et de de Man sont diametralement
Dans un sens, on peut regretter que dans ses essais, Poulet n'analyse
pas des textes litteraires particuliers, qu'il n'insiste ni sur la fa~on dont la
juxtaposition de mots, de phrases, de chapitres engendre des significations, qu'il ne parle ni de l'esthetique d'une oeuvre d'art ni de sa rhetorique. n explique d'ailleurs longuement son attitude a l'egard de l'oeuvre
d'art dans une lettre aMarcel Raymond qui date du 9 decembre 1960. Il
s'y retere a la phrase de son ami qui avait declare: L'oeuvre d'art a sa
structure propre, son espace et son temps a eile. Pour Poulet, l'oeuvre
d'art devient avant tout un lieu mental [qui] n'est rien d'autre que l'esprit
lui-meme (celui de l'auteur, du lecteur) se donnant le spectacle et l'usage
de ses propres pensees dirigees, comme Dieu en lui-meme deroule le
spectacle et realise l'idee interne de ses creationsI6. Et un peu plus tard,
il va jusqu'a dire: Admirer une oeuvre d'art comme objet d'art, c'est flechir le genou devant le veau d'or11. Detruire la litterature, ce serait pour
lui la considerer comme un objet, ce serait la priver de ce qui fait son essence, la conscience (ou la pensee) qui seule pennet au lecteur de s'y retrouver. Or pour eviter de detruire la litterature, il faut, selon Poulet, percer les mots, detruire la coquille du langage: A la limite des plus merveilleuses reussites verbales de la poesie, il n'y a plus de mots, plus de
chair, plus rien qu'une transparence qui, portee a son point de diaphaneite supreme, aboutit al'immaterialite, l'invisibilite totale du poeme,
comme il y a invisibilite du cristal le plus purts.
Comment reconcilier une approche de la litterature de ce genre avec
l'approche rhetorique? Dans un sens, il n'est pas faux de dire que Poulet
ne lit pas les textes en tant que textes, et ceux qui visent une grammaire
generale de la litterature ne manqueront pas de lui faire ce reproche, car

Hommage a Georges Poulet. Publie dans Modem Language Notes, Volume 97/5,
Decembre 1982, pp. 6-7.
16 Marcel Raymond- Georges Poulet, Correspondance 1950-1977, op. cit. (note 2), p.
17 /d;, p. 60.
18 ld., p. 64.


ils ne peuvent lui pardonner son mepris de l'analyse objective des

harmonies, des astuces, des trucs comme etant les vraies beautes 19.
Cette critique est tout a fait justifiee de leur point de vue. A Cerisy-laSalle, Poulet s'etait encore propose de se rapprocher des methodes plus
formelles, mais en fait, son dialogue avec Marcel Raymond, son amitie
avec Jean Rousset, ses conversations avec Paul de Man probablement,
l'ont pousse dans une autre direction: celle de l'indetennine qui precede
la pensee et 1a prise de conscience.
Oll est la verite de Poulet, Oll mene la voie de de Man en 1969?
Quanta celui-ci, la reponse a deja ete donnee dans les autres exposes de
ce colloque. Quant a la pretendue destruction de la litterature, un essai
relativement peu connu de Poulet donne une sorte de reponse au dernier
alinea de l'essai de de Man, sans que l'auteur s'y refere de fa~on explicite.
Le texte est intitule Lecture et interpretation du texte litteraire et a
paru dans un volume collectif edite par Edmond Barbotin sous le titre

Qu'est-ce qu'un texte? Elements pour une hermeneutique20.

Poulet y part d'une definition du texte qu'il a trouvee chez Albert
Beguin a propos de Du Bos: Un texte, c'est d'abord cette matiere
indissociable et complexe, ce tissu qui non seulement 'traduit' une pensee
ou un evenement interieur, mais qui desormais les contient et les retient
en lui-meme; [... ] i1 n'y a plus moyen de separer l'idee ou l'experience de
cette apparence concrete, de ce tissu de paroles, de cet ensemble
rythmique Oll elles sont definitivement incluses par une veritable
Pour Poulet, le texte, plutt que d'etre un objet materiel, est emotion,
il est pensee communicable. Et ceci signifie que pour le lecteur, le texte
litteraire en tant que langage se presente sous forme d'obstacle (nous
nous souvenons de la phrase de de Man qui disait que le critique
detruit le texte), et Poulet parle meme d'une effraction, car il veut
atteindre la realite subjective qui palpite et se refuse au fond du
texte22. Il est vrai que pour sa part, Poulet ne retrace pour ainsi dire jamais cette danse et ce combat dont la visee est l'approfondissement
dans le centre de la pensee d'autrui. Il n'en parle pas puisque l'objectif et
le subjectif sont pour lui separes par un fosse infranchissable, et par Ia il
19 ld., p. 60.
20 Paris, Librairie Corti, 1975.
21 Art. cit., p. 64. Le texte de Beguin se trouve dans A.B., Creation et Destinee, Paris,
Le Seuil,, La Bacoruriere, 1973, p. 220.
22 Qu'est-ce qu'un texte? op. cit., p. 67.


met en question toute la methode stylistique teile que l'a pratiquee par
exemple Uo Spitzer, car un texte ne doit pas etre seulement analyse et
examine de l'exterieur; il doit etre encore ressenti et repense dans un
mouvement mental ou se confondent la pensee creatrice et la pensee lectrice23. Dans cette optique, il ne reste d'autre solution que de concevoir
le texte comme une realite avant tout subjective. C'est que - toute proportion gardee - le rapport entre le je lecteur et le texte est rapproche de
la relation intersubjective ou existentielle entre des hommes qui essaient
de se comprendre. Une fois de plus, on pense aux philosophes du dialogue24.
Certes on ne peut pas dire qu'un texte soit du meme ordre qu'une personne, mais rien ne nous interdit de concevoir le rapport lecteur-texte
sous forme d'analogie avec le rapport JE-TU. Ni le texte ni le TU ne se
revelent jamais dans leur totalite: de 1a notre raison d'etre en tant que _critiques litteraires. Si de Man s'en tient plutt au developpement syntagmatique des figures, Poulet, tout en ayant conscience de l'importance du
contexte, se degage souvent de la texture continue et pratique le procede de la recurrence des motifs et des themes25. Cette methode paradigmatique etablit un rapport intertextuel, comparable a ce qui se passe
dans la Recherche, et fait apparaitre une verite supra-temporelle26
Ce que de Man a developpe a propos de la tension entre le point de
depart et le centre en tant que principe organisateur, trouve une analogie
dans la tension entre la lecture syntagmatique et la lecture paradigmatique. A la lecture rectiligne et progressive en vue de la reconstruction de
la duree de l'oeuvre s'oppose chez Poulet la juxtaposition de certains
moments paradigmatiques de l'oeuvre, ce qui finit par creer un type de
temporalite base sur la reconnaissance de la similitude des differents
Dans ce cas, ce n'est plus l'ordre seriel qui compte, mais l'emanation
d'une pensee coherente. Voif pourquoi Poulet peut parler de centre,
du cogito du texte: il s'agit d'une activite spirituelle centrale de qui le
texte, et tout ce qu'il contient, pensees, sentiments, actions, dependent de

23 ld., p. 69.
24 Cf. ace propos Hans-Jrg Braun/Peter Grotzer, Wege zur existentiellen Kommunikation: Karl Jaspers, Gabriel Marcel, Martin Buher, Philosophisches Seminar, Seminar fr Vergleichende Literaturwissenschaft der Universitt Zrich, 1989, 46p.
25 Qu'est-ce qu'un texte? op. cit., p. 74. Poulet se rerere aWagner et a Proust.
26 ld., p. 75.


la fa~on la plus etroite27. L'un des buts de notre lecture de l'oeuvre de

de Man n'est-il pas de voir par quelle voie, par quelle methode on peut
arriver a quelque chose qui en serait la verite? Si l'on pouvait arriver a
trouver la zone Oll S'incame cette presence d'un Sujet a lui-meme et a
ses objets28, nous aurions la preuve que l'oeuvre de de Man est aussi de
l'ordre de la litterature, telle que le con~it Poulet. Ou s'agit-il plutt
d'une interrogation philosophique? La difficulte decrite dans le texte
The Resistance to Theory29 provient peut-etre de cette alternative.
Dans un texte a caractere scientifique, les Objets sont au centre de
l'interet (et si je dis objets, je pense aussi aux figures du langage vers
lesquels de Man s'est toume apres son essai sur la methode et la verite de
Georges Poulet, essai dont le titre suggere un autre adieu, celui a HansGeorg Gadamer). Reconnaissons avec les deux auteurs que la difference
entre un texte litteraire et un texte philosophique n'est que graduelle et
que l'experience de tous les grands textes est au fond la meme: Celle
d'un sujet qui se decouvre dans son activite propre, et cette activite
meme dans le mouvement par lequel elle s'exerce sur elle-meme et sur le
rrionde3o. Dans ce sens, Paul de Man n'a pas tort quand il dit dans son
essai que le sujet vulnerable et fragile du critique trouve un refuge
dans l'ecriture; on se demande si malgre sa mise en relief des
COntrarietes impliquees dans la methode de lecture de Georges Poulet,
Paul de Manne s'est pas, apres un grand detour, rapproche de fa~on
paradoxale du defenseur de la conscience critique quand il remarque
dans Resistance to Theory: To stress the by no means self-evident
necessity of reading implies at least two things. First of all, it implies that
literature is not a transparent message in which it can be taken for
granted that the distinction between the message and the means of
communication is clearly established. Second, and more probiematically,
it implies that the grammatical decoding of a text leaves a residue of
indetennination that has to be, but cannot be, resolved by grammatical
means, however, extensively conceived31. De tout cela il resulte la
necessite de continuer a lire et a s'interroger sur la portee de cette activite
qui est la seule fa~on de faire vivre les textes.
27 ld., p. 78sv ..
28 lbid.
29 Dans le volume posthume qui porte ce titre, presente par Wlad Godzich, Manchester
University Press, 1989 (fheory and History of Literature, Volume 33).
30 Qu'est-ce qu'un texte? op. cit., p. 81.
31 Paul de Man, Resistance to Theory, art. cit. (note 29), p. 15.



Paul de Man und Georges Poulet waren whrend ihrer gemeinsamen Lehrttigkeit an der
Universitt Zrich (1964-69) gut befreundet und trafen sich mehrmals in der Woche. De
Mans Aufsatz Verite et methode dans l'oeuvre de Georges Poulet (Critique, Juli 1969)
ist zugleich Hommage und Abschied, bzw. Infragestellung. De Man scheint die sichere
Methode der Bewusstseinskritik zu bewundern, preist jedoch gleichzeitig die stndige
(Selbst-)Befragung als Weg zum Verstndnis der Figur der fliehenden und Widerstand
leistenden Wahrheit literarischer Werke. Es war fr de Man die Zeit, da er sein Interesse
und seine Bewunderung von Gadamer auf Genette, besonders Derrida bertrug. Zu jener
Zeit przisierte Poulet dann seine Suche nach dem Cogito in verschiedenen Aufstzen, in
denen er die Ausgangsposition des kritischen Lesers als jene der Schwche und der Unsicherheit beschreibt und demzufolge seinen kritischen Nachvollzug der je entscheidenden Bewusstseinserfahrungen in einem literarischen Werk an gewissen Kategorien orientiert (vor allem Zeit und Raum). Dass der Punkt sowohl als Ausgangspunkt wie als
Zentrum aufgefasst wird, ist fr de Man, der vor allem Poulets Arbeiten ber Proust im
Auge hat, nur anzunehmen, wenn sie dialektisch in Bezug gesetzt werden. Fr de Man ist
die Literaturwissenschaft, wie er in einem spter publizierten kurzen Hommage a Georges Poulet bemerkt, the criticism, the devalorization, and destruction of literature
itself. Fr Poulet ist das Entscheidende das Bewusstsein hinter den Wrtern, fr de Man
ist es die syntagmatische Lektre der Figuren. In beiden Arten des Lesens bleibt eine
Zone der Ungewissheit bestehen: vielleicht ist die Wahrheit der Literatur gerade dort zu
erahnen. De Mans Gleichsetzung von Bewusstsein und Sprache bleibt im Zusammenhang mit Poulets Literaturverstndnis ein Missverstndnis, mglicherweise von de Man


Neil Hertz


Latein Allegories of Reading, towards the end of bis chapter on the Social Contract, Paul de Man writes, "We have moved closer and closer to
the 'definition' of text, the entity we are trying to circumscribe"l, and
many of bis readers must lean closer to the page: here it comes!
We can think of other such points in works of criticism where analytic questioning is, for a moment, made to feel like a quest-narrative, the
critic pausing to invite the reader to share a sense of mounting anticipation, of getting "closer and closer". There is Maurice Blanchot, on the
oi>ening page of L'Espace litteraire2, gesturing at what he takes tobe the
region towards which bis writing is moving, the pages on the gaze of
Orpheus, on Orpheus' trip to the Underworld, bis climb upward and then
his turning to look back at Eurydice. Those pages have been frequently
cited in recent years. Less familiar is an oddly apposite moment in
William Empson's writing, another descent into Hell towards another
woman. He is beginning his entry into the zone of maximum - that is,
Type VII - ambiguity and has been discussing what Freud called the antithetical sense of primal words, citing Freud on how the primitive
Egyptians use the same sign for "young" and "old"3. Now he catches
Paul de Man. Allegories of Reading: Figural Language in Rousseau, Nietzsche,
Rilke, and Proust, New Haven, Yale University Press, 1979, p. 268. Hereafter, references to this volume, designated as AR, will be given in the text. Other work:s of
de Man's will be similarly referred to. They are: Blindness and lnsight: Essays in the
Rhetoric of Contemporary Criticism, 2nd ed., Minneapolis, University of Minnesota
Press, 1983 (BI); The Rhetoric of Romanticism, New York, Columbia University
Press, 1984 (RR); and The Resistance to Theory, Minneapolis, University of Minnesota Press, 1986 (R1).
2 Maurice Blanchot, L'Espace litteraire, Paris, Gallimard, 1968 [1955], p. 5. Hereafter
references to this work, abbreviated as EL, will appear in the text.
3 Empson goes on to cite Freud's claiin that the Egyptians "only gradually learnt to
separate the two sides of the antithesis and think of the one without conscious comparison with the other'', then comrnents: "When a primitive Egyptian saw a baby he


himself up, mock-apologetically, grants that all this talk about the Egyptians is "in some degree otiose" and writes:
I have been searching the sources of the Nile less to explain English verse than
to cast upon the reader something of the awe and horror which were feit by
Dante arriving finally at the most centrique part of earth, of Satan, and of hell.
Quando noi fwnmo f, dove la coscia
Si volge appunto in sul grosso dell'anche,
Lo Duca con fatica e con angoscia
Volse la testa ov'egli avea le zanche4.

Empson leaves untranslated these lines from the last canto of Inferno:
"When we were where the thigh tums, just on the swelling of the
haunch, the Leader with labour and strain brought round his head where
his legs had been."5 Drawing out the analogy, Empson then adds cheerfully: "We too must now stand upon our heads, and are approaching the
secret places of the Muse" (7T, p. 196).
No one sounds less like Blanchot than Empson, but the two passages are
thematically comparable. lt is common, in end-of-the-line texts6 of this
sort, for movement towards one's goal tobe allegorized in terms of the
pull of desire, the fear of the law. So Blanchot can write, "Regarder
Eurydice, sans souci du chant, dans l'impatience et l'imprudence du desir
qui oublie la loi, c'est cela meme, l'inspiration" (EL, p. 231), and
Empson can, still more insouciantly, superimpose "the secret places of
the Muse" on the shaggy loins of the character Dante calls "the Emperor
at once thought of an old man, and he had to leam not to do this as his language became more civilised. This certainly shows the process of attaching a word to an object as something extraordinary; nobody would do it if his language did not make
him [ ...]" (7I, p. 194). The passage, for all its throwaway quality, is worth comparing with de Man's linking of the "advent of theory" in literary studies to the
"introduction of a linguistic terminology ... that considers reference as a function of
language and not necessarily as an intuition" (KJ', p. 8).
4 William Empson, Seven Types of Ambiguity, New York, New Directions, 1966
[1930], p. 196. Hereafter referred to, in text, as 71'.
5 Dante Alighieri, The Divine Comedy, /:Inferno, with translation and comment by
John D. Sinclair, New York, Oxford University Press, 1961 [1939], p. 425 [Canto
xxxiv, lines 76-79].
6 On the movement of a text towards "the end of the line" see Kenneth Burke, The
Philosophy of Literary Form: Studies in Symbolic Action, rev. ed., New York, Vintage Books, 1957 [1941], pp. 56-75, and my own discussion of such scenarios in the
"Afterword" to The End of the Line: Essays on Psychoanalysis and the Sublime,
New York, Columbia University Press, 1985, pp. 217-39.


of the dolorous realm". But what of that moment in Allegories of Reading: can a similar figuration be read there as weil?
At first glance, it would seem not: the pages we are concemed with
(AR, pp. 266-70) move back and forth between lengthy citations of the
Social Contract and careful inteipretive paraphrase. The argumentation
is compact, the diction a combination of Rousseau's own and de Man's
conceptualized linguistic vocabulary. What govems the movement of
these paragraphs is not a quest-narrative but the elaboration of an analogy between law and grammar with respect to the categories of the general and the particular. De Man begins with Rousseau's Statement that
"there can be no fundamental Law that is binding for the entire body of
the people" and glosses this as implying that "the meaning of the contractual text has to remain suspended and undecidable" (AR, p. 266). He
then notes Rousseau's insistence on the separation of the particular
members of the body of the state from the state as a whole, noting the
paradox that "to the extent that he is panicular, any individual is, as individual, alientated from a law that, on the other hand, exists only in relation to his individual being" (AR, p. 267). This is paraphrastic, dry and
demonstrative. But what de Man concludes from this passage of argument is put more intriguingly:
From the point of view of the legal text, it is this generality which ruthlessly
rejects any particularization, which allows for the possibility of its coming into
being. Within the textual model, particularization corresponds to reference,
since reference is the application of an undetermined, general potential for
meaning to a specific llllit. The indifference of the text with regard to its referential meaning is what allows the legal text to proliferate, exactly as the preordained, coded repetition of a specific gesture or set of gestures allows Helen to
weave the story of the war into the epic. (AR, p. 268)

I will take up the allusion to Helen shortly. For the moment I would note
that nothing in Rousseau - or in de Man's earlier argument - would seem
to require that the legal text here be characterized, anthropomotphically,
as "ruthless" or "indifferent". One effect of this diction, however, is to
link a discussion of the law with de Man's later consideration of excuses
in the final chapter of Allegories of Reading. There Rousseau's claim that
his lie about Marion was unintentional, an elfet machinal of his timidity,
leads de Man to note that the excuse is both fiction and machine, referentially detached implacable in its repetition of a preordained pattem,
comparable in this respect to Kleist's marionettes, "capable of taking on


any structure whatever, yet entirely ruthless in its inability to modify its
own structural design for nonstructural reasons" (AR, p. 294). That de
Man had the later discussion in mind while he was working on the Social
Contract is clear from another reference to Rousseau's lie in the pages
we're considering: "as we know from the Confessions", de Man writes,
"we never lie as much as when we want to do fll justice to ourselves,
especially in self-accusation" (AR, pp. 269;.270). In both chapters of Allegories of Reading and elsewhere - passim - de Man's language personifies the agency of the text, but personifies it as impersonal, often
fiercely so.
But to retum to Helen: where did she come from? what is she doing
dans cette galere? Like Rousseau's lying excuse, de Man associates her
with both fiction and machine. Years earlier, in "Criticism and Crisis",
he had written:
All literatures, including the literature of Greece, have always designated themselves as existing in the mode of fiction; in the Iliad, when we first encounter
Helen, it is as the emblem of the narrator weaving the actual war into the
tapestry of a fictional object. Her beauty prefigures the beauty of all future narratives as entities that point to their own fictional nature. (BI, p. 17)

There Helen seived as a figure in a mise en abtme, hovering somewhere

between standing in for the narrator and standing in for the aesthetic object. In the later text she figures less as an emblem of the narrator than as
the locus of a set of gestures, her own agency enabled by, but also submitted to the imperious control of the text. But I am flattening out what
is in fact a more interesting formulation in de Man: the analogy his sentence insists on is between "the legal text" and "Helen" - the ruthless indifference of the one rubs off on the other, and we are reminded that
Helen is at once the accomplished artist of the beautiful, the weaver passively rehearsing the weavers' code, and the woman whose beauty initiated the Trojan War and "brought death to so many brave men".
That last quotation is not from the Man, but from what I would speculate was de Man's source for this allusion to Helen, a remarkable paragraph in Empson which I shall cite, in the interests of thickening up the
texture, and raising the stakes, of this discussion.
We know that when de Man read Seven Types of Ambiguity in the
1950s he found its first and seventh chapters particularly compelling.
Appended to the first chapter is an "Annex on Dramatic Irony" illustrative of what Empson calls "a sort of dramatic ambiguity of judgement

which does not consider the characters so much as the audience" (7T, p.
43). In this case the characters are Portia and Bassanio in The M erchant
of Venice - Empson is writing of the scene of the three caskets - and his
point is that the song Portia sings while Bassanio is deciding which casket to choose ("Teil me where is fancy bred ... ") hints in its rhymes at the
proper choice without exactly suggesting that Portia is intentionally tipping him off. The scene allows us to entertain some doubt as to Portia's
honesty, but leaves the question unresolved. Empson goes on:
Irony in this subdued sense, as a generous scepticism which can believe at once
that people are and are not guilty, is a very normal and essential method; Portia's song is not more inconsistent than the sorrow of Helen that she has brought
death to so many brave men, and the pride with which she is first found making
tapestries of them; than the courage of Achilles, which none will question, 'in
his impregnable armour with his invulnerable skin undemeath it'; than the
sleepers at Gethsemane, who, St. Luke says, were sleeping for sorrow; than the
way Thesee (in Racine), by the use of a deity, at once kills and does not kill
Hippolyte. This'sort of contradiction is at once understood in literature, because
the process of understanding one's friends must always be riddled with such indecisions and the machinery of such hypocrisy; people, often, cannot have done
both of two things, but they must have been in some way prepared to have done
either; whichever they did, they will have still lingering in their minds the way
they would have preserved their self-respect if they had acted differently; they
are only tobe understood by bearing both possibilities in mind. (7T, p. 44)

