Navigation – Plan du site
La plausibilité d'une traduction : le cas de La disparition de Perec

La Disparition: Problem Translations. The Rough and the Smooth

John Lee
p. 137-157


John Lee propose ici une analyse des paradoxes de deux approches de la traduction: les facilités d'une version d'apparence lisse, sans aspérités, et pourtant faite d'excroissances (ajouts inutiles) et de trous (omissions et autres insuffisances), à quoi il oppose les complexités d'une traduction autrement polie, c'est-à-dire respectueuse du grain du texte (les réseaux, notamment auto-représentatifs) et de l'intelligence du lecteur (sa capacité à appréhender ces complexités).

Haut de page

Texte intégral

1To hand (on its way to the wastepaper basket) a request for biographical details for a possible entry in Who’s Who in the World. Clearly there is some amusing mistake, apart from being on someone’s hand-me-down mailing list I am mainly unknown even to English readers of Perec’s lipogrammatic novel La Disparition, although being the first of so far three English translators of that work — Ian Monk being the second — is, I suppose, my chief claim to “fame”.

2No doubt many people think Gilbert Adair is the one and only perpetrator of this task — anyone who managed to wade through that recycled tree, the Sunday Times, to his column on the subject will certainly have noted that he claims therein and thereby to know the book better than anyone else in the world, especially as there is no time for dozing on this job (I summarize from what came to us in French in Courrier International (11-17 November 1994).

3This is a daring boast at the best of times, but in the circumstances rather a rash one since it hints that two, never mind three translators, is definitely a crowd. And as it also provokes me to remember that by the same token I too know a thing or two about this novel, I shall finally succumb to the temptation to catch him napping. Not so much a Who’s Who, shall we say, as a What’s What, for, beyond the readership of this particular novel, something more fundamental is, I feel, at stake.

4The scruples I have taken nearly three years to overcome to do this arise not merely from difficulty in airing a grievance, but also from two types of bias involved in comparing with something by one’s own hand. Not only the natural selfish one, but no less an opposite tendency to play down one’s own achievement in the face of the expected trouvailles and felicities of the competing version. This latter phenomenon is further mixed up with the inevitable desire to change and improve upon rereading.

5Perec once mentioned dreaming of finding La Disparition full of e’s, a nightmare that has visited me on several occasions: recognizing literally one’s own text and yet spotting ever more e’s disfiguring it like cancerous cells. This impossible dream is the nocturnal counterpart of the very real fact that the book is indeed crammed with e’s in every other possible guise, and in this respect translation may be seen as a never-ending, self-critical process of incorporating occurrences that had previously slipped through the net. In this respect, the importance of retaining possibly obscure proper names cannot be overstated.

6This sometimes governs one’s treatment of an entire section. The obvious example of this is Perec’s quoting of five poems from an anthology by the late President Georges Pompidou and which a miscellaneous collection of English poetry fails to render. This is because the Pompidou book happens to place under each author’s name a decorative pattern the e-full potential of which may not be obvious to everyone (especially if they do not take the trouble to look), but clearly did not escape the notice of Perec, that “gaucher contrarié” or right-handed southpaw.

  • 1 John Lee, N°3, TLS, 2-8 Sept. 1988, p. 958, excerpt below; PN Review ° 68, 1989, p. 18-19, poems by (...)

7Some preliminary information is required before we proceed. My translation was undertaken with a certain amount of backing and the real prospect of finding a publisher at a time when, unbeknownst to me, a translation of La Vie mode d’emploi was under way under the auspices of Collins Harvill. As I awaited acceptance from another publisher, which eventually came, I complied with a request to show the first few chapters to Harvill, who rejected it and promptly bought up the rights. Having gone so far, I completed my translation with little hope of publication, although several excerpts were published along with critical apparatus in the TLS, PN Review, Parcours Perec, Etudes littéraires, and more recently Palimpsestes, this time illustrating an article by Sara Greaves based on her master’s dissertation analysing my translational strategies1.

8Whilst Harvill were still casting about for a translator, I predicted that any alternative version would of necessity have certain properties, a prediction which in the fullness of time turned out to be correct. This was not prejudice on my part, but an unavoidable conclusion based on close familiarity with the text. The number System, clearly a major stumbling block, is a case in point. Of two possible solutions to the problem, one appeared to me considerably less unsatisfactory than the other. Having rejected my stance on this, the publishers left their translator no alternative but to recourse — doubtless willingly however — to what I consider the inferior, not to say soft, option, involving maybe an English alternative (e.g. thirty for twenty), or passages like this:

  • 2 Honesty dictâtes that any quote I make from Adair should be accompanied by the corresponding passag (...)

I put 25 francs on Scribouillard.
Out of 26 original nominations, only 25 now stand at Longchamp’s starting post, Whisky 10 (No. 5) having withdrawn. Initially Whisky 10 was thought a cinch to win, although, surprisingly, at official odds of 18 to 1.
(A Void, p. 65)2

9The problem is as follows: Perec’s System makes exhaustive use of numbers that are lipogrammatic in French but not in English (where, as in other areas, exhaustiveness is out of the question):

  1. to designate the lipogrammatic process itself (3, 5, 6, 25, 26, and arguably 1);

  2. to hold his chronology together (18, 20, 28). Thus in 1968, it is 20 years since Douglas Haig Clifford died aged 20. This sets his birth in 1928, the year the original Douglas Haig died, and neatly coinciding with his mother Anastasia’s non-appearance in the first talking movies. Neither of these networks can be safely tampered with;

  3. to link together the remaining miscellaneous numbers (10, 23...). One example, “un million” is cleverly divided by “vingt-huit” in a currency exchange calculation to give “thirty six thousand” dollars (Perec’s English). Since 36,000 x 28 = 988,000, the margin of error is negligible: a mere 1.2 %: Elsewhere, lacking a suitable large number, Perec writes “vingt-huit fois vingt-huit”. The question is: can we get away with altering these?

