Nothing but the Worst

Michael Wood

  • The Paul de Man Notebooks edited by Martin McQuillan
    Edinburgh, 357 pp, £80.00, April 2014, ISBN 978 0 7486 4104 8
  • The Double Life of Paul de Man by Evelyn Barish
    Norton, 534 pp, £25.00, September 2014, ISBN 978 0 87140 326 1

‘How often in my life have I said those words, and yet?’

John Banville, Shroud

‘I had jumped,’ Conrad’s Jim says of his abandonment of his ship, adding a moment later: ‘It seems.’ Marlow, the narrator of the novel who is listening to Jim’s story, says: ‘Looks like it.’ This is one of many instances where Marlow’s language is drier and tougher than his thought, which is unwilling to condone Jim’s act but very sympathetic to the difficulty of living with the memory of it. ‘And in what was I better than the rest of us,’ he says, ‘to refuse him my pity?’ But he does refuse it in words, again and again. A little later Jim claims: ‘There was not the thickness of a sheet of paper between the right and the wrong of this affair.’ Marlow murmurs: ‘How much more did you want?’

Jim can’t forget his guilt and can’t face it either, and Marlow arrives at a philosophical conclusion: ‘The truth seems to be that it is impossible to lay the ghost of a fact.’ Marlow can’t have known that he was sketching a theory of deconstruction as it was later elaborated by Paul de Man, and he was certainly hesitant enough about his proposition. But de Man’s suggestion, in Allegories of Reading, that ‘excuses not only accuse but they carry out the verdict implicit in their accusation’ is surely a version of Marlow’s thought. The more we pretend the ghost is gone or not really a ghost at all, the more insistent the haunting gets. We can see it in Jim’s language. Not ‘I jumped’ but ‘I had jumped.’ Even in narration he avoids the moment, and the ghost sneaks into something as simple as a change of tense.

But then de Man seems to contradict himself. ‘Excuses generate the very guilt they exonerate, though always in excess or by default … any guilt … can always be dismissed as the gratuitous product of a textual grammar or a radical fiction: there can never be enough guilt around to match the text-machine’s infinite power to excuse.’ Always? Any guilt? Never? The text-machine is the ghost’s busy attendance on us; the fact becomes whatever the ghost says it is. There is a bit of confusion here but not necessarily a contradiction. The argument that excuses are accusations appears to be de Man’s own, a version of Paul Valéry’s dictum that to confess is to lie. The argument that excuses generate guilt seems to be de Man’s voicing of Rousseau’s writing practice – he is talking about the Confessions. All excuses accuse, but a crafty confessor will allow them to proliferate so that even real guilt becomes improbable because there is too much of it.

But what do we make of the passage that leads to the first, stronger claim about excuses? The claim turns out to be only half the story.

It is always possible to face up to any experience (to excuse any guilt), because the experience always exists simultaneously as fictional discourse and as empirical event and it is never possible to decide which one of the two possibilities is the right one. The indecision makes it possible to excuse the bleakest of crimes because, as a fiction, it escapes from the constraints of guilt and innocence. On the other hand, it makes it equally possible to accuse fiction-making … of being the most cruel [of activities] … Excuses not only accuse but …

Something very strange is happening here. A lucid, impersonal account of one of the relations between language and reality, or between ghost and fact, slips into a dubious proposition about what is ‘never possible’. It’s true that we can always pretend the ghost doesn’t exist – that’s why it’s only a ghost – and true too that the ghost is always likely to come back and make more virulent accusations. But how could it be impossible to decide which is happening at any given time? We can make excuses for the bleakest of crimes, which is perhaps what de Man means, but we can’t simply excuse it, even in fiction.

