 | |  |
Notes
James Joyce, A Portrait of the Artist as a Young Man. The definitive text, corrected from the Dublin holograph by Chester G. Anderson and edited by Richard Ellmann (1964).
Chester G. Anderson, "The Text of James Joyce's A Portrait of the Artist as a Young Man." Neuphilologische Mitteilungen 65 (1964), 160-200. (Quoted as Anderson.) This is a but slightly revised version of the Introduction to Chester G. Anderson, "A Portrait . . . Critically Edited . . .," unpubl. Ph.D. dissertation, Faculty of Philosophy, Columbia University, 1962.
It is described in much detail, which often corrects assumptions by Anderson (pp. 190 ff.), in chapters 5 and 6 of Jane Lidderdale and Mary Nicholson, Dear Miss Weaver (1970). (Quoted as Lidderdale.)
Contrary to the nomenclature in John J. Slocum and Herbert Cahoon, A Bibliography of James Joyce, 1882-1941 (1953), it is bibliographically correct only to speak of one first edition, the first impression of which was published in two separate simultaneous issues variant merely in the two distinct states of the title-page, and issued in two different bindings. The variant title-pages are both conjugate in their sheets and were probably printed by stop-press alteration from separate plates for which the identical typesetting of author and title was used. There is distinct type-damage to four separate letters (the 'e' in 'the', the 'a' in 'a', the 'M' in 'Man' and the 'C' in 'JOYCE') which positively secures the identification. Owing to the absence of B. W. Huebsch's publisher's device from the title-page of the London issue, its typographical lay-out differs in the wider spacing between the two lines each for author and title: the lines 'BY | JAMES JOYCE' have as a block been moved further down the page. Beyond that, the variance of the title-pages is merely in the alternative imprints. This first edition was never corrected but for a few minor alterations in its plates and ran out in 1950 in its 44th impression, while a second American edition, editorially corrected by Harry Levin, began its run in 1947 and went into many impressions and several separate issues, American and English. The edition named by Slocum-Cahoon 'The First English Edition, English sheets' (1918) is in truth the second edition of the novel, and it is the 'first English edition' only in so far as it is the fountainhead of the authorially corrected English line of the text. Of the fifth impression of the original American first edition there was in 1921 once more a separate issue for The Egoist Ltd. in London. This is not the 'third edition' (nor, of course, the fifth). Bibliographically, the reset Jonathan Cape publication of 1924 is the true third edition. As it is reset from the London edition of 1918, it might under the special circumstances governing the textual transmission of this novel be termed the second English edition. It had numerous impressions until it was replaced in 1956 by the reset Jonathan Cape illustrated edition (the third in England).
Lidderdale, pp. 92, 99, and 103 gives vivid accounts of when and how Miss Weaver was forced to give in to the demands for excision; with respect to the sentences omitted near the end of Chapter IV, Harriet Weaver herself wrote in the margin of Joyce's letter to her of July 24, 1915: ". . . the managers of the firm objected to certain expressions. . . . That was why the Egoist changed printers." James Joyce, Letters, II, 355, fn.
The latter correction refers to Chapter V. These were perhaps the two errata of which Harriet Weaver enclosed a slip in her letter to B. W. Huebsch of July 24, 1916.
Partridge & Cooper Ltd., whose name appears in the Egoist colophon, were a subsidiary of James Truscott and Son. These were the managers whom Harriet Weaver had to contend with. Cf. above, note 7, and Lidderdale, p. 91.
With reference keyed to the Viking [Anderson] text, the passages in question are: p. 7.13-14; 43.11-18; 44.32-35; 137.24-30; 138.6-9; 151.5-9; 192.8-11; 200.3-5; 205. 22-28; 206.30-32; 211.29-31; 212.5-7; 242.27-30.
P. 98.35-99.10 and p. 115.31-116.7. The same (?) indelible pencil has bracketed a part-column on p. 71, Feb. 16, 1914 (i.e. Viking [Anderson] p. 22.6-37), but there is no link in contents between this passage and the other two.
The full sheet either was cut from the beginning, serving as a divider between the chapters and a protective end cover, or else was used by Joyce as a folder for Chapter II, in which case the British Museum binder cut it apart and inadvertently turned fol. 27 upside down.
