Editing the corpus of documents associated with Walt Whitman might be described as a series of compromises or as a quixotic task, depending on whether you are of a practical or a despairing inclination—a challenging effort to impose order on a notoriously unruly set of materials. An infamous picture of Whitman at the end of his life captures what appears to be chaos: the elderly poet, in a chair in his house on Mickle Street in Camden, surrounded by the debris of his writing life, a tossed sea of papers that were—loosely at best—organized into piles.[2]

The efforts of the Walt Whitman Archive over the course of its twenty-year history have been devoted to trying to make sense out of these piles and many others. Early editorial decisions that were aimed toward maximizing the intelligibility of the Archive led to the adoption of genre as a central organizing principle. One product of that framework is that in reproducing documents on the Whitman Archive, we also reproduce genre, in both text encoding and interface. Genre—defined broadly as a category or type of writing “characterized by a particular form, style, or purpose”—is one way of organizing the information in both physical and digital archives.[3] There are others. Chronology, for example, also informs the approach of the Whitman Archive and of many other collections, archives, and editions engaged in organizing archival materials.[4] Whitman Archive staff members continually make decisions about boundaries between genres, documents, and texts in order to decipher the materials and to facilitate users’ encounters with the Archive. But it is crucial as we impose orderliness to also honor messiness, not just because that is how Whitman left—and lived in—his personal archive, but also because his writings seem designed to make a mess of boundaries, smudging any clear lines we (or anyone else) might wish to impose by separating documents based on genre or era.[5]

Whitman’s correspondence presents a useful opportunity for thinking about matters of genre and order. Living at a time of rapid development in the scope and uses of the postal system, Whitman offers one example of the range and intensity of letter writing in the nineteenth-century United States.[6] Over the course of his life, the poet communicated by post with correspondents worldwide about family, personal, and professional matters. His correspondence often has complex relations to published works. In some cases, Whitman would cut up incoming letters, paste them back together, and write notes toward poems or prose pieces on the backs of them. In other cases, letter and publication merge. Whitman’s powerfully moving letter to the mother of corporal Frank Irvin following Irvin’s death, for instance, is known only as it appears in Memoranda During the War and Specimen Days; no draft or original copy of the letter has been found. In this case, we encounter this letter as an integral part of larger, later prose works.[7] What is a letter from or to Walt Whitman? How should such documents be edited and presented to readers in a digital age? What is necessary to the understanding of Whitman’s correspondence and its relationship to his published and unpublished poetry and prose, and what types of annotation and contextualization are necessary? These fundamental, simple-sounding questions have proven to be enormously complex for those of us working at the Walt Whitman Archive.

The question of “what is a letter” is a generic and editorial question, but it is also an ontological and interpretive one.[8] For editors of printed editions of correspondence, the classification often determined whether or not something made it into a published volume. For Whitman Archive editors working to digitize Whitman-related materials, classifications like genre can determine whether or not something gets published earlier rather than later; in what category it is grouped; and in which part of the Archive it appears. In both cases, classification shapes how the letter is experienced, the limitations of its reach and relation, and the conventions of its representation.[9] In the digital environment of the Whitman Archive, classification also becomes a matter of what markup we use and which parts of the letter we choose to transcribe and encode.[10] In our treatment and representation of Whitman’s correspondence, we have borrowed from the conventions of printed volumes (providing transcriptions that retain, as much as possible, the original spelling and punctuation of the letters; using brackets to mark illegible or unclear words; and adding footnotes as explanatory apparatus). But we expect to do more in future years to update our editorial policy and our display and search functionality to begin to account for the complexity of many of the archival objects associated with Whitman. This process has returned us to a long history of correspondence editing as well as the tension between classification or order and randomness or serendipity—letter or litter—as we consider what lies ahead for the Whitman Archive and its ongoing engagement with what Jerome McGann has called “the problem of knowledge representation.”[11]

In this essay, we focus on genre as an example of one seemingly coherent organizing principle that can actually disorient or needlessly limit the representative possibilities for an archival object. We discuss some of the edge cases of Whitman’s correspondence—documents whose generic boundaries are difficult to determine—and explore their significance for editorial purposes and the ways that our thinking about these objects has affected how we have conceptualized and continue to conceptualize the structure of the Whitman Archive more generally. The epigraph to this essay represents one of several times over the course of Whitman’s late-life conversations, transcribed immediately afterward by Horace Traubel and eventually published in nine volumes, when Whitman and Traubel joke about a letter or another document surfacing from the litter of papers in the poet’s room in Camden, New Jersey. What is the relationship of the letter to the litter?[12] It is important not to make negative assumptions about litter, given Whitman’s fondness for “compost,” “debris,” “leaves-droppings,” and the like. Whitman’s room, like the Whitman Archive, was a broader context, an environment that impacted the material form of his papers and that structures the ways they can be interpreted.

