When a young man came up to him in Zurich and said, ‘May I kiss the hand that wrote Ulysses?’ Joyce replied, somewhat like King Lear, ‘No, it did lots of other things too.’ (JJII: 114)

In the wake of the Joyce 2002 Papers acquired by the National Library of Ireland, attention has been refocused on the composition of Ulysses. Pre-fair-copy drafts are now known to be extant for all but eight of the eighteen episodes of Ulysses and this paper examines one of the episodes for which there is still tantalizingly little prepublication material known to be extant, “Wandering Rocks”. As such it offers a speculative genetic analysis of the missing documents, by focusing on their echoes, antecedents and precedents in the extant avant-texte. A survey of the paratextual apparatuses that contributed to the episode (newspapers, a children’s board-game) is offered and the impingement of these elements on the composition of “Wandering Rocks” will be explored.

“I cannot dictate to a stenographer or type,” Joyce wrote to John Quinn in May 1917. “I write all with my hand” (LII: 396). Three years later the manuscript of the “Wandering Rocks” episode, sent to Quinn from Paris, was to refute this claim, as it is written in both Joyce and Frank Budgen’s hands. The letter was written at a time when Quinn, keen to assist in the launching of a writer whom he suspected of talent, first began accumulating a collection of Joyce material. In March 1917 he quarrelled with Edmund Byrne Hackett over the Egoist sheets, which B. W. Huebsch had used to set up Portrait. Hackett, a Clongowes schoolmate of Joyce’s, had brought the sheets to Huebsch and after publication retained them. In response to Quinn’s offer of “ten or twelve pounds” on the sheets, Hackett set a price of $100 (£20) on the manuscript; half to be paid to Joyce, half to him. Eventually dispelling Hackett’s claim to the sheets or to any part of “the fine piece of philanthropy” he had in mind for Joyce, “the Godly J. Quinn” cabled Joyce directly on March 13, doubling his initial offer of ten pounds.² Joyce promptly accepted the offer writing that “Mr Pound will cable you asking you to send the money through him, as that seems to me the shortest way” (March 19: LI: 100). Quinn was nettled by what he saw as Joyce’s suspicion of him when Pound’s cable (“Joyce accepts. Money to be sent via me. Pound”) arrived weeks before the letter. On April 11, when the letter finally arrived, he wrote conciliatorily, however, inquiring after Joyce’s health and congratulating him on the success of A Portrait of the Artist as a Young Man. He also registered a continued interest in Joyce’s manuscripts:

Is that [the MS of Exiles] in typewritten form? I should like to read it if it is not too inconvenient. Of course it has not been published yet. How do you do your plays? Do you dictate them on the typewriter to a stenographer, or do you write them on the typewriter? Or do you write them and have them copied? What I am getting at is that occasionally I collect manuscripts (but not as generally as I used to). And if your play is in manuscript or the “Portrait” is in manuscript, I should be glad to buy the MS. if you care to sell it. That is partly an interest in MS. but more an interest in James Joyce.

Here is the context for Joyce’s description of his writing processes in the May letter. The £20 remitted through Pound reached him on April 30 and ill with glaucoma he responded, via Nora, that “it seems […] unfair that you should now pay $100 instead of $50 you first offered” (LII: 395). He ‘begged’ Quinn to accept the pages of corrections for Portrait and Dubliners as part of the transaction. In mid-May, recovered from the glaucoma attack, he himself wrote outlining both his compositional practises and the dispersal of his manuscripts and hence their unavailability for sale: “I have here with me only the MS of Exiles. The MSS of Dubliners […] and of A Portrait of the Artist are in my desk [in Trieste] and so I cannot easily get them for the moment” (LII: 396). The brevity of his letter and his reliance on Nora as a scribe Joyce put down to glaucoma, adding “it seems better and I can write again.” In addition to the description of his compositional methods, he also explained, “When the fair copy is ready I send it to a typist” (as if to instantiate this process the version of the letter extant in the New York Public Library is a typed copy Quinn had prepared for an American eye specialist).³ The end result of the exchange of letters was that Quinn bought the manuscript of Exiles for £25. He complained of the document’s illegibility to Pound, who agreed with him, writing that Joyce was “quite cracked in suggesting that his unutterable scrawl COULD be deciphered at ANY price whatsoever.”⁴ Expressing himself less forcefully to Joyce, Quinn observed, “your handwriting is not the clearest in the world.”⁵