10Adair tackles this latter instance with impeccable arithmetic, “translating” it as “784” and raising a number of more specific questions in the process:

  1. is this number important in itself, i.e. maybe, since we are talking in years, something relevant occurred in 1184 (1968-784)?

  2. if so, given that he has allowed himself to write any figure he likes, why did he not at least leave 282, which carries more information?

  3. if not, why not write “a thousand and sixty-six” or some other lipogrammatic figure?

11My own solution illustrates a major guiding principle when translating this text: ascribe as much importance as possible to what is actually written, even if you do not know why, for you will doubtless find something if you only look hard enough. As indicated above, Pompidou provides one instance of this. A second example, which also happens to illustrate the principle “if it ain’t broke, don’t fix it”: Adair replaces “axolotl” with “anchovy”. Now an anchovy is something you may find on the edge of a plate of someone who has had pizza, but “axolotl” is a beautiful e-less word in English and French alike, and means “water doll” in Nahuatl, a language mentioned by Perec.

12I long suspected that its place here had something to do with its being a distinguished example of neoteny, or the retention of immature characteristics in adulthood, a mechanism thought to explain the current flatteringly elevated status of homo sapiens sapiens — here in the form of external gills. However it was only very much later, on seeing a photograph in New Scientist, that I spotted the three red, lacelike protuberances forming a capital E on either side of the head. This is what has done for Douglas Haig Clifford’s carp Jonah and now twice been drained out of his fishpond.

  • 3 Even as I write, I detect that Perec is within 0.006% of 2 Mayan Baktuns, equal to 40 Katuns (288,0 (...)

13Likewise, although to this day I still have not worked out the possible relevance of the 12th century to our story3, I thought it wiser to attempt a close approximation to 28 within the terms of my apparently unacceptable numerical System. This System involved two devices. The first of these was recourse to a dictionary of Scottish dialect to obtain, not one, but a whole string of alternative lipogrammatic spellings of “twenty", successively lining up all the other vowels.

14The second was to improvise a nonce word “fivorsix” to cover the countless occurrences of “cinq ou six", of which more later. What, one may ask, is wrong with a nonce word? The dictionary is full of them, and so is La Disparition. And here for once, necessity being the mother of invention, it is justified. With these two devices, the entire numeric System became available in English, together with an endless supply of English lipograms, curiously enough all even: 2, 4, 30, 32, 34, 36, 40, 42, 44, 46, 50 etc., 60 etc., 1066, 2040, etc. etc.

  • 4 The reader is now in a position to appreciate my awaited quotation without distraction; in addition (...)

15The only vital number still missing was 28 in the two above-quoted instances. The second of these I got round by suggesting “thirty by twinty-six", which comes to 780, an unpardonable imprecision of 0.5 %4:

  • 5 My pocket calculator when held upside down gives an interesting resuit had he only taken the closer (...)

16To return to the earlier example, when converting the million “hrivnas” (the price on a criminal’s head) into dollars, I said the dollar was “up two from 26", which does no more than suggest “a floating dollar” subjected to market forces. Adair’s solution is rather more devious, since he quotes the hrivna at not thirty but thirty-two to the dollar5. This comes out as a roundish figure of $31,250, which unfortunately he cannot use because, although he systematically fails to acknowledge Perec’s English, he leaves it at “thirty-six thousand”.

La Disparition

Qui fournit un appui probatif à la position du dollar ($)

Un dollar valant vingt-huit hrivnas, Albin fit, tambour battant, un calcul approximatif, puis contrôla dans un journal du soir la fluctuation du cours.
- Thirty-six thousand, dit-il plutôt faraud.

A Void

Which will furnish a probationary boost to a not always almighty dollar ($)

As a dollar was worth about thirty-two hrivnas, Albin briskly ran up a rough calculation, taking inflation into account, and said, just a bit boastfully: “thirty six thousand".


Which just shows how much a floating dollar ($) is worth

  • 6 The comment “[GP’s Franglais in bold italics]” followed the first occurrence of this.

With your dollar up two from 26 hrivnas, Albin did a quick long division and had a look in a twilight tabloid to monitor any fluctuation in world gold standards.
- Thirty-six thousand6 said Albin smugly.

17Adair has introduced “inflation” as a device to make his figures look right, forgetting the newspaper exchange rates in the process. However one thing that is clear from my above remarks is that there is nothing “approximatif about Perec’s calculation, something to which I have tried to alert the reader with an oxymoronic “quick long division", whereas Adair’s sums are easily as “rough” as he claims. Furthermore, I am no expert in money matters and so am wary of “brisk” sales patter, but inflation I believe is what happens to prices, wages, tyres, rubber ducks, frogs-that-want-to-be-as-big-as-oxen and the universe at the age of a little over 1043 seconds, whereas currency rates usually only fluctuate. The word Perec uses, Adair’s hrivna, has indeed fluctuated violently from 28 to 32, which actually means less dollars, not more. There is another explanation for the missing $4,750: the usual term, I believe, is commission, in this case a hefty 15.2 %:

18Thus, unbeknownst to the reader, exchanging Perec’s French for Adair’s English is costing him dearly, for he continues metaphorically to hand over $36,000 for a million heavily devalued hrivnas (anagram: va[r]nish...), with a substantial eut going to the “just a bit boastful” middle man who doctors his chapter heading to “boost” the dollar and cover his tracks. What a pity he steps outside of the probity of “probatif” (probative, substantiating), exposing himself to the “probationary” treatment meted out to usually young offenders:

19Constant supervision is something both Perec and Adair call for, as they are both always up to something, although, I would suggest, in very different ways.