To Professor Giorgio Melchiori of Rome, who was most conveniently at hand in the British Museum reading room when I made this discovery, I am grateful for confirming my guess as to the nature of the document and for supplying the further information here given.
Letters, I, 80; 'the first half' can refer only to Chapters I and II, as no more had yet been published in July 1914.
Late in March 1915 (c. 29 March) Ezra Pound wrote: "Dear James Joyce: I took the final chapter of your novel to Grant Richards this a.m." (Pound/Joyce Letters, ed. F. Read, London 1967, p. 33). In a letter of April 22nd to Joyce, Harriet Weaver specifies that "Mr Pound sent Mr. Grant Richards . . . the part of the M.S. of your novel which has not yet been set up, together with a complete set of the numbers of 'The Egoist' in which it has appeared up to date. I asked for the M.S. to be returned by May 20th. This would give Mr. Richards two months in which to consider it." For access to those of Harriet Weaver's letters to James Joyce which concern the publishing of A Portrait, in photostats of the holograph originals, I am grateful to Miss Jane Lidderdale. While quotations from them here and below are according my own transcription, reference should be made to the edition of John Firth, "Harriet Weaver's Letters to James Joyce 1915-1920", SB 20 (1967), 151-188.
A renewed scrutiny of the inserted galley slip reveals an ink stroke in the margin about half-way down the column which does not stand against a correction to be made. It looks like the line-count strokes of Chapters I-IV, but divides off line 2254 of Chapter V as printed. However, if it may be assumed that the cumulation of 53 lines of italicized verse in the preceding sections of the chapter was disregarded in the count, the marking would be seen to stand against line 2201 of the regular text, reflecting a next to faultless line-count in hundreds. The observation would help to argue for the correctness of our assumptions about the fates of EC-W.
Harriet Weaver had submitted the novel to printers before, while Pinker was still searching for a publisher (see Lidderdale, chapter 6 passim); refusals from printers were coming in ever faster, so she may for some time have been circulating two copies of the text.
Miss Lidderdale's discussion of the dates and events leading up to Huebsch's publication of A Portrait (Dear Miss Weaver, chapter 6 passim) differs radically from Anderson's hypotheses (Anderson, pp. 190ff.). Being much more fully based on documentary evidence, her account serves as my frame of reference.
Anderson, p. 188 f. is aware of the facts. Curiously, he does not recognise the bearing they have on determining the provenance of EC-A.
The reference is here repeatedly to the unpublished Weaver-Huebsch correspondence. I gratefully acknowledge being given permission to use it.—It seems safe to say that Huebsch indeed never received the Marshall copy. For had it passed into his hands, he would, even though not printing from it, have handed it over to John Quinn to whom Joyce in 1917 sold all material relating to the first book edition which Huebsch held, and it would now be found in the Slocum Collection at Yale.
See the account of the events in London on September 6th and 7th, 1916, in Lidderdale, p. 125; and compare with the letter from Harriet Weaver to Huebsch of September 7th as quoted by Anderson, p. 190. This letter is now found attached to a complete set of tearsheets known as EC-B. Nowhere is there any documentary evidence, however, that 'EC-B' was an integral set of Egoist tearsheets from the outset. Rather, from the facts as they now begin to emerge I would infer that the one and only copy which ever became an identifiable unit was the printer's copy. Not even this copy, however, secured integrity until the printers stamped its sheets with serial numbers. It had none before and did probably not then, for example, contain the holograph insert (leaf no. 35). The other sets (EC-B and EC-C) had no natural integrity as physical objects until they became identifiable as catalogued units in the Slocum Collection. I suggest that, at various times, portions of the text (chapters and inserts, sections annotated and not annotated) were shuffled and reshuffled between them, the last time probably by Mr. Slocum himself. For the Quinn sale catalogue still speaks of three sets, each 'containing manuscript corrections by the author and Miss Weaver' (quoted by Anderson, p. 187). But EC-C now has no corrections.
Incidentally, she also gave her own complete run of The Egoist which the British Museum library did not possess before. Cf. Lidderdale, p. 425.