It was not until 26 June 1919, when the episodes written amounted to those up to and including “Sirens”, that Quinn wrote to Joyce asking how he cared to “dispose of the manuscript of Ulysses.” With cash readily available for manuscripts, Joyce was, from very early on, attuned to the appearance of his papers as potential sale copies (that he knew the full extent of Quinn’s dissatisfaction with the appearance of Exiles is evident from his limerick to Sykes: LII: 406).⁶ While the scope of this paper does not extend to a treatment of the Rosenbach Manuscript sale-copy portions (“Wandering Rocks” is after all in-line) the concern for the appearance of his manuscripts, borne out of his dealings with Quinn, offers an explanation for the coda Joyce appended to the end of the draft: “pp. 32–48 were written by my friend Francis Budgen at my dictation from notes during my illness January – February 1919 | James Joyce” (U-syn 546.28n).⁷

As is well known, “Wandering Rocks” is unique among the sections of the Rosenbach Manuscript in that it is written in two hands. “Nestor” and “Proteus” both contain marginal additions in, presumably, Claud Sykes’s hand, copied from Joyce’s postcard instructions. The verso of a “Circe” sheet, fol. [29v], on the other hand, contains an approximation of Sarastro’s aria from Mozart’s Die Zauberflöte in what may be Giorgio Joyce’s hand, but “Wandering Rocks” is the only portion of the manuscript produced, in part, by inditer and scribe. Despite the claim in his letter to Quinn, the use of amanuenses was not unknown to Joyce. Among the prepublication Ulysses materials there are notes in Italian in Lucia’s hand in NLI 36,639/5B, one of the notebooks that make up the Joyce 2002 Papers. Over the years, as his eyesight failed, Joyce’s reliance on amanuenses increased: in Nora’s hand there are the “St Kevin” sketch in Buffalo notebook VI.B.3.42–45 (see Buffalo VI.B.3, pp. 46–48) and the composite “Tristan and Isolde” / “Mamalujo” sketch among the Joyce 2006 Papers. The composition of the Wake in particular involved a coterie of “anticollaborators” (FW 118.25–26) the recollections of whom range from Stuart Gilbert’s acerbic diary-account of taking dictation from “The Great Man” “curled on his sofa […] pondering puns” to Beckett’s most-likely apocryphal “Come in” (JJII: 662).⁸

Budgen has left us his own account of the process in James Joyce and the Making of Ulysses. While he evidently witnessed the composition of “Wandering Rocks” at first hand, he does not allude to his involvement in the preparation of the manuscript. Instead the reader is treated to an early version of the metaphor of Joyce the engineer “at work with compass and slide-rule, a surveyor with theodolite and measuring chain or, more Ulyssean perhaps, a ship’s officer taking the sum, reading the log and calculating current drift and leeway.”⁹ How much of this is merely figurative is uncertain since Budgen goes on to describe the paraphernalia of “a map of Dublin before [Joyce] on which were traced in red ink the paths of the Earl of Dudley and Father Conmee. He calculated to a minute the time necessary for his characters to cover a given distance of the city” – a concern with the niceties of verisimilitude that can only have given succour to Clive Hart as he wandered one-legged-sailor fashion around Dublin with a stopwatch and copy of Ulysses.¹⁰ We know that for a period of at least “five weeks” in January and February 1919 Joyce could do “little or nothing except lie constantly near a stove like a chimpanzee” (LII: 437) but nowhere in his testimony does Budgen mention the drafts or “notes” to which sore-eyed Joyce might have directed him during this period.

Rather we hear of the children’s game “Labyrinth” (see figure) bought at Franz Carl Weber’s toyshop on the Bahnhofstrasse (Weber’s 1914 catalogue gives the name of the game as “Labyrinth-Spiel”) which Joyce played “every evening for a time with his daughter Lucia.” The playing board for this game, on proud display at the Zürich James Joyce Foundation, is a honeycomb of interlocking but unclosed hexagons that form a multicursal route from one side of the board to the other. The object of the game was to traverse the field and then return to the starting point, the player being occasionally obliged to miss a go or two when his (or his daughter’s) token was moved into a hexagon marked with one or two dots. Budgen records that as a result of winning or losing at the game Joyce “was enabled to catalogue six main errors of judgement into which one might fall in choosing a right, left or centre way out of the maze.”¹¹ Since Joyce had Budgen assisting him, we are bereft of any letters he may have cared to write to the latter apropos the composition of “Wandering Rocks” and in the absence of substantial prepublication material what we have is an avant-jeu de société.