I set down this paragraph in part as a contribution to the recent polemic

about Paul de Man's behavior during the war, about which I shall have
more to say later in this paper. One could wish that Empson's "generous
scepticism which can believe at once that people are and are not guiltf'
were as widespread as he generously believed it to be. But I also imagine
that, coming upon this soon after the war, de Man may very weil have
been struck - both entertained and gratified - by the matter-of-fact
shrewdness with which Empson could take up loaded questions of innocence and guilt. That those questions were on de Man's mind as weil,
and remained there as questions tobe taken seriously,nobody who has
read Allegories of Reading can doubt.
But my more immediate reason for citing Seven Types is to suggest
that the way iQ which Empson links the image of Helen weaving to the
thematics of indecision, machinery and guilt may account for the allusion's surfacing where it does in de Man's essay, in the middle of a consideration of machinery, guilt and indecision - what de Man calls undecidability and figures as suspension. Helen, he s~ys, is like the legal text,


and that text, as we noted before, "has to remain suspended and undecidable" (AR, p. 266). The paragraph that begins "We have moved closer
and closer to the 'definition' of text ... " concludes in this way:
But just as no text is conceivable without grammar, no grammar is conceivable
without the suspension of referential meaning. Just as no law can ever be written unless one suspends any consideration of applicability to a particular entity
including, of course, oneself, grammatical logic can function only if its referential consequences are disregarded.
On the other hand [the next paragraph begins], no law is a law unless it
also applies to particular individuals. lt cannot be left hanging in the air, in the
abstraction of its generality. (AR, p. 269)

That final image, of the law "left hanging in the air", coming as it does
just after the repetition "suspension", "suspends", could have been included in the essay called "Lurid Figures"7, where I commented on the
recurrence in de Man's writings of what I took as a kind of idiosyncratic
punning, in which certain tenns he relies on conceptually - words like
"suspension" and "disfiguration" - often turn up in proximity to images
of hanging or of physical defacement or mutilation, producing an odd
but characteristic pathos. I labelled it the "pathos of uncertain agency",
and, since I need to say more about that notion, and cannot assume a
reader's familiarity with the earlier essay, I shall briefly rehearse some of
its claims here.
I tried to show that lurid figuration in de Man was not willful or gratuitous but a necessary by-product of his theoretical concems, a function
of his attempt to dwell, speculatively, on what he called a "pennanent
disjunction" (RT, p. 92) in language, the "radical estrangement between
the meaning and the perfonnance of any text" (AR, p. 298). I cited what
was already a much-cited pronouncement of his - "No degree of knowledge can ever stop this madness, for it is the madness of words" (RR, p.
122) - setting it alongside some bizarre writing of his own that seemed
to me to qualify as sufficiently mad to illustrate his point. Tue idea that
the particular pathos in de Man's text was linked to a sense of uncertain
agency I derived at first from a reading of a paragraph in his essay on
Walter Benjamin's "The Task of the Translator", which I shall set down
again here:

7 In Lindsay Waters and Wlad Godzich (eds.), Reading de Man Reading, Minneapolis, University of Minnesota Press, 1989, pp. 82-104 (Hereafter referred to as RDR).


All these activities - critical philosophy, literary theory, history- resemble each
other in the fact that they do not resemble that from which they derive. But they
are all intralinguistic: they relate to what in the original belongs to language,
and not to meaning as an extralinguistic correlate susceptible of paraphrase or
imitation. They d.isarticulate, they undo the original, they reveal that the original
was always already d.isarticulated. They reveal that their failure, which seems to
be due to the f act that they are secondary in relation to the original, reveals an
essential failure, an essential disarticulation which was already there in the
original. They kill the original by discovering that the original was already
dead. (RT, p. 84)

What's noticeable here is that a familiar deconstructive turn - the claim

that the activity of disarticulation is discovered to have always already
taken place - is given an unfamiliar twist: the construal of disarticulation
as murder and murder as, paradoxically, "discovering that the original
was already dead". This, it seemed to me, adds, if only fleetingly, a
pathos of uncertain agency. The prose conjures up a subject - "perhaps a
killer, perhaps only the discoverer of the corpse - who can serve as alocus of vacillation: did I do it? or had it already been done?" and thus introduces another version of undecidability (or suspension), this time
between judgements of a subject's passivity or activity, innocence or
guilt (RDR, p. 86).
lt may be worth recalling that in the pages we are considering, pages
en routeto a definition of "text", the ruthless or indifferent activity of the
law - or the text - is set over against a counter-force, a form of covert or
subversive agency attributed, in the legal model, to the individual and in
the textual model to the referent. De Man notes that "the logic of grammar generates texts only in the absence of referential meaning, but every
text generates a referent that subverts the grammatical principle to which
it owed its constitution". Similarly, glossing Rousseau's argument that
"all citizens constantly desire the well-being of each [because] no one
exists who does not secretly appropriate the term each and think of himself when he votes for all", de Man characterizes this as an "act of deceit", a theft which "steal[s] from the text the very meaning to which [ ... ]
we are not entitled, the particular I which destroys its generality"; and it
is at this point that he links this discussion to the Confessions, to the motifs of guilt, lying and self-accusation (AR, p. 269). The basis for that link
should now be clearer: it is in tbe traces, in the language of this page of
argumentation, of the figure of an undecidably guilty subje~t and/or referent who can be imaged as facing off in specular fashion with a power-


ful but vulnerable (that is, flawed, disarticulated, already dead) law and/
or text.
This section of "Promises (Social Contract)" concludes with the
promised "definition" of text as an "entity" that must be considered from
a double perspective: from the perspective of grammar it is a "generative, open-ended, non-referential" system; from the perspective of figure
it is "closed off by a transcendental signification that subverts the grammatical code to which the text owes its existence" (AR, p. 270). We can
see that half the chapter's title refers to the grammatical system, to the
structure of promising, the other half to the particular "allegorical narrative of [the text's] impossibility", the paradoxes of contractual society
that de Man has teased out of Rousseau. The indirection of allegory was
necessary, de Man had argued, because "what remains hidden in the
everyday use of language, the fundamental incompatibility of grammar
and meaning, becomes explicit when the linguistic structures are stated,
as is the case here, in political terms" (AR, p. 269).
In "Lurid Figures" I tried to demonstrate that yet another narrative
could be read out of de Man's writings about textuality, one that had
been suggested to me by a biographical anecdote, the story of de Man's
mother's death. Because I had been told this story in confidence the
anecdote does not appear in "Lurid Figures"; however, one result of the
recent interest in Paul de Man's youth has been the publication of the
story, or a version of it, so I feel free to discuss it here. Tue story, as a
university friend of de Man's, Edouard Colinet, teils it, goes like this:
Paul's father, Robert (Bob) de Man, was a businessman, manufacturing and
selling medical instruments and x-ray equipment in Antwerp. Paul's elder
brother died in an accident and, after that, bis mother committed suicide: Paul
had the bad luck of being the first to find her hanged - he was about 15 years
old at the time. Paul's father was so disturbed by these two violent deaths that
for a time Paul had tobe taken care of by bis uncle Henri8.

I would stress the fact that this is an anecdote, one account of how
something may have taken place; there are other versions. According to
Paul de Man's cousin Jan - Henri's son - it was Robert who found his
wife's body, and it was not the case that Paul "had to be taken care of by
his uncle". That never happened, says Jan de Man: Paul was seventeen,
8 Edouard Colinet, "Paul de Man and the Cercle du Libre Examen", Responses: On
Paul de Man's Wartime Journalism, edited by Werner Hamacher, Neil Hertz, and
Thomas Keenan, Lincoln and London, University of Nebraska Press, 1989, p. 427.


not fifteen, at the time and didn't need taking care of9. Jan de Man has
the dates right, that much can be confinned: the elder brother was killed
June 20, 1936, when Paul was 16; his mother hung herself exactly a year
later. But as to who found her body, all that can be said is that the story
Paul de Man told Edouard Colinet and some other Belgian friends is the
same story he told, in later years, to various members ofhis family.
What he chose to teil provides the genn for what I find obliquely. inscribed at various points in his writings - a tableau of uncertain agency,
of someone confronting a suspended body, himself suspended between
feelings of matricidal guilt and of the intensified innocence of the bereft,
immobilized in the act of having "kill[ed] the original by discovering
that the original was already dead". Its appearance in de Man's texts may
be read :is another example of end-of-the-line figuration, to be contrasted
with the passages from Blanchot and Empson with which we began. As
each of these theorists approaches an elusive center - named variously as
text (de Man), as the articulation of l'oeuvre and l'ombre (Blanchot), or
as "the most complicated and deeply-rooted notion of the human mind"
(Empson) - they find themselves spinning gendered allegories of law
and desire. In this passage of Blanchot's the object of desire and the law
that forbids its realization seem distinct: Orpheus, in the impatience and
imprudence of his desire to look at Eurydice, must forget the law, we are
told (EL, p. 231)~ In Empson, things are less clear-cut: Satan and the
Muse don't occupy the same imaginary space, but rather are juxtaposed
metonymically, seemingly at the critic's whim. More remarkably, at the
equivalent point in de Man's text, the law - at once ruthless and vulnerable to subversion - is impossible to differentiate from the object of ambivalent desire: the law is like a hanging woman. The structure is not triangular and static but specular and unstable, and, precisely because of
that instability, it is available to figure the textual operations that de Man
will follow in one essay after anotherlO.
9 Letter from Jan H. de Man, June 27, 1988.
10 Compare de Man's account, in "Lyric and Modemity" (1969) of the non-filial (or, at
least, non-Oedipal) relation of Mallarme to his predecessor Baudelaire: ''The truly
allegorical, later Baudelaire of the Petits Poemes en Prose never stopped haunting
Mallarme, though he may have tried to exorcize his presence. Here was, in fact, the
example of a poetry that came close to being no longer representational but that remained for him entirely enigmatic. Tue darkness of this hidden center obscures later
allusions to Baudelaire, including the Tombeau poem devoted to the author of the
Fleurs du Mal. Far from being an older kinsman who sent him on his way, Baude-


I am proposing that Paul de Man's writing bears the traces of a particular, thoroughly contingent event in bis life, but the fonn those traces
take is bound to be overdetennined. The figure of the hanging woman
has a long history that pre-dates de Man's encounter with it in 1937; indeed, we can be sure that de Man's own relation to that figure must predate its actualization for him at the time of his mother's death. For the
figure recurs in fantasies of matricide and abjection that inform texts
from classical times to our own11. If we would speculate that de Man's
laire, or, at least, the most significant aspect of Baudelaire, was for hirn a dark zone
into which he could never penetrate" (BI, p. 184). In Mallarme's "Le Tombeau de
Charles Baudelaire" that "dark zone" is named "un immortel pubis". De Man is engaged in contesting the appropriateness of thinking in terms of genealogical lines of
descent, whether of critics of modern lyric (e.g., Otto Friedrich as "father" of HansRobert Jauss, who, in turn, "fathered" Karl-Heinz Stierle) or of poets (e.g., Baudelaire as "father" of Mallarme and "grandfather" of the surrealists). Here representational poetry is aligned with genealogy and a reassuring, if conflictual, pattem
of Oedipal lineage, the "trulyallegorical" with "a dark zone into which [one] could
never penetrate": a figure that blurs differences - of position and of gender, of figures of law and desire - much as does the passage in AR, or Empson's evocation of
Satan at the "most centrique part of earth" which is somehow also the site of "the
secret places of the Muse".
11 On the particular association of death by hanging - and of ~ther forms of suspension
- with women in classical texts, see Eva Cantarella, "Dangling Virgins: Myth, Ritual and the Place of Women in Ancient Greece" in Susan Rubin Suleiman (ed.) The
Female Body in Western Culture: Semiotic Perspectives, poetics today, 6(1985), pp.
91-101. Cantarella cites among other sources the work of Nicole Loraux: "Le corps
etrangle" in Actes de la Table Ronde: "Du chlltiment dans la cite. Supplices corporels et peines de mort dans le monde antique", Rome, Ecole Franyaise de Rome,
1984, pp. 195-214, and Fafons tragiques de tuer unefemme, Paris, Hachette, 1985.
Marc Redfield has pointed out to me an intriguing modern instance in Great Expectations, where the reiteration of the word "figure" loads a macabre scene with a further, rhetorical burden In Chapter 8 Pip wanders into an abandoned brewery on
Miss Havisham's property, sees Estella in the distance, then suffers this hallucinatory moment:
"lt was in this place, and at this moment, that a strange thing happened to my fancy.
I thought it a strange thing then, and I thought it a stranger thing long afterwards. I
tumed my eyes - a little dimmed by looking up at the frosty light - towards a great
wooden beam in a low nook of the building near me on my right hand, and I saw a
figure hanging there by the neck. A figure all in yellow white, with but one shoe to
the feet; and it hung so, that I could see that the faded trimmings of the dress were
like earthy paper, and that the face was Miss Havisham's, with a movement going
over the whole countenance as if she were trying to call to me. In the terror of seeing
the figure, and in the terror of being certain that it had not been there a moment before, I at first ran from it, and then ran towards it And my terror was greatest of all


witnessing an extemalized version of one such fantasy was traumatic, we

must also add that such a witnessing would necessarily place him in a
different relation to the contents of the fantasy, a relation we can know
nothing about but which need not have been merely disabling 12.
when I found no figure there." (Charles Dickens, Great Expectations, New York,
Harper Bros., 1961 [1861], p. 77.) Dickens will have Pip recall this moment twice
later in the novel, each time in connection with bis unsuccessful attempts to grasp
the secret of Estella's face: whom does she remind him of! That the answer turns out
tobe Magwitch, Estella's literal and Pip's figurative father, complicates the bearing
of the hallucination: a dimly discemed admonitory man's face is superimposed on
the vision of a hanging woman, the law blurred into the lineaments of the ghastly
bride of desire, and all under the sign of resemblance and figuration.
The writings of Thomas Hardy display the interrelation of suspension and disfiguration in similarly remarkable ways. The quasi-hallucinatory memory of a hanging
woman - a murderess whose execution Hardy bad witnessed as an adolescent - is
disseminated throughout his fiction., som.etllnes in elaborate scenes, som.etllnes in

brief descriptive figures, often in conjtmction with the motif of a mask or a closeclinging garment that conceals/reveals the outlines of a person's face or figure. The
surfacing of this material invariably signals a reflection on the relations of narrator
and character, .author and text. Eve Kosofsky Sedgwick drew my attention to the
fllest account of the biographical anecdote, in Robert Gittings, Young Thomas
Hardy, Boston and Toronto, Little Brown, 1975, pp. 32ff.
De Man has written of the appearance of the word "hangs" in Wordsworth's texts,
characterizing it as "the exemplary metaphor for metaphor, for figuration in general"
(RR, p. 89), and in "Lurid Figures" I discussed the ways that word gets caught up in
gendered scenarios of specular encounter (RDR, pp. 95-102). That discussion is am
plified in the last pages of this essay.
12 De Man's disputing the primacy of psychoanalytically based theories of motivation most explicitly in "Excuses (Confession)" (AR, pp. 278-301), but also in bis earlier
discussion of Yeats (RR, p. 231; cf. RDR, p. 89) - should make one cautious about
bringing terms like "fantasy" and "trauma" to bear on his writings. But his point was
never the irrelevance of psychoanalysis, merely (et encore!) the difficulty of its articulation with a theory of textuality like his own. Hence one finds him insisting, for
example, that "the moment of dispossession", the moment when a writer loses control of his text, "from the point of view of the subject [...] can only be experienced as
a dismemberment, a beheading or a castration" (AR, p. 296) and one would like to
hear him develop more fully the implications of that "can only": what necessity,
other than textual, is he granting to the f antasy of castration?
With respect to the.pervasiveness of fantasies of matricide, the pertinent psychoanalytic writings are those developing the insights of Melanie Klein, and developing
them with an alertness to the implication of subjectivity in language. I have found
suggestive formulations in Nicolas Abraham's work on introjection and incorporation and in Julia Kristeva's discussions of abjection. Here, for example, is Abraham
on the origins of the sense of guilt: ''Coupable sera donc celui qui n'a pas echappe a
la duplicite, qui se sert du langage. [...] n s'agira de considerer l'origine de la dupli-


If one is persuaded that there is this strand of figuration in de Man's

writings there still remains the question of what can be made of it. Can it
be shown to function in ways that invalidate the general bearings of de
Man's work or, altematively, can it serve to deconstruct his more abstract
theoretical arguments? Here the model would be de Man's remarks on
The Birth ofTragedy:
[T]he deconstruction does not occur between statements, as in a logical refutation or a dialectic, but happens instead between, on the one hand, metalinguistic
statements about the rhetorical nature of language and, on the other hand, a
rhetorical praxis that puts these statements into question. (AR, p. 98)

Are we dealing with such a rhetorical praxis here? De Man would say we
are not, and offer another name for what we are pursuing: he would call
it an obsession and he would have no trouble accommodating it to his
eire et du langage. Or nous savons que ce qui rend la duplicite ineluctable c'est la
rupture de la symbiose qui lie d'abord l'enfant a la mere. A l'oppose de toute autre
theorie, je tiens pour acquis que cette rupture n'est ni un fait d'abandon, de frustration, de scansion, ou de sevrage, mais le resultat nature! d'un processus d'introjection
c'est-a-dire d'interiorisation de la relation d'abord innocente a la mere. Le resultat de
cette introjection est le dedoublement du ple objet de la relation innocente en objet
interne et objet externe. C'est ainsi precisement qu'advient la duplicite et son acolyte
le langage. La culpabilite premiere se trouve ainsi inscrite dans l'etape la plus archai'que de la constitution du Moi. [... ] Le fantasme du 'crime' ne serait alors rien
d'autre qu'une rationalisation retrospective de la culpabilite inherente a l'acte meme
de l'introjection." ("Le 'crime' de l'introjection" in L'Ecorce et le noyau, Paris,
Aubier-Flammarion, 1978, p. 126). I would set this passage on the guilt that language necessarily imposes on the subject alongside the lines (discussed below) with
which the Man closes his essay on Rilke, lines in which a sense of primal duplicity
is projected onto an encountered face: "Masque? Non. Tu es plus plein,/ mensonge,
tu as des yeux sonores" (AR, p. 56). A passage from K.risteva on the constitution of
phobic objects describes the hallucinatory transformation of that face: "[O]n est en
droit de supposer que toute activite de verbalisation, qu'elle nomme ou non un objet
phobique ayant trait a l'oralite, est une tentative d'introjecter les incorporats. En ce
sens, la verbalisation est depuis toujours confrontee acet 'ab-ject' qu'est l'objet phobique. L'apprentissage du langage se fait comme une tentative de faire sien un 'objet'
oral qui se derobe, et dont l'hallucination forcement deformee nous menace du dehors." (Pouvoirs de l'horreur: Essai sur l'abjection, Paris, Seuil, 1980, p. 52). lt
should be stressed that neither Abraham nor K.risteva is seeking to describe unusual
or pathological states; on the contrary, they are offering hypothetical accounts of the
routine violence of the speaking subject's entry into language. Abraham's "crime",
Kristeva's menacing "ab-ject" are components of what, to borrow a phrase of Santayana's, might be called "normal madness", what de Man called "the madness of


understanding of how texts work. He knows about obsessions, bis own

and other people's, and can be wrily witty on the subject, as in these remarks about Michael Riffaterre:
lt would be all too facile to point to the psychological irnplications of Riffaterre's model, in which the mathematical as weil as the matemal irnplications
of the "matrix" are obvious, or of bis literary exarnples, with their obsessional
stress on death, on sarcophagi, on a not altogether simple sexuality, on hallucination and on obsession itself. H morbidity happens to be on.e's measure of theoretical audacity, Riffaterre is second to non.e. (RT, p. 40)

De Man's reader, picking up the allusion to the matemal, noting de Man's

own interest in "death... sarcophagi...obsession itself' might be tempted
to read this as veiled autobiography, but that would not dismay de Man.
He could acknowledge writing of this sort as telling - telling of bis own
obsessions - and still insist that obsession-in-general plays only a secondary and derivative role in the motivation of texts13. Obsessional con-

cems will find expression in what I have been calling lurid figures, and
the covert narratives of violence or eroticism these imply are, de Man
has argued, "defensive motion[s] of the understanding", ways of imposing intelligibility on otheiwise baffling operations of language.
That is the burden of the last pages of "Anthropomorpbism and
Trope in the Lyric", where de Man reads the relationsbip between
Baudelaire's sonnets "Correspondances" and "Obsession" as the
"construction and undoing of the mirror-like structure that is always involved in a reading" (RR, p. 252). After the symmetries between the two
have been established, de Man goes on to demonstrate the several ways
"Obsession" translates "Correspondances" into "psychological and
therefore intelligible" equivalents for elements in the more enigmatic
poem. Although he had begun bis reading by warning that to arrange the
two sonnets "into a valorized qualitative bierarchy" would be "more
convenient than it is legitimate" (AR, p. 254 ), he will nevertheless insist
that the more "perfectly and quickly" understood "Obsession" is the less
13 See the Foreword to the Revised, Second Edition of BI: "I am not given to retrospective self-examination and mercifully forget what I have written with the same
alacrity I forget bad movies - although, as with bad movies, certain 'scenes or
phrases retum at times to embarrass or haunt me a guilty conscience. When one
irnagines to have feit the exhilaration of renewal, one is certainly the last to know
whether such a change actually took place or whether one is just restating, in a
slightly different mode, earlier and unresolved obsessions." (BI, p. xii)


commanding text. Even if, contrary to fact, it had been written first, it
could not be taken as either the origin or the cause of "Correspondances". "'Correspondances' implies and explains 'Obsession' [he
goes on] but 'Obsession' leaves 'Correspondances' as thoroughly incomprehensible as it always was." What emerges from these pages is
that obsession (in Baudelaire's poem andin general) will always operate
as a "defensive motion of the understanding" - like figuration, it is a
mode of intelligibility. De Man concludes:
Whenever we encounter a text such as "Obsession" - that is, whenever we read
- there always is an infra-text, a hypogram like "Correspondances" undemeath.
(RR, 262)

If the specular figures we have been noting are thus obsessive, if, following de Man, we can not hope to find in them the source or origin of
the more difficult turns of his text, how then might they function? Perhaps as signs of a defensive wish for just such a source and origin, the
confirmation of self (purchased at whatever cost) that a traumatic recollection could provide? De Man would seem to imply something of the
sort in the last chapter of Allegories of Reading when he considers the
threat that a writer may lose control of bis text, a threat that contains
within it the possibility of "the radical annihilation of the metaphor of
selfhood and the will" (AR, p. 296):
This more than warrants the anxiety with which Rousseau acknowledges the
lethal quality of all writing. Writing always includes the moment of dispossession in favor 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.

