The fight over the new Ulysses, like all academic arguments over commas, is a fight between two ideas of human nature, two visions of judgment, two images of eternity*.

On one side of the textual arena Hans Walter Gabler offers an ‘ideal text’ from which all error, especially the author’s, has been burned away in the refining fire of bibliographical theory. On the other side John Kidd champions an imperfect but historically authentic Ulysses, a book produced partly by genius, partly by accident, partly by the exigencies of the printing-shop. In matters of wording and punctuation, the differences between one Ulysses and the other are fairly trivial: the debate is now focused on the question of whether Joyce copied the Christian name of a certain Conolly Norman from a page in Thom’s Directory on which it was spelled with one ‘n’ or from a perhaps more significant page on which it was spelled with two. But the differences between the Joyce who wrote Gabler’s Ulysses and the Joyce who wrote Kidd’s are so extreme that it seems impossible to imagine that each published a book with the same title on the same day.

These Joyces are the product of two rival models of authorship: the fading coal and the gothic tower. Every editor, every reader, more or less consciously assumes that one or the other is the true model of the act of writing. The clearest statement of the fading-coal model occurs in Shelley’s Defence of Poetry: ‘The mind in creation is as a fading coal, which some invisible influence, like an inconstant wind, awakens to transitory brightness ... Could this influence be durable in its original purity and force, it is impossible to predict the greatness of the results; but when composition begins, inspiration is already on the decline, and the most glorious poetry in the world is probably a feeble shadow of the original conception of the Poet.’ This model is flattering to authors. They can assure themselves that their forgotten original conception was a work of genius, and that the faults in the finished work result merely from a lapse of memory. For an editor, this model suggests that authors forget their inspiration as they re-copy and revise their work, and offers an almost irresistible challenge to look beyond an author’s forgetfulness and recover, if not the grand original idea, at least the first bright representation of it. The editor becomes the author’s intimate advisor, pointing to the mistranscribed or ill-chosen word in the fair copy, hearing the great man’s voice break with emotion as he thanks the editor for recalling to life the long-forgotten flame.

The gothic-tower model is less gratifying and more realistic. The original conception is merely a rough sketch. The finished work is the product of accumulated details, most of them chosen by a more sage and experienced artist than the one who started the work. Some of those details result from chance events. A fax from Porlock interrupts one set of revisions and starts a train of thought which leads to another. A sub-editor capitalises a word, and makes a character more pompous than the author intended and the author, pleased with the effect, adds a few more words to emphasise the pomposity. The compositor’s eye skips over ten words in the typescript, and the author realises that a single word can fill the gap more effectively than ten. The finished work is an act of collaboration as well as of inspiration. The author expects the printer to correct some of his bad habits of spelling and punctuating that he is too lazy to unlearn. And he expects the publisher’s reader to alert him to any detail that might cause him to make a fool of himself in public.

Two different models – but they issue in three different Joyces: two invented by Gabler, one invented by Kidd. Gabler imagines a Joyce who made Ulysses and another Joyce who marred it. The first was a romantic artist-hero inspired to brightness like a burning coal – or, in Gabler’s words, ‘borne on a crest of creative activity’. The second was a mere ‘scribe’, an inattentive drudge who prepared defective fair copies of the inspired drafts composed by the first. This same scribal Joyce, perhaps for the sake of professional solidarity, tolerated or ignored the omissions and errors of his fellow bunglers who prepared the typescripts and proofs. What complicates the distinction between the author and the scribe is the fact that some invisible influence repreatedly moved the scribe to transitory brightness while copying a draft or reading a proof and transformed him into the artist who inserted a melodious sentence or sharpened an ironic phrase. Gabler’s editorial mission is to rescue the artist’s fire from the scribe’s ashes, to re-create the ideal Ulysses that is the sum of Joyce’s inspirations but which Joyce the author never saw.

Gabler’s Ulysses is a longer book than the 1922 version because Gabler restores all the bits that disappeared through the inattention of the scribes. No matter how often Joyce passed over defects in the typescripts and proofs, he did so as a scribe and not as an author. Any suggestion that these defects received ‘ “passive authorisation” may be confidently rejected’. All writers who have noticed after a second or third reading that one of their sculpted sentences came back from the printer as a mangled torso will sympathise with Gabler here. So will all readers who find that Gabler’s restoration of lost phrases turns vague nonsense into luminous precision. Gabler’s pursuit of the ideal, inspired Ulysses has at least produced, in dozens of places, a more lucid and coherent one.

