My editorial decisions were further complicated by not all the necessary evidence being available, although fortunately the gaps are mostly from material associated with the earlier books. In looking at the wealth of material that is available, I have consulted the manuscripts of the various parts of the poem held by Kent State University, the University of Virginia, the Houghton Library of Harvard University, SUNY Buffalo, and the Beinecke Library at Yale. I have also consulted a number of private archives, including the New Directions files in New York and Norfolk, and the papers of figures in some way connected to the poem, where I was able to track down the family or literary executor. Through these materials much of the history of the poem’s composition can be traced.
Kent State holds a version of the Book I typescript that went to the printer, and that is dated by Williams March 1, 1945. The library holds the galleys that were set up from this version, and upon which Williams made many excisions, cutting about a tenth of his text. The library also holds a set of page proofs set up from these edited galleys. These materials came through a private donor, and originally through the New York dealer House of Books. (Vivienne Koch notes in the preface to her 1950 study William Carlos Williams that she possesses “the emended galley proofs of Book I given to her by James Laughlin.”) In addition, Kent State holds a late typescript of Book III, dated May 18, 1949, by Williams, and inscribed by him to David Ignatow.
The University of Virginia received the material sold in 1964 by Kathleen Hoagland, Williams’ friend, novelist, and local historian, who typed some of the late drafts of Books I-IV. The library holds the carbons of these drafts, with valuable annotations added by Ms. Hoagland, probably at the time of the sale. The Hoagland material also includes some of the poem’s source materials—letters and historical summaries—with which she was involved.
The New Directions materials at The Houghton Library include the printer’s typescript of Book IV, for Book V the printer’s typescript and a typescript for magazine excerpts, page proofs of the first edition of Book V, and the galleys of Book V’s reset text for the 1963 collected Paterson.
Most of the poem’s manuscripts are with the major repositories of the poet’s material, at Buffalo and Yale. In general, Buffalo holds most of the material associated with the first three books (including another set of page proofs for Book I), and Yale the material for the last two, although Yale holds some early drafts of the first three books, and Buffalo early drafts of Book IV. In the Neil Baldwin and Steven Meyers catalogue of the Buffalo collection, the Paterson material covers El to E30. At Yale the Paterson manuscripts are numbered Za186 to Za190, although a few early drafts of Book V material are filed with some currently uncatalogued papers donated by John Thirlwall in the late 1960s.
The major gaps in the material for preparing an edition are the lack of a late draft, galleys or page proofs for Book II, the lack of galleys or page proofs for Book III, or galleys for Book IV. Some of this material could be in private hands, although my study of sale catalogues and queries to booksellers produced no leads. Fortunately, I was able to consult one copy of the page proofs of Book IV at Dartmouth College Library, where Richard Eberhart had donated the set he received for a New York Times book review.
Also unavailable to me were some of the source materials for the poem’s prose, and while this complicated a few editorial decisions, the main effect of these gaps was to limit the record I hoped to provide of the differences between the source materials and the poem’s version of them. The letters from Alva N. Turner and Edward Dahlberg excerpted in Book I are not with the collections of their correspondence to Williams at Buffalo and Yale, and the originals of the letters from Marcia Nardi used in Book II are not filed with her other letters to Williams in the two collections. I was also unable to find the sources of a few of the prose passages. But generally I was able to record the source, thanks to the excellent pioneer work of such scholars as Benjamin Sankey, Mike Weaver, John Thirlwall and Ralph Nash. In addition, Williams sometimes provided clues himself in his notes on the manuscripts, and I was further helped by being able to consult the copies of the poem owned and annotated by Thirlwall and Norman Holmes Pearson, both close friends of the poet, and by the annotations of Kathleen Hoagland on her carbons at Virginia. With the help of the manuscripts and these sources, I was also able to clarify and correct some attributions that have appeared in earlier scholarship.
