To the Editors:

In The New York Review of Books dated June 30, 1988, though available on London book-stalls around June 10th and distributed free of charge in the cloisters of san Giorgio during the week of the 11th International James Joyce symposium from 12 to 18 June in Venice, you publish an essay entitled “The scandal of Ulysses” by Dr. John Kidd. The article is couched in the allegation, continuously reemphasized, that the editors did not compare all their work against the original documents. This is not true. They did. Dr. Kidd knows and has known this since the early 1980s when we first met. For him to assert otherwise scandalously rebounds upon himself.

The condemnation of the critical and synoptic edition of Ulysses (New York: Garland Publishing Inc., 1984), as well as of the current trade editions in England and the Us (Bodley Head and Penguin in England, Random House in the Us; all 1986) which print the reading text from the critical edition, spreads over some 18 columns of newsprint. Yet it rests on a mere 15 examples of readings cited in evidence. Neither individually nor together do they attain the degree of significance capable of sustaining a critique, let alone of carrying the weight of Dr. Kidd’s wholesale indictment of the method, scholarship and theory of the editions.

Three of the readings cited are name forms. One of them catches the edition misreading the hand of Frank Budgen, the friend to whom Joyce dictated part of the novel’s 10th episode. The second, if Dr. Kidd is right, may be an editorial failure to pick up a peculiarity in the original proofsheet. The third is a form that happens to differ from that of the name of the director of the Richmond District Lunatic Asylum in Thom’s Directory of Dublin. Yet it is the form Joyce unambiguously wrote in the fair copy from which the first chapter directly descends. By a representational standard which Dr. Kidd otherwise fervently endorses whenever he believes that the edition does not meet it, it should therefore have found his praise.

This leaves us with twelve examples. For ten of these, the complaint goes against two commas, two colons, one capital letter, one case of the author’s underlining in one manuscript against his not underlining (for italics) in another, one instance of a row of dots for an ellipsis, and three recording problems that affect not the text, but the apparatus in the critical and synoptic edition only.

We are then left with two examples: a spelling (“tway” as against “twey”) and a complex substantive reading—the only example, except for two of the name forms, to touch at all upon the words of Ulysses. First, the spelling variant: “tway,” whether we like it or not, is a spelling introduced in the first English trade edition of 1937, the only edition in Joyce’s lifetime which, as bibliographic scrutiny has ascertained (though Dr. Kidd would insinuate that such work to support the editing was not undertaken) incorporates authorial corrections. Clamouring for bibliographic analysis, Dr. Kidd yet wilfully closes his eyes to the editorial consequences of its results. What hypothetical “someone attempting to correct the text” in London in 1936/37 should have had a reason sufficiently to doubt the quaint archaism of “Watchers they there walk” (the reading of 1936) so as to locate the authentic replacement for “they” as “tway/twey” in a dictionary? No, the “tway” correction, together with other significant changes, helps to substantiate the interpretation of the bibliographic evidence. Joyce himself corrected the text. since he did so, the edited text incorporates the forms he employed in 1937 as his latest choices—even though for the first and only time he spelled “tway,” not “twey.”

The only example touching the word forms among the 15 given by Dr. Kidd receives an extra illustration in the article. This is a complex case that requires extended arguing according to the several options that the editorial hypothesis (i.e., the outline of the document relationships and of the edition’s principles and rules of procedure) holds in store. Tractable as it may be according to more than one of these options, it is a particularly bad example on which to indict the edition’s theoretical foundations. On the contrary, a perceptive discussion of the editorial potential inherent in variously handling it would help to set into proper relief the edition’s system of principles. How casuistically Dr. Kidd dismisses or accepts what little he allows himself to understand of them is neatly proven when, within a few paragraphs, he turns round implicitly to advocate constituting the text according to the very editorial rules he rejected for his one substantive example, or for printing the name of Johannes Jeep in roman. I would love to emend the title of Jeep’s song to “Youth here has an end” on the strength of a fragmentary draft manuscript—and I thank Dr. Kidd for pointing out the missed opportunity.