As we work to reorder Whitman’s piles and reproduce them in a digital environment, we revisit the question of the difference between letter and litter and the usefulness and intelligibility of genre-based organization. The Whitman Archive is moving toward an editorial approach that embraces the potential for multiple forms of categorization and display, a structure that allows objects to be both letter and litter, and many other things besides. Documents that resist order point to things other than genre, such as the complexities of composition and the effects of Whitman’s environment and his paralysis on the shape and substance as well as the subjects of his compositions in his later years. Reading Whitman’s correspondence as woven into his prose and poetry illuminates both. Such complexities suggest that we may need to revisit editorial commonplaces in the digital era. Ultimately we see the future of the Whitman Archive as an effort to reach beyond the limitations of genre or temporality or any other single organizing principle by developing the framework to represent a document in multiple ways, within multiple sections, with options to navigate between one document and the next. In doing so we leverage new media capacities to represent process as well as product, document as well as text, and to attend to the history of an archival object in all its sometimes frustrating ambiguity and multitudes.

The keys to a genre are the particulars of its form, style, or purpose—the characteristics or the protocols associated with it as a category of writing. We might take as our starting point the definition of a letter by Edward Vanhoutte and Ron Van den Branden as “a physical channel through which a communicative situation is established,” defined by the distinctive features of time, receiver, and sender.[13] Certainly in the case of Whitman there are plenty of letters that fit such a basic definition. Still, Whitman’s correspondence, like that of many literary figures, can sometimes be difficult to distinguish from other kinds of writing.[14] Though there are many documents that take the conventional form of a letter—date, salutation, body, signature—there are also fragments of letters, as well as iterations of letters in copies or drafts. On one end of the spectrum are jottings or loose notes that seem to be directed to a third party or that have some of the formal features of a letter.

Figure 2: Fragment of a Whitman letter, written on the verso of an undated letter to Whitman from Joseph B. Marvin. Charles E. Feinberg Collection, Library of Congress, Washington, DC.

The signature at the bottom in pencil, combined with the instructions to a third party—“Remember me to Coley, John Towers, Jim Sorrell, David Stevens & all the boys”—seem like unmistakable proof that this fragment represents a line toward a letter. And, interestingly, these words appear again, in a draft of a nearly complete letter from Whitman to Peter Doyle dated October 14.

[Click on image to enlarge] Figure 3: The second leaf of a draft letter from Whitman to Peter Doyle, October 14, [1868]. Charles E. Feinberg Collection, Library of Congress, Washington, DC.

In the draft, the line from the fragment appears, seemingly in a later iteration, but has been crossed out. We do not have a copy of the final letter, if one exists. The only evidence that this particular fragment ever made it into a letter that was sent is its presence in a printed transcription by Whitman’s friend and executor Richard Maurice Bucke, who edited and published a collection of letters from Whitman to Peter Doyle in 1897.[15] This collection, one of several published after Whitman’s death that focus on the letters between the poet and a single correspondent, is framed with an introductory mention of Whitman’s “Calamus” poems, and what Bucke calls the “exceptional and remarkable” friendship between Whitman and Doyle.[16] In the introduction, Bucke notes Whitman’s tendency in his letters to send his love to a group of friends and acquaintances known by Doyle.

It is unclear whether Bucke, whose collection of letters was based on a set given to him by Doyle, possessed a final version of this letter that included the line, or if he simply restored the line to the transcription in spite of its deletion in the draft.[17] The line is consistent with Bucke’s expressed intention in his edition to emphasize and provide evidence of Whitman’s “calamus” friendships, and so he may have had this reason to restore it if he was working from the draft. Conceivably, the line was crossed out later, though neighboring corrections in Whitman’s hand in the same pencil as the overstrike would suggest that Whitman was responsible for the deletion. In any case, the collection of Doyle letters was republished by Whitman’s executors in The Complete Writings of Walt Whitman in 1902, with the excised line still included in the October 14 letter.[18]

This document, beginning “More about William Blake,” may represent a partial draft of a draft of a letter. Or it could be that Whitman copied the line onto this note when he crossed it out of the draft letter to Doyle, in order to save it for later. Or perhaps he wished to reproduce or remember the line in another epistolary context. Is “More about William Blake” a letter, a note, a deletion, or all of the above? Presumably a draft of a letter doesn’t establish a communicative situation—at least not until an editor comes along to read the draft! In the case of Whitman, who received at least one letter from an auctioneer about the prices for which his manuscripts had sold at auction, it is possible to speculate that, chaotic though the papers in his room may have seemed, his decisions to retain manuscripts or hand them to his friends were made with an eye to readers and editors to come.[19] On the Whitman Archive, we have not yet begun editing drafts of letters, unless they are the only remaining copy of a letter,[20] but in the future we plan to present all drafts, so that users can see iterations, mark the progression from draft to final copy, and sometimes better grasp nuances in his writing and his relationships.

