Archives-Based Digital Projects in Early America
The use of quantitative methods to analyze the past in history departments coincided with the introduction of machine-readable cataloging to replace the card catalogs in libraries in the middle of the twentieth century. The more recent application of computational tools and methods to humanities scholarship relies on a similar return to the archives, so that scholars can apply new methodologies to old records and deduce and understand anew. Recent digital humanities work productively challenges the professional specialization that separated scholars working in libraries from those working in academia. The scholarship of historian Charles McLean Andrews and bibliographer Jacob Blanck, conducted before these professional divides were solidified, is in many ways comparable to digital humanities’ reliance on critical classification systems and on inclusive models of scholarly production. In a world of collaboration, transparency of process reigns supreme as methodology has become a subject of scholarship. This article, based on the 2018 WMQ-EMSI workshop, “Archives-Based Digital Projects in Early America,” offers a state-of-the-field report on how scholars working in academia and in libraries are remediating the early American historical record through digital tools and methods.
This labour of classification was both Herculean and depressing: it could last months or more, without the slightest glimmer of an intellectual discovery on the horizon. . . . A number of researchers in Cambridge and Paris are now working out programmes so that the computer can take care of this thankless preliminary phase, from the initial data-collection to the reconstitution and statistical analysis of the family files. The historian will then have virtually nothing to do but apply thought: which should after all be his or her proper task.—Emmanuel Le Roy Ladurie, The Territory of the Historian1
In what has become a sort of manifesto for history's quantitative turn in the middle of the last century, Emmanuel Le Roy Ladurie's The Territory of the Historian figures the work done by computers as separate from that done by historians. The computer is, Edmund A. Bowles tells us in a [End Page 451] similar vein, "an instruction-following machine."2 In these formulations, the historian and the computer are separated, operating in two distinct spheres. Minutiae and drudgery characterize the "thankless" "labour of classification," which Le Roy Ladurie specifies as "initial data-collection," "reconstitution," and finally "statistical analysis." This computation enables scholarship, which Le Roy Ladurie contrasts to the "labour" of the computer. Historians wait for the computer to do its work, so that they can "apply thought" and thereby perform their "proper task."
At the same time as Le Roy Ladurie and the quantitative turn took hold in Continental and then U.S. history departments, scholars based in libraries framed their work in similar terms as the potential of the "instruction-following machine" began to be explored. A small team of catalogers and programmers at the Library of Congress conceived of MARC, or Machine-Readable Cataloging, in the mid-1960s.3 The name most associated with its creation and perhaps most notably its implementation is Henriette D. Avram, who understood MARC as a way to "relieve humans of the drudgery of preparing the catalogs, make the information more rapidly available, and provide the facility for additional access points to the data."4 The Library of Congress completed the pilot project in 1968, and beginning in the early 1970s, with the support of the American Library Association, they sent MARC ambassadors to libraries throughout the United States and increasingly abroad to train catalogers in its use and to consider with them how their card catalogs might be transformed into the new, computerized system. This process of transforming the catalog from one based in index cards to one that lived on a computer became known as retrospective conversion, or RECON. Though critics lamented the losses of handwritten notes on such index card, as well as inevitable mistakes that crept in during the RECON process, MARC took hold, and its numerical [End Page 452] fielding system remains the dominant mode of both creating and sharing bibliographic data in and across cataloging systems.5
Those who work in the historic bibliographic record—or "rare book catalog[ers]," as the profession has dubbed them—were not the first to adopt MARC, but eventually, if a bit begrudgingly, they too came to rely on computers to aid them. Field leaders Terry Belanger and Stephen Paul Davis perhaps best summed up their colleagues' sentiments when they wrote in 1980: "The rare book community . . . prefers jumping on the computer bandwagon to being run over by it."6 Central to this leap was the English Short Title Catalog (ESTC), which got off the ground at the British Library in the late 1970s. Rare book catalogers and librarians, whose practices grew out of nineteenth-century bibliography, had long fixated on how to describe materials that more often than not defied contemporary classification terminology. They did not wholeheartedly celebrate the release from drudgery that Avram promised. In describing the ESTC's use of MARC to create a union catalog of all imprints published in English before 1800, Earl R. Taylor of the Boston Public Library gave poetic voice to the fear that humans might no longer be necessary to create bibliographic data. Historians and bibliographers are scared, Taylor explained, because "no one knows what shapes shadows may take in the dark."7 Rare book catalogers [End Page 453] and bibliographers might then, in the words of eminent Jonathan Swift bibliographer William B. Todd, end up being "mere mortals, [who] stand mute before" computers.8 True to the eighteenth century that he knew and loved so much, Todd's use of hyperbole was somewhat satirical, and yet he gestured at the question underlying his colleagues' fear: what is the place of human agency in the organization of the historical record?
Le Roy Ladurie's assurances about the historian's "proper task," Taylor's fear of the unknown, and Todd's concerns regarding powerlessness all register today as rather quaint. Their configurations of the relationship between scholar and computer, what is now more commonly referred to as "digital," came and went, as intellectual trends tend to do. In their wake, scholars have not been left idle with "virtually nothing left to do," nor have they been left "mute." And yet the distance between past fears and contemporary scholars' lived reality notwithstanding, current archives-based digital projects are indebted to the moment when computation first arrived on the scene of preserving and animating history. Both that time and our own promise a return to the archives so that scholars can apply new methodologies to old records and look at, deduce, and understand anew. In both, methodology becomes the subject of scholarship. Even though digital scholars might not now recognize the separation between the work of the computer and that of the historian, the current remediation of the historical record in digital environments invites us to once again pause and consider how our relationship to both the computer and the historical record defines our projects as well as our professional trajectories. The archive, far from being computerized in the way Le Roy Ladurie imagined, follows a parallel and sometimes antagonistic track to the development of digital methods in historiography. Indeed, the archive, as a sight of competing classifications, offers an opportunity for us to reflect on what historians do that machines do not; what archivists do that machines do not; and most importantly for my purposes here, whether the professional distinctions between what historians do and what archivists do still hold.
After all, early formulations that divide human labor and machine labor are in some ways barely recognizable to the contemporary scholar using digital tools to remediate the historical record. We work with, not before or after, computers; or more precisely, an understanding of what computers can and cannot do intimately informs our scholarship. We have adjusted our labor to the strictures of what computers can do, and that adjustment is where our toil now resides. Moreover, scholars create and critique classification systems because in our computer-aided efforts to recover history's forgotten voices and to listen to archival silences we are endlessly reminded [End Page 454] that classification has consequences.9 Or, as Safiya Umoja Noble, who writes at the intersection of Information Studies and African American Studies, has recently argued in Algorithms of Oppression, "those who have the power to design systems—classification or technical—hold the ability to prioritize hierarchical schemes that privilege certain types of information over others."10 We might acknowledge, like Le Roy Ladurie, that classification can be "both Herculean and depressing" and that it brings little reward, but the idea that something outside the human that interacts with the historical record in a meaningful way can be trusted and can remediate it impartially no longer holds.11 In our present moment, this shift in the way historians understand how to work with computers necessitates and indeed causes an evolving understanding of scholarship.
The change that is underway is perhaps best summarized as an attention to positionality, both to instruments of use and to the nature of our work. We are not only refiguring what we do and how we do it but also, and perhaps most importantly, why. The parallels between the rhetorics in the academy and the library that I began with are not coincidental; instead, they illuminate how twentieth-century professional constructions created an artificial distinction between those working on either side of the reference desk. Roy Rosenzweig traced the divide between the historian and the archivist to the 1936 decision by the Conference of Archivists to break from the American Historical Association (AHA) and become the Society of American Archivists. This break reflected a schism that had been underway since the first few decades of the twentieth century. Where there had been one profession, there were now two: one "charged with writing about the past" and the other with "preserving the records of the past."12 The historian who organized, cataloged, and recorded became largely extinct in the fields focused on either early modern history or the more modern era, exactly the subjects that dominated U.S. history departments. Within such a department during the remaining decades of the twentieth century, the work that someone like the formidable Charles McLean Andrews did to index in incredible detail the Colonial Office archives at the Public Record Office (PRO) in London was rarely recognized as academic scholarship. And yet, [End Page 455] in the twenty-first century, his efforts seem powerfully familiar. A return to Andrews's understanding of his labor offers a glimpse into a professional relationship with the archive that, as Rosenzweig bemoans, was lost in 1936.