But in the pages that follow an interesting complication is developed,

one I was unable to take account of in "Lurid Figures" but which bears
on our earlier discussion. The threat just described is not, it seems, the
most threatening of threats, for it remains tied, de Man argues, to the
metaphor of text as body. He will then go on to treat of the "more directly threatening alternative of the text as machine" (AR, p. 297). Recalling the moments in the Fourth Reverie that conjure up a dangerous
machine, he adds:


Tue threatening element in these incidents then becomes more apparent. Tue
text as body, with all its implications of substitutive tropes ultimately always
retraceable to metaphor, is displaced by the text as machine and, in the process,
it suffers the loss of the illusion of meaning. Tue deconstruction of the figural
dimension is a process that takes place independently of any desire; as such it is
not unconscious but mechanical, systematic in its performance but arbitrary in
its principle, like a grammar. This threatens the autobiographical subject not as
the loss of something that once was present and that it once possessed [that is,
the threat is not one of castration or beheading -NH], but as a radical estrangement between the meaning and the performance of any texl (AR, 298)

In "Lurid Figures" I had simply conflated these two accounts, and,

glossing what I took to be a "moment of madness", had cited both of

them as weil as a third, similar fonnulation of de Man's:
[F]iguration turns hallucinatory [I wrote] in an attempt to render intelligible
what, according to de Man, cannot be rendered intelligible, the "radical estrangement between the meaning and the performance of any text" [AR. p. 298].
Or again in the language of Allegories of Reading, it is the moment in which
"the writer severs himself from the intelligibility of his own text," one that "has
to be thematized as a sacrifice" [AR, p. 205-207], or that, "from the point of
view of the subject ... can only be experienced as a dismemberment, a beheading or a castration" [AR, p. 296]. (RDR, pp. 99-100)

But it now seems clear that only two of these citations - the last two describe the same moment, a moment that is figured as sacrifice or castration, and that these in turn are meant to be contrasted with a more
dangerous moment, that of the radical estrangement of meaning and performance. If that's so, how are we now to read the passage on the Social
Contract with which we began? With its allusions to the "fundamental
incompatibility of grammar and meaning", to the "impersonal, machinelike systematicity" of the Social Contract (AR, p. 268), it asks to be read
as a figuring of the text as a motiveless machine. Yet threaded through
these pages, I would argue, in the allusions to the social body and its
members, in the specular struggle between a quasi-personified ruthless
generality and a deceitful, thieving particular /,andin the images of the
hanging text or law, is just the sort of figuration de Man associates with
text-as-body. Could the more lurid figures - because lurid and because
figural - mark a defensive motion of understanding? would the simultaneous conjuring up of text-as-body shelter the subject from the more
threatening aspects of text-as-machine? That would be like saying one
finds the sonnet "Obsession" threaded through "Correspondances"; and


indeed that is one way of reading de Man's pages on those poems. Or it

would be like saying that the appearance of Helen in the text - her
beauty, her equivocally active and passive relation to the motions of
weaving - similarly conflates the object of desire and the law so as to
veil the threat to meaning in luridly attractive metaphoricity. lt may be
possible to proffer a theoretical distinction between these two metaphors
'for text, body and machine, but it may not be possible to write of the
text-as-machine without drawing on the idiom of text-as-body, that is,
without calling upon lurid figures.
lt may seem that the only interest of the appearance of these figures
in de Man's writings lies in the ways they might serve as clues to the unfolding of a drama of engagement and defense - engagement with problems of textuality and defense against the risks attendant on such
thought. But obsessions may serve as promoters of thought as weil, and I
think that can be demonstrated by looking at pages de Man devoted to
two thematically related poems, one by Rilke ("Quai du Rosaire" [AR,
pp. 4-43]) the other by Hugo ("Ecrit sur la vitre d'une fenetre flamande"
[RT, pp. 45-51]). Both concern the motif of a Flemish carillon, the
"Glockenspiel, das in den Himmeln hngt", in Rilke's plangent final line,
and de Man will refer to the Rilke poem, nine years later, in "Hypogram
and lnscription", where he offers a reading of the Hugo poem to give focus to his critique of Riffaterre's semiotics.
De Man turns to "Quai du Rosaire" in the course of a discussion of
figuration in Rilke, and concludes his brief account of it by locating the
poem's "true interest" not in its "thematic statements" but rather in "the
intricacy and wealth of movements triggered by the original chiasmus".
One can acknowledge this and still note that the thematic statements,
which de Man does not ignore, are loaded ones for him. Or rather, that
de Man's paraphrase and selective citation of the poem loads it with a
secondary, lurid interest that is neither dissimulated nor fully confronted.
The poem, written in Gennan, is a description of the city called Brugge
in Flemish, but by referring to the francophone Flemish poet Georges
Rodenbach, de Man can name the city in French, not once but three
times in the space of two pages, as "Bruges la morte". lt is as if the
sound of that feminine ending - la morte - were particularly compelling,
set as it is in an interpretation that dwells on "the seductive but funereal
image of a temporal annihilation which is enjoyed as if it were a sensuous pleasure, 'der Sssen Traube/ des Glockenspiels', which actually is


the death knell that reduced the city to a ghostly memory"14. De Man
will locate the poem's thematic appeal in its combining "the audacity of a
14 References in de Man's texts to authors other than the ones he is directly concemed
with are rare enough to prompt some curiosity when they occur. In this case there is
an "immediate warrant for the allusion: as de Man points out in a footnote (AR, p. 42,
n. 22), Rilke was familiar with Rodenbach's novel Bruges-la-morte. lndeed, turning
to the novel, one finds that much of its action takes place in a house on the Quai du
Rosaire. There are other resonances of Rodenbach in de Man's work, however, that
are not mediated by Rilke, but seem rather to have been effects of Rodenbach's importance as a cultural touchstone, especially for francophone Flemish readers. As a
participant in the literary revival associated with the review La Jeunl! Belgique in the
1880s, then as a Belgian Symbolj:St poet and novelist living in Paris and associated
with Mallarme, Rodenbach represented the possibility of high art that was at once
cosmopolitan in its modemity and rooted in the particularities of Flemish history and
geography. De Man's articles in Le Soir testify to his own youthful investment in
this ideal.
But there are still other ways in which Rodenbach's writings reverberate in de Man's
texts, particularly, but not exclusively, in his discussions of ''Quai du Rosaire" and
"Ecrit sur la vitre d'une fenetre flamande". Rodenbach's novels Bruges-la-morte
(1892) and Le Carillonneur (1897) play out fin-de-siecle dramas that link the sound
of Bruges's churchbells and a sense of a city poised between the powerful memory
of its medieval prestige and the possibility of its modern revival to narratives of
erotic madness and death. Tue later novel ends with its hero climbing the town's
main belltower to hang himself within the largest of the bells: "Et il entra dans la
cloche comme la flamme dans l'eteignoir" (le Carillonneur, Paris, Bibliotheque
Charpentier, 1913 [1897], p. 325). In Bruges-la-morte (Paris, Flammarion, 1904
[1892]) a widower settles in Bruges precisely because the city was, like the wife,
dead, and he was haunted by "un sentiment inne des analogies desirables" (p. 94).
He falls in love with a woman he takes for his dead wife's double, then, increasingly
tormented by the differences between them, ends by strangling her with the long
tress of his wife's hair he had saved as a relic: "Les deux femmes s'etaient identifiees
en une seule. Si ressemblantes dans la vie,. plus ressemblantes dans la mort, ... il ne
les distingua plus l'une de l'autre - unique visage de son amour" (pp. 271-72). A
moumer/murderer, haunted by what Rodenbach. names explicitly as "le demon de
l'Analogie" (p. 64). Tue lurid thematics and uncanny coincidences in each novel are
in the service of a very canny and controlled exploration of the nature of resemblance.
One last "resemblance'' is worth citing here, as further possible evidence of the hold
of Rodenbach's imagery on de Man's imagination. In "Lurid Figures" (RDR, p. 92), I
quoted de Man's reading, in "Shelley Disfigured", of the disappearance of the "shape
all light" in The Triumph of Life: "There is no doubt that, when we again meet the
shape (11. 425ff.) it is no longer gliding along the river but drowned, Ophelia-like,
below the surface of the water" (RR, p. 111 ), and I noted that there were no signs of
a drowning in Shelley's lines. Nor is there any allusion to Ophelia, and it is a puzzle
how she made her way into de Man's text. One possibility is that she had migrated
from Bruges-la-morte, where she figures in a hallucinatory moment, as the widower


paradox with a promise of beauty or even, in the image of the grapes, of

sensuous gratification on the far side of the grave", then, in a characteristic gesture, draw back from the seductions of his own paraphrase: "Yet
the true interest of the poem does not stem from these thematic Statements ... ". What is missing from de Man's critical account is any thematization of the reader-critic's own fascination. In the tenns he would later
adopt, de Man refuses to "give a face" to the suspended carillon15; hence
he can find no way of inscribing his own face in this text, as he will in
the later essay, with the help of Hugo's remarkable image of the mind as
"l'esprit, ce veilleur fait d'oreilles et d'yeux" (RT, p. 49).
But the need for just such an inscription is signalled further along in
the Rilke essay, in its concluding paragraphs. There, after commenting
on the way in which certain of Rilke's poems represent a "renunciation
of the euphonic seductions of language", a "denunciation of the ultimate
figure, the phonocentric Ear-god on which Rilke, from the start, has wagered the outcome of bis entire poetic career, as error and betrayal", de
Man cites, as a valediction, these lines from one of Rilke's French texts:
Masque? Non. Tu es plus plein,
mensonge, tu as des yeux sonores.

"At the moment of its fulfillment", de Man comments, "the figure announces itself by its real name", that is, as mensonge (AR, pp. 55-56).

walks by the canals of the dead city: "Dans l'atmosphere muette des eaux et des rues
inanimees, Hugues avait moins senti la souffrance de son coeur, il avait pense plus
doucement ala morte. Il l'avait mieux revue, mieux entendue, retrouvant au fil des
canaux son visage d'Ophelie en allee, ecoutant sa voix dans la chanson grele et
lointaine des carillons." (p. 19) - A linked series - la morte, visage, Ophelie, carillons - is available here for further linking to motifs of hanging, suspension, face,
figure, figuration, when the topic at issue is, as it is at key points in de Man's work,
the power and limitations of metaphorical structures.
15 De Man's insistence on the importance of the trope of prosopopeia is first formulated
in a summary fashion in "Autobiography as De-Facement": "Our topic deals with
the giving and taking away of faces, with face and deface, figure, figuration and disfiguration" (RR, p. 76). On de Man's use of the term "face", more particularly on his
reading of Wordsworth, see Catherine Caruth, "Past Recognition: Narrative Origins
in Wordsworth and Freud", Modern La.nguage Notes, 100, 1985, pp. 935-48, and
Cynthia Chase, "Giving a Face to a Name'', in Decomposing Figures: Rhetorical
Readings in the Romantic Tradition, Baltimore and London, Tue Johns Hopkins
University Press, 1986, pp. 82-112.


If we had any doubts that here, at the end of the Rilke essay, de Man
is belatedly and surreptitiously addressing the suspended carillon, they
disappear when, in "Hypogram and Inscription", he writes, of Hugo's
poem, that the "carillon's relation to time has tobe like the relationship
of the mind to the senses: it is the sonorous face,. the 'masque aux yeux
sonores' (Rilke) which, by metonymic substitution, links the sound of the
bells to the face of the clock" (RT, p. 48). Precisely by dwelling on his
own version of the figure Wordsworth calls the "speaking face" (RR, p.
89), by untying the knot named "masque aux yeux sonores", a lmot by
which, in the language of "Shelley Disfigured", "knowledge, oblivion
and desire hang suspended" (RR, p. 107), de Man has been able to press
his understanding of figuration past where it was in the Rilke essay to the
explicit and rewarding discussions of reading-as-prosopopeia that characterize his work after Allegories of Reading. In "Hypogram and Inscription" the gain is registered in two ways: in the critic's embedded
signature - the mystery guest signing in as "a bizarre waking monster",
"l'esprit, ce veilleur fait d'oreilles et d'yeux" - and in the lucidity of the
discussion that accompanies this signature, in which de Man articulates
the link between the hallucinatory aspect of prosopopeia and the arbitrary or "catachretic" imposition of meaning (RT, pp. 48-49)16.

With this in mind, I want to return to the question of de Man's wartime
writings and their relation to his later work, a question that takes on a
somewhat different look (or ring) when it is posed against the background I have been sketching in. And I want to link it to another historical question, that of de Man's interest in the work of William Empson.
More particularly, I shall offer a speculative account of what de Man
found in Seven Types of Ambiguity when he set out to introduce Empson's work to the French reading public in 1956.
16 Appropriately, the critic's signature is inscribed in what de Man calls a "seduction
scene", a scene in which questions of cognition are played out "in the erotic mode of
'mere' sense perception" (RT, p. 49). Tue scene prompts de Man to append an uncharacteristically La Rochefoucauldian footnote: "Rather than being a heightened
version of sense experience, the erotic is a figure that mak:es such experience possible. We do not see what we love but we love in the hope of confirming the illusion
that we are indeed seeing anything at all" (RT, p. 53, n. 23).


De Man's discussion of Seven Types concentrates on Empson's remarkable seventh chapter, for it is there that he finds both the boldest
theoretical formulations and the most intriguing illustrative material. We
have already noticed Empson's way of figuring his path through his
book: coming to the Seventh Type, that is, coming to terms with radical
ambiguity is like reaching Satan, like Dante's reversing direction at the
center of the earth, like approaching "the secret places of the Muse". As
the chapter goes on the implications of this playful language are developed into a darker thematics of incest, sacrifice, hanging and gender confusion. For example, glossing the last lines of Keat's "Ode on Melancholy" - "His soul shall taste the sadness of her might/ And be among
her cloudy trophies hung" - Empson writes:
Her trophies (death-pale are they all) are cloudy because vague and faint with
the intensity and puzzling character of this fusion, or because already dead, or
because, though preserved in verse, irrevocable. They are hung because sailors
on escaping shipwreck hung up votive gifts in gratitude (Horace. III,i.), or because, so far from having escaped, in the swoon of this achievement he has lost
life, independence, and even distinction fi:om her. (7I, p. 217)

This swoon into indistinction, into a hanging which is also a fusion with
"Veiled Melancholy", resonates with the lurid figuration I have been
tracing in de Man's writing. Moreover, in these pages, the swoon is made
to feel like a mode of sacrifice, as, further along, Melancholy is replaced
by the figure of Christ depicted, in Empson's account, as a "monstruous
hermaphrodite deity". The phrase occurs in the course of a discussion of
an epigram of Crashew's improvising on the Biblical verse "Blessed are
the paps that thou hast sucked" (Luke 11 :27). Crashaw had written:
Hee'l have his Teat e're long (a bloody one)
Tue Mother then must suck the Son,

and Empson comments:

Tue [...] couplet is 'primitive' enough; a wide variety of sexual perversions can
be included in the notion of sucking a long bloody teat which is also a deep
wound. The sacrificial idea is aligned with incest, the infantile pleasures, and
cannibalism; we contemplate the god with a sort of savage chuckle; he is made
to flower, a monstruous hermaphrodite deity, in the glare of the short-circuiting
of the human order. (7I', p. 221)


What Empson is tracing is what he calls "something weird and lurid" in

the seventeenth-century mystics" "apprehension of the sacrificial system", which he will characterize as "a true sense of the strangeness of
the 'mind's world" (7T, p. 222). Tue drift of the chapter is towards a
reading of George Herbert's "Tue Sacrifice", in which the Crucifixion is
held up as the privileged manifestation of a Type VII ambiguity. Empson
moves through the poem slowly, glossing various stanzas, stressing the
"fusion of the love of Christ and the vindictive terrors of the sacrificial
idea" (7T, p. 228) until he reaches what he calls "the final contradiction":
Lo here I hang, charged with a world of sin
Tue greater world of the two ...

as the complete Christ; scapegoat and tragic hero; loved because hated; hated
because godlike; freeing from torture because tortured; torturing bis torturers
because all-merciful; source of all strength to men because by accepting he exaggerates their weakness; and, because outcast, creating the possibility of society. (TI, p. 232-33)

Left out of this litany of paradoxes is an odd equivocation conceming the

gender of the voice that speaks in Herbert's poem, one Empson bad
noted earlier in bis commentary. Tue refrain with which each stanza ends
- "Was ever grief like mine?" - and whose repetition contributes more
than any other verbal device to what Empson calls the poem's "strange
monotony of accent" is a quotation from the Old Testament that, as
Empson remarks, "refers in the original not to the Saviour but to the
wicked city of Jerusalem, abandoned by God, and in the hands of her enemies for her sins" (7T, p. 227). Empson's pronouns convey the shift of
gender, though he does not comment on it; bis remark is made very
much in passing. But it is worth our dwelling on for a moment. lt is one
thing for Jesus on the Cross to knowingly echo the Psalmist - "My God,
my God, why hast thou forsaken me?" (Matt, 27:46; Ps 22:1) - and another, somewhat more puzzling thing for Herbert (or the liturgical tradition he was drawing on) to have the hanging Christ ventriloquize the
lament not f or but of a sinful and abandoned woman 17. Here are the
17 Rosemund Tuve would contest this statement. In "On Herbert's 'Sacrifice"' (Kenyon
Review, 12, Winter 1950, pp. 51-75) she had taken Empson to task for ignoring the
poetic and liturgical traditions that informed both the figuration and the tone of Herbert's poem. Empson had replied briefly in "George Herbert and Miss Tuve"
(Kenyon Review, 12, Autumn 1950, pp. 735-38), and Tuve had gone on to extend
and further document her disagreements with Empson in the opening section ("'The


verses from Lamentations (1:8-9, 12, 16), first the words of the poet, traditionally taken tobe Jeremiah, describing Jerusalem, then the plaint of
Jerusalem herself:

Jerusalem hath grievously sinned; therefore she is removed: all that honoured her despise her, because they have seen her nakedness; she sigheth, and
tumeth backwards.
Her filthiness is in her skirts; she remembereth not her last end; therefore
she came down wonderfully: she had no comforter. 0 Lord, behold my affliction: for the enemy hath magnified himself.

Is it nothing to you, all ye that pass by? behold, and see if there be any sorrow like unto my sorrow, which is done unto me, wherewith the Lord hath afflicted mein the day of his fierce anger.
For these things I weep; mine eye, mine eye runneth down with water, because the comforter that should relieve my soul is far from me: my children are
desolate because the enemy prevaileth.

With bis usual offhandedness, Empson doesn't quote the passage, merely
identifying it as "a quotation from Jeremiah" (7T, p. 227), which may
weil have led some readers to the wrong book of the Bible. But Paul de
Sacrifice' and modern criticism") of A Reading of George Herbert, London, Faber
and Faber, 1952, pp. 19-99. There, in the fuller version, she addresses, among many
other passages, Empson's reading of "Was ever grief like mine?". Adducing a medieval lyrical geme, the Monologue or complaint of Christ, she writes: "One group
of these monologues is formed by the 0 vos omnes qui transitis poems. Christ's first
words, as in Herbert, will be some variant of these phrases from Lam. i.12, where, of
course, they are said by the city of Jerusalem. They are twice repeated, as if by
Christ, in the Good Friday and Holy Saturday responsories, and it is liturgical convention, not Herbert, which makes 'any grief like unto my grief apply to Christ's
sorrow. [...] Both in liturgy andin lyric the conventionalizing of a situation, for one
thing, fixes certain words and ideas inextricably in a particular frame of reference as the 'anendite, et videte Si est dolor similis sicut dolor meus' become inescapably
Christ's own words, for Herbert or for Handel, so that we must be chary of interpretations based on a lively sense of their being rather said by the city of Jerusalem 'in
the original'. [... ] Herbert's 'original' was not a verse in Lamentations, but a wellknown and effortlessly accepted tradition which made a double reference to both
Old and New Testament, with all the resulting implications, absolutely inescapable."
(p. 34) - Although her stress is on how "inextricably" or "inescapably" certain
meanings are fixed within "a particular frame of reference", Tuve seems to be
granting that "double reference ... with all the resulting implications" is also
"inescapable"; the question seems to come down to whether one's sense of a particular implication was "lively" or subdued. My claim in the following pages is that
there are signs de Man responded to Empson's ..lively" reading of was ever grief
like mine?" when he encountered it in 1956.


Man, in 1956, seems to have taken the trouble to track down the allusion
to Lamentations, to give the appropriate reference and to quote one of
the verses, which then figures at the conceptual center of his praise of
Seven Types. Empson's "less serene mind", he writes, was not content
with LA. Richards' "reassuring notion of art as the reconciliation of opposites", for he understood, de Man goes on, that
the text does not resolve the conflict, it 'fllll1ll!S it And there is no doubt as to the
nature of the conflict. Empson has alreacly prepared us by saying that it is "at
once an indecision and a structure, like the symbol of the Cross", and ends his
book on George Herbert's extraordinary poem entitled ''The Sacrifice", a
monologue uttered by Christ upon the cross, whose refrain is drawn from the
"Laments of Jeremiah" (1,12).