The deep interest of de Man’s criticism – it is in the end less theoretical than speculative, a form of close reading full of philosophical questions – lies in its attention to the chance of uncertainty where we are most convinced of having got rid of it. In rhetorical questions, for example. We know they’re not literal or straight – that’s what the phrase means. But what if they were literal, what if their grammar overcame their usage? I can say, ‘What do you mean?’ to signify ‘How dare you?’ or ‘What are you referring to?’ or ‘What is the sense of that statement?’ You will understand me through the context of the remark, but contexts can shift. De Man’s leading cases come from the television show All in the Family and Yeats’s poem ‘Among Schoolchildren’. Archie Bunker says, ‘What’s the difference,’ meaning, ‘Who cares,’ and Yeats asks, ‘How can we know the dancer from the dance?’, meaning we can’t. But what if we really wanted to know the difference, as Archie doesn’t? De Man is right to say that literal meanings always dog rhetorical practice, and vice versa. The meaning that isn’t there remains, in its silent way, a part of the meaning that is there, and this state of affairs has all kinds of consequences for our thinking about reading and interpretation. But it is not true that we cannot ‘in any way make a valid decision as to which of the readings can be given priority over the other’. ‘Decision’ here plays the role of ‘never possible’ in the argument about excuses, it mixes high theory and daily practice. It’s true that we may not be able to make a final decision, or any decision at all in the abstract, but we can’t understand the sentences in the abstract either. The fine point of deconstruction is not that we can’t decide but that we have to keep deciding. In Yeats’s case, for example, we can tell the dancer from the dance because we know how much hard work has gone into the performance, and Yeats’s poems are full of thoughts along these lines. But in the perfection of the achievement we may forget the hard work and even the person, and that is what the question means in its rhetorical mode. The dancer is the dance – for now.

De Man’s enemy is our habit of submitting ‘uncritically to the authority of reference’, of assuming that language always points away from itself and points clearly. We are devoted to ‘the myth of semantic correspondence between sign and referent’, caught up in ‘the fallacy of reference’. These are all grand overstatements, as de Man himself indicates when he says, ‘the notion of a language entirely freed of referential constraints is properly inconceivable,’ but they have a job to do, and they allow us to see why de Man thinks that ‘far from being a repression of the political, as Althusser would have it, literature is condemned to being the truly political mode of discourse.’ Literature in this view is what helps us to think twice, a confirmation of Brecht’s view that all thought is on the side of the oppressed. Or at least is against the myths and fallacies the authorities want us to believe are true. ‘There should be no fundamentalists in criticism,’ de Man says.

De Man’s writing is at its most subtle and inventive when raising these questions; at its weakest when it lapses into what I think of as a form of intellectual elegance, the too tidy or too lofty closing note, a version of the finality he himself works so hard against. This is where interesting difficulties become radical impossibilities, as we have already seen, and further examples are everywhere. As in de Man’s notion that the deconstruction of the word ‘nature’ has to be ‘endless’, to create ‘an eternally repeated pattern of regression’. Why would it be endless, wouldn’t a few strategic strikes do most of what needs doing? Or in his translation, noted by Stanley Corngold, of Nietzsche’s phrase ‘error takes place’ or ‘error happens’ (‘der Irrthum stattfindet’) as ‘error reigns.’ Nothing but the worst, or the most extreme.

We could also think of de Man’s rather beautiful evocation of Peirce’s ‘so suggestively unfathomable definition of the sign’. You have to like the unfathomable a lot to write like that. Evelyn Barish catches this note very well in an awkward but telling mixed metaphor. De Man’s ‘approach to literature … took as its bedrock a subversive doubt concerning the truthtelling power of language’. Even subversive doubt will become a bedrock if you hold to it steadily. But then it isn’t doubt, or subversive. Barish herself is struggling bravely to sort out two claims: that deconstruction for de Man was the theory of an ‘ability to deny’, that it was a matter of ‘dealing by denial with the valency of past, compromising acts’; and that de Man’s ‘practice insists that … there is no constructive way to dissolve painful memory’. No deconstructive way either.

Writing in 1979 de Man regrets that the word ‘deconstruction’ has become ‘a bone of contention’, or ‘a war cry’, as he says elsewhere, because ‘no other word states so economically the impossibility to evaluate positively or negatively the inescapable evaluation it implies.’ A lot of labour (and an expert control of irony) has gone into that sentence, but surely the impossibility and the inescapability are both too much. It’s worth recalling that ‘deconstruction’ is not an old word, not a synonym for ‘destruction’ or ‘demolition’ and that for a long time it was spelled with a hyphen. But then the identification of a failure to evaluate an evaluation is not an act of deconstruction, it is a shrewd and useful observation (this kind of failure does indeed happen all the time) leading to a perfect conceptual impasse, the zone of the impossible and inescapable. That’s what I am calling elegance. De Man is less elegant but clearer, in a 1984 interview reproduced in the Notebooks volume: ‘There is a negative moment in [deconstruction], as there is a negative moment in any critical reading that is not simply, shall we say, nihilistic. I don’t want to be too sanguine about this. However, we are doing something positive.’