After his first visit to the PRO at the behest of the Pennsylvania Historical Society in 1893, Andrews spent the next two decades—the height of his career—intermittently compiling the first guide to these manuscript materials. A student of Herbert Baxter Adams's at Johns Hopkins University, Andrews was part of the wave of self-consciously professional historians who sought, according to Alfred F. Young and Gregory H. Nobles, a "new scientific history—[through] mastery of the documents."13 Soon after his initial visit, Andrews completed his first monograph, The Historical Development of Modern Europe, but the archive he encountered in London redirected his scholarly trajectory entirely; upon his return to England a decade later, Andrews dedicated himself not only to the study of early American history but also to making that history accessible to scholars in his wake. He considered a guide to the PRO to be essential; without it, "a complete history of our colonies will not be written," he told the AHA in his 1898 address. In the place of the "loose generalizations" that Andrews saw afflicting early American history, he aimed to uncover "the exact details, not of the laws on the statute books, but of the actual working of the system in the field, [so] that reliable conclusions may be made possible."14
In an essay found among his papers and published posthumously in 1944 by the William and Mary Quarterly, Andrews advocated for an archival turn that would produce a more impartial, less bombastic historical narrative by returning to the historical record itself. This methodological shift is at once about homing in, as the tedium of archival work necessitates, and expanding out, a move beyond the parochial. He described his "desire to discover, if possible, the place of our colonial history in the larger history of the world of its time; to treat all subjects with regard to their proper proportions and their relative importance in the general scheme; to authenticate and interpret the course of events only as far as the evidence warranted; to preserve balance and perspective; and to follow to the end of the period the operation of the many factors involved." He continued, "Because for one hundred and seventy-five years the colonies were colonies and not independent states, they must be treated as such; otherwise the true meaning of our early beginnings is perverted and lost."15 The PRO appealed to [End Page 456] Andrews not only because it was a treasure trove of the evidence he sought but also because it allowed for a colonial history that could be told without teleological and nationalist narratives that distorted, in Andrews formulation, our understanding of "the true meaning of our early beginnings." This early, formative transatlantic moment in historiography is marked by a return to the archives to perform a task that looked more like the work of an archivist than of a historian. In his study of the "concept and contours" of Atlantic history, Bernard Bailyn gives Andrews as one of his examples of "immensely creative archival scholars" and credits him with the "discover[y]" of "the Anglo-American archives" through the diligent way he "catalogued them, indexed them, and put them to use."16 Karin Wulf likewise identifies Andrews as part of the origin story of what she has come to call #VastEarlyAmerica. Wulf describes his vision as "vast indeed," adding that it "looks particularly interesting from our vantage so many decades later."17 This "interest" lies in Andrews's more capacious understanding of both what early American history includes and what a historian's job entails. And in both areas, Andrews's immediate successors did not follow his lead.
Andrews's vision not only sent him to the archive but also drove him to pass a decade in the middle of his career listing and compiling, noting and assessing, indexing, and arranging. These are activities most associated with archivists, in contrast to the writing and publishing most associated with historians, to recall Rosenzweig's explanation of the post-1936 divide. Andrews worked at a time before the professional distinction between historian and archivist, academic and librarian had been institutionally [End Page 457] solidified. Our present modes and methods of scholarly production are fruitfully challenging these professional distinctions and revealing them to be twentieth-century constructions, not reflections of the work that archives require of us. More than a century ago, professionalization began placing different modes of inquiry and meaning making into disciplinary buckets that are now proving porous, if their bottoms have not dropped out altogether.
The two-day workshop "Archives-Based Digital Projects in Early America," hosted by the Early Modern Studies Institute (EMSI) and the William and Mary Quarterly (WMQ), brought together scholars who share a focus on early American archives and their remediation but whose professional trajectories differ. Each scholar presented a digital project and, to prepare for the seminar, produced a white paper that reflected on the origins, trajectories, and methodologies of their work making digital projects from early American archives of various sizes, materials, and scopes. Previous WMQ-EMSI workshops focused on the scholarship of midcareer academic historians embarking on second or subsequent book projects. This year, however, a combination of scholars employed in academia and those working in libraries and university centers, as well as, at times, independently, all at various stages in their careers, reflected the larger professional transformation I am tracing here.
As the convener of this event, I sought out scholars with a heterogeneity of experience in navigating the intersections between archival and digital work so as to surface the lived experiences of those with different levels of institutional, grant, or other kinds of professional support. Not only were the participants at various stages in their careers, but the projects they represented also exist at different points along trajectories of progress, completion, and support. For example, the Reading the First Books project is in one sense complete: its grant funding has ended, it has published articles related to its research and findings, and the Internet Archive Wayback Machine preserves the project's website. Yet because of the many technical obstacles Reading the First Books encountered, it only transcribed fifty of the three hundred books printed prior to 1601 in the Americas that its organizers had planned for in their grant application. The Early Novels Database (END) is also not complete, but it is not over in the same way that Reading the First Books is. Like Reading the First Books, END has been inactive for two years, but it remains a work in progress. Though it has received considerable institutional support from the University of Pennsylvania and Swarthmore College, it currently lacks funding for further growth and therefore remains a static site that makes clean, well-formed data available but has no interface in which to query or otherwise interact with the significantly enhanced metadata it created to describe the 2,002 [End Page 458] novels in its database. The George Washington Financial Papers Project has not yet completed its data entry, but it developed its interface in conjunction with the creation of its data, so its progress looks different from that of END. The George Washington Financial Papers Project has completed the goals outlined in its initial grant: developing a working system, transcribing and fielding data for three major ledger books, and adding an additional Gouverneur Morris account book. But the project continues, as students are currently transcribing two accounts books and entering them into the system.18
At the other end of the spectrum of completion, the Jesuit Plantation Project is in its nascent stages and promises to continue because its director's institutional home recognizes the project as scholarship. Its interface is also a work in progress, and its data have yet to be completed, though with a finite number of people to record, the end is in sight. In contrast, it is hard to imagine that the Yale Indian Papers Project, which has been ongoing since 2002, will ever be complete, as its goal is to publish nearly four centuries' worth of primary source documents written by, for, or about the Indigenous people of the American Northeast. Fortunately, it has sound (if basic) institutional support at Yale, but in addition to being moved from one center to the next at Yale, it must constantly vie for grant funding to incorporate new materials and to improve its technological base. Like the Yale Indian Papers Project, the Georgian Papers Programme (GPP) is grounded in institutions with vested interests in its success, and its work continues apace. Of course, questions around platforms and interfaces continue to vex it, but active research fellows regularly report on their findings and the project's progress on social media and the project's blog. Unlike any of the other projects, the Pobladores Project grows out of two completed digital history initiatives, the Early California Population Project, which traces the baptism, marriage, and burial of 14,000 settlers, and the Early California Cultural Atlas. The data entry needed to trace settlers' origins to the 136 Mexican towns and 78 global locations from which they came is not yet finished, and so the project to date has no public interface.19 [End Page 459]
The workshop's participants therefore operate on different scales of support, with different metrics of progress, and as we regularly returned to in our conversations, within different systems for professional advancement. Given that diversity of settings, what defines success is up for grabs in ways that it is not when monograph publication and the resulting professional status are the end goal. In the hybrid field that is part traditional humanities, part library science, and part digital innovation, fixed metrics of success and failure no longer hold. Our two-day discussion touched on the frustration produced by that situation, as the group repeatedly addressed how existing institutional and grant-funding structures value and, at times, devalue or fail to recognize our work.