And de Man copies out the Biblical verse ("ls it nothing to you, all ye
that pass by, etc.") before going on to assimilate Empson's theory of poetic ambiguity to a Hegelian account of the Crucifixion:
This conflict can only be resolved by the supreme sacrifice: there is no stronger
way of stating the impossibility of an incarnate and happy truth. Tue ambiguity
poetry speaks of is the fundamental one that prevails between the world of the
spirit and the world of sentient substance: to ground itself, the spirit must turn
itself into sentient substance, but the latter is knowable only in its dissolution
into non-being. Tue spirit cannot coincide with its object and this separation is
infinitely sorrowful. (BI, p. 237)

The propriety of de Man's translating Empson into the idiom of Hegel's

Unhappy Consciousness has been questioned recentlyls, but more interesting than this debate, from my point of view, is the fact that the language de Man chooses to give voice to by introducing the Biblical verse
into his text is not to be found in either Empson's chapter or Herbert's
poem. The voice de Man summons up, in an act of prosopopoeia, is that
of an allegorical figure in Lamentations, an affiicted woman called
"Jerusalem", and it echoes that of another afflicted woman the central
character in a poem de Man reviewed in Le Soir in 1942, a poem about
the ravages of war in Belgium which takes as its epigraph these verses
from the Gospel of Matthew:
18 See Terry Eagleton, ''The Critic as Clown", in Cary Nelson and Lawrence Grossberg (eds.) Marxism and the Interpretation ofCulture, Urbana and Chicago, University of Illinois Press, 1988, pp. 619-31, and Christopher Norris "Some versions of
rhetoric: Empson and de Man", in The Conjlict of Faculties: Phi.losophy and Theory
after Deconstruction, London and New York, Methuen, 1985, pp. 70-96.


Then was fulfilled that which was spoken by Jeremy the prophet, saying,
In Rama was there a voice heard, lamentation, and weeping, and great
mourning, Rachel weeping for her children ... (Matt, 2:17-18; allusion to Jer,

The poem, ''The Massacre of the Innocents", was by Hubert Dubois, a

Belgian Catholic poet de Man had written of admiringly before19. lt is, in
the words of de Man's review, "a meditation on the guilt that has led
humanity to the dreadful state in which it finds itself for the moment", a
meditation structured as a narrative of the abjection and redemption of
that figure of Rachel who, the poet claims, is to be found within all his
readers as they cry out, with him, against the horrors and injustices of
war. Because de Man's review is interested exclusively in the poem's
ethical themes and its fonnal mastery, a brief rehearsal of its plot is necessary, if the connection I would propose, between this piece of wartime
joumalism and de Man's later writings - on Empson and Herbert as weil
as, more generally, on the economy of sacrifice- is to make sense.
After an introductory stanza personifying the Rachel "within each
one of us" - representing her drunk with grief, cradling a dead child in
her anns and demanding justice of the Lord - her pitying and powerfully
accusatory voice is heard until it is interrupted, first by the poet, shocked
by the "impudence" of her address to God, then by a downpour of blood
and a chorus of voices from heaven, identified as the dead voices "of
children or of angels":
Assez crier, assez mentir! Allons, Mere, agenoux!
Assez fonder sur Dieu ton injuste courroux.
[ ... ]
Allons, Rachel, tu sais que si l'on prend tes fils,
Toi-meme les perdais, toi-meme en fis jadis
Perir [ ... ]
n pleut leur sang sur toi, coupable mere.

Tue voices go on to recount why Rachel bears this blood guilt, and the
accusation tumed against her reads like an indictment of heedless loose19 De Man's review of "Le Massacre des Innocents" appeared in Le Soir, September 1,
1942; it is reproduced on pp. 265-66 of Paul de Man, Wartime Journalism, 19391943, edited by Werner Hamacher, Neil Hertz, and Thomas Keenan, Lincoln and
London, University of Nebraska Press, 1988. His review of Dubois' earlier work,
"La poesie du bois dormant'', appeared on September 9, 1941; it is reproduced on
pp. 143-144 of the same volume.


living during the entre-deux-guerres period, the sort of indictment familiar to readers of right- and left-wing anti-plutocratic, anti-parliamentarian tracts of the 20s and 30s. Rachel has turned men away from virile
and arduous pursuits towards "les biens caressants [... ] les biens donnes,/Les biens charnels". She is responsible, moreover, for their losing
the desire to transmit life, to propagate sons - and here the poem takes
on a tone of pro-life fervor familiar to us these days but not always present in pre-war denunciations of capitalist decadence: Rachel is blamed
for denying her vocation as mother, and for counseling young brides not
to bear children, hence for murdering the unborn, the very children
whose voices are now denouncing her. Tue voices press Rachel to abandon her hypocritical self-pity and to acknowledge her implication in evil,
and this set of stanzas closes with a chilling celebration of war - "Temps
du meurtre, 6 temps pur!" - as an unmasking of the death that had been
dissimulated during a corrupt peacetime.
Now Rachel can be seen for what she is, not a victim but a drunken
Amazon, "plus sombre que la mort et comme eile, brulante", and
abruptly Rachel confronts this image of herself mirrored in a blood-red
fiery sky. This is the poem's turning-point, a moment of specular
apotheosis that is also Rachel's salvation: the instant she sees herself "la
mort re~oit sa face" and when death recognizes itself it is, says the poet,
by the grace of God no longer Death. Tue poem ends with Rachel transformed, singing of the promise of peace in an apocalyptic landscape still
bloodied with war: "il pleut toujours le sang sur mon temps dechire;/
Mais unjour i1 sera le ciel en verite"20.
20 Hubert Dubois (1903-1965) was to publish another poern during the Occupation
similar in its apocalyptic setting, in its contempt for pre-war decadence and in its
voicing of an acceptance of Belgium's defeat: Le Chant dans /es ruines, Bruxelles,
Editions de la Toison d'Or, 1944. I have been unable to discover whether or not
Dubois was prosecuted for collaboration, although reviews of his works in the 1950s
suggest that he may have been. One critic, reviewing his career, after referring to
"La poesie au bois dormant", writes of him: "Que cette prescience trop lucide se soit
ensuite egaree dans le temporel, SOUS la poussee d'evenements exterieurs, importe
peu sur le plan superieur ou nous entendons a present juger une oeuvre qui a finalement retrouve son vrai sens. [... ] C'est que nous avons depasse, par une etrange purification, le massacre et les ruines" (Andre Gascht, "Le danseur du sacre", Le
Thyrse, Bruxelles, 1953, pp. 496-97). The last sentence alludes discreetly to Dubois'
wartime publications, titles that are sometimes included in, sometimes omitted from
his post-war bibliographies. (My thanks to Ortwin de Graef, Tom Keenan and
Chantal Kesteloot for their help in locating copies of Dubois' poems and of writings


Like "The Sacrifice" (as Empson read it), this is a poem in which the
plaintive and unjustified voice of a fallen woman is redeemed. On the
evidence of his seeking out and citing the verses from Lamentations in
1956, I would guess that the memory of his involvement with Dubois'
text was stirred when de Man came upon Empson's discussion of Herbert. More interesting than this historical footnote, however, is the fact
that none of the various aspects of Dubois' poem that I have been drawing attention to - the figure of the victimized but also guilty mother who
is both abjected and recuperated, the ways in which sexual, familial and
socio-political elements are blended in the poem's plot and imagery, the
fable's turning on a scene of specular reversal - had, in 1942, found its
way into de Man's review of "Le Massacre des Innocents". Neither the
lurid setting nor the particular actions of Rachel and her accusers is
commented on; instead, de Man insists on reading the poem as a rigorous
movement of thought, a "logical reflection" aimed at bringing out
"principles and abstract tendencies". The fundamental element of
Dubois' poetics, de Man claims, is "the word with its rational content";
the miracle of the poem - and de Man is unstinting in his praise - is that
Dubois can make his thought come alive "without recourse to any of the
standard poetic artifices - allegory, symbol or metaphor", relying rather
on "that most difficult of fonns, the direct expression of his thought".
And that thought is, as de Man summarizes it, comprehensive and generous in its elevation, ethical in its content:
Complaint and lamentation cannot be justified, even in a situation as pitiable as
this. For all that is happening now is not the blind and pitiless action of destiny,
but the consequences of a fault, or an accumulation of moral faults, committed
down through the ages. Tue utility of such an ordeal is to oblige people to become conscious of this guilt, to make the masses see that they have acted badly.
Consequently, the more severe the punishrnent, the greater the hope of at least
witnessing the growth of those true values which should allow life to be lived
harmoniously, in the place of the false facilities which have led to bis catastrophe.

This is powerfully wishful writing, as any fonnulation of the "utility" of

sacrifice, the salutary or expiatory or compensatory value of suffering, is
bound to be. Equally wishful, I believe - and obscurely related to the
wish for sacrifice to serve some purpose - is de Man's odd insistence on
the rational, direct, non-figurative nature of Dubois' poetic language. In a
closed sacrificial economy, there would be no leftovers. no pains unac-

counted for, nothing that couldn't be subsumed under a governing aim; in


the closed economy of the text of reason, the sort of text de Man emphatically wants Dubois' to be, the "standard poetic artifices - allegory,
symbol or metaphor", with the possibilities for ambiguity or equivocation attendant on their use, must also be ruled out. And it is just these
questions - of sacrifice, of figurative language, of a lurid thematics - that
surface again when de Man reads Empson, and which are there taken up
in a significantly different way.
But there is something eise missing from de Man's discussion of
Dubois' poem, and missing, so far, from our own consideration of it: the
word "Jew". Like other readers of de Man's writings for Le Soir, when I
first came across the headline on the Chronique litteraire for September
l, 1942, "Le Massacre des Innocents", I read on to see if it represented
the sort of subversive gesture I had occasionally found elsewhere in de
Man's articles. I knew that the deportation of Belgian Jews had begun in
the summer of 1942 (I subsequently learned that raids resulting in mass
arrests had taken place in Antwerp in mid-August); it seemed possible
that de Man's choosing to review this particular poem at that particular
time was a surreptitious act of protest and solidarity, an encoded naming
of the crime he saw taking place around him. Other readers have entertained that possibility21, but I found this hoped-for interpretation didn't
survive my reading of the article or, later, of Dubois' poem. Both poem
and review, although each acknowledges the horrors of wartime Belgium, shift the responsibility for those horrors away from the Nazi occupiers - in Dubois' case, onto the pre-war sinfulness Rachel is made to
represent; in de Man's, onto a more diffusely characterized "accumulation of moral faults committed down through the ages" or "repeated
crimes against the human person". Neither Dubois nor de Man appears
to make any attempt, coded or not, to differentiate the degrees of suffering feit by various elements of the population, unless the simple title of
each text, "The Massacre of the Innocents", could be counted on tobe
read with such a message in mind. I'm not convinced it could and find
both poem and review disconcerting to read. If either is in any way about
21 Shoshana Felman, for example, in "Paul de Man's Silence" (Critical Jnquiry, 15,
Summer 1989, pp. 704-44), argues for the "latent resistance connotation'' of the
poem and of de Man's review of it {pp. 714-15). That seems mistaken. Other readings of Felman's, however, in particular her use of de Man's discussion of
Rousseau's Confessions to illuminate his own refusal to speak of his wartime activities {pp. 729-34), strike me as persuasive and valuable contributions to the current


the Belgian Jews, it would seem to testify to a blank disregard for what
was happening to them in 1942, or - if the concem was there but forced
to dissimulate itself - to what from our current perspective looks like
astonishingly bad timing22.
22 Yet another, more sinister interpretation needs tobe considered. There is a possibility that, in writing about Dubois' poem when he did and as he did, de Man was participating, intentionally or inadvertently, in a campaign to counter the sympathy and
pity non-Jewish Belgians were expressing for the Jews during the summer and fall
of 1942, as the brutality of the Nazis' treatment of them became harder to ignore. In
his detailed history of the ordeal of the Belgian Jews, Maxime Steinberg cites passages from the collaborationist press inveighing against "toute fausse pitie" and denouncing those Belgians who were moved to protect the Jews (cf. pp. 159-162 of
1942.'. Les cenJ jours de la deportation des Juifs de Belgique, Tome II of Maxime
Steinberg, L'Etoile et le fusil, Bruxelles, Vie Ouvriere, 1983-87, 4 vols.). Having
read these pages, I was curious to know where Steinberg would place de Man's review. I wrote him, enclosing copies of the review and of Dubois' poem, and received
the following reply, which I transcribe at length as a document to be added to those
appearing in Responses: "[ ...] Du point de vue de l'historien, la question de Manne
se pose pas dans les termes d'une alternative morale; il ne lit pas non plus l'article
retrospectivement. La question est de savoir si le ler septembre 1942, le lecteur du
"Soir" decouvre dans "Le Massacre des innocents" une allusion, quelle qu'elle soit,
la tragooie juive en train de [se] jouer en Belgique occupee. De Man, quant a lui,
n'est pas "innocent". 11 suit l'actualite de l'occupation, y compris sa persecution antisemite. Son article de 1941 sur "Les Juifs dans la linerature actuelle" n'appliquait
pas seulement une grille de lecture antisemite a l'histoire litteraire. L'antisemitisme
de de Man etait aussi politique: l'article persuadait le lecteur qu'il n'y avait pas lieu
de s'inquieter des mesures que preparait l'Occupant. La conclusion de 1941 est significative: de Man s'attache aY demontrer qu'une Solution du probleme juif qui viserait
a la creation d'une colonie juive isolee de l'Europe, n'entrainerait pas, pour la vie litteraire de l'Occident, de "consequences deplorables".
A tout le moins, l'auteur du "Massacre des innocents" n'etait pas dispose a
"deplorer" la prochaine deportation des Juifs. Publiant son article alors qu'elle venait
de commencer, il y a moins d'un mois, sur un rythme paroxystique (le convoi VII
part precisement le ler septembre), le critique lineraire ne la deplore pas. "La
metaphore" du poete s'adresse, selon de Man, "l'homme capable de sublimer la
souffrance qui tord joumellement l'humanite en guerre, capable de voir, malgre une
immense pitie, que cette douleur est salutaire parce qu'elle fait expier des crimes
repetes contre la personne humaine". Le propos est caracteristique de pessimisme
ideologique de la "droite revolutionnaire" ralliee, des 1940, a l'Ordre nouveau. Le
sentiment refletait alors l'air du temps dans un pays defait - dans tous les sens du
terme. En 1942, l'opinion est, du point de vue allemand, franchement "hostile" a
l'Occupant. Le sentiment public est dans l'attente d'un debarquement imminent et la
tentative avortee des Anglo-Canadiens du 19 aoiit a Dieppe n'entame pas
l'optimisme du public. Dans son pessimisme, de Man exprime le ler septembre un
sentiment qui n'est plus porte en dehors de la mouvance d'Ordre nouveau. En cet ete


1942, le Sentiment populaire se dresse contre l'Occupant et Se prete d'autant mieux a

denoncer la "guerre a l'outrance que les Boches livrent aux malheureux juifs". Un
temoignage, et non des moindres, donne la mesure du sentiment public: les deportations de l'ete, ecrit encore en 1942 le cardinal V an Roey dans sa correspondance
vaticane, "ont ete executees avec une brutalite et meme une cruaute qui ont revolte
profondement la population beige". Le pouvoir militaire d'occupation, inquiet de la
"sensation", s'emploie a ce que l'action antijuive en cours "eveille le moins possible
l'attention du public et ne suscite pas de sympathie pour les Juifs dans la population". La section de propagande, service politique place sous la tutelle des militaires,
ne commet pas l'erreur psychologique d'attirer encore plus l'attention en faisant donner la presse censuree contre le sentiment populaire. Les militants les plus frenetiques de l'antisemitisme se gaussent, en revanche, de ces "benets de gon touJours
prets as'emouvoir lorsqu'il s'agit d'un meteque au nez crochu": auxiliaires ardents de
la police SS, ils s'efforcent de premunir leur mouvance contre "l'enjuivement" qui
gagne l'opinion belge.
"Le Soir" n'appartient pas ala presse radicale d'Ordre nouveau. Le joumal "vole" est
reste, au temps de la censure allemande, dans sa fonction de "grand" quotidien de la
capitale belge. Le joumal ne participe pas, au moment de la deportation, aux timides

tentatives d'intoxication qui se manifeste dans la presse la plus militante. L'article de

de Man ne s'y inscrit pas. Le lecteur habituel du quotidien n'y trouve pas la moindre
allusion aux Juifs. La "metaphore" du poete est une reference bien trop lointaine,
meme pour le lecteur catholique de 1942. Au reste, le public ne con~oit pas encore
cette "barbarie nazie" qu'il denonce en termes de "massacre des innocents". Le mot
de genocide n'existe toujours pas parce que la chose reste inconcevahle. L'echo qui
en parvient de l'Est touche a peine des Cercles restreints, en cet ete 1942. A l'Ouest.
la "barbarie nazie" reste, aux yeux de ses contemporains, le scandale de l'arrestation
massive d'hommes et de femmes, d'enfants et de vieillards voues aetre deportes vers
l'Est mysterieux. Le public ne sou~onne nullement le "massacre" de ces "innocents" pendant "les remous" de l'ete 1942.

De Man lui, decouvre dans la "metaphore" qu'il apprecie, une occasion de philosopher sur "les remous de cette epoque" et d'y "lancer un coup d'oeil comprehensif et
genereux". Le critique 1.itteraire ne pratique pas - il le disait en 1941 - "l'antisemitisme vulgaire". Le sien est d'ordre intellectuel. 11 l'a dispense de porter le moindre
regard "comprehensif et genereux" sur la tragedie juive dont ses yeux sont temoins.
En 1942, au temps de la deportation, ce "regard" se porte dans la clandestinite.
En tant qu'historien de la solution finale en Belgique occupee, il ne me parait pas
possible d'aller au-def de cette lecture de l'antisemitisme du jeune de Man." (Letter
of 27 September 1988).
Steinberg's analysis is instructive for the care with which it weighs de Man's words and its own. Neither exculpatory nor prosecutorial in tone, it may be contrasted with
some other attempts at the historical contextualization of Paul de Man's articles. I'm
thinking in particular of "Fascist Commitments" (Responses, pp. 21-35), in which
John Brenkman, after announcing that he will adopt a "juridical stance and a prosecutorial attitude" because "the prosecutorial stance establishes the aggressiveness
required of such an inquiry" (p. 21), seeks to demonstrate de Man's complicity in
specific propaganda campaigns undertaken by the collaborationist press in 1941 and


Without seeking to soften this judgment of de Man's review of

Dubois, it is still possible to pursue the question of how that article may
have come to take the fonn it did. I would approach it by focussing on
what I noted had been excluded from de Man's discussion of the poem call it, emblematically, Rachel, in her three aspects: Rachel as Jew (that
is, as possibly giving voice to the plight of the Belgian Jews in 1942),
Rachel as coupable mere (that is, as embodying a complex of feelings
about mothers and sons, and the real and fantasmatic exchanges between
them), and Rachel as metaphor (as the central figure of Dubois' poem,
standing in turn for the incomprehending, pitying reaction to war, the
self-pity of its victims, the self-indulgence of the pre-war democracies,
Death, Hope, and so on). lt is very unlikely that de Man, in 1942, could
have read "Le Massacre des Innocents" without both experiencing and
bracing himself against what he would later call "the seductive power~ of
identification'' (AR, p. ix), in this case the pull of Rachel, the temptation
to a pitying and self-pitying, thoroughly ambivalent act of identification
with a victimized women. That drama of seduction and resistance would
be played out in silence: the connotative resonance of Dubois' figurative
language, along with its possible bearing on the Situation of the Jews,
'";'Ould go unrecognized and not appear in the text, suppressed along with
the figure of the guilty mother. Instead language would be found to stave
off that temptation, first by analyzing the poem as a "logical reflection",
then in the fonn of a final, idealizing tribute to Dubois:

1942 by showing how "circumstances gave even de Man's most empty phrases very
precise and concrete meanings" (p. 30).
Brenkman assumes that if, through archival research, he can establish the existence
of a set of such "circumstances" contemporaneous with a particular article of de
Man's, then he has effectively demonstrated de Man's ("fascist") engagement in that
aspect of the Occupation. Steinberg, too, is interested in placing de Man's articles in
their microhistorical context, but his sense of context is at once fuller and more nuanced than Brenkman's, and it allows him to discriminate differing intentions and
degrees of involvement. lt allows him, in this case, to state confidently that de Man's
review of Dubois was not written as a contribution to an anti-pity campaign - a
campaign that was indeed being waged, in one section of the collaborationist press,
in 1942 - and (of equal importance) that it would not have been read as such by its
contemporary audience. An investment in scrupulous analysis of this sort is finally a
more trustworthy attitude for the historian to adopt than the "juridical" or
"prosecutorial" stance Brenkman believes to be "required of such an inquiry".


Tue man capable of sublimating the suffering that daily wrenches humanity at
war, capable of seeing, despite an immense pity, that this pain is salutary because it expiates repeated crimes against the human person, exhibits that fundamental superiority of being that is proper to all true artistic talent.