20The beauty of Perec’s lipogram is of course that you only have to dip into one short section to gain access to the whole. And the thing about Adair’s effort is that wherever you look, you are rewarded with some insight. All published sections of my own work have invariably involved the more difficult passages since they epitomize the problems that need to be addressed. I shall continue therefore to adopt that selection criterion here.


21The sample I now propose to take I have chosen as part of the wider context published in the TLS of 2 Sept. 1988, to which the reader is referred (Adair’s p. 271) and is actually not by Perec at all but by Queneau. Whether or not Adair’s “Raymond Q. Knowall” is a relevant pun — if so, Raymond K. Nowall might have been a neater solution — for present purposes we shall take it that his snippet is well enough informed as to the workings of the text as a whole.

La Disparition

Ondoyons un poupon, dit Orgon, fils d’Ubu. Bouffons choux, bijoux, poux, puis du mou, du confit; buvons, non point un grog: un punch. Il but du vin itou, du rhum, du whisky, du coco, puis il dormit sur un roc. L’infini bruit du ru couvrit son son. Nous irons sous un pont où nous pourrons promouvoir un dodo, dodo du poupon du fils d’Orgon fils d’Ubu.
Un condor prit son vol. Un lion riquiqui sortit pour voir un dingo. Un loup fuit. Un opossum court. Où vont-ils? L’ours rompit son cou. Il souffrit. Un lis croit sur un mur: voici qu’il couvrit orillons ou goulots du cruchon ou du pot pur stuc. Ubu pond son poids d’or.

A Void

“I’m going to rock this child in his cot”, sighs Orgon, son of Ubu. I’m going to wolf down mutton, broccoli, dumplings, rich plum pudding. I’m going to drink, not grog, but punch.” Orgon drinks hock too, rum, Scotch, plus two hot brimming mugs of Bovril to finish up with, which soon prompts him to nod off. Running brooks drown out his snoring. I stroll to rocks on which I too will nod off, with Orgon’s dozing son, with Orgon, son of Ubu. Condors swoop down on us. Poor scrofulous lions slink out, scrutinising dingos with scornful looks. Chipmunks run wild. Opossums run too, without stopping. North or south? I wouldn’t know. Plunging off cliff tops, bison split limbs in two. It hurts. Ivy grows on brick, rising up from stucco pots to shroud Windows or roofs.
From Ubu’s bottom drops his own bulk in gold.


I’m christ’nin’ this doll, quoth Orgon, son of Ubu. I’m thirsting for fungi, octopi, foci, for lights or comfits. I’ll not drink grog, but punch: so plonk is drunk, or rum or whisky or coco; dossing down on this rock, his snoring drown’d by this brook’s boring sound without bound. I’ll go sub its bridging for this bit of kip, kip for this kid of Orgon son of Ubu’s son’s. Condor wings off in flight. Lion’s out without its might to sight this dingo.
Wolf is flitting. Opossum stops sitting. How this rush? Bruin’s bust his nick. Suff’rin’ for it too. Lilium growing on brick: now it’s stuck on lug or spout of jug or pot in top stucco.
Ubu sits on his pounds in gold.

22The first thing that leaps off the page is that Adair is about thirty per cent longer than the original: 145 words and 813 characters/spaces as compared with 117 and 622 respectively in the original (my own figures being 126 and 670). This may mean he feels particularly at ease with this additional constraint of ostensibly eschewing not only e’s, but a’s as well and using only one y. Then again it may indicate the opposite. Quality, not quantity, is what we are after. And quality we shall assess in terms of formal equivalence (compliance with the lipogrammatic constraint), semantic adequacy and poetic harmony.

23Let us start by seeing how Adair tackles the lipogrammatic constraint, which in my book is only the tip of the stylistic iceberg, but, as Perec’s reference to the Titanic reminds us, is too often more than many people see. He fails on this count in two respects, first by falling for Perec’s deliberate mistake, and secondly by missing his hidden agenda.

24Deliberate Mistake: The next thing that strikes the reader here is of course that “whisky”, an extremely interesting word whose main function here is to carry that single y, has been “scotched”, presumably because the translator cannot do without “ivy” to render “lis”. We shall see later why perhaps he turned his nose up at the literal translation, “lily” — which I Latinize into “lilium”, having for my part kept “whisky”. The point here is that the French contains not one, but two y’s, the other being in the very first word, “ondoyons”! In my version, the second y appears in the preposition “by” — interestingly, TLS readers will note that this was changed to “in” by a perspicacious but overzealous proof-reader: Adair’s prose however, as it states, delivers “a solitary y” (although unfortunately this is only one of three “solitary’s":), maybe because with “ondoyons” his mind was entirely on avoiding “a”.

