Mary Wack, Professor
Department of English
Washington State University
Last year I taught an experimental course nicknamed “Electronic Chaucer” that may offer a glimpse of the sorts of resources to be found in Chaucer classrooms in 2001. I’d like to briefly indicate some of the electronic tools I used, including an image archive, and then discuss some of the issues of publishing such an archive.
The undergraduate seminar in the Canterbury Tales made use of a variety of electronic tools available over the Stanford computing network to supplement our printed text and conventional printed resources. The course was taught in a networked classroom in the undergraduate library, where 18 Macs were linked to each other and to the university network. The tools we used over the network included a text search and concordancing program known locally as “The Searcher,” an on-line version of the OED, and the MLA Bibliography and Art Index on networked CD-ROMs. These were prepackaged tools: I did no development of the software or databases. What I did develop was a prototype of the Stanford Humanities Image Archive, a database of digitized images (or “imagebase”) with associated texts. This became an interactive resource for the classroom, a research tool for the students, and a project that they contributed to through their own research and writing.
My use of computer tools to teach Chaucer grew out of specific pedagogical problems that I faced. Though I have always felt that computer technology will inevitably transform teaching and scholarship, I felt no personal call to pioneer methods of using computers in the classroom. Like St. Augustine, I said: “O Lord, let me learn Hypercard–but not yet.” Then a series of impasses in my teaching led me to wonder whether computer technology could offer a way out.
One problem was large class size: I often lectured to classes of 100-200 students. Years of running these courses in the standard way left me dissatisfied with the passivity that large lecture courses seemed to foster, and with the logistical problems of encouraging undergraduate research with so many in the class. The library couldn’t handle presentations on rare materials or research tools for that many students in a large group, and didn’t have the staffing to schedule presentations for 5 or 7 smaller groups from a single course.
A second problem vexing me was at the graduate level. While many of my students were brilliant and well read in any number of theoretical texts, a disturbing number (to me, at least) who took my medieval seminars could not read the text closely. New Criticism and the study of poetry had apparently been so thoroughly routed from the field in favor of prose fiction and a variety of post-structural theories that very few could analyze a line or a passage of Middle English in any detail–a skill, it seemed to me, still worth having.
A third problem involved the use of slides. I wanted to improve upon slides–which I use heavily as a teaching tool in medieval courses–for a couple of reasons. Since there was no slide library available to non-Art department faculty at Stanford, I had to rely on my personal collection for classroom needs, and my students had access to nothing at all. Until recently at Stanford, there were also significant costs involved in showing slides to large classes–it cost over $100 a shot for someone to unlock projection booths in the large classrooms and set up the machines, making visual presentations very costly.
I also began to feel that I needed to evolve teaching methods that appealed to students’ visual literacy as a means of cultivating greater verbal literacy. If it is true that we are moving toward a post-literate society–as the recent report on national literacy would suggest–then those of us whose business involves literate skills need to rethink our teaching strategies in fundamental ways.
The time had come to see whether computer technology could help me solve some of these pedagogical problems, and over the course of a couple years, the “Electronic Chaucer” course evolved. One exercise I used was based in the playful possibilities offered by powerful word-processing programs. Students chose a portion of the General Prologue to the Canterbury Tales and then “illuminated” the text using the typographical palette offered by their word-processing program. Figure 1 Differing fonts, type styles and sizes, and formatting options allowed them to gloss the text through layout and typography. In this way they could offer an interpretation of the text using visual cues, which then served as a starting point for class discussion or for their own more extensive written analysis. The virtues of the exercise are many: it can be a boon to shy students; it appeals to students’ creativity; it can be used in collaborative learning groups; it can encourage students to “hear” the text even as they gloss it visually; it prods them into thinking about texts not just as strings of words, but as systems of meaningful signs, some of which are verbal and some not; it can serve as a starting point for a historical discussion of the layouts in medieval manuscripts, and how the disposition of the text on the manuscript page was a part of the communicated meaning. Such an exercise offers an attractive way for students to work actively with the text, allowing them to grapple with the relations between its visual and aural texture and its meaning; it is a concrete way of introducing the idea of form as meaning.