The white papers, as well as the two days of discussion around them, consistently revealed that, though we may look backward when we devote ourselves to digitally based archival projects, the current reparative archival turn propels us forward. In academic discourse, historians of indigenous peoples have long been attuned to erasure in the archives, as demonstrated by Jean M. O'Brien's cautionary tale of southern New England nineteenth-century local histories that, to varying degrees, make what she terms the "extinction claim."20 Likewise, in the context of African American history, Marisa J. Fuentes has delved into not only what this erasure—this silence—means for the creation of history but also what it means for the researcher. She asks those of us who work with the historical record to consider "our own desires . . . to recover what might never be recoverable and to allow for uncertainty, unresolvable narratives, and contradictions" so as to become attuned to "an ethics of history and the consequences of reproducing indifference to violence against and the silencing of black lives."21 In the practices of library science, efforts to redress the racist library classification systems that lumped literature by, about, and for black people in all places and in all times under one category have been underway since Dorothy Porter's groundbreaking work at Howard University in the middle of the twentieth century. Laura E. Helton has recently shown that Melville Dewey's proprietary regimes coupled with the gendering and therefore depreciation of women's library work left much of Porter's efforts unnoticed. Helton adeptly demonstrated that Porter's "task of ordering a [End Page 460] library enacted critical claims about the place of blackness in systems of knowledge," transforming the library catalog from "a site of retrieval" to the producer of "a new black imaginary."22 Contemporary library scientists such as Michelle Caswell and T-Kay Sangwand continue in the legacy of Porter, calling attention to and innovatively circumventing the "symbolic annihilation" found in inherited information classification systems.23
Those convened at the Huntington work in an environment that has been created by the legacy of these scholars in academia and those in libraries. The critical lenses of race, labor, and gender studies guided us from the start as we decided which archives merit our attention and why; how different forms of representation will function computationally and how they will make meaning; how to negotiate our work, both as individuals and in collaboration; how our accomplishments will be legible (or not) as professional achievement; how the projects will be funded and—all too often a completely different question—how they will be sustained; and how the data we create or re-create will interact with other data sets and how that interaction makes meaning. The list of considerations is long, and yet the point is simple: trained humanists have to consider not just what and how to do things with archives, but why we do them. Scholarly inquiry and intellectual zeal draw us to certain archives, and Fuentes's "ethics of history" [End Page 461] surface constantly in the questions about data designation and design, interoperability, and interface that digital scholarship poses.
The group assembled at the workshop testified to the erosion of professional strictures whose distinctions are increasingly arbitrary and whose scaffolding no longer supports what we do. Less a rupture than a return, what we embodied, in the spirit of Rosenzweig's understanding of the profession, is a reanimation of a nineteenth-century conception of what scholarship entailed, when the work of the historian and that of the archivist had not yet been made distinct. The habits of mind, the intellectual impetuses for scholarly work have not changed. What initially propelled many of these projects are time-honored and tenure-case-tested "They Say / I Say" academic moves.24 But the ways in which we rebuke claims or fill in gaps have changed. For example, in her Jesuit Plantation Project, Sharon Leon investigates the lives and experiences of 275 enslaved people who resided on Jesuit-owned estates in southern Maryland until 1838, when Father Thomas Mulledy sold them to Louisiana. The scholarly attention on this sale and on Jesuit slaveholding more generally, Leon tells us, has "focused on the implications for the religious community and the moral universe in which these men made their decisions about slavery."25 In response, Leon has returned to the archives, and since 2016, she has immersed herself in the records of this sale at the Maryland Province Archives and a number of other related collections in search of clues about the enslaved community.26 Her rejection of previous scholars' claims that we need to do more to understand the enslavers rather than the enslaved takes the form not of a polemic in a journal article or a monograph but instead of a carefully assembled and linked data set published in Omeka S.
In a similar move, Steven Hackel responds to the lacuna of scholarship on the early Hispanic families of California. According to him, when social historians acknowledge them at all, they offer little more than "disparaging caricatures of early Californians as fandango-loving hedonists inhabiting dusty pueblos and decadent ranchos."27 To better understand these people and their lives, Hackel has set out to amass a large set of data and to create the Pobladores Project Database, which maps the movement of the "thousands of settlers [who] came from Mexico to colonial California . . . nearly all [of whom] have been forgotten, eventual losers in the contest for [End Page 462] California and in the subsequent battle for a place in American history."28 Leon, Hackel, and the others in the workshop have gone to the archives, and instead of synthesizing and publishing an article or a monograph on what they have read, observed, compiled, and come to know, they amassed data in sophisticated and complex ways, and then, with commensurate thoughtfulness, shared that data. The labor, the doing of scholarly work, has therefore changed, but fortunately for those of us committed to a life of the mind, the desires that bring us to scholarship, the habitus that defines it and to some extent defines us, has not. And the participants in the workshop were fortunate to have two days to reflect on our practices and to learn from each other.
In the course of our conversation, the participants continuously returned to questions about standardization. We all related moments in our projects in which we paused to ask: Is what I am doing sui generis? Should I be calling upon an established data model or vocabulary? If I use a standard model or vocabulary, will I inadvertently reproduce its inherent biases? Perhaps my project should create a new model or format? If I create one, how can I make it discernible for end users as well as potentially useful for future projects? Can I ensure that others will use it? In many cases, in our remediation of collections from analog to digital, from paper to screen, we reenact Nicholas Popper's vivid description of early modern work in archives as moving along "a familiar arc from frenetic compilation to taxonomic classification."29 It is through "taxonomic classification" of vocabularies and data formats that standards are formed and our work becomes interoperable, rendering it legible both to people and to machines. Workshop participants are all acutely aware that those classification systems matter as much as the data used to build them, that our decisions about when to adopt and when to innovate, or how to find the balance between those two possibilities, will—in many ways—determine our outcomes. The larger the project and more complex the data, the more challenging these considerations become.
Take, for example, the GPP, a ten-year interdisciplinary project to digitize, conserve, catalog, transcribe, interpret, and disseminate 425,000 pages in Britain's Royal Archives and Royal Library relating to the Georgian period, 1714–1837.30 As James Fisher, metadata assistant for the GPP, recently reflected on compiling lists of subject headings for indexing the Georgian Papers, "This is not nearly as straightforward as it might sound. [End Page 463] It requires a detailed knowledge of the papers themselves, a broad awareness of the trends in eighteenth-century scholarship, and a sense of how to mediate between them."31 Mediation is key, and examples abounded in our two-day conversation to illustrate that this is now "the proper task"—or at least the one we find ourselves constantly doing—when we interact with the historical record. Thus, Lindsay Van Tine explained that END aimed to generate high-quality metadata about novels published between 1660 and 1850 so as to reveal, as the project webpage states, "how early novels instruct readers about themselves." END's staff chose to expand on a data model that has been the bibliographic lingua franca for half a century, the aforementioned MARC.32 The team selected this format for sound reasons: the additional metadata that END seeks to add to bibliographic records could be built upon MARC's existing bibliographic data and so there was no need to start from scratch. In fact, doing so would be an imprudent use of time and resources. The data added to the MARC records, created in the same format, could then potentially be re-ingested into an online public access catalog (OPAC) or another front-end system that serves up MARC data, such as HathiTrust. If END did not use MARC, its work would be forever siloed, unable to communicate technologically with existing bibliographic data.
And yet, as is inevitable when using a data model for a purpose other than that for which catalogers initially created it, END project leaders had to identify and correct unintended consequences. In fact, the very premise of END came under scrutiny from its inception because its conceivers predicated the project on a genre classification—the novel—that is contested at best and unstable at worst. Genre categories are troubled, for they exist on a continuum, are historically and materially contingent, and yet determine expectations and interpretations in many ways. Given all of that, it is not a surprise that literary critics have long grappled with what we mean when we talk about genre. Northrop Frye obsessed over the taxonomies of literature; E. D. Hirsch saw verbal meaning as dependent upon it, as "necessarily genre-bound."33 Though literary critics have wrestled with the theoretical implications of genre, library scientists began seeing it as a necessary component in their own work. At a 1979 meeting to discuss the need to identify genre in rare book catalog records, the Independent Research [End Page 464] Libraries Association recommended that the Association of College and Research Libraries produce the handbook Genre Terms: A Thesaurus for Use in Rare Book and Special Collections Cataloguing.34 After its creation, rare book catalogers had a standard vocabulary to describe genre, but this does not mean that they would or could apply that vocabulary consistently and uniformly. The messiness inherent in that process, combined with the difficulty of meshing genre terms with the Library of Congress subject headings that were integral to MARC, meant that identifying the "novels" on which END depends required thoughtful consideration.35 Catalogers and bibliographers have not uniformly applied the category of "novel" that the project espoused. END's creators, therefore, found that the project's challenges transcended merely the construction of technological crosswalks from one data format or platform to another. They had to scrutinize the data itself and be attentive to past catalogers' interpretation of the "novel."
