During a 2011 Northeast Modern Language Association (NeMLA) panel dedicated to the works of eighteenth-century French-American author J. Hector St. John de Crèvecoeur, panelists and attendees participated in an engaged discussion regarding the rewards and challenges of teaching Crèvecoeur to 21st-century college students. Among the biggest challenges discussed, audience members noted the surprising lack of available teaching editions of Crèvecoeur’s most celebrated work, Letters from an American Farmer. American literature anthologies, for example, typically include only small sections of Letters as a way to introduce students to the style and subject matter of Crèvecoeur’s text. Trade publishers such as Dover, Penguin Classics, and Oxford Classics have published complete editions of the text, but — with their lack of limited (if any) annotations and other supplementary aids for textual, historical, cultural, or critical interpretations — these editions are not ideal for classroom use and have not been published with educators and students in mind.[1] Given the literary and cultural significance yet textual complexity of Crèvecoeur’s narrative, panel attendants remarked, there was a decisive need for a teaching edition of Letters.

Since meeting as presenters on this NeMLA panel, we have teamed up to try to identify a workable solution to what we saw as a challenge to professors of early American literature: finding a way to allow faculty to more fully integrate Crèvecoeur’s Letters into the college classroom. This article will outline our pedagogical concerns with current print editions of Crèvecoeur’s work (and other early American texts); lessons learned from our early research on alternative digital critical models; our attempts as faculty to begin to develop and create a digital critical edition of Letters; our experiences using this digital edition in the classroom; and the broader implications for the growth of digital editions of early American (and other) texts. Finally, we will discuss an initially unanticipated shift in our project: transferring the editorial responsibilities over to our students and teaching them to curate the text themselves as a way to help them become active readers, to deepen their engagement with material, to educate them about the editorial process, and to move them towards the professional.

Crèvecoeur’s Letters from an American Farmer is a complex, comprehensive canonical American text studied across disciplines, including English, History, and Political Science. What makes the text so well suited for classroom discussion at all levels are the questions it raises about American identity and utopian idealism, as well as the rich and varied ways it does so. In asking that now famous question, “What then is the American, this new man?”, Crèvecoeur’s Letters became the first and most important text to openly examine early American identity, culture, politics, and social structures (Letter III). The narrative seeks to address all facets of American colonial life on the eve of the American Revolution. Caught between a utopian view of the colonies shaped by Enlightenment thinking and the growing chaos and conflict around him, the narrator simultaneously highlights both early idealism and disillusionment in the American project. In the process, he touches on fundamental issues that were prevalent in late eighteenth-century America and are still of critical significance today: immigration, commerce, the frontier, the environment, Indigenous populations, the transatlantic slave trade, and disparate class structures. In addition to its approach to important social and historical issues, the text — as an epistolary novel that skirts the boundaries between fiction, history, and autobiography — poses fundamental questions about literary form itself. Letters offers students, teachers, and scholars alike a critical lens through which to examine American history, politics, culture, literature, and language.

Despite the popularity of, the importance of, and the classroom opportunities provided by Crèvecoeur’s text, it is difficult for students to access and interpret for a number of reasons. The late-eighteenth-century language is challenging in its own right. The question of the text’s genre is also convoluted: Letters is an epistolary narrative comprised of twelve letters written by a fictional character, an Orange County New York farmer James, to a British subject named Mr. F.B., describing — as James says — “our American modes of farming, our manners, and peculiar customs” (Letter I). Although Crèvecoeur himself was French born and only immigrated to America in the late 1750s, he shares enough qualities with James — as a farmer, living in Orange County New York, struggling to navigate the tenuous socio-historical dynamics of rural life on the cusp of the American Revolution — that the text ultimately blurs the boundaries between autobiography, non-fictional prose, and the novel. The ambiguous genre often makes it difficult for students to ascertain the distance between the author and the narrator. This distance is further complicated by sudden narrative shifts between and even within letters, where, for example, notes or brief histories from James or even from another writer are occasionally inserted into the narrative. Finally, and most importantly, the narrative itself is engrained in the socio-historical background of the late Colonial period in America as well as the complex philosophy of the Enlightenment (and, arguably, the ultimate breakdown of that philosophy) and, as such, includes references to various eighteenth-century subjects — botany, farming, hunting, whaling, slavery, Indigenous culture, religion — that often elude twenty-first century readers.

To fill the gap left by anthologies that publish fragments of Letters — but fail to illustrate the complexities of genre, the development of the narrative, the narrator’s emerging story, as well as its radically shifting tone — and trade editions that publish the complete text — but lack the necessary detailed annotations and developed ancillary materials to make sense of the context — we set out to publish a print critical edition of the complete text, replete with those necessary supplementary reading materials. We reached out to a couple of publishers, such as Norton and Broadview, who specialize in critical editions particularly developed for and marketed towards the college classroom. We received interest in our initial inquiries and requests for more developed proposals, but ultimately the consensus seemed to be that — although a teaching edition of Letters was, indeed, both timely and valuable — given the trade editions currently available, the market could not accommodate another print edition.

Because of the limited print publishing options and the growing rise in the digital market, we decided to shift course and research digital publishing possibilities. As we did, we quickly realized that, while more challenging, the digital might provide a better opportunity to achieve our primary goal: to make Crèvecoeur’s text not only available but also accessible to the twenty-first century student. Given the very intertextual nature of Crèvecoeur’s writing, a digital approach to his work began to make the most sense, particularly for today’s students. Not only are twenty-first century students woefully uninformed about much of the history and culture of early America but they are also not inclined to do research (even using an online dictionary) to look up information. Unfortunately, printed footnotes are not especially useful either, since most students tend to skip them: today’s students are used to having information at their fingertips — and lots of it. The elements of Crèvecoeur that students, especially non-English or History majors, often tend to find dull (the sections on botany and whaling, for instance) could now be made visual and relevant to, almost tactile for, the contemporary reader. In addition, as an open-source edition, this new format would speak to potential print edition market concerns, would expand open access to early American literature for students and other readers alike, would allow us to engage in the growing field of the digital humanities, and could eventually provide the opportunity to further develop our project to include other early American texts. In general, an online edition would allow us to make use of current developments in technology that would maximize the learning potential for readers and open up the textual world of early America to those outside academia.