That sentence may strike the ear now as both hyperbolic in its celebration of the artist's superiority and callous in its willed sublimation of suffering; but it is worth noting that it is also an act of fervid identification.
We can take it as a reminder of the difficulty of resisting one sort of
identification without falling into another. This is by no means Paul de
Man's problem alone, but it is very much a problem he tumed his mind
to thereafter. His post-war writing can be thought of as an extended
study of the stakes and the mechanisms of identification, which is another way of saying that the puzzling knot that binds uses of figurative
language to specular structures and to gestures of sacrifice would explicitly occupy his attention23.
One can see signs of this in de Man's writing immediately after the
~ar. Notions like that of the utility of sacrifice, the compensatory rewards of suffering - notions that govemed his reading of "Le Massacre
des Innocents" - have disappeared from articles like ''The Dead-End of
Formalist Criticism". There, Empson's choice of Herbert's poem as his
climactic example is read not as a transcendent reconciliation of the conflicts he had been exploring but as a demonstration of the impossibility
of reconciliation, hence of the delusory nature of those poetics de Man
labels "salvational". De Man's language has its own pathos but his interest in sacrifice is less in its pathos than in its structure and dynamics:
"sacrifice" is read as a figure of incommensurability, a thematic gesture
at a linguistic problem that won't go away. That's why the first sentence
of de Man's gloss on Herbert's poem - "This conflict can be resolved
only by the supreme sacrifice" - is followed immediately by one shifting
the emphasis away from any conceivable resolution: "there is no
stronger way of stating the impossibility of an incarnate and happy truth"
(BI, p. 237). If we were to ask why these two atheists, Empson and de
Man, should meet at the foot of the Cross, de Man would say that they
were led there not in the imitation of Christ but through their shared in23 On the appearance of the concepts (and words) renunciation, sacrifu:e, temptation,
seduction, and identifu:ation in de Man's writings, sec Minae Mizumura,
"Renunciation". in P. Brooks. S. Felman. J.H. Miller (eds.), The Lesson of Paul de
Man, Yale French Studies, 69 (1985), pp. 81-97.


terest in radical linguistic ambiguity, in the sort of contradiction that

could be thought of, in Empson's words, as "at once an indecision and a
structure, like the symbol of the Cross". An indecision and a structure that is, now taking up de Man's idiom, an impasse, or dead-end, or aporia, a moment of suspension or, in the lurid thematics I have been tracing, a specular encounter with a hanging figure.
A hanging "figure"? or a hanging woman? The figure on the Cross is
a man, isn't he? We have seen that in Herbert's poem the answer to that
is yes and no. The plangent refrain of the poem is, equivocally, that of
Christ and Jerusalem. And I have been suggesting that that equivocation
is a productive one, an element in the sort of ambiguity both Empson and
de Man are concemed with. A final glance at the divergence of Empson's
path and de Man's after their imaginary meeting in 1956 should bear this
Empson spent a great deal of wit and energy, most memorably in
Milton's God (1961), but throughout his post-war writings, inveighing
against what he called "neo-Christian lit. crit.", a complicity - sometimes he made it seem like a conspiracy - of theologians and close-readers to foster piety and subservience to a system of unnatural values centered on God the Father, whom Empson liked to call "the torture monster"24. His chief objection to this god was that he took "satisfaction" not just satisfaction in the acknowledgment of a redeemed debt, but the
sensual satisfaction of a sadist - in the crucifixion of his son, thus setting
an example of finding delight in human suffering that, Empson thought,
had perverted Western behavior for two millenia. This stance would
seem to have obliged him to reconsider his discussion of ''Tue Sacrifice'\ which he did, in 1950 and again in 1963, chiefly in response to
Rosemund Tuve's criticisms. Characteristically, he both does and doesn't
I put "The Sacrifice" last of the examples in my book, to stand for the most extreme kind of ambiguity, because it presents Jesus as at the same time forgiving
his torturers and condemning them to etemal torture. lt strikes me now that my
attitude was what I have come to call "neo-Christian"; happy to find such an
extravagant specimen, I slapped the author on the back and egged him on to be
even nastier [...] Rather to my surprise, Miss Tuve agreed that the poem carries
24 Empson's quarre! with Christianity is usefully summarized by John Haffenden in his
Introduction to a posthumously published collection of articles: William Empson,
Argufying: Essays on Literature and Culture, lowa City, University of Iowa Press,
1987, pp. 21-40.


the major ambiguity, which was traditional and noble; after rejecting most of
my illustrations, she seemed disposed to treat me as a pagan stumbling towards
the light Clearer now about what the light illumines, I am keen to stumble
away from it. (A, p. 257)

There .was a time when Empson could describe "what the light illumines", his primal scene of Christianity, in language that captured the
disorientation - of feeling and thought - that his encounter with radical
contradiction had provoked:
The sacrificial idea is aligned with incest, the infantile pleasures, and cannibalism; we contemplate the god with a sort of savage chuckle; he is made to
flower, a monstrous hermaphrodite deity, in the glare of the short-circuiting of
the human order. (71', p. 221)

By the 1950s, this flashing vision had been discarded, the "monstrous
hermaphrodite deity" replaced by "the torture monster", a more comprehensible figure of the Law as a sadistic father engaged in tormenting an
iqnocent son. Empson has stabilized the vacillation implicit in the earlier
scene - a vacillation of position and of gender - that had made the scene
adequate to the bewildering nature of the linguistic problem he was engaging. Of the "hermaphrodite deity" only a trace remains in the later
writings. lt occurs in an unpublished piece ("The Satisfaction of the Father" [c. 1972]), in the course of a summary of Thomas Aquinas' discussion of the Crucifixion: "Aquinas plainly knew", writes Empson, "that
the most intimate place of the religion was a horrible one" (A, p. 624), a
sentence in which we can hear a flickering acknowledgement of what it
was that had earlier led Empson to associate radical ambiguity not only
with sacrifice, but also with Satan and with "the secret places of the
I have quoted enough of Paul de Man's post-war writings to suggest
that his understanding of sacrifice took a different turn from Empson's, a
turn that precluded any stabilization of the concept. Consider, as a final
example, the last pages of "Wordsworth and the Victorians" (RR, pp. 8892) which contain, coincidentally, another endorsement of Empson's
work. This time de Man cites "Sense in the Prelude''25, the "one essay
[that] stands out from the fundamentally harmonious consensus that
25 Originally published in Kenyon Review, 13 (Spring, 1951), pp. 285-302, then as
Chapter 14 of The Structure ofComplex Words, London, Chatto and Windus, 1951,
pp. 289-305.


unites, for all apparent disagreement, all contemporary writers on

Wordsworth". Specifically what de Man admires is Empson's arguing
that if one traces Wordsworth's uses ofthe word sense one discovers that
"the emerging confusion cannot be reduced to any known model of trope
that would control an identifiable semantic field"; "it is impossible", de
Man concludes, "to make sense out of Wordsworth's sense". As a reader
capable of sophisticated rhetorical analysis, Empson is then compared to
Wordsworth himself who, in one of his Prefaces, had produced an analysis of the word hangs equally astute but with absolutely opposite results:
hangs, de Man notes, accurately, is "by Wordsworth's own avowal, the
exemplary metaphor for metaphor, for figuration in general", hence precisely for the possibility of making sense.
Having thus positioned hangs and sense at opposite poles, de Man
now introduces a key word of his own, face, a term that, as we have
seen, has been implicit in his thinking from the earliest written chapter of
Allegories of Reading through his latest texts on prosopopoeia:
Masque? Non. Tu es plus plein.
mensonge, tu as des yeux sonores. (AR, p. 56)

The force of de Man's reading of "face" in The Prelude is developed out

of his understanding of the tensions between the cognitive and performative aspects of language, tensions which are both stated and enigmatically enacted in the climactic sentence of his essay: "How are we to reconcile the meaning of face, with its promise of sense and of filial preservation, with its function as the relentless undoer of its own claims?" In
"Lurid Figures" I discussed the strangeness of this sentence, pointing out
the way its language draws at once on the lines of Wordsworth's de Man
was analyzing- the Blessed Babe passage- and on Yeat's rhetorical (or
maybe not so rhetorical) question "How can we know the dancer from
the dance?" (RDR, pp. 98-99).
But the sentence can also be juxtaposed instructively with some of
the language of Allegories of Reading I cited earlier, for example, with
de Man's characterization of "Quai du Rosaire" as combining "the audacity of paradox with a promise of beauty or even [... ] of sensuous
gratification on the far side of the grave" (AR, p. 43). The two promises
(of "sense and filial preservation", and of "sensuous gratification"
beyond the grave) resemble one another, but we can note a telling
difference in tone between the earlier lugubrious account of "Quai du


Rosaire"'s imagery as "seductive but funereal" and the more difficult

language and explicit violence with which the facing-off in Wordsworth
is described.
Or we might align this allegorizing of meaning-and-function as
mother-and-child with the pages on the Social Contract with which we
began: "There can be no text without grammar: the logic of grammar
generates texts only in the absence of referential meaning, but every text
[read: mother] generates a referent [read: child] that subverts the grammatical principle to which it owed its constitution [read: relentlessly undoes its own claims]."
In each of these passages - from "Wordsworth and the Victorians"
and from "Promises (Social Contract)", as in his later writing generallythe language of Demanian theory coexists in a state of high tension with
the obsessive figuration I have been following. Positioning himself so as
to feel the force of that tension - as if acknowledging that he could occupy no other position - was clearly productive for de Man. Its value to
him can be read obliquely out of the last paragraph of the Wordsworth
[O]ne can find, in Wordsworth's text. lexical continuities which are perfectly
coherent; despite the somewhat ominous overtones of the literal predicament it
invokes, the word "hangs" is a case in point. Other words, such as "sense" in
Empson's essay, lead instead to near-total chaos. Somewhere in between, at the
interface of these contradictory directions, words such as "face" can be said to
embody this very incompatibility. (RR, p. 92).

To embody an incompatibility: the phrase resonates with de Man's praise

ofEmpson, years earlier, for having so forcefully demonstrated that what
poetry was about was "the impossibility of an incamate and happy
truth". The difficulty and the interest of de Man's later work is in its idiosyncratic refinement of a theory and pathos of sacrifice: the "incamation" or "embodiment" his later texts invoke is not that of a Christian
god-in-man or of a Hegelian spirit-in-substance but of "incompatibility"
or "incommensurability" in words. The reception of de Man's work suggests that such an understanding of sacrifice - and the strange combination of control and disorientation it exacts from its subject - may be considerably harder to take in than the sustaining paradoxes of theology.



Dieser Aufsatz, Fortsetzung einer frheren Untersuchung zu den "Lurid Figures" in Paul
de Mans kritischen und theoretischen Schriften, geht aus von dessen Beschftigung mit
William Empsons Seven Types of Ambiguity in den 50er Jahren. De Mans Lektre von
Empsons Begriff des Opfers werden verschiedene andere Texte zur Seite gestellt - eine
biographische Anekdote, eine von de Man 1942 in Le Soir publizierte Rezension und
einige seiner spteren theoretischen Schriften - im Hinblick auf einen sonderbaren, aber
charakteristischen Aspekt von de Mans Prosa: seine eigenwilligen Wortspiele, in denen
zentrale Konzepte wie jene der Suspensionen und Disfigurationen eng verbunden sind
mit Bildern eines wrtlich-wirklichen Hngens und Entstellens. Der Autor stellt fest,
dass das Erscheinen dieser "grellen Figuren" in de Mans Text mit seinen produktivsten
Einsichten in die Mechanismen der literarischen Sprache einhergeht.


William Flesch


"Aesthetic Formalization. in Kleist" argues with some passion against

Schiller's concept of "the perfect aesthetic society", a society formed,
Schiller says, like a "weil executed English dance, composed of many
complicated figures and tums [... ] Such a dance is the perfect symbol of
one's own individually asserted freedom as weil as one's respect for the
freedom of the other"l. In an essay on Kant and Schiller de Man makes
the argument most explicitly - Schiller finds a terrible echo in Goebbels'
Michael, where the Fhrer is likened to an artist molding a plastic society: "Politics is the plastic art of the state". This idea de Man derives
from Schiller: in an "aesthetic education" "Schiller would be taught, because it is a popularization, a metaphorization of philosophy. As such,
the aesthetic belongs to the masses; as we all know, and this is a correct
description of the way in which we organize those things, it belongs to
culture, and as such it belongs to the state, to the aesthetic state, and it
justifies the state, as in" Goebbels' novel2. In the Kleist essay de Man
makes the same argument, one that implicitly, if ironically, valorizes his
own role in education as an opposition to aesthetic education. "Tue politics of the aesthetic state are the politics of education" (p. 273) he writes,
and locates the political stakes this way: "Tue 'state' that is here being
advocated [by Schiller] is not just a state of mind or soul, but a principle
of political value and authority that has its own claims on the shape and
the limits of our freedom" (p. 264 ).
De Man uses Kleist to tease out the implications of Schiller's writing.
Kleist puts forward a series of parables, all tending to contrast aesthetic
gracefulness with the anguished self-mutilating self-consciousness that

1 Rhetoric of Romanticism, New York, Columbia University Press, 1984, p. 263.

2 Forthcoming in de Man's Aesthetic ldeology, ed. Andrzej Warminski, University of
Minnesota Press.


belongs to the human life it organizes. In the parable that de Man con-

siders last, the dancer Herr C. describes for the narrator the gracefulness
of marionettes as far outstripping that of any human dancers; this comes
as less of a surprise since he has just shocked the narrator's sensibilities
by describing amputees with artificial legs who dance with an astonishing grace because of their prostheses. De Man's comments on this parable are important, since they make his strongest case against a Schillerian aesthetic ideology:
We have travelled some way from the original Schiller quotation to this mechanical dance, which is also a dance of death and mutilation. Tue violence
which existed as a latent background in the stories of the ephebe and of the bear
now moves into full sight. One must already have felt some resistance to the
unproblematic reintegration of the puppet's limbs and articulations, suspended
in dead passivity, into the continuity of the dance: "all its other members [are]
what they should be, dead, mere pendula, and they follow the law of pure gravity."
Tue passage is the harder to assimilate since it has been preceded by the
briskly told story of an English technician able to build such perfect mechanical
legs that a mutilated man will be able to dance with them in Schiller-like perfection. [... ] Tue dancing invalid in Kleist's story is one more victim in a long
series of mutilated bodies that attend on the progress of enlightened self-knowledge, a series that includes Wordsworth's mute country-dwellers and blind citybeggars. Tue point is not that the dance fails and that Schiller's idyllic description of a graceful but confined freedom is aberrant. Aesthetic education by no
means fails; it succeeds all too well, to the point of hiding the violence that
makes it possible. (pp. 288-89)

These sentences are powerful. They uncover the violence hidden beneath
an aestheticism fonned through its repression, "the potentially violent
streak in Schiller's own aesthetic theory" (p. 280). As always, de Man's
tone expresses no shock at his own shocking conclusions, but something
more like a resigned patience with the untopplable ruses of aesthetic ideology. lt is this tone which distinguishes de Man from Benjamin, to
whom de Man's defenders have been comparing him. Benjamin too
spoke of the violence attendant upon what he called "the aestheticization
of politics". But his tone was of impassioned resistance, not de Man's
world-weary demystification.
lt is de Man's tone that contributes most to his power as a writer. (His
philosophical, political, and aesthetic ideas are not so extraordinary or
original as to deserve the unique hagiography, or the notoriety, which his
tone has eamed for him.) That tone (which I think he leamed from Blan242

chot though I would say that he lacks Blanchot's depth), is almost always
impersonal. The passage above is typical, with its talk of how far "we"
have travelled, or the resistance that "one" must feel.
In his prefaces, it is true, he does speak about himself, usually affecting amused embarrassment, as though surprised that his fragmentary
essays should have cult status. But in his essays he maintains the impersonal tone intact; he never speaks of his own life, of himself in any way,
unlike his Y ale colleagues. Tue impression he gives, the impression he
sought to give, was one of pure intelligence, uncolored by empirical life.
What has become in de Man a jargon tenn - "reading" - means engaging
in something like a purely transparent opacity, opacity untinted by anything extraneous to its own inherent difficulty.
Moments of autobiography then would be as precious as they are
rare; the Kleist essay contains such a moment, and de Man's. students
must have treasured this exception to the anti-autobiographical rule. And
this is not just any exception. De Man actually purports to describe his
own fall from grace, as an illustration of the fall from grace Kleist describes. The text holds a mirror up to de Man and shows his face. And at
what a moment: his fall from grace! This must be interesting just because the end of the essay puns manically on the word "Fall". But in
light of the recent revelations, this description of a fall from grace must
be interesting indeed:
We can all remember personal versions of such a fall from grace, of such a loss
of innocence. (I for one remember trying to drive down a Swiss street after
having just read, in a local newspaper, that for every 100 metres one drives one
has at least thirty-six decisions to make. I have never been able to drive gracefully since.) (p. 277)

Of course the joke is in the deflating example. Nevertheless, the example

is not perhaps entirely jejune. Why a Swiss street? Why the reference not
to just reading, or just .leaming the fact, but to reading a newspaper?
Kleist's story of course appeared in the Berliner Abendbltter, so that the
whole of "Aesthetic Fonnalization in Kleist" consists of a response to
reading a newspaper. Nota Swiss newspaper, however, but a German
one, and perhaps the fact that de Man's fall from grace supposedly occurred after this canonically neutral reading is meant itself to neutralize
whatever culpability might actually consist in for de Man. For de Man's
culpability was not that he wrote in a neutral Swiss newspaper, .but in a
collaborationist Belgian one, whose name, Le Soir, echoes that of the

Gennan Abendbltter Kleist wrote for, whose name might now be more
appropriate for Le Soir.
Is this overreading? Maybe, though there's still the question about the
specificity of the example. Why not any street - why a Swiss street? And
driving in Switzerland is itself a necessarily charged context, since de
Man's uncle Hendrik de Man died there when a train crashed into his car.
He had taken refuge in neutral Switzerland from the sentence passed on
him by the Belgian govemment after the war. Hendrik de Man was the
person who counselled Leopold II to capitulate to the Nazis, and as the
head of the Labour Party he urged collaboration (and probably got Paul
his job with Le Soir). He died on June 20, 1953, probably a suicide, as
Tom Keenan suggested to me: Paul's brother Hendrik had died in a bicycle accident, also at a railroad crossing, on the same date in 1~36 (hence
perhaps the significance of thirty-six decisions), and his mother Magdalena committed suicide on the date of the first anniversary of her elder
son's death, by hanging herself. De Man apparently found her body3; this
is a far likelier candidate for de Man's first loss of innocence. I think
Kleist's newspaper article evoked that hanging body for him: "Caught in
the power of gravity, the articulated puppets can rightly be said to be
dead, hanging and suspended like dead bodies: gracefulness is directly
associated with dead [sic] 4, albeit a dead cleansed of pathos" (p. 287). lt
is after all the contrast between these dead puppets, or at least the gracefulness of the dance, and the "heavy breathing" (p. 279) of the real self
that is forced into the dance (most vividly under fascism) that de Man is
intent to underscore.
De Man, in his response to the anonymous accusation sent to Harvard, strangely identified Hendrik de Man as his father, rather than his
uncle. There has been some speculation about this (adoption?). Certainly
this identification makes the family more suffocating. Tue deaths of the
"father" Hendrik, the son Hendrik and the mother Magdalena Maria implicate Paul more closely than he otherwise might have been if Hendrik
is only an uncle. Tue uncle replaces his own brother as father, and so the
fratemal relation is as specular as the generational relation. Tue next
generation mirrors fratemal specularity, and so the living Paul mirrors
3 Neil Hertz has said that this isn't certain.
4 The essay is taken from a transcript of a lecture de Man delivered at Comell; apparently there was a vote arnong the transcribers as to whether de Man said "dead," or
"death" here and immediately after. I hear it as "death", but will acquiesce to that


his dead brother Hendrik, and so mirrors his father's brother Hendrik as
weil. The father and the son die at railroad crossings, on the same date:
another specular relation. The son's brother must now be part of the
chain of specularity. In other words, he may be the smviving son, but he
might as easily have been the dead son Hendrik, double or mirror image
of the dead father, whose death in turn mirrored his son's. Thus, the allusion to the possibility of repeating the suicidal road accident of both
brother and uncle (responsible for the deaths of mother and aunt as weil)
makes for a charged moment in de Man's essay.
Did de Man know it was a charged moment? I think he did, since
"Aesthetic Fonnalization in Kleist" makes another analytical move of a
sort exceedingly rare in de Man's writing. Not only does it contain an
autobiographical moment, it also uses biography (in Empson's phrase) to
elucidate Kleist's essay. De Man's style of close reading almost always
eschews this referential moment; but this essay, even as it announces the
fictionality of reference, ..the dubious status of referentiality" (pp. 27273), actually engages in a referential argument. I am going tobe quoting
the whole of this argument, paragraph by paragraph, over the next few
pages. I begin simply by establishing the thematization of biography and
autobiography in a text which itself contains a highly unusual autobiographical moment:
Tue only explicit referential mark in the text is the date of the action, given as
the winter of 1801. Now 1801 is certainly an ominous moment in a brief life
rich in ominous episodes. lt is the year when Kleist's self-doubts and hesitations
about his vocation culminate in what biographers call his "Kant crisis." lt is
also the year during which Kleist's engagement to Wilhelmine von Zenge begins to falter and during which he is plagued by doubts similar to those which
plagued Kierkegaard in his relationship to Regina and Kafka in his relationship
to Felice. Between the two events, the Kant crisis and the forthcoming breach
of promise with Wilhelmine (the final break occurred in the spring of 1802),
there seems to be a connection which, if only he could understand it, would
have relieved Kleist from his never resolved self-desperation. To uncover this
link would be the ground of any autobiographical project.
Tue link actually and concretely existed in the reality of Kleist's history,
but it took a somewhat circuitous route. (pp. 283...:84)

I too will be taking a somewhat circuitous route in this reading, but it's a
route that de Man actually invites. lt may not require more than thirty-six
decisions per hundred meters. De Man has asked whether Kleist himself
can say "with fll authority, that his text is or is not autobiographical" (p.


