Editorial Approach
A consideration of the EMVR’s potential interventions in literary scholarship prompts a problem of definitions: is the project fundamentally a digital tool, a work of scholarly editing, or something else? Though I have often thought of it (and described it) as a “reading platform,” in looking at the structure of the application, it seems only to partially fit the definitions of “tool” and “digital edition.” As a tool, the EMVR is capable of working with multiple texts and multiple authors. It is tool-like because the interface provides a unique way of engaging with texts, especially historically situated texts, within a visual context. Texts in this sense are interchangeable: the system in future iterations will be modular, with the program drawing on and loading any number of encoded texts as discrete files, with a large database of images stored separately. The EMVR (in this future form) simply takes the encoded texts and selects appropriate images from the database against which to juxtapose it based on user interaction. That core functionality, as well as its interface design, is what makes it a tool.
At the same time, however, the EMVR is the Early Modern Visual Reader, which means that it seeks to juxtapose specific texts and images from a certain period, a goal that requires choices that are somewhat editorial in nature (e.g. choosing keywords and asserting a certain reading of them based on context in order to select relevant images, as I explain in the methodology section). The EMVR is not, ultimately, a scholarly edition, especially in the traditional sense.[1] It is much closer to the hybrid “knowledge site” of The William Blake Archive (a useful comparison as both are concerned with pairing images and texts) as identified by Philippe Régnier (68). As Régnier notes, “The inclusion of a biography, a glossary, and a chronology, as well as critical articles and a kind of discussion forum, and putting specific tools at the disposal of visitors, make [the Blake Archive] at once a scholarly edition, a special library collection, a museum, and a research center” (68). As I will discuss in Chapter Five, there are certainly editorial choices being made in the EMVR now, even if it is—like the Blake Archive—something other than a scholarly edition of a text or texts. Even if in the future some version of tagging may automate parts of this process—by allowing images to be paired automatically with parts of the text—scholarly editing work is always going to be essential for encoding literary texts and for the process of selecting appropriate images. We might look, then, to digital editions as a model for how to approach the theoretical work involved in encoding texts and selecting images for the EMVR. We should not forget that for a digital edition “editorial judgment is still what the editor has to offer, no matter what tools are available. The editor’s voice should come over strongly” (Haugen and Apollon 56). With this in mind, future versions of the EMVR will seek to establish editing guidelines drawn directly from traditional scholarly editing practices, where applicable, and scholarly oversight of the functionality of the application will remain a part of its design.
What other lessons might the EMVR take from digital editions? Creating digital editions has been a central area of practice for the digital humanities since it began. A digital edition is not simply a digitized text—instead, it is specifically a text that supports research and/or is the result of research (Nyhan 118). A digital edition has been tested for accuracy as a part of the process of digitization. Working on the Edith Wharton digital edition, to use an example from my own experience, meant ensuring that the text of House of Mirth was completely accurate by carefully entering and re-checking each chapter of the novel. This focus on accuracy is one thing that sets a digital edition apart from other digitized texts.
In addition, it is widely agreed that a digital edition should have several core goals. First, it should “provide the essential data that the scholar needs, such as bibliographical data about the base text” (119). As for the EMVR project, it simply draws its texts from scholarly editions, but it does make the source clear and accuracy is key to its function to facilitate scholarly exploration and research. A digital project should also “describe editorial decisions and workflow.” With my following chapter set to explain this process and ultimately to be made available in some form on the EMVR website, it seems that the platform fulfills this expectation of disclosure. The methodology behind the project defines limits, based on historical research and the best practices of similar scholarly works, within which images may be selected and connected to words and phrases in the text. Finally, scholars seem to agree that the EMVR should “use non-proprietary and Open Source technologies and software” for the purposes of future-proofing and transcoding mentioned in the previous section, as well as accessibility concerns. The EMVR achieves this goal, as well, using only pure HTML, CSS, and JavaScript, without any external libraries, specifically to make the platform as future-proof and accessible as possible. Unlike dependencies such as Flash, Quicktime, or Real Player, the core capabilities of the browser, while not truly evergreen, are much more stable. JavaScript is only growing in popularity and now being used for full-stack development, with a majority of developers now recommending the use of pure JavaScript whenever possible. At the moment, the EMVR does not utilize TEI markup, because its only focus in terms of identifying discrete units of text is placing a single <span> tag around key words and phrases that will be tied to images, but it would be possible to incorporate TEI into the project easily to further enrich the text.
Nyhan notes three more expectations for a digital humanities text. First, it should be made clear how the core theory that underlies the work should be expressed in its design. As mentioned above, the EMVR’s theory and methodology section will provide this information: the platform is built specifically to communicate the value of relationships between text and visual culture, and encourage exploration of that relationship, in its user interface and user experience (UX) design. Moreover, a digital edition should enable work that could not be otherwise done without digital tools—in other words, digital editions should be born-digital scholarly works. Conceptualized as a digital platform, the EMVR moves farther from print than a “standard” digital edition (if there is such a thing): the work of juxtaposition it enables relies heavily upon the resources and affordances of the digital environment. The EMVR’s born-digital nature positions it differently than print texts for access and exploration by a wide variety of users. One might imagine a series of print volumes that could provide similar juxtapositions of words and images, but these books would still not approach the capabilities of the EMVR, which capitalizes on numerical representation, modularity, and variability as an iterative and extensible platform. It is also networked and accessible for the purposes of Nyhan’s final point, that digital editions should interact with other digital texts and digital ecosystems. When the EMVR moves to modular textual ingestion, encoding those texts with TEI will make them machine-readable for the purposes of interfacing with other scholarly digital projects.
In conclusion, because the principles of digital media inform a new environment for scholarly projects, most digital humanities projects are—not unlike e-lit works—“hopeful monsters,” in that they must navigate these principles as well as the goals and expectations placed on scholarly work. In other words, they often must find their own footing, but they do not do so in a vacuum. The EMVR may be at its core a new digital tool, but it shares many goals with scholarly digital editions of texts, projects that have already been grappling with the affordances and limitations of the digital space for born-digital scholarship. Attending to those goals and noting how digital editions are conceptualized as born-digital scholarly works provides a framework for understanding the EMVR’s contribution to literary studies, the space it occupies as a digital humanities project, and its limitations as well as its advantages for scholars and other users.
Continue to Methodology.
[1] Interestingly, Haugen and Apollon assert that digital editions are not fundamentally new—the technology may enable new methodologies in the future, they acknowledge, but as of yet the only thing that has changed for the production of critical editions between print and digital is the nature of the tools, not the nature of the work itself (56). While I agree with this premise to an extent, things may already have shifted more than Haugen and Apollon are ready to acknowledge: the decidedly nontraditional Blake Archive won a Modern Language Association prize as a “scholarly edition” over fifteen years ago (68). I do see more projects like the EMVR (ones that incorporate editorial choices with other approaches and methodologies) emerging based on the accessibility and prevalence of experimentation with new kinds of digital scholarship. I think of the experiments in Vectors, in which the capabilities of Flash drove innovation as established scholars considered new ways to approach the work of scholarship altogether.