The Texts of Pilgrimage

The Collected Edition of Pilgrimage was published in 1938, reprinted in 1967 with March Moonlight added, its 12 and then 13 books gathered into 4 volumes, its text corrected and proof-read by the author herself. Was that text reliable? How did it compare with the texts of the 10 individual books issued by Duckworth between 1915 and 1931 and the 1 book issued by Dent in 1935? And further, how did the English editions of the first 6 novels, from Pointed Roofs to Deadlock, compare with the reset texts published by Knopf in New York between 1919 and 1921? And which first editions were used as copy text for the Collected Edition? These questions, rarely mentioned by Richardson scholars, lingered in my mind for years as I explored other aspects of the author’s life and work. Then before A Reader’s Guide was finished and while Notes on ‘Pilgrimage’ was at a fairly early stage, curiosity got the better of me.

On 28 April 1995 my wife, Dorothy F. Thomson, and I began comparing the text of the 1967 edition of Pilgrimage (identical to the 1938 edition, apart from a few minor anomalies) with the English First Edition of each of the 11 individually published books of the series. To this we added a comparison of chapters 1 to 3 of March Moonlight with their periodical version as "Work in Progress," Life and Letters, 1946. We finished that enterprise on 25 August 1996. Next, we compared the English First Editions with the independently set American Editions of the 6 novels, Pointed Roofs, Backwater, Honeycomb, The Tunnel, Interim, and Deadlock. And finally, we tackled a comparison of the Little Review version of Interim with the English First Edition. This last endeavor, guaranteed to put any marriage under severe strain, so radical were the differences between the two texts, we finished around mid 1997, by which time I had ceased to keep records of precise dates.

I want to be quite specific about the method my wife and I used. Each evening, Dorothy read aloud from the First Edition, noting formatting and punctuation as she went along. I followed, reading silently in the Collected Edition. Each time I found a difference, the reading stopped while I recorded the change. The next morning Dorothy typed up the changes; then I proofed these against the two texts. From here we went on to the first editions, Dorothy reading from the American Edition while I followed in the English First Edition. Finally, Dorothy read aloud the English First Edition of Interim; I followed stumblingly in the Little Review text. Then twice, over the next two years, the page and line number of each variant was checked for accuracy. During this process of reviewing, a number of omissions turned up. But it will be obvious that, given the methods used to compare the texts, other omissions must be inevitable and errors unavoidable.

By the 1930s Richardson believed readers of Pilgrimage were losing contact with her and she with them on account of the years intervening between one book and another. With Duckworth she was trapped. He had the rights to her earlier volumes, selling a steady trickle of them each year, but because they had been set in different types he would not reissue them in a combined format. Since Knopf had dropped her in America, however, she hoped there might be a chance with a new publisher there. On 4 October 1931 she wrote to the American agent, I. R Brussel: "The only sound way of dealing with these short chapter-volumes is to issue them in sets of two or three together & indeed I hope some wise man may be found who will adopt this plan." But no wise man came forth, though she tried several possibilities. Thus she confesses, in December 1933, after rehearsing the situation to S. S. Koteliansky: "A real edition by a real publisher would greatly comfort me & might make it possible for me to finish the book." See Windows on Modernism, Selected Letters of Dorothy Richardson (Athens: University of Georgia Press, 1995), 255. Koteliansky, who was with the Cresset Press, exerted himself. Finally, on May 21st 1935--after eighteen months of intermingled hope, despair, and anxiety--she could report to Bryher: "I have now signed Dents contract, swearing to be ready by July 1st [with Book xi, Clear Horizon], & hope I may" (Windows, 293). Richardson understood this contract to mean that after Clear Horizon had been published in the fall of 1935, Dent would then go ahead and issue the whole eleven parts of Pilgrimage in groups of two or three books, with a short time interval between each of the cumulated volumes. And with the income from Dent, she would be able to go on with Volume V of Pilgrimage, beginning with Dimple Hill.

