Ever since Dalton, in 1966, on the occasion of the 25th birthday of Finnegans Wake, in the Festschrift Twelve and a Tilly, called for a complete and thorough overhaul and restoration of the entire text, there has been much talk about the possibility of a corrected Finnegans Wake. Now, almost forty years later, the Wake having reached its retirement age, without showing any signs of aging – on the contrary, it is getting younger by the day – there is still no sign of a revised edition.

Why? There are many causes. First there is the Estate. It seems it has had some bad experiences with corrected texts, dating back from the Ulysses-controversy in 1984. When Hans Walter Gabler published his three-volume critical edition of Ulysses he was widely and critically acclaimed, but at the same time, as is usual in the dantesque masonic academic circles, so we are told, not by everybody. Especially when the critical edition was followed by a real reader-friendly Ulysses, without the scholarly emanations and apparatus, but with on the cover in big widescreen lettering the words 'the corrected text'. Critics fumed from all orifices and in the New York Review of Books and elsewhere a scholary and somewhat undignified wrangle ensued that went on for many months and still crops up now and then. The critically inclined reviewers stated that Gabler had been sloppy, his textual decisions were dubious, his criteria for including or excluding emendations were wrong and if they were right he didn't follow them strictly. It wasn't a corrected text, it was a downright corrupted text, some suggested half in jest but more than half in earnest. Now, Stephen Joyce wants to avoid such a disaster with Finnegans Wake. He officially proclaimed to the Joycean community during a congregation in the good city of Zürich in 1996, while he waved his hands in midair in patriarchal fashion: "This is the book my grandfather gave you and I don't want anyone mucking about in the text." And his lawyer added, judiciously, juridically, on as well as off the record: "Once bitten, twice shy."

The second obstacle for a brushed-up Finnegans Wake are the fundamental doubts and dissensions in the Wakean world concerning a project like that. There is no consensus, not only on what a new edition should look like, but also over the question of the possibility and even the desirability of the restoration of the text. After the initial enthusiasm of the late sixties and seventies, doubts have crept in, and a revised Wake seems further away than ever. Not only Stephen Joyce seems to have learned something from the Gabler project, also those who wanted to furbish the Wake and to rid the text of mistakes seem to have lost faith.

On the basis of a genetic in-depth survey of the Kevin-episode (FW 605.04-606.12) Jack Dalton in 1966 estimated a number of 7000 emendable mistakes, mistakes that were made in the transmission from one stage to another by typists and printers. Apparently Dalton excluded the inadvertences made by the author himself when he fair-copied his drafts or he filled his typescripts to the brim with fresh textual matter. If Dalton had included these in his estimate, he would have found at at least twice that number. For each passage there are on average twelve to fifteen stages, from the first note to the final page proofs, and Joyce generally only added material, so there was plenty of opportunity to make ample mistakes, and the printers and typists didn't miss almost any of these opportunities. How could it have been otherwise with a text as inscrutable and impenetrable as the Wake? The miracle really is that so many things went right. Jack Dalton energetically and voraciously started counting typos but he quickly lost interest and never fulfilled his promise. This late nephew of the Dalton gang was usually quick on the draw but fired mostly blanks (as in his supremely and ridiculously unjust review of David Hayman's A First-Draft Version of Finnegans Wake in A Wake Newslitter, November 1963).

First we have to look into the question: is it at all possible to make a corrected Wake? Arguments against a restoration are manifold and weighty. We will classify them now, using as guides a number of outspoken opponents, Roland McHugh (in The Finnegans Wake Experience, 1981), Bill Cadbury and others (in debates at the electronic forum fwake-l in 1992 and 1995) and Sam Slote (in his Soundbite against the Restoration, 2001).