Since the emergence of the Joyce 2002 Papers and other recent unearthings, there are still eight episodes of Ulysses for which we have no early documentation other than the Rosenbach Manuscript , namely “Telemachus”, “Nestor”, “Calypso”, “Lotus Eaters”, “Hades”, “Aeolus” “Lestrygonians” and “Wandering Rocks”. In the case of “Telemachus” a related document, the early Martello Tower fragment, is extant (located at the British Library). For several of these episodes the situation is complicated further by the fact that their one element known to be extant is the out-of-transmission Rosenbach Manuscript portion. But despite its inline manuscript, the avant-texte of “Wandering Rocks”, that chapter of surfaces, comprises very little material for the genetic critic’s depth-plumbing analysis. One explanation for this paucity might be that “Wandering Rocks” was the last episode to be planned. When Joyce wrote to Weaver in May 1918 he described the broad structure of the book as he then saw it: “the Telemachia [sic], that is, the three first episodes. […] The second part, the Odyssey, contains eleven episodes. The third part, Nostos, contains three episodes. In all seventeen epiodes” (LI: 113). Unless Joyce had simply made a numerical error, it seems that one episode of the book was not conceived of prior to May 1918; “Wandering Rocks” a centrepiece added to Ulysses that offsets the possibility of a centre by bringing the number of episodes up to an even number.¹² There is no centre to the labyrinth of Ulysses, only the difficulty going in (or labor intus to give the standard medieval etymology).¹³ Structurally outside of Homer’s Odyssey, explicit evidence for “Wandering Rocks” cannot be found in Joyce’s early notebooks for Ulysses.¹⁴

The earliest Ulysses notebook known to be extant, the so-called “Subject Notebook” NLI 36,639/3, does in fact contain a small cluster of “Wandering Rocks” entries; a scant four lines of notes headed “Names and Places” on p. [12r] contains the notes “Sherlock” (U 10.1011), “bQuigley” (U 10.1125) and “bO Connor (Wexford) palmnut meal”. The first two notes appear in the pages of the Rosenbach Manuscript written by Budgen (fols. 35 and 40 respectively). The third note was transferred to a page of “Wandering Rocks” notes in a later notebook, NLI 36,639/5A, as “O Connor Palm Nut Meal Wexford”. There it is crossed out in blue and enters the text as a second-round addition to the typescript as “with sacks of carob and O’Connor’s palm nut meal, O’Connor, Wexford” (TS V.B.8, fol. 9, JJA 13:15; U-syn 496.9–10, now U 10.434–35). While a cluster of “Wandering Rocks” notes occurring in four-lines of notes is suggestive of early planning, they can be found on a page containing, as the title indicates, proper names and business premises. It is hardly surprising that several such notes end up in the episode dedicated to the city itself (other notes on the page can be found in episodes as diverse as “Calypso”, “Circe” and “Ithaca”).¹⁵ Postulated elements of the “Lost Notebook” that end up in “Wandering Rocks” include details from the entry on Silken Thomas in the Dictionary of National Biography and an account of the General Slocum disaster lifted from The Times.¹⁶ The next notebook known to be extant, Buffalo commonplace notebook VIII.A.5, with its entries garnered from Joyce’s Zentralbibliothek reading around the subject of the Odyssey, does not contain material for “Wandering Rocks” to quite the same degree that it does for the other adventures (though Herring does trace the blue-crossed entries “blub lips” on p. [10v] and “ZYZI blub lips” on p. [17r] to “Wandering Rocks” U 10.1273, which is present in the Rosenbach Manuscript, fol. 47.¹⁷ The duplication of “blub lips” found in “Lotus Eaters” (“sitting round in a ring with blub lips, entranced”) incidentally, is much later; added around 30 June, 1921 on the first setting of gathering 5; JJA 22:247, U-syn 160.16, now U 5.335–36). There are no “Wandering Rocks” notesheets known to be extant (there may never have been any) and the four late notebooks Buffalo V.A.2, NLI 36,639/4, 36,639/5A and 36,639/5B contain among them only three pages headed “Wandering Rocks”: 36,639/4, p. [5v]; 36,639/5A, p. [7r]; and V.A.2, p. [10r].

If one brackets off the additions made to the typescript exemplars and proofs in Joyce’s hand, then really the only substantial body of autograph material available for “Wandering Rocks” is the Rosenbach Manuscript. Because of the particular appearance of the manuscript, however, it is stratified in a manner that lends itself to genetic criticism. Joyce wrote the sections dealing with the “Roman” up to the end of the fourteenth section, U 10.954; Frank Budgen wrote Martin Cunningham’s exit from the Castle to the “British ensign” ending the episode. Not only are fols. 32–48 written in Budgen’s hand at “dictation from notes”, but together he and Joyce went back and revised the entire manuscript, including the earlier portion in Joyce’s hand, making additions and corrections. Then, when he had recovered from his illness, Joyce went back over the entire Budgen-revised manuscript. As later scenes were written, earlier pre-emptive intrusions of the same material had to be updated. So, while no earlier draft survives, by stripping away the later layers of additions in Budgen’s hand, a draft within the draft of “Wandering Rocks” can be discerned. By focusing on the earlier stratum of the intrusions, pre-Rosenbach Manuscript versions of the corresponding sections emerge. Within the plane of the manuscript errant wanderings across the board are inscribed. While the issue of why there are no early drafts of the episode remains unresolved, by examining the interpolations in this way, intimations are given of fragments of a lost draft or set of notes.