I found, in other words, that the notorious fluidity of electronic text, which presents significant problems for print-based notions of copyright, intellectual property, and publishing, was a great pedagogical boon because it could be made to approximate some of the conditions of pre-print manuscript culture. Like the scribes of medieval manuscripts, each of which presents a unique version of the text, my students were able to manipulate the electronic text to create their own unique versions. While the notion of scribe as author is no longer strange to medievalists, the notion of the student’s authorial role with respect to the text will require rethinking customary power relations among teacher, student, and text. I might point out that whereas some of these issues and their implications have been raised by theorists of hypermedia, it doesn’t require full hypermedia to arrive at them rather quickly: only Microsoft Word or a similar program.
A more challenging assignment required the students to perform and analyze the results of on-line searches of an electronic version of Chaucer’s tales. I should point out that the students used both print and electronic editions of the Riverside Chaucer. They bought the print version, published by Houghton Mifflin, while Stanford purchased a site license for the electronic version from Oxford University Press, which had acquired the rights to the electronic version from Houghton Mifflin. Thus the print and electronic editions enjoyed a peaceful coexistence in my class–or so I thought, of which more in a minute. While I have elsewhere discussed the pedagogical success of these assignments–which was notable–that is less important for this forum than the institutional infrastructure that allowed me to offer on-line searching in my classes. The physical location of the class was in the undergraduate library; the machines were obtained through a grant application by the Freshman English program. The physical and institutional location of the electronic edition of the Riverside Chaucer was the Academic Text Service of Information Resources. This group was historically and culturally separate from the libraries, though recently it was merged with them organizationally. It took the lead in developing a library of electronic texts, in training faculty to use the search software, and in providing the logistical support for integrating the technology in the classroom and teaching students how to use the software. For me they were the gateway, to use one of the themes of this symposium, to considerably improved teaching conditions.
I also discovered, rather unexpectedly, the students’ potential role as gatekeepers in my future choices of editions. When I introduced the Searcher into my undergraduate course, I asked the students to keep journals so that I could track their responses to the program. I discovered–to my dismay–how much they hated the seven-pound Riverside Chaucer, which I and other Chaucerians consider a scholarly edition of the highest quality, absolutely packed with notes, glossary, critical apparatus, and bibliography. To them, its physical form conveyed heavy drudgery, the oppressive weight of an obscure language, the mental paralysis of facing a tradition of learning that overwhelmed them. When it was electronically dematerialized, however, they felt that they could approach Chaucer’s texts as poetry, and not as a dead language to slog through.
Knowing, as I do now, that the very monumentality of the print edition I have been using works against the goals of the course, I will very carefully reconsider what type of text I will use the next time that I teach an undergraduate Chaucer course. Now that I teach in a state institution, economic issues will figure into the decision as well. At $65, the Riverside Chaucer may be a luxury my students cannot afford. While the $100 price tag of the Oxford electronic version is no bargain now, the $300 site license is. I could well envision requiring the students to use a networked electronic text in conjunction with a cheap paperback edition of the Canterbury Tales, and then asking them to buy several supplementary books.
When I teach such a course again in 2001 Oxford University Press will have competition for provision of the e-text of Chaucer. By then, Peter Robinson’s electronic edition of the Canterbury Tales, published by Cambridge University Press in CD-ROM form in a new Cambridge Electronic Editions Series, will be well advanced. The first edition is scheduled to be released in late 1994, with 55 MSS and 4 pre-1500 printed editions of the Wife of Bath’s Prologue. It will also contain digitized images of all eleven hundred pages of the MSS and editions, as well as full collations and analyses of the manuscript relations, databases of spellings and variants, and collation and analytic software. Though designed as a research tool, it will have a place in the classroom because it will offer students easy and practical access to primary sources–manuscripts and early printed editions–that are beyond their reach and that of their institutions.