Leon, too, discusses the need to scrutinize her data sources, to not accept that the archives leave no trace of the enslaved people's lives she is trying to uncover. Though Leon's sources—account ledgers rather than END's cataloging records—are analog, not digital, she too has to read carefully and knowledgeably to extract "every instance of an event involving an enslaved person."36 Even still, Leon reflects in a recent blog post, "the data set can never stand on its own without the support of additional interpretative framing."37 In another example of eternal vigilance to the consequences of classification, Erica Cavanaugh explains how the George Washington Financial Papers Project, which makes Washington's business and household accounts freely accessible in a digital edition, also had to navigate decisions about labeling its data, choosing data-generated, sui generis vocabularies in some instances and existing, established vocabularies in others.38 Given the goal of this project to be useful in secondary schools, the project team made decisions with pedagogical purposes in mind. As Cavanaugh explains, "This editorial decision was made in order to allow users to know what terms were used by Washington and his associates while also providing a list of words that could eventually be used when searching the transactions. [End Page 465] The only time this approach was not followed was on the subject of donkeys and mules. In an effort to prevent the distraction of younger users, it was decided not to include jacks or ass in the title."39 The work we do as standardizers, which becomes especially salient in linked open data's push for Unique Resource Identifiers (URIs), means that we are forever dealing with hypothetical scenarios. This kind of postulating mirrors the writer's consideration of audience: Which journal should I pitch this article for? Which scholars' work do I cite? Will the audience I imagine understand my verbiage, syntax, and references? But in creating the sorts of projects discussed at the workshop, we are no longer only writing essays to give context and contours to the archive; instead, we are refiguring and reconceptualizing that archive. We are acting upon our sources and hoping to make them newly accessible.
And if the output of our work has changed, so has the nature of the work itself. The workshop participants acknowledged the deeply collaborative nature of the projects. They presented themselves as synecdoches for a community of catalogers, transcribers, professors, technology specialists, managers, directors, principal investigators, and others—the list goes on—who all contribute to our projects. There are, of course, scales of collaboration on which digital scholars work that are often shifting, but we become uneasy when the romantic notions of authorship and the ownership inherited from traditional scholarship attribute accomplishments to a single person. This departure begs several questions: Are our reflexive relationships with the archive dependent on romantic constructions of authorship that the digital, with its turn to collaboration and interdependence, gradually but effectively makes obsolete? Can we imagine a communal, affective relationship to the past that still compels us toward the archive, or does our inherited lineage of practice inevitably preclude the collaborative, iterative models of production on which digital scholarship depends? In other words, is the return to the archive that the profession inherited from the nineteenth-century legacy of antiquarians such as Walter Scott, Hannah Mather Crocker, and Washington Irving (just to name a few) still figured as a "rational reflex"?40 Or, with our renewed focus on methodology and [End Page 466] training, do scholars now rely on "learned behavior" to know how to interact with the archive?41
And what is the fate of these "projects" that are all too often devoid of the comforting teleological nature of publishing in the relatively fixed medium of print? As Popper observed in the course of the workshop's conversations, a print monograph or article, by its very definition, has an end, whereas a digital project does not necessarily have one.42 Does the work we do—which entails administering to our duties, navigating the ethics of collaboration, and seeking grant funding—satisfy our intellectual curiosity and scholarly desires as well as lead to professional reward? Perhaps the Georgian Papers Programme knows such potential hurdles best. Arthur Burns describes the nature of the collaboration as "unusually ambitious," efforts that extend across institutional and national boundaries, including the British Royal Household, whom Burns diplomatically describes as having "protocols, expectations and risks [that] are inevitably very different from those of archives more generally."43 For the GPP, archivization in the digital age becomes, even in its internal workings, a constant negotiation with state power, family politics, and interinstitutional dynamics. Fortunately, the scholars at the helm—Burns, Patricia Methven, and Karin Wulf—see the long-term value in such negotiations, and they have the institutional support from Royal Collection Trust, King's College London, the Omohundro Institute of Early American History and Culture, and William & Mary to support their vision. [End Page 467]
Models of how best to address obstacles that come with collaboration abounded in our workshop. For example, the Yale Indian Papers Project—with its goal of offering a wide audience "visual and intellectual access to significant historical knowledge [of New England's earliest cultures] for the purposes of teaching, scholarly analysis, and research"—has implemented guidelines for ethical collaboration as outlined in the Protocols for Native American Archival Materials and in the National Endowment for the Humanities Code of Ethics Related to Native Americans.44 Paul Grant-Costa and Tobias Glaza explain how "communication with and involvement of Native communities in the editorial process, especially to share information and to review documents for culturally sensitive information and evaluate them for final publication" is a "critical component of our professional and ethical standards." They readily admit that these standards hamper efficiency: "the review process can be time-consuming, especially for individuals with full time jobs, family and social obligations, and tribal responsibilities."45 And yet the standards cannot be jettisoned in the drive to completion as they are paramount to accomplishing a central part of the project's mission: "To re-inscribe indigeneity into a collection of documents that represents a shared history between Americans, Native Americans, Britons, and the Atlantic World by fostering the participation of Indian scholars and tribal members in the editing process and in discussions by acknowledging them as colleagues, scholars, intellectuals, and representatives of the Native voice."46 Practitioners such as Grant-Costa, Glaza, and others who so carefully consider how best to achieve both the intellectual goals and the ethical imperatives of their projects have to be patient, but they also need to justify the pace of their labor through careful articulation of their methodology and its rationale in grant applications, scholarly journals, or online apparatuses for the site. The return to methodology and its ethics, thus, is ubiquitous, as is the link between practice and politics, not cordoned off as the sole province of those working with culturally sensitive [End Page 468] materials. After all, the NEH Code of Ethics reminds us that though these protocols are "rooted in relations with native peoples of North America," the code should be applied liberally: "researchers are urged to follow it whenever living cultures and peoples are involved."47 Relationships and involvement do not develop with the speed associated with the digital—they are not documents to be scanned, records to be processed, or data to be wrangled—yet they are as vital as these steps in the creation of digital projects from the archives.
In a world of collaboration, transparency of process reigns supreme. How a scholar has done something becomes almost as important as what the scholar has done because, in part, the time scales of intellectual labor have shifted. The doing is not finite, the moment of publication no longer marks an end; instead we sail toward an ever-receding horizon of completion. The good news, of course, is that we do not sail alone. Scholars no longer toil in isolation, inputting archival materials in their brains, receding into dark offices and library carrels, and outputting definitive articles and monographs. Instead, that time in darkness, the work behind the curtain, is now in the spotlight as we are increasingly asked to account for how we know what we know. The progenitors of this methodological turn can, as I began this article exploring, be found in the both the academy and the library. It is important to recognize that in our current moment it is not only academic scholars infiltrating the way scholars in libraries work but vice versa: intellectual labor in libraries offers us both many instructive examples of best practices and a lineage for our work.
Consider that the scholars who first embarked on the systematic study and recording of books for the field of bibliography wondered from the start about their work's relationship to academic disciplines. In 1912, Shakespeare scholar and bibliographer W. W. Greg described bibliography, which he wanted to be at once more capacious and purer, as a science insofar as it is defined by how work is done rather than the material to which it is applied—the what of the work. He wrote, "It is the method itself, not the object to which that method is applied, that gives [bibliography] unity."48 Like historical narratives at certain moments, bibliography also relies on an explanation of method, what in digital humanities parlance is often referred to as "documentation." One of the most eloquent examples of documentation of process comes from bibliographer Jacob Blanck's preface to his Bibliography of American Literature (BAL) in which he describes, in exacting detail, how he constructed a methodology to "escape the fault of prejudice, both personal and sectional," so that in his decades of research he could [End Page 469] meet each author he encountered "on common ground."49 Though he did not flag it as such, this was the mission statement of his ambitious undertaking, and the documentation of process he offers in the preface to his first volume testifies to the efficacy of his method.