25Hidden Agenda: So much for the deliberate mistake. What about the hidden agenda? Although unannounced, in symmetry with the a, there is no z in the French, nor in my English. Adair has one, in a self-referential “dozing”. This being so, one suspects that the absence of j and q, far from being an alternative exploit, is merely an accidental overlooking of the fact that this polylipogram also happens to be a “defective pangram” containing all but a, e and z. Hence, although “sighs” is a very neat cockney lipogram for “says”, it fills what is otherwise a useful slot for introducing a q with “quoth” (also more accurate semantically).

26Rough draft: We come to a number of mistranslations, not all of which appear intentional and cannot be entirely ascribed to difficulty with the constraint. Incidentally, much of the above-noted extra material is involved here. This would seem to indicate that long-windedness is not so much revelling in lipogrammatic overkill as struggling to come to terms with the text.

271 - “Ondoyer” means to baptize in an emergency. However “baptism” is ruled out (contains an a) and so is “christen” (contains an e). Here my method diverges from Adair’s. On the principle that there is no limit to semantic freedom once it takes hold, I preferred the (doubly) apostrophized form “christ’nin”‘. If this solution was not acceptable to Adair, he could always have tried “dousing”, which would have had the additional advantage of assonancing with “dozing". However assonance does not seem to be an acceptable device either. Instead we have the first of several circumlocutory approximations: “I’m going to rock (this child) in his cot”. Here “in his cot” is not only superfluous, but actually takes us further away from the original sense.

282 - “Sous un pont” is rendered by “rocks on which”, a curious prepositional inversion necessitated by this second dose of that favourite interjection of Molly Bloom’s, “rocks”, although no translation of “roc” is forthcoming: My solution, “bridging", is also a second occurrence, being taken in a huge spanning movement from the very first page of the book which carries several echoes (six to be precise, here italicized) to this penultimate page from the penultimate chapter: I refer to the poem by Jacques Roubaud, a link that obviously could not have occurred to Gilbert Adair who for some reason has quite simply overlooked this somewhat hermetic poem.

La Disparition

Un corps noir tranchant un flamant au vol bas
bruit fuit au sol (qu’avant son parcours lourd
dorait un
son crissant au grain d’air) il court
portant son sang plus loin son charbon qui bat

Si nul n’allait brillant sur lui pas à pas
dur cil aujourd’hui plomb au fil du bras gourd
Si tombait nu grillon dans l’hors vu au sourd
mouvant bâillon du gris hasard sans compas

l’alpha signal inconstant du vrai diffus
qui saurait (saisissant (un doux soir confus ainsi
on croit voir un
pont à son galop)

un non qu’à ton stylo tu donnas brûlant)
qu’ici on dit (par un trait manquant plus clos)
l’art toujours su du chant-combat (noir pour blanc)
J. Roubaud


A black thing wings a flamingo, low flying
Bound along ground (which, prior to flight, not light,
Brown’d a grinding
sound in flood of air), plying,
Carrying its blood afar, coal carrion in fright

If nobody was coming braggingly to pass,
Galling brow now, plumb on sagging arm a bind,
If, falling, stark cicala, out of sight, out of mind,
Moving, gagging, gray sick luck, out of compass

Alpha, inconstant sign of truth’s diffusion,
Which might know (grasping (on a night of calm confusion
So you think
to sight its bridging footfall)

This No, flaming gift to your plumbago, writing)
That thus is said (by missing mark, most shut of all)
That long-known art of wordplay-swordplay (black for whiting).
J. Roubaud

29“Lack of poetry” would indeed appear to sum up much of the criticism I am making about this translation. Queneau’s text is basically a prose poem that has had all the music wrung out of it.

303 - With “I too will nod off with Orgon’s dozing son” for “promouvoir un dodo, dodo du poupon”, we have an extra, first person singular, sleeper.

314 - With “with Orgon” for “d’Orgon”, we have yet another English language sleeper. What a dozy lot:

325 - “Condors swoop down on us” for “prit son vol” marks a curious change of direction aimed somewhat aggressively at a first person plural. “What dreams may come”...

336 - “With scornful looks”: if this refers to the “riquiqui” lions scrutinising a potential meal of dingos, they would surely rather have a lean and hungry look; either way, this phrase has no equivalent in the French. Or maybe it is a throwback from the condors? I wouldn’t know.

347 - “Chipmunk” is a rather far-fetched translation of “loup”, standing in for “wolf, already used as a verb. If I had had the bright idea of using “wolf as a verb, I should have put “wild dog” here.

358 - “Run too, without stopping” certainly translates the primary sense of “court”, but “without stopping” is incorrectly self-referential, since “court” is short indeed and stops immediately.

369 - “Où vont-ils?”: “North or south? I wouldn’t know” is somewhat laborious and, it is my contention, off target, since the book is principally about going east and going west. Thus Ahab’s fateful voyage in Moby Dick, combining the one, literally, and the other, metaphorically.

3710 - Having got this far, and until we see what they get up to, these “bison” appear to be a relatively acceptable rendering of “l’ours”, since my “bruin” would admittedly sound odd in Adair’s plural, of which more later.

3811 - “Rompit son cou”: the connotation in this novel of a fatal injury to the neck is of course “avocalisation", through the severing of the vocal cords, here marking the removal of one of six animals. This is somewhat undermined however by pluralizing them all, and then further subdividing, instead of drowning, these Gadarene bison “plunging off cliff tops, (...) split(ting) limbs in two”.

3912 - “Ivy” for “lis” instead of white “lily” and, as stated above, containing the y that really should have been kept for “whisky", I can only attempt to motivate by some desire for verisimilitude in a paragraph where clearly from the outset (“bouffons... bijoux”) verisimilitude is not the order of the day.