When the course is offered at Washington State in 2001, if all goes according to plan, the students will have access to resources that were unimaginable just a short time ago. From networked classrooms or from their dorms they will be able to search the on-line library catalogue for the entries on electronic editions of Chaucer. They will then be able to call up those editions and search them interactively in the same session. If, say, while reading the Miller’s Tale they were struck by the image of Nicolas playing his instrument, they could–in the same session–search the catalogue for digitized images of medieval musical instruments and view them. If they were curious about the song that he played, Angelus ad virginem, they could search for the relevant catalogue entry and play the song as they studied the images and text.
This will be a full multimedia catalogue and database. The imagebase that I developed was, in comparison, only two dimensional–text and image. However, that relative simplicity allowed me to design and implement the system in a few months with the help of several grants and with technical support from Stanford’s Libraries and Information Resources. I used a pair of beta-version commercial software programs called ArtAccess and Image Access, developed for collections management at museums and art houses. These programs have the ability to display and manipulate high-resolution color images of manuscripts, art works, maps, and the like. The quality and flexibility in the reproduction of images goes far toward reducing the logistical problems of access to the sorts of objects that medievalists often study (manuscripts, objects in European collections). Such a databases promises both to deepen and to democratize the study of medieval culture in the sense that it opens to the many–undergraduates and the public–possibilities for concentrated engagement with medieval objects more typical heretofore of advanced academic research. Students at geographically isolated institutions, from less cosmopolitan backgrounds, and those with limited library resources, have a chance to work with historical, cultural, and visual materials not otherwise available to them. Unlike the Stanford students, students in Pullman–like most of the rest in the country–don’t have the opportunity (usually) to hop down to L.A. for the weekend, and drop in to the Huntington Library to see the Ellesmere manuscript of Chaucer. But an image archive can bring significant pieces of the Ellesmere manuscript and Canterbury Cathedral to them, and expand their understanding and appreciation of medieval poetry and culture–or any other non-local culture.
The “imagebase” allows the storage and retrieval of images with up to 35 pages of information, commentary and bibliography attached to each one. The images were digitized from my personal slide collection using PhotoCD and Photoshop. Any kind of image can be entered: manuscripts, architecture, artworks, maps, documentary photos, and so on. The program displays thumbnail images on the screen as though they were slides on a light table. Figure 2 Selecting an image for viewing brings up a record known as “artwork info”: a larger image with identifying information.Figure 3 Paging through the record by means of the icon at the bottom middle allows you to select any one of the fields on the second “page” of the record for a greater depth of information about the image. Figure 4 If you wish to learn more about the artist, you click on the button “artist info” and the screen in Figure 5 appears. The imagebase thus allows students access to more information about an image and its context than does the usual classroom slide show given by a non-art historian like myself.
The “image record” containing “artwork info” can also be used to indicate spatial or conceptual relations among images. For example, the record for Trinity Chapel at Canterbury includes “details,” which appear as thumbnail images of the stained glass in the chapel. By clicking on the appropriate icon, students can view the stained glass. In a similar fashion, the pilgrim portraits in the Ellesmere manuscript can be viewed either as details of the full-page view, or as full-record images in themselves. This associative feature could allow an instructor to structure sequences of images for students–for example, a rudimentary guided tour through Canterbury Cathedral.