The scanty array of examples provides not even the flimsiest of foundations for a critique, let alone a condemnation, of the critical and synoptic edition, or, particularly, the current trade editions—not to speak of the slurs on the competence and scholarly integrity of the editor and editorial team—such as they are flourished in Dr. Kidd’s rhetoric of allegations, insinuations and sweepingly generalising assertions. Within the evidence adduced, close observance of details may optimize the edition in five or six cases. Three of these concern the apparatus (and thus the critical edition) only. For the reading text (and thus the trade editions), the article may have helped to change for the better one or two name forms, and one comma (with “Youth here has an end” thrown in as an extra).

Two basic misconceptions in Dr. Kidd’s assumptions about the editions of Ulysses, finally, need to be pointed out First, “The Corrected Text” is not the editors’ name for the edition. It is a phrase coined for the trade publication of the critical edition’s reading text by agreement between the James Joyce Estate and the publishers. second, the critical edition is a text edition, not a manuscript edition. It nowhere seeks or claims the diplomatic or topographic fidelity of a manuscript edition: i.e., it does not presume to render Ulysses “as Joyce wrote it” in the sense of recreating the writing and layout of manuscripts. As a text edition, rather, it telescopes textual developments into a synopsis, using diacritics and symbols to refer its textual and editorial results to the manuscripts and other documents where the writing acts implicitly or explicitly took place. The majority of textual documents for Ulysses are fortunately extant, though some are inevitably lost, as would be the case in any conceivable textual situation. It is a state of affairs that skilled editorial professionalism is specifically designed to meet. On top of its analytical display of textual developments which the edition most owes to common principles and practice according to a German school of editing, it follows received Anglo-American traditions of critically eclectic editing in establishing a reading text as an ideal text, i.e., as the sum of the acts of writing that shaped Ulysses. This reading text is the editorial suggestion—as any and every edited text is an editorial suggestion—of a valid text for Ulysses.

Dr. Kidd’s argument against the edition of Ulysses, then, is seriously flawed by an elementary failure to distinguish its critically editorial functions before a background of documentary referentiality which he tends to mistake for its representational aim. Of course he is right in implying that other editions of Ulysses than that of 1984/86 are conceivable. If one considers the example of shakespeare’s plays, there is no telling how many scholarly editions of Ulysses the future will see. But that precisely is the way: to realise, one day, an edition that can hold its own beside the first. To profess, in the end, to be unable to deal with the critical scholarship internationally invested in the editing of James Joyce’s novel of the century in any other way than by scrapping its evidence and returning to the highly inauthentic printing of 1961 is to admit utter helplessness before the challenge of Ulysses.

Hans Walter gabler

University of Munich

Munich, West Germany

To the Editors:

John Kidd’s “The scandal of ‘Ulysses’ ” does not mention the most scandalous “improvement” in the revised text of 1986: the setting of all dialogue, prefaced with a dash in the French manner, flush left. This typographical eccentricity is so unusual as to be disfiguring, and it does not appear in the text of which Joyce repeatedly read proofs, which is conventionally indented. His manuscripts do show the dash without indentation, but what we have here, as in the almost unreadable Thomas H. Johnson transcriptions of Emily Dickinson’s manuscript poems, is a mistaken scholarly fidelity to holograph mannerisms that were never meant by the author to be translated into type.

John Updike

Beverly Farms, Massachusetts

To the Editors:

Though it’s some years since I worked closely with the text of Ulysses, and only one major interpretive issue is involved (the “word known to all men,” which seems relatively easy to resolve), I am persuaded by Mr. John Kidd’s evidence that the Gabler edition is fatally flawed. It’s bad enough to find changes inflicted on the names of real people—like H. Thrift, Captain Buller, and Conolly Norman—all correctly spelled in previous editions. It’s worse to be unable to find, in the thicket of variants recorded by Gabler’s expensive Garland edition, authority for only one of these changes, and that based on a misreading of a facsimile. If this sort of error lies on the surface, what will be the total count when the edition is thoroughly worked over?


So it seems to be back to square one. The job of a proper—so called “definitive”—edition will be enormous. In many instances the better reading of the text will almost certainly be a judgment call, involving neither right nor wrong, but an estimate of balanced probabilities. That cannot be properly represented by accumulating variants at the foot of the page under peremptory sigla. A proper edition of the book will be cumbersome, pedantic, and encyclopedic—beyond the powers of any but a team of trained and disciplined researchers. It will take a long time, and by itself it may not be worth the effort; but it will provide an indispensable basis for a reading edition of the book in which the common reader can feel at least minimal confidence.