364 is the number of corrections counted by Cahoon in Joyce's manuscript list (Y). I count entries (364) and corrections to be made (365; two separate instances of 'public-house' > 'publichouse' are given one entry). These corrections sometimes involve more than one change. Anderson counts 373 changes (cf. footnote, p. 162). Attempting to apply his criteria, I count at least 379 changes. Yet I believe we are all describing the same body of corrections.
Anderson has no real ground for assuming (footnote, p. 197) that YW was written before Y: Y was completed by April 10th, YW was compiled between April 18th and May 2nd, arriving in New York on May 15th (see below).
The date they were mailed to Huebsch, which is authentic, is suggestive: on August 16, the Brighton printers who at first were going to print the English edition retracted their offer. Thus, Harriet Weaver—despairing momentarily that the book would ever be printed in England—may have wished that all corrections were in the hands of the publisher who alone thus far had the text of A Portrait in print. In the event, of course, only the second Egoist Press edition and its descendants ever incorporated Joyce's 'nearly 400' corrections.
As far as the events go which lead up to finding a printer in England for the second edition, Miss Lidderdale (Dear Miss Weaver, pp. 139 ff.) has drawn upon the Weaver-Huebsch correspondence and largely recorded the relevant details.
Pinker must have kept the corrections on file in yet another typescript copy; for the manuscript original (Y) was sold to John Quinn sometime in June, 1917; by July 10th, Joyce had received Quinn's acknowledgement (Letters, I, 104).
HB is apparently not identical with the copy marked up and given before August 16th, 1917, to the Pike's Fine Art Press in Brighton to print from. They returned a book with 'passages marked in blue pencil' to be 'modified or removed'. (Lidderdale, p. 142). There are no traces of bluepencil markings in the Bodley volume. The discrepancy in the number of corrections between the handwritten list sent to Huebsch on August 16 (15 corrections plus removal of one typing error) and the additional entries in pencil in YTW (17 corrections plus removal of two obvious typing errors) may have its explanation here. The fifteen corrections in the handwritten list may have been the result of annotating the copy for Pike's, the two additional ones may have been added to YTW in preparation of the printer's copy for Johnson's of Southport. As Harriet Weaver then spotted another six misprints and hyphenation errors in the course of annotating HB which were never entered in YTW, it must remain an open question —until all relevant documents can be reexamined in preparation of a critical edition—whether it is merely a happy coincidence that YW and YTW concur in the number of 17 corrections in excess of Joyce's authentic 365.
To complete the record, a set of pencilled notations on the back flyleaf (verso) should be observed: '26-41 234-246 280-292 for Sesame book 1942'. If taken as page references, '26-41' comprises the greater part of the Christmas dinner scene in Chapter I; '234-246' the conversation between Stephen, Davin and Lynch until just before the esthetic theory section in Chapter V; and '280-292' the final conversation with Cranly. I have not investigated the relevance of these jottings.
The corresponding section of the typescript which served as printer's copy for The Egoist happens to survive. Curiously, the typist first spelled 'revery' according to the manuscript, but the final 'y' was altered in ink to 'ie' by an undeterminable agent. Joyce himself did sporadically enter corrections in ink in the typescript, but the 'ie' does not appear to be in his hand. The spelling 'reverie' occurs several times in The Egoist. It was successfully eradicated by Joyce himself in all instances but the present one. It is highly probable that the failure to observe his Y instruction at B: 87.9 was spotted and amended by him when he proofread J.
The texts in the editions of Harry Levin (in The Portable James Joyce and elsewhere; cf. Anderson, p. 167) and of Anderson/Ellmann (1964) are both conflated texts. The latter in particular, which draws on the manuscript, albeit in a not readily controllable manner which on analysis proves to be unsystematic, provides—in the true technical sense of the word—a contaminated text.
It should be noted here that no scholar with bibliographic and paleographic expertise has yet investigated the Dublin holograph. Until it has been fully described and analysed, neither the above dates can be given with full assurance, nor is it possible to say whether or not our present conception of how the text of the novel evolved will need to be modified.
The figures for the Egoist variants here and below derive from double collation (D-E, D-J) which, though done with all possible care, has not been counter-checked. They should therefore be taken as approximations to indicate relations.