4013 - This leads to uncharacteristic verbal economy, with one word — “pur” — being actually omitted altogether, but in this overblown translation the poor lily turns into an inflationary beanstalk: “rising up from stucco pots to shroud Windows or roofs” for “couvrit orillons ou goulots du cruchon ou du pot pur stuc”.

  • 7 Georges Perec, “Histoire du lipogramme” dans Oulipo, La littérature potentielle, Paris, Gallimard, (...)

41Food for thought: If I have so far passed over the food and drink section, it is because more is going on here than would appear to meet the eye and it has little to do with semantics. Thus, whilst Adair neatly translates “choux” with “broccoli”, his ravenous appetite otherwise leads him to “wolf down” indiscriminately only food, albeit quite as heavy on the stomach as jewels. What is actually on the menu is in the first instance a set of irregular plurals (bijou, joujou, caillou, chou etc.), followed by a rhyme and a preserved foodstuff, literally “made with” here fat (lipos). Perec wrote tongue in cheek “lipogramme ne veut pas dire ‘lettre graisseuse’ et encore moins ‘un gramme de graisse’”7, but of course for him it does: Working backwards through this list, I have certainly lost a little of this with my comfits” (sugar for fat), my rhyme “lights” is somewhat weaker too, although here again the translation is accurate (pet food for Leopold Bloom), and my first trio offers a set of irregular English plurais — all the more irregular since, although used in the 1995 Booker Prize, my “octopi” is unfamiliar to the OED which prefers “octopodes” or “octopuses” — however “pi” certainly makes it sound more palatable.

42There is nothing much to say about Adair’s fare, except that there is not a plural in sight, instead a collection of collective nouns. This could be a rather interesting ploy to offset his recourse to a multiplicity of animals lower down for French singulars, but try as one may, one fails to see how this might work.

43There is however one item of note in Adair’s menu, “plum pudding”. “Crocus and plum pudding” is an oath used by police officer, Ottavio Ottaviani, which Adair independently attributes specifically to Paris cops from Corsica, although as always failing in the process to identify Perec’s English. I do not know where Perec got the crocus from. However I suspect he got the plum pudding from Moby Dick, which at least is where the dictionary seeks explanatory help:

1851 H. Melville Moby Dick. viii. 64 Plum-pudding is the term bestowed upon certain fragmentary parts of the whale’s flesh.
plum-pudding voyage, a short voyage for which a supply of fresh provisions is carried
1851 H. Melville Whale xvii. 94 Some sailors who had just come from a plum-pudding voyage as they called it.

44Fragmentary, comfortable, riskfree, short-distance travei, this “rich” plum-pudding surely says something about our journalist’s translating style...

45Pulling no punches: “Du coco” is rendered a s “two hot brimming mugs of Bovril to finish up with”. This is a recipe for punch, not a list of successive drinks, hardly Bovril therefore (incidentally “to bovrilize” means to condense, or epitomize, rather the opposite of what we have here). As a perfectly acceptable word of English in its own right (Longman’s Dictionary of the English Language, p. 281), “coco” bridges neatly with the network of cocos — egg, communist — that I for one went to some lengths to reconstruct in English.

46If this solution were to be acceptable, “brimming” and “to finish up with” are clearly surplus to requirements and reinforce a “mugs” game that introduces a “two” which is also supernumerary, and damagingly so, as we shall see. Unfortunately, coco (here coconut) can never suggest the cocoa (cacao) that lies behind this choice. The OED — which admittedly does state this usage to be obsolete — will have no truck with anyone who thinks otherwise:

Coco remained the established spelling in the 18th c, till the publication of Dr. Johnson’s Dictionary, in which the article Coco was (app. by some accident, for Johnson in his own writings used coco, pl. cocoes) run together with the article Cocoa (= Cacao); this gave currency to a confusion between the two words which still prevails, although careful writers have never ceased to use the correct form coco.

47It is however an alternative spelling in phrases like “I should coco(a)” (rhyming slang for “I should say so”). Either way, deliberate use of this obsolete word or variant spelling strikes me as better than dropping out of the circle of “careful writers”.

48Cinq ou six: We come back to the unfinished number question. The number of words is one way of indicating a forbidden number or similar word. Perec does this with “Idiot, idiot, idiot:” rewriting “triple idiot”. At the risk of mixed operatic pathos, Adair does something similar when he has Douglas Haig Clifford as Mozart’s commendatore singing five of Rodolfo’s six “mi”s, three “Mimi”s to be precise, at the end of La Bohème (my own translation of the fatal mi or E was a tempered equivalent, “F flat”). It comes therefore as something of a surprise that he fails to notice the fact that five “eatables” are followed by six beverages since, unfortunately, he only offers four of the former and, with his two brimming mugs of Bovril, seven of the latter.

  • 8 “The disparition of three final stars”, James Joyce, Ulysses.
  • 9 That this should be at the heart of the lipogram is borne out incidentally by the value of log e, 0 (...)
  • 10 Georges Perec, W ou le souvenir d’enfance, Denoël, 1974, p. 57.

