The Argument for DigiTextual Pedagogy
Despite the growing prevalence of digital reading technologies, university literature courses still largely hold the codex as king and traditional papers as the loftiest form of critical engagement. In an era when pedagogical theory insists, with clear benefit, that instructors “meet the students where they are,” the continued teaching of traditional scholarly methods problematically ignores this pedagogical dictum and demands that students use increasingly antiquated technologies of textual engagement, ones which are admittedly foundational for but lack clear demonstrable application to the majority of the reading and analysis they perform in their daily lives. This suggests an inappropriate division between students’ textual engagement in academic and non-academic settings.
This continued reliance on traditional, hard-copy texts (and, thereby, traditional methods of reading and criticism) is particularly problematic when we consider the rapidly-changing reading methods people commonly use outside of university classrooms. Despite the complaints about dwindling sales of printed artifacts (newspapers, books, magazines, etc.), people read and write more now than they ever have in history. From social media to electronic news sources, from email to comment sections (increasingly found on websites from The New York Times to Merriam-Webster), and particularly evident in the nigh-replacement of audio phone conversations by texting, there has been an explosive proliferation of the written word, and this explosion is primarily digital. But, as Christian Vandendorpe notes, the methods of reading in these new spheres is inherently altered by their paratextual elements and technologies of representation and navigation. Drawing on Mark Heyer’s “The Creative Challenge of CD-ROM” – in which Heyer breaks reading down into Grazing (slow careful reading of a single text), Browsing (skimming over text, looking for snippets of interest), and Hunting (looking for specific pieces of information) – Vandendorpe notes that digital reading favors the latter two. He suggests that the former, Grazing, is still primarily a codex-heavy practice, largely because of “the rigidity of the screen and the technological barriers posed by the mouse and keyboard” for navigating large swathes of text. First, while this is currently true, that does not mean it will remain that way forever; second, and more importantly, we need to establish what it is we intend as the take-away learned knowledge or skills for our students. The statistics from Nielsen show that most long-form texts which are sold by booksellers are hard-copy, but these are also (according to Nielsen) primarily long-form fiction – that is, novels. Yes, we (some of us) often use novels as the primary texts in our English courses, but it is problematic if the take-away from our courses for students of engineering, business, law, and medicine is how to read a novel critically. If this were so, then surely we should use the form of text which students will most commonly read thereafter, and stick to the codex (or do so 70% of the time, to be literal about it). But if this were so, then the fate of English departments is even more grim than people currently suspect, as teaching people how to do a leisure pastime well is more appropriate courseware for public community centers than universities. Ideally, as a discipline in the humanities, English courses should rather teach logic and critical thinking through analytic approaches to reading and writing text –all text, any text, be it fiction, poetry, drama, legal proof, memorandum, or text message. And, as previously noted, the vast majority of the text people currently read and write in the present moment – text which, to whatever degree, needs to be read and responded to critically and analytically – is NOT long-form fiction. Thus what we are (or should be) doing is teaching textual engagements (reading and writing) for those forms and technical environments which students will primarily encounter; and as these texts are primarily going to be electronic, it only stands to reason that we primarily teach electronic textual engagements.
And electronic textual engagements are undoubtedly different, requiring a different familiarity and different skill set. Putting aside all differences paratextual (web browser “framing” vs. book margins and covers, the tactile physicality of the codex, etc.) and navigational (keyboard, mouse, and find functions vs. flipping back and forth through pages), the hyperspace environment changes the way text can be read. Hyperlinks, in particular, provide rapid movement between texts; but even when hyperlinks are absent, additional browser windows can be opened for rapid research using hypertext reference resources (dictionaries, encyclopedias), critical commentary (“I wonder if there are any articles on sexism in the ‘Molly’ chapter of Ulysses”), or general clarification (typing in “32 feet per second per second” in a search engine). These windows can be positioned adjacently or layered palimpsestually on the same screen, wherein the user can leap back and forth from one layer to the other with the immediacy of an alt-tab stroke. The theoretical effects on the semiotics of this have been well-examined – e.g. Jerome McGann’s “radiant textuality,” Ananda Mitra’s “reciprocal intertextuality,” and Kevin Kelly’s imagined “universal library” where individual books essentially lose their insulation and the textual space, not individual texts, is what one reads.
But the practice of using the internet to access reference materials is not limited to digital texts. With any computer linked to the internet, one can perform any of these activities just as easily while reading a codex alongside; or one could read a codex and then pull out another codex (a hard-copy encyclopedia, for instance) and look up information. What digitized texts do is speed up the process, and thereby make the process easier. Digitized texts do not create this new method of reading, they enable it. One is less likely, while reading the epigraph of T.S. Eliot’s The Waste Land: 1922, to move out of a codex-published form and look up (either physically or digitally) who Petronius was, what the plot of the “Cena Trimalchionis” is, a translation of the Silver-Age Latin, or critical commentary on how the ostentatious parvenu Trimalchio relates to the rest of Eliot’s poem, if only because such research requires putting down one form of media and picking up another, and thus takes greater time and effort. Even if a digital form of the text is not hyperlinked to such information, a reader does not need to remove her hands from the keyboard and mouse to open up new browser windows or tabs and access this information with amazing rapidity, and then immediately move back to Eliot’s text and continue on. There is no medium-change, there is almost no time delay in access, and the consequent speed makes such research mid-text more appealing for the reader. Hyperlinks further accelerate this practice, and thereby further encourage extra-textual reading movements.
