Call Me Digital

February 17, 2006 Premium

Technology is reshaping literary scholarship on Herman Melville through recovery of his lost annotations

Imagine, at the end of Herman Melville's Moby-Dick, that Captain Ahab and the crew of the Pequod kill the white whale instead of the other way around. That Ishmael is not alone in his escape. Steven Olsen-Smith, an associate professor of English at Boise State University, has reconstructed textual evidence that strongly suggests that Melville, whose 1851 novel stands as one of the great achievements of American literature and an enduring study of doomed monomania, entertained just such a scenario.

Mr. Olsen-Smith is part of a new wave of Melville scholars who are combining old-fashioned textual scholarship with new digital technology to track the writer's creative process, hunting down evidence that lurks in the margins of his source material and in the rare drafts that survive. "Melville scholarship is hampered by the lack of primary evidence," Mr. Olsen-Smith laments. "So many manuscripts have been lost or destroyed, so many letters are unrecovered."

In Melville: His World and Work (Knopf, 2005), Andrew Delbanco, humanities professor and director of American studies at Columbia University, observes that "detailed reconstruction of Melville's revisions of Moby-Dick is impossible since no manuscript or notes survive." But that assessment turns out to be an overly bleak one. No, Mr. Olsen-Smith doesn't have a long-lost draft of Moby-Dick up his sleeve. But he has recovered the next best thing — the notes Melville made in his copy of a critical source for Moby-Dick: Thomas Beale's 1839 book, The Natural History of the Sperm Whale.

In January, Mr. Olsen-Smith made the results available on a new Web site, Melville's Marginalia Online. On February 1, another piece of pathbreaking textual scholarship went live: a multilayered, online edition of the extant fragments of the manuscript of Typee: A Peep at Polynesian Life, Melville's first book. Published by Rotunda, the electronic imprint of the University of Virginia Press, the Typee project is the brainchild of John Bryant, a professor of English at Hofstra University and the editor of the Melville Society, which publishes Leviathan: A Journal of Melville Studies.

These projects continue a decades-old tradition of Melville scholarship that dates back to 1919, the centennial of the writer's birth, when, after years of neglect, Melville and his literary remains began to attract the close attention of scholars. But Mr. Olsen-Smith's and Mr. Bryant's projects also mark a departure from traditional ways of handling and editing manuscript material, one that takes advantage of new technology and recent turns in scholarly editing.

Decline, Fall, and Rise

The paucity of primary sources derives in large part from the downward trajectory of Melville's career. When Typee came out in 1846, he was only 27 years old. A best seller in its day, the book "made him as famous as he would ever be when he was alive," says Samuel Otter, an associate professor of English at the University of California at Berkeley and the author of Melville's Anatomies (University of California Press, 1999).

"The name died before the man," Mr. Olsen-Smith says. "Compare Melville to Mark Twain, for instance — a man who remained beloved throughout his life and after, up to the present. People saved every scrap. ... It's a different story with Melville."

Indeed, Melville spent most of his own lifetime being neglected by readers. The author of Moby-Dick (1851), "Bartleby the Scrivener" (1853), and The Confidence Man (1857) spent the last decades of his life writing poetry that nobody much wanted to read. His last great prose excursion, Billy Budd, was unfinished at his death and published posthumously.

In some ways, Melville didn't help his own cause. He demonstrated what Mr. Olsen-Smith calls "a kind of artistic exclusiveness" that shied away from attempts to make himself accessible to the general public or to posterity. He claimed, for instance, to have burned his friend Nathaniel Hawthorne's letters to him.

Through his early fame and later obscurity, Melville was a passionate reader and annotator. In the absence of manuscript material, scholars have learned to look for compositional clues in the margins of his books — when they can find them. When he died in New York City in 1891, nobody tried to keep his library intact. "Melville's reputation was washed up," Mr. Olsen-Smith points out, "and it would not have occurred to anyone to preserve the evidence."

Scholars estimate that the writer owned about 1,000 books at the time of his death. Some went to friends and family; the rest were dispersed to secondhand booksellers in Manhattan and Brooklyn, and from there made their way into public library collections and the hands of private collectors. The whereabouts of 285 titles have been tracked, which means that more than 700 could still be extant somewhere, waiting for scholars to find them.

Lost and Found

Melville's copy of Beale's The Natural History of the Sperm Whale is hardly a new find. It surfaced in the 1930s, by which time someone already had erased Melville's check marks, underlinings, and scribbles. The volume has been in the possession of Harvard University's Houghton Library since 1960.

