None of the contributors to Responses can take the venom out of that passage. By comparison with many other instances of anti-Semitism in European writers in the Thirties and Forties, it is mild, but there is little merit in measuring degrees of malice. Nor is this article the only wounding one: seventeen months later, de Man published much the same view, though in milder terms. Faced with these articles, Derrida tries to express and to relieve his distress by saying that “one must have the courage to answer injustice with justice,” i.e., justice to de Man. He finds some comfort in the fact that de Man starts the more offensive article by mocking “vulgar anti-Semitism” (“L’anti-sémitisme vulgaire…“), meaning Céline, I think. But Derrida’s attempt here seems to me hopeless. By deriding vulgar anti-Semitism, de Man is merely laying claim to a high-minded form of the sentiment, according to which Jews are of little significance in any consideration: he is not disavowing the prejudice.
Derrida quotes from Henri de Man’s Au delà du marxisme (1926) a distinction between “pure Marxism” and “vulgar Marxism,” the former a “dead truth,” the latter “a living error.” “I despise all forms of vulgarization,” Henri de Man says, “of truth put within reach of those who prefer ersatz goods, radio and phonograph music, champagne for democratic banquets.” But this doesn’t help, it merely shows that a distinction between ostensibly high-minded and obviously low-minded forms of an ideology was in Paul de Man’s vicinity.
In fact, none of Paul de Man’s wartime sentiments was original. The article (Le Soir, January 20, 1942) in which he ridiculed the French writers who dreamed of saving man before saving the world (“qui, tous, se preoccupent de sauver l’homme avant de sauver le monde“) could be compared with many sentiments of the day that fostered collective salvation (“des réalisations collectives“) at whatever cost to particular persons. The only notable feature of de Man’s articles is their self-confident, unrueful tone, as of a young man who believes that history has chosen him to speak its truth.
One can only speculate why de Man stopped writing in March 1943. He later claimed that the Germans were turning the screw on free speech. But he may also have decided, after the entry of the United States into the war, the surrender of the German army at Stalingrad on February 2, and the successes of the British army in North Africa, that he had misread the future: Germany wasn’t going to win, after all.
Responses gives many of the arguments on both sides of the dispute. The most vigorous adepts of denunciation are John Brenkman, Stanley Corngold, and Jeffrey Mehlman: what they have to say may be added to indictments voiced elsewhere; by Frank Lentriccia, for instance, who has been quoted as saying that de Man’s theory of reading is a stratagem for dissolving historical evidence and nullifying its moral consequence; by Tzvetan Todorov, who has maintained (Times Literary Supplement, June 17–23, 1988) that “there is no real discontinuity between the antisemitism and pro-Hitlerism professed by de Man during his youth and his ideological options”; and by Jon Wiener, who has referred to de Man (The Nation, January 9, 1988) as “something of an academic Waldheim.”
These writers speak for themselves; my paraphrases would be misleading. But the case for the prosecution, roughly summarized, is that de Man was from the start of the war a willing collaborator. His opponents also argue that the theory of reading which he practiced in his later years was an elaborate alibi, a device to evade personal responsibility by undermining the very terms in which moral responsibility may be incurred and denoted. I am surprised that Samuel Beckett has not yet been mentioned in the debate—a writer whose work in the French Resistance, for which he was later awarded a medal, testifies to a quite different way of responding to the war.
It would be equally misleading to attempt a paraphrase of the defense. Some critics, notably J. Hillis Miller and Geoffrey Hartman, bear witness that in many years of friendship they never saw the slightest sign of anti-Semitism in de Man. Miller also claims that de Man’s wartime writings, however jejeune and deplorable in their anti-Semitic character, are the folly of a young man, and that they have nothing to do with his mature work. Derrida makes much of Henri de Man’s influence upon his nephew, but neither Derrida nor anyone else has provided strong evidence that Henri de Man had much influence after Paul went to work for Le Soir in December 1940.
I have spoken of this episode as if it were a case in law. In a sense it is, even though the defendant is dead and can’t speak for himself. Derrida and other critics have asked that the matter should not be discussed as a case of crime and punishment. It would be agreeable if someone could propose an alternative to the idiom of prosecution and defense, but I can’t see how this is possible.
The second question raised in Responses is clearly the more tendentious one: Is de Man’s wartime writing continuous with the form of Deconstruction he practiced in his later work, or irrelevant to it?
When we think of de Man, we think of Deconstruction, but it is not self-evident that we should. Now that we have his entire work, collected and uncollected, at hand, we see that he was in many respects a man of his time. He lived through the heyday of phenomenology, existentialism, the disputes between Sartre and Camus, the rise of French Structuralism, and the questioning of Structuralism that led to post-Structuralism and Deconstruction. He did not arrive at Deconstruction as by one leap from Brussels. Traces of postwar sentiment, enthusiasm, disillusionment characteristic of the period can be found in his essays. If there is a crucial turning point in his intellectual career—as I think there is—it comes with his reading of Heidegger; or rather, with his reading of Heidegger’s reading of Hölderlin. If we need a date, I would settle for 1954, although de Man had by then read Heidegger’s Being and Time and Letter on Humanism.*
When de Man met Derrida, at a colloquium at Johns Hopkins in 1966, the turn toward Deconstruction, a philology that takes linguistic failure for granted, had already been taken. But it was important for de Man to find that he and Derrida were thinking about Rousseau in ways similar as well as different. Derrida and de Man now seem different more than alike; Derrida is far more playful than de Man, though equally serious in and beyond his play with words.