De Man didn’t invent deconstruction and never said he did. He said the opposite: ‘I consciously came across “deconstruction” for the first time in the writings of Jacques Derrida’; ‘Derrida used the term and put it on the map.’ And still less did he, in Evelyn Barish’s words, ‘create a new philosophy, a way of looking at the world that redefined America’s cultural point of view’. But several of the writings collected in the Notebooks indicate that he did intuit very early the critical motion characteristic of deconstruction. In a remarkable 1960 essay on the role of symbolism in 20th-century thought about literature – by ‘symbolism’ he means not the relatively minor organised movement of that name but pretty much everything we associate with Baudelaire, Rimbaud and Mallarmé – he talks about the quest for the autonomy of artistic form, and says we can perceive this theme only ‘if we place ourselves both inside and outside the form, if we can feel both the persistent need for its construction and its inevitable undoing’. This is 17 years before de Man would have read the word ‘deconstruction’ in Derrida. There is even something of the later reaching for a point of no return in the word ‘inevitable’. We can track too, in this early work, a very broad genealogy for what would become a recognisable and influential style of thinking. A man experienced in literary journalism and trained in the French method of explication de texte meets the American New Criticism. He writes rather negatively of this meeting in his book Blindness and Insight (1971), but in the Notebooks interview he is more grateful:

Then in this country I was exposed to a new kind of reading, a much closer mode of reading, basically a new critical mode of reading. I had always had an attraction for, and was interested in, the writings of the French critics in the moment of close reading that seemed to raise certain problems that they were not expressing all too successfully.

Add a touch of Heidegger to this, and you begin to get something of the deconstruction that de Man taught so well. ‘A correct critical vocabulary has to be philosophical before it can become historical.’ We should note too that since the text I have quoted is a transcription of an interview, it can’t distinguish, or can distinguish only too clearly, between a new critical mode of reading and a New Critical mode of reading.

If we were not to talk of elegance in de Man’s writing, we might talk of obliquity. This was a word he himself liked, and he felt philosophers were too uncomfortable with it. ‘That things are not as they seem to be,’ a student note on one of his seminars reads, ‘has been more available in literature departments.’ And on Hegel: ‘If you go to the Aesthetics directly, you will be confused. I’m not going to de-confuse you. I’m going to give you a more oblique reading.’ Martin McQuillan, editor of The Paul de Man Notebooks, has certainly taken this tip. These are not, as the title might suggest, Paul de Man’s notebooks. A set of such books exists relating to a series of seminars de Man taught in Zurich, Cornell, Johns Hopkins and Yale from 1963 to 1983, and is in the library at Irvine. They are listed by their names and (mostly) dates on pages 309 to 314 of this volume. None of them is reproduced here, although we do learn that the book ‘takes the seminar notebooks as metonymic of wider archival schema’. What’s the difference? Didn’t we always know that Archie Bunker was wrong to take the question so lightly? Actually, once we have got over the slight preciosity of the idea, it makes very good sense. This selection of essays, translations, academic reports, book outlines and notes for classes at Yale rather than in Zurich gives us a strong sense of a mind in process, looking not towards the inevitable monograph but in various different directions at once, as its curiosity takes it. The essay on symbolism, for example, looks like an excellent start on a book on this subject. But it led de Man to Romanticism and Rousseau, long preludes to the symbolist music he had seemed to be after.


Paul de Man was born in Antwerp in 1919 and died in New Haven in 1983. He came from a well-to-do but none too happy family – his mother committed suicide in 1937, his brother was a rapist who died at the age of 21, hit by a train. He excelled at school in Brussels and flopped at the university. He didn’t like either engineering or political science, but he seems also to have developed a kind of phobia for formal scrutiny. Much later he managed to get out of most of the examinations for the doctorate at Harvard, and to fail an elementary test that much less gifted students could hardly fail if they tried. During the war he wrote a great deal of literary journalism for a Nazi-controlled newspaper in Brussels, including plenty of celebrations of German virtues (as distinct from French decadence) and at least two pieces where the anti-Semitism was unmistakable. The most interesting element in the longest of these pieces is the mention of Kafka – spelt Kafha, but that is probably a misprint – along with Gide, Hemingway and Lawrence as happily exempt from Jewish influences. Did he not know that Kafka was Jewish? Was he using some sort of high modernist argument that if you were any good you wouldn’t be Jewish any more? Or was he secretly baiting his employers? It is true that his sly piece appeared on a page that was full of crass and obvious expressions of hatred and prejudice. I’m not sure slyness is much of an excuse, but de Man’s argument about excuses is way ahead of me anyway.