Is it necessary? Finnegans Wake is no Ulysses. Ulysses has seen quite a few editions while Joyce was alive, some with the author's corrections, some with new typographical errors. It was about time that somebody brought some order into the house. That is not to say that everybody was happy with the gablerized version, as we already mentioned. As a reaction even the first 1922 edition was republished, because it may not have been ideal, but at least it was the milestone that changed the face of literature by tearing it like a roadmap to pieces. Finnegans Wake however has seen only one edition, all things considered, the 1939 one, in which Joyce corrected 866 typos that gradually found their way into the main body by glueing them in. The text was never reset (until our bilingual Dutchification - we're not allowed to call it a translation). The Finnegans Wake we know is a fullblown genuine Ausgabe letzter Hand, that stands like a rock on our plate. An immovable rock. Joyce checked and doublechecked every single word and punctuation mark, and if something went missing or was published in a horribly mutilated form or genetically modified beyond recognition, it was Joyce's last will. It received his imprimatur and nihil obstat. We have his final intentions in black and white. Who are we to know better? This is the book that the grandfather of Stephen James Joyce gave us and that's all there is to it. That's all ye need to know. It says what it says and it was meant to be that way. Now go and muck yourselves, or do something useful for society, is the implication.

Is it possible? The indisputable fact that Joyce gave his blessings to the final outcome, at the same time makes it impossible to decide whether he overlooked obvious typos or silently approved them. Everything can have happened, on purpose or by accident. And if Joyce nodded, he may have nodded 'yes'. We know his eyesight wasn't too good and that it bothered him. We know too that he didn't particularly care for proofreading. (Who does?) He stopped the postproduction proofreading of Ulysses halfway through. When proofreading, he never consulted the old text; he relied on his elephantine memory. We can see that he corrected a lot of mistakes in each stage of compostion, but he left in quite a few as well. But the Wake is such a text that each and every letter and punctuation mark can be full of meanings. What may seem a typo may very well be a intentional and meaningful distortion, even if it was perpetrated by a typist or a printer and afterwards approved of by the author. We possess almost all stages of composition of every Finnegans Wake-passage in the James Joyce Archive, it seems, but when you delve into the Archives, you soon discover that a lot is missing. Not all changes were in writing. Sometimes Joyce dictated. In other words, there is no unbroken chain of material witnesses to rely on. On page 89.19 we have the famous 'skilllies' with three l's where two of these lampposts would have been more than sufficient. An oral correction by Joyce? An anal retentive joke by the typist, a posteriori approved by Joyce? Or still just genital selfindulgence by the printer and not spotted by the author? The Wake is not a text where typos stand out like sore thumbs in the eyes of the beholder, to say the least. But what if the three l's really mean something? (The 'what if' in textual criticism.) Dictionarymaker Eric Partridge maintains that Shakespeare is so witty and funny that even now there are jokes in his plays that nobody has ever noticed or even laughed about. The same might apply to the three l's. Maybe somewhere in a distant future somebody on the planet Mars will laugh one of his intergalactic heads off because of the l in the middle. And this 'skilllies'-business is a relatively clearcut case. This is the fundamental Heisenbergian uncertainty principle uniquely inherent to the Wake, which we will dub and christen The Skilllies-Principle: we never never can be sure of any anything whether it was meant like that or not.

Is it desirable? The reader gets attached to orthographic outsiders. Danis Rose, in his 'Reader's edition' of Ulysses takes exception to Stephen's musing at the end of the Eumaeus-chapter. He emends Und alle Schiffe brücken into the correct German Und alle Schiffe brechen. Fritz Senn deplores this (in his review in the James Joyce Quarterly, 34:4), because brücken has the potentiality of brechen, but not the other way round: there are no brücken anymore in brechen. He feels robbed, and rightly so. Once you start this kind of home-improvement, it never stops: will the next editor correct Bloom's budget, or change Barclay & Cook's address in Talbot Street from 18 (as Bloom remembers it in Ithaca) to 104 (as it was in the real reality)? Will he teach Molly Bloom how to pronounce Greek words or teach the cat not to say mrkgnao, but miaow, as all cats do, as every child knows? For the same reason Roland McHugh disapproves of Jack Dalton's emendation of 'Siker of calmy days' (FW 237.31) in 'Siku of calmy days'. Siku is the Swahili-word for day and probably what Joyce actually wrote on the Wake-galleyproofs. The typescript however shows 'Siker' (JJA 51:417, 51:416). McHugh is ready to read 'siku' in 'siker', but finds it impossible to see 'seeker' in 'siku'. Finnegans Wake is like Ulysses the image of an imperfect world in which, to quote the author's alter ego Stephen in Scylla and Charybdis, 'all errors are volitional and are the portals of discovery.' Even if it wasn't meant like that, it was meant like that.