The first indisputable interpolation, Denis J. Maginni “walking with grave deportment” past Dignam’s Court (U 10.56–60), does not appear in the Rosenbach Manuscript. It was one of the last passages to be added to the chapter, as Fritz Senn has pointed out,¹⁸ in September 1921 (JJA 24:20). In any case, the three instances of Maginni (all of which were added on the page proofs), exist in a marginal relation to the rest of “Wandering Rocks” in that they point outside of the episode, for the reader to “Lestrygonians” U 8.98 (this, the first appearance of Maginni, was the last written, however: early October 1921, JJA 23:140) and for the genetic scholar to the earliest extant draft of “Circe” (where Professor “Maghinni” first “presents himself” in the text of Ulysses: Buffalo V.A.19, p. 21r, JJA 14:235). Similarly, the three instances of the crumpled throwaway’s progress down the Liffey, in sections four (U 10.294–97), twelve (U 10.752–54) and sixteen (U 10.1096–99) exist in a vacuum in relation to the rest of the episode (they also index “Lestrygonians” U 8.57). They are divorced from the action, acting as tokens of time’s passage rather than its simultaneity. All three instances appear on the Rosenbach Manuscript as additions in Budgen’s hand. The first two seesaw about a chiasmus: “A skiff, a crumpled throwaway, Elijah is coming” passes the quays (U-syn 486.34–487.2) and then the quays further east are passed by “a skiff, a crumpled throwaway, […] Elijah is coming” (U-syn 514.21–23). In the third instance this constant element undergoes condensation: “Elijah, skiff, light crumpled throwaway” (U-syn 534.36). The second occurrence of the throwaway replaces what was the first element of another set of interpolations: the cockle-pickers’ return from Sandymount Strand. Now mentioned only in sections thirteen (U 10.818–20) and nineteen (U 10.1274–77), the two women feature in section twelve of the manuscript, in an interpolation which was heavily revised before being discarded altogether.

Joyce caught the mismatching number of cockles but it was not until Budgen was revising the manuscript that the word “bag”, presumably lost in the transferral from an earlier document, was restored to the earlier interpolation. Joyce also had Budgen correct the motion of the cockles in line with the second interpolation and the final appearance of the cockle-pickers at the end of the episode: “two dusty ^sanded^ women halted themselves, an umbrella and a bag in which eleven cockles rolled” (U-syn 546.19–20: in Budgen’s hand). As an interpolation in the second Stephen section, the old women’s progress through Irishtown is accordingly cast in the language of “Proteus”, the women are “weary” (U 3.464, 468) and their bag is “a midwife’s” (U 3.32: “Wandering Rocks” also supplies an authorial gloss on “gamp”).¹⁹ The discarded interpolation in section twelve, however, while it may offer a glimpse of an earlier version of this paragraph, does not give a neutral vision of the cockle-pickers and their “bonnets” (U 7.935), refracted as it is through the “Aeolian” lens of Stephen’s Parable of the Plums (itself indebted to the scene on the beach). Perhaps this explains its ultimate absence from all but the synoptic edition of Ulysses.

Let us return to the first section of the episode; it features two versions of Father Conmee boarding the outward-bound tram. The second of these is present in the body of the Rosenbach Manuscript (in Joyce’s hand: U-syn 476.14–16) though the bridge is initially given as “Annesley bridge”. The interpolation of this action into section two is also in the manuscript (U-syn 482.12–13) where Conmee steps, again, onto a tram at “Annesley bridge” as Corny Kelleher lolls against a doorframe. It was not until the first portion of the manuscript was being revised in Budgen’s hand, however, that the verbal and visual chiasmus of two Christian clergymen boarding and disembarking on Newcomen Bridge was established (U-syn 476.8–13). Joyce had caught the error of “Annesley” for “Newcomen” in the passage interpolated into section two and made Budgen change it. The Conmee section remained unchanged, however, until Joyce himself was revising the manuscript after he recovered from his illness. This oversight suggests that it was the interpolation that prompted the additions in Budgen’s hand and not the U 10.113–14 account of Conmee boarding the tram (the spatial arrangement of the material on the folio makes it clear that Joyce’s emendation postdates the scribal addition: Facsimile MS “Wandering Rocks”, fol. 3). The second bridge, where the old woman alights, is given as “Newcomen” in the draft and was not corrected until Joyce was revising the typescript for serial publication (the relevant typescript page is not extant but the Little Review text gives “Annesley”, Little Review vi.2, p. 37).²⁰