Two additional sources for the poem’s prose that might have been very helpful for the edition are also missing. John Thirlwall told both Norman Holmes Pearson (in a letter of October 27, 1953, private collection) and Benjamin Sankey (see p. 228 of his Companion to Paterson) that Williams owned a bound file of The Prospector. Within the pages of this newspaper, which reprinted articles on local history, Williams and Kathleen Hoagland found some of the historical sources used in the poem. Sankey, citing Thirlwall, records Williams’ holdings as “for 1935–1936,” whereas Thirlwall told Pearson “Bill has bound together the issues from Friday, July 3, 1936 to Thursday, November 12, 1936.” The more exact description covers exactly the run of The Prospector during which its contents were devoted almost entirely to local history, rather than contemporary local news. But whatever the particular issues, I have been unable to track down Williams’ copy. Equally fruitless has been my attempt to find Kathleen Hoagland’s copy of the paper (which could be the same item that Thirlwall refers to). When she offered her Williams material for sale in 1964, she also offered a run of The Prospector. Unfortunately, Clifton Waller Barrett, who purchased the Hoagland material for the University of Virginia, chose not to purchase what might have seemed at the time a publication only tangential to Paterson, and my attempts to trace the fate of the item through dealers’ records have been to no avail. While I have been able to study other copies of the newspaper, Williams’ and Ms. Hoagland’s own copies may well have had a particular value for tracing and recording the sources of the poem.
The missing Marcia Nardi letters in Book II brought their own textual and annotation problems. At the Harry Ransom Center, University of Texas, among the papers purchased from Ms. Nardi in 1966 is a version of the long letter that, in a revised form, is the source of the prose ending Book II. The question here is, who did the revising? Although some scholars have argued that Williams revised Nardi’s prose, citing what appears to have been a copy of this Texas version, supplied by Nardi herself, Williams’ practice generally in Paterson is to cut and edit letters, but very rarely to make verbal changes, and even then of only a word or two. My full reasons for treating the Texas material as an earlier draft by Nardi and not as a copy of a source document subsequently revised by Williams, are spelled out in the textual notes. In the absence of the original letter, and of the other Nardi letter excerpted in Book II, I treat the earliest Buffalo transcription of them as a source from which to record subsequent changes (chiefly excisions). This is also the route I take with other missing source documents—turning to what appears to be the earliest version available, although of course that version may well already have been shaped by Williams.
A further complication in my attempt to record the relationship of the prose sources to the final text concerns the missing papers of the late Bloomfield historian and artist Herbert Fisher. Mr. Fisher’s involvement with the poem was uncovered by Mike Weaver in preparing his 1971 study William Carlos Williams: The American Background. Fisher wrote to Weaver that c. 1938–1940 Williams “stopped by our house … with a friend, said he had heard of me and wanted to make my acquaintance.” At that time Williams
borrowed some notes, not a manuscript, of mine which he used in his Paterson. He was making an outline for his poem. When his book came out I was surprised (not angry) to find he had copied word for word. However, the material was not original with me! I had obtained the material from William Nelson, William Scott, Mr. Longwell, and other historians.
Kitty Hoagland typed the manuscript for Mr. Williams. She had borrowed my notes and recognized them as min
e and mentioned the fact to Mr. Williams.
Later, a rumor got around that Mr. Williams had stolen the material for his book from me. He did not. The material he used he could have obtained from the original sources and I never considered it as stolen material. I freely loaned it to him. I hope this will put the record straight on that score.
Mr. Fisher’s manuscripts had disappeared by the time he was writing to Weaver, evidently the result of his 1963 move to Hot Springs, S.D. My enquiries to Mr. Fisher’s friends in Hot Springs, and my visits to the Bloomfield Public Library and the Bloomfield Historical Society failed to turn them up.