Up till it now we have seen that redoing the Wake is not necessary, not possible and not desirable. Now for the practical side of life: Is is practicable? Is it doable? If emending 'skilllies' with three l's gives rise to so many objections, real hell will break loose when lost words or sentences are to be reconstructed. There is many a passage screaming out for emendation, correction, restoration, revision, especially when entire lines have been skipped in the process of printing or typing, causing major syntactical disruption and collateral damage in the sentence. An observant and attentive Sam Slote in his Soundbite notices how the Wake galley-printer in the Ondt and the Gracehoper-fable accidentally jumps one line too far down, from a 'he' on line 4 to a seemingly identical but still very different 'he' on line 5 (JJA 57:364). In the Wake it ends up on page 414.35, not as the intended 'Or, if he was not done doing that, improbably he was always striking ...' but as 'Or, if he was always striking ...', which makes it into an ungrammatical hash, an amputated, disappointed sentence. Nothing more simple than to reinsert the lost words, one would think, but Sam Slote fears that once you start emending, the floodgates of 'Pandorra' will be opened, as we say in Holland. Quote: 'If this error is corrected then, obviously, other comparable errors should be emended as well.' Unslote. And where do you draw the line? It is a slippery slope smeared with soap and before you know there will be exceptions to rules which are already exceptions to rules. And also: whó draws the line? Who lays down the criteria? Every selfrespecting Wakeologist will make up his own rules. And without verifiable, strict criteria textual scholars will devour each other like howling wolves in muddy waters. It is simply not possible to lay down criteria for revision for the Wake, along the lines of for instance Gabler's 'continuous manuscript text' for Ulysses. The 'copy text' that is our starting point is already complete. Every emendation is necessarily an on-the-spot decision, a judgement call to be made by a fallible human or Joycean being (homo joyceanus erectus) and cannot be reproduced by a computer. Emendation will be a tissue of educated or not so educated guesses, as far as possible based on knowledge and experience or insider trading. Moreover, it is well known that once you start correcting mistakes, you've already started making new ones. Better leave bad enough alone.

There is also a 'logopoeic' reason for leaving things as they are and not to disturb the wild life of the Wake, according to Sam Slote. One of Gabler's more important discoveries is 'the word known to all men' in Ulysses. Someone at some moment dropped or skipped some lines, causing readers for many decades to guess what the word could be that they should know but didn't. But when Gabler put the lines back – 'Do you know what you are talking about? Love, yes. Word known to all men' – an outcry of indignation was heard: maybe Joyce had crossed out the passage deliberately and in his right mind to supply the reader with an extra riddle and to keep the scholars busy for another three centuries. We know Joyce seldom deleted, but still we can't reinsert words or passages that were lost in an early stage and even less so in Finnegans Wake. The lost words, from their moment of demise or disappearance, do not form part any more of the texture and the process of fermentation. They are of a different order than the words that remained and that Joyce tinkered and jostled with and kicked about for many years in the writing process. Moreover, if something was lost or went wrong, Joyce on many occasions worked his way around the difficulties or made up something entirely new. There is no such thing as an 'invariant context' in which we could resuscitate lost words in Finnegans Wake. The context was everchanging, up till the last moment before publication. Emending means changing the context. It can't be done. The Wake is too closely knit.

In conclusion: restoration of the text of Finnegans Wake is not permitted, it is not necessary, it is not possible and it is not feasible. It is not even a possibility in the actual actuality of all possibilities. It is, in one word, an impossibility, in theory and in practice, in the full, logical, only sense of the word.

And yet, we did it. Not in the English text, but in our translation. And the only reason we did it, is that we simply want to read each and every word Joyce added to his book in the seventeen years of writing Work in Progress. And we expect everybody who loves Joyce and Finnegans Wake to want the same.