While the digital option was compelling, it created a number of immediate challenges. The most self-evident obstacle was that, while both of us had various levels of expertise and experience utilizing technology in the classroom, neither of us at the time could claim to be “digital humanists.” In essence, we “fell into” the digital humanities. Rather than a conscious dive into the digital — its practical applications, its theoretical underpinnings, and the nuances of its scholarship — we were led there, essentially as a publishing refuge when we realized that the analog could no longer accommodate what we had initially hoped to produce. Not only were we not coders, but to cross over into the digital we would have to begin, largely as novices, to investigate whether there were any publishing platforms that could accommodate our Platonic ideal of this edition; and then we would have to consider the theoretical implications of this publishing paradigm shift, implications that we could not then even fully anticipate. Although this circuitous route toward the digital humanities is no doubt far more typical in academia, this alternate perspective — the attempt to penetrate the digital humanities after years of producing more traditional scholarship — is not readily reflected in DH scholarship. Matthew K. Gold and Laura F. Klein’s 2012, 2016, and 2019 Debates in the Digital Humaities, for example, each presents a near exhaustive approach to the “debates” in the digital humanities — defining, theorizing, critiquing, teaching, and envisioning the future of DH (in the 2012 edition, for example) — but by and large each edition still focuses on those who already understand the terms of the debate. In what follows, we hope to add to the rich discussion of the digital humanities by offering a slightly different perspective: a process that involved familiarizing ourselves with foundational digital humanities scholarship, exploring relevant digital platforms, and engaging with other digital humanities projects and pedagogies to begin to carve a space for our digital edition. In short, we hope to share our occasionally painful, frequently tortuous and challenging, but also deeply rewarding process of “going digital.”

The first step in switching our perspective from the print to the digital was confronting new theoretical questions, including critical and pedagogical issues that were not evident when our goal was simply to make a print critical edition of Letters from an American Farmer. For one thing, the existence of and increasing emphasis on digital texts over their print counterparts has raised, and continues to raise, questions about the very nature of scholarship, criticism, and textuality as well as how these elements intersect. In their “Introduction” to Text Editing, Print and the Digital World, Marilyn Deegan and Kathryn Sutherland provide an overview of some of these critical issues. For example, can the standards and methods used to create print texts also be considered valid when creating digital editions, and how does the nature of the text impact the way it is read and used? In short, they observe, “Our ideas of what constitutes a literary work are under revision: what factors determine its boundaries and shape, what we mean by ‘text’ and what features define it” [Deegan and Sutherland 2009]. The questions that guided their 2009 volume in many ways remain at the heart of debates over digital editions today; we had to consider several of these questions as we began to investigate digital options.

The most important and immediate question we had to confront was the one about form: what exactly did we mean when we said we were compiling a “digital ‘critical edition’”? In Digital Scholarly Editing: Theories, Models and Methods, Elena Pierazzo, referencing the work of Edward Vanhoutte, observes that “a great theoretical effort has been deployed in order to distinguish digital archives from digital editions and documentary editions from critical editions” [Pierazzo 2015]. Our first task was to tease out these differences. We had in mind the Norton editions as a generally accepted and respected model of a print “critical edition”: a text with an approved version of the text; annotations that explain terms, concepts, and references to help guide the reader; documents that set the text in its historical and cultural context; scholarly articles that introduce the reader to the critical conversation taking place around the text over time; and a bibliography.

When translating this model of a critical edition to the digital, we generally subscribed to the type of ideal digital model Peter Shilingsburg outlines in “How Literary Works Exist: Convenient Scholarly Editions”:

The scholarly electronic edition of the future…will be convenient: it will be as cheap as a paperback book, with a user-friendly interface…with bookmarks, highlighting, space for marginal notes, and the ability to annotate… In order to avoid the down sides of paperback books, the electronic edition must give accurate access to representations (images) of specific historical forms of the text and specific critical editions of the text and to the ancillary materials that contextualized the texts at the time of origin and the times of reception that we care about. It would be even more convenient if the accumulation of scholarship related to the work were also at hand.  [Shillingsburg 2009]

While we wanted to make all of these digital materials “convenient” and “at hand” for the reader [Shillingsburg 2009], we were also cognizant of the tension regarding the digital that Jessica Pressman and Lisa Swanstrom cogently articulate in “The Literary And/As the Digital Humanities,” specifically the tension between data/information and interpretation in DH. We wanted to utilize the digital to provide that critical information for readers (annotations, ancillary materials, accumulation of scholarship) but simultaneously prevent the digital from turning readers into passive recipients of knowledge. In order to ensure that these technological tools would be used to help readers become engaged participants, to make sure, in the words of Pressman and Swanstrom, “literary critics don’t take data at its word,” we wanted to incorporate interactive features, such as integrated reading questions and the and the capacity for commentary [Pressman and Swanstrom 2013].

Most importantly, however, we agreed with Patrick Sahle’s definition in “What is a Scholarly Digital Edition?” that “a digitized edition is not a digital edition” [Sahle 2016]. We did not want to create a digitized facsimile of an analog edition. Rather, we wanted to take the work that digital archive projects such as Project Gutenberg had done and make these unedited digital texts not only available but accessible to students. Acknowledging that we would not be able to create that “ideal” edition of the future that Shillingsburg outlines, we aimed to approximate a version of it. Our initial plan was to create a digital edition of Crèvecoeur’s Letters that would include critical annotations as well as links to appropriate supplemental materials as hypertext; students would be able to move their cursor over or click on a word to locate definitions, reading questions, historical and cultural references, videos, and hyperlinks to relevant web sites. We wanted to create an interactive experience for the reader that included the kinds of dynamic features Sahle uses to define digital editions: “browsing paths,” “real hyperlinks,” and “integrated technical tools” [Sahle 2016].