At times the existence of a draft alone can be highly illuminating, as in the case of Whitman’s 1890 response to John Addington Symonds’s repeated inquiries about the meaning of the “Calamus” cluster of poems.[21] The final copy is no longer extant, but if it were, that final copy would almost certainly convey less about Whitman’s response than does a mere image of the draft. This is the famous letter in which Whitman claimed: “Tho’ always unmarried I have had six children.” Equaling that whopper is his final sentence, described by Edwin Haviland Miller as “artfully deceptive”: “I see I have written with haste & too great effusion—but let it stand.”[22]

[Click on image to enlarge] Figure 4: Recto of a draft letter from Whitman to John Addington Symonds, dated August 19, 1890. Charles E. Feinberg Collection, Library of Congress, Washington, DC.

Painstakingly crafted and pasted together, bearing witness to numerous deletions and additions that led to the final version of the letter, this draft belies Whitman’s description of writing “with haste.”[23] The material composition of this draft letter, with its paste-ons and revisions, is as revelatory as the content, and the two are mutually explanatory even as they contradict each other. Here an image of the object, a link to Symonds’s letter, and an explanation that this is a draft as well as a transcription of the text of the letter are necessary to understanding Whitman’s response to Symonds’s query about “Calamus.”[24]

Other explanation is necessary, too. Many have discussed the front of this draft; few have mentioned the back of it.[25] The verso of the draft letter reveals that the draft has been pasted together out of various materials, both printed and handwritten.

[Click on image to enlarge] Figure 5: Verso of a draft letter from Whitman to John Addington Symonds, dated August 19, 1890. Charles E. Feinberg Collection, Library of Congress, Washington, DC.

One scrap is a printed poem titled “Address to the Veterans,” attributed to “William Errickson.” Another is a letter from Richard Maurice Bucke that can be dated between February 5 and 15, 1890, describing the death of Bucke’s sister-in-law Matilda Gurd. Still another is a partial envelope with the return address of Dora Warnecke, a resident of San Pablo, California.[26] A series of other letters and objects have been recycled into the draft of Whitman’s letter to Symonds. How are we to treat this new composite object? Typically, for publication on the Whitman Archive, we would separate out the various parts, publishing each letter individually and linking them in the annotations, providing images of both recto and verso. But a robust system of relation, created and implemented through both the annotations and the encoding, is needed to fully represent the complexity of this object, which singlehandedly muddies the distinctions between incoming and outgoing letter, as well as draft and final copy.

Calamus, the volume of letters from Whitman to Doyle edited by Bucke, itself begins with an item that may or may not be classified as a letter: Bucke reproduces a note written on the flyleaf of a copy of Specimen Days that Whitman sent to Doyle in 1883.

[Click on image to enlarge] Figure 6: Walt Whitman, Calamus: A Series of Letters Written During the Years 1868–1880 by Walt Whitman to a Young Friend (Peter Doyle), ed. Richard Maurice Bucke (Boston: Laurens Maynard, 1897), iii. The copy of Specimen Days with the original inscription to Doyle is at the Harry Ransom Center, University of Texas at Austin (shelfmark PS 3220 A1 1882, copy 1).

Does this inscription, headed “Pete” and signed “W.W.,” originally appearing at the front of a volume that traveled, as a letter would, through the mail, constitute a letter? The placement of the message in Specimen Days means the flyleaf inscription could be read as an enclosure or simply as an introduction or a paratext to Specimen Days. Thinking of this inscription as a letter is useful because it is clearly one in a series of communications with Doyle. As Ted Genoways observes in a footnote to his transcription of the flyleaf in volume 7 of The Correspondence, “this note is significant, because it constitutes the first correspondence from [Whitman] to Doyle since July 1880. It appears that writing Specimen Days stirred [Whitman]’s memories of the times he shared with Doyle in Washington” (69). Not thinking of the inscription as correspondence, within a history of correspondence, puts the reader at a disadvantage in interpreting the meaning of this inscription.