To assert that in E "printinghouse stylesheets triumphed almost completely over the copy in punctuation, hyphening, capitalization, and other accidentals" (Anderson, p. 185) is much too sweeping a statement. As regards hyphenation and capitalization it is not true.
I count at least 106 instances of such omission, equalling almost exactly one-third of the total of D-E substantive variance: 19 instances in Chapter I, 28 in II, 22 in III, 9 in IV and 28 in V.
Page/line references are to the 1964 Viking printed text as used in its 1968 reprint in: James Joyce, A Portrait of the Artist as a Young Man. Text, Criticism, Notes. Ed. Chester G. Anderson, 1968 (The Viking Critical Library). The quotations, however, give the manuscript readings unless otherwise indicated.
This is a fascinating variant. Stephen also notes "the proud conscious movement of her perfumed head" which accompanies the undoing of the gown. Moreover, the next paragraph in the text makes it clear that Stephen and the young woman stand apart in her room: ". . . she came over to him and embraced him. . . ." "Watching her as he undid her gown", therefore, which very clearly is the manuscript reading, appears to be a genuine Freudian slip of the author. It need of course not have been corrected by him. A typist or compositor would have been capable of spotting the inconsistency.
This revision, to avoid a quibble on Dixon's signet ring, is comparable to the E→H authorial change of 'trunk' > 'body' at 201.14 to avoid a pun on the preceding "whinny of an elephant".
These are provisional rules, as the facts and inferences concerning the subsequent rounds of authorial correction by which they must be augmented and modified have not yet been discussed. Yet they are also the central rules for establishing a critical text, as the results of a comprehensive analysis of the body of D-E variance must form the basis for any editorial hypothesis and procedure. It may be appropriate therefore in their light to indicate statistically whether "the definitive text, corrected from the Dublin holograph . . . published in 1964 by The Viking Press, Inc." has a claim to being definitive. Of the total of substantive variants, that is 317 by our count, 29 (by our count) are corrections in E of manuscript error. This leaves 288 instances on which editorial decision must operate. Giving the editors the benefit of the doubt in the case of the possible six authorial cuts in Chapter III, we find that in 158 out of the 288 instances editorial decision follows the rules here postulated, while in 130 instances it goes against them. The ratio of (what we would regard as) correct to incorrect is thus 55%:45%. Taking into account that several 'correct' decisions were in fact anticipated by the author in the course of his repeated subsequent corrections of the text, this is tantamount to a flat 50:50 ratio of hit and miss. Corresponding figures for the treatment of accidentals—which by reason of Joyce's fairly systematic later restyling of hyphenations and capitalizations could in any case not be based on the D-E variation—have not been worked out. Nor has the treatment in the Viking text of substantive variance in the later editions (H, B, J) been systematically analysed. The impression that it, too, is somewhat haphazard—and in particular so with regard to the J variants on which hypothetical inference must again operate—stems from cursory observation only. It would seem that these facts and conditions are due to the lack of a comprehensive and logically consistent hypothesis of the transmission of the text of A Portrait from the Dublin holograph (1913/14) to the third book edition (Jonathan Cape, 1924). This lack prevents the 1964 Viking edition from fulfilling the standards of a definitive text.
In particular, the typewriter used seems to have had no key for underlining. The underlinings in T which copy underlinings in D to indicate italics in print were done by hand and are missing in all those places in T where E fails to italicize.
This discussion is based on xerox copies of the extant pages which were kindly put at my disposal by Mrs Elaine Bate who now owns the originals, and to whom I am grateful for permission to use the copies for the purposes of this article. Since the article went to press, I have had the opportunity of inspecting the fragments themselves. They are the ribbon-copy typescript originals, on white typewriting paper 22.5 x 28.4 cm., uniformly watermarked CROXLEY MANIFEST BANK | LONDON under a rampant lion, facing left, which holds in its front paws a standard, unfolding over its head and bearing the inscription LION BRAND. The ribbon ink, originally of a blackish colour, has been affected by damp and has largely turned purple. Each of the chapters separately was once stapled together very close to the left paper edge. Every sheet shows holes in the upper right-hand corner which appear to be the marks of the compositor's copy-holder. The rust-marks of paper clips indicate that the fragmentation of the typescript is of an early date, in all probability resulting from the manuscript's handling in the printinghouse. The authorial corrections can without much difficulty be isolated from the several stages of annotation in evidence. Page II.17, which is the first leaf of the last of the extant fragments, bears the inscription in watery-blue ink "Portrait of The Artist as a young Man | by | James Joyce." The handwriting is Ezra Pound's. The Dublin holograph, which I have also meanwhile been able to inspect, shows a series of pencil marks throughout Chapter I which in ten instances divide off page-beginnings of the typescript, and in two instances specify an (authorial?) re-paragraphing of the text, carried out in the typescript.