The program provides other interesting new possibilities for the organization of a body of knowledge. An instructor can create related sets of images for students (called “portfolios”) in a matter of minutes; they can be modified or deleted at the click of a button. If students found a certain topic particularly engaging, the instructor could create a custom portfolio on that subject by selecting appropriate items from the full imagebase. Or students could create such portfolios for themselves, to work on over the semester. My current portfolios include, for example, Canterbury, the Knight’s Tale, the Ellesmere manuscript, and one for materials contained in Stanford’s Department of Special Collections. Figure 6 My notion in creating this portfolio was that the imagebase could be a way of mediating between the sheer numbers of large lecture classes and the limited resources of special collections departments. Students can do preliminary browsing and research on the computer, and then follow up with more directed research in the collections themselves. It could also be a way of developing collaborative research between faculty and students in the humanities. Take, for example, the Stanford Chaucer portrait: unknown to Chaucer scholarship, it is a historical mystery whose creation, use, and provenance I am trying to solve. I can place the digitized image at students’ disposal, as well as information they can use to develop their own investigations, and perhaps collectively the mystery could be solved more quickly than otherwise. There is an additional benefit: though, technically speaking, the digitization is not of archival quality, it has nonetheless proved to be highly useful since it was discovered that the original was damaged enough to need immediate conservation and little future handling.
Students can also organize knowledge for themselves or take a more active role in their study of the images through the program’s search capabilities. The simplest among them, the “QuickSearch” facility, contains a variety of search fields (title, artist, classification, keyword, etc.) with pull-down menus that list the available options. Figure 7 Selecting “woman” from the keyword option produces a “light table” containing the images in the database that I have tagged with the keyword “woman.” Figure 8 It can serve as a quick visual essay on the representation of woman in medieval and Renaissance culture. Users can also search by classification, say 15th century British manuscripts, or by collection, such as the British Museum or the Louvre. Having obtained the search results in a matter of seconds, students can then use the bibliography and commentary in individual records as springboards for their own further research.
As the final course assignment, I had students choose an image, research the information to go into the record’s fields (provenance, bibliography, commentary, description, artist biography, etc.), and then actually write the record themselves. In this way they practiced (or learned) research skills not often emphasized in English classes; they had to synthesize information into an attractive one or two screen presentation (no one likes to read computer text extensively); and they practiced writing for a public audience of future users of the Image Archive, not merely for the private reading pleasure of the instructor. I wanted them to develop the sense that research is collaborative, interactive, and an on-going process as much as a final product.
If the imagebase is available in a classroom, as this was in a networked Macintosh classroom in the undergraduate library, it can provide powerful, immediate visual information in response to students’ questions or problems, making classroom time more visually interactive and productive than is customary now. One of the assignments for the course had students select a page of the Ellesmere manuscript and discuss the literary implications of its layout or ordinatio. Though they had read the classic article on the topic, they really didn’t understand the concept of ordinatio, as became clear during class discussion. With the workstation in the room, I was able during class to show them a page from the Ellesmere manuscript juxtaposed with pages f”rom both the Kelmscott Chaucer and an Ovid manuscript. Figure 9 On the spur of the moment I could illustrate by comparison and contrast how the elements of page design contribute to a reader’s interpretation of the text. Without the Image Archive it would have taken me at the very minimum several days to track down the volumes, obtain permission from Special Collections for xeroxing the Ovid manuscript and Kelmscott Chaucer, and then to get the copies back and duplicate them.
The program’s ability to juxtapose images on the screen (known as “tiling”) allows a flexibility of display beyond the tandem projection of slides. Figure 10 shows four images representing some aspect of erotic love in the Middle Ages. I have found that my students can “read” such images–that is, decipher their conventions and feel their power–more quickly than they can master the intricacies of equivalent conventions and variations (love as wounding, vulnerability, sickness, death) in a set of four texts. I can thus use such a display to prime their visual imaginations before or as we study literary texts.
The larger file size available in this program allows detailed study of the manuscript as well. You can see details of pen strokes, rulings, painting, etc. This level of image reproduction (4 megabyte file) will allow more serious study of manuscripts than is currently possible in microfilm and most facsimiles.