It is true, in any event, that de Man’s final position was that of Deconstruction; final if only because he did not live long enough to do more than practice it; how he might have gone beyond it, we will not know. Some of the constituents of his Deconstruction were already commonplaces in Structuralism: the removal of the personal subject—call it consciousness—from the center to the periphery of experience; or rather from the center, where it is invoked as will or voice, to the periphery, where it is a mere function or consequence of the linguistic codes it is deluded enough to think it commands.
So far as Structuralism is an attack on the axioms of bourgeois liberalism, especially upon liberalism’s complacent gratification in the possession of a self, identity, will, spontaneity, the authentic power of choice—so far, it makes common cause with Deconstruction. But Deconstruction, both in Derrida’s version and in de Man’s, attacks Structuralism for harboring a suppressed nostalgia. Whether in the version of Saussure, or Roman Jakobson, or Barthes, the claims of Structuralists to find largely unintended underlying patterns, or codes, which organize and help give meaning to texts, show a longing for a stability the Structuralists claim to have renounced. According to Deconstruction, Structuralists want stability, however squeamish they are about finding it in a personal source, as in individual consciousness or will.
Deconstruction attacks Structuralism for wanting to come to rest upon the very codes—linguistic, signifying codes—it takes grim pleasure in describing. Structuralism has no quarrel with language, and expresses merely the dismay of one imprisoned in it. Deconstruction has many quarrels with language; not with particular languages, English, French, German, and so forth, but with language as such. Many writers, from Locke to T.S. Eliot and Borges, have denounced language; it is a liar, a deceiver, it bewilders the mind with metaphors, dreams, mere dreams. Words, as Eliot wrote in “Burnt Norton,”
Decay with imprecision, will not stay in place,
Will not stay still.
But Eliot believed that at some congenial moment in the past, perhaps in the first years of the seventeenth century in England, the language was reliable, holding out the possibility, as in Shakespeare and Donne, of unity of experience. De Man’s Deconstruction, on the contrary, refuses to be appeased: if language appears to hold out such a promise of words and experience truly converging, the appearance is a deceit, the promise a trap. We are naive if we expect language to serve our purposes and desires. Deconstruction is the practice, in reading, of one who refuses to be lulled into the complacency of self, into believing in the stability of reference, or in the appearance of a seamless web of meanings. It is a debunking of the delusory tokens of meaning and reference. In one of the essays on Rousseau in Allegories of Reading, de Man writes:
The deconstruction of a system of relationships always reveals a more fragmented stage that can be called natural with regard to the system that is being undone.
This ostensibly “natural” stage then substitutes its own system of relationships for the one it helped to undo. It conceals, therefore, the fact that it is itself merely one system among many: it has no better claim to being natural than the system it undermined. The pretentions of each and every system to be the true one can be endlessly uncovered. But the work of Deconstruction is indeed endless, because each stage or pattern tries to take possession of the entire ground, and must be shown to be specious.
The simplest example of a deconstructive reading is de Man’s account of Yeats’s “Among School Children.” Normally, the last stanza of that poem is taken to be a celebration of a possible unity of being. The condition is represented by the chestnut tree, “great rooted blossomer,” thriving upon the order of natural life, organic unity of leaf, blossom, and bole. The poem culminates in another symbol, that of the dancer, “O body swayed to music.” Again, most readers interpret the last line, “How can we know the dancer from the dance?” as claiming the felicity of not having to know one from the other, since they have converged. As if to say: Isn’t it wonderful that one can reach a state of being, a condition of unity, so complete that it is not necessary to distinguish between the dancer and the dance? But de Man rejects this felicity. He insists that it is not possible to be sure whether the stanza says what most readers think it is saying, or the reverse: that in some circumstances or in some respect one needs to distinguish between the dancer and the dance.
De Man on Heidegger on Hölderlin can't easily be paraphrased, but the rough gist of the matter is that Hölderlin's poetry "is critical of its own certitudes, their illusory character unveiled." De Man concludes that "the poetic word" cannot establish Being; it lives only in the desire of Being, and therefore in its fateful absence. It cannot establish Being, for "as soon as the word is uttered, it destroys the immediate and discovers that instead of stating Being, it can only state mediation." Language cannot reunite what it has separated by naming: Being and Becoming. "Their unity is ineffable and cannot be said, because it is language itself that introduces the distinction."↩
De Man on Heidegger on Hölderlin can’t easily be paraphrased, but the rough gist of the matter is that Hölderlin’s poetry “is critical of its own certitudes, their illusory character unveiled.” De Man concludes that “the poetic word” cannot establish Being; it lives only in the desire of Being, and therefore in its fateful absence. It cannot establish Being, for “as soon as the word is uttered, it destroys the immediate and discovers that instead of stating Being, it can only state mediation.” Language cannot reunite what it has separated by naming: Being and Becoming. “Their unity is ineffable and cannot be said, because it is language itself that introduces the distinction.”↩