Many DH theorists note that this change in reading methods can present problems. In Humanities Computing, Willard McCarty notes what he terms the “danger of distraction” (104): when the reading act is repeatedly broken by jumps into external texts, it can break down the progressive building of meaning in the primary text; and, potentially, a reader may abandon the primary text mid-way through for indefinite amounts of time while she follows link after link in subsequent pages down a rabbit hole. (I am reminded of just such an occurrence I recently experienced, in which I left Ulysses in search of a photograph of Nelson’s Pillar and ended up, half an hour later after numerous logical movements, investigating the differences between Arabica and Robusta coffee beans.) However, this could easily be understood not as a problem in hyperspatial textuality, but as a more intertextual understanding of “text.” Is the “text” I am reading Ulysses, or is it a blending of meaning drawn in pieces from Joyce, then Britannica on Nelson’s Pillar and on the Dublin Spire, then Google Maps on the location within the city, then back to the Pillar, then contemporary actions of the British navy… and finally to financial effects of early coffee trade and the differences between the two kinds of beans? It is, undoubtedly, a different kind of reading than sitting in a coffee house reading Ulysses one sentence at a time without break, but this is less of a “problem” and more of a paradigm shift in the way people read – similar to, as Vandendorpe outlines, the way methods of reading changed dramatically in the movement from scroll to codex: portability (changing where one could read), tables of contents and indexes, punctuation and spacing (allowing easier silent reading), all changed the way people stored, organized, and retrieved information from these two forms; and, I would add, pagination makes slow reading over extended periods of time dramatically easier, as saving one’s place in a scroll progressively over a series of days or weeks is incredibly difficult.
Furthermore, if this is the kind of reading practices students are engaging in most of the time, and if this is a potential danger of this kind of reading, then this should be a subject for discussion in university literature courses. Students in this setting should, quite specifically, be thinking critically about their own methods of engagement with texts. That’s (theoretically) why they are there.
Bertrand Gervais also argues that hyperlinks break down the common hermeneutics of the reading process, from a “logic of discovery” (“truths stemming […] from a quest for information”) to a “logic of revelation” (“having [truth] revealed without any effort”) – the former an active reading engagement, the latter a passive absorption of pre-dictated relation which calcifies the much-needed slippage in the signifier/signified relationship.
Gervais’s argument, however valid it is conceptually, is flawed functionally. First, Gervais assumes that a reader will necessarily follow a hyperlink, and will think of its linked text as the definitive meaning rather than a suggested meaning. Neither of these are given. Second, if a reader comes across material which she does not know and wishes to find some definitive material for it, the kind of movement to do so is the same. If she comes across the word “autotelic” in a Norton anthology, a gloss will likely be provided, which is very similar to a link which provides some definition thereof. If no gloss exists, she will (if she cares to “discover” meaning) look in a dictionary, and this movement of “discovery” through finding and using a dictionary, albeit slightly more laborious and time-consuming, is not necessarily that different from a hyperlink to a dictionary definition or merely typing the word into a Google search. The hyperlink “reveals” meaning, but so does the OED, and both only do so after a reader decides she wants to “discover” the meaning there. All of these could be argued to be discovery, all could be argued to be revelation. The only difference is who is providing the information and how hard it is to acquire it.
Hence my skepticism toward concerns of excessive “authoritative/editorial control over interpretation” through hyperlink annotations: it is temporally, not theoretically, different from a non-embedded text. Someone coming across the word “swyved” in Chaucer’s “Miller’s Tale” can easily look up what that word means in any number of dictionaries in any number of media-forms (a person’s mind, of course, also presenting a sort of dictionary). Or, with the Broadview Anthology, one could look at the gloss stating it means “made love to.” In a digitized version, there could be a hyperlink on this word pointing to the OED page for the word or giving an OED definition – we’ll stick with the euphemism and go with “copulated with” for sake of parallelism. The important part of all of this is that, though the gloss and hypertext undoubtedly provide an easy answer which a reader is much more likely to rely on, in no way is the reader limited to accepting this definition as the definition. Such glosses are suggestions by the editor, not commandments from some transcendental signified on high. Hyperlinks are no different.
These are just two examples among many, but the point is these are the very kinds of discussions which should be taking place in university literature classrooms. For better or worse, this is the kind of textual engagements students are and will be using in their personal and professional lives, and thus practice with and discussions about digital textual engagements – indeed, including discussions of the theoretical problems and implications of digital textuality – should be deeply ingrained in the pedagogical methodology of English courses. I am still rather unconvinced that long hyperspace texts are not potentially going to replace codex texts (particularly with inventions like tablets to increase mobility and E Ink’s “electronic ink” to reduce eye fatigue); but even if the codex forever lives on as the method of reading long-form texts, the codex still isn’t (and isn’t likely to be again) the primary medium for reading or writing, which means that English pedagogy has a responsibility at least to incorporate digital textual engagement into its course objectives. It is becoming increasingly clear, as Peter Shillingsburg argues, that “scholarly editions should be constructed and published electronically” (82).
 I would rather argue that it has more to do with eye fatigue from illuminated screens and imperceptibly-fast flickering from screen refresh rate, but either way, the point stands that full-length codices are preferred over a digital counterpart – Nielsen’s report shows that e-books only represented 30% of book purchases in 2014 (2).
 For an argument which relies heavily on intertextual theory, I should note that my use of “hypertext” herein always refers to digital texts in a hyperspace environment rather than in the Gerard Genette sense of relation to a “hypotext.”
 The apparatus to critical editions can function similarly, but for longer explanatory materials (e.g. a plot summary of the “Cena Trimalchionis” and explanation of Trimalchio’s character therein), this material is too long for footnotes, thus requiring readers to investigate special sections in the codex to find this information. Particularly for heavily intertextual High Modernist works, this would make a critical edition excessively long (Ezra Pound’s Cantos would likely quintuple in length) and thereby more difficult to navigate quickly.