Perhaps the most famous letter associated with Whitman, extraordinary in content if traditional in form, was sent to the poet by Ralph Waldo Emerson after the publication of the first edition of Leaves of Grass in 1855. To Emerson’s dismay, Whitman reprinted the missive with his own extended reply in the “Leaves-Droppings” section of the 1856 edition of Leaves, going so far as to use part of Emerson’s letter as a blurb on the spine of the book. Even before doing this, Whitman had the Emerson letter printed in a newspaper, turning a private communication into a public one. Cast out of the insular network of person-to-person correspondence and into the public role of endorsement, Emerson’s letter changes in form and function, becoming both paratext and provocation to its sender (and its numerous unforeseen recipients).[27] Interestingly, however, the 1856 Leaves of Grass as a whole is framed in Whitman’s response as part of a letter—an enclosure, as it were: “HERE are thirty-two Poems, which I send you, dear Friend and Master, not having found how I could satisfy myself with sending any usual acknowledgment of your letter.”[28] The genre of Whitman’s extensive response letter is anything but straightforward, with even the speech-act words that seem to perform gift-giving or presentation (“Here are thirty-two poems”) coming after the poems in the volume. The document is labeled “Letter to Ralph Waldo Emerson” and contains the formal features of a letter—date, location, salutation, body, and signature—but the prose of the letter, despite its frequent address to “Master,” is akin to an essay along the lines of the first preface to Leaves of Grass, and of course its appearance within a book (the letter was never sent as a manuscript to Emerson) takes it out of a context of personal correspondence.

These documents pose a challenge for the editor seeking to situate them within some kind of coherent system, whether that system is based on genre, chronology, or some other organizing principle. Each document can be interpreted as a letter or part of a letter that was composed or sent at a particular date or time. But this only tells one part of the story of the document. Each segment of a repurposed object or series of notes could be classified and dated differently. The publication and circulation of Emerson’s letter after it was sent and received are central to the meaning of that letter. Representing these objects online without collapsing the range of interpretive possibilities they introduce requires both creativity and diligence. A digital environment offers new potential to represent single objects in multiple ways, but also to create more robust links among objects, allowing the user to move, for instance, from the draft of a letter to the final version, or from a transcription of the letter as Emerson sent it to its position within the 1856 Leaves of Grass. Digital scans or facsimiles also help to clarify the material complexities of some of these documents, even as searchable transcriptions offer another way of navigating them tailored more specifically to user interests.

In fact, the digital scans of the letters on the Whitman Archive might come as a surprise to readers familiar with the printed editions of Whitman’s correspondence. A world quite apart from the neat typeface and standardized spacing of the printed letters, the manuscripts feature all manner of messiness, including multiple hands, smudges, deletions, markings of the holding repository, postmarks, and notes written by owners. Here the messiness reflects the history of the individual document and also the history of previous editors. Many of the letters include a note at the top written by Horace Traubel, locating the letter within his With Walt Whitman in Camden, where Traubel frequently includes complete transcriptions of letters Whitman hands to him in the course of their conversations.

Figure 7: Traubel’s note at the top of a letter from James Redpath to Whitman (October 5, 1886), “see notes Aug 26 & 30, ’88,” points to entries in With Walt Whitman in Camden dated August 26, 1888, and August 30, 1888. The latter includes a transcription of the letter, with Whitman’s commentary. Charles E. Feinberg Collection, Library of Congress, Washington, DC.

In their interactions, Traubel often reads aloud a letter to or from Whitman, providing the occasion for reliving a historical moment and eliciting commentary from Whitman about the sender, the situation, or the letter itself, as the epigraph to this essay suggests.[29]

Just as Whitman preemptively shaped the critical conversation about his poetry through placement of his own anonymous reviews of Leaves of Grass, here again we find him beating us to the punch: in tandem with Traubel, he became the first to provide extended commentary on the correspondence of Walt Whitman. In fact, Traubel’s extraordinary years-long interview, peppered by transcriptions of letters, might be said to render him one of the first editors of Whitman’s correspondence.[30] It is worth noting in this context Traubel’s process of editorial selection, which is guided by Whitman himself. Letters seemingly appear in With Walt Whitman in Camden organically, as they emerge from the litter, or as they are given to Traubel by the poet. After one visit, for instance, Traubel writes, “W. handed me a stained letter which he wished me to read back to him.”[31] Acting as axes of conversation, the letters provoke reflections on the past, on mutual acquaintances and friends of Whitman and Traubel, on the poet’s work, and on many other topics. Set into the flow of the discussion, the letters help give rise to a rich context of day-to-day rituals, reminiscences, and interactions. This is no ordinary “edition” of letters, and yet in its very resistance to cordoning off the archive from daily life, its openness to the combination of factors that led to the unearthing of any given letter, and its careful representation of the social scene of the letter’s second life and rearticulation as well as the text of the letter itself, it offers a provocative view of what might emerge out of an editorial approach that takes seriously the conjunction of letter and litter, or that resists the ready logic of genre- and chronology-based categorization.[32]