283). The "autobiographical project" which seeks to uncover a hidden

link is hence as much de Man's as Kleist's. To contribute to Kleist's biography is for de Man to engage in something like a form of autobiography, and we've seen already that he has done just that. This provides a
counterweight to the complaint many make against de Man, that his arguments for the non-phenomenality of history, for history as untotalizable iteration, enumeration, and stutter, made for a kind of special
pleading, for an anticipatory self-exculpation. "The reality of [... ] history": that phrase has resonance in de Man now that it didn't in 1984, but
that resonance is not beside the point. For once, de Man is narrating history, and not just evoking its materiality.
Not that Kleist is a mirror image of de Man (or vice versa); but de
Man's text thematizes the narcissism that Kleist describes. Kleist's
speaker (whom de Man calls K solely on the basis of the initialed signature H. v .K., as though one were to call the person who initials the letters
P .d.M. at the base of his articles in Le Soir M) - Kleist's speaker teils C
the story of a young man he knew. Drying himself in front of a mirror
both the speaker and the young man noticed his resemblance to the famous statue of the boy pulling a thom from his foot. (This statue is '"to
be found in German collections"', as de Man stresses [p. 278]; why this
stress? Some allusion to cultural hegemony must be intended here, and
again the wartime writing in:flects this stress.) Tue resemblance delights
the young man, but the speaker, in a challenge to his narcissism, denies
seeing it. Tue young man can never again reproduce his unconscious
grace, and this is for the young man the moment when self-consciousness means a fall from grace.
De Man charts in some detail the non-binary specularity that his
parable evokes: "what the young man confronts in the mirror is not himself but his resemblance to another. [... ] Tue structure is not specular but
triangular. [... ] As for the teacher's motives in accepting to enter these
displacements of identity, they are even more suspect than those of the
younger person" (p. 278-79). This analysis of triangular specularity must
invite the comparison not only between de Man and Kleist, not only
between Kleist and the situations of de Man's own past (such as the
breach of contract to his first wife, when he committed bigamy), but also
between Kleist and other specular figures for de Man. One such figure
would be the pseudo-father Hendrik, for whom Heinrich von Kleist
might stand.


I suggest that Heinrich von Kleist is for de Man another name for
Hendrik de Man, mirror image of bis son Hendrik, and of bis triangulated son Paul. This at least is what de Man's essay, with its own extraordinarily ingenious use of biography, invites us to consider. lntentionally?
Randomly? That's undecidable. But the random difference between the
random and the intentional is the Mallannean point at issue in all of de
Man's work. You can't totalize that difference, even under the anti-totalizing name of the random - you can't be sure that what seems random is
not in fact densely overdetennined. "Aesthetic Fonnalization in Kleist"
repeats just this lesson, in the continuation of the long passage I started
quoting above:
The link actually and concretely existed in the reality of Kleist's history, but it
took a somewbat circuitous route. For wben Kleist next met bis bride~to-be, in
1805, in Knigsberg, sbe was no longer Frulein Wilhelmine von Zenge but
Frau Professor Wilhelmine von Krug. Dr. Wilhelm Traugott Krug was Kant's
successor in the latter's cbair in philosopby at the University of Knigsberg.
Kleist, wbo bad wanted to be, in a sense, like Kant and wbo, one migbt conjecture, bad to give up Wilhelmine in order to achieve this aim, found himself replaced, as busband, by Krug, wbo also, as teacber philosopber, replaced Kant.
What could Kleist do but finish writing, in the same year 1805, a play to be
called - what else could it bave been -Der zerbrochene Krug? (p. 284)

Tue specular structure that de Man's essay has already invoked makes a
dizzying appearance here. De Man doesn't tease out all the possible connections -how could he? But it matters that_Kleist is in a sense a witness
to a narcissistic relationsbip from wbich he is excluded: Wilhelmine
marries Wilhelm, and not (Bernd Wilhelm) Heinrich von Kleist.
Do I need to say that Wilhelm Krug - William Jug - is not so very
different from William Flesch - William Flask? Especially since Kleist,
in the essay de Man alludes to approvingly at the very end of "Aesthetic
Fonnalization", writes of political action - the f ormation of nations - as
discharging randomly, like a so-called "Kleistische Flasche" (later
known as a Leiden J ar) (Kleist, cit., II, p. 321 ).
De Man at any rate would have seen there another moment of narcissistic specularity in the text, as Kleist alludes to the random appearance
of bis own name as a detenninate image of the random. And there's
something a little random, a little wild, or zerstreut about de Man's allusion to this text, since it nowhere contains the tenns - Beifall, Einfall,
Zurckfall - that de Man seems tobe quoting from it (p. 290). This is
another invitation to be suspicious, and to think that what matters is that

Kleist writes there of Kleist, in this case Ewald von Kleist, its inventor,
and may seem to de Man in some sense accidently but inextricably
bound up with the mirroring play of the accidental, with what excludes
all his intent either to belong or to cease to belongs.
On de Man's reading, at any rate, he is also excluded from the mirrored substitution of K sounds, to which he should belong, when Krug,
and not Kleist, replaces Kant. De Man too spent the last years of his life
writing about Kant, as weil as Kleist, and seeing in Schiller a bastardized
and aestheticizing substitution for Kant. So that de Man mirrors something like the oblique mirroring that the narrator of "ber das Marionettentheater" describes in his relationship to the narcissistic young man.
That triangular specularity also haunts the relationship among "ber
das Marionettentheater", Der zerbrochene Krug, and de Man. Der zerbrochene Krug begins with a description of Judge Adam's fall from
grace, a literal one which leaves him bloody, but which his assistant
Licht reads, as I have been doing, genealogically (because of the descent
of the name), and so allegorically:


Ihr stammt von einem lockern ltervater,

Der so beim Anbeginn der Dinge fiel
Und wegen seines Falls berhmt geworden;
Ihr seid doch nicht -?
Gleichfalls -?6

Tue pun on falling here (in the word Gleichfalls) anticipates de Man's
analysis of the word at the end of his essay: "when, by the end of the
tale, the word F all has been overdetennined in a manner that stretches it
from the theological Fall to the dead pendulum of the puppet's limbs to
the grammatical declension of nouns and pronouns (what we call, in English, the grammatical case), then any composite word that includes Fall
(Beifall, Sndenfall, Rckfall ( 46 [of Hegel's Encyclopedia]), or Einfall) acquires a disjunctive plurality of meanings" (p. 289), much like the
overdetennined plurality of allusion I have been following de Man in
tracing. Tue importance of Der zerbrochene Krug may be seen further in
5 What may also matter in that essay is that it lapses into French to tell a fable from
La Fontaine in which the improvising fox randomly comes up with an innocent
scapegoat to be sacrificed in his stead and in the stead of his ruler the lion, whose
dirty work among the lesser animals he has to do.
6 Smtliche Werke, Munich, Knaur, 1962, p. 211.


the fact that the broken jug had on it (among other things) a picture of
Brussels, now destroyed. Its importance to "ber das Marionettentheater" comes from its anticipation of another of the later text's
themes. Judge Adam is clubfooted, and hence a clear avatar of the boy
with the injured foot that the narcissistic young man sees himself doubling in "ber das Marionettentheater;" but Adam claims (somewhat
like C.) that his Klumpfuss is more graceful than his other foot. Adam like de Man (at least in this essay) - tums out Oike the swollen-footed
Oedipus) to be the object of the judgment that he symbolizes: the crime
being tried in his courtroom, the destruction of the jug and the shattering
of its picture of Brussels, was committed by him. Judgments directed at
others now turn out to be self-judgments. And de Man, in Allegories of
Reading, had already hinted at the importance of such a self-reflective
reading: "in the plays of Kleist, the verdict repeats the crime it condemns" (p. 245). What Adam fears throughout is that he will end up in
the position of another guilty judge, one Judge Pfaul (readable as a portmanteau of Paul and Fall?). And given his continual invocation of the
reading and misreading of Kantian judgment, de Man, who reads and
judges these texts would have to see himself in the position of the aesthetic ideologists he condemns:
Wr's wahr, gestrenger Herr? Der Richter Pfaul,
Weil er Arrest in seinem Haus empfing,
Verzweifhmg htt' den Toren berrascht,
Er hing sich auf? (p. 220)

De Man attempts to take control of his encounter with Kleist, perhaps,

by taking control of the letters in the passage: as though he is more able
to articulate and disarticulate Kleist's fate than Kleist himself. As in
Stevens' The Comedian as the Letter C, the play of K-sounds, of course,
is part of the point in the passage, not only in the proper names, but also
in the language which asserts that a link actually and concretely existed
between the Kant crisis and the breach of promise; but this link is linked
also to the break between Kleist and Wilhelmine von 7.enge; link and
break form a kind of Derridean brisure, even if this is conjecture and
even if it took a somewhat circuitous route.
But how does this bravura display of signifiers help de Man's reading
of "ber das Marionettentheater"? Does it? My account of the narcissistic nature of Krug's relationship to his wife and to his predecessor might
shed some light on Kleist's story, but de Man does not himself make that

point. What link does his circuitous conjecture look like it constitutes or
at least evokes? The next paragraph promises an answer. Is there one? I
continue the quotation:
All this, and much more, may have been retained, five more years later, in
1810, when he wrote ber das Marionettentheater, in the innocuous-looking
notation: winter of 1801. But he may just as well have selected this date at ran, like Mainz, although he was to go to
dom, as he wrote city of M
Mainz only in 1803. Who is to say that this notation is random while the other
isn't? Who can tell what terrible secrets may be hidden under this harmless
looking letter M? Kleist himself is probably the one least able to teil us and, if
he did, we would be well-advised not to take his word for it. To decide whether
or not Kleist knew his text to be autobiographical or pure fiction is like deciding
whether or not Kleist's destiny, as a person and as a writer, was sealed by the
fact that a certain doctor of philosophy happened to bear the ridiculous name of
Krug. A story that has so many K's in it (Kant, Kleist, Krug, Kierkegaard,
Kafka, K) is bound to be suspicious no matter how one interprets it. Not even
Kleist could have dominated such randomly overdetermined confusion.

This is all very dazzling, this "dismembennent of language by the power

of the letter" (p. 290); but what does it say about "ber das Marionettentheater"? To me, the link to K.leist's history is still hidden. In fact de
Man means the passage as a demonstration of the impossibility of mastering the difference between the random and the detennined, and not as
even a conjectural account of the origin of K.leist's essay. As conjecture
it falls completely apart; it explains nothing whatever.
Nevertheless, I want to continue reading it in the way it authorizes.
The story, with its invocations of Kierkegaard and Kafka, "is bound to
be suspicious". But this anachronistic story is of course de Man's, not
K.leist's. So de Man is effectively challenging his readers to see his own
essay as suspicious ..
Weil, what terrible secret does hide under "this hannless looking
letter M"? De Man might as weil have alluded to the hannless looking
letter G in "Herr von G ... ", the noblem an who owns the bear that stars in
the last anecdote of K.leist's story. In fact, the link might have been better, since all the K's that de Man cites are persons, while M is a place.
This is all the more striking, since in his list - "(Kant, Kleist, Krug,
Kierkegaard, Kafka, K)" - de Man seems (intentionally? randomly?) to
have left out the other proper name he'd earlier mentioned: Knigsberg
(does this perhaps faintly recall King Leopold, and so require repression?). Why then the anention on M, perhaps short for Mainz? I think


that the parallel set up between K and M (limbed letters nearly graphical
rotations of one another) cements the connection: von Kleist/de Man. (I
don't know whether the de Mans derived their name from Mainz, but the
pun is certainly there, and de Man points to it.)
wtiat does de Man think of the possibility of this kind of conjectural
interpretation? Is it a ruse, or is it accurate? I do not claim that the clues
the text is littered with are none of them red herrings. But the allusion to
bis own life is clear, whether intended or not, whether random or not. Indeed, not even de Man could dominate such randomly overdetermined
confusion, and so he must perforce approve of this conjecture. But I
think he did more than approve - he planted the clues in order to turn bis
reading of Kleist into its own allegory.
De Man has always seemed to bis followers a super-reader. One can
excuse the desire then to interpret the allegory of reading that de Man
reads into Kleist: the allegory of the super-reader. C. tells Kleist's narrator of his trip to Herr von G.'s estate. Herr von G. has two sons, the elder
of whom challenges C. to a fencing match. C. wins, and the son peevishly anounces that he has an opponent for C. that C. will never match.
And indeed C. does not match bis opponent, the chained bear. This prelude de Man does not consider, although it has interesting consequences
for the allegory of reading he is about to set up. C. goes from winning a
fencing match to losing it, and in de Man's terms this means that something like a switch in position is occuring between reader and writer, engaged in "the fencing match of interpretation" (p. 290): fencing with the
bear "I was almost in the state of the young Herr von G''7. Interesting too
is the minutely described situation: G. makes no appearance in the story;
but he has two sons, of whom only the elder matters to the anecdote. The
familial pattern may be typical, bot it also recalls the oddly triangulated
specularity I've been ascribing to de Man's relation to bis own family.
At any rate, de Man's interest is in the way the chained bear can infallibly tell feint from thrust. C. is absolutely unable to touch the bear,
and not only that, he is unable to get the bear to react to a feint. He never
hits the bear, but he doesn't even get the secondary pleasure of getting
the bear to evince a mistaken and unnecessary defensiveness. The bear's
imperturbability C. finds absolutely frustrating, the way he could do
what no human fencer could, and utterly ignore feints, "bis eye on mine,
as though he could read my soul in if' (p. 830).
7 Kleist, p. 830.


De Man sees in this bear "the figure of a super-reader who reduces

the author to near-nothingness" (p. 281), and goes on to say that
the relationship between author-reader and reader-reader now becomes in a
very specific sense antagonistic. For [in a hermeneutic, as opposed to imitative,
model of text] the meaning that has to be revealed is not just any meaning, but
the outcome of a distinction between intended and stated meaning that it is in
the author's interest to keep hidden. (p. 281)

So that super-reading uncovers links or meanings that an author hides.

De Man continues, with a justification of red-herrings as a way for the
author to maintain control over his (or her) hidden meanings: "Hence the
need to mislead the reader by constantly altemating feints with genuine
thrusts: the author depends on the bewilderment and confusion of his
reader to assert his control'' (p. 282). Is this a fair account of de Man's
own will to power over the reading of his texts? Yes and no - it's both
genuine and feigned. As is the next sentence, where de Man once again
elides the difference between writing and reading as a difference between biography and autobiography: "Reading is comparable to a battle
of wits in which both parties are fighting over the reality or fictionality
of their discourse, over the ability to decide whether the text is a fiction
or an (auto)biography, narrative or history, playful or serious" (p. 282).
Again, this account of reading Kleist reading invites what it also attempts to foreclose in a reading of de Man reading.
The bear is a reader who makes reading superfluous. He can always
teil a feint from a thrust, and so experiences none of the resistance that
constitutes reading: ''The entire hermeneutic ballet is a display of waste:
either we master the text, and then we are able but have no need to feint,
or we don't and then we are unable to know whether we feint or not. In
the first case, interpretation is superfluous and trivial, in the second it is
necessary but impossible" (pp. 282-83). Yet something odd has gone on
here, signaled a few lines earlier in fact: "Kleist puts his own text en
abyme in the figure of the super-reader or super-author made invincible
by his ability to know feint from" thrust (p. 282). Super-reader or superauthor. The switch in position that I mentioned above occurs here, because suddenly in de Man's text, the bear is associated with author and
not only reader. In what follows, the bear annihilates not the author
(reduced to "near-nothingness") but the reader, who knows less about his
(or her) own reading of the text than the author who controls it. That

reader tumed writer is of course de Man, specular image of Kleist. What


more apt description for bis current position (or that of bis spokespersons) than de Man's description of the possibility that the author has become a super-reader, de Man as super-man? De Man is a super-reader
and so.becomes a super-author; and then look what happens if you follow t:Q.e specular logic of this reading, by replacing "Kleist" with "de
Man" in the following quotation, so that anyone who criticizes de Man
becomes only part of "a harrassed pack of snipers beaten in advance":
And h~w about Kleist's own text? By staging the figure of the super-reader, has
he hlrn.self become like the bear and achieved the infallible discrimination of
genuine seriousness - "der Ernst des Bren" - reducing bis commentators to a
harrassed pack of snipers beaten in advance? Can he say, for example, that bis
text is or is not autobiographical? (p. 283)

Why the military tenn "snipers"? Does it matter, in view of the later revelations about de Man's past, that Kleist had been a Prussian officer?
That Wilhehnine von Zenge was the daughter of a Prussian officer? Is
the fencing match of interpretation one more allegory for Paul de Man's
war? He continues bis reading in a way that seems to suggest this, precisely by attempting to ward it off, by arguing that we cannot ever read
as infallibly as a reading like mine might seem implicitly to claim tobe
doing. But at the same time he goes on to hint that Kleist is not quite in
control of bis own text. I continue the long passage that is the focus of
this analysis:
Not even Kleist could have dominated such randomly overdetermined confusion. The only place where infallible bears like this one can exist is in stories
written by Heinrich von Kleist.
Why did Herr C, once he bad discovered, as we can assume he bad, that
the bear could tell feint from thrust, persist in trying to feint? Could he not have
matched the bear's economy of gestures by making all bis attacks genuine,
forcing the bear to tak:e them seriously? Grantecl it would have been tiresome,
but not more so than the actual situation, and the fatigue would have been
shared. Both would have sweated instead of C alone...,... and, for all we know, he
might have scored. Such a commensensical solution however is logically possible only if one concedes that C is free to choose between a direct and an oblique
attack. But this is precisely what has to be proven. lt is only an hypothesis, and
as long as it has not been verified, C can never unambiguously attack. From the
point of view of the bear, who knows everything, he always feints and, as
seems indeed to be the case, the bear hardly ever has to make a move at all.
From C's own point of view, which is deluded, no thrust ever goes where it is
supposed to go. His blows are always off the mark, displaced, deviant, in error,
off-target. Such is language: it always thrusts but never scores. lt always refers
but never to the right referent. (pp. 284-85)


This is fast, but I gather that "he always feints" because even bis thrusts
may be meant only to set up bis feints; then they wouldn't be real thrusts.
At any rate, no attack on the super-reader tumed super-author can succeed. In every way, he has anticipated bis readers. This anticipation is
one of the lessons of the essay on Kleist. Sniping commentaries, interested in de Man's own bistory, will find themselves always prevented by
de Man. What is perhaps strangest about this is the odd good humour of
the whole account. And yet this is in keeping with de Man's argument.
Tue risk that the tone meets is the risk that playfulness will be slain in
the serious defensiveness that this essay puts forward. But de Man, in
this section at any rate, is just not serious, unlike the bear with whom he
otherwise identifies:
Why then indulge in reading (or writing) at all since we are bound to end up
looking foolish, like the fencer in the story, or to become the undoer of all pleasure and play, like the bear has become by the end of the story, when he has
killed off all possibility for play by scoring whenever he deigns to enter the fray
- which he does out of defensive necessity. No one is hurt, for the bear never
attacks, except for the game itself, forever slain in the unequal contest between
seriousness and play. (p. 283)

No one is hurt, but the potential for hurt comes out of the seriousness of
readers who would ignore de Man's playfulness to the peril of all flexibility. De Man here scapegoats as overserious any reader who (like me)
would do to him what he nevertheless enables her to do, simply by mirroring him. Defensive necessity would be de Man's apology for the
lighthearted tone of the essay, and presumably also for the "reality [of
bis own] bistory". De Man here undoes the difference between bis absolute imperturbability as the super-reader/super-author and the levity he
displays, levity capable of parrying any attempt to motivate de Man's
imperturbability of tone, or iciness of style.
And yet this levity may have a source in the reality of Kleist's bistory
as wen. For under the innocuous-looking notation 1801 Kleist may be alluding to another duel he leamed of at that time, a duel involving another
Kleist, who was saved from death and became a poet through the deadly
serious triumph of levity over seriousness. Thus once more de Man's reflections on this specular text lead to another specular moment. In this
one Kleist might see himself, or a wishful version of himself, in a succesful kinsman - like Wilhelmine's father a Prussian officer who (unlike
Heinrich) bad not quit the military. In a letter to Wilhelmine von Zenge


Heinrich writes of visiting the poet Gleim in Paris. For Gleim had been a
friend of the poet Ewald von Kleist, a distant relation of Heinrich's, who
"feil at Frankfurt". He writes Wilhelmine:
Kleist was at that time wounded in a duel, and lay sick in bed at Potsdam.
Gleirn was then the regimental quartermaster and visited the sick man, without
knowing anything more about him. Ah, said Kleist, I have the most terrible
boredom, since I can't read anything. 1'11 teil you what, answered Gleim, I want
to come here from time to time and read something to you. [Hemi Thomas, in
his novel about de Man, Le Parjure, will have his narrator reading to the
bedridden de Man character as weil.] At that very time, Gleim had just written
comic poetry, in the style of Anacreon, and read him among others an ode about
death, which ran something like this: Death, why do you elope with my girl?
Can you fall in love with someone else? - And it goes on in this way. At the
end it says: what will you do with her? lndeed with your lipless teeth you can
bite the girl, but not kiss her. - At this idea, how death with his naked, angular
teeth vainly pressed into the soft red lips, in quest of a kiss, Kleist laughed so
hard that through his shaking the bandage on the wound on his hand jumped
off. They called a surgeon, who said, lt's a good thing that you sent for me,
since gangrene is developing insidiously, and tomorrow it would have been too
late. - Out of gratitude Kleist dedicated to poetry the life that poetry had rescued for hims.

