The Author Machine

CHARLTON HINMAN HAD a problem. For his study of the printing history of Shakespeare’s First Folio, he needed to collate all 79 copies held at Folger Library in Washington, DC. This meant comparing each line, each word, each letter, to identify the many changes introduced into the work through the process of its publication. Hinman did a bit of back-of-the-envelope math. The task would take him over 40 years.

Hinman then tried a new tack: instead of collating by hand, he built a machine to assist him. The Hinman collator was a boxy, gunmetal-gray behemoth. Its user would sit in front of the machine with a platform to their left and right, upon which lay seemingly identical copies of a book, each opened to the same page. A series of mirrors and flashing lights revealed to the person peering into the Hinman’s binocular viewer that the pages were often not identical after all. Any differences between them would flicker or blur, announcing themselves as variants without the painstakingly slow work of side-by-side comparison.

The collator was born of Hinman’s experiences in World War II, but the machine would go on to be enlisted into the Cold War as part of a massive effort to shore up the cultural credibility of American literature abroad. In the early 1960s, a new organization called the Center for Editions of American Authors (CEAA) deployed the Hinman in the gargantuan task of editing and publishing the complete works of 15 canonical writers in hefty multivolume definitive editions. The CEAA quickly persuaded the US government that the Hinman could “purify” the works of celebrated American authors from the “corruptions” of their publishing milieu. Once variants were identified with the Hinman, teams of editors would classify them as “authorial” or “non-authorial” changes. Purging non-authorial variants would create a text representing the author’s original intentions. For the first time in history, the thoughts of Hawthorne, Melville, Emerson, Whitman, and Thoreau would reach readers unimpeded by the impositions of editors, publishers, typesetters, and others. Their thoughts would then sit bound in deluxe volumes in US Information Service libraries around the world, displaying American literary and cultural supremacy.

The Hinman’s ability to generate data about literary objects twinned humanities scholarship with technoscientific progress at a crucial moment. The CEAA’s seemingly objective approach to the study of literature put it in an advantageous position to become the first flagship initiative of the National Endowment for the Humanities in 1965. In doing so, the organization established a funding model for humanities scholarship that split scholars and critics into vociferously contentious camps. Simultaneously celebrated as a tool to help purge the American canon of errors and decried as the symbol of the humanities’ subservience to technology, the Hinman was the locus of a controversy that looks very much like the ongoing debate over the digital humanities.

Charlton Hinman was of an analytical cast of mind, having received his PhD from the University of Virginia in 1941 under the preeminent American textual scholar of the 20th century, Fredson Bowers. Bowers was a second-generation practitioner of the New Bibliography, a discipline pioneered by British scholars who wished to bring a scientific approach to a field that they saw as regrettably amateurish. Though Hinman’s Shakespeare project was interrupted by World War II, his time in the military did not prevent him from mulling over the problem of machine collation. He served as a cryptanalyst in a code-breaking unit with several other future bibliographers—and with Bowers as his commanding officer, no less—giving him plenty of opportunity to discuss the complexity of large-scale textual comparison. He heard of reconnaissance missions in which aerial photographs were taken of enemy targets before and after bombing runs. These photographs would be viewed in rapid succession through a projector, creating a flickering effect where bombs had done damage. The story turned out to be false, but its underlying principle would stay with Hinman. His collator would use a series of mirrors to superimpose images of the same page from two copies of the same edition of a work. Illuminating these pages alternately causes the user to see them like an infinitely repeating flip-book, with areas of difference dancing about.

A Hinman collator is about six feet tall and five feet wide and looks like it was stolen from the set of a 1950s sci-fi flick. Covering the University of Virginia’s acquisition of a Hinman, the Richmond Times-Dispatch wrote that it had “a row of toggle switches, flashing lights and a generally sinister appearance” and called Hinman himself “Frankenstein Redivivus” with “a new creature.” A photograph in The Columbus Citizen-Journal on August 3, 1961, shows the delivery of a Hinman to William Charvat and Roy Harvey Pearce of Ohio State University. “Detecting differences in copies of the same edition of a published work by microscopic comparison is the job of this machine,” says the caption, not quite correctly, before boasting of its $5,000 price.