While critics such as Sahle stop short of distinguishing between a critical and scholarly edition, our goal was to move beyond form and also to distinguish our digital critical edition of Letters from a scholarly edition that focused primarily on textual criticism. Dennis D. Moore has published several collections of Crèvecoeur’s writing that pay careful attention to textual sources and composition and are critical for academic scholarship. We hoped to take on the work that Kathryn Sutherland claims has been deemed less glamorous: the kind of textual annotation that is “most in favour with student readers, general readers and commercially minded publishers, to all of whom it is perceived as adding value” [Sutherland 2009]. And perhaps this was one place where digital editions could start to differ from their analog counterparts: that they take seriously the place of textual annotations and the readers who value them. At the same time, we were concerned with maintaining the accuracy of the edition and demonstrating that this type of innovative and interactive digital edition could be as viable and reliable as the more traditional printed one we had hoped to create at the outset. This is why we were careful about the edition of Letters we chose to use and about creating that edition ourselves, not depending on another digital edition already in existence, although there are several. We ultimately chose the 1783 Thomas Davies London edition of Letters as the most authoritative, and transcribed the text from a digital facsimile of the original into a .doc file to ensure its accuracy. In addition, any minimal emendations that were made would be noted in the edition. The focus would be on editing the apparatus over the text, but at the same time we wanted to ensure textual precision.

Beyond the theoretical underpinnings, the challenges we faced were also quite practical: while we have both used technology in various forms for our teaching (such as learning management systems, presentation tools, group communication tools) and recognize its pedagogical potential in the classroom, neither one of us had extensive experience with digital publishing platforms. Furthermore, neither of us knew or was prepared to learn website-building languages such as HTML, CSS, or Javascript. Given our own technological limitations, we began to explore what types of digital critical editions had already been published online, either early American or other literary texts. This research would give us a general sense of the scope of similar existing and ongoing digital projects, specifically critical editions geared towards college-level classroom use; it would allow us to ascertain the technologies that might have been used to create such projects (as well as provide us with valuable contacts); and it would provide us with a better visual sense of the parameters for our project — the kind of architectural, stylistic, and pedagogical opportunities, as well as the boundaries such various technologies might afford us.

What surprised us initially was not how many but rather how few digital critical editions we were initially able to locate easily and readily online, and in particular how few online editions we found with interactive, digitized annotations and hyperlinks, new textual models constructed with the contemporary student in mind. Extensive and invaluable progress has clearly been made in recent decades in the digital humanities. Well-known digital archives, such as Project Gutenberg and the Internet Archive, have done immeasurable work digitizing texts and making them available online for teaching and scholarship purposes, but the purpose of these important projects has been primarily to archive these texts rather than to annotate or edit them. More recently, smaller non-profit projects have contributed critical work in this area. Just Teach One, sponsored by Commonplace: The Journal of Early American Life, is one example of a digital recovery project that is more specifically suited for the classroom, as it not only offers digital transcriptions of texts (in this case “neglected or forgotten” early American texts) but also annotates and edits them. While Just Teach One and projects like it provide editions with a basic apparatus, they are still not “digital critical editions” in the sense that they are digitized PDFs, which is to say that they do not offer those “integrated technical tools” Sahle characterizes as an essential part of such digital editions [Sahle 2016].

Many other digital projects are working on that same important process of recovery and/or preservation but also offer various levels of dynamic integration with those critical “technical tools.” Other sites, such as Digital Library of Medieval Manuscripts and American Transcendentalism Web, as well as single-author archival projects, such as the Willa Cather Archive, The Walt Whitman Archive, Jane Austen Manuscripts, Digital Thoreau, and The William Blake Archive, are representative of the type of digital textual scholarship and digital reading experiences being created online. Each of these sites offers distinctive digital models with unique features particularly valuable for those starting out in the digital humanities. For example, Internet Shakespeare offers the ancillary material (an Introduction, Critical Survey, Bibliography, etc…) fundamental to a critical edition, various versions of each text, and the type of digitally-integrated annotations we planned to include. The William Blake Archive integrates multimedia to recreate exhibitions of Blake’s work. Digital Thoreau makes manifest a “fluid text” edition of the seven existing manuscript versions of Walden. While each of these sites, and the majority of those we examined, tend to be geared towards micro-literary communities, we did identify other interdisciplinary projects, such as The Vault at Pfaff’s: An Archive of Art and Literature by the Bohemians of Antebellum New York, sponsored by Lehigh University, which provide a cornucopia of primary and secondary sources, a digital arcade of hyperlinked sources, but which are focused on a very limited theme, in this case the fascinating 1860s New York City bohemian world of Charles Pfaff’s Manhattan beer cellar. Finally, we discovered some for-profit digital publishers, such as Touch Press, which work with publishers to create digital content; Touch Press's version of T.S. Eliot’s The Wasteland includes notes, various audio recordings, and critical and dramatic interpretations of the poem.

Each of these projects was valuable in helping us to conceive of possibilities within the framework of digital publishing; we did, however, learn two valuable, interrelated lessons in our initial investigation. First, the sites we found most readily tended to be geared to scholarly rather than critical editions of work. Digital Thoreau, for instance, is more thoroughly focused on what Kathryn Sutherland calls “the establishment of the text, its variants and transmission history” as opposed to the interdisciplinary, multimedia critical edition we hoped to create. As discussed, scholarly editions of Crèvecoeur’s Letters of an American Farmer currently exist, and we were not looking to create another. Rather we sought to create an open-source digital edition for the student for whom those unmediated, scholarly print editions were intimidating and difficult to penetrate. In short, although the sites we found were helpful in showcasing the array of options the digital might provide (integrated multimedia, interactive comment tools, hyperlinked annotations), the projects were far larger in scope (e.g. exhaustive single-author archives) and seemed framed with a different audience in mind.

The most important lesson we learned from this hands-on, primary research, however, was not what we found but rather how hard the material itself was to find. We were struck by the fact that these projects were not catalogued anywhere; instead, they seemed to float in separate silos rather than being gathered together in a single or even several classified repositories. To find a print critical edition, one can search on Amazon or ask a local book seller; to find a digital critical edition, one either needs to be a scholar in a specific literary field or have some similar entry point. There are online directories, such as the Omeka Directory and the Open Access Directory, which gather all types of digital sites and projects, but they do not specifically identify critical editions as such. By contrast, we had hoped to locate a repository of digital editions much in the same way one can go to Project Gutenberg to locate a collection of unedited electronic texts.