This Kleist is wounded by a pistol duel, not a sword. And the military
context is clear, even if its seriousness is relieved by the luridly comic
text that saves Kleist's life. The grotesque undoing of prosopopoeia
would, I think, have appealed to de Man.
But why did de Man choose Kleist as the occasion of parodic confession? Or why did Kleist choose him? In the last sentence of "Aesthetic
Formalization" de Man coordinates history with "the fencing match of
interpretation". That fencing match is not the same as the one-sided one
between chained bear and attacking but powerless human. De Man performs here his own indifference to the past, precisely in the way he anticipates all criticism, and suggests that none of it can possibly hit its
mark. And yet it was the same indifference to the hideousness of Gennan
hegemony that allowed de Man first to accede in and then to obliterate
his part in the disarticulation of the Low Countries during World War II.
These bodily terms are nearly the same as those he uses on 25 February
1941 when he praises the Belgian army for its resistance to the Germans:
praise designed to denigrate Belgium's allies, not the Germans (and also
8 My translation from the letter which appears in Kleist's Smtliche Werke und Briefe,
Munich, Hanser, 1961, II, 656-57. This is the edition and volume that de Man uses.


to produce propaganda demonstrating that Belgium could still think herself free even though she was in chains): "Among all the armies placed
on the line, the Dutch, the English, the French, and the Belgian, only this
last maintained control of its movements and was not dislocated by the
Gennan attacks." That dislocation and loss of motor control, that disarticulation, began with a military conquest spearheaded by a Panzer Division whose invasion was so precisely coordinated that it might have reminded one of a well-executed German dance, composed of many complicated figures and turns9. Much as the young de Man wanted to read
this invasion as a mock combat, led by "highly civilized invaders", for
others it was all too serious.
Tue Panzer Division which most effectively disarticulated the Low
Countries - or perhaps articulated them, made them limbs of Gennany was commanded by a general named Ewald von Kleist 10. Tue French
name for that division was le corps Kleist, the Kleist body. De Man became one of the willing puppets of the masters of le corps Kleist, sublimely indifferent to "the suffering that daily twists humanity at war", an
indifference for which "Aesthetic Fonnalization" is perhaps mirror, perhaps apology.

9 See B.H. Liddell Hart, The German Generals Talk, New York, Morrow, 1948, pp.
111-136, for example this: "[General] Blumentritt revealed that on eleven occasions
between November and April the armies received the order to 'fall in' - to be ready
to attack in forty-eight hours. 'Each time it was cancelled before the time expired.
These repeated cancellations led us to think that Hitler was merely bluffmg, and was
only using the threat of attack as a means of prompting the Allies to consider his
peace offer.' But when the twelfth order came, in the month of May, events took
their fatal course." (p. 111 ).
10 He was a collateral descendent of Heinrich.



A la fin de sa vie, Paul de Man proposa ses reflections sur la forme, le developpement et
les effets de ce qu'il a appele "l'ideologie esthetique". II fit remonter l'une des formes
prises par cette idoologie en Europe a l'erreur que (selon lui) Schiller commit en lisant
Kant, une erreur qu'il mit en parallele avec celle commise par Goebbels. lorsqu'il s'inspira
de Schiller. "Aesthetic Formalization in Kleist's ber das Marionettentheater represente
l'un des plus importants de ces exposes. La maniere dont l'essai est construit est neanmoins curi~use et semblerait dependre d'une lecture dont l'excentricite reproduit, pour le
moins, celle que cet essai decouvre et analyse chez Kleist Suivant lme pratique deconstructionniste devenue courante, j'essaie de soumettre cet essai au meme genre de lecture
que cet essai lui-meme applique aKleist. II semblerait, d'apres cette lecture, que de Man
ecrit une "crypto-autobiographie" faisant allusion aux activires collaborationnistes
auxquelles il s'etait engage en Belgique, de 1940 a 1943. La raison pour laquelle de Man
aurait produit une telle autobiographie reste mysrerieuse. II esperait peut-etre offrir, au
cas ou ses activites apparaitraient au jour, une justification qui ne se decouvrirait que si
ces activites elle-memes etaient decouvertes; ou peut-etre peut-il lui-meme etre accuse
d'entretenir une relation detachee., voire esthetisee., avec sa propre histoire_


Jonathan Culler


As my title implies, it seems to me appropriate both to initiate a concluding discussion that might bring out what we have leamed or accomplished in our engagement with the work of Paul de Man at this conference and to attempt to direct our thinking toward the future. When Paul
de Man died in 1983, he left behind a legacy of difficult texts, many of
them still unpublished. lt seemed apparent that an important activity of
criticism and theory in America, at least in the immediate future; would
be to interpret de Man's critical and theoretical writings, exploring their
implications, and' especially working out their possible relationships to
other contemporary critical discourses, such as psychoanalysis, feminism, and revi~ionist marxisms, which have frequently engaged those
versed in deconstruction but did not interest de Man.
The discovery of de Man's wartime journalism by Ortwin de Graef
and of further early joumalism by Thomas Keenan has posed a different
set interpretive tasks, which many of the contributors to the volume of
Responses, edited by Werner Hamacher, Neil Hertz, and Thomas
Keenan, have ably taken on, reconstructing historical and theoretical
contexts and bringing out dimensions of de Man's mature writings that
resonate with the earlier. The discussion of these wartime writings has
only illustrated more clearly a principle of which we were already aware,

that meaning is context bound but context is boundless, and our surprise
at the contextual materials that have become relevant demonstrates with
a force thathas often been dismaying that the principle of the boundlessness of context insures the impossibility of mastering the meaning even
of discourses one knows weil.
This project of interpretaton is scarcely complete or completable, but
, I fear that the attempt to complete it may frequently involve a surrender
to the temptations of narrativization, as we produce from the array of interpretive materials and connections offered in the Responses volume
and elsewhere, an intelligible story, whether it be the tale of a conver259

sion, of de Man's progress from the wickedness of an European ideology

to the analytical detachment of American close reading, or a turn from a
disastrous political engagement to an evasion of politics, only belatedly
mitigated by bis "critique of aesthetic ideology" and never-fulfilled project of writing on Marx and Kierkegaard, or a tale of sinister continuities
and the return of the repressed. Tue seductions of these fictional narrative schemes are harder to resist now - when there is a lurid "before" one
ineluctably seeks some sort of "after." De Man writes "no one in bis
right mind will try to grow grapes by the luminosity of the word 'day',
but it is very difficult not to conceive of the pattem of one's past and future existences in accordance with temporal and spatial schemes that
belong to fictional narratives and not to the world"l. The problem, of
course, is that while these schemes do not "belong" to the world, they
structure it, and within the circumstances that they generate, the most
powerful counter to one narrative scheme usually seems to be another
scheme which takes in more of the facts. lf in our theoretical sopbistication we thought we had learned to avoid the tendentious constructions of narrative, we find the dependence of intelligibility on narrative figures forced imperiously upon us - and thus all the more important to resist in the interests of reading.
In looking toward the future one might ask many questions about the
future of Paul de Man - about the future of that name, of that oeuvre, and
the future that bis writings offer those who engage them. My question is,
what seems especially valuable or productive in de Man's work for the
future of literary criticism and theory? In responding to such questions
one's tendency is to focus on things one thinks one understands - something that can be presented as an acbievement, an acquis. In the present
critical and polemical climate it may be especially tempting to take certain ends as given and to claim that de Man's writings are means towards
those ends. The <langer of thinking in this way is of forgetting that reading and research are valuable insofar as they remain open to the unknown or unexpected. Discovering what one already believes one knows
is a tautological sort of research and writing. One should try, therefore,
to avoid the temptation of treating de Man's work as a set of solutions in
order to preserve the possibility of learning something new. lt is quite
possible, even likely, that this work will sutprise us - that previous
Paul de Man. The Resistance to Theory, Minneapolis, University of Minnesota
Press, 1986, p. 11. Henceforth cited as RT.


readings will come to seem misreadings, that this work may prove to
have interesting things to say about precisely those matters it is reputed
to ignore, for instance. De Man writes of Rousseau that "the existence of
a particularly rich aberrant tradition" of misinterpretation of a writer is
"no accident.but a constitutive part of all literature, the basis, in fact, of
literary history''2. We can say that what makes an oeuvre live is its tantalizing availability for reading and misinterpretations, its resources for reversing what it is thought to have achieved.
On an earlier occasion, for a volume entitled The Future of Literary
Criticism, I outlined five areas in which it seemed to me that de Man had
made signal contributions3. I want to take up each of these (while adding
a sixth topic) in terms of the somewhat different perspective I have been
seeking to develop, thinking less of achievements than of future problems and prospects.
1. First, there is de Man's revaluation of allegory, which criticism in the
wake of Coleridge and Goethe had treated as an undesirable and unsuccessful type of figuration, a product of the operations of fancy rather than
imagination. An assumed superiority of the symbol underlay literary
taste, critical analysis, and conceptions of literary history. Looking at the
supposed shift from allegorical to symbolical imagery in late-eighteenthcentury poetry in ''The Rhetoric of Temporality," de Man challenges the
view that romantic literature produces through the symbol a reconciliation of man and nature and instead identifies the allegorical structures at
work in its most intense and lucid passages. Allegorizing tendencies
"appear at the most original and profound moments [... ] when an authentic voice becomes audible," in works of European literature between
1760 and 1800. He writes:
Tue prevalence of allegory always corresponds to the unveiling of an authentically temporal destiny. This unveiling takes place in a subject that has sought
refuge from the impact of time in a natural world to which, in truth, it bears no
resemblance [...] Whereas symbol postulates the possibility of an identity or
identification, allegory designates primarily a distance in relation to its own origin, and, renouncing the nostalgia and the desire to coincide, it establishes its
language in the void of this temporal difference. (BI, pp. 206-7)

2 De Man, Blindness and /nsight, enlarged edition, Minneapolis, University of Minnesota Press, 1983, p. 141. Henceforth cited as BI.
3 Jonathan Culler, "Paul de Man's Contribution to Literary Theory and Criticism," in
The Future of Literary Theory, ed. Ralph Cohen, New York, Routledge, 1989.


This account of the relation between symbol and allegory, and its revaluation of allegory, has been central to recent work on romantic and postromantic literature in America, but the implications of de Man's reflection on allegory are not exhausted here. We can now see, as Minae
Mizumura writes, that "the tension between symbol and allegory is already another name for the tension between a temptation of assuming the
readability of a text, that is, of reconciling sign and meaning, and a renunciation of this temptation"4 But de Man also uses the term allegory
in Allegories of Reading for texts' implicit commentary on modes of signification, implied second- or third-order narratives about reading and
intelligibility. Foregrounding the way texts function as allegorical statements about language, literature, and reading, Allegories of Reading
poses a question about the relation of figuration to interpretation that
needs to be pursued.
But the further question that now may pose itself for us more pressingly is the relation between allegory and history. In the conclusion of
the "Promises" chapter of Allegories of Reading, while arguing that the
"redoubtable efficacy" of Rousseau's Social Contract is due to the
rhetorical model of which it is a version, de Man writes "textual allegories on this level of complexity generate history ," as if the historical
effect or productivity of a text were an allegorical power, a power of allegory5. Tue relationship seems more intimate yet difficult to grasp in the
last essays where allegory seems an incomplete narrative of a non-figurative occurrence which de Man associates with the materiality of actual
history or historical modes of language-power. Kevin Newmark's difficult essay "Paul de Man's History" in Reading de Man Reading helps
trace the elaboration of these terms6.
2. One of de Man's achievements has certainly been the revaluation of
romanticism, the demonstration through studies of Rousseau, Hlderlin,
Wordsworth, Shelley, Keats, and Baudelaire that it includes the boldest,
most self-conscious writing of the Western tradition. Tue early romantics, Rousseau, Wordsworth, and Hlderlin, are "the first modern writers
to have put into question, in the language of poetry, the ontological pri4 Minae Mizumura. "Renunciation," Yale French Studies, 69, 1985, p. 91.
5 De Man, Allegories of Reading, New Haven, Yale University Press, 1979, p. 277.
Henceforth cited as AR.
6 Wlad Godzich and Lindsay Waters, eds., Reading de Man. Reading, Minneapolis,
University of Minnesota Press, 1989.


ority of the sensory object," for which later romantic and post-romantic
literature and critical discussions of it would remain nostalgic7. lt is now
apparent that other things are at stake in de Man's focus on romanticism,
that the focus on it is crucial to an understanding of our recent past and
our cultural situation. For instance, there is the problem of what Philippe
Lacoue-Labarthe in La Fiction du politique calls the "national aestheticism" that issues from a reading of romanticism but to which the work of
a writer such as Hlderlin provides a divergence of crucial, critical force.
A critique of the reception of romanticism has been an activity of
"deconstruction in America." An aestheticizing and monumentalizing
interpretation of romanticism, institutionalized in the teaching of
Wordsworth in American universities, has been challenged andin some
measure dismantled by the deconstructive readings produced by de Man
and his students8.
De Man insists that the question of romanticism is not just one of
characterizing a period or a style. Discussion of romanticism is particularly difficult, he suggests, because it requires a coming to terms with a
past from which we are not yet separated, a past whose most intense
questioning involves precisely this interpretive relation to experiences
become memories - that is, the very structure on which our relation to it
depends. Descriptions of romanticism always miss the mark, for reasons
which are structural rather than due to failures of intelligence. A further
complication is introduced by the fact the genetic categories on which
literary history depends - the models of birth, development, death - are
most decisively promoted but also exposed by the romantic works that
they would be used to discuss: "one may weil wonder what kind of historiography could do justice to the phenomenon of romanticism, since romanticism (itself a period concept) would then be the movement that
challenges the genetic principle which necessarily underlies all historical
narrative" (AR, p. 82). As a result, he writes, "the interpretation of romanticism remains for us the most. difficult and at the same time the
most necessary of task" (RR, p. 50).

7 De Man, The Rhetoric of Romanticism, New York, Columbia University Press,

1984, p. 16. Henceforth cited as RR.
8 Tue key role of this critique in de Man's own changes - the turn towards a linguistic
terminology above all - emerges clearly in the dual versions of his "Time and History in Wordsworth," published for the first time by Andrzej Warminski and Cynthia Chase inDiacritics, Winter 1987.


3. Third, there is de Man's identification of the relationship between

blindness and insight. In Blindness and Insight he argues that critics
"owe their best insights to assumptions these insights disprove," a fact
which "shows blindness to be a necessary correlative of the rhetorical
nature of literary language" (BI, p. 141). A famous passage describes the
way the New Critics' concentration on language (rather than authors, for
example) was made possible by their conception of the work as organic
fonn but led to insights into the role of irony that undennine the conception of literary works as hannonious, organic wholes. For them, as for
other critics, an
insight could only be gained because the critics were in the grip of this peculiar
blindness: their language could grope towards a certain degree of insight only
because their method remained oblivious to the perception of this insight. Tue
insight exists only for a reader in the privileged position of being able to observe the blindess as a phenomenon in its own right - the question of his own
blindness being one which he is by definition incompetent to ask - and so being
able to distinguish between statement and meaning. He has to undo the explicit
results of a vision that is able to move toward the light only because, being already blind, it does not have to fear the power of this light. But the vision is unable to report correctly what it has perceived in the course of its joumey. To
write 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 insights
that it unwittingly provides. (BI, pp. 105-6)

I should add, parenthetically, that this is not de Man's attempt, as some

have claimed, to make bis own youthful blindness an ineluctable necessity - at least not unless one can show some brilliant insight of his
wartime journalism that was made possible by its blindness to the significance of anti-semitism, for example. De Man's is a theory about the
dependency of truth upon error, not simply about the pervasiveness of
This relation is structural, not psychological, for de Man. Tue blindness is not a product of the distinctive individual histories of critics. And
although "blindness" seems to belong to a phenomenological vocabulary
of consciousness, perhaps we should rather construe it in a more mechanical way, as a predictable disruption of a perceptual mechanism. As
Barbara Johnson's and Hans Jost Frey's papers have indicated, de Man
speaks of what others would call the unconscious in tenns of mechanisms of language: what happens independently of any intent or volition
of subjects. He would, as Neil Hertz stressed, interpret psychological ac-


counts as defensive ways of creating intelligibility, of countering the

threat of the random and of mechanical unintelligibility. An important
question here, which Barbara Johnson's discussion of the "inhuman and
impersonal" has broached, is the possible impact of this way of thinking
on a post-structuralist psychoanalytic criticism which explores how texts
are structured by psychic conflicts or operations they theorize. As Neil
Hertz suggested, de Man's thought may link up with the explorations of
Nicolas Abraham and Maria Torok. At a time when psychoanalytic
readings may become the refuge of a certain humanism, as in American
Ego psychology, which sees us as most human in our "unconscious
selves," insistence on impersonal mechanisms may prove salutary.
4. This leads to my fourth and I think most important topic, what Barbara
Johnson calls de Man's developmentof a materialist theory of language
and Hans-Jost Frey investigates as the "madness of words." One might
say that what de Man first described as the division at the heart of Being,
and then as the complex relation between blindness and insight that prevents self-possession or self-presence, is in his later work analyzed as a
linguistic predicament, the figural structure of language that insures a division variously described as a gap between sign and meaning, between
meaning and intent, between the perfonnative and constative or
"cognitive" function of language, and between rhetoric as persuasion and
rhetoric as trope.
Although literary theory has to a considerable extent assimilated the
demonstration that reading should focus on the discrepancies between
the perfonnative and constative dimensions of texts, between their explicit statement and the implications of their modes of utterance, criticism has not yet explicated and worked with the more difficult and unsettling aspects of de Man's writing on language and occurrence. In emphasizing certain non-semantic aspects of language, from the indetenninate significative status of the letter, as in Saussure's work on anagrams,
to the referential moment of deixis, as in Hegel's "this piece of paper,"
de Man stresses that language is not coextensive with meaning, and
rhetorical reading becomes in part an exposure of the ideological imposition of meaning as a defense we build against language - specifically
against the inhuman, mechanical aspects of language, the structures or
grammatical possibilities that are independent of any intent or desire we
might have, yet which are neither natural nor, in fact, phenomenal.


There are, in de Man's accounts, two levels of imposition. First there

is the positing by language, which does not reflect but constitutes, which
simply occurs. De Man speaks of "the absolute randomness of language
prior to any figuration or meaning" (AR, p. 299). This does not mean, as
some commentators affect to believe, that somehow agents are not responsible for their words or actions; on the contrary, the possibility of
their being responsible depends on the randomness of language itself, the
blind occurrence of its positing. De Man writes, ''The positing power of
language is entirely arbitrary, in having a strength that cannot be reduced
to necessity, and entirely inexorable in that there is no alternative to it"
(RR, p. 116). Then there is the conferring of sense or meaning on this
psiting, through figuration - as in allegorical narratives of law and desire, lurid figures of castrating and beheading, and less lurid figures as
weil. Positing does not belong to any sequence or have any status; these
are imposed retrospectively. De Man asks, "How can a positional act,
which relates to nothing that comes before or after, become inscribed in
a sequential narrative?[ ... ] it can only be because we impose, in our turn,
on the senseless power of positional language the authority of sense and
meaning" (RR, p. 117). We transfonn language into historical and aesthetic objects, or embed discursive occurrences in ~arratives that provide
continuities, in a process of troping that de Man calls "the endless
prosopopoeia by which the dead are made to have a face and a voice
which tells the allegory of their own demise and allows us to apostrophize them in our turn" (RR, p. 122). "We cannot ask why it is that we,
as subjects, choose to impose meaning, since we are ourselves defined
by this very question" (RR, p. 118).
Neil Hertz's "More About Lurid Figures," discussing passages in
which de Man describes the law in Rousseau, observes that for de Man
the divergence between grammar and meaning becomes explicit when
the linguistic structures are stated in political tenns (AR, p. 269). De Man
writes of "an unavoidable estrangement between political rights and laws
on the one band, and political action and history on the other. The
grounds for this alienation are best understood in tenns of the rhetorical
structure that separates one domain from the other" (AR, p. 266). That
rhetorical structure is the discrepancy between language conceived as
grammar and language as reference or intentional action, and the ineluctability and indetenninacy of this structural relationship is what de
Man calls "text." "The structure of the entity with which we are con-

cerned," writes de Man in bis exposition of The Social Contract, "(be it


as property, as national State, or as any other political institution) is most

clearly revealed when it is considered as the general fonn that subsumes
all these particular versions, namely as legal text" (AR, p. 267). The
problematical relationship between the generality of law, system, grammar, and its particularity of application, event, or reference is the textual
structure ~ousseau expounds in the relationship between the general will
and the particular individual, or between the state as system and the sovereign as active principle. The tension between grammar and reference
is duplicated in the differentiation between the state as a defined entity and the
state as principle of action or, in linguistic terms, between the constative and
performative function of language. A text is defined by the necessity of considering a statement, at the same time, as performative and constative, and the
logical tension between figure and grammar is repeated in the impossibility of
distinguishing between two linguistic functions which are not necessarily compatible. (AR, p. 270)

What is the significance of that aporia between performative and constative? lt emerges clearly in Rousseau's question of whether "the body
politic possesses anorgan with which it can enoncer [articulate] the will
of the people." The constative function of stating a preexisting will and
the perfonnative positing or shaping of a will are at odds, and while the
system requires that the organ only announce what the general will detennines, the action of the state or "lawgiver" will in particular instances
declare or posit a general will. This is especially so in the founding of the
state, for though, as Rousseau writes, "the people subject to the Law
must be the authors of the Law", in fact, he asks, "how could a blind
mob, which often does not know what it wants [promulgate] a system of
Law." The structural tension between perfonnative and constative here
in what de Man calls the text is detenninative of history, with the violence of its positings, its tropological substitutions, and their "eventual
denunciation, in the future undoing of any State or any political institution'~ (AR, pp. 274-5).
5. One might, then, insert an additional rubric to bring my five to six: de
Man's writings, contrary to what has frequently been suggested, offer a
particularly demanding reflection on the nature and structure of history9.
9 See, for example, Cynthia Chase's discussion of de Man's Kleist essay in her
"Trappings of an Education," in Responses, ed. W. Hamacher, N. Hertz and T.
Keenan, Lincoln, University of Nebraska Press, 1989.