And yet, in the production of BAL, as in the production of archives-based digital projects, intentional collaboration seldom means easily found consensus. In his recapping of Blanck's process of selecting authors, Michael Winship, who, upon Blanck's sudden death in 1974, worked with Virginia L. Smyers at Harvard's Houghton Library to complete BAL, detailed the consensus model for an author's inclusion that Blanck developed. Blanck and the Bibliographical Society of America (BSA) formed an advisory committee and actively solicited suggestions beyond the small group convened, but Blanck was nonetheless often resolute once a choice was made. The BAL archive, a 43-box collection that is a monument to methodology, contains much of the back-and-forth that resulted from Blanck's open call and reveals that he was perhaps less open to outside critique than this method suggests. Blanck came to his list of 281 American authors who lived between the Federal period and 1931 only after publishing a prospectus and tentative author list of 270 names, together with an appeal for additions and deletions.50 Lyle Henry Wright, a prominent bibliographer and librarian at the Huntington Library, responded a month after Library Journal published Blanck's list with a note next to William H. Brewer's name, "I should prefer to see men like . . . William H. Brewer included rather than many of the third rate novelists and poetasters in the list."51 Despite Wright's urging, Brewer, who wrote more natural history than belles lettres, did not make the final cut. Such contrary opinions abound in the correspondence, as do records of Blanck's engagement with his critics, and not just about the authors he would include in his list. He wrote, for example, to Fredson Bowers that the description he set forth to define "edition" might not square with what Bowers was formulating for his work Principles of Bibliographical Description (1949), but Blanck found confidence in the committee established by the BSA, who backed his project. He responded to Bowers, "The definitions agreed on by my committee were arrived at after long (and sometimes heated!) discussions and I must confess that I am entirely willing to accept their findings without argument. At the beginning discussions, [End Page 470] I realized that universal acceptance and agreement were impossible, but agreement within this group has been reached and I accept the results."52 Transparency of process enabled Blanck's success, and the project's own archive shows that he took the time to engage with his detractors, even when he did not heed their advice.
Blanck's work offers us an instructive predecessor, a forebear for our own work, not only because we have a record of how he did—and did not, at times—build consensus, but also because he sought expertise beyond the hallowed halls of scholarship found in special collections and in academia. Blanck's professional training began neither in the library nor in the academy, but instead in the buying and selling of rare books. He did not spurn commercial enterprise as inherently anathema to his work; instead, he embraced the expertise held by those who bought and sold books for profit. In accounting for Blanck's considerable success in his undertaking, Winship explains that "without question the roots of BAL are in the book trade, not the academy."53 In a letter from BSA President Randolph Greenfield Adams to Blanck as he was starting out, we learn why. "The collector of Americana is way ahead of the professors of English and literature," writes Adams.54 One might say the same of the proprietary digitization companies that have built robust digital products out of the archives: these products offer models of what can (and at times cannot) be done successfully in the digitization of early American archives. These companies' objective—profit—might differ from humanists' goals, but digital scholarship ignores their products and potentials for collaboration at a considerable loss to its own objective of ultimately making the historical record more accessible.
Despite all the lessons that might be taken from Blanck's work, it is easy, in our current cultural moment of increased attention to marginalized voices in American literary history and changes in our understanding of the literary canon, to see how it fell short. Recent scholarship in black bibliography and Native American bibliography demonstrates that in many ways Blanck failed to capture "American literature," but not for lack of trying.55 [End Page 471] From his documentation it is evident how and why he used the sources and resources available to him. His detailed description of methodology invites future generations of scholars, in this age of digitization and recovery, to intervene. This process—both its transparency and its recording, in print and in his own documentation that Houghton Library has sagaciously preserved—is his success, not his inclusion or exclusion of certain authors.
Failure is having its own renaissance in our new knowledge economy, and though theoretically we might celebrate "more undisciplined knowledge, more questions and fewer answers," in the words of cultural critic Jack Halberstam, it is hard to build a career on failure.56 To figure out how to be stewards of and thinkers with the historical records, scholars must have space to test, to play, to fail, and then—borrowing from Samuel Beckett—to fail better. And yet the tenure case has no positive category for failure, neither the National Endowment for the Humanities nor the Mellon Foundation award money for it, and search committees do not seek it. Those institutional realities notwithstanding, a welcome collateral consequence of the prominence of methodology in scholarship is an increase in transparency, so that in a moment at the workshop that I deeply appreciated, Hannah Alpert-Abrams bluntly proclaimed that Reading the First Books—a project to "develop tools for the automatic transcription of books printed in multiple languages, using variable orthographies, during the first centuries of the printing press"—was a project "born out of the failures."57 She was referring to failures of transcription, but I think that most of our projects have had or will have a phoenix-from-the-ashes moment. Still, digital projects rarely fit neatly into "success" or "failure" boxes but frustratingly hover somewhere in between, as when Alpert-Abrams discussed the fewer transcriptions completed than planned and the possibility that the University of Texas might host the project, or Cavanaugh discussed the foray into the platform DocTracker that was abandoned when it did not grant the Washington Financial Papers Project the affordances needed to [End Page 472] transition from flat spreadsheets to a more dynamic relational database.58 How to make room for failure, or at least for experimentation, in the new knowledge economy has yet to be determined. Lasting impact, one measure of success, is as simple and as complicated as preservation when it comes to archives. A brick-and-mortar institution succeeds when it keeps its collections safe.59 What is the scholar's role in this legacy work in the digital environment?
Archives-based digital projects are a form of cultural heritage work, digital scholars hope, and yet we must recognize that the technology on which they rely does not have the same investment in longevity. Our projects exist across time, but they also exist in time. Products of the tools available, they are inevitably somewhat technologically determined, and the one constant in technology is that it is always changing. The Yale Indian Papers Project needed Mukurtu to develop before the project could adopt the tool that would put the theory they committed to into practice; the Washington Financial Papers Projects encountered endless frustration every time Drupal upgraded; the Pobladores Project Database's visualizations were not fully functional until days before the workshop began when the University of California, Riverside, Information Technology team found a workaround for glitches in the Google Earth functionality on which certain visualizations relied. In our working with the past, examples abounded of the ways in which we are irrevocably tethered to the present.
But our projects are not only synchronic. They are, of course, also diachronic as digital scholars inherit legacy data and data structures. Sometimes we welcome these predecessors, such as Hackel's work transforming the Early California Population Project and the Early California Cultural Atlas or Van Tine's discussion of legacy formats in the MARC record with all its [End Page 473] affordances. At other times, these predecessors must be jettisoned, as is the case with Grant-Costa and Glaza's discussion of legacy information systems, which presented them with a dilemma like the one Dorothy Porter encountered at Howard University almost a century ago. "The standard Library of Congress subject classification system does not adequately accommodate the American Indian perspective," they explain, concluding that it fails to "meet the research needs of a new generation of scholars and students."60 That legacy is in part traceable to the language used to describe analog materiality, and though the conundrums encountered might not be unique to the digital, they take on different urgencies as we make decisions about how to reorganize these digital archives. Cavanaugh offers a telling example of this in her discussion of "the document": "When documentary editors deal with correspondence, it is easy to define an individual letter as the document. But this is not the case with financial records. Is the entire account book a document? The individual pages within the book? Or are the distinct transactions considered a document?"61 Alpert-Abrams confronted a similar obstacle when thinking about how optical character recognition (OCR) software might disrupt the inherited hierarchy of the page as it transforms the image of the page into data that can be processed, searched, and analyzed. Alpert-Abrams asks, "What if OCR was not just a transcription system, but also a tool that analyzed the properties of a written page?"62 The workshop participants then wondered if "archive" is another example of the nomenclature used to describe analog materiality that no longer serves either a descriptive or an operational function in relation to our projects.
And then of course we must think of legacy in terms of the future, of what our projects will leave behind. How do scholars anticipate the needs of a new generation of researchers and students? As someone who has spent almost a decade working on both sides of the reference desk in archival institutions of varying size and scope, I have awoken in the middle of the night in a cold sweat thinking that someone a hundred years from now will shake their fist at me for some wrong step in the same way that I shake my fist at the stewards who brought glue and acid-laden folders to their archival practices, who threw out newspapers because they had been microfilmed, and who trimmed books' edges so that they looked tidier, marginalia be damned. "It is always with the best intentions that the worst work is done," Oscar Wilde tells us.63 How, if at all, can we know the legacies we are creating? Linked Open Data has become the latest effort to ward off the [End Page 474] ravages of irrelevancy of our data, data structures, and work, and I applaud Leon's use of it and her accompanying description of its utility in establishing the "spousal partnerships" among enslaved people, relationships that are meaningful and therefore worthy of recording, despite and because of their defiance of Euro-American domestic and familial norms. But what if tabular data and XML become Ozymandias's "vast and trunkless legs" and "shattered visage," that "colossal wreck, boundless and bare" stretching over "lone and level sands"?64 That prospect leads me to again interrogate how our work aligns with and differs from that of archival institutions.