After the war de Man founded a publishing company and stole virtually all of its money. This is the episode in his life that baffles his biographer most, as well it may. What did he want the money for? Why was he so intent on ruining his own reputation? It is true that even in his later life, if Barish is right, one of his ways of saving money was not to pay the rent, but the Belgian episode goes well beyond this. When the trial ended, long after de Man had taken off for America, he ‘was sentenced in absentia … to a total of five years in prison, plus heavy costs and fines’. Barish notes de Man’s extravagance repeatedly (‘He never had enough money’; he couldn’t satisfy ‘his own extravagant needs’; he ‘could never manage his spending’) and his then wife seems to have been pretty demanding too. But none of this catches the compulsive note of the stealing and refusing to pay. The obvious thought is that de Man must have needed the money for drugs or to respond to blackmail, but this seems not to have been so. We are left with the idea of some kind of temporary, all too clever pathology, and this is where Barish settles: ‘Perhaps at some level, he also knew that when the day of reckoning came, while it would mean disgrace, he would be free. One day he would simply fall out of the orbit of the bourgeois world and disappear. It would all be over.’

There was plenty of deception in the later life, it seems, but nothing as far out as this. De Man in America was something else, a character out of Melville or Thomas Mann, suave, charming, unreliable, resilient, always walking on some kind of moral and financial tightrope. He claimed that his well-known uncle, Hendrik de Man, a socialist who became a National Socialist, was his father, and he invented several publications well in advance of their (non-) appearance. He taught at Bard College, got his doctorate at Harvard in spite of various visits from the Immigration and Naturalization Service regarding his status as a visitor and a mysterious letter to the university administration denouncing him for his wartime activities; held senior positions at Cornell, Johns Hopkins and Yale. He also found his way to the late critical style I discussed, and to a real eminence in the American academy.


Barish has some difficulty in establishing her tone and approach. Much of the time she wants to prove that de Man was an idol with feet of clay, but for that she has to make him more of an idol than he was, ‘a cultural giant of epic proportions’. ‘To have been his student,’ she says, ‘virtually assured a candidate of one of the few good university jobs available.’ Not even Midas had such a Midas touch, even in the bad old days of the timely phone call. To say nothing of the fact that in those quarrelsome times being de Man’s student could just as easily have disqualified you from getting a job. I don’t think he seemed ‘to personify the highest … ethical achievement’ either, not because anyone suspected his ethics, but because ethics was not part of the critic’s portfolio, any more than it was of the poet’s. But Barish also thinks de Man is now forgotten, ‘no longer seems to exist’, so she has to resurrect him in order to bury him again.

He’s not forgotten, any more than he was a giant. Quarrels still arise almost every time his name or his legacy comes up. But as the book goes on – it has been, Barish says, twenty years in the writing – her patience and research overcome these simplifications, and she manages to be truer to her surprise than to her moralising. ‘Initially,’ she says, ‘I found it impossible not so much to believe what I was learning as to comprehend it.’ At its best, this book signals its continuing failure to comprehend but recounts the details of the learning very well. Writing of de Man’s lack of interest in his first son, who ‘forgave his father’, Barish says: ‘whether de Man forgave himself, or indeed whether he ever blamed himself, is impossible to know.’ Not only is it impossible to lay the ghost of a fact, it is impossible to know how the ghost talks to other people.

Barish, like many other writers, wants to connect de Man’s late work to his early career, and especially to his collaboration with the Germans. The most common line in 1987, when the old articles were resurrected, was to say they proved that deconstruction and indeed any sort of fancy literary theory was bad news, because it attacked the very idea of the truth – as if the truth was always on the side of virtue, and wouldn’t be in any trouble if those annoying sceptics didn’t keep showing up. De Man’s defenders thought his persistent suspicion of the defections of language was a form of atonement, a way of teaching us not to do what he did.

I remember reading the passages I quoted at the beginning of this piece, about excuses and the bleakest of crimes, quite some time before there were any revelations about de Man’s past, and wondering what this man had done to know so much about the intricacies of apology. Now I feel the weight of history in those sentences all the more, but I can’t tell whether he was with Conrad’s Marlow or against him. Barish is surely right to say ‘there is a profound connection between the man who secretly fled from Belgium … and the one we knew for generations later as our intellectual and cultural leader,’ even if ‘leader’ isn’t quite the right word. But what was the connection? Did de Man think he could divert the ghost of a fact (the ghost of many unappealing facts) by analysing its ghostliness? Had he got to a stage where he knew the ghost wouldn’t leave but wouldn’t do him any harm either? Or did he think the ghost really was laid, and he was free to use past experience for his study of the elegant, purely theoretical certainties of error?