For seven itching years we translated the Wake from the non-English into the non-Dutch. On a given moment it became clear to us, thanks to hints from the Antwerp Joyce Center, that to understand what we were translating, we had to have recourse to the complete history of the Wake in the making. As early as late 1966, A. Walton Litz, we might even say The Walton Litz, already knew it: 'In many cases the understanding of a passage in Finnegans Wake is virtually a re-enactment of the process of composition.' Amen! Because indeed, such, to us, proved to be exactly the case. Fortunately, almost the entire extant composition process was published in the 63 volume James Joyce Archives. The Ur-Wake, or Work in Progress comprises the last twenty volumes or seven thousand nine hundred and twenty-six wellfed pages. We also consulted the later findings, such as a typed version of chapter I.1, some early sketches (published in the James Joyce Broadsheet in June 1989) with additions and changes, letters sent to Sylvia Beach with instructions for changes, a facsimile page reprinted in transition, and the almost complete run of the very final page proofs that turned out to be lovingly stowed away, unbeknownst to almost everybody, in the MacFarlane Library of the Tulsa University, in the middle of Oklahoma. And fortunately David Hayman, in his A First-Draft Version of Finnegans Wake, had already deciphered those parts with the most obscure handwriting, almost unreadable to human eyes. Through Joyce's accretive way of writing we could see the book take shape under his very pen. We saw nodes grow into chapters. We saw words busy being born and busy dying. We saw sentences grow from mere words to more pages. We saw syntaxes swell until, by an unsollicited intervention from outside, they burst. We saw the precise intention of Joyce go to waste because of accidental sabotage by typists and printers. Many times we wanted to cry out: Take care, Joyce! Watch out! Something's going wrong there! Keep your hands on the weel, for chrissake! Look in your rear mirror! Over there! You're losing something! O my God, this can't be really happening! An entire sentence off the road! An entire paragraph into the gulley! But the car scribbledehobbled rambling on, through puddles and potholes, rainshowers and hailstorms of critcism and ignorance, and one after the other essential car parts fell off to remain behind on the rocky rough country road from Dublin.

We saw cristalclear phrases being ruined and disjointed by an accidental loss of punctuation marks, letters, words and sometimes whole lines. We saw Joyce make the most of typographical errors by concocting something new out of the muddle. We saw how he desperately tried to correct accidental mistakes, but more often than not we saw how he had to admit defeat and lay down his arms in the face of the inevitable inky, murky sea of mistakes his typists and printers made, and by neglecting them, continue them. In short, we were biting our nails in sorrow and impotent rage, howling at the moon of the inevitable course of history.

A revised edition would have helped us - simple translators and by no means scholars o blimey no - enormously in our translation, but there was none, so we had to locate most of the missing parts all by ourselves - with more than a little help from some of the genetic scholars and faultfinders. Strangely enough, we discovered during our meagre seven years of translating, only 2228 'real' mistakes. Not the seventhousand that Dalton promised, but 'only' 2228 unaccounted for textual derailments from one stage into another, or transmissional departures as the scholars cautiously prefer to call them. 2228, that is three and a half per page, including blank pages. One third of these concerns spelling, twelve percent is punctuation, sixteen percent is missed words or phrases (and redundant words), four percent is mislocations and there are eighteen cases of skipped lines of text, that is less than one percent. In seventy cases we have stumbled on a combination of one or more elements. In several instances Joyce manages, within the timespace capsule of a mere few lines, to make all mistakes in the book or to prefer not to notice them. On page 516 this even happens twice. On lines 19 and 20, in a hallucinating recap of the incident with Herr Betreffender, we find in the printed text: 'he kept forecursing hapscuth's foul Fanden, Cogan, for coaccoackey the key of John Dunn's field'. Yawn answers the four inquisitors, telling them that a Shem-like figure is telling about a HCE-figure who is cursing the devil (Danish: fanden) because he can't find the key. An archival plunge taught us that all kinds of things went wrong in the transmission, and most of all in the transition-pages for the galley-printer, the magazine that should be renamed transmition. The printed HCE stammerword sounds like 'koeakoeakomekiki' but it is not Wakean Averellesque Daltonese for 'cuando se come aqui', 'what time do we eat around here' (as uttered in Lucky Luke nr. 31 Tortillas for the Daltons). No, it is a transmissional departure, for which we can now deliver the transmissional arrival. On the transition-pages Joyce only adds 'coacoac' between 'the' and 'key' (JJA 61:77), but the galleyprinter misunderstands. Also he overlooks the addition 'and ravid' and puts commas around the misplaced 'Cogan', that belongs eight lines up, near Miles, so as to form Miles de Cogan, the Anglo-Norman 'governator' in 1172, and at the same time 25th generation forefather of Winston O'Churchill. Improved, but by no means ideal (because not everything is clear yet), the phrase would run: 'he kept forecursing hapscuth's foul Fanden and ravid for the coacoac key of John Dunn's field'. The transmission can boast of the following Class A-departures: omitted words, misplaced words, wrong punctuation, switched words, and various misspellings. And all within six words! Truly a festival of faults, a comedy of errors, a jawbreaking, riproaring rampage of mistakes!