Legacy was, and in many ways still is, the defining value of brick-andmortar archives, which traditionally strive for preservation of the past to access it in the present and ensure its future. This is indeed a mighty goal, and I selected the workshop group in part because of the mutual dependencies their projects have thoughtfully created between scholars working in the academy and those in libraries. One of the most common missteps I have observed in archives-based digital humanities work is the slippage between data that represents a moment in time versus that which reflects an archive itself, its institutional history, and its practices of collecting, however fraught these histories might be. The projects presented in our workshop avoid that trap in two ways. First, they refrain from bombastic claims and acknowledge institutional histories as formative in the making of their data. Second, they thoughtfully articulate how histories of collecting and processing the archival record can tell us as much about a given moment in time as the archival object can tell us about the moment it was created.
And yet we remain unsure of how to square models of legacy work that rely on fixity and rigidity, as exemplified in brick-and-mortar archives, with those that prize the fluidity and interoperability of digital projects. It is tempting to look to the pre-1936 divide between the historian and the archivist as a prelapsarian moment, a halcyon day to which we can hope to return. Though I am surely somewhat guilty of that here, the difference in medium between then and now reveals the limits of such nostalgia. Print archival guides and bibliographies mirror the fixity of the institutional holdings they described in ways that the digital does not, and yet they are instructive nonetheless. They point to other professional models than those currently available for working with the archive. Contemporary parlance often frames these models as alternatives to the norm, as part of the "Alt-Ac" movement. Resituating them in the history of the discipline normalizes them, but not with normalization itself as the end goal. Instead, as the participants repeatedly returned to in the workshop, we must continue to [End Page 475] pursue longevity because when carefully executed, digital projects in early America are the new legacy work. Or rather, with new professional models and sustained institutional support, they can be. [End Page 476]
Molly O'Hagan Hardy is currently an independent scholar who consults on special collections digitization. She would like to thank the organizers and participants in the workshop. She is also grateful to Matthew Brown, Matt Cohen, Will Slauter, and Michael Winship for their insights and suggestions for the essay.
Editor's note: The following essay grew out of the 2018 WMQ-EMSI workshop, "Archives-Based Digital Projects in Early America," the thirteenth in an annual series sponsored by the University of Southern California-Huntington Early Modern Studies Institute (with financial support from the USC Dornsife College of Letters, Arts and Sciences, the Mellon Foundation, and the National Endowment for the Humanities) and the Omohundro Institute of Early American History and Culture, hosted by the Huntington Library and the University of Southern California on May 18–19, 2018. The workshops are intended to foster intellectual exchange among a group of scholars approaching a general historical question from diverse chronological, geographic, and methodological perspectives. Molly O'Hagan Hardy acted as the workshop's convener. The participants were Hannah Alpert-Abrams, Arthur Burns, Erica Cavanaugh, Tobias Glaza, Paul Grant-Costa, Steven Hackel, Sharon Leon, Patricia Methven, and Lindsay Van Tine. Each supplied a precirculated paper linked to their project's website, and each offered a formal comment on another's essay and website. For the 2018 WMQ-EMSI workshop website, see https://oieahc-cf.wm.edu/workshops/emsi/2018/index.cfm.
1. Emmanuel Le Roy Ladurie, The Territory of the Historian, trans. Ben Reynolds and Siân Reynolds (Chicago, 1979), 5.
2. Edmund A. Bowles, "Towards a Research of New Dimensions," in Computers in Humanistic Research: Readings and Perspectives, ed. Bowles (Englewood Cliffs, N.J., 1967), 11.
3. The change that Emmanuel Le Roy Ladurie foresaw of course developed unevenly. A decade later and in the context of early American history, J. McAllister lamented that "British North America is the final refuge for statistical incompetents studying a period of modern American history." McAllister predicted that despite "social history blossom[ing] like a green bay tree," change was afoot. He concluded his extensive and in many ways prescient literature review of scholarship in early America with the provocation, "wherever 'statistics' exist can itchy-fingered computer buffs be far behind?" McAllister, "Colonial America, 1607–1776," Economic History Review 42, no. 2 (May 1989): 245–59 ("British," "wherever," 255, "blossom[ing]," 251). For MARC, see Matt Schudel, "Henriette D. Avram; Transformed Libraries," Washington Post, Apr. 28, 2006; repr. as Schudel, "Henriette Avram, 'Mother of MARC,' Dies," Library of Congress Information Bulletin, May 2006, https://www.loc.gov/loc/lcib/0605/avram.html.
4. Henriette D. Avram, MARC: Its History and Implications (Washington, D.C., 1975), 18.
5. For examples of the criticism of RECON, see Nicholson Baker, "Annals of Scholarship: Discards," New Yorker, Apr. 4, 1994, 64–86; and, more recently, Timothy Messer-Kruse, "How Google Scrambled the Academic Mind," Chronicle of Higher Education, May 29, 2019, http://www.chronicle.com/article/How-Google-Scrambled-the/246382. Opinions vary widely on MARC's longevity. The current moment in the technology of the catalog is commonly referred to as "a post-MARC environment" because, beyond its "postness," no one really knows what this period will look like. The most tenable new platform seems to be one also initiated at the Library of Congress. A decade ago, the Library of Congress began to consider how library data could be made more accessible and interoperable by making it more active in the semantic web. In 2011, the Library of Congress set out to build a framework that could be widely implemented. The result was BIBFRAME, a replicable Linked Data pattern used by libraries and archives to define web-accessible resources and to promote sharing and collaboration. Published in structures that search engines such as Google and Bing can understand, the bibliographic data could now be consumed by the open web. Just as with MARC in its early days, BIBFRAME—which has yet to be widely adopted—has been received by the bibliographic and especially the rare book community with mixed emotions. For more on BIBFRAME, see "Bibliographic Framework as a Web of Data: Linked Data Model and Supporting Services," Library of Congress, Nov. 12, 2012, http://www.loc.gov/bibframe/pdf/marcld-report-11-21-2012.pdf. For a salient critique of BIBFRAME, see Karen Coyle, "If It Ain't Broke," Coyle's InFormation (blog), Apr. 12, 2017, http://kcoyle.blogspot.com/2017/04/if-it-aint-broke.html.
6. Terry Belanger and Stephen Paul Davis, "Rare Book Cataloguing and Computers—II," AB Bookman's Weekly, Jan. 14, 1980, 187–204 (quotation, 187).
7. Earl R. Taylor, "Cataloguing and Computers: Librarians and Cyberphobia," Papers of the Bibliographical Society of America (PBSA) 75, no. 4 (Fourth Quarter, 1981): 392–400 (quotation, 395).
8. William B. Todd, "The ESTC as Viewed by Administrators and Scholars," PBSA 75, no. 4 (Fourth Quarter, 1981): 389–92 (quotation, 392).
9. I am borrowing here from the title of an influential work by Geoffrey C. Bowker and Susan Leigh Star, Sorting Things Out: Classification and Its Consequences (Cambridge, Mass., 2000). In this book, they argue that "systems of classification (and of standardization) form a juncture of social organization, moral order, and layers of technical integration. Each subsystem inherits, increasingly as it scales up, the inertia of the installed base of systems that have come before"; ibid., 33.
10. Safiya Umoja Noble, Algorithms of Oppression: How Search Engines Reinforce Rac ism (New York, 2017), 138–39 (quotation).
11. Le Roy Ladurie, Territory of the Historian, 5.
12. Roy Rosenzweig, Clio Wired: The Future of the Past in the Digital Age (New York, 2011), 24 (quotations).
13. Alfred F. Young and Gregory H. Nobles, Whose American Revolution Was It? Historians Interpret the Founding (New York, 2011), 24.
14. Charles M. Andrews, "American Colonial History (1690–1750)," in Annual Report of the American Historical Association for the Year 1898 (Washington, D.C., 1899), 47–60 ("complete," 56, "loose," 53); Andrews, The Historical Development of Modern Europe: From the Congress of Vienna to the Present Time, 2 vols. (New York, 1896–98).
15. Charles McLean Andrews, "On the Writing of Colonial History," William and Mary Quarterly, 3d ser., 1, no. 1 (January 1944): 27–48 ("desire," 27–28, "Because," 29). I do not mean to suggest that Andrews's archival turn was a wholly original self-fashioning. As Anthony Grafton traces in The Footnote: A Curious History, nineteenth-century German Leopold von Ranke in many ways invented scientific history by "creat[ing] and dramatiz[ing] a new practice, based on a new kind of research and made visible by a new form of documentation." Though Grafton shows how Ranke "considerably exaggerated the archival component of his work," Ranke succeeded because, writes Grafton, he brought "the flavor and texture of the documents into his own text. . . . He made his book into a sort of archive. He enabled the reader to share something of the impact of his own direct encounter with the sources"; Grafton, The Footnote: A Curious History (Cambridge, Mass., 1997), 56 ("creat[ing]"), 61 ("considerably"), 57 ("flavor"). In the context of early American history, Herbert Eugene Bolton published his Guide to Materials for the History of the United States in the Principal Archives of Mexico in 1913, doing for archives south of the continental United States what Andrews had done for those to its east. Thanks to Steven Hackel for bringing Bolton's work to my attention.