Five months later, with Clear Horizon published, this was still what she expected when she wrote to Bryher in October about the winter of 1935-36: "For which I have no plans, beyond recklessly & ruthlessly beginning xii & allowing myself fifteen months in which to complete it together with the work to be done in relation to the new edition & upon the chaos created in the old by a concert of carelessness" (Windows, 298). When November came and the sales figures for Clear Horizon proved disappointing, she imagined–so she told J. C. Powys--that Dent "will now feel disposed to postpone, indefinitely, the new edition" (Windows, 303). What followed was more complicated than that. In March 1936 Richard Church, on behalf of Dent, told her that if her reputation was to be established and made secure, she and Dent must give the public a "finished" Pilgrimage. He stressed "how important for us all will be the fact that the great book has been drawn to a conclusion" (Windows, 306). That she was in no position to draw her great book to a conclusion she well knew. Though she had made a beginning with Dimple Hill, she viewed it as the first book of a new Volume V. Even if she finished Dimple Hill it could in no way stand as the conclusion to Pilgrimage. For the next two years and more, she agonized, temporized, protested and pleaded on behalf of her incomplete creation. The drama of its coming into being, its incompleteness obscured, is described by Gloria G. Fromm in Dorothy Richardson, A Biography (Athens: University of Georgia Press, 1994), pp. 297-312, and documented in the letters to and from Richard Church, and in the letters from Richardson to Koteliansky and Bryher (Windows, 306-47).

And at the moment of its coming forth, October 1938 in England, November 1938 in the United States, the eyes of the English-speaking world were turned away, fixed on Adolph Hitler and the shadow of another Great War.

No sooner had Richardson told Bryher of her signing the contract with Dent, than she added: "After that I must read every syllable & comma of my ten vols." (Windows, 293). In the same vein she wrote to J. C. Powys on 19 November 1935: "Meanwhile I hopefully attack the chaos, of those volumes, resultant on Duckworths efforts to standardise my punctuation & my counter-efforts to recondition it" (Windows, 303). Three months later, in a letter to Koteliansky, the theme is continued: "I will not enquire for C.H. But I am moved to record the deepening, in the course of reading & correcting all the little volumes, of my sense of the relative futility of publishing them separately. They support & amplify & clarify each other."

Since no further comment is made during 1936 or the first half of 1937 about correcting the little volumes, it would seem probable that she had completed her corrections of the 10 earlier books during 1936. And it would seem, too, that once there was a settled prospect of a new edition, she tackled revisions to the 10 first edition texts promptly and without complaint. After all, for years she had lamented the misprints and errors in her texts. The revisions extended as well to the new books. As she tells Owen Wadsworth on 27 January 1938: "I labour. Having carefully corrected the whole round dozen . . . I now receive almost daily a bunch of proofs for more reading." The "round dozen" now includes not just Clear Horizon, published in 1935, but Dimple Hill, the twelfth book, presumably corrected in typescript. For the final plan, devised by Dent and Richard Church, was to add the previously unpublished Dimple Hill to Volume IV of Pilgrimage as an added incentive for readers to purchase the Collected Edition.

Yet the process of revision for this new edition was neither clear nor simple. For Richardson it rarely was. On 28 August 1937, in an unpublished letter to Bryher, she observed in referring to Dent: "Not on any account can I entrust them with corrections. The whole is peppered with errors, many of them typists mistakes which make sense, of a sort." This suggests that her initial reading of the First Edition texts had not been as thorough as it should have been, for which reason she hoped to spot further errors from the past as well as new misprints in her scrutiny of the Collected Edition proofs. The actuality, however, is exposed in her unpublished letter of 27 January 1938 to Wadsworth. While proofs snow down upon her, she strives "to get on with the first chapter of Vol 5 [i.e. March Moonlight]." Richardson here lays bare her priorities. She gives to a prospective Volume V time that could have been devoted to proofreading. The result is predictable. In a letter of 21 December 1942 she says of Volume IV: "When Dent’s edition was going through, I couldn’t manage much proof-reading, & some of the p[rinters]. e[rror]’s I found, even on dipping, made me squirm" (Windows, 454-55). Her good intentions, it would seem, were undermined from all sides, by the press of daily preoccupations, by boredom and impatience with proofs, and by the lure of going on with Volume V.