All our findings we added to the bilingual edition of our dutchification in a list of 'textual variants', but in fact we consider them to be in most cases the final word of the master himself. That's why we opted for an as-complete-as-possible translated version, that is, a text in which we incorporated all lost and found objects without any ado in the story, wherever and whenever possible. And it proved to be possible in around 90 percent of the cases. The Wake is sturdy enough to take some extras. Additions, as long as they are by Joyce, don't hurt the book. Even if they are not yet fully-fledged Wakese or if Joyce later adds new lines with the skipped lines as a basis. Yes, even if he seems to have deliberately crossed out some lines, usually they are so good that they have to be at least mentioned.

The same all-inclusiveness we applied in our Hamlet-translation (De Harmonie, Amsterdam, 2000). The only passage that the Arden editor, Harold Jenkins, didn't include in the play itself but relegated to the Longer Notes, we managed to include as well. It is a passage at the beginning of the second scene of the third act, when Hamlet admonishes the players to play well and not to lose themsleves in unnecessary digressions that are not called for and that slow the story down without adding anything to it whatsoever. The First Quarto shows another ten or so lines of Hamlet doing exactly the thing he warns the players not to do: endlessly elaborating and harping on about bad actors making the same jokes over and over again not knowing where or how or when to stop. Jenkins doesn't think it is genuine: 'it provides, ironically enough, an instance of the thing complained of.' Exactly, and that is why to us it sounded genuinely Shakespearean, and deserved to be included and translated. Our Hamlet was now more complete than all other Dutch translations, more complete even than Jenkins' own edition! Now we really could read every word by Shakespeare, being the literary gourmands that we are. We wanted to be just as complete in Finnegans Wake. Enough or too much! One for all and all or nothing!

However, at that moment Dirk Van Hulle stepped in with a subtle word of warning: in this way, would you then not be translating Work in Progress instead of Finnegans Wake? A nice and useful dinstinction, indeed, but we simply had no choice but to continue doing what we were doing. The alternative would be translating typographical errors and ungrammatical sentences while knowing full well that they were wrong and how they should be. As long as we knew what we were doing it was okay. ('Don't worry, I know what I am doing.' - famous last words by somebody.)

Moreover, if we really would have been translating Work in Progress, we would have needed as many translations for each episode as there are stages, an average of fifteen stages. We were crazy (to try to translate the Wake) but not that crazy!

We felt we were translating an ideal Wake, a Wake without the external distortions that had crept in. Could we reverse the process, we wondered, and rebuild an ideal English text out of our Dutch translation?

Tout au monde est fait pour aboutir à un livre, Mallarmé wrote in an early draft of his famous maxime le monde est fait pour aboutir à un beau livre. One book, and not a hundred or thousand books. Joyce did not write Finnegans Wake intending that there would be dozens of different versions. Nor did Shakespeare write Hamlet or King Lear assuming that an early version would be just as good and worthwhile to publish or play as a later one. (However interersting it is for textual scholars!) In fact, no writer would like to publish first, second or third draft versions as equal to the real thing. As an example of what a corrected English text of the Wake could look like, we used Harold Jenkins exemplary Hamlet edition of 1982. Jenkins worked for twenty-eight years and it is probably the last Grand Unification Hamlet that is going to appear in the next who knows how many years, the last edition to fuse the existing versions into a more or less ideal Hamlet. The new Ardens are being made with a poststructuralistic decontructivistic tendency to destroy. The new Arden Hamlet has the first and second Quarto and the Folio seperately published in one book.