Whitman did more than read and comment on his correspondence. Particularly once he was less mobile, after his paralysis, he would cut some letters into pieces, paste the fragments back together, and use them as the scaffolding for poetry or prose manuscripts. One example, currently held at the Walt Whitman House in Camden, is a draft of the poem that would eventually be published as “You Tides With Ceaseless Swell.”

[Click on image to enlarge] Figure 8: Recto image, manuscript draft of “You Tides With Ceaseless Swell.” Courtesy of Walt Whitman House, Camden, NJ. Reproduced with permission.

As he had with the Symonds draft, in order to create this manuscript, Whitman cut apart and pasted back together several other pieces of paper. In this case those pieces included four different letters: one from Whitelaw Reid, dated July 17, [1878]; another from Richard J. Hinton, dated September 10, 1871; yet another from William J. Sewell, dated January 8, 1884; and a final letter, undated, from an unidentified sender.

[Click on image to enlarge] Figure 9: Verso image, manuscript draft of “You Tides With Ceaseless Swell,” constructed out of a series of letters sent to Whitman. Courtesy of Walt Whitman House, Camden, NJ.

Scholars have described the importance of collaging to Whitman’s poetic and composition practices, but at first glance this seems to be a pasting-together that functions at the level of conscious composition only depending on which side of the document one examines.[33] That is, the piecing together of the poetry lines seems highly purposeful, while the shards and remains of the fragmentary letters are pasted together in apparently random fashion. The content of the letters may have contributed to the poetry manuscript, but if so, it is not immediately apparent. The letters were probably selected not because of any content but because of the available white space left on a leaf. Spanning over a decade, the letters also reflect Whitman’s practice of keeping papers and letters for extended periods of time, even as he eventually used them for scratch paper.[34] What literary sense can be made of this practice? Certainly the letters constitute bibliographical evidence. Based on the letters, we can date Whitman’s composition: if, as Geoffrey Sill argues, the title and opening two lines were written first on the back of the 1878 letter, the poetry draft was almost certainly composed between that date and the months after January 8, 1884, the latest date of one of the letters used in composition.[35] The 1871 Hinton letter seemingly acts only as the scaffolding for the bits as Whitman pasted them together.[36]

This manuscript is striking for many reasons, not least the problems it poses for editorial treatment on the Whitman Archive. As a material object, it is a study: in one part of the poem draft, Whitman appears to have used a piece of paper to overwrite previous lines with newly composed ones, but the bulk of the manuscript appears to have been written in one order, then cut to pieces, shuffled, and glued back in a different order. The manuscript in its current order reads[37] :

Figure 10: Transcription of manuscript draft of “You Tides With Ceaseless Swell.” At the top left corner of the pasted-down piece of paper that includes the first line under the title and the next one, there is a partial note and a series of numbers written in pencil by someone other than Whitman. We have presented the line starting "What clue" with a gray background because it has been crossed out with two vertical lines in the original manuscript.

It is difficult to tell what the original order of the lines was, in part because additional pieces of paper have been pasted on top of the letter fragments used to form the bottom layer. In any event, it seems clear that the first two lines—beginning with “You”—have been added later, in pencil, pasted on top of the original first two lines, which like the rest of the manuscript were written in ink. Further corrections have been made in purple pencil. To form the foundation for the clipped lines, the letter from Hinton was cut into two pieces, and the letter from Sewell pasted in between.[38]