Thus, in addition to the challenges already named, we soon realized that if we were to move to the digital, one of our challenges would be to figure out how to make our open-source edition available to those beyond our own students. If well-resourced projects such as Internet Shakespeare Editions or Jane Austen Manuscripts were not immediately evident to us as professors of English and newly-inquiring digital humanists, how could we ensure that our edition would ultimately be generally available to students and other educators alike? We began to recognize that we would also need to navigate the important issue of a repository, a digital cooperative where readers could go to access the edition itself as well as other digital critical editions being done in the shadows. As a result, our search for an appropriate platform expanded to include one that could serve as a repository while also being readily available to others.

Initially, we had hoped to find an online press or nonprofit scholarly platform that was developing these types of digital critical editions and might be able to share information on emerging technologies appropriate to creating such an edition and potentially even hosting the Crèvecoeur edition we planned to produce. Although we failed to find that source in our early research, we discovered a promising publishing platform called “Scalar” when we attended a workshop entitled “Critical DH (Digital Humanities) Interventions in Scholarly Communications and Publishing” at the 2015 Modern Language Association (MLA) Convention in Vancouver, Canada. Scalar describes itself as “a free, open source authoring and publishing platform that’s designed to make it easy for authors to write long-form, born-digital scholarship online. Scalar enables users to assemble media from multiple sources and juxtapose them with their own writing in a variety of ways, with minimal technical expertise required” [About Scalar]. We had some basic concerns with the publishing platform, specifically that we would be constructing this edition using a technological infrastructure that was uniquely built by Scalar and housed by the Alliance for Networking Visual Culture. Once we designed the edition within the framework of Scalar’s platform, we would not be able to move it. Should its funding source fail or should we find a more appropriate platform, we would have to start from scratch. Given the 75,000-word length of Letters, this prospect was daunting.

That said, in all other respects Scalar was almost exactly the type of platform we were looking for. Like WordPress the program was user-friendly, but unlike WordPress, Scalar was designed by academics for the specific purpose of developing innovative digital online scholarship, and as such each new project was appropriately and reassuringly called a “book.” The open-source platform had partnered with various humanities centers that permitted its authors to integrate media more seamlessly, and its design allowed for a clean but flexible and varied display of interactive digital features — from basic notes to internal tags and paths to annotated audio or visual images to reader comments. Given that our initial research pointed to ongoing work in the digital humanities and, more specifically, work on digital critical editions that was still quite piecemeal and dispersed, Scalar seemed like an excellent option: the project would be housed on an academic platform whose mission was linked to scholarly work within the field of the digital humanities, and it offered us the technology we needed without the immediate requirement of technical expertise.

Before committing to Scalar, however, we spent some time looking at other platforms, such as Omeka, an open-source web publishing platform. Two major differences made us realize that Scalar was clearly the better choice. First, Scalar describes itself as a platform for those who want to publish “book-length works” [About Scalar], and Omeka specializes in hosting “digital collections and…media-rich online exhibits” [Omeka]; since we were focused on creating a book-length work, Scalar was ideal in this regard. Second, the free Omeka Classic version requires users to have an external server to host their material, while Scalar offers a server as part of their platform. In short, Scalar is self-contained and free, and for those who are beginning in the digital humanities, this makes it a more user-friendly option. Finally, however, and most importantly, we didn’t really find Scalar; Scalar in essence came to us. For two academics who “fell into” the digital humanities and had a specific project in mind, being able to engage with Scalar at an MLA workshop, hands-on, with other academics, and having access to Scalar representatives with whom we could keep in touch to ask specific technological and strategic questions, this platform was clearly the right place to start.

Once we had identified a suitable platform and began to build the edition, some strategic questions related to annotations and hyperlinks became apparent. The issue of hyperlink “stability” is, of course, one that is repeatedly discussed when developing online materials. For example, in “URLs Link Rot: Implications for Electronic Publishing,” a study of articles published by Emerald Publishers between 2008 and 2012, D. Vinay Kumar, B. T. Sampath Kumar, and D. R. Parameshwarappa “found that 48.53 per cent of URL citations were not accessible and the remaining 51.47 per cent of URL citations were still accessible” [Kumar, Kumar, and Parameshwarappa]. A related issue involves open-source accessibility. We had decided to use the Oxford English Dictionary (OED) as our source for annotated definitions; we not only wanted to offer contextually-appropriate eighteenth-century definitions but also to include the link to the original online source so that students could peruse the various definitions further as well as other information, such as the word’s pronunciation, etymology, and origin. The problem is that the OED requires an account, and we expressly wanted to avoid subscription-based sources and the need to sign in to retrieve annotation links. As a result, we chose the less ideal course: to construct our own definitions and provide a link to the online Merriam-Webster Dictionary for ancillary information. Similar challenges arose regarding historical, philosophical, scientific, and other related hyperlinks. The contextual relevancy, the scholarly value, the factual accuracy, and the stability of the link were of primary importance. More largely, however, we wanted to build a conscious plan about the types of sources we privileged. What open sources would count as scholarly enough? Would the History Channel, for example, be a source academic enough for college-level students? If so, would we want to try to use that source as consistently as possible? Would it be better to locate the most suitable source for any given textual reference, or might the edition then seem like a random hodge podge of hyperlinks without a specific purpose, targeted audience, and directed coherence? This issue was compounded by exactly what had drawn us to this project in the first place: a sufficient number of the references in Letters from an American Farmer were abstruse enough not to offer us a wealth of dynamic online reference options.