A Grand Unification Finnegans Wake should be built on a Genetic Unified Field Theory in which all levels from notebook note to printproof print are taken into account. Not all variants can or should be included in the main body, but the emendations that are made are to be explained in the footnotes. Like Jenkins, we use two kinds of footnotes. The upper to note the differences with the published text, as Jenkins notes the variance with or between the Folio or Quartos. The column footnotes explain the editorial matters in plain words, what possibly went wrong where and how in the composition. It is not an explanation of the text itself or its meanings, but in this way the reader receives insightful information about the way Joyce wrote and the problems he faced. For instance here you will read where he used an Armenian wordlist and in which notebook he took the list down. You will get to know at which moment in the (de)compositional process the word appeared and disappeared or changed its outward appearance without Joycean authority. You can see where Joyce drove on in his rambling wondercart instead of stopping and picking up the lost pieces and parts. The notes mention the prepublications in transition and other periodicals, as well as the galleys and the many revisional stages. A certain passage may appear to be 'clean' and spick and span and spotless, but that may only be because there are some stages missing in the transmission. This can be noted too. It is possible to give the grammatically still intact version of longer sentences that have syntactically broken up. Extra light can be shed on doubtful cases. All sources are given by their British Museum number and their JJA-page. More often than not, the manuscript pages are so densely scribbled on that references to lines are useless, so the place on the page to look for will be indicated more precisely by a entirely new improved formula in which the page is considered as a clock and the location is referred to in time by the small hand of the clock, as airplane pilots pinpoint locations in the sky. For instance you will find the words 'born engles' on JJA 46:31 easily and in a jiffy on 430-500 hours.

Whether such an edition will be a more correct Finnegans Wake, is in the end undecidable, but at least it will be a more corrected one. And all in all a genetic edition along these lines will also be a good introduction to the content of Finnegans Wake itself, even apart from the fact that almost all emendations and relocated additions are very informative for the course of the story. For instance in this particular fragment the three additions that were sadly lost on three different occasions, 'it being all rock', 'petrified' and 'ouxtrador', all three offer a clear view of the situation in which Earwicker finds himself at that precise moment: chased into a room extra-door, outside the door by Herr Betreffender, Earwicker feels as if he is lain into a stoney grave.

In this way, we still can't look inside Joyce's head to find out how he really meant everything, but we are able to look over his shoulder on his writing desk, or rather on his bed where he lay writing, clad in white and with a bottle of mineral water within reach. Of course, a bookshelf can be written about every lost comma, but we have to start somewhere. The scholarly ideal would be a total genetic transcription of all levels, a synoptic-critical variorum-edition without any concessions to the reader, without any reader-friendly reading text to go with it, but instead full of pointed brackets, daggers and double-daggers, deletions and even colours. And the whole lot published in a priceless series of academically priced dvd's. For that we cannot wait. And who can? Sooner will be found out which words in the Bible are from the hand of God and which ones are later interpolations of the scribes.

Finnegans Wake is a book that intrigues. It intrigued you, dear Wakeologists, and that thou shalt never forget. But the Wake is not the toy of Wakeology. It belongs to the readers. The Wake scholars have a duty to produce a text that is as good and correct as possible and not to dream on and get bogged down in endless arguments about the impossibility of such a project. All reasons against a restoration are not based on respect and love, but on fear and awe, and fear and awe are bad counselors. Thou art not a secret society, although it sometimes seemeth like that. The knowledge, the know-how is there, and with knowledge it is the same as with guns: once they exist, they will be used. Every nonproliferation-treaty about emendations of transmissional departures is doomed to fail. So, you'd better prepare a good revised edition, or else moonlighting bunglers like ourselves will do the job for you and jump in the gap in the common Joycean market. With a fresh edition of the Wake, you can give the Joyce manuscripts a real new future. Finnegans Wake, the next generation!