Sill has made a case for the interpretive importance of this reordering, which resituates the subject of the line beginning with the word “holding” and demonstrates Whitman’s close attention to the structure and prosody of his lines. Another case could be made for the interpretive importance of the ostensibly unrelated base materials used in Whitman’s late-life poetic “collage.” Reading this poetry manuscript in relation to its material features highlights its multi-generic, multi-temporal, and multidimensional characteristics. Whitman had shuffled poetic lines as early as the very beginning of his work as a poet, putting clipped subjects or lines in an envelope or on a string and flipping them around until he was satisfied with the order.[39] In his later, less mobile years, this shuffling drew on the paper that surrounded him, and so letters sent to him became part of his poetic constructions, recycled but not thoroughly composted, visible still at least in part in the artifacts that are his poems, prose, and notes toward essays, letters, and lectures. His final product here reveals a temporal and material conjunction: decades-old letters are joined to pieces of recent letters to form the foundation for a new composition. In the case at hand, the poem has a striking resonance with its material manifestation. Like the tides with “ceaseless swell and ebb,” the poet revises, acting as “unseen force, centripetal, centrifugal,” pasting together a chain not in this case of “sun, moon, earth, and all the constellations,” but rather of near-to-hand letters, creating a new archive out of the old one that surrounds him.

While the content of the letters does not inform the poetry in obvious ways, it does suggest the resonance of the letters as the base materials out of which the poetry is wrought. Brought together to form the physical foundation for Whitman’s poetry manuscript, these letters represent an intriguing cross-section of nineteenth-century society. The partial letter from Sewell, a railroad official, former Union general, and U.S. senator from New Jersey, reads like a form letter, probably sending Whitman his annual pass to the West Jersey Railroad, of which Sewell was then vice president. The letter from Hinton, a journalist who met Whitman in the hospitals during the Civil War, alerts Whitman to an article about the poet published in the Springfield Republican. The letter from Reid, editor of the New York Tribune, who, like Hinton, first met Whitman in the hospitals during the Civil War, is difficult to decipher but seems to forward a check and to thank Whitman for sending him a volume. These letters are practical transactions that navigate major nineteenth-century industries—railroads, newspapers—speaking to the poet’s transportation, publication, and distribution practices, all now quite literally constituent of his poetic composition practices. The world becomes the word. In pasting the letters together, the poet mimics the tides in acting as “boundless aggregate,” with scissors and glue “holding the universe as one.” He makes the written words and voices of his correspondents integral parts of his poetic project—physically, of course, but also intellectually, serving as they do as evidence of the everyday concerns out of which he wove his poetry. Letters, litter, and verse are all but inseparable.

Again, it goes without saying that this practice of Whitman’s, fascinating though it may be for contemplating his methods and materials of poetic composition, complicates the project of editing his correspondence. Here, as in the case of the Symonds draft, the letters used in the construction of the manuscript have been reduced to fragments. Sender, recipient, and date have been removed in one case. This object involves a total of five authors, writing in five different hands and at least two genres on multiple scraps of paper spanning almost fifteen years. In a number of provocative ways, this object contains multitudes. The challenge becomes how to represent this object as a number of different, mutually informing things—letters, litter, poetry—while facilitating the user’s access to it through multiple channels and across multiple forms of display. At stake is the “subtle indirection and significance” of such documents—the ways in which seemingly disparate items, assembled through cutting, pasting, and repurposing, might speak to each other and to Whitman’s composition practices more generally, and how we can leverage our own processes of representing them on the Whitman Archive so that such meanings become accessible or at least do not get lost in remediation.

In editing, as in letter writing, one engages a set of people as well as a set of conventions. In the case of letter writing, as Whitman’s correspondence demonstrates, those people can be contemporaries, a larger public, or imagined audiences of the future. The engagement can take many forms, to which drafts and material characteristics sometimes offer clues: intimacy, adulation, studied nonchalance—even destruction and reconstruction, as Whitman’s repurposing suggests. If the Collected Writings of Walt Whitman had dealt with Whitman’s poetry manuscripts, as was the original intention, the letters on the back of “You tides with ceaseless swell and ebb” would have warranted a footnote; perhaps a facsimile image would even have been included, because the materiality of this document has interesting implications for Whitman’s composition practices and the structure of this poem. More challenging may have been the question of what would have happened had the Collected Writings included incoming in addition to outgoing correspondence. Would these incoming letters, partial, cut up, and repurposed, have been included in a printed collection of Whitman’s correspondence? Every correspondence project makes decisions about what constitutes a letter, in the face of any number of ambiguities. That so few discussions of Whitman’s draft of his 1890 letter to Symonds mention that it was pasted together out of other things partly results from previous editorial treatment: Edwin Haviland Miller, working within the constraints of a printed volume, did not provide a facsimile of the letter, nor did he mention the materials out of which it was made, and he limited his transcription of it to the text on the recto.[40]