Robert M. Adams

Santa Fe, New Mexico

To the Editors:

May I add a note to John Kidd’s “The scandal of Ulysses” on the related ruckus over textual corruptions in the Gilbert and Ellmann edition of Joyce’s letters, though of course the publishing histories are not parallel and the corruptions not similar in kind—in the novel the disappearance of an aposiopesis and a dieresis, dittography replacing what had been mistaken as haplography, etc., in the letters misreadings of words, but sufficiently numerous and bizarre to mandate rescension of the entire Joyce correspondence.

Gilbert and Ellmann include only 13 of the 211 known communications from Joyce to sylvia Beach. On this sampling, Gilbert, though less circumspect than Ellmann, paradoxically does not stray as far from the plausible. Joyce’s “Yeats told me” becomes Gilbert’s “Yeats says,” Joyce’s “if it has not gone” Gilbert’s “if he has not done so.” But Gilbert is gratuitous, offering “can you [not]” for Joyce’s “can you,” with no effect on the meaning. Gilbert, moreover, “corrects” Joyce’s date on a letter though the postmark confirms it.

Some of Ellmann’s guesses are wild. He turns Joyce’s “[a dog] fond of raw postmen” into “[a dog] fond of raw posteriors,” Joyce’s “ought” to “expect,” Joyce’s “I think” to “luckily,” Joyce’s “amicably” to “presumably,” Joyce’s “saw” to “have.” Ellmann also omits the concluding sentence in one letter, a word in each of two others, and a parenthesis.

So much for the Ulysses authorities most familiar with Joyce’s handwriting. But Melissa Banta, editor, with Oscar silverman, of James Joyce’s Letters to sylvia Beach, who exposed these errors, does not inspire absolute confidence, referring to Joyce’s “financial straights” and passing over his—though surely not his—“rue Guy-Lussac.” Nor does Mr. Kidd. The title of sweelinck’s famous variation cycle is Mein junges Leben hat ein End (not eine Ende). A few lines above that Mr. Kidd writes “neither…are mentioned,” and, some paragraphs further above, “The ‘Eumaeus’ episode is cliché, euphemism, redundancy, and mashed metaphor ambered in a gooey sentimentality.” To amber in goo?

Robert Craft

New York City

To the Editors:

Professor Kidd’s critique of the Gabler text of Ulysses inspires me to try my own hand at the same sport. The work by sweelinck is Mein junges Leben hat ein End’.* Ende is a neuter, not a feminine, noun and requires the neuter form of the definite article.

Jon N. Elzey

Demarest, New Jersey

To the Editors:

Although the study of James Joyce is not my specialty, the firebomb John Kidd flung at the “corrected” text of Ulysses sparked my intention to complain someplace about a small misdemeanor in the Gabler edition. On page 227, line 791, in “sirens,” the flood of musical quotation and allusion may be what led Gabler and Company to render cord, as in vocal cord, incorrectly as chord, as in minor or lost chord. The word is spelled correctly in the Modern Library edition (1934, 1942) on page 272, line 39. John Kidd is to be commended generally, and specifically for forbearing to anagrammatize Gabler’s name as Gartle.

Beverly Fields

University of Illinois at Chicago

Chicago, Illinois

John Kidd replies:

In the year 2088, the general editor of selected Papers of the Joyce Wars has her hands full. she finds the documents (paper, electronic, and plasmic) surviving from 1988 incomplete, contradictory, error-prone, stylistically archaic and a touch comic. Letters in one issue of The New York Review keep Professor Futura and her textological lexica busy for weeks.

John Updike’s protest against the 1984 quotation dashes (where the style of all Ulysses editions approved by Joyce is overthrown) strikes Professor Futura as long overdue. In her master’s thesis on twentieth-century typography she argued that the 1922 first edition of Ulysses put through the press by Joyce strongly influenced world literature. Novelists of all nations had by 1988 so often played with Joyce’s 1922 style (which he repeated in Finnegans Wake seventeen years later) that the Corrected Text fell outside the very tradition Joyce inaugurated.