But the role of Mr. Fisher’s notes in even the earliest existent drafts of the poem is unclear. Mr. Fisher’s account, and the commentary he provided to Weaver over a number of letters, led Weaver to speculate in some of his own annotations to the poem that the verbal differences between the poem and some of its sources were the result of Mr. Fisher’s mistranscription or Williams misreading his handwriting (see Weaver 118–119). But Williams, to my knowledge, never mentions Mr. Fisher in his accounts of the poem’s development, and although Kathleen Hoagland annotates Mr. Fisher three times in her notes on the Virginia typescripts, the references are all to the opening pages of Book I.
Mr. Fisher’s letters to Weaver are replete with a deep knowledge of local history and lore, and in the early 1960s Mr. Fisher published a long series of newspaper articles on the subject. But my study of the Paterson typescripts has led me to conclude that the connection here is looser than he recalled writing to Weaver twenty-five years later. While many of the historical narratives that appear in the poem are repeated almost verbatim in various histories of Paterson and New Jersey, including the sources cited by Mr. Fisher in his letters, a close study of Williams’ earliest available transcription usually reveals which particular source he used, and this is not always the version Mr. Fisher cites to Weaver. Sometimes Mr. Fisher recalls prose passages as his particular synthesis from a variety of sources, but Williams’ typescripts and revisions clearly reveal a different source and development. Williams’ changes from his sources can be traced through the various retypings, as noted above. It appears that Mr. Fisher, along with Kathleen Hoagland and probably others, provided Williams with direction and what amounted to a bibliography of historical sources, that he then, for the most part, went on to study himself. This part of his thinking on the poem thus took its place alongside the collage-like letters of David Lyle, a correspondent from Paterson, Ms. Hoagland’s historical writing, and his own trials and experiments, to bring about his growing conception of the poem’s final form.
Within the limitations imposed by what I was able to find, I record the differences between the prose sources and the Paterson version in the notes, and record any evidence of authorial revision behind the changes. Most of the small verbal differences that are introduced into the various retypings are probably transcription slips by Williams or the typists who helped him. On the whole in the first four books I have let them stand, since their presence reflects Williams’s compositional methods. The typescripts also provide ample evidence that if a retyped sentence contained an obvious error, Williams revised for sense rather than going back to the source or even to the previous typescript (see, for example, the prose from the first Marcia Nardi letter in Book I, or the first Allen Ginsberg letter in Book IV). I have, however, corrected a handful of proper names and dates, where Williams’ own transcription accurately followed his source and the error crept in in subsequent retypings (for example, I have corrected the name of the Rev. Hooper Cumming, the date in the first line of Book I’s eel harvest, and the date of Sam Patch’s fatal dive—the last example discovered by Weaver and corrected for the 1968 printing).
I concluded, however, that the prose of Book V had to be treated as a special case. This was also the conclusion of the New Directions staff, guided by a number of well-intentioned readers, when Book V was reset for the collected Paterson in 1963. Williams’ own typed first transcription drafts are usually close to his source, although as was his practice, some sentences or paragraphs are omitted. But more than any other book, the prose of Book V goes through verbal changes in subsequent retypings that display all the characteristics of mistranscription. (Mrs. Williams commented to James Laughlin in a 1963 letter suggesting changes to Book V’s extract from Mezz Mezzrow, “the typist… I remember did that part under pressure” [New Directions archives].) As noted above, Williams’ capacity to check the work of his typists was severely diminished by 1957, and correspondence at the Ransom Center reveals that among those who helped him review the typescripts was Louis Zukofsky. The main attention of both poets clearly went into checking the accuracy of the poetry.
I felt that I had to reject the solution arrived at for Book V’s prose in the 1963 and subsequent printings, which amounted to a rather piecemeal and inconsistent restoration of the language of the original sources in Book V (and in some of Book IV). These changes were made possible by the source letters from Allen Ginsberg, Josephine Herbst, and Ezra Pound having turned up among the Williams papers at Yale. In addition, Gilbert Sorrentino forwarded a few revisions for the prose sketch of his included in Book V, to make sense of some garbled sentences. But all these changes went unrecorded, sometimes added new printing errors, and were evidently made without consulting the typescripts. The hope, reasonably, was to try to rescue Williams’ intentions in those parts of the poem prepared when his physical handicaps impaired him, but the result was that identified prose sources and letters from well-known writers were thus altered, largely in Books III, IV, and V, whereas the prose in the other books remained untouched—although it, too, differs from its sources. And little notice was taken of possible spacing errors in Book V or in the reset versions of the earlier books.