16. Bernard Bailyn, Atlantic History: Concept and Contours (Cambridge, Mass., 2005), 5.
17. Karin Wulf, "#VastEarlyAmerica and Origins Stories: WMQ 1:1," Uncommon Sense (blog), Omohundro Institute of Early American History and Culture, Feb. 22, 2016, http://blog.oieahc.wm.edu/vastearlyamerica-origins-stories-wmq-11/ (quotation). For a more recent explanation of #VastEarlyAmerica, see Wulf, "Vast Early America: Three Simple Words for a Complex Reality," Humanities 40, no. 1 (Winter 2019), https://www.neh.gov/article/vast-early-america.
18. "Reading the First Books: Multilingual, Early-Modern OCR for Primeros Libros," University of Texas at Austin, https://sites.utexas.edu/firstbooks/; for the Internet Wayback Machine archive of this site, see https://web.archive.org/web/20190207212803/https://sites.utexas.edu/firstbooks/; "Database," Early Novels Database, https://earlynovels.github.io/; George Washington Financial Papers Project (GWFPP), Washington Papers, University of Virginia, http://gwpapers.virginia.edu/editions/financial-papers-project/.
19. Sharon Leon, "Life and Labor under Slavery: The Jesuit Plantation Project," Jesuit Plantation Project, http://jesuitplantationproject.org/s/jpp/page/About; Yale Indian Papers Project, Yale University, https://yipp.yale.edu/ (which has begun transitioning to the Native Northeast Research Collaborative, https://www.thenativenortheast.org/); Georgian Papers Programme (GPP), https://georgianpapers.com/about/; Steven Hackel, "Towards a Spatial History of Colonial California: The Early California Cultural Atlas and the Pobladores Project Database," paper presented at the 2018 WMQ-EMSI workshop.
20. Jean M. O'Brien, Firsting and Lasting: Writing Indians out of Existence in New England (Minneapolis, 2010), xvi (quotation).
21. Marisa J. Fuentes, Dispossessed Lives: Enslaved Women, Violence, and the Archive (Philadelphia, 2016), 12 (quotations). See also Jennifer L. Morgan's search in the archives for records "on the subaltern, on people and places that are understood as outside of or marginal to the archival project of nation building"; Morgan, "Archives and Histories of Racial Capitalism: An Afterword," Social Text 33, no. 4 (December 2015): 153–61 (quotation, 154).
22. Laura E. Helton, "On Decimals, Catalogs, and Racial Imaginaries of Reading," PMLA: Publications of the Modern Language Association of America 134, no. 1 (January 2019): 99–120 (quotations, 112). For more on Dorothy Porter, see Thomas C. Battle, "Moorland-Spingarn Research Center, Howard University," Library Quarterly 58, no. 2 (April 1988): 143–51; Melissa Adler, "Classification along the Color Line: Excavating Racism in the Stacks," Journal of Critical Library and Information Studies 1, no. 1 (2017): 1–32, esp. 4, http://libraryjuicepress.com/journals/index.php/jclis/article/view/17; Zita Cristina Nunes, "Cataloging Black Knowledge: How Dorothy Porter Assembled and Organized a Premier African Research Collection," Perspectives on History, December 2018, repr. as Nunes, "Remembering the Howard University Librarian Who Decolonized the Way Books Were Catalogued," Smithsonian, accessed Nov. 27, 2018, https://www.smithsonianmag.com/history/remembering-howard-university-librarian-who-decolonized-way-books-were-catalogued-180970890/. For more on the neglected labor of women in bibliographic and library work, see Barbara A. Mitchell, "Boston Library Catalogues, 1850-1875: Female Labor and Technological Change," in Institutions of Reading: The Social Life of Libraries in the United States, ed. Thomas Augst and Kenneth Carpenter (Amherst, Mass., 2007), 119–47; Molly O'Hagan Hardy, "Digitization," Early American Studies 16, no. 4 (Fall 2018): 637–42.
23. See Michelle Caswell, "Seeing Yourself in History: Community Archives and the Fight Against Symbolic Annihilation," Public Historian 36, no. 4 (November 2014): 26–37 (quotation, 27); T-Kay Sangwand, "Preservation Is Political: Enacting Contributive Justice and Decolonizing Transnational Archival Collaborations," KULA: Knowledge, Creation, Dissemination, and Preservation Studies 2, no. 1 (November 2018). See also Christian Kelleher, "Archives Without Archives: (Re)Locating and (Re)Defining the Archive Through Post-Custodial Praxis," in "Critical Archival Studies," ed. Caswell, Ricardo Punzalan, and Sangwand, special issue, Journal of Critical Library and Information Studies 1, no. 2 (2017): 1–30.
24. I am referring here to the foundational text for learning how to write an academic argument, now in its third edition, Gerald Graff and Cathy Birkenstein,"They Say / I Say": The Moves That Matter in Academic Writing (New York, 2017).
25. Leon, "Life and Labor under Slavery."
26. For more on Sharon M. Leon's methodology, see "Thinking with Linked Data; Representing History," [bracket] (blog), Nov. 1, 2018, http://www.6floors.org/bracket/2018/11/01/thinking-with-linked-data-representing-history/.
27. Hackel, "Towards a Spatial History of Colonial California," 3.
28. Ibid., 4 (quotation). For the Pobladores Project, see Steven Hackel, Jeanette Zernecke, and Nat Zappia, "Origins of California Settlers," Early California Cultural Atlas, 2015, http://ecai.org/ecca/Settlers/SettlersByPlace.html.
29. Nicholas Popper, "Archives and the Boundaries of Early Modern Science," Isis 107, no. 1 (March 2016): 86–94 (quotation, 93).
31. James Fisher, "Carving up the Georgian Papers: Metadata and Subject Indexing," GPP (blog), Aug. 28, 2018, https://georgianpapersprogramme.com/2018/08/28/carving-up-the-georgian-papers-metadata-and-subject-indexing/.
33. E. D. Hirsch, "From Validity in Interpretation," in Theory of the Novel: A Historical Approach, ed. Michael McKeon (Baltimore, 2000), 14–33 (quotation, 17); Northrop Frye, "Rhetorical Criticism: Theory of Genres," in Anatomy of Criticism: Four Essays, ed. Harold Bloom (1957; repr., Princeton, N.J., 2000), 243–340.
34. Genre Terms: A Thesaurus for Use in Rare Book and Special Collections Cataloguing, 2d ed. (Chicago, 1991).
35. The Library of Congress began developing Library of Congress Genre/Form Terms for Library and Archival Materials (LCGFT) in 2007, but it was not until May 2015 that the library approved a first set of terms for use with literary works. For more details, see J[anis] Young, "Summary of Decisions, Editorial Meeting Number 15—Genre/ Form Terms for Literary Works (Part I)," Mar. 2, 1015, http://www.loc.gov/aba/pcc/saco/cpsoed/psd-1515.html.
36. Sharon M. Leon, "The Jesuit Plantation Project, Redux," paper presented at the 2018 WMQ-EMSI workshop, 5.
37. Leon, "Thinking with Linked Data."
39. Erica Cavanaugh, "The George Washington Financial Papers Project: Project Description," paper presented at the 2018 WMQ-EMSI workshop, 8.