In treating the correspondence, the Whitman Archive has benefited from editorial predecessors. The volumes of outgoing Whitman correspondence compiled by Miller and supplemented by Genoways have stood the test of time, representing largely accurate and editorially responsible collections, and they have been invaluable resources, saving Whitman Archive staff many of the headaches of identifying locations of particular letters. The most apparent shortcoming of those editions was something over which the editors had little control: the print medium. The drawbacks of print are some of the primary limitations lamented by twentieth-century editors of correspondence, and indeed there are extraordinary benefits to digital collections of correspondence. These include the ability to add newly discovered letters as they surface, and to put them in the desired order (chronological or by correspondent, for example) rather than including them in an appendix or a supplement. A digital environment also affords us the space to include both incoming and outgoing letters and to render enclosures in full, simplifying to some degree the process of selection by expanding the bounds of publication to items that simply wouldn’t have been feasible to include from the perspective of a print publisher, for whom each extra page costs money.

In the first section of this essay, we raised the question of what constitutes a Whitman letter. From the standpoint of reading and editing in the digital era, we might ask why it matters. Restrictions of space, paper, and ink no longer lend the same kind of urgency to the process of classifying and selecting by genre as they did when a limited number of printed volumes could be produced. On the web, given enough server space and an existing, functional architecture, the number of letters to be treated can multiply quickly, and anything that seems at all like a letter associated with an author could conceivably be included. There are several factors that render classification and selection still relevant in a digital era, however. One is the coherence and navigability of the interface. Another is the structure of grant applications for the letters (organized in the case of the Whitman Archive, for instance, by era), which can determine publication insofar as which letters appear first, and which (like undated letters) seem likely to wait until the end of the project. And it is perhaps worth considering that, where space is no object, time begins to loom large: waiting until the end of a complete edition of Whitman’s incoming and outgoing correspondence is wait enough; waiting until the end of an edition of all Whitman’s writings, including the correspondence, may well be an endless proposition.[41]

In keeping with the loosely genre-based editorial practice we have developed for the Whitman Archive, the documents published on the Archive have been sorted into categories, including “Books by Whitman,” “Manuscripts,” “Notebooks,” and “Letters.” Genre lends one form of coherence to a project, offering limitations, direction, and selection criteria for work designed to meet a particular goal or to specify the obligations associated with grant funding. As a result, Whitman Archive grant proposals have also often been organized by genre or chronology (or both). Early grant applications submitted by the Archive focused on poetry manuscripts. More recently, the Archive has been awarded grants devoted to digitizing, cataloging, and editing prose manuscripts, fiction, journalism, marginalia, and correspondence. But the examples discussed in this essay, and many other such cases, have continuously presented a challenge, because genre distinctions are themselves sometimes fuzzy, and individual documents may be multi-generic in nature. Preserving the “subtle indirection” of the documents, the messiness as well as the order, may require thinking differently about genre, and instating documents within their longer and sometimes whimsical histories.

The archival objects discussed in this essay fit into a category of correspondence, but many of them also fit comfortably into other categories—poetry manuscript, published book, printed poetry, note. These documents resist efforts to sort and silo by genre. Correspondence draws attention to the centrality of sociability, intimacy, and connection, even as it also points to publicity and mediation. Without the image of Whitman’s room in Camden, it is more difficult to understand objects like the Symonds letter and the “You Tides” manuscript, constructed by the poet out of the papers that surrounded and at times seemed to engulf him. Without Traubel’s careful accounts of his conversations with Whitman in With Walt Whitman in Camden, most of Whitman’s commentary about the letters he sent and received would be lacking. An advantage and a challenge of working on the Whitman Archive is that, whatever limited intentions we may have had when we set out to edit a set of documents like Whitman’s correspondence, we have ended up thinking much more broadly about the range of documents that comprise an author-centered digital archive, and that test its boundaries and coherence.

As the Archive expands, it becomes increasingly clear how difficult much of Whitman’s archive is to categorize definitively. One way we plan to respond to this challenge is to move toward an interface or user experience that is predicated on users’ questions or searches and that allows for a single object to appear in multiple sections or categories. Another way forward is to think about ways to create and optimize a system of relations, representing the fact that a letter can also be a poetry manuscript, as well as several other letters, or to represent the publication and revision history of a document, showing, for instance, that a fragment that seems like a letter could point both to a draft letter, from which it has been excised, and to one of the earliest edited collections of Whitman’s correspondence. Editors of the past have often focused on specific items or collections of items, or on specific authors; objects like the ones discussed here encourage us to think instead about ongoing, transformative connections between items and authors, between authors and their friends and family, and between authors and editors. Such links remind us that any given item did not exist in isolation, and that a long history of editing continues to affect how documents are read and re-edited in different media.