Charvat and Pearce would use that machine to edit Ohio State University Press’s Centenary Edition of the Works of Nathaniel Hawthorne, perhaps the most important collected works in the history of American literature. Bowers served as textual editor for the Hawthorne edition. Despite his background in Renaissance drama, he updated the editorial methodology of his New Bibliographical forebear W. W. Greg and adapted it to the publishing environment of the American 19th century. Put simply, to follow the Greg-Bowers method, one first identifies the copy-text, or the earliest surviving text of the work—ideally a manuscript, but often the first edition. Then, after collating every edition published within the author’s lifetime, those variants determined to be authorial are inserted into that copy-text. This procedure creates a new text that combines elements from every other version, a single “eclectic” text that includes the changes that the author wanted to see in print while discarding the rest. Of course, this process would be impossible without what Bowers calls the Hinman Collating Machine.

¤

The centenary edition of The Scarlet Letter came at an auspicious time for the study of American literature. The fortunes of American literature as an academic discipline were greatly improved by the nationalist imperatives of the early Cold War, but scholars were stuck with cheap, error-riddled paperback versions of classic texts. At a conference on “Editions of American Authors” in 1962, participants came to the grim conclusion that there existed serviceable editions of only two authors, Sidney Lanier and Emily Dickinson. The Centenary Hawthorne’s use of the Hinman collator suggested a way forward. The machine would be at the center of a mammoth effort to purify the American canon.

These discussions resulted in the founding of the CEAA, an organization housed within the Modern Language Association, the major professional association for the academic study of language and literature. The Center was headed by a committee of English professors who distributed its NEH funding to the major edition projects located in university English departments across the country. They adopted Greg-Bowers as an official methodology, recommended use of the Hinman for collation, and designed a seal of approval to be printed on those editions that met the CEAA’s standards. Editions produced under the Center’s supervision, from the Iowa-California Works of Mark Twain to the Harvard Collected Works of Ralph Waldo Emerson, are still the standard texts in American literary scholarship and are still reproduced (without the baggy scholarly appendices) by publishers like Penguin, Norton, and the Library of America.

The CEAA was a patriotic endeavor as much as a literary one. American literature professors William Gibson and Edwin Cady, in 1963, imagined the outcome of their fledgling organization in explicitly nationalistic terms:

Suppose many of the public, school, and university libraries of the United States and all the chief United States Information Service libraries abroad had on their shelves good complete editions of fifteen American literary masters. What then? We venture one prediction. American readers of these texts would know more clearly than ever before “the curious fate” of being an American. Readers abroad would come to understand more discriminatingly and to respect more justly the country that bred such men.

With its rigorous editorial methodology, its Cold War rhetoric of cultural supremacy, and the impressively scientific-looking Hinman collator, the Center was perfectly primed to become the NEH’s first major investment. The problem of funding the humanities without appearing either propagandistic or censorious was a serious one. In her 2015 book Proofs of Genius: Collected Editions from the American Revolution to the Digital Age, literary scholar Amanda Gailey remarks that “[g]overnment funds had tended to support scientific knowledge, but humanistic inquiry presented a thornier subject, because it often dealt directly with studying and interpreting ideology.” In its early years, the NEH needed “a humanist enterprise that was seemingly free of ideology but that affirmed American history and cultural accomplishment.” It may be a coincidence of history, then, that just as the CEAA was in search of funding, the NEH was in search of exactly such a large-scale, public-facing, seemingly objective literary project.

In its first annual report to the president, dated January 15, 1967, the NEH explained its decision to fund the Center using language similar to Gibson and Cady’s:

Another of the Endowment’s initial actions was to devise a program of support for the production of “pure” texts of the major American authors. The rationale for such a program seemed manifest. Emerson is the most revered non-Asiatic philosopher-writer in Asia; in Europe, Whitman is credited with freeing poetry from the shackles of verbal and technical conventions; Twain and Melville are counted among the world’s greatest novelists. That uncorrupted texts of the works of such men do not exist is a diminution of their stature and an impoverishment of the American cultural and literary heritage. Their works are monuments; yet the monuments are defaced and eroded.

The language of the cultural Cold War is here intertwined with the language of Greg-Bowers. The terms “pure” and “uncorrupted” as applied to edited texts are adapted from official CEAA statements. The emergence of American literature as a global phenomenon—Emerson in Asia, Whitman in Europe, Twain and Melville around the world—bolsters the argument for the importance of presenting these authors in authoritative editions to foreign audiences. And finally, these editions are imagined as monuments: both figurative monuments that serve to memorialize the American literary tradition and literal monuments in the form of imposing multivolume editions that will represent that tradition in international libraries.