40. Popper, Isis 107: 89 (quotation). Jacques Derrida first described the relationship between the archives and feelings in Derrida, Archive Fever: A Freudian Impression, trans. Eric Prenowitz (Chicago, 1996). Ann Cvetkovich explores this relationship more fully in Cvetkovich, An Archive of Feelings: Trauma, Sexuality, and Lesbian Public Cultures (Durham, N.C. 2003). For the legacy of antiquarians, see [Walter Scott], The Antiquary (Edinburgh, 1816). For the work Hannah Mather Crocker wrote on the occasion of gifting her family's books and manuscripts to the American Antiquarian Society, see Crocker, "Antiquarian Researches, Pleasant and Easy by an Original Antiquarian," Mather Family Papers, 1613–1819, box 10, folder 4, American Antiquarian Society, Worcester, Mass; repr. in Observations on the Real Rights of Women and Other Writings, ed. Constance J. Post (Lincoln, Neb., 2011). Washington Irving depicts the early nineteenth-century antiquarian in [Irving], The Sketch Book of Geoffrey Crayon, Gent. (New York, 1819). For an extensive discussion of the figure of the antiquarian in early America, see Lindsay DiCuirci, Colonial Revivals: The Nineteenth-Century Lives of Early American Books (Philadelphia, 2018). For the reanimation of the figure of the antiquarian in twentieth-century American bibliography, see Molly O'Hagan Hardy, "Bibliographic Enterprise and the Digital Age: Charles Evans and the Making of Early American Literature," American Literary History 29, no. 2 (Summer 2017): 331–51.
41. I borrow this formulation from Nicholas Popper because while I discuss current and future archival relationships, he looks back to the pre-romanticized figuring of archival work through an examination of how English statesmen organized their papers in the second half of the sixteenth century. Popper writes that "archivization has been presented as an organic development rather than a strategy for negotiating political wilds, as a rational reflex rather than a learned behavior"; Popper, Isis 107: 89.
42. Matthew G. Kirschenbaum brought a similar question to the fore in "Done," a special cluster of articles in Digital Humanities Quarterly he edited a decade ago, asking, "What is the measure of 'completeness' in a medium where the prevailing wisdom is to celebrate the incomplete, the open-ended, and the extensible?" See Kirschenbaum, "Done: Finishing Projects in the Digital Humanities," Digital Humanities Quarterly 3, no. 2 (2009), http://www.digitalhumanities.org/dhq/vol/3/2/000037/000037.html.
43. Arthur Burns, "The Georgian Papers Programme," paper presented at the 2018 WMQ-EMSI workshop, 3.
44. "About the Project," Yale Indian Papers Project, accessed Nov. 30, 2018, https://yipp.yale.edu/about (quotation); Protocols for Native American Archival Materials, Apr. 9, 2007, accessed Nov. 27, 2018, https://www2.nau.edu/libnap-p/protocols.html; "Code of Ethics Related to Native Americans," National Endowment for the Humanities (NEH), accessed Nov. 27, 2018, https://www.neh.gov/grants/manage/code-ethics-related-native-americans. For other models of working with indigenous archival materials in the digital environment, see the Traditional Knowledge (TK) Labels provided by Jane Anderson and Kim Christen's Local Contexts, an initiative to support Native, First Nations, Aboriginal, and Indigenous communities in the management of their intellectual property and cultural heritage; "Traditional Knowledge (TK) Labels," Local Contexts, accessed Dec. 20, 2018, http://localcontexts.org/tk-labels/.
45. Paul Grant-Costa and Tobias Glaza, "Editing the Native Northeast Archives," paper presented at the 2018 WMQ-EMSI workshop, 11.
46. "Mission Statement," Yale Indian Papers Project, accessed Nov. 30, 2018, https://web.archive.org/web/20170705100845/http://yipp.yale.edu/mission-statement.
47. "Code of Ethics Related to Native Americans," NEH, accessed Nov. 27, 2018, https://www.neh.gov/grants/manage/code-ethics-related-native-americans.
49. Jacob Blanck, "Preface," Bibliography of American Literature (New Haven, Conn., 1955), 1: xiv ("escape"), xi ("common").
50. Michael Winship, "Introduction to the Reprint Edition," in Bibliography of American Literature Compiled by Jacob Blanck for the Bibliographical Society of America, vol. 1, Henry Adams to Donn Byrne (New York, 2003), .
51. For Brewer's name and Wright's comment on it, see Bibliography of American Literature Papers, 1943–1991, box 34, Houghton Library (HL), Harvard College Library, Harvard University, Cambridge, Mass. For a reprint of Blanck's list, see "The New Bibliography of American Literature," Library Journal 69, no. 12 (June 1944): 549–50.
52. "Letter from Jacob Blanck to Fredson Bowers," June 26, 1946, Bibliography of American Literature Papers, box 32, HL; Fredson Bowers, Principles of Bibliographical Description (Princeton, N.J., 1949).
53. Winship, "Introduction," in Bibliography of American Literature, .
54. "Letter from Randolph G. Adams to Jacob Blanck," Oct. 6, 1945, Bibliography of American Literature Papers, box 34, HL.
55. "Black Bibliography Project," Meredith L. McGill: Information, Writing, Work-in-Progress (blog), accessed Nov. 30, 2018, https://mlmcgill.com/current-projects/black-bibliography-project/. "About DIBB," DIBB: The Digital Black Bibliographic Project: Creating Next Generation Cultural Data, Oct. 15, 2018, accessed Nov. 30, 2018, http://dhrace.net/DIBB/about; "What We Do," Black Book Interactive Project, accessed Nov. 30, 2018, http://bbip.ku.edu/what-we-do; "Digital Atlas of Native American Intellectual Traditions: Creating a National Network," accessed Nov. 30, 2018, https://danait.wordpress.amherst.edu/.
56. Judith [Jack] Halberstam, The Queer Art of Failure (Durham, N.C., 2011), 10 (quotation). An early paean to failure in digital humanities can be found in the artful articulation of John M. Unsworth more than two decades ago, "If an electronic scholarly project can't fail and doesn't produce new ignorance, then it isn't worth a damn." See Unsworth, "Documenting the Reinvention of Text: The Importance of Failure," Journal of Electronic Publishing 3, no. 2 (December 1997), https://doi.org/10.3998/3336451.0003.201. Through open ballot and without institutional affiliation, the Digital Humanities Awards have included a number of atypical categories for achievement; since 2014, they have included the award category "Best Exploration of DH Failure." For more, see "Digital Humanities Awards: Highlighting Resources in Digital Humanities," Digital Humanities Awards, accessed Mar. 4, 2019, http://dhawards.org/.
57. "Reading the First Books," accessed Dec. 22, 2018 ("develop"); Hannah Alpert-Abrams, with Maria Victoria Fernandez, Bryan Tarpley, and Megan Scarborough, "People's Paper: Reading the First Books: Multilingual, Early Modern OCR for Primeros Libros," paper presented at the 2018 WMQ-EMSI workshop, 3 ("born").
58. Erica Cavanaugh, "The George Washington Financial Papers Project: Project Description," paper presented at the 2018 WMQ-EMSI workshop, 5
59. For examples of explicit mention of preservation in the mission statements of special collections libraries and archives, see "Mission Statement," American Antiquarian Society, accessed Dec. 18, 2018, https://www.americanantiquarian.org/mission-statement ("Our mission is to collect, preserve, and make available for study the printed record of what is now the United States of America from first European settlement through the year 1876"); "About the Kislak Center," Penn Libraries, accessed Dec. 18, 2018, https://www.library.upenn.edu/kislak/about ("to make our collections accessible; to use technology in innovative and meaningful ways; to enhance teaching and research; and to preserve our cultural resources for future generations"); "JCB Online," John Carter Brown Library, accessed Dec. 18, 2018, https://www.brown.edu/academics/libraries/john-carter-brown/jcb-online ("to preserve and collect books and other recorded information on the discovery and settlement of the Western hemisphere"); "Mission + Vision and Aspirations for 2027," Beinecke Rare Book and Manuscript Library, accessed Dec. 18, 2018, https://beinecke.library.yale.edu/mission-vision-and-aspirations-2027 ("The Beinecke Rare Book & Manuscript Library collects, catalogs, preserves, and makes accessible rare books, manuscripts, and other formats from ancient to modern times").
60. Grant-Costa and Glaza, "Editing the Native Northeast Archives," 3.
61. Cavanaugh, "George Washington Financial Papers Project," 2.
62. Alpert-Abrams, "People's Paper," 11.
63. Oscar Wilde, "The Critic as Artist: Second Part: With Some Remarks upon the Importance of Discussing Everything," in The Prose of Oscar Wilde (New York, 1916), 141–200 (quotation, 188).
64. Leon, "Jesuit Plantation Project, Redux," 6 ("spousal"); Percy Bysshe Shelley, "Sonnet.–Ozymandias," in The Poetical Works of Percy Bysshe Shelley (London, 1874), 552 ("vast").