There were other challenges as well that involved not only strategy but time: what became immediately evident was that this digital project was going to be far more time consuming than our initial plan to construct a print edition. In addition to struggling through the ins-and-outs of Scalar’s architecture and learning the various facets of its technological possibilities (and limitations), every attempt to insert an annotation, include a hyperlink, or embed media involved several steps and a careful system of organization. Given that Scalar’s infrastructure is not (unlike WordPress, for example) based on hierarchical relationships but has, instead, what it calls a “flattened hierarchy” (where every “node” is treated as a “page” on the same level, and the editor can connect the pages as desired), each new page would need to be given a careful title and description that would allow us to connect it to the text as desired and, as importantly, be able to relocate it as needed later amongst hundreds and (given the length of this text) potentially thousands of other “pages.” And once a page was created and the annotation or embedded media image or hyperlink was constructed, the page would then need to be internally linked to the appropriate word or phrase in the text itself. The academic and pedagogical process was only a portion of the work; the technological work was also a challenge, and without the help of graduate assistants or the support of grants, the project was going to take more time than we had hoped. More importantly, the question of scalability — which is to say the ability eventually to create other early American teaching editions and to create a repository for such digital editions — would prove daunting without locating other digital humanities partnerships.[2]

For this reason, we decided to pilot a portion of the project. Before we determined if we were going to dedicate what might be — given our teaching and other research commitments — a year or more of editorial work to finish the whole edition, and before we engraved all twelve letters onto Scalar and became locked into that platform, we chose to complete only a strategic selection of the letters first while simultaneously presenting what we had completed at national conferences and beginning to use this portion of the project in the classroom to start to get feedback from the reader’s perspective. We therefore began working on the four most seminal letters in the text: Letter I, which establishes the narrative context and framework; Letter III, which offers the most salient line and thematic discussion in the text: “What then is the American, this new man?”; Letter IX, which addresses the profligate lifestyle of plantation life in Charleston, South Carolina as well as an extended and powerful diatribe on the evils of slavery; and Letter XII, which presents our narrator and protagonist — once steeped in the idealism and reason of Enlightenment doctrine and confident that the American colonies were “the most perfect society now existing in the world” — now driven to despair by the chaos and danger of the impending American Revolution (Letter III).

Although we have made our Scalar digital edition of Crèvecoeur’s Letters public for the purposes of limited classroom use, conference presentations, and publications, the edition is very much a work in progress. For that reason, we attach the “live” link to the digital edition here — http://scalar.usc.edu/works/crevecoeur--letters-from-an-american-farmer/index — with a certain reticence. In his essay “The Book, the E-text, and the ‘Work Site,’” Paul Eggert helps to explain this reticence when he discusses the relatively “complete” nature of a print publication as a “unified piece of scholarship” — “every part of the volume enlightened by every other part, all of it seamlessly interdependent and unobjectionably cross-referenced, nothing said twice, all of it as near perfectly balanced as you could ever make it” — as opposed to the open-ended, never complete nature of the digital edition, in which errors can be made (or addressed) at any point [Eggert 2009]. For Eggert, the concern is primarily that such lack of closure “will prove an opportunity to drop [our] standards” [Eggert 2009]. One might equally point, however, to the anxiety with regards to digital publication of never attaining perfection or completion. We are still in the early stages of creating the edition (at minimum, we plan to have Letters I, III, IX, and XII annotated and replete with an Introduction, Emendations, and an extended Bibliography); however, we wanted to make the edition available for public commentary as a way to turn the project into a communal critical and pedagogical venture.

In many ways, our students have played a foundational role in this communal venture. Initially, our plan was simply to pilot the in-progress edition in our undergraduate and graduate classes in order to get a sense of how students engaged with our edition: to get feedback on their experience reading the online edition versus the traditional Norton print text and also on their experience navigating the various types of annotations and hyperlinks in the Scalar interface. As we began to contemplate how the edition might be made available and accessible for students, and as we began to pilot the edition in class, however, it became clear how students could use Scalar to become active participants in their own editorial work. As Brett Hirsch suggests in Digital Humanities Pedagogy, “the digital humanities is about learning by doing,” and as we ourselves learned by “going digital,” we wanted to explore how the digital might not only enhance the literary experience for our students but also enhance their own critical thinking experience as well ([Hirsch 2012], original emphasis). Thus, in what follows, we discuss the various ways in which we have integrated the project into the classroom: first by piloting our edition in the classroom and then increasingly asking our students to partake in textual ownership, by using Scalar to annotate their own versions of Letters and then ultimately by using the platform to curate their own texts and create their own digital editions using our edition as a model.

In my Early American Literature class — a 300-level undergraduate course that combined English majors with general education students — at Southern New Hampshire University (SNHU) during fall semester 2017, I had students read versions of Crèvecoeur’s Letters from an American Farmer in three different formats: portions of the text from the print Norton edition (which I had assigned for the class), portions of the text from our online edition of Letters that had been edited and annotated, and portions of the text from our online edition that had not yet been annotated. I asked for informal feedback, most importantly regarding whether they preferred reading the text online or in print; whether they preferred the online portions that had been annotated or left unannotated; and, more generally, whether they found the digital edition easy or difficult to navigate and whether the annotations and/or hyperlinks were helpful or intrusive.

Numerous studies have been done examining the effects of reading digitally versus in print, but the results have been far from conclusive. As Ferris Jabr observes in “The Reading Brain in the Digital Age,” most studies conducted before 1992

concluded that people read slower, less accurately and less comprehensively on screens than on paper. Studies published since the early 1990s, however, have produced more inconsistent results…. And recent surveys suggest that although most people still prefer paper… attitudes are changing as tablets and e-reading technology improve and reading digital books for facts and fun becomes more common.  [Jabr 2013]

Interestingly, I noticed that the preference for print over digital in my informal survey was clearly influenced by disciplinary major: it was my English majors who noted a preference for print over online texts, and I attribute this — given my personal knowledge of the English majors at SNHU — to their having grown up in a culture of print books; they tend to be the students who display in multiple ways (several of whom have tattoos of their favorite novels and poems, for example) a deep investment in the traditional culture of the humanities. Just as interesting, however, the majority of my general education students noted a specific preference for the digital and several mentioned the same reason: the prohibitive cost of print books versus the open access nature of digital books. Having required my students to buy the Norton anthology made them particularly aware, as one student said, that “books tend to be expensive.”