Richardson’s failure to detect errors in her texts is apparent at a more fundamental level. In "Appendix 1: Misprints & Errors & Oversights in the Collected Editions of Pilgrimage," Notes on ‘Pilgrimage’: Dorothy Richardson Annotated (Greensboro: ELT Press, 1999), I have recorded 25 instances in which the MS readings of Pointed Roofs, Dawn’s Left Hand, Dimple Hill and March Moonlight appear superior to the readings of the printed text. Many of these seemingly inferior readings of the first edition texts were the result of "typists errors," having their origin in misreadings of the MSS., misreadings Richardson failed to detect in revising for the Collected Edition. No doubt she also overlooked misprints and typist’s errors "which make sense, of a sort" in the other 9 books comprising Pilgrimage. But for these texts no MSS survive.

A Collected Edition of Pilgrimage: the idea of bringing together in one place, in proper sequence, and in four compact volumes the individual books chronicling the life of Miriam Henderson was an exciting prospect for Richardson. But the three years from March 1935, when she signed her contract with Dent, to March 1938 when proofs "ceased snowing" (Windows, 345), were for her a strenuous and exacting time. She was just recovering from a nervous breakdown when she was confronted by the demand in Dent’s contract for a MS of Clear Horizon by July 1. She met this deadline and later in 1935 read proofs for the new book, at the same time beginning her corrections to the First Edition texts. The process of revising the "little books" continued through much of 1936. Meanwhile in October 1935 she began the next book in the series, Dimple Hill, daunting in its demand for an entirely new setting. At Christmas of 1935 she burned the MS in despair. In the New Year she began again, going very slowly (Windows, 309). Finally, more than 18 months later, in August 1937, she submitted the MS of the novel to Dent, two months late (Dorothy Richardson, 306). In 1936 she commenced a struggle even more prolonged than that with Dimple Hill, the attempt to write an Introduction or Foreword to the new Pilgrimage. Though she told Koteliansky in August 1937 that she had sent in a Foreword and proofs had come back, she was clearly insecure about it (Windows, 336). As late as 27 January 1938 she told Wadsworth she was struggling "to write some sort of preface, the hardest job I ever attempted." She was still fussing over her four-page Foreword in April 1938 (Windows, 345). For my analysis of the reasons underlying her difficulties, see "Dorothy Richardson’s Foreword to Pilgrimage," Twentieth Century Literature 42 (Fall 1996), 344-59.

Though Richardson’s energies were consumed by these literary preoccupations, she was, throughout this period, burdened by the domestic and social responsibilities which typically plagued her later years. Under such circumstances, it will not surprise that Richardson, in carrying out her revisions of the first edition texts, did not introduce any major creative or stylistic changes. Her lack of time and resources precluded it. But even in better circumstances her revisions would not in all probability have been substantial. Her visionary procedure of entering into the past and recreating it as a lived present was an exacting discipline which did not lend itself to later tinkering. So apart from eradicating misprints, she concentrated on "correcting the chaos D’s reader made of my punctuation, which I failed to proof-read into order" (Windows, 280). Or as she put it to Bryher in an unpublished letter of 28 August 1937: "Meanwhile Duckworth’s reader, systematically, but humanly irregularly, regulated, in vol. after vol., my punctuation, & I re-regulated, also irregularly, his regulations. Result, my obstructive reputation for unreadable prose."