If it could be assumed that the conditions under which the typescript came about remained fairly constant throughout the five chapters, one would expect to find those corrections of incomplete manuscript readings in Chapter V which appear to be clearly authorial similarly to have been written into the typescript. This would agree with the evidence from authorial correction/revision and definitely indicate one round of authorial attention to Chapter V only.
Proof, that is, as distinct from 'first-degree' inference from facts such as the inscription of the words on the manuscript pages which were observed earlier to render plausible the explanation that the omissions were typist's errors; or 'second-degree' inference to the same effect, taking the form of conclusions drawn by analogy from an analysis of a different group of variants which were inferentially also declared to be non-authorial.
As it happens, 'there' seems on literary grounds to be a definite improvement over 'in the square', referring as it does to both outdoor and indoor locations in that square. The evidence of T creates a paradoxical situation, even an editor's dilemma. 'There', it is true, is confirmed to be authorial, as was hypothetically assumed before the T fragments came to light. At the same time, it is revealed not as a revision undertaken in view of the original reading, but as a spontaneous correction of a typist's error. As an instance of alternative phrasing it was never intended to replace, but rather to restore the reading which had accidentally got lost. What on first sight appears to be an authorial second thought is really an attempt to recover the wording of a first thought. Editorially this means that the original reading 'in the square' should be given preference over the authorial correction 'there', contrary to the general rule by which authorial corrections should replace original readings in a critical text, and despite the subjective judgment that 'there' be preferable to 'in the square' in its context. The editorial decision to adhere to textual logic over literary judgement is inevitable because the situation is not unique in the course of the transmission of A Portrait: when reading the E text for H, and H for B, Joyce belatedly spotted several further typescript errors of omission which he corrected without recourse to his manuscript. The results of his correction always differ from the original readings, though never except in the present case for the better.
These seven instances are: father > uncle > father (26.11); he had got > got > he had got (28.8); or > [om.] > or (32.5); Let it > let us > let it (32.8); opinions > opinion > opinions (34.14); MacManus > MacManns > MacManus (38.27); about it > about > about it (53.1). The variant chestnuts > chestnuts > limes occurs at 24.10.
If it were assumed that as much text as made up the first three Egoist installments was set up at once by the printers on receiving Chapter I in manuscript, then it would be possible to imagine that proofs of installments two and three were passed to and fro between England and Trieste in the last days of January/early days of February, 1914, while the first Egoist installment of February 2nd was already appearing. The change chestnuts > limes is involved, which is identically repeated in Chapters II and III. To ascertain when, where, and by what agent it was introduced into the final Egoist text in its first instance would carry implications for the timing of the completion of both the authorial faircopy and of the typescript for all of Chapter II, and the beginning, at least, of Chapter III.
It is well to emphasize, however, that it appears to have been James Joyce's own concern for a pleasing typographical appearance of his text in print which motivated his changes of such accidentals as capitalizations and word divisions. With this in mind, it may be recalled that the question was raised and left open above (p. 27) whether consistency in the treatment of accidentals should in fact be extended to the adoption of the manuscript system of dashes in direct speech. On pragmatic grounds, I would defer a decision in this matter until, by way of a practical experiment, a few sample pages were set up in type so as to show whether both printers and readers could within reason be asked to cope in full with the Joycean unorthodoxy of punctuation in dialogue. Robert Scholes in "Some Observations on the Text of Dubliners: 'The Dead'" (SB 15 [1962], pp. 200 f.), discusses the matter of Joyce's dashes in Dubliners and the later works. He has reduced Joyce's usage to initial dashes only in his Viking edition of Dubliners (1969).
| | |