The mutual dependence of the Center and the NEH reached its peak in 1967­–68 with that year’s grant of $300,000 (over $2.5 million today). This was also the year that the most space was dedicated to the CEAA in the NEH’s annual report. Dated January 13, 1969, the report dedicates four pages to a section entitled “Literature of the Young Republic,” which argues for the urgency of presenting accurate texts of American authors as American literature becomes more important on the world stage. “The project is in part recognition of the change in the seriousness with which American writers have come to be taken,” the report explains. “Earlier in this century British authors were considered the standard masters of the English language, even in American schools. […] But since World War II, ‘American studies’ have flourished not only in the United States but abroad.” Following this explanation of the importance of American literature, there is a photograph of an NEH staff member stacking volumes of Whitman, Hawthorne, Emerson, and Twain—a visual testament to the Center’s productivity.

Perhaps not coincidentally, 1968 was also the year that the CEAA most needed to be defended from criticism. In a controversy reminiscent of today’s debate surrounding the prerogatives of the digital humanities, historian Lewis Mumford and critic Edmund Wilson lambasted the Center for its apparent scientism and careerism. In the New York Review of Books, Mumford attacked the Harvard edition of Emerson’s Journals and Miscellaneous Notebooks, calling it emblematic of “the preconceptions and the mock-scientific assumptions governing the pursuit of the humanities today.” For Mumford, the failure of the Journals was a sign of the “technological extravagance and human destitution” that constitutes “the fashionable mode of our day.”

Wilson wrote a letter to the editor approving of Mumford’s criticism and then took up the battle flag himself with his own two-part essay in the NYRB. For him, this was a personal vendetta, as his own proposed series of reprints of American classics was superseded by the CEAA’s more technoscientific approach. Wilson, too, understood the role that literature could play in the Cold War, writing in a letter addressed to President Kennedy, W. H. Auden, Robert Lowell, Perry Miller, John Crowe Ransom, Allen Tate, Lionel Trilling, and others that “[i]f we can squander billions of dollars on space rockets, nuclear weapons and subsidies to backward countries, why should not the United States government do something to make American literature available?” His proposal was met with approbation by its recipients, and he “was told later that a substantial sum of money had been set aside for this purpose by the National Humanities Endowment,” as he tells it in his NYRB piece. However, the Center’s own proposal, with its detailed editorial methodology and its promise to “purify” the American canon using the imposing Hinman collator, was more attractive. “The next thing I heard about it was that this money had somehow been whisked away, and my project ‘tabled,’” he complained. “The Modern Language Association had a project of its own for reprinting the American classics and had apparently had ours suppressed.”

Greg-Bowers intentionalism and the Hinman collator were therefore particular targets for Wilson’s wrath. He calls Bowers the “great Demiurge behind all this editing” and opines that he has “found no reason to believe that he is otherwise much interested in literature.” Regarding Northwestern University Press’s Writings of Herman Melville, Wilson writes:

[T]he project in the case of Typee has been so relentlessly carried out in the technical language of this species of scholarship—of “substantives,” “accidentals,” and “copy-texts”—that a glossary should be provided for readers who are not registered union members—if there are any such readers—of the Modern Language Association.

Finally, he criticizes the edition’s “Historical Note,” somewhat confusingly bemoaning its authorship by literature professor Leon Howard instead of “an anthropologist who knows Polynesia.” But, he laments, he shouldn’t “expect that the MLA would care to humiliate its Hinman Collating Machine by associating it with a raw anthropologist.”

Mumford’s and Wilson’s articles inspired furious letters from proponents and detractors alike, including one from William H. Y. Hackett Jr., the vice president of publishing company Bobbs-Merrill. The Center’s editors, in Hackett’s view,

hold that passing a public domain text such as The Scarlet Letter through the Hinman Collating Machine gives them copyright protection; they choose to forget their action restricts the free use of a national treasure that is our common property. […] Is the Hinman Collating Machine alive? Does it breath? [sic] Is it an author?

For Wilson and Hackett, the Hinman collator had become a symbol of the “technologically extravagant” humanities, an author-machine who could be humiliated and who thus must be protected.