Another critical question involved the issue of navigating the edition itself. Jabr points to an experience many readers have when trying to recall where they saw something in an analog text as an example of one of the advantages of print books: “Both anecdotally and in published studies, people report that when trying to locate a particular piece of written information they often remember where in the text it appeared.” In addition, he notes,

An open paperback presents a reader with two clearly defined domains — the left and right pages — and a total of eight corners with which to orient oneself. A reader can focus on a single page of a paper book without losing sight of the whole text: one can see where the book begins and ends and where one page is in relation to those borders… …In contrast, most screens, e-readers, smartphones and tablets interfere with intuitive navigation of a text and inhibit people from mapping the journey in their minds.  [Jabr 2013]

It was for this reason that, in our digital edition, we decided to keep each chapter of Letters as a single document, rather than creating separate pages as one would in a book or in trying to mimic an e-book. At its most basic, this allowed students to scroll ahead and see how long the text was and how much more they had to read, something which is not as effective when clicking ahead through numerous, individual pages. This kind of grounding in terms of text length is one way that reading the digital text can be made reminiscent of reading on paper. One can also still orient oneself visually, since for the reader the single “page” chapter can be similar to a book page; instead of “eight corners with which to orient oneself,” the reader “can focus on a single page of a paper book,” even if that page is quite long [Jabr 2013]. While this grounding certainly proved effective with my students, one issue became paramount: without standardized pagination, we had no easy way to locate text for communal discussion in class. Despite the fact that the digital allows for searchability and the textual annotations provide visual landmarks in the text, there were no quick indicators to reference. This experience verified the need for those critical paragraph indicators as a way to maintain a shared sense of orientation, and it became clear that moving forward with the edition we would need to add these visual landmarks.

Beyond this issue of visual landmarks, one of our main concerns about Scalar was the visible presence of annotations in the form of notecard symbols and highlighted words or phrases, which could prove to be a distraction, snagging the eye and interrupting the reading process. Despite this initial concern, students overwhelmingly found the annotations to be “very helpful.” They noted that the annotations helped them “read faster,” provided context and definitions, and allowed them to stay focused. What was interesting, however, was that while students unanimously found the note-type annotations (notes linked directly to the text on the page) beneficial, they were more split on their assessment of the videos and hyperlinks (those links that took them to external pages). Some noted the videos as particularly illuminating while others found them to be distracting, taking them away from the flow of the text; they generally appreciated the videos and hyperlinks and wanted them to be available, but some suggested the links be offered, unlike the annotations, at the end of a section rather than integrated in the middle of the text itself. Interestingly, it seemed that while many students were accustomed to reading texts that contained visible hyperlinks and other signs of authorial “interference,” picking and choosing what they want to look into as they go along and ignoring the rest, others felt more compelled to engage with each interactive feature, which ultimately diverted them from the flow of the narrative.

Piloting the edition in class provided a couple of key indicators: it confirmed that the work we were doing on the edition itself was pedagogically valuable, and it offered some important feedback on how to improve the reading experience. What also became clear, however, was that — as valuable as students found the annotations and hyperlinks that we had created for them — the digital benefits of Scalar meant that students could move beyond being just readers of the text. The flip side of the “unified piece of scholarship” Eggert notes in print publications is that, with the digital, students could now interact with the text with more ease and explore their roles not just as passive recipients of the text but as interactive interpreters, as curators of the text [Eggert 2009].

It was this role of students as “interactive interpreters, as curators of the text,” that I chose to explore in my approach to the Crèvecoeur project in two graduate classes in early American literature. The first, in spring 2016, was “Hybridity/ies in Early America”; the other, in spring 2017, was “From New Netherland to New York.” In part, I chose a project format that took advantage of the advanced status of these students, but I was also interested in the communal potential of digital texts and seeing how this might be experienced with different groups of students. With a digital text, readers can actively participate in the construction of information, either creating new material or adding to material that already exists. In some cases, readers can edit the existing text (Wikipedia is probably the best known example of this), and the platform keeps track of edits so that the text becomes a living history of change and response; other texts allow for commentary alone, but the comments still become part of the user experience for readers. Using Scalar for our edition of Letters from an American Farmer meant we actually had multiple options available. New material could be added to an existing text; an existing text could be elaborated upon or commented on; and these elaborations could themselves be edited. The project as I envisioned it would engage students in the creation of their own, communal version of Letters, one that would span two different classes, and possibly more in the future.

Given the advanced skill level of the students, having them actually annotate the text seemed approproiate; however, I did not want students adding annotations directly into our working digital edition of Letters. Fortunately, one of the benefits of Scalar is that it is easy to create additional books from existing material. This meant I could create an independent edition of the text specifically for the students, which would avoid potential problems of having someone accidentally change or remove material from our original edition, or our having to eradicate material afterwards. At the same time, I told the students that at the end of the process I would review their annotations and those of the highest quality could be incorporated into our edition, giving credit in the acknowledgements for their contributions. Since Scalar assigns a number to each editor of a text, it would be easy enough to determine the creator of any of the annotations. Thus, the students were potentially participating in our professional project, doing “real world” work instead of work simply for a class, and possibly adding to their scholarly credentials. Having students do their own annotations also introduced them to digital humanities skills and a tool, Scalar, that could prove useful for their work later on, especially for those who planned to pursue an advanced degree.

For the first course in which I included the digital editing project, “Hybridity/ies in Early America,” I set parameters that would make the work manageable and keep the stakes relatively low, especially since it was my first attempt to have students use Scalar. Each of the students in the course was assigned a 500- to 750-word selection of one of the letters to annotate. I did not specify the exact number of annotations they had to create; instead, they were advised to be as thorough as possible without overwhelming the text with commentary. Because Letters was to be the final text in the assigned readings for the semester, we would read their annotated version instead of the print edition I would normally have assigned. Since each assigned segment would vary in terms of the number of annotations that made sense, the students were advised that their work would be assessed on the quality of the information provided and whether essential words, phrases, or concepts had been explained and illustrated. They were also advised to think of their audience as undergraduates as opposed to other graduate students as they made decisions about what to annotate and how to do so. Finally, in another form of communal engagement, I incorporated an in-class peer-review session, which took place the week before we were to start reading their annotated version of Letters. Students were paired off to review each other’s work and provide formal feedback: Were the annotations clear? Were they factually and mechanically correct? Did all the links work? Had anything important been left out? Students then had the subsequent week to make revisions.