Section nine, Lenehan and M‘Coy’s walk through Temple Bar, is punctuated by numerous interpolations and is the sectional origin for several in the other sections. One of the former category, an interpolation from section eighteen, describes Patrick Dignam on William Street. In Joyce’s first version Master Dignam stands at Mangan’s counter waiting for pork steaks (U-syn 502.6–8). The sectional origin of this, in Budgen’s hand, sees Dignam having left the butcher’s and walking along “warm Wicklow street dawdling” (U-syn 536.26–27: a cancelled ‘F’ in the manuscript implies that Budgen almost restored the “late” name to the premises). When he had recovered from his illness Joyce went back over the entire Budgen-revised manuscript and worked the language of section nineteen into the section nine interpolation. Mangan’s became “late Fehrenbach’s”; young Dignam carries “a pound and a half” of pork steaks. While the interpolated version is more simplistic, it cannot with conviction be called a genetic antecedent to the opening of section nineteen. For a more compelling example of this phenomenon we turn to an interpolation in section eleven, the meeting of Simon and Dilly Dedalus outside Dillon’s auction-room. Simon pauses, tongue in cheek, father and daughter on a spotted hexagon, as it were, while the Rosenbach Manuscript narrator tells us, “Mr Kernan, pleased with the order he had booked, walked boldly along Thomas street” (U-syn 510.8–9). This sentence is repeated word for word at the opening of the following section. For “five weeks” or so (LII: 437) the sentences remained the same until, with his eyesight restored, Joyce made a series of revisions specifying place, “From the sundial towards James’s gate”; Kernan’s destination and employer, “Pulbrook Robertson”; and changed the street name from Thomas to James’s Street (Facsimile MS “Wandering Rocks”, fol. 23).²¹ The earlier genetic state, the interpolation to section eleven with its mention of “Thomas Street”, remained in the text until Joyce was correcting the typescript for serial publication (Little Review, vi.3, p. 33; the change is extant on Darantiere’s copy, JJA 13:20). The two occurrences of this sentence appear on the rectos of adjacent folios in the manuscript (fols. 22–23); one wonders whether the positing of an exterior document (a draft or a notebook) would be justified. Rather, the sense is of the manuscript of “Wandering Rocks” acting as its own note repository, as a notebook whose elements can move ‘upstream’ to be redeployed in later sections. Granted, the fluidity of the Ulysses notebooks is absent in the manuscript (and of course transferred elements are not struck through); we do not find disparate entries speculatively constellated under Homeric rubrics, but rather fragments deliberately introduced (the only word for it is ‘interpolated’) under the asterisk-string section dividers in accordance with whatever timetable Joyce had devised. For certain portions of the text, it seems, the draft acts as its own storehouse of material.

Simon Dedalus and Father Cowley’s meeting on Lower Ormond Quay in section fourteen (U 10.882–83), interpolated into section twelve (U 10.740–41), provides a comparable example. In this case the repetition abides in the final text but to arrive at that concordance involved a degree of jury-rigging. To Cowley’s salute, as it appears in the manuscript interpolation, Dedalus merely “answered” (U-syn 514.8–9). In the section he answers “stopping” (U-syn 522.21–22). To this Joyce adds, on the typescript copy for Darantiere, Cowley’s question “How are things?” (TS V.A.8, fol. 18, JJA 13:24.) The passage interpolated into section twelve is now out of sync on two counts. This was rectified on the first setting of the passage as placard 26 (JJA 18:270). Such word-for-word correspondences in the final text indicate events which are happening simultaneity. But “Wandering Rocks” also contains intrusions which act as tokens of time’s passage in the episode: Maginni’s progress from Great Britain Street to Grafton Street; the wending of the throwaway down the Liffey; and the two bicycle races around the closed circuit of Dublin University College Park.

In “Eumaeus” Bloom reads the result of the Gold Cup in the “pink edition extra sporting” (U 16.1232) of the Evening Telegraph for Dublin, 16 June 1904, “on page three, his side” (U 16.1276–77). If he had looked over at the next column he would have found himself reading the “Wandering Rocks” episode of Ulysses for, as is well known, the two accounts of the wheelmen’s gyring are lifted from the last pink.²²