Perhaps most importantly, the Hinman represented a robotic replacement that threatened both Wilson’s job as a literary critic and editor and Hackett’s job as a publisher of American literature. As the controversy spilled out of the pages of the NYRB and began to receive coverage elsewhere, the CEAA felt it had to respond in an official statement. Gordon Ray, Thackeray scholar and head of the Guggenheim Memorial Foundation, retorted in the Center’s pamphlet Professional Standards and American Editions: A Response to Edmund Wilson that “this attack derives in part from the alarm of amateurs at seeing rigorous professional standards applied to a subject in which they have a vested interest.” Of course, this was exactly the issue for Mumford and Wilson: professional standards were being set by the CEAA, but without graduate training in scholarly editing and access to a Hinman collator, those standards could not be met. Editors, publishers, professors, and critics looking to publish new scholarly editions faced the decision to follow Greg-Bowers or go without funding.

The CEAA faced criticism even from its own editors. Greg-Bowers is a slow methodology that requires meticulous collation, emendation, and proofreading, but the NEH funding schedule pressured editors to speed up their work. Director Matthew Bruccoli “was very good as an administrator in NEH’s terms and in the MLA’s terms, because he was relentless about requiring updates and progress reports on volumes,” David Nordloh, an editor for Indiana University Press’s A Selected Edition of William Dean Howells who has published articles critical of the Center, explained to me over the phone. “Our deadline date was at the end of the summer and by early summer we were getting constant notes and telephone calls from Bruccoli saying, ‘What’s the stage of this volume? We’ve got to get some more things approved and under seal’—as he called it—‘in order to make a good enough case to the NEH for the next stage of funding.’ And the NEH was mostly coordinating its support for our edition and the others through the CEAA.”

The NEH’s support of the CEAA expired on August 31, 1976. Beginning on September 1, the organization was reconstituted as the Committee on Scholarly Editions, which continues today. The CSE differs from the Center in two important respects: it does not receive or allocate NEH funds, and it grants its seal of approval to editions of any country of origin and in any language, not only those of American authors. Rather than solving Wilson’s problem, however, the CSE made it more intractable. Aspiring editors would have to petition the NEH directly for grants relating to scholarly editions, but the NEH looked to the CSE for guidance. The CSE, of course, recommended Greg-Bowers. The method’s insistence on authorial intention and its singular focus on purity and corruption spread from the Center to the NEH itself and began to affect not only American authors of the 19th century but also authors around the world and from any era.

Despite this widened scope, CSE-approved editions continued to be mainly reserved for white male authors. Though it was presented as an objective, universal method, Greg-Bowers is best suited to authors with large bodies of work featuring multiple print runs across many editions. Editors who wished to republish the work of a more obscure author were left with little recourse. W. Speed Hill, a chairman of the CSE and a panelist for the NEH, explained in 1991 that proposed editions of neglected women writers were often rejected because their authors were, at least by that time,

uncanonical, or because the applicants were insufficiently skilled in the preparation of complex, institutionally-based grant applications, or because the applicants were frankly not interested in the niceties of copy-text, historical collations, emendations, substantives and accidentals, and all the other minutiae [that Greg-Bowers involves].

Scholarly editions of minority or women authors meant to sit alongside editions of Hawthorne, Emerson, and Melville were to be deferred until the author in question had achieved canonical status outside of the editorial establishment. Even then, Greg-Bowers would be necessary, so that the editor must be “interested” in its “niceties”: copy-text, emendations, substantives and accidentals, and collation with the Hinman collator. A machine used to promulgate the American canon of the early Cold War was being used to maintain that canon even as the Cold War drew to a close.

¤

Mumford and Wilson’s critique of the humanities as voiced through their attacks on the CEAA—that the field had become too specialized, too insular, too invested in methodology, and too uninterested in the intellectual labor of interpretation and criticism—anticipates more recent concerns raised about the digital humanities’ “redefinition of technical expertise as a form (indeed, the superior form) of humanist knowledge.” Critics of the digital humanities might see Mumford as eerily prescient for his claim, made in 1968, that American literary scholarship was “on the eve of its surrender to the computer and to those limited problems that computers so deftly and swiftly handle.” Though the Hinman collator is an analog machine, Mumford foresaw the data-driven approaches to literature that it enabled and the contention within humanities disciplines that these approaches would cause.