49There is however more to this than merely the lipogram, which primarily reflects Perec’s autobiographical concern and linking of the number 11 (5 + 6) with his mother’s death, probably at Auschwitz, the original “Disparition” officially dated 11.2.438. “Ou” often translates as “and” in everyday French, and particularly so of course in La Disparition whence “et” is banned. This was partly my reason for listing the ingredients of this punch with “or”s, when “with”, though longer, might have been better — the other reason being the link with "or” (gold) in the last line. This last is no idle comment, for as I have noted elsewhere, one of the more lucrative and sinister aspects of the death camps was the extraction of the victims’ gold teeth. “Cinq ou six” goes further, actually completing the fateful date with 2 numbers of 4 and 3 letters. This element is of such crucial importance that it has to be repeated with monotonous regularity throughout the novel. The same goes for “vingt-cinq” in combination with “vingt-six”, where the pair of twenties underlines the forty of 43, exactly 25 years previous (68 - 25 = 43)9. I therefore make no apologies if my “fivorsix” jumps off the page even more loudly; given the mixed potential of English — four and three letters, but an unwanted e — it just had to be done. The alternative is a rather tasteless cover drawing by Mark Moran with a pair of tombstones in flagrant contradiction (RIP and let RIP) with the key to so much of Perec: “Ma mère n’a pas de tombe”10.

50Buvons un coup: Before we deal with the final line, it may be noted that “buvons” at this point harks back to a previous occurrence transcribed Ba va, which Adair misses entirely. Here is the passage, referring to a so-called Katun:

La Disparition

La complication naït surtout du fait qu’il s’agit ici d’un jargon avocal, n’utilisant pas la vocalisation, donc impliquant contradiction quant à sa prononciation. Mais choisissant, par imitation, à l’instar du connu:

Ba va sa ka ma sar pa ta par da
Bi vi si ki mi si pi ti pir di
Bo va sa ka ma sor po ta par do...

Un gabarit simulant la transcription, nous allons, par la raison, l’intuition ou l’imagination, aboutir à un brouillon moins approximatif. Il s’affaira aussitôt, traçant vingt-cinq signaux au crayon blanc sur un placard noir. Voici l’inscription qu’on obtint:

Ja gra Va Sa La Da La Ma Tâñ
A Ma Va Jas A Ta Krat’ Dâ
La Pa Sa Ya Ra Da Ra Cha

A Void

Its complication arising principally from its “a-vocal” quality as a jargon (that’s to say, it’s a jargon with no history of vocalisation), and thus from an implicit contradiction with any notion of pronunciation, by choosing, in imitation of that which I do know:

Ba va sa ka ma sar pa ta par da
Bi vi si ki nù si pi ti pir di
Bo vo so ko mo sor po to por do...

A matrix simulating transcription, I think that I can, by logic, intuition or imagination, polish up my original rough draft.“Vowl instantly got busy, chalking up his 25 signs on a blackboard and coming up with this:

Ja gra Va Sa La Da La Ma Tâñ
A Ma Va Jas A Ta Krat’ Dâ
La Pa Sa Ya Ra Da Ra Cha


My difficulty I put down mainly to our facing an avocat lingo avoiding vocalisation, which accounts for so many contradictions as to its pronunciation. But f, as in that famous kid’s song Buvons un coup, which is sung thus:

Ba va sa ka ma sar pa ta par da
Bi vi si ki mi si pi ti pir di
Bo vo so ko mo sor po to por do...

I pick a mould to hold our transcription, I can, by ratiocination intuition or imagination, finish up with a not so rough draft.

I don’t follow, said ABC; with Haig I only got as far as “Humpty Dumpty sat on a wall” and “Baa baa black ram”, but not “ On my third day of Christmas”.

— All right. It’ s as if you sang:
Hamp Ta Dam Pta Sa Ta Na Wal
Himp Ti Dim Pti Si Ti Ni Wil
Bo Bo Block Shop Hov Yo Ono Wol
Bu Bu Bluck Shup Huv Yu Unu Wul

to Gallic words signifying: “I’ll drink to that: my billhook’s lost but I got its shaft back”. V got going forthwith chalking up fivantwanty signs on a blackbord, and his inscription was as follows:

A Hav Gra Va Nat Wa Than Tha Hals
An Da Hav Gra Van Ma Van Jan Sa
Pan Tha Dast Wa Than Tha Rac


  • 11 Although of Polish origin, in La Disparition, Perec usually uses “poli” in the sense of “politesse” (...)

51The first point to note here is that Adair’s text doubtless reads better than mine: for example, he uses “polish” when I use his “finish up with” in a spot of Oulipian “anticipatory plagiarism”11. This, however, is an argument I am not prepared to embark upon: first, there are plenty of counter-examples; secondly, the point at issue is where polish stops and gloss begins; thirdly, it is worth remembering that we are comparing a published text with work that was mostly completed when all hopes of publication were fading fast, and, had it not already been condemned largely unread, not being, I quote the publishers, sufficiently “plausible” (another two-edged word), would no doubt have come in for some further “polishing”, although within the terms I have stated above, for as Jacques Roubaud correctly pointed out, mere avoidance of e’s in English is too easy to be of any interest.