The account concludes with a two and three mile handicap and the “One Mile Bicycle Club Championship”. For the half-mile race, U 10.651–53, an addition made to the Rosenbach Manuscript, fol. 21, in Budgen’s hand, Joyce conflated the four winners of the two half-mile handicap heats. For the quarter-mile (i.e. 440 yards), U 10.1258–1260, a marginal addition on the Budgen-inscribed portion of the Rosenbach Manuscript (in Budgen’s hand), fol. 47, the names of the race’s winners and also-rans are listed off in the order printed until, coming to “J. J. Comyn” and perhaps wishing to avoid an avatar of himself, Joyce retreated to the top of the column and plucked Huggard from the first unmined race, the first heat of the 120 yard handicap. As has been observed elsewhere, the newspaper’s “Greene” became “Green”, “Jones” “Joffs” and “Adderley” “Adderly” (U-syn 546.1–3) in the transfer to the Rosenbach Manuscript. But those who would replenish the ranks of the collegiate cyclists and restore errant initials from the Evening Telegraph report must be wary of literals and compositors’ errors. Maurice Cherry Greene appears variously as “C. M. Greene”, “M. C. Greener” and finally “M. C. Greene”. George Newcomon Morphy, if one were to judge from his presence in the first heat of the half-mile and in the “Three Miles Flat Handicap”, was a spud Murphy. Small potatoes perhaps, considering that the uncommon name Morphy was accurately recorded in the last pink’s account of the 440 yards handicap and from there worked into “Wandering Rocks”, but also an instance of the newspaper’s “bitched type” (U 16.1263) problematising attempts (whether by Joyce or his editors) to bring the text of Ulysses in line with the historical reality of that day in June 1904.²³ Did Joyce know that this was one of the only races reported in the newspaper that actually got the participants’ names right? I doubt it. More likely, he was interested in indicating the passage of time by juxtaposing the opening race with one further down the programme (the 440 yards is also one of the few races that doesn’t feature cyclists who also competed in the half-mile handicap). Otherwise, whither “Green” and “Adderly”? Whom are we to blame for the dropped silent es and the transformation from “Jones” to “Joffs”? The hand is Budgen’s but the inditer is Joyce. It is difficult to imagine him, however, reading the miniscule print of the newspaper to Budgen “during my illness” and given the deft use of his source, it seems unlikely that Joyce merely directed his scribe to the column. An intermediary first-order or later notebook or a draft can be posited between the newspaper and Budgen’s caligraphy.²⁴

John Kidd’s labyrinthine spiel about “the old boy of quad and green” Harry Thrift, while it provided convincing rhetoric for his “first salvo” in the Joyce Wars, rests on an error that has nothing to do with Gabler’s editorial practices or the use of facsimiles.²⁵ Consulting the manuscript (or its facsimile) reveals that a graphological peculiarity of Budgen’s hand is responsible for the error: two fundamentally different capital Ts appear almost side-by-side in the words “H. Thrift, T. M. Patey” (Facsimile MS “Wandering Rocks”, fol. 47). The first T of “Thrift” is a cursive whereas the spelt-out initial letters result in a two-stroke tau cross. The Munich team concluded that these were two different letters. They plumbed for an S in the former case and scandal ensued. However, the first word of this same folio of the manuscript, “Thither”, should have resolved this crux for them: it begins with a cursive capital T similar to that found in “Thrift” (and there are any number of other examples in the seventeen pages of the manuscript that are in Budgen’s hand).

While “Shrift” is demonstrably wrong because Budgen wrote “Thrift”, how are we to approach the other wheelmen? Are the names “Green” and “Adderly” wrong because these were real men whose names were x-ed in the transfer into notebooks or the Rosenbach Manuscript? In short, the issue is the degree of verisimilitude we demand of Ulysses. The same episode of the book moves the Mirus Bazaar inauguration to 16 June from 31 May 1904 and calls the Grand Canal the Royal (U 10.1273). If plausible reasons can be put forward for these particular distortions (part of the “disingenuous simplicity” of the episode’s narrator, in Clive Hart’s words) what of apparent slips of the pen?²⁶ To be “nettled not a little” by Greens and Adderlys “(as it incorrectly stated)” (U 16.1262) is to expect of Ulysses a fidelity to minutiae while at the same time condoning the more overt liberties it takes with historical events and places. The appearance of “L. Boom” (U 16.1260) in the fictional Evening Telegraph would seem to sanction such newspaperly or mediatory divergences from accuracy. “These are not misprints but beauties of my style hitherto undreamt of” (LI: 187) Joyce wrote to Weaver in September 1922, a tone which may suggest, in the light of the several months he had spent compiling errata, a softening of his attitude towards certain types of error. For, more so than towards the underpining reality of Dublin, 16 June 1904, Ulysses is textually orientated; first outwardly to Thom’s and documents like the last pink (and from the former of which it inherits a botch of misprints) but, more fundamentally, its emphasis is on an internal textuality. W. E. Wylie is not, after all, William Evelyn Wylie another “man in the bicycle race in Ulysses” for most readers so much as the brother of Gerty MacDowell’s erstwhile sweetheart, wily Reggy.²⁷