Professor Futura regrets having to correct Updike’s statement that Joyce’s “manuscripts do show the dash without indentation, but what we have here…is a mistaken scholarly fidelity to holograph mannerisms.” Updike is misled by Gabler’s cunning phrase “without indentation.” Gabler refuses to say that no manuscript of Ulysses has his own system of flush-left dashes. Rather, all the manuscripts have the dashes well into the margin, a design nowhere illustrated or mentioned in the 1984 or 1986 editions. There is no “fidelity to holograph mannerisms” in the Gabler system. That version agrees with no manuscript and no edition approved by the author, but is an innovation disguised as restoration. Even the master typographer of the Arion Press, Andrew Hoyem, was led to imagine Joyce was behind the unsightly page devised by Gabler. No volume of protest from the advisers to the November 1988 Arion edition (illustrated by Robert Motherwell) could undo the damage once the Gabler page had inspired Hoyem to tinker with the Joycean system. Hoyem’s page is more attractive than Gabler’s, but is without authority.

Robert Martin Adams is far too modest, Professor Futura concludes. His surface and symbol: The Consistency of James Joyce’s “Ulysses” (1962) ought to have been a model for the editors of the synoptic Edition. Instead, they seem unaware of the misnamed characters (not all from the hand of Joyce) that populate their text, while they “correct” a name here and there at random. Thrift to shrift is the most notorious goof. That Conolly Norman’s first name should have only one “n” seems to have caught Gabler by surprise. He demands that Kidd accept the erroneous Connolly because a single manuscript has it. Yet Joyce revised the typescripts, and his change is reflected in the correct spelling found in all texts prior to 1984. As Adams suggests, an editor seeks “balanced probabilities.” (The “absolute confidence” sought by Robert Craft being unattainable.)

From her archival vantage Professor Futura supplies a long list of mistaken identities in The Corrected Text which Kidd cautiously withheld from his New York Review article until the danger had passed that his work would be further misappropriated. Gabler never answered the charge that the majority of the changes between the 1984 and the 1986 editions were heisted without acknowledgement from Kidd’s articles, papers, and interviews. Taking his cue from an offhand remark Kidd made to The Washington Post (April 2, 1985), Gabler added in 1986 a period after “bottom of the ashpit” (page 624) in the structural middle of Molly Bloom’s otherwise unpunctuated soliloquy. The English mix became Latin nux for the first time in any edition, umlauts of 1984 were removed from Hurhausdirektorpresident and Ueberallgemein at Kidd’s insistence, the Irish Gaelic was recast (after Kidd, Irish Literary supplement, Fall 1985), a spanish phrase gained an accent, the innovative 1984 saltgreen was restored to Joyce’s salt green, and other unauthentic compounds and spellings were undone in 1986. This and more without the courtesy of a citation to Kidd inside or outside the edition.

A paradox of historical linguistics is neatly solved by Professor Futura: she explains that Robert Craft, Jon Elzey, and John Kidd are each correct, though one spells the German noun End, another End’ with an apostrophe, and another Ende. Working from different sources for the title of sweelinck’s song, each chose a form attested in Grimms Wörterbuch—end, end’, or ende. With pride in her archival spadework she reports that five torturously revised drafts of Kidd’s “The scandal of Ulysses” unearthed at the University of Virginia all read ein Ende. The feminization of ein into eine was a keyboarding error at The New York Review. Editing the article for the selected Papers, she alters eine to ein (as Kidd wrote it) and lets stand his Ende.

Robert Craft’s accusation of widespread corruption in the edited Letters of Joyce, while true, demands of Professor Futura an extensive apparatus showing where Ellmann himself slipped, and where he merely followed the sloppy transcriptions typed by sylvia Beach. she notes that in the years before cheap xerography, work on Joyce depended on typed transcriptions sent by stanislaus Joyce to Herbert Gorman, stuart Gilbert, and others. Textology was less sophisticated in the Pre-Post-Modernist age. The Joyce Wars of Eightyeight changed all that.

