Authored by: Marilyn Deegan, Department of Digital Humanities, King’s College London
Textual criticism and scholarly editing are fundamental activities in the text-based humanistic disciplines and are practices that have their origins more than two millennia ago. Texts have been copied, interpreted, annotated, discussed and analysed in the search for meaning, and every generation of scholars brings new tools, techniques, perspectives and interpretations to textual artefacts: that texts that have been worked on for generation after generation (like the Bible and the classical texts) can bear the weight of so much meaning is astonishing.
The advent of the digital age brought new tools to the service of scholars. Initially the computer was used to make easier and faster many of the component tasks of editing, with the final output remaining a conventional print edition. In 1946, for example, the great pioneer of digital scholarship, Father Roberto Busa, approached Thomas Watson of IBM to ask for his help in the analysis of the works of Thomas Aquinas (one and a half million lines, nine million words). Watson replied that the computer could not do what Busa was asking: it was a calculating machine not a word analysis machine. Busa quoted to Watson IBM’s slogan “What’s difficult we can do straight away, the impossible takes a little longer” and Watson accepted the challenge. According to his obituarist Stefano Lorenzetto (Busa died in 2011 at the age of 98), ‘Father Busa breathed into it [the computer] the gift of words. This happened in 1949.’
Text analysis remained a primary activity for computing humanists, and gradually other functions that we now take for granted were adopted: word processing, text searching, electronic communications, electronic typesetting, electronic publishing, etc. Not all of these developments have been welcomed as benefits by scholars: publishers increasingly request electronic submissions to journals, and camera-ready copy for editions; some of the efficiency gains offered by the tools are offset by the time and effort expended fulfilling these new requirements.
In 1965, Ted Nelson propounded the theory of hypertext as a model for creating and using linked content (he called this ‘non-sequential writing’), and this was taken up in the 1970s and 80s, with early hypertext systems (Hypercard, Guide, Intermedia) being developed. The increasing capacity of the hardware to handle images, sounds, and video, and Tim Berners-Lee’s realization that if you add hypertext to the internet you can exchange documents via external links led to his invention of the World Wide Web.
At the end of the 1980s, scholarly editors began to think of the use of computer tools not just as means of producing printed texts, but as means of displaying editions electronically. For what is an edition if it is not a hypertext, a complex web-like system of linked transcriptions, variants, glosses, notes, and all the other apparatus we associate with the printed form? Non-sequential writing, indeed! Textual critics like Jerome McGann, Kathryn Sutherland, Peter Shillingsburg, Peter Robinson and others began both to theorize about literary works and editions as hypertexts and to develop hypertext and multimedia systems around these ideas. Initially, ideas (as they do) ran faster than practical possibilities. The fluidity of the electronic medium was seen as a benefit: how wonderful to be able to spot an error and correct it instantly, to be able to add a reference, to constantly add new transcriptions or images to online editions, to interlink internal and external referents and create complex paths through the materials. Peter Robinson proposed a new model of editing where the edition is made by the reader from whatever is available, the reader determines what is read and how it is presented, the reader controls the choice of materials and anyone can alter any word and invite others to read the altered text. He called this model ‘fluid, collaborative and distributed editions.’ (P.M.W. Robinson, ‘The Ends of Editing’, Digital Humanities Quarterly, Summer 2009, 3 (3), http://digitalhumanities.org/dhq/vol/3/3/000051/000051.html). But not all scholars see this as adding value. One of the key benefits of a scholarly edition is its stability of citation. As Gordon Campbell pointed out (‘Scholarly Citation and the Value of Standard Editions’ blog post on this site, 15 March 2012), standard editions are absolutely crucial for other scholars and students: accurate quotation matters if they are to be able to interpret the works based on accurate renditions of the texts.
In the late 1980s and early 1990s, there was also much enthusiasm for texts with complex interpretative interlinkings and the addition of sound, images, video clips etc. If the technology could do it, we were trying to use it! Links were hardwired into editions. Before the widespread adoption of the web and the sophistication of web-based programs, these editions were produced on CD-ROMs—a nightmare for libraries to deliver to readers, and very difficult to update. When an operating system was upgraded, the CD often no longer worked. Such was the fate of the excellent and expensively-produced CD Word, which appeared in 1990 and ran on Windows 2.0. CD Word was the first digital library of serious Bible study tools with commentaries, lexicons, support for Greek, Hebrew, and extensive hypertext linking. It was reported to have cost $1 million to develop, and ran wonderfully well under Windows 2.0 and 3.0, but was unable to run under later versions and wasn’t upgraded. This brings up a crucial issue: developing complex functions makes editions much harder to preserve for the long-term, as does developing on one stand-alone platform. Another feature that editions must have is durability, something print has always achieved. Take for example R.W. Chapman’s 1923 edition of The Novels of Jane Austen published by the Clarendon Press. In 1965-6 the so-called ‘third edition’ was revised by Mary Lascelles and continues to be reprinted. But the text has not changed. Set in 1923, it continues the same through all subsequent editions. Page numbers have never changed and citation remains utterly stable.
Electronic editing has been much discussed over the last twenty years, but it remains true that there has not been widespread adoption of electronic practice by editors. And what scholars want, we have found over the years, is something that gives us all the benefits of print, with many of the benefits of the electronic. So we want well-edited, reputable texts of authors with scholarly apparatus, notes, etc, as well as easy access, searchability (and cross-searchability), good bibliographic references, and we want critical mass. We also want it to be easy to use (we do NOT want to become computer experts, just as we did not want to become typesetters). Offering online versions of existing printed texts gives us all these benefits: we still have what print does best, and we also now have what the digital does best – in short, Oxford Scholarly Editions Online is what scholars have been waiting for over the last twenty years.