52Fivantwainty blackboards: My complaint here is that, despite a personalized claim to the contrary (“that which I do know”), Adair clearly has not a clue of what is going on here and which I have spelt out for the English reader, namely that the words “Buvons un coup ma serpette est perdue, mais le manche m’est revenu” are given successive monovocalic treatments. This being so, it is plain that the next set of 25 syllables strictly coincide with Baudelaire’s rendering of the 26th and final chapter of Poe’s Narrative of Arthur Gordon Pym of Nantucket: “j’ai gravé cela dans la montagne et ma vengeance est écrite dans la poussière du rocher". (One “error” in Perec’s transcription — as usual probably deliberate — the accent on the first syllable of “montagne” has slipped onto the second: Ma Tân for Mâ Tañ.) It is also then evident that the decoded message is a lipogrammatic approximation of this. This is what Adair in turn proceeds to translate, which is fine, except that he has lost his reader on the way. My solution, for what it is worth, has brought a return to Poe’s text, which happens to be closer to the monovocalic version than Perec’s is to Baudelaire’s, close enough in fact to enable the original quotation to be restored without actually looking it up — this, I suspect, would be rather more difficult in French. In other words, it would be churlish to find fault with Adair’s inscription given his method; however, being creditably close to Perec’s, it is on the wrong side of Baudelaire and perhaps needlessly distant from an English original with which many readers might well be familiar.

  • 12 “docks” is the word I was looking for and now prefer.

53Bottoms up: I come now to Adair’s punch line, “From Ubu’s bottom drops his own bulk in gold.” The word we are looking for of course is “lays”, for the whole book is about dropping French eggs or œufs/e — hence my allusion to the third day of Christmas, on which occasion, it will be remembered, my true love sent to me an appropriate monovocalism, “three French hens” “a-laying”, one supposes, three French eggs. My second-best solution is next in chronological sequence: “sits” in the sense of “broods” (a term I found too ambiguous, the metaphorical usage being far more likely in such a context)12. Ambiguity is a major problem for Adair too, since one suspects that for most people, if anything is going to drop from “Ubu’s bottom”, it is not going to be a golden egg but his infamous “Merdre”!

54I should never have brought this up, you understand, had not Adair himself drawn my attention to this distasteful matter by spontaneously mentioning, pardon his French, “a furious ‘shitr:’, a notorious oath from Jarry’s Ubu Roi that’s still fairly common (...) in Cantal” (p. 213), spontaneously, I say, because what Swann actually says is “Fouchtra pour la Catarina:”... (this I left in French, on the understanding that the people of Cantal do sometimes swear in that language). Whether this unpronounceable English is the usual translation or a lipogrammatic equivalent, my suspicion is therefore that, far from vowel movement, what we have here is a consonant bowel movement. This reading is unfortunately confirmed by “bulk”, for which sense e in the OED is “roughage”, or indigestible fibrous matter in food and metaphorically elsewhere.

  • 13 Marcel Proust, Du côté de chez Swann. Paris, Gallimard, coll. “Folio”: p. 403.

55I shall refrain from imitating Swann and Mme de Guermantes’s cruel witticisms on Mme de Cambremer’s unfortunate patronymic, “cette double abréviation”13, nor shall I pause on the overtones of “scrofulous lions”. I shall merely conclude that what is at stake here is bulk in the everyday sense of volume, or if you will, more technically, the thickness of a book without its covers, inasmuch as a certain amount of extra material is required to bridge the gaps in this translation, another, but neither the last nor the least, falling unavoidably on this very same page 213:

La Disparition

... non sans aplomb, il improvisa tout un pot-pourri (alla podrida), fignolant surtout un air fort connu, la Polka du Mitard, chanson du jour dont voici la coda:

Alors qu’il dit, on fait quoi
J’y dis Va savoir faut voir
D’ac pour voir mais pour voir quoi
Ça j’sais pas j’y dis faut voir

Hors pour voir, il faut la cour
Faut qu’ça cass’, ou sans ça scions
J’y pourvois par la scansion

Mais j’suis pas plus affranchi
Pour autant qu’il dit j’y r’dis
Cours au mur si tu l’franchis
Où qu’y’a un mur qu’il dit

Hors tout autour y a la cour
Faut qu’ça pass’ faut qu’ça partout
J’l’y convois dans la Chanson:

Bravo: Bravo: Bravissimo: applaudit la Squaw.
— Il suffit:

A Void

... actually starts improvising, and not without aplomb, a potpourri of military music.
Squaw wildly applauds him. “Bravo: Bravo: Bravissimo:
“All right, all right, that’ll do”


... aplomb to risk improvising a right potpourri (alla podrida), notably turning his lips round a famous ditty, a hit song of that day, JailBird-Song of Alcatraz, which tails off as follows:

So, says him, what’ll I do?
Says I, I dunno, go an ‘ look;
OK, I’ll go an’ look, but look at what?
Dunno, do I? says I, go an ‘ look

What:go an’ look out? I wanna court
You gotta quash or I’m a squash
This I scan, him I taught

But I’m still not out
Says him; says I agin
Run to t’wall, you can jump
What wall? says him, t’silly lump

Now out ‘tis nowt but t’ard
You gotta rush you gotta dash
This you can, sang I his Bard

- Bravo: Bravo: Bravíssimo: said Squaw, applauding him warmly.
- That’ll do now:

56To bridge and not abridge: One last comment on “fill-ins", taken from E.V. Wright’s Gadsby, quoted, doubtless with some self-referential intention, on page 63 of La Disparition. This I scan, as you please, from Perec’s “anglais", my own “Midatlantic” or Adair’s “Pastoral” (an odd school subject which again implicitly offers original English as a translation):

57It is a story about a small town. It is not a gossipy yarn; nor is it a dry, monotonous account, full of such customary “fill-ins” [...] No...