A twenty-five-page typescript was prepared from the manuscript and sent to Pound “some days” before February 25 (LII: 436); portions of two of at least three exemplars are known to be extant (U-syn 1739). A further page of typescript (which Hans Walter Gabler lists as lost: U-syn 546.18n) is item 59 in Bernard Gheerbrant’s “Catalogue de l’exposition ‘James Joyce et Paris’”:

Les Errants (p. 244 de l’édition originale d’Ulysses [i.e. U 10.1247 to the end of the episode]). Aux dernières lignes dactylographiées Joyce ajoute des corrections et cinq lignes nouvelles autographes.²⁸

Gheerbrant had earlier curated and catalogued the “James Joyce: Sa Vie, Son Œuvre, Son Rayonnement” exhibition in the Librairie La Hune, Paris but the page of “Wandering Rocks” typescript was exhibited at “James Joyce et Paris” in 1975; long after the University at Buffalo had acquired the two extant typescript exemplars as part of the Sylvia Beach collection in the winter of 1959. The longer of the two fragments, TS V.B.8a, is missing only its third and final page and it is clear that Gheerbrant’s item 59 is the final page of this exemplar. The Archive reproduction of the 8a typescript ends on fol. 24 with Mr Eugene Stratton’s “blub lips” (U 10.1274, JJA 13:30) and in the Rosenbach Manuscript the typist, presumably, underlined the next word “agrin” to denote the start of a new page (Facsimile MS “Wandering Rocks”, fol. 47). The end of fol. 47 and entirety of fol. 48 – the final nine lines in the Gabler edition – thus make up a twenty-fifth typescript page; the “cinq lignes nouvelles autographes” refer to additions made to this page which are not present in the Little Review (vi.3, p. 47) but which do appear in the first setting of this passage in proof as placard 28 (JJA 18:305). These five lines, which Gabler accordingly gives as second-level typescript overlays, can be inferred as Lansdowne/Landsdowne Road (which is thus either Joyce or the compositor’s error) and “the house said to have been admired by the late queen when visiting the Irish capital with her husband, the prince consort, in 1848” (U-syn 546.22, 25–26. In the first setting of placard 28 the year of the queen’s visit is changed to “1849”, JJA 18:305). This final page is now in a private collection in the UK.

No indication as to the whereabouts of the missing fol. 3 of Darantiere’s copy of the typescript is forthcoming, but through collation of the Little Review and relevant placard its autograph contents can be inferred. Five discrete additions were made to this missing page, additions not in the Little Review (vi.2, p. 37) but which are all found in the first setting of this passage in proof (placard 24, JJA 18:235). Therefore they must have been added on Darantiere’s copy of the typescript (Gabler has them all down as second-round typescript additions: see U-syn 476.8–478.31 for his synoptic edition of the missing page). While the typescript may be irretrievably lost, each of the additions has an extant material antecedent: each can be traced to a source in NLI 36,639/5A, p. [7r]. All of the blue-crossed entries on this page, which is headed “7. Wandering Rocks”, seem to have been added on the second-round typescript. The five additions made to fol. 3 are reproduced below; bold indicates the words that derive from entries in the notebook.

The first of these additions, and the longest, can be traced to two spare notes on p. [7v] of the notebook: “bFr Conmee & ticket inspector” and “bIvy church (N. Strand)”, the first of which stood as a mnemonic for the longer scene. The last addition, by contrast, is almost entirely composed of material already present in the notebook. It is built out of a cluster of blue-crossed entries occupying four lines around the bottom of the page; Joyce needed only to add a few connecting phrases in the transferral from notebook to typescript.

One possible feature of the missing page of typescript, not attested to by any autograph copy, is the alteration of “Mr Eugene Stratton grimaced” at U 10.141 to “grinned”. Hans Walter Gabler gives the former reading, going on the authority of Rosenbach Manuscript “Wandering Rocks” fol. 4 (U-syn 478.8 and note) but “grinned” is present in both the Little Review (vi.2, p. 37) and on the first setting of placard 24 (JJA 18:234). Therefore, the alteration must have been made on at least two of the typescript exemplars. The entire marked-up typescript, with its two missing pages shadowily restored, differs significantly from the Rosenbach Manuscript in two places: at points on fols. 44–45 of the manuscript. Neither U 10.1196–97 “From its sluice in Wood quay wall under Tom Devan’s office Poddle river hung out in fealty a tongue of liquid sewage” nor U 10.1220–23 “A charming soubrette, great Marie Kendall” &c. is present in the Rosenbach Manuscript yet both are found in the body of the typescript (TS V.B.8a fols. 23–24, JJA 13:29–30). Again, no autograph copy attests to these additions. However, caret marks have been inserted at the appropriate points in the manuscript, presumably by the typist (Facsimile MS “Wandering Rocks”, fol. 43–44). One can envisage Joyce sending a postcard or two to his typist, like those he sent to Claud Sykes for the Telemachiad episodes, giving the text of the addition and its point of entry in the manuscript.