A few additional posthumous changes to Book V occurred through suggestions by John Thirlwall, who was solicited for his comments. Some further revisions, such as the Sam Patch date noted above, occurred in 1968 as a result of a note from Mike Weaver to James Laughlin following his study of the manuscripts. These changes were mostly in the nature of correcting dates and proper names, changes which I have retained, for the reasons noted above.
Guided by my attempt to try to remain sensitive to the way the poem’s compositional history, its author’s own history, and his concerns and attitudes, arc built into the fabric of the text, I have sought a compromise for the prose of Book V that I hope comes closer to these aims than the 1963 solution, and I have included that compromise within my larger general treatment of the poem’s typescripts and printed versions. If Book V’s prose were to be returned to the language of its sources, then the same treatment would need to be applied to the other four books, and this would clearly be far too much editorial intrusion. James Laughlin voiced a similar concern as coming from Mrs. Williams, when writing to Mike Weaver in January 1967: “Correcting Bill’s texts is a rather tricky business. I always consult with Mrs. Williams when a correction is suggested, and generally speaking, unless it is an obvious ‘typo,’ she prefers that things should remain as Bill wrote them. She feels, as I understand it, that his idiosyncrasies were very much part of his style, and I can see her point” (carbon in New Directions archives). Of course, with a poem like Paterson, the question of what parts of the poem “Bill wrote” is itself an issue, both editorial and critical.
For Book V’s prose I concluded to use as copy text the version as first transcribed and typed by Williams, while accepting later changes that carry evidence of his having authorized them. The result is to bring the prose of Book V in the present edition much closer to the sources than in the first edition, but not as close as in the collected edition printed in 1963 and subsequently. For verbal and spacing issues in the poetry of Book V I have been guided much more by the late typescripts and the first edition.
I have given more authority to the reset editions of Books I and II than to the reset editions of Books III and IV, and as indicated above, still less
to the posthumous reset printing of Book V. In a small number of examples in the later books I restore the reading of the printer’s typescript where no evidence exists that Williams directed a change in the printed version (for example “the tame sea” for “the time sea” of all editions, Book IV, iii, and Phyllis’s missed “to” in one of her Book IV letters). I have not interfered, as the posthumous printings did, with such features of the poem as Williams’ reproducing an error that appears in his source (the “Jonatan” of Book IV, iii). I have allowed all Williams’ typescript revisions to stand, even where they are based erroneously on the assumption of a typing error in the retyped script, for example his changing Allen Ginsberg’s “seldom dig exactly what you are doing” to “seldom did exactly what you are doing,” no doubt because he was unfamiliar with the usage. A further example is Williams’ cutting in the prose of Book V a repetition of Gilbert Sorrentino’s included for stylistic purposes.
In cases where the retyping of the prose produces a substitute word, either directly or through Williams’ response to a typing error, I leave the word, but I correct typographical errors that are not words. Mike Weaver drew New Directions’ attention to the “castor hat” of Cornelius Doremus in Book I that became a “Kastor hot,” but the text was not changed. This transformation happened a letter at a time in the typescripts, and could serve as a good example of the changes the prose undergoes. I have given Cornelius back his hat. But I have let “throsh” stand in Alva N. Turner’s letter in Book I, even though the earlier drafts, and probably the missing letter itself, read “thrash.”
The general principles governing the selection of textual and background annotations can be found at the head of Appendix C. Limitations of space and the complexity of recording non-verbal variants have made it impossible to record all differences between printings or all non-verbal corrections to the text. I have generally not recorded the following kinds of variants, unless they appear to have important critical significance:
Paterson (Revised Edition) Page 19