Richardson, in preparing for the Collected Edition, seems to have inserted revisions on a first edition copy of each book of Pilgrimage, after which Dent reset the text and returned proofs to her for further revision. This was complicated in the case of the 6 earlier novels by two possible variables. Was the English First Edition (E) used on which to insert revisions or was the American (A) edition used? With one exception, a comparison of the two editions offers little evidence that the reset American Editions of the first six novels, issued between 1919 and 1921 under Knopf’s imprint, have textual authority. The English Editions were used as copy text, except in the case of Interim. Details are set out in Chapter V. But if in 5 instances an English Edition was used, was the text a copy of the first issue or of a later issue? We know that some if not all of the earlier volumes of Pilgrimage were reissued by Duckworth. A copy of Pointed Roofs, for instance, announces "2nd edition, 1921" and a copy of Backwater offers the following information: "First Published July 1916 / Second Impression September 1916 / Third Impression June 1930". And in an unpublished letter to Bryher of 16 [April] 1934 Richardson says that Duckworth "has consented, in regard to a volume just going into its second edition [Dawn’s Left Hand ?], to remove the most glaring of its errors." It is therefore possible that occasionally Duckworth made one or more minor but significant revisions to a reissued novel and provided the author with a copy of the reissue. Richardson, then, might have inserted her corrections in such a copy. Whether she did or did not is impossible to say, because the task of comparing the texts of the various issues of individual novels remains to be done. Meanwhile, first issues of the individual novels have been cited in all textual comparisons.

That is true as well in Chapter V where the English First Editions are compared with the independently set American Edition of each of the first 6 novels of Pilgrimage; and also in Chapter VI where the English First Edition of Interim is compared with the Little Review version. In every case the English first issues are cited.

In searching for a style that would accommodate her stream of consciousness approach, Richardson was content at first to cultivate sentence fragments, a sprinkling of suspension points and, in her more formal sentences, a lightness of punctuation. Sentence fragments and participial constructions had the special advantage of being free from the rigidities of past, present and future. They could convey Miriam Henderson’s moment to moment experiences informally and flexibly.

With The Tunnel Miriam’s style of reflection becomes a little more complex and mature, which opens the way for the author to omit many traditional commas. But Richardson’s most daring experiment since her launching of stream of consciousness narrative in Pointed Roofs was yet to come. When Interim began to appear in June 1919 in the Little Review, the text followed Joyce, as well as many continental authors, in shedding quotation marks for dialogue. (Ulysses, continuing this practice, was already well under way in the same Little Review.) Here Richardson utilized the dash to mark speeches and sometimes ran a series of verbal exchanges together within a single paragraph. On occasion, at the beginning of a paragraph, she omitted the dash altogether. Then just over two thirds of the way through the narrative in Chapter VIII (Little Review 6, December 1919, 20-28), she reverted without warning to traditional quotation marks. The same sudden shift occurs in the English First Edition published in December 1919, the change taking place on page 198 of the 293-page novel. But the transition was not complete until page 218, in the midst of Miriam’s conversation with the Canadian doctors, paralleling the Little Review 6, January 1920, 37. One may suppose that in the latter part of 1919 Gerald Duckworth or his representative discovered what she was up to and nixed it. Or one may even suppose that Richardson herself concluded her new method was not a success. Whatever the case, it is remarkable that Duckworth allowed the first two thirds of the novel to appear under his imprint in its radical and inconsistent Little Review format.

In the Collected Edition this shift in convention would have occurred in volume II at page 401, with the full transition established at page 411; instead Richardson revised chapters I through VII, inserting standard quotation marks and paragraphing throughout, to bring them into conformity with chapters VIII through XI.

These revisions for the Collected Edition text are only the last stage of Interim’s transformation. In preparing the English First Edition in 1919, Richardson made radical and extensive alterations to the Little Review text, alterations far more pervasive than any she ever undertook in preparing the 11 First Edition texts for the Collected Edition. It is as though for the first and only time we are allowed to glimpse in the Little Review a Richardson MS in an earlier state of development. The many revisions made to this periodical text in preparing the English First Edition are set out in Chapter VI.