Its influence on the author of Moby-Dick has also long been recognized. Dennis C. Marnon is administrative officer of the Houghton Library and a Melville enthusiast who has assisted several scholars, including Mr. Olsen-Smith, in the hunt for Melville's missing library. The Beale, he says, "puts us pretty close to Melville composing Moby-Dick. He's reading it at the time, and some of the marginalia not only find their way into the actual text ... but all the passages that are incorporated freely or in a modified way ... are also marked in this copy."

Scholars had assumed those markings, once erased, were lost for good. "None of them realized that the marginalia was as recoverable as it is," Mr. Olsen-Smith says. "It's been quite common to write about Melville's use of Beale without consulting the book."

It was in 1998 that he realized that at least some of Melville's markings were recoverable. To make them out, he first tried a highly sophisticated technique: squinting. Once he realized he could in fact decipher some of the characters, he used a combination of techniques to recover what he could. He subjected the erasures to different degrees of light and shadow, read with a high-powered magnifying glass, took digital photos that he could enlarge on his laptop computer, and did word searches in the text of Moby-Dick to confirm guesses about what a word or phrase might be.

"The moments of lightning striking were very few," Mr. Olsen-Smith says. "Recovery took a long time. It was letter by letter, sometimes parts of letters." He has only recently completed the process. Not every mark could be deciphered, but he says he is confident that he has recovered all that it was possible to recover.

To understand why Melville's marginal scribbles matter so deeply to a scholar that he would spend years of his life squinting at them, one has to understand the kind of relationship the writer enjoyed with his library. "Melville was extraordinarily dependent on the writings of other men," says Mr. Otter. "He has an incorporative imagination."

So, when writing, he might evoke another writer's image or argument or incorporate passages from another's work, responding to it "in detail and intimately," Mr. Otter says. "That's a significant part of the way his mind worked. ... Melville responds on the level of diction, syntax, image. That's why source study is so important in Melville studies."

On page 182 of the Beale book, for instance, in a chapter called "Chase and Capture of the Sperm Whale," Melville jotted this simile alongside a detailed description of the death throes of a harpooned whale: "As when the water issuing ... off from a fountain ... [word unrecoverable] & slowly lowers ... so the dying spout of the whale." In Moby-Dick the image is reworked and amplified: "As when by unseen hands the water is gradually drawn off from some mighty fountain, and with half-stifled melancholy gurglings the spray-column lowers and lowers to the ground — so the last long dying spout of the whale."

On page 184 of the Beale, Melville made a verbal sketch of an image of whale killers and their captain dragging the carcass of a dangerous whale, who has stove in their ship, away from the vortex caused by the ship's sinking.

Both passages can be examined on the Melville's Marginalia site. The note that pops up alongside the second annotation describes it as evidence of "the quite startling fact" that Melville at one point entertained the idea that Ahab (or an early version of him) would kill off his nemesis and that many of the Pequod's crew, not just Ishmael, would survive the hunt.

Making a List

Mr. Olsen-Smith was brought up in the scholarly tradition of Melville scholars who immersed themselves in the particulars of the writer's biographical and textual legacy, most notably Hershel Parker, emeritus professor of English at the University of Delaware and author of the two-volume Herman Melville: A Biography (Johns Hopkins University Press, 1996 and 2002), and the late Merton M. Sealts Jr. The younger scholar has worked closely with both Mr. Parker and Mr. Sealts.

In 1948, Mr. Sealts put together a "Check-List of Books Owned and Borrowed" by Melville. For fifty years, that "served as the authoritative record of title and edition information for books Melville is known to have owned and borrowed over the course of his reading life," the introduction to the Melville's Marginalia site notes.

Melville's Marginalia Online aims to continue the record begun in the "Check-List" and in editions of Sealts's book Melville's Reading. It also builds on the work that Wilson Walker Cowen did in his 1965 Harvard dissertation, Melville's Marginalia, a serious but flawed and undercirculated attempt to bring Melville's notations to the wider attention of the field.

Just as he is carrying on work begun before he entered the field, Mr. Olsen-Smith expects Melville's Marginalia to outlive him. This spring an updated version of the "Check-List" will go online, along with PDF versions of two volumes of Matthew Arnold's poetry that Melville annotated. Next up will be facsimiles of Melville's annotated copies of Aristotle, William Alger, the King James Bible, and works by Emerson and Shakespeare.