Editions of correspondence are themselves a genre of sorts, with their own conventions, forms, purposes, and expectations. Whitman’s correspondence on the Whitman Archive is, ultimately, an editorial palimpsest. It, too, is an act of reconstruction. We draw heavily on and update the notes and the transcriptions produced by Miller and Genoways, but we also derive transcriptions from nearly all the many editions of letters to and from Whitman that have been published since the nineteenth century. Like Whitman, we have recycled and remediated, and like him we hope to push back against the apparent rigidity of category and transform how users experience these objects by creating new pathways for them to quickly shift contexts, locating items in different parts of the Archive by clicking from one letter to another, pulling up the digital images or our encoding, accessing encyclopedia entries about particular correspondents, or navigating to the version of a letter that appears in With Walt Whitman in Camden. But we also intend with our reproductions of these materials to extend into the future the many forms of sociability enacted by the letters, inviting others to reproduce our transcriptions and the interpretations to which they give rise, using the TEI encoding and other data we have created to pursue still more pathways, continuing to compost and participate in the centripetal and centrifugal force that is historical research and discovery.

In 1888 Whitman said to Traubel, “I suppose I have done a lot less general letter-writing than most men—I was not a voluminous letter writer—when I wrote at all it was mainly with a very definite notion of something very practical that needed to be said. . . . nothing ever came to me in a hurry: even my storms came taking their time.”[42] Later he would remark that he was “not ‘a correspondent by nature.’”[43] While Whitman was a less “voluminous” correspondent than some writers, this comment does not account for those who wrote to him; we have identified over 3,000 incoming and outgoing letters from 1860 to 1887 alone, ranging from postal cards to sixteen-page effusions, and we expect to add many more as we embark on editing Whitman’s early and late-life correspondence.

McGann has called editing a creative act, and so it is—but it is also a political act, with its own long history. One important feature of our work on Whitman’s correspondence is that, despite the continued exertion of a gravitational pull by Whitman, the incoming letters introduce a range of new voices and hands to the Whitman Archive. Readers now can experience the volubility and intensity of William D. O’Connor, bent on identifying and reckoning with those who would attack or censor Whitman’s Leaves of Grass. They can read the thoughtful and engaging letters of Whitman’s close friend and English admirer Anne Gilchrist. They can peruse the reflections of Alabama farmer and philosopher John Newton Johnson, who had a lengthy correspondence with Whitman and came to visit him in Camden in 1887, or they can explore the connection between Whitman and various friends, acquaintances, and admirers from all over the world.

Editing Whitman’s correspondence for publication on the Whitman Archive turns us back to our working definition of correspondence, but it also, perhaps more importantly, prompts us to consider what digitization allows and requires of editors of archival materials. In one vision of the future, the Whitman Archive would represent the overlap between nineteenth-century (composition, revision, and archival) practices and twenty-first century (editorial) practices as part of the editorial product, that is, the documents that appear on the Whitman Archive. These practices might be represented in how such objects are defined and explained, how they are related to other documents on the Archive, and how they are navigated within the broader structure of the site. The evolution in the treatment of documents by the Whitman Archive over time is moving toward an emphasis on the extent to which many of these objects defy the apparent order of categories of genre and question the neat hierarchy of TEI/XML tags that helps to enforce that order by marking distinctions between handwriting and print, poetic line and paragraph. As the digital humanities continues to work on the forms of insight that can result from interpreting documents as data, the acts of qualifying the apparent order of the machine-readable text and emphasizing the messiness—and unpredictability, complexity, and creativity—of writing and editing (and being human) become ever more pressing.

As we confront the piles that were once in Whitman’s room in Camden and reorder them in digital space, one way of honoring the messiness in the new medium is by multiplying the representations. We have begun to move in this direction by offering multiple views of a document, including facsimile images, searchable transcriptions, and a view of the XML encoding. Another way is by encouraging an understanding of documents and texts in relation to other documents, texts, and conversations, as Traubel’s With Walt Whitman in Camden insisted on doing. It may be that the seemingly marginal conjunctions of editorial phenomena and oral and documentary history represented by Traubel’s notes on Whitman’s letters, signaling the replication of the archive in more than one place, offer a productive model for thinking about how to edit Whitman documents in digital space. In the end, the Whitman Archive hopes to create a richer environment for scholars, balancing the project of making sense of Whitman’s archive and of thinking about our own desires to make sense of the digital archive, and looking for opportunities to further expand access to Whitman-related documents by creating multiple avenues for users to explore the transforming and transformative relationships between letters, litter, and literature.