Interim, we may say, was the end of Richardson’s purely technical innovations. That she remained in an assertive frame of mind until 1921, the year Deadlock was issued, is clear from a letter to H. G. Wells accompanying a signed copy of the novel: ". . . the more I write, the less I punctuate. What I really like . . . is the absence of anything but full stops . . . But I did, over my pf. [proof] pepper in quite a lot of little marks, of wh. D’s double-damned reader made a cheap selection." (Quoted by Bertrand Rota in item 593 of a catalogue offering the Wells copies of Richardson’s first eight volumes of fiction. My Xerox copy does not permit me to date the catalogue, but presumably it was circa 1947.) Then just three years after her letter to Wells, Richardson penned an eloquent concession speech to traditional punctuation, which she published as "About Punctuation," The Adelphi (London) 1, April 1924, 990-96. The tone of her opening paragraph was elegiac:

Only to patient reading will come forth the charm concealed in ancient manuscripts. Deep interest there must be, or sheer necessity, to keep eye and brain at their task of scanning a text that moves along unbroken, save by an occasional full-stop. But the reader who persists finds presently that his task is growing easier. . . . It is at this point that he begins to be aware of the charm that has been sacrificed by the systematic separation of phrases. He finds himself listening. Reading through the ear as well as through the eye. . . . in the slow, attentive reading demanded by unpunctuated texts, the faculty of hearing has its chance, is enhanced until the text speaks itself. And it is of this enhancement that the strange lost charm is born. . . .

Only the rarest of modern prose can thus arouse and affect. Only now and again, to-day, is there any strict and vital relationship between the reader and what he reads. (990)

After stating her case for freedom and after celebrating its varied manifestations in Rabelais, Sterne and the early H. G. Wells, Richardson moves on to the elaborate periodic sentences of Henry James which demand--if they are to be followed at all--a sustained immersion by the collaborative reader (992-93). Such recreative collaboration, demanded also by the unpunctuated MSS of the past, is Richardson’s ideal. But alas, "it is dangerous in these days of hurried readings to ask for the re-scanning even of a single phrase" (996). Hence, "the importance of the comma cannot be exaggerated. . . . In prose everything turns upon it" (994).

During the time between 1921 when she sent H.G. Wells his copy of Deadlock and 1924 when she published this essay, some person or reviewer planted in the author’s mind the notion that lack of conventional punctuation, and more especially a paucity of commas, was contributing to her "obstructive reputation for unreadable prose." The thought is vigorously expressed by Alfred A. Knopf in an unpublished letter of 17 February 1920 to William Heinemann (quoted by Fromm, Dorothy Richardson, 119). Speaking of Mrs. Dawson-Scott, Knopf complained that her new style was "suffering from an extreme of Richardsonitis," for she has with "unflagging pertinacity kicked out every punctuation she came across and stuck in its place a trio or quartette of periods."

The most basic logic of stream of consciousness narrative is its freedom from restrictive rules of grammar and punctuation. Such freedom is relative however, dependent upon the character of the experiencing subject and the quality of the life encountered. One can believe that Richardson’s shift to a more traditional style after Interim was well-timed. Her new punctiliousness in the 1920s and 1930s paralleled her prose narrative as it became the stream of a more and more superbly articulate and organized and elaborative mind. Here the flowering of commas could help the reader to follow the ins and outs of Miriam’s increasingly lengthy periodic sentences.

Each of the following chapters sets out in table format the textual differences between the various editions of the Pilgrimage text. Chapter I compares Volume I of the Collected Edition with the English First Editions of Pointed Roofs, Backwater, and Honeycomb. Chapter II compares Volume II of CE with the English Editions of The Tunnel and Interim. Chapter III compares Volume III of CE with the English Editions of Deadlock, Revolving Lights and The Trap. Chapter IV compares Volume IV of CE with the English Editions of Oberland, Dawn’s Left Hand and Clear Horizon. No independent First Edition exists for Dimple Hill, which concluded Volume IV in 1938, nor for March Moonlight which was added to Volume IV in 1967; however Chapters 1 to 3 of March Moonlight in CE are compared with the parallel text in Life and Letters, 1946, at the conclusion of my Chapter IV. Chapter V compares the English and American First Editions of the first 6 novels of Pilgrimage; and Chapter VI compares the English First Edition of Interim with the Little Review text.