The Melville's Marginalia site makes rare archival material freely available to anyone who has access to a computer, but it has its limitations. Because of the difficulty in photographing rare books, and because of restrictions that collections like the Houghton currently have on handling such material, it cannot yet give readers a facsimile of Melville's actual copy of Beale. Instead it takes a copy of the same edition, turns it into a browsable, searchable PDF file with check marks, underlinings, and notations inserted in the spots where Melville had them, and provides an annotation for each that one accesses by positioning the cursor over the spot. (Users' computers must have Adobe Reader 7.0 installed to view the site properly.)

For scholars obsessed not only with what Melville noted but also with precisely how he noted it — what Mr. Otter calls "a language of marking" — that approximation may not be good enough. Mr. Olsen-Smith hopes that, as technology improves and restrictions are lifted, he will be able to reproduce the actual editions.

Still, he says, projects like his and Mr. Bryant's Typee mean that researchers anywhere can obtain material once available only to those lucky enough to be able to visit special collections rich in Melvilliana. "The more and more primary evidence that goes online and becomes available, the more scholars and students are going to become obliged to become engaged in archival research," he says. "There's no excuse not to anymore."

Flow and Ebb

Glimpses into the mind of the writer at work are fascinating in their own right. But once scholars have netted what they can from this kind of close archival and textual archaeology, the question becomes how to usefully integrate those revelations into new editions of Melville. Previous generations of scholarly editors sought to "combine versions of texts in order to come closer to the editors' conception of the author's final intention," Mr. Bryant writes in his book The Fluid Text: A Theory of Revision and Editing for Book and Screen (University of Michigan Press, 2002).

Mr. Bryant sets aside the idea of final intention and focuses instead on the stages of a text's evolution. As he defines it, a fluid text is "any literary work that exists in more than one version." He goes on to argue that "all works — because of the nature of texts and creativity — are fluid texts." Building off the work of scholars like Jerome McGann, who have put the emphasis on writing as "social text" rather than the individual product of genius, Mr. Bryant shifts the editorial emphasis away from one "definitive" version and onto "the multiplicity of versions" that come about as an author revises and as editors, printers, and other "collaborators" make their own changes to a manuscript.

Hence Mr. Bryant's new online edition of Typee is not a static reconstruction, but rather a series of layered texts. Its intent is to let the reader follow Melville through the drafting and revising of Typee, at least what can be reconstructed from the three extant manuscript chapter fragments. The site allows a user to compare and contrast — to lay the manuscript evidence alongside a reading text or "base version" and the text of the first print edition, in various combinations. What Mr. Bryant terms "revision narratives" explain what happened or what changed at certain points in the text. He has a critical, fluid-text analysis of the writing of Typee based entirely on manuscript evidence, Melville Unfolding, forthcoming from University of Michigan Press later this year.

The Rotunda site draws on more sophisticated technology than does Melville's Marginalia. Instead of PDF files, it uses an XML platform with a powerful and flexible search function and the ability to display different versions of the text side by side in a split-screen format, with changes marked in different colors. As part of the Rotunda series, it is not free; one must buy access to it at rates that range from $195 for an independent scholar to $545 for a university library, according to Mark H. Saunders, the imprint's manager and assistant director for marketing and sales at University of Virginia Press.

The site's flexibility and multilayered approach make it hard for the novice to navigate at first, although an extensive introduction and notes explain the particulars of the Web site.

Mr. Marnon, of the Houghton Library, describes Melville's hand as "crabby and difficult and unique," an assessment borne out by a look at the Typee manuscript. Mr. Bryant had to wrestle with the author's often challenging penmanship to decide such matters as whether an adjective used to describe young island boys is "striking" or "stripling." Scholars of a more classical textual and biographical bent, such as Mr. Parker, are certain to quarrel with some of Mr. Bryant's editorial choices and readings.

"You have to know Melville's vocabulary," says Mr. Parker, who in addition to writing his two-volume life of Melville is the longtime associate general editor of the Northwestern-Newberry editions of Herman Melville's Writings. He has also taken up the question of textual variations in his book Flawed Texts and Verbal Icons: Literary Authority and American Fiction (Northwestern University Press, 1984). "You have to know the context to know when something that looks like evidence isn't evidence. ... Part of what's involved is a sort of historical decorum, where you need to submerge yourself in the writer's milieux, plural, and vocabularies, plural."