Having begun with a theory designed to serve the particularities of Joyce’s prose, Gabler ends with an edition which sacrifices them to a textual theory. Gabler’s ideal Ulysses is a bibliographical utopia. Like all utopias, it sounded marvellous when it was planned in cafés and conference rooms, but turned arbitrary and ugly when put into practice. Gabler infers (on dubious grounds) that Joyce wanted the opening dashes of his dialogue to be set flush with the left margin rather than indented as in the French style-which was also the style of Finnegans Wake and the destroyed first edition of Dubliners. So Gabler’s dialogue appears in unreadable, uninterrupted blocks of type. Gabler finds no evidence that Joyce wanted the large boldface type used in 1922 for the headlines in the newspaper episode, so this strident piece of Modernist typography is reduced to polite capitals. Meanwhile, Gabler’s own theoretical prose gives pause even to his sympathisers. Do you really want a Ulysses shaped by the judgment of someone who writes that his edition ‘places authorial variants within the invariant context of the text-in-writing’ and ‘transforms the act of writing into successive writing results in the editorial synoptic presentation’? Emendation requires an ear for metaphor. Can you rely on someone who writes that Joyce’s drafts ‘give every appearance of being in volatile compositional gestation’?

Kidd’s most telling attack on Gabler’s method involves a paragraph, typical of many, where a typist altered the tone by miscopying a word and Joyce then emphasised the new tone by inserting a similar word in the same sentence. True to his theory, Gabler corrects the scribal error but keeps the authorial addition, making the whole sentence fall apart. And he throws in another word which had disappeared from an earlier draft and probably has no business in the sentence as Joyce finally revised it. Every word in Gabler’s version of the sentence was written by Joyce, but Joyce never wrote, saw or imagined the sentence that Gabler prints.

Whenever someone challenges Gabler’s methods, he replies that they are based on an established tradition of German bibliographical theory. If you don’t agree with the method, clearly you don’t understand the theory. Furthermore, the theory was put into practice largely by a computer which sorted through text that had been fed to it in the form of ‘error-free input’. This computerised, theoretical Ulysses is therefore a better book than anything that could be produced by muddled, empirical human beings. In Gabler’s bibliographical argument, as in other forms of criticism, the relation between theory and literature resembles the relation between the government and the people in Brecht’s poem on the 1953 Berlin uprising: ‘The people have lost the confidence of the government ... Would it not be simpler to dissolve the people and elect a new one?’

Unfortunately, Kidd’s alternative to Gabler may not be much of an improvement, except to the degree that the nonsense produced by Joyce’s typists and printers is superior to the nonsense produced by an editor and a computer. Kidd hasn’t said how he might proceed if he were to edit Ulysses, but someone else could easily put together a text based on the principles implied in Kidd’s essays, lectures and interviews. The resulting Ulysses would look a lot like the 1922 edition, perhaps with the same pagination. Kidd argues that the 1922 printing, the only one that Joyce supervised and corrected, is a lot less defective than Gabler claims. Kidd describes Gabler’s text as ‘marbled with the fat of ... pseudo-restorations’. A Kidd-inspired edition would presumably keep the leaner contours of 1922. In conversation, Kidd has convinced sceptical listeners that Joyce added or altered words in the page proofs in order to make the new words refer to their physical location on the page: for example, a line would be altered to include its own line number. The only new edition that could respect the intent of these changes is an edition which preserved the page numbering and lineation from 1922. If you start weaving in Gabler’s restorations, the physical reality of the printed book, which Joyce exploited at the last minute as one of the subjects of the book itself, disappears into the insubstantial vapour of an ideal text.

Kidd’s Joyce is no romantic artist-hero but a puckish and obsessive joker who delighted in any accidental detail. This Joyce threads together the fragments of Ulysses by alluding to Byron’s Hours of Idleness near the start of a chapter and inserting near the end of the same chapter the name of a Dubliner who, when you look him up in Thom’s Directory, turns out to have lived in Byron Lodge. Kidd’s Joyce doesn’t care about the phrases lost from his early drafts, but improvises on any oddity he discovers in the typescripts and proofs.

This is a plausible Joyce who has much in common with the Joyce who worried over women’s infidelity and therefore married a woman named Barnacle. But anyone tempted to believe that Kidd’s Joyce is the true one will pause over Kidd’s style as much as over Gabler’s. For Kidd, as for Kidd’s Joyce, the arbitrary and extravagant detail is all. In a letter to the press, Kidd offers this eyebrow-raising comment on a comment by Gabler: ‘Irony abounds. What redounds to Dr Kidd rebounds. On several grounds, it sounds, he’s out of bounds.’