My text, it will be observed, is arranged back to front in respect to chronology. The Little Review publication of Interim, from June 1919 to May-June 1920 (Chapter VI), began running before the publication of the English and American First Editions of Interim in December 1919 and June 1920 (Chapter V), which in turn are earlier than the Collected Edition Interim, 1938 (Chapter II). Moreover, Chapters I to IV are arranged with the chronologically later text, the Collected Edition, cited on the left and the earlier text, the First English Edition, on the right. Likewise, Ch. VI places the later text, the First English Edition of Interim, on the left and the earlier Little Review text on the right. The purpose of this organization is not to stand chronology on its head, for the order of the texts is obvious enough. The purpose is to place at the left margin for immediate reference the page citations to the edition of Pilgrimage the reader is most likely to possess.

For Pilgrimage I have cited the Collected Edition, 4 volumes, London: J. M. Dent, 1967. Except for the addition of March Moonlight, this 1967 edition is a reissue of the 1938 edition by Dent. The New York: Alfred A. Knopf editions of 1938 and 1967 are from sheets supplied by J. M. Dent. In essentials, then, and with the exception noted, all Collected Editions of Pilgrimage are textually identical, including later ones by Popular Library and Virago.

Each table, setting out the differences between the two texts being compared, is introduced by a "Summary of Variants," with illustrations of some of the more important differences. These summaries employ the terms defined below.

Total Variants In theory every textual change constitutes a variant. In practice certain common sense rules have been applied. Thus when Richardson inserts or deletes a group of words at one point in the text, this insertion or deletion counts as 1 variant even though 10 words may be in question. A multiple hyphenation of words counts as 1 alteration. Also when a cluster of dots followed by a lower case letter is changed to a period followed by a capital letter, the combined changes--since a period implies a following capital--normally count as 1 variant. But when a single word is capitalized, italicized, and fitted out with a missing accent, that entry counts as 3 variants.

Substantive and Non-Substantive Variants In an age when authors proofread their texts before they go to press, the old distinction between substantive and accidental variants breaks down. Even the most innocuous-looking comma may by its insertion convey some subtle intent on the part of its author. In these circumstances, the best a critic can do is to lay down certain rules to guide his judgment in assessing which variants in a text are more significant and which less significant. Even so, the standard used to define variants as substantive or non-substantive is bound to be in part arbitrary, especially so when judgment is used to differentiate between different instances of the same type of variant. In the present study, all verbal differences, however slight they may be, are treated as substantive; also all changes in the placement of quotation marks, of paragraphs, and of section breaks. (But see the Note to Interim.) On the other hand, words placed in italics are counted as substantive when the addition of italics changes the emphasis, but not when the addition simply confirms an emphasis implicit in the rhetorical construction. The same principle holds in evaluating changes to question marks and exclamation marks. Many differences in capitalization and in punctuation, including hyphenation, do not rank as substantive; nor do minor variants in spelling (like realise | realize and forgone | foregone), incidental foreign accents, or casual aberrations in choice of quotation marks for a passing phrase or for an indented piece of verse. Substantive variants are marked >.

Page and line numbers Each page and line number points to where that variant occurs, though text serving as a tag for the variant may sometimes fall on another line or even on a following page.

Misprints and Errors The record here is incomplete because those errors and misprints found in both the English First Edition and the Collected Edition do not register as variants. For a more complete account of misprints and the like in CE, see "Misprints & Errors & Oversights in the Collected Editions of Pilgrimage," Notes on ‘Pilgrimage’, Appendix I. Notice also that minor irregularities in the English First Editions, such as the absence of a circumflex, are treated in the present catalogue as simple variants rather than as errors.

The titles of books, plays, magazines, journals, art works, and musical works of all kinds are normally in italics in CE and in ". . ." in the English First Editions (E) and the American Editions (A). The exception is the English First Edition of Clear Horizon which, like CE, uses italics for these titles. For all the other books, however, italics in CE are the equivalent of double quotation marks in the first edition. These differences, being systematically applied, are not recorded as variants. But any deviations from this standard are recorded as variants. It will be noticed in particular that Richardson’s citations of art works and musical works are far from consistent. Paintings for example sometimes appear in CE with italicized titles and in E with capitals, but without either ". . ." or italics. Such technical irregularities have been recorded but have not been designated as misprints. As a result, significant misprints and errors stand out, each signaled by an *.