My treatment of the “Wandering Rocks” avant-texte only highlights the paucity of material that is known to be extant for the episode. While we have a few pages of late notes, the dearth of evidence of early planning can be attributed to the short period of time which elapsed between Joyce’s first conception of “Wandering Rocks”, sometime after May 1918 (LI: 113), and the drafting of the episode at the beginning of 1919 (the Rosenbach Manuscript of “Scylla and Charybdis” was finished on “New Year’s Eve | 1918”, U-syn 468.23n). As we have seen from the use of the Subject Notebook and the Alphabetical Notebook, this period of six or seven months was buttressed by material collected at earlier times. Joyce compiled notes on Dublin places and people as a matter of course, not with a “Wandering Rocks” episode in mind, but once the episode was conceived, the notes were parsimoniously put to that end.

Arthur Walton Litz, in his treatment of the Martello Tower fragment at the British Library (JJA 9:1219–1222), describes the “fascinating career of revisions” undergone by the passage describing the Dedalus kitchen.³⁰ Prose in the British Library fragment is whittled down to laconics in the notesheets: BL “Cyclops” 1:18–20 reads “bDilly’s kitchen: oatmeal water, cat devours charred fishheads and eggshells heaped on square of brown paper, shell cocoa in kettle, sootcoated” which is pared down further on BL “Eumaues” 5:55–57: “sDilly’s kitchen. Oatmeal water cat devours charred fishheads & eggshells heaped on brown paper. Shell cocoa in sootcoated kettle.” From the latter the passage was worked into “Eumaeus” (U 16.269–78) but the Dedalus kitchen also features in the fourth section of “Wandering Rocks”. While it does not appear to have any textual origin in “Dilly’s kitchen” the chequered career of the latter sets a precedent for the process by which material for “Wandering Rocks” was likely gathered. A “mosaic of movements” (U 15.4100) from stray pages and notebooks into “Wandering Rocks”-rubriced note repositories created the storehouse of elements necessary to begin drafting. Whether shorthands like “Fr Conmee & ticket inspector” or “Dilly’s kitchen” were derived from scenes written out or stood as mnemonic triggers for scenes composed mentally, the episode became such a catch-all that disparate material could be shoe-horned into its nineteen sections.

The absence of pre-Rosenbach drafts can be subsumed under the general question of why so few drafts survive for the early adventures. This lacuna in the prepublication Ulysses dossier has never been satisfactorily accounted for and perhaps additional manuscripts remain to be discovered; until the discovery of the Joyce 2002 papers, the breach stretched all the way from “Calypso” to “Wandering Rocks”.³¹ But for the tenth episode the matter is complicated by Frank Budgen’s involvement. If we assume that Joyce did not start drafting “Wandering Rocks” until the beginning of 1919, he cannot have spent long at the episode before his eye trouble obliged him to seek Budgen’s assistance. The Rosenbach Manuscript is not a first draft but what antecedent can be posited for it? Such a draft was likely to have been written in a copybook and, like the early drafts of some of the other episodes, must have been fragmentary in form (that is to say, not covering the entire episode as well as comprising discrete blocks of text). Unlike the early drafts of the other episodes however, where the fragmentary form is only a compositional peculiarity and one to be removed in subsequent drafts, the fragmentary form abides in the final text of “Wandering Rocks” (I am thinking here of the “Proteus” NLI 36,639/7A, “Cyclops” Buffalo V.A.8 and “Circe” Buffalo V.A.19 drafts).

But perhaps a lost copybook does not even need to be posited. Each of the nineteen sections of “Wandering Rocks” is short (Father Conmee’s is about two hundred lines in the Gabler edition) and in composing them Joyce could have strung together sets of laconics on a par with “Dilly’s kitchen”. The genetic traces of these associational shorthands may be discernible in the first stratum of interpolations. As such, the first stratum would not be ‘interpolations’ in the strict sense of the word but rather a trellis or array of nodes about which the rest of the episode was subsequently built. On a larger scale the fragmentary nature of the episode is a consequence of this. Whether or not Joyce ever intended to impose a unity on “Wandering Rocks” comparable to that of the other episodes is unknown, but the ease with which he could dictate the “initial style” (LI: 129) to Budgen in early 1919 paved the way for its disruption and eventual abandonment.

Notes