In order to prepare the students for the project and give them as much time as possible to complete their annotations, I conducted a Scalar workshop during the second class meeting (we met once each week). During the workshop, each student was asked to annotate the same passage from Letters; we then compared the results to see what each person had decided to annotate, why, and how, so they could see the results of individual editorial choices. In addition to the hands-on workshop, I posted directions for the basic Scalar functions — how to create a notecard, how to create a web link — in the online “shell” for the course (Seton Hall University uses Blackboard as its learning management system). I started each class session by asking how the project was going and regularly included that question in my weekly email updates to the class. Because they were all working on the same text and with the same digital tool, I found that the students were able to help one another much of the time. I would come to class and one of them would be explaining how to credit a source or how to edit a video with Scalar. Thus, in addition to learning the challenges of textual editing they were also learning to work collaboratively, which is not the norm for literature scholars. In addition to giving the students familiarity with one aspect of the editing process and a useful web tool, my goal was to get them to think outside the box, to use as many different resources as possible in their annotations, and to be creative. They could include any material that would enrich the experience of reading Letters for others, but they also had to think about how much they annotated, especially given the potentially crowded viewing field in Scalar, with its notecard symbols and colored hypertext links.

Besides showing the students how to use Scalar, I had an opportunity to provide them with a professional perspective on scholarly editing. I invited Tiffany Potter, who had edited the University of Toronto print edition of another text we were reading for the course, Ponteach, or the Savages of America: A Tragedy, to a Skype interview with the class, which she graciously accepted. The students were able to ask questions about her editing process, about the kinds of challenges she had faced, and about how she had dealt with some of the problems they were facing: how much to annotate, how much to take for granted, and how best to explain complicated matters in a brief space.

When the time finally arrived to read Letters for the course, the students were able to do two things: first, they were able to better understand certain aspects of the text as a result of the annotations and to comment on their efficacy; second, they were able to compare their reading of the annotated as opposed to the unannotated parts of Letters (since they had not annotated the text in its entirety), which led to a productive discussion about the value of annotations and what they added (or not) to the reading process. Having a personal stake in this particular version of the text gave the students an additional investment in the reading and discussion. The project evaluations (I administered a separate one from the regular course evaluation) were uniformly positive. Some students were frustrated with Scalar, as one would expect with any new technology tool, but overall they were able to comment intelligently and even passionately about what it meant to edit a text; most of them agreed that it was something they could see themselves doing again. All of them agreed that they would never look at an edited edition of a text in quite the same way again.

I incorporated a modified version of this project into a subsequent graduate course in spring 2017, “From New Netherland to New York.” The primary reason for the changes I made was that I had several students in the class who had taken the prior early American graduate class. In order to provide a different experience for them, in this later course students were given the option to work on one of three different digital projects, with the annotation of Letters being one of those. In this iteration I expanded the amount of text to be annotated. Whereas the first class had been asked to annotate only a 500- to 750-word section of Letters, each student in the spring 2017 course who opted to do the annotation project was given a full letter to edit. However, rather than having this second group of students work on a new, “clean” version of the text, I had them use the same version of Letters that the spring 2016 students had annotated, adding yet another dimension of collaboration. Their first directive was to review the existing annotations made by the students in the previous class and make sure they were factually correct and that all the links worked; they were also required to document any changes they made and to explain the reason for any changes using the Comment function in Scalar.

Giving students options can increase their investment in the work they are asked to do, and the Letters project in the spring 2017 course bore this out. The students who opted for this project were exceptionally committed to it and to the quality of the material they added; they were also critical (in the best sense) of the annotations that had already been added by students in the spring 2016 course. This kind of digital text project generates a complex reading community consisting of 1) those who have read the text without being involved in creating the annotations but who have benefitted from the work others have done; 2) those who have commented on the annotations, thereby influencing future editorial decisions; and 3) those who have actually annotated the text. This latter group can then be divided yet again into those who annotate the text at a particular moment in time (i.e. spring 2016 versus spring 2017). Since Letters was an edition in progress, students had a chance to think critically about what others had chosen to add to the text, change what was done previously, and then add to it themselves. They thus became curators of the text as opposed to simply readers, and, hopefully, readers with a different relationship to the material. Students involved in this type of project also gain a new awareness of audience, those for whom the annotations are being created, and have to ask of their work: What do those readers need to know? How much information is too much? How can the information best be conveyed? The result is a text that is interactive in the best possible sense. The annotations reflect what students themselves thought it would be useful to know as they and other students read, and it provides information for readers like themselves about terms, historical references, and cultural references. This graduate student edition remains a text that can be added to and updated by other classes, providing a communal experience for future students as well. Most importantly, being able to have students participate in the annotation process gives them a different level of engagement with the text, and it can ultimately — we hope — make them more active participants in a scholarly community. Finally, if their annotations are included in our “official” online edition of Letters, the students can be understood to be participating in an even larger community of readers and helping to “author” a text that is organic in the way it is shaped and develops over time.