He decries what he sees as "a certain amount of fantasy" in some of the Typee transcriptions. For instance, if Melville really did mean to use the word "striking," must we read that as his comment on the sex appeal of the young male islanders?

Mr. Bryant says that Mr. Parker's alternate suggestion, "stripling," makes some sense. One way to decide, he says, would be to search for all words ending in "-king" in the manuscript and compare them to the disputed instance of "striking/stripling."

The flexibility of the Rotunda edition allows one to take a disputed bit of text and examine the manuscript evidence for oneself, along with the editor's transcriptions — which, Mr. Bryant says, may subsequently be emended as scholars weigh in. And in fact it's deuced hard for someone unfamiliar with Melville's difficult penmanship to tell whether the word in fact is "striking" or "stripling" (although "stripling," as Mr. Parker points out, could be argued to have a more Melvillean ring to it). Turn to that section of the first print edition, which in the Rotunda edition can be set alongside the manuscript passage, and one discovers that the adjective disappears altogether.

Navigating Ambiguity

Mr. Olsen-Smith finds Mr. Bryant's approach to the Typee material "very impressive" and says he envies his ability to work with a facsimile of the actual texts. But, he says, "I do think he takes this idea of fluidity a little too hard, reading ambiguity into things where it doesn't need to exist. ... One of the dangers of a fluid-text analysis is that it favors ambiguity over decision making, over the sense of an authorized meaning or interpretation. In some cases, we have enough information to decide with some authority."

Berkeley's Mr. Otter calls such disagreements "natural and healthy. This is what scholarship is about — making public your findings and having them subject to emendation." No one knows more about Melville's life and texts than Mr. Parker, he says, but "I will never again write about Typee without consulting [Mr. Bryant's] site. ... In giving us a sense of how Melville wrote that book, it's invaluable."

Mr. Bryant's fluid-text editorial principles will also be on display in a forthcoming print edition of Moby-Dick, co-edited with Haskell Springer, a professor of English at the University of Kansas. That version will include important revision narratives alongside the text, relegating smaller editorial points and notes to a more standard scholarly apparatus at the back of the book. It will be published as part of the Longman Critical Editions series this fall.

In that venture, too, Mr. Bryant's philosophy will be in competition with the editorial approach embodied by Mr. Parker, whose Northwestern-Newberry Moby-Dick, co-edited with Harrison Hayford and G. Thomas Tanselle, has long been the standard scholarly edition. It also served as the basis for the Norton Critical Edition of Moby-Dick, which appeared in 1967, went into a second edition in 2001, and remains the most commonly used teaching text.

If a partial draft manuscript of Typee causes intense debate, imagine the variant readings possible with Moby-Dick. The differences between the first British edition and the American edition have obsessed scholars for decades. For instance, the British version put the famous opening "Extracts" at the back of the book, and dropped the epilogue in which Ishmael tells the reader that he alone survived, an omission that left British reviewers wondering whether the story was being told from beyond a watery grave.

In some ways, for teachers and scholars hoping to understand not only how Melville wrote but how he has been read in different places and eras, the more editions the better. "I believe that there's such a thing as an author's intention, but I believe that editors' efforts to restore that intention are always a combination of evidence and speculation," Mr. Otter says. "Ambiguity is a good thing. It was one of Melville's favorite words."


Web-based technology is making inroads into scholarship on the American author Herman Melville in two new projects.

  • The new multilayered online edition of the existing manuscript fragments of Melville's first book, Typee: A Peep at Polynesian Life, published by Rotunda, the electronic imprint of the University of Virginia Press (, allows the reader to follow Melville through the drafting and revising of that text. Users can compare and contrast versions — with the manuscript evidence placed alongside a reading text or "base version" and the text of the first print edition. The project is the brainchild of John Bryant, a professor of English at Hofstra University and the editor of the Melville Society, which publishes Leviathan: A Journal of Melville Studies.

  • No manuscript of Melville's classic 1851 novel, Moby-Dick, survives. But on the Web at Melville's Marginalia Online (, Steven Olsen-Smith, an associate professor of English at Boise State University, has recovered the notes that Melville made in his copy of a critical source for Moby-Dick: Thomas Beale's 1839 book, The Natural History of the Sperm Whale. At the Web site, an edition of Beale's book has been transformed into a browsable, searchable PDF file with check marks, underlinings, and notations inserted in the spots where Melville had them. The annotation for each instance is accessed by positioning the cursor over the spot at which Melville scribbled. Section: Research & Publishing Volume 52, Issue 24, Page A14