Andrzej Warminski, in bis contribution to the volume on de Man's

wartime joumalism, when puzzling over what might have caused the
frenzied viciousness of newspaper responses to the de Man affair,
shrewdly offers the hypothesis that precisely when de Man seemed dead
and buried, he comes back - bad enough - but "comes back in a way that
makes it forever impossible for 'supporters' and 'enemies' alike to mouth
the old stupidities about bis work" as 'anti-bistorical' or 'apolitical' - revealing that in truth it is "nothing but a sustained, relentless mediation on
bistory and the political"lO.
Indeed, I think this is so, and it will be valuable to demonstrate it in
critical readings - not just of de Man's late works, but also of the earlier
writings, given the inclination of both opponents, such as Frank Lentricchia, and friendly commentators, such as Christopher Norris, to read de
Man's essays of the 1950s as evasions of bistory and retreats to inwardness11.
Norris is altogether mistaken, for instance, when he insists that "what
de Man always sets up in opposition to bistory is a certain idea of the poetic, of poetry as a deeper, more authentic knowledge, undeluded by the
claims of merely secular understanding. Perhaps," Norris continues, "the
most striking example is "Wordsworth and Hlderlin," where de Man
raises questions of bistorical belatedness, of poetry's relation to politics,
and specifically that kind of revolutionary politics that preoccupies
Wordsworth in The Prelude"12.
This is indeed a striking example, for the argument of this essay is
precisely the opposite of what Norris takes it tobe. Far from opposing,
as Norris claims, an inwardness of poetic imagination to an exteriority of
historical action, de Man specifically links revolutionary historical action
to the poetic imagination; such action in this case falls through "excess
of interiority" - a commitment to the imperious autonomy of the imaginations and a disregard for material resistances. In Wordsworth's Prelude, the revolutionaries who attack the cloister of the Grande
Chartreuse, like the travellers setting out to cross the Alps, "are driven
by the same, almost divine wish, and stand under the influence of the poetic faculty. This gives them the power to direct themselves decisively
10 Andrzej Wanninski, "Terrible Reading," in Responses, cit., p. 389.
11 Shoshana Felrnan's "Paul de Man's Silence," Critical lnquiry, 15, 1989, particularly
pp. 722-44, provides a compelling demonstration of this sort.
12 Christopher Norris, Paul de Man: Deconstruction and the Critique of Aesthetic ldeology, London, Routledge, 1988, p. 5.


towards the future. But it is just as certain that in this same instant, this
faculty is conscious. neither of its powers nor its limits, and that it errs
through excess" (RR, p. 58). Elsewhere de Man speaks of Marxism as
"ultimately a poetic thought" in its attempt to imagine a future that corresponds to its own convictions (BI, p. 240).
History and poetry are thus in a very complicated relation in
Wordsworth, Hlderlin, and de Man. First, "History is, to the extent that
it is an act, [only part of the notion of history] a dangerous and destructive act, a kind of hubris of the will that rebels against the grasp of time.
But on the other hand it is also temporally productive since it allows for
the language of reflection to constitute itself' (RR, p. 57). Tue language
of reflection and inteipretation, which manifests itself when the traveilers in the Prelude recognize they have missed the crossing of the Alps,
is that of poetry, but it is also the language of history, since as de Man
notes at the end of the essay, "the poet and historian converge in this essential point to the extent that they both speak of an action that precedes
them but that exists for consciousness only because of their intervention"
(RR, p. 65). Act and inteipretation are linked but divided (for the historian as weil as poet and critic), as in the crossing of the Alps, "in which
the coming-to-consciousness is in arrears of the actual act."Moreover, if
the act of the revolutionaries or the traveilers fails from its excess of interiority, "poetry partakes of the interiority as weil as the reflection: it is an
act of the mind which allows it to turn from one to the other" (RR, p. 59).
Note that interiority and reflection are the opposites distinguished here,
not synonyms. Poetry is not opposed to history but includes both the
imaginative projection and the self-conscious reflection on the relation
between an occurrence and the signification it acquires.
In Wordsworth's complex interrogation of the relation between poetry and history, "the imagination appears as the faculty which allows us
to think of our striving for action as a need for a future, as a maladie
d'idealite (as Mallarme put it) that projects us out of the everyday present into the future" (RR, p. 57). lt is the vital, dangerous, productive energy of historical action as weil as the source of poetr)'. Although Norris
writes that for de Man "the history of poetic consciousness from
Wordsworth to Hlderlin is a passage marked by a growing disenchantment with the idea that poetry, or imaginative thought of any kind, might
actively engage with issues of real-world history and politic~", de Man's
reading of Hlderlin argues rather that "the direct opposition between
Titanism and poetry that has been maintained, explicitly or implicitly, by

so many interpreters", is a mistake, and that in Hlderlin "a dimension

similar to Titanism can reside within the poetic act, although it represents at the same time a turning back [from the excess of interiority]
through which consciousness transforms the excess into language" (RR,
p. 63).
I suspect that a reading of those passages or essays which are supposed to demonstrate de Man's aversion to history will in fact show a
powerful reflection on the gap between action and knowledge, which it
may be awkward for those who live by claiming to study literature politically to acknowledge. Despite the overwhelming evidence history affords that one cannot control the historical outcome of one's actions, and
despite reflection on this problem in the Marxist tradition, critics such as
Lentricchia persist in treating de Man's discussions of this problem as a
rejection of history or of "the political," as it is called, even as a rejection
of action. They might do weil to look at de Man's critique of Malraux in
"Tue Temptation of Pennanence" for "a nationalistic conservatism" that
emerges from the attitude of one for whom history has become painful
and who sees it "only as a shapeless fatality" or his attack in "The Inward Generation" on "a preconceived and reactionary view of history as
indifferent and meaningless repetition"13.
A challenge for serious interpreters of the problem of history is to
relate such remarks to de Man's later discussions of history as power and
occurrence, as in "Kant and Schiller", where he writes, in a passage discussed by Marc Redfield, "History is therefore not temporal, it has
nothing to do with temporality but [with] the emergence of a language of
power out of a language of cognition." History is productive occurrence,
and therefore not meaningless, but it emerges from a breakdown of cognition and of the tropological system on which cognition depends. Historical occurrence is unnameable, but gives rise to naming, through such
figures as prosopopoeia, which figuratively reinscribe it into a system of
meaning14. This persistent approach to the relation of act and interpretation is a rich vein of thought that we have scarcely begun to mine.
6. Finally, de Man's late essays, collected in Aesthetic Ideology, undertake a critique of an aesthetic ideology which imposes, even violently,
13 In de Man, Critical Writings, 1953-1978, ed. Lindsay Waters, Minneapolis, University of Minnesota Press, 1989, pp. 33-4 and 17.
14 See Sam Weber, "The Monument Disfigured,'' in Responses, cit, p. 422.


continuity between perception and cognition, fonn and idea, and wbich
reading, pursued to its limits (as it occurs in texts), is always undoing.
Retrospectively, we see this project in earlier writings as weil,
in de Man's discussions of Heidegger, and in bis critique of the
"salvational poetics" wbich sees poetic imagination as a way of overcoming contradictions, and of the "naive poetics" wbich "rests on the
belief that poetry is capable of effecting reconciliation because it provides an immediate contact with substance through its own sensible
form" (BI, p. 244). The critique of aesthetic ideology does not somehow
undo or compensate for collaboration, and it is important not simplistically to conflate aesthetic ideology and fascism, but to distinguish, as
Ortwin de Graef does in bis careful discussion, between aesthetic ideology and specific political positions. But we can say that much of de
Man's mature work is staked on the premise that close reading attentive
to the working of poetic language will expose the totalizations undertaken in the name of meaning and unity.
The late essays in Aesthetic l deolo gy, as Marc Redfield has explained, find in Kant's work on "aesthetic" a critique of the ideology of
the aesthetic developed, for instance, by Schiller and applied, or misapplied, both in humanistic conceptions of aesthetic education and in fascist conceptions of politics as an aesthetic project. Traditionally, the
aesthetic is the name of the attempt to find a bridge between the .phenomenal and the intelligible, the sensuous and the conceptual. Aesthetic
objects, with their union of sensuous form and spiritual content, serve as
guarantors of the general possibility of articulating the material and the
spiritual, a world of forces and magnitudes with a world of value. Literature, conceived here as not as literary works but as the rhetorical character of language revealed by close reading, "involves the voiding rather
than the affinnation of aesthetic categories" (RT, p. 10). So, for example,
the convergence of sound and meaning in literature is an effect wbich
language can acbieve "but wbich bears no relationsbip, by analogy or by
ontologically grounded imitation, to anything beyond that particular effect. lt is a rhetorical rather than an aesthetic function of language, an
identifiable trope that operates on the level of the signifier and contains
no responsible pronouncement on the nature of the world - despite its
powerful potential to create the opposite illusion" (RT, p. 10). Literary
theory, in its attention to the functioning of language, thus "raises the
question whether aesthetic values can be compatible with the linguistic
structures from wbich these values are derived" (RT, p. 25). Literature it271

self raises this question in various ways, offering evidence of the autonomous potential of language, of the uncontrollable figural basis of
forms, which cannot therefore serve as the basis of reliable cognition, or
as de Man argues in the essay on Kleist in The Rhetoric of Romanticism,
allegorically exposing the violence that lies hidden behind the aesthetic
and makes aesthetic education possible.
De Man's essay "Kant and Schiller" concludes with a quotation from
a novel by Joseph Goebbels, already cited by Barbara Johnson and Marc
Redfield, which casts the leader as an artist working creatively on his
material: ''The statesman is an artist too. The leader and the led ('Fhrer
und Masse') presents no more of a problem than; say, painter and color.
Politics are the plastic art of the state, just as painting is the plastic art of
color. This is why politics without the people, or even against the people,
is sheer nonsense. To shape a People out of the masses and a State out of
the People, this has always been the deepest intention of politics in the
true sense." De Man's argument is that this aestheticization of politics,
which seeks the fusion of form and idea, is "a grievous misreading of
Schiller's aesthetic state," but that Schiller's conception is itself a similar
misreading, which must be undone by an analysis that takes us back to
Kant. Kant had "disarticulated the project of the aesthetic which he had
undertaken and which he found, by the rigor of his own discourse, to
break down under the power of his own critical epistemological discourse"15. We discover here another instance of the structure Neil Hertz
has discussed, of later readings revealing that the original was already
disarticulated. De Man seeks to demonstrate how the most insightful literary and philosophical texts of the tradition expose the unwarranted violence required to fuse form and idea, cognition and performance.
This aspect of de Man's work has already begun to receive attention,
for instance in Christopher Norris's book Paul de Man: Deconstruction
and the Critique of Aesthetic ldeology, andin Marc Red:field's work of
which we have an excerpt here. Norris, for all the strategic importance of
his early exploration of this aspect of de Man, seeks to assimilate de Man
to Adorno, and, as Marc Redfield argues in a review of Norris and J.
Hillis Miller entitled, "Humanizing de Man," Norris displays scant understanding of the "materiality" ofthe letter or of inscription which in
such essays as "Phenomenality and Materiality in Kant" is what resists
15 De Man, "Kant and Schiller," Aesthetic Ideology, ed. Andrzej Warminski, Minneapolis: University of Minnesota Press, forthcoming.


transfonnation "into the phenomenal cognition of aesthetic judgment"16.

A good deal of work remains to be done on this difficult topic of the prosaic materiality of the letter or inscription and its relation to bistory and
the aesthetic, as when de Man writes, for instance, that "the critique of
the aesthetic ends up, in Kant, in a formal materialism that runs counter
to all values and characteristics associated with aesthetic experience, including the aesthetic categories of the beautiful and the sublime"17. That
fonnal materialism of the letter or inscription when considered non-teleologically - that is not as sign but as blank, indetenninately significative
mark - is a puzzling concept, what de Man calls on the one band "all we
get" yet on the other ~and impossible to experience as such, except as
what gets transfonned when we confer sense and meaning. Despite successful moves in a few recent essays in explicating the critique of aesthetic ideology, there is much more work to be done.
De Man's writing grants great authority to texts - a power of illumination
wbich is a power of disruption - but little authority of meaning. This
highly original combination of respect for texts and suspicion of meaning will give bis writing a continuing power, though its effects are not
easily calculable. His essays commit themselves to major literary and
pbilosophical works for their relentless undoing of the meanings that
usually pass for their value. His cumbersome yet memorable writing,
with its tone of authority and elusive yet resonant key tenns, effectively
teaches suspicion of meaning and "the danger of unwarranted hopeful
solutions," wbile demanding (in a paradox Barbara Johnson discussed),
as the price of possible insight, a commitment to the authority of the text.
Especially important is de Man's insistence that we not give into the
desire for meaning, that reading follow the suspensions of meaning, the
resistances to meaning, and bis encouragement of a questioning of any
stopping place, any moment that might convince us that we have attained
a demystified knowledge. This frequently puts us in an uncomfortably
precarious situation, precisely at what might seem a programmatic moment. "More than any other mode of inquiry, including economics," de
Man writes, in a sentence quoted by Barbara Johnson, "the linguistics of
literariness is a powerful and indispensable tool in the unmasking of ideological aberrations, as weil as determining factor in accounting for their
16 Marc Redfield, "Humanizing de Man," Diacritics, 19 (Summer 1989), p. 35.
17 De Man, "Kant and Schiller," Aesthetic ldeology, cit..


occurrence." That fonnulation - "detennining factor" - seems to me to

carry a waming: while exploring possible links between de Man's thinking and the resources of other contemporary theoretical discourses, such
as psychoanalyis, feminism, and marxism, we ought to remain alert to
the possibility that the tools for unmasking may also be determining
factors, factors that determine and thus help account for ideological aberrations. As so often with de Man, one cannot be sure whether this formulation is a subtle waming or a grammatical ambiguity. Tue linguistics
of literariness is an important factor in accounting for ideological aberrations but to call it a determining factor - may this not suggest that it
determines them and accounts for them because it produces them, as
weil as helping to analyse and explain them? As so often, when con.fronted with the indetenninately significative dimensions of language on
which we cannot but confer sense and meaning, we are left with that
more than grammatical problem.

Quelles sont les contributions les plus importantes de l'oeuvre de Paul de Man pour
l'avenir de la theorie et de la critique litreraires? Six domaines ou cette oeuvre apporte
non seulement un acquis mais aussi un prograrnme ou une problematique pour la
recherche et pour la reflexion theorique peuvent etre identifies: (1) la theorie de
l'allegorie et son rapport avec l'histoire; (2) la revalorisation du Romantisme, dont
l'interpretation est une tache des plus necessaires et des plus difficiles; (3) l'explication du
rapport entre "blindness" et "insight"; (4) le developpement d'une theorie marerialiste du
langage; (5) une reflexion sur la nature de l'histoire dans ses relations avec l'imagination
poetique et le probleme du rapport entre la connaissance et l'evenement; (6) la critique de
l'ideologie esthetique, entamee dans ses demiers ecrits.



Jonathan Culler: Cornell University, Society for the Humanities, USAIthaca, NY 14850
William Flesch: Brandeis University, Department of English, USAWaltham, MA 02254
Hans-Jost Frey: Bergstrasse 18, 8044 Zrich
Reiner Weidmann: Im Hgeler 8, 8910 Affoltern am Albis
Norbert Gabriel: Gennanistisches Seminar der Universitt Bonn, Am
Hof 1 D, D-W-5300 Bonn
Ortwin de Graef: Mindebruderstraat 1, B 42, B-3000 Leuven
Peter Grotzer: Seminar fr vergleichende Literaturwissenschaft der
Universitt Zrich, Plattenstrasse 47, 8032 Zrich
Christiaan L. Hart-Nibbrig: Chemin des Pierrettes 20, 1025 SaintSulpice
Neil Hertz: Johns Hopkins University, Humanities Center, USABaltimore, MD 21218
Peter Hughes: Englisches Seminar der Universitt Zrich, Plattenstrasse
47, 8032 Zrich
Barbara Johnson: 61 Fresh Pond Parkway, USA-Cambridge, MA 02138
Laura Quinney: Wellesley College, Department of English, USAWellesley, MA 02181
Marc W. Redfield: Tue Claremont Graduate School, Department of
English, 170 E. Tenth Street, USA-Claremont, CA 91711
Claude Reichler: Rue des Channilles 5, 1203 Geneve



Henry H. H. Remak

The Situation of Comparative Literature in the Universities

Ram6n Sugranyes de Franch Realite de l'au-def et illustrations des sens. Vision de la

culture baroque
Liliane R. Furst

"The Happiest Days your Life"

Renate Bschenstein

Anmerkung zum Artikel von Lilian R. Furst

Ernest Giddey

Isabelle de Montolieu et Jane Austen

Peter Grotzer

Narr, Gaukler, Hungerknstler als Allegorie des


Jrgen von Stackelberg

Die deutsche Edelrose. Anmerkungen zur bersetzung von Umberto Ecos Norne del/a Rosa


Ein Brief von Max Frisch zum Don Juan-Thema


Yves Chevrel

Litterature comparee et tradition classique: situation

des etudes comparatistes en France

Approches du texte litteraire - Wege zum literarischen Text - Interpretazioni

de/ testo /etterario:

Peter Grotzer


Philippe Jaccottet
Jean-Luc Seylaz

Au petit joun
"Au petit jour". Lecture d'un poeme de Philippe

Jacques Geninasca

La clairvoyailte attente de !'ignorant

Conrad Ferdinand Meyer

Werner Stauffacher
Christiaan L. Hart-Nibbrig

Lyrisches lcij., ins Bodenlose starrend

Torquato Tasso
Guglielmo Gorni
Georges Gntert

C.F. Meyers :,,Mwenftug" oder der Abstand des

Textes zu sich selbst. Ein Annherungsversuch
Da/ Canto XII del/a "Gerusalemme Liberata ":
11 chiasmo d~ Clorinda
11 combattiniento di Tancredi e Clorinda


Actes du Colloque sur la traduction litteraire

(Universite de Lausanne, 31 mai et 1er juin 1985)
Walter Lenschen



De Chamonix a Courmayeur.
(Que veut dire l'impossibilite de traduire ?)

Friedhelm Kemp

Form durch Freiheit. Ein Pldoyer

Hans-Jost Frey

Die Beziehung zwischen bersetzung und Original

als Text

Giorgio Orelli

Tradurre poesia

Madeleine Santschi

La traduction, corps physique: a partir d'une experience de traduction de Pasolini

Marcel Schwander

Westschweizer Literatur in deutscher bersetzung

Eugene Guillevic

Traduire la poesie

Wolfgang Hildesheimer

Der Autor als bersetzer - der bersetzte Autor


La parola mancante. Traduzione del Mozart di


Guiseppe Bevilacqua

Traducendo Celan postumo

John E. Jackson

Traduire Celan: raisons d'un echec

Traugott Knig

Probleme der Rckbersetzung deutscher philosophischer Begriffe im Werk Sartres

Jean-Rene Ladmiral

Traduire les philosophes allemands


Harald Pricke

Zur Rolle von Theorie und Erfahrung in der Literaturwissenschaft

Rita Mill er-Isella

11 ,,campo associativo" comme metodo di confronto

testuale nel processo critico della traduzione letteraria

Verena Ehrlich-Haefeli

Secularisation, langue et structure familiale: le pere

dans le thetre de Lessing et de Diderot

Markus Winkler

Madame de Staels Bemerkungen zum Idyllischen in

Literatur und Leben der Deutschen

Susanne Wehrle

bersetzbarkeit und Unbersetzbarkeit der Lyrik.


Akten der Studientagung ber das literarische Paradox

(Universitt Zrich, 28. und 29. November 1986)
Peter Grotzer

Paradoxa in Litteris

Andnis Horn

Zur Paradoxie der Metapher

Jean-Jacques Marchand

Le discours paradoxal dans le Prince de Machiavel.

Caracteristiques et fonctions

Pier-Giorgio Conti

G. Leopardi, L'infinito, V.15 e dintorni o: del paradosso

come figura e come principio

Charles Mela

Un paradoxe litteraire: le lai du Lecheor

Eleonore Frey

Das Paradox des Unsglichen bei Georg Trakl

Fritz Gutbrodt

"Tue words are wild": Das Paradox der Wortwildnis

bei Hopkins


Manfred Gsteiger

Les relations litteraires entre la Suisse alemanique

et la Suisse romande.
Appendice: Les ecrivains contemporains et la Suisse
plurilingue: A propos d'une enquete recente.

Giovanni Parenti

La poesia pastorale come poesia artificiosa. Origine e

fortuna del Summationsschema.

Ernest Giddey

Madame de Stael et le romantisme anglais.

Michele e Antonio Stuble

La Gran Bretagna del Settecento vista da un

viaggiatore italiano.

Marc Elikan

Les langages dans le Pantagruel de Rablelais chapitre IX.


Peter Grotzer

Imagologie: Problemes de la respresentation litteraire -

Studientagung der Schweizerischen Gesellschaft fr

Allgemeine und Vergleichende Literaturwissenschaft
vom 29./30. Juni 1987 in der Hochschule St. Gallen.

a l'imaginaire

Daniel-Henri Pageaux

De l'image

Hugo Dyserinck

Zur Entwicklung der komparatistischen Imagologie

Henri Quere

Le Spectre de la figure

Gustav Siebenmann

Das Lateinamerikabild der Deutschen - Quellen,

Raster, Wandlungen.

N 8, 1988

Yves Giraud

La Belle Matineuse. Histoire d'un theme poetique.

Sophie Megevand

Un phenomene original: trois adaptations simultanees

d'une piece espagnole.

Maura Formica

Das Mrchen im Mrchen. Zur Rezeption von

Gian Battista Basiles Pentamerone in Clemens
Brentanos Italienischen Mrchen.

Alfred Sprde

Zwei Projekte ,,moralistischer" Literatur. Die Bedeutung des Aphoristikers und Sprachphilosophen
Ludwig Wittgenstein fr den Schriftsteller Aleksandr

Jean-Yves Pidoux

Grandeur nature.

Marc Elikan

XIP congres de l'ICLA/AILC

a Munich


Je3:n Malaplate:

Un plagiat ignore: la traduction de Faust par Gerard de


Peter Schnyder

Nietzsche in Frankreich: Aspekte seiner Wirkungsgeschichte

Mark Elikan

Un aspect du probleme de la traduction litteraire:

l'exemple d'Isaac Bashevis Singer


Antonio Lara

Gongora - Darmangeat - Jaccottet: La traduction des


Maria Deppermann

Protest und Verheiung: Zur Bedeutung Friedrich

Nietzsches fr die Kultur des Fin de siecle in

Jean-Marc Moura

Une etude d'imagologie: L'image du tiers monde

mena9ant dans le roman fran9ais contemporain