After I give this demo, the first question people ask is: How can we buy this? That is, when and how are you going to “publish” it? Given their perception of the value of the enterprise, and their lack of resources to replicate my work, the question is reasonable. After all, why should the work be duplicated? Why not conserve collective energies and resources? My answer to the question is that I cannot “publish” it as things stand, because the time investment required to resolve the copyright and software licensing issues is, for me, prohibitive–not even to speak of potential fees involved. Where is the gateway to disseminate such a project? Are there publishers willing to work with faculty authors to negotiate with software companies and with holding institutions for electronic image rights in order to “publish” such a project?
I was much encouraged when I discovered recently that among the consortium of companies backing the New Media Centers Initiative is Prentice Hall. According to a press release: “Prentice Hall, the consortium’s founding publishing partner, plays a valuable role in steering debate over practical ways to deal with intellectual property issues, distribution, and royalties in the context of new media, and sharing expertise in the process of turning prototypes developed at New Media Centers into viable products for the academic or commercial marketplace.” Part of the good news for me was that publishers are getting in on the act. In my limited view of the field, what I have been seeing is that software companies are the ones buying up image rights, developing products, and generally dominating the market: Microsoft’s “Art Gallery for the Macintosh” is a case in point. For $80, you can buy a CD-ROM with “access to over 2000 pieces of art in the collection” of the National Gallery, London. There are guided tours of the collection’s highlight from art experts, a historical atlas, a “comprehensive glossary” of art and historical terms, the “correct pronunciation of over 750 artists’ names,” and last but not least, you can “see artists’ brush work and other techniques through unique animation sequences.” I haven’t seen this product yet, and my response is divided: on the one hand, it is better to have it than not to have it. An individual or a library can afford to buy it whereas the comparable art book version would be prohibitively expensive, and it would lack sound and animation. On the other hand, is the quality of the product sufficient for academic purposes? The color is 16-bit, and most users for the next few years will play it on 8-bit monitors; yet the industry standard for serious digital artwork is at least 24-bit color. Will such products shift the role of faculty from being the primary producers of courseware, with technical support from software engineers, to being consultants to software design teams whose primary market is commercial? And given low faculty salaries and high consultants’ fees, the market may drive that shift in more ways than one.
Let me throw in another complication to the prospect of “publishing” such an archive: traffic in images over Internet. With the increasing availability of public domain images over Internet, such as the Library of Congress exhibits, why should we bother with the hassles of publishing a select corpus of images from a wide range of institutions? Why not settle for whatever will be available free over Internet?
My demo audiences have posed a second set of questions related to publishing an image archive, questions that focus on gatekeeping roles rather than on gateways. Why, they ask, bother with such a project; why not publish a book instead? How can such a project possibly be linked to your research and publication program? In other words, how can such a project be justified in terms of career rewards, when these rewards in the humanities–hiring, tenure, promotion, salary increases–are based on the cornerstone of book publication? Other concerns are related: how does the academy evaluate the quality of a multimedia publication when the criteria for evaluation, once again, are deeply rooted in the culture of the book? As one colleague put it, a book has a title page with the author’s name, the date of publication, and the name of the press. With these the book can be contextualized historically and socially, and the quality of its scholarship can be evaluated or contested in those terms. But how is that done for multimedia publications, which appear to be authorless, dateless, collaborative, open-ended and fluid?
These are the questions of the gatekeepers at the top echelons of the academic hierarchy. Given the assumptions of these questions, it would be a fool or a visionary who published a multimedia work and submitted it for tenure evaluation where such assumptions reign. But that reign may be dated sooner than we imagine, as the young visionaries on our faculties challenge us with such publications more and more frequently, forcing a choice between losing rising stars or re-forging the professional consensus on what constitutes work that can be evaluated and what the criteria of quality are. As Vartan Gregorian put it in his keynote address last year at the Irvine conference “Technology, Scholarship, and the Humanities” (which I read as an electronic publication of the ACLS over Internet): “The relevant organizational structure must change to adapt to the new technology. Until that happens, the real revolution of technology in higher education will not have happened.” In order to make that revolution happen, we need dialogue and partnership with publishers as they work out their own procedures and criteria for evaluating the quality of multimedia publications. Then, were a young Chaucerian to come up for tenure in my department in 2001, we could confidently assess the merit of her multimedia publication, “The Road to Canterbury.”