Haut de page


1 John Lee, N°3, TLS, 2-8 Sept. 1988, p. 958, excerpt below; PN Review ° 68, 1989, p. 18-19, poems by Mallarmé, Hugo and Rimbaud; Parcours Perec, Colloque de Londres - Mars 1988, “La Disparition, problèmes de traduction”, followed by a translation of pages 22-25, Lyon, PUL, 1990, p. 109-126; “Ce qui dit art maniaque dit oui”, Études littéraires, Vol. 23, N° 1-2, Été-automne 1990, Université Laval, p. 79-86; Sara Greaves, “De La Disparition de Georges Perec à Vanish ‘d! de John Lee: la traduction traduite”, Palimpsestes, N° 9, Paris, PSN, 1995, p. 105-115; Textes de référence, p. 34-47; Sara Greaves, La Traduction d’un lipogramme, Étude de la traduction anglaise de La Disparition de Georges Perec intitulée Vanish’d!, et réalisée par John Lee, mémoire de maîtrise dirigé par Anne Roche, Université de Provence, 1989, 174 p.

2 Honesty dictâtes that any quote I make from Adair should be accompanied by the corresponding passage from my own version. However, for reasons which will become clear, I am deferring this until the reader has read on.

3 Even as I write, I detect that Perec is within 0.006% of 2 Mayan Baktuns, equal to 40 Katuns (288,000 days), with my own margin of error 0.01%. This is probably no coincidence in the context (Squaw praying to the “Grand Manitou” in a story divided into 20-year katun-like periods) as I shall explore elsewhere, and meanwhile suggests the two above as alternative translations. This analysis is further borne out by the following dating from the Chilam Balam (referred to in connection with a matter discussed below on p. 197 of La Disparition): “8 Ahau fut le Katun (1184-1204) où ils arrivèrent, les survivants des Itzas...” Les prophéties du Chilam Balam, version et présentation de J.M.G. Le Clézio, Gallimard, Le Chemin, 1976, p. 108.

4 The reader is now in a position to appreciate my awaited quotation without distraction; in addition to my homemade System, it combines a variety of ploys — currency exchange, non-translation, Adair’s Arab numerais and proportional multiplication: ‘“I got two-fifty on Quillpush.’ You had had twainty-six running, so only fivantwinty in starting stalls, what with Whisky Dix, carrying a 5 on its back, scratching. Whisky Dix had had many punting folk’s fancy, although against long odds of thirty-six to two.” Here my “long” merely reinforces “although”, as does Adair’s “surprisingly”; his “official” is his own addition (“quoiqu’il cotât”).

5 My pocket calculator when held upside down gives an interesting resuit had he only taken the closer figure: EEE’EEEEE.

6 The comment “[GP’s Franglais in bold italics]” followed the first occurrence of this.

7 Georges Perec, “Histoire du lipogramme” dans Oulipo, La littérature potentielle, Paris, Gallimard, 1973, p. 78. Cf. also John Lee, “Vingt-trois shan du grand Li Po” dans Nuova corrente, N° 108, Genoa, Tilgher-Genova, 1991, p. 317-347.

8 “The disparition of three final stars”, James Joyce, Ulysses.

9 That this should be at the heart of the lipogram is borne out incidentally by the value of log e, 0.4343.

10 Georges Perec, W ou le souvenir d’enfance, Denoël, 1974, p. 57.

11 Although of Polish origin, in La Disparition, Perec usually uses “poli” in the sense of “politesse”. Cf. John Lee, “NdT”, lecture given at the University of Provence in May 1991: “Puisque le principe du blanc produit du bancal, avatar du clinamen, il devient urgent pour le traducteur, en tant qu’organisateur second de l’écrit, de différencier ce phénomène du n’importe quoi, de gérer cette difficulté plutôt que de la masquer derrière une prose trop lisse. Le lipo est le contraire de poli: c’est d’ailleurs pourquoi le roi est mat: Ce qui n’exclut pas la courtoisie; au contraire, les remarques qui vont suivre peuvent se comprendre comme les politesses de l’auteur envers son lecteur, et traiteront de la difficulté spécifique qu’elles présentent au traducteur soucieux de respecter cette qualité, trop rare, du texte source. Une telle courtoisie réunit au moins deux aspects: d’une part elle consiste à ne pas prendre le lecteur pour moins malin que soi; d’autre part, il s’agit de ne jamais oublier que le lecteur est un autre et qu’il a besoin d’une inscription sur la page pour comprendre ce qui se passe dans votre tête.”

12 “docks” is the word I was looking for and now prefer.

13 Marcel Proust, Du côté de chez Swann. Paris, Gallimard, coll. “Folio”: p. 403.

Haut de page

Pour citer cet article

Référence papier

John Lee, « La Disparition: Problem Translations. The Rough and the Smooth », Palimpsestes, 12 | 2000, 137-157.

Référence électronique

John Lee, « La Disparition: Problem Translations. The Rough and the Smooth », Palimpsestes [En ligne], 12 | 2000, mis en ligne le 01 janvier 2000, consulté le 24 février 2018. URL : ; DOI : 10.4000/palimpsestes.1649

Haut de page


John Lee

John LEE vit depuis 1975 en France où il est traducteur-interprète. Après quelques traductions littéraires (Georges Perec, Henry James), il s'est surtout spécialisé dans des domaines aussi variés que les engins de manutention, le débarquement en Normandie, les chaudières, la contrefaçon, la mémoire chez la seiche, l'art contemporain, les hommes en noir ou encore la santé mentale.

Articles du même auteur

Haut de page

Droits d’auteur

Tous droits réservés

Haut de page
  • Logo Presses Sorbonne Nouvelle
  • OpenEdition Journals