The letter of H.W. Gabler is in a prose unfamiliar to Professor Futura. Her plasma syntactic collator cannot identify the dialect. she inserts a sample Gablergraph for a plasma search under “academic prose” and retrieves only “Thomas Carlyle SEE Professor Diogenes Teufelsdröckh.” she supplies a note about Gabler’s repetition of “name form” (meaning “spelling”) in his letter and edition—his attempt at translating German Namensform. she chooses two passages from 1984 Ulysses page 1745 as the epitome of the dialect:

The text of the final working draft thus logically postulated may be reconstituted, as the substratum of the textual development displayed synoptically, from the fair copy and the typescript in conjunction. In the restitution, the fair copy and typescript texts emend each other mutually in accidentals and non-revisional substantives…. standard usage does not provide a safe norm for adjudicating errors in this chapter of linguistic experiment. Reduplication is a mode of self-generative morphology….

Try as she might, Professor Futura can find no evidence for Gabler’s claim that the editors checked “all their work against the original documents.” Both the 1979 prototype chapter (Munich, privately printed) and the 1984 three-volume edition explicitly state that facsimiles were used in lieu of originals. she quotes page v of the 1979 introduction:

These non-authorial adjustments to the printer’s copy, at present localized only from the photoreprint in the James Joyce Archive, remain to be double-checked against the original document[s].

How Gabler could “localize” whether changes were authorial or not without knowing which were in Joyce’s ink and which in the printer’s pencil mystifies her. In Kidd’s article she learns that the disputed accents in ink and blueish pencil in Crème de la crème on a typescript at Buffalo are recorded identically on prototype page 29 and final synoptic edition page 370. since both fail to record the author’s inked circumflexes beneath the pencilled grave accents she assumes that the facsimile transcriptions were checked neither before nor after 1979. More important than the range of errors detailed by Kidd is the convoluted and ambiguous admission that when they exist facsimiles were used in place of originals:

All variants revealed in the first machine collation of each group were checked against the original documents or document reproductions in facsimile (for the Rosenbach Manuscript) and photo-reprint (The James Joyce Archive for scribal transcripts, typescripts and proofs) to eliminate all input error. With error-free input… [1984, page 1906].

Gabler and his supporters came to regret and even deny that early claim of “error-free input.”

Not only the first, but the last check of transcriptions was done against facsimiles:

Claus Melchior…verified this edition’s emendation and historical collation lists one final time after typesetting against the Archive reproductions of the originals [page 1909].

The final 1984 statement on facsimile use confirms that only spot checks were made:

The libraries holding the original materials for Ulysses have been helpful on my repeated visits to resolve doubts arising from the Archive reproductions [page 1910].

If the facsimile was flagrantly inadequate a check was made. If the facsimile appeared satisfactory, the original was passed over. That is how within the space of the shortest episode, the seventeen leaves of “Nestor,” ten erasures went unrecorded in 1984. No “doubts arising” from the facsimile of the Rosenbach Manuscript, Gabler thought it unnecessary to hold each leaf up to a lamp in Philadelphia to supplement the record of erasures by his graduate students transcribing the facsimile in Munich. Joyce’s “Buller” on Harvard proofs was transcribed as “Culler.” With no “doubts” to investigate, the unambiguous original went unexamined.

Convinced by Kidd’s examples from documents in Philadelphia, Buffalo, and at Harvard that the Gabler team could not possibly have checked all editorial work against originals, Professor Futura wonders why Gabler insists in his letter to The New York Review that:

They did. Dr. Kidd knows and has known this since the early 1980s when we first met. For him to assert otherwise scandalously rebounds upon himself.

[Irony abounds. What redounds to Dr. Kidd rebounds. On several grounds, it sounds, he’s out of bounds.]

The registers of the libraries with major Joyce holdings, which patrons are required to sign daily, also yield no evidence of Gabler or any co-editor spending the long months in reading rooms necessary to verify facsimile transcriptions. Professor Futura is aware that the Center for scholarly Editions of the Modern Language Association would not grant its seal of approval until an editorial team affirms that every original is checked twice, each time by a different editor or assistant. An editor unable to give names and dates for every document verification would be shown the door with little ceremony.