I adapted an exercise called “Illuminated Texts,” which I found described in a handout from Ohio State’s Program for Computing in Composition and Literature.
For a discussion of the problems presented by electronic text, see the essays in The Politics of the Electronic Text, ed. Warren Chernaik, Caroline Davis, and Marilyn Deegan, Office for Humanities Communications Publications Number 3 (Oxford: Office for Humanities Communication, 1993); Paul Delany and George Landow, eds., “Managing the Digital Word: The Text in an Age of Electronic Reproduction,” in The Digital Word: Text-Based Computing in the Humanities (Cambridge, Massachusetts: MIT Press, 1993), pp. 3-28; and Nancy Kaplan and Stuart Moulthrop, “Seeing Through the Interface: Computers and the Future of Composition,” ibid., pp. 253-70.
For example, by George Landow, Hypertext: The Covergence of Contemporary Literary Theory and Technology (Baltimore and London: John Hopkins University Press, 1992).
Ed. Larry Benson et al. (Boston: Houghton Mifflin, 1987).
”Chaucer and Hume Meet the Searcher,” Speaking of Computers, July 8, 1991 (Stanford: Libraries and Information Resources), pp. 1-2.
In “The Places of Books in the Age of Electronic Publication,” Representations 42 (Spring 1993), pp. 13-37 Geoffrey Nunberg offers a wide-ranging assessment of the implications of electronic publication for the future of the book.
Described in The Canterbury Tales Project, Newsletter no. 1 (Oxford: Office for Humanities Communication, August 1993), pp. 3-4.
In point of fact, the programs no longer exist. The original company has split into two: AXS handles ImageAXS, meant primarily for photos, though any digitized image can be used. Digital Collections, Inc. is developing Ark, the successor to ArtAccess, still intended as a collections management tool.
Christinger Tomer, “Emerging Electronic Library Services and the Idea of Location Independence” in Landow and Delany, eds., Digital Word, pp. 139-161 surveys some of the issues of “location independence” in a networked “information omniverse.”
For an excellent introduction to the issues surrounding digitization, see Peter Robinson, The Digitization of Primary Textual Sources, Office for Humanities Communications Publications, No. 4 (Oxford: Office of Humanities Communication, 1993).
For a discussion of archival quality digitization versus “transmissive” level quality, see Robinson, Digitization, pp. 14-15 and the projects on pp. 82-94.
Malcolm Parkes, “The Influence of the Concepts of Ordinatio and Compilatio on the Development of the Book,” in Medieval Learning and Culture: Essays presented to Richard William Hunt, ed. J. J. G. Alexander and M. T. Gibson (Oxford: Clarendon, 1976).
Note Kiernan’s project for 21 mb files of Beowulf.
In order to reproduce materials covered by copyright (i.e., not yet in the public domain) without any risk of incurring litigation, I wrote letters to each copyright holder. The permission was sought and granted for the course at Stanford. Obviously, publishing the materials in order to distribute them widely would require seeking permissions again, a substantial chore. For a view of the future of copyright, see Jane C. Ginsburg, “Copyright Without Walls?: Speculations on Literary Property in the Library of the Future,” Representations 42 (Spring 1993), pp. 53-73.
Press release 10/17/93, Cincinnati, OH. Also described in “White Paper.”
Microsoft product information sheet, “Features & Benefits,” for version 1.0.
Transcript of Keynote address for “Technology, Scholarship, and the Humanities: The Implications of Electronic Information,” Irvine, California, Sept. 30-Oct. 2, 1992, Summary of Proceedings (Internet: ACLS and J. Paul Getty Trust, 1993), p. 9.