In 1996, literary scholar and editor Charles Ross made a series of predictions about the effects that digital editing would have on the discipline of editorial theory. First, he said, it would kill Greg-Bowers. As with so many who try to predict the future, Ross was wrong; digital editing has not killed Greg-Bowers, which continues to be applied to large-scale critical editions. In fact, many of the problems of funding that were introduced with the Hinman collator and the Center have persisted into the era of digital editions. Because of institutional constraints and funding requirements, the new digital canon is in danger of becoming as conservative as the print-based canon of the 1960s.

As was the case with edition grants in the CEAA and CSE eras, funding for large-scale digital archives is tied to the already established reputations of authors. And, as one of the most remarkable canon-building exercises of the 20th century, the Center itself has been a boon to the reputations of a select group of white male authors. Ken Price, an editor of the digital Walt Whitman Archive, writes that “past decisions often become reinscribed in the present: projects funded by government agencies to produce print editions in previous decades solidified their academic standing in ways that leave them in advantageous positions in a new electronic environment.” He remarks that the WWA directly benefited from the Center-endorsed Collected Writings of Walt Whitman in its grant applications, as they presented it as proof of “Whitman’s firm canonicity.”

In her 2015 book Traces of the Old, Uses of the New: The Emergence of Digital Literary Studies, literary scholar Amy Earhart observes that, quite simply,

NEH grants, which fund a majority of the digital literary projects, are often judged by impact and impact is most recognized by numbers of hits to the site. We know that canonical writers have a greater chance of a large following than little known writers. So, it follows that a good number of the archive and edition projects are focused on canonical writers, such as Rossetti, Melville, or Whitman.

This in effect replicates the problematic tendencies that began with the Center’s Cold War project of canon formation. Of course, the Hinman collator is not solely responsible for this effect; rather, it arises through the collective inertia of an alphabet soup of institutions: NEH, MLA, CEAA, and CSE. But as the catalyst of the particular technoscientific approach to the humanities that informed the NEH’s earliest days, the Hinman is key to understanding our current moment.

The days of the Hinman are waning. It now has to compete with other collation machines like the Lindstrand Comparator and the McLeod Portable Collator, not to mention digital collation programs. Divorced from battles over government grants and debates over the future of the humanities, the machine now looks more like a curiosity than a threat, but it is still in use today for both research and pedagogy. Over the phone, David Nordloh argued that, “in certain ways, the Hinman, as clunky and expensive as it was, […] remains the ideal mechanism for doing text collation.”

University of Virginia professor David Vander Meulen frequently uses a Hinman for his research in early books as well as in his teaching—so much so that he has his own Hinman in his garage. The Hinman “has sometimes been viewed as a totem of the dehumanization of scholarship,” he writes via email. “But I encourage students to think of it as a tool, like a pair of eyeglasses, that helps them to see things they might otherwise overlook.” Though the Hinman helps to generate data, it is up to his students to interpret that data. “Their work is fundamentally humanistic, for by it they are excavating a record of the author at work and are bringing to light the process of artistic creation,” he says.

There are some tasks for which the Hinman is better suited than digital collation. Nicole Gray, project specialist at the University of Nebraska–Lincoln Libraries and editor of the variorum edition of the 1855 Leaves of Grass, used a Hinman collator to identify “spatial variants,” or “differences in spacing that didn’t affect the text but that provided evidence about how the books were printed,” as she explains via email. Such differences cannot be detected by text-based collation programs but can be spotted with physical comparison aided by a Hinman. “My experience with the Hinman for the Whitman project makes me wonder if such machines might not continue to make us see and experience books in ways that computational methods sometimes obscure,” Gray says. “When you run a collation script, you get output, possibly combined with the kinds of visualization that have been built by developers to showcase the output. But the Hinman is an example of a kind of engagement between human and machine that requires both to participate.”

Perhaps the future of the Hinman lies in just this type of encounter, a close physical engagement with printed material that is all too rare in our current digital environment. But we can also use the Hinman to better understand our digital environment itself—the structures of funding and the problems of representation and canonization that this strange machine inaugurated.

¤

Matthew Blackwell is a writer based in Tenerife, Spain. He is currently working on a book manuscript, The Uncorrupted Canon: The Postwar Rise of American Scholarly Editing.

¤

Featured image: Horatio C. Forjohn. Idle Governor, 1940. Smithsonian American Art Museum, Transfer from General Services Administration. www.si.edu, CC0. Accessed April 26, 2022.