Some of the most forceful arguments in Gabler’s letter she finds plausible, yet searches in vain for the Kidd passages Gabler purports to refute. Powerful blows strike against the air:

The critical edition is a text edition, not a manuscript edition. It nowhere seeks or claims the diplomatic or topographic fidelity of a manuscript edition: i.e., it does not presume to render Ulysses “as Joyce wrote it” in the sense of recreating the writing and layout of manuscripts.

Professor Futura is unable to locate any book, article, paper, interview, lecture, draft, panel discussion, radio broadcast, jotting, or doodle in which Kidd used the terms manuscript edition, diplomatic fidelity, topographic fidelity, or layout of manuscripts. The “basic misconceptions in Dr. Kidd’s assumptions” that haunt Gabler are lost to history.

Still, the phrase “Ulysses as Joyce wrote it” is found in Gabler’s article of that title from 1979 (omitted from the 1984 Ulysses bibliography), in the Afterword to the 1986 paperback (page 649), and in a paper read June 17, 1988, in Venice, only three days before Gabler’s New York Review letter disclaiming his own words:

An edition, then, that establishes the text of Ulysses critically as Joyce wrote it… [“The Editor Reviews the Reviews,” typescript, page 8].

Why Gabler chose to say that Kidd had “cited in evidence” only fifteen variants puzzles Professor Futura. The New York Review article states unambiguously that in two thousand places there is no manuscript support whatever for the Gabler readings. That four hundred changes of spelling and punctuation were adopted by Gabler from editions after 1922 for which not so much as a note to the printer in Joyce’s hand exists surely implies more than fifteen queries.

For the discredited 1937 changes adopted by Gabler (such as Gabler’s tway for Joyce’s twey), Professor Futura cites Kidd’s well-known paper (read April 26, 1985, society for Textual scholarship, New York) arguing that most of the changes were caused not by Joyce’s intervention but by proofreading against the 1932 Hamburg setting (e.g., 1932 “15 june,” 1936 “15 June,” but 1937 “15 june”). Kidd also is in print (James Joyce Literary supplement, Fall 1987) showing that the very first 1937 change (Joyce’s “Hellenise” to un-Joycean “Hellenize” on 1937 page 5), along with another dozen variants, were undetected when the 1984 editors made sloppy last-minute collations in the spring of 1983—rather late for double checking and clear thinking. Not knowing the full range of 1937 variants, how could Gabler say which—if any—come from Joyce? From copies of the handwritten collation sheets provided to Kidd by Gabler in 1985, Professor Futura calculated a 41 percent failure rate in detecting plate variations (such as 1922 versus 1925) during the hasty work of 1983. she is not surprised that the lists remain unpublished.

Gabler admitted collating neither the first American edition of 1934, nor the Matisse edition of 1935, nor the large-paper format of the 1936 (a signed copy of which she examined in the Kidd collection). Nor did Gabler publish any account of the three textual states of 1932, nor the new readings of 1933, 1935, 1939, and 1940—all within the lifetime of Joyce. Professor Futura allowed herself a smile at an editor lauding his own “skilled editorial professionalism” while he adopted 150 dubious variants first cropping up in the 1930s. For that period the printing history supplied by Gabler is a virtual blank.

For even the most elementary particles (quirks, quarks, etc.) the Gabler claim of no more than fifteen errors is wrong. Kidd’s article states that twenty-seven rows of dots for ellipses which Joyce wrote in the “Eumaeus” episode were removed without explanation. Professor Futura patiently lists those missing twenty-seven. In the Corrected Text, one ellipsis written by Joyce is absent from pages 508, 509, 516, 525, 537, 541, 542, and 543; two each are absent from pages 532, 534, 539; three from 538; and five ellipses by Joyce are absent from 526 and five from page 533. By any known natural number system, more than fifteen specific points, in fact, more than one hundred specific points, can be extracted from Kidd’s article to correct the Gabler calculus.

Further, Kidd lists in his article thirteen classes of unexplained textual tampering—for each Professor Futura cites in her notes a score of exempla. Gabler defended not one of those categories of editorial excess in his thousand-word letter.

In her distinguished career as chief textologist to the United scholarly Academies, Professor Futura was honored and enriched for her monumental selected Papers of the Joyce Wars. Few combatants of the wars anticipated the fame (and infamy) which came to redound and rebound upon them.

This Issue

August 18, 1988