As a result of the various ways in which our students had both used the Crèvecoeur edition and participated in developing new digital content, I was encouraged to consider yet another type of digital exercise as a way to explore the possibilities of digital texts. As I was scheduled to teach an upper-level American Literature Seminar in spring 2018, I decided to use the opportunity to have my students engage in a semester-long project dedicated to constructing their own short digital critical editions using Scalar. I chose not to have students work on Crèvecoeur’s Letters from an American Farmer for several reasons. Most importantly, some students in my seminar had taken the Early American Literature course I had taught that previous fall when I piloted the Crèvecoeur digital edition. While the Crèvecoeur edition would prove to be a good model for them, I was concerned that if they were also asked to construct a digital edition of Letters, there would be a temptation to replicate elements of the edition we were producing; I wanted them to enjoy the freedom of a “tabula rasa.” In addition, my students were undergraduates, and I knew that the exercise itself would be particularly challenging. Given the difficulty of Crèvecoeur’s writing, I felt that choosing a more approachable early-American writer — both in form and theme — would be a better option. Therefore, I assigned them, instead, a choice of Benjamin Franklin essays. Over the course of fifteen weeks, my students were asked to incorporate their own selection of key Franklin essays into a digital edition that would include: digital annotations within the text (and related ancillary materials, such as maps, word clouds, videos, etc.); a Note on the Text and list of Emendations; an Introduction; and an Annotated Bibliography. What I realized was how quickly I had to adjust my expectations, not so much in terms of the digital but rather in terms of what my students understood about more traditional concepts associated with the print form (e.g. the critical edition and textual criticism). For example, while I was initially concerned about teaching my students the Scalar platform, I found they picked up the basics with ease. What they struggled with, however, was the very concept of what a critical edition is. In an age where raw, unedited literary texts are more readily available online, students seemed totally unfamiliar with the model of the critical edition and the editor’s role in textual production. Even more than the critical edition itself, students were utterly confused by the study of textual criticism. The notion that what they read is not automatically the “authoritative text,” that there may be versions that are more or less authentic, that an editor may have emended the text, and that as scholars their responsibility is to choose the most valid textual source: all of these ideas seemed foreign to them. When we began working on emendations, one student said: “Wait, I’m supposed to ‘edit’ Benjamin Franklin? I can’t play God. I can’t do this.” After repeated discussions about textual criticism, I was relieved to hear this “aha!” moment: they were beginning to realize the import of what I was asking them to do.

Given my students’ confusion and their own uncertainty, asking them to be “start-to-finish” curators of a digital edition was particularly unnerving. The idea that my students would be using Yale’s Franklin Collection (what we had chosen as the starting point for the “authoritative text”), and then — as undergraduates — emending, annotating, and digitally “publishing” that work, highlighted the concerns Eggert notes in comparing print and digital publishing forms and underscores the ease with which texts can be manipulated and endlessly altered online. That said, having my students create their own editions from beginning to end — creating a “Note on the Text,” emending that text and constructing an “Emendations” page, including an Introduction, providing annotations and hyperlinks, and offering an Annotated Bibliography — allowed them the chance not only to actively engage with and manipulate the text using the digital but also, ironically, to understand the traditional models associated with the print form. First, they were able to get a sense of the kind of detailed and pain-staking work that goes into textual criticism and, more philosophically, to ask about the ontological relationship between author and editor in the construction of the text. Second, they were able to locate that “state of prolonged anxiety” Eggert notes when trying to complete a print edition [Eggert 2009]. This project was not just another college exercise. Their work mattered. And I made clear that because Scalar was an “open source online and publishing platform” any edition they constructed for my class they could ultimately make public and even searchable for portfolios and jobs. They would be responsible for the work they produced, beyond the classroom. In many ways, therefore, this digital assignment gave students the opportunity to experience professionalism and the kind of high-stakes perfectionism that Eggert associates with print rather than digital publishing.[3]

Our various experiences with digital critical editions have made us even more convinced of their value, whether for scholarship or for teaching or both. The Crèvecoeur edition has proven beneficial for precisely those reasons we set out to produce it: it would not only make Letters from an American Farmer available to students and a larger public audience, but it would, as importantly, make the text accessible to that audience. In the process, the digital medium would avoid current cost concerns from the publisher and the buyer associated with the print market and open up new ways for the reader to engage with the text and its context through interactive technological tools.

As our experience has revealed, however, there are several caveats when it comes to starting a digital edition. For one, an edition needs a repository, a stable place for the text to reside. While we have opted for Scalar, and have confidence that the tool will be functioning for some time to come, there are still risks that a non-profit platform like Scalar might lose its funding or that platform support may be limited. Given this inherent lack of stability, it is important to have a strategy for dealing with changes in tools, platforms, and applications. It also seems critical that, moving forward, more repositories are created for such editions. As discussed earlier, archives do exist, although they seem either to be reserved for a single author (The William Blake Archive, for example) or the work of certain editors (Just Teach One, for example). No doubt there are some clearinghouses that we simply could not locate. Ideally, however, any repository for such digital editions will be easily located and easily accessible by and for public readership, in much the way that archives for unedited texts currently are, such as Project Gutenberg or the Full Text Archive. To be effective, we would argue, such repositories should also — by nature — be open-source.

Of course, the most important lesson we have learned from our own experience creating a digital critical edition is the time it takes to engage in these projects (and, of course, this might account for their limited availability). Beyond helping to define and distinguish the medium, Sahle’s discussion of digitized vs digital editions helps explain the complexity involved in their creation; “browsing paths,” “real hyperlinks,” and “integrated technical tools” all require — for their survival — the interdependence of other “real,” dynamic paths [Sahle 2016]. The nature of this interdependence means that the initial construction is time-consuming, as is its upkeep. While the original text may remain stable, the apparatus constructed does not and for this reason the result is a digital edition that is forever in flux.

This flux, what many critics note as its instability, is disconcerting; however, it can also be seen as a pedagogical and scholarly boon. The analog text is certainly more stable. Conversely, that stability comes at a price: it isn’t long before the analog edition becomes outdated, and what was the definitive version — with the most current theoretical essays and contextual material — is now out of step with contemporary and even cutting edge critical thought. The digital critical edition, because it can contain both past and present materials, by contrast, represents a host of perspectives across time. The reader can then see both where the text has been and where it is going, and contribute to those new directions. Thus, while this lack of stability creates fundamental challenges — increased initial time and continued labor and maintenance — its mutability also helps to ensure its continued relevance.

Finally, what we have found is the inherent communal nature — both for our students and for ourselves — in these types of digital projects. Ideally, the digital edition can reflect various perspectives. It can reflect the current cultural moment, it can retain historical elements, and it can incorporate the work of scholars at various stages in their learning. The edition incorporates, as Sahle says, “integrative technical tools” and thus is by its nature integrative; it integrates the perspectives of communities of contexts and communities of scholars [Sahle 2016]. Therefore, it is not only important to consider how much time such a process will take but to find good partners to work with. We were fortunate to have been brought together at that 2011 NeMLA conference, and that our interest and work styles are so similar. Not everyone starting a project will be so lucky. For now, we plan to finish the Crèvecoeur edition, continue to explore future digital humanities pedagogical opportunities with students, and look ahead to possible future collaborative projects, perhaps housed at a site of our own making.