The 1922 edition is the one that Joyce supervised, but the Joyce who accepted the faults of that edition for the sake of its line numbering is as much a fantasy as Gabler’s romantic fantasy of author and scribe. Writers with sharper eyes than Joyce’s skip over a typo in one word while revising the word next to it, and hope the proof-reader will clean up after them. In the months after the book appears the author busily lists the faults that he and his friends notice. Then he gets distracted by the next book and learns to tolerate the faults in the earlier one. But the fact that the author shrugged his shoulders doesn’t mean that an editor should imitate him. Kidd is right to question the notion of an ‘obvious error’; errors exist none the less. The insoluble but unavoidable problem is that of deciding which errors to correct. If you fix everything that doesn’t fit your theory, you introduce the kind of nonsense often produced by Gabler. If you correct nothing, you endorse the nonsense introduced by typists and printers. The true, correct Ulysses isn’t Gabler’s or Kidd’s. Nor, unfortunately, is it any of the infinite varieties of Ulysses which fall in between.

Every editor needs an inner Gabler and an inner Kidd to make his life difficult by arguing against all his decisions and insuring that he doesn’t suppress one version of the author in favour of another. When I edited the Auden-Isherwood plays the debate of those two inner voices repeatedly erupted into a shouting match. I managed to resolve some disputes to the satisfaction of both voices. Auden drew many passages in The Dog beneath the Skin from an abandoned play called The Chase. The typist of The Chase had trouble reading Auden’s hand and left some blank spaces which Auden then filled in with words that he misremembered from his manuscript. By printing the original versions in the text of The Chase I satisfied the inner Gabler, while I satisfied the inner Kidd by printing the later versions in The Dog. Other plays were less easy to apportion between them. I prepared two entirely different editions of The Dance of Death, one for each voice, before deciding on a third version which each seemed willing to accept with reservations. Now that I’ve sent back the final proofs. I wake to the loud complaints of one or the other, each insisting that it speaks for the absent author, each insisting that the other be banished from the main text in the front of the edition to the wilderness of notes and appendices in the back.

Auden wrote: ‘What every author hopes to receive from posterity – a hope usually disappointed – is justice.’ It would be much easier to give an author justice if we could be reasonably certain who the author was.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 10 No. 21 · 24 November 1988

One aspect of the controversy over the text of Ulysses which was not dealt with in Professor Mendelson’s brilliant exposition of the respective positions of Gabler and Kidd (LRB, 27 October) concerns copyright. Bearing in mind that Ulysses will fall into the public domain in 1991, it may be surmised that the publication of a ‘corrected’ text represents an attempt to start a new copyright period running, so that rival publishers could be warned off from poaching for another fifty years. Other recent instances of a similar nature come to mind – of publishers hiring scholars to edit the texts of authors whose copyright is about to lapse, with the same object in mind. An extreme example was brought to my notice in which the textual editor himself is claiming personal copyright in the emendations, however trivial, he has made to his author’s work. It is time to scotch these ploys by bringing a test case, in which the court would be asked to rule (as I believe to be the law) that, when an already published text is re-edited, no new copyright period is established unless the re-edited text differs so substantially from the earlier one as to amount to a new work. Clearly the Gabler text of Ulysses would not satisfy this criterion.

John Whitehead
Munslow, Shropshire

Vol. 11 No. 1 · 5 January 1989

Until reading the Diary of 27 October by Edward Mendelson, it had been my impression that the controversy surrounding Gabler’s Ulysses was not theoretical but historical. John Kidd’s allegations, voiced in ‘The Scandal of Ulysses’ (New York Review of Books, 30 June 1988), appear to be directed at Gabler’s violations of his own editorial principles and not the principles themselves.

Mesmerised by theory, nowhere does Mendelson reveal that Gabler has made some monstrous blunders. In 1984 Gabler labelled his Ulysses as the 11th typesetting; it was actually the 18th. While Mendelson muses over Kidd’s discovery of the Byronic allusion lodged in a certain Dubliner’s name, he neglects to inform us that the Dubliner, Captain Buller, makes a unique and unprecedented appearance as ‘Culler’ in Gabler’s Ulysses. Likewise, Dublin’s Harry Thrift has been altered to ‘Shrift’.

Kidd also balks at the two thousand emendations by Gabler that have no authorial basis. (That is, Gabler follows the typists and printers instead of his purported ‘Ulysses as James Joyce Wrote It’.) Since Gabler gives authority to any variant that appears in Joyce’s hand, his heavy dependence on facsimiles seems enough to undermine his own approach. Mendelson laments the loss of the Modernist typography in the ‘Aeolus’ episode and is justified. Gabler ignored this message, in Joyce’s hand, to the French printers: Est-il possible d’employer des types plus foncé …? Incidentally, the type Joyce commented on was already significantly darker and larger than Gabler’s headlines.

Kidd does not champion the 1922 Ulysses for history’s sake (a nightmare from which he is trying to awake), but requires that ‘errors’ remain only where Joyce actively incorporated them in his ongoing revision, as in the paragraph referred to by Mendelson. And while Mendelson appropriately protests Gabler’s liberties with the passage, he neglects to mention Kidd’s point of last June: on that page alone Gabler overrules Joyce’s final manuscript seven times.

Charles Phillips
New York

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences