Time Will Tell, But Epistemology Won't: Part I

In planning the conference to celebrate the launch of philosopher Richard Rorty's archive, which includes twenty-four plus linear feet of conventional paper documents and over a thousand "born-digital" computer files that are being made available to scholars on the Internet, we decided to structure the day very differently from a more conventional academic homage that might have treated Rorty's life's work as a stable and static monument to his intellect, one to be defended by unquestioning acolytes unwilling to engage in debates or discussions that might be ongoing or explore the possibilities of "insouciance" -- a word that comes up dozens of times in Rorty's electronic correspondence and remarks.

Instead, as my welcome to participants emphasized, the conference was about work to be done rather than work already done. It was also a day-long interdisciplinary conversation between philosophers, cultural critics, rhetoricians, and archivists about technologies, platforms, media, publics, and scholarship that spoke to Rorty's legacy in a time that we all recognized as transitional.

What was particularly remarkable about the planning of this event was that no one was more supportive of a potentially irreverent approach than Mary Varney Rorty, Rorty's widow and literary executor. When Bruce Sterling suggested in a posting in his Wired blog that the conference might treat Rorty as a mere "digital theory object" or his memory in the era of computational media as a "pig" in "the era of sausage machine," she expressed good humor and said that Rorty would have appreciated being cited in such a publication.

Indeed, Mary Rorty's comments on "Memory, Ethics, and Literary Custodianship in the Era of Computational Media" were the first to be posted on the Internet, and the other papers from the conference will soon be available online. She noted that Rorty "never had a thought that he didn’t write down and publish" and believed that "if you haven’t expressed those first thoughts, that you can build upon or retreat from, your intellectual trajectory remains a private amusement rather than a public good." (She also observed that academics are "socialized to believe that it has to be perfect, just right, deathless and eternal before we can commit it to print," a finding that also holds true in the case of the "conservatism of young scholars," which is described in a recent Mellon Report on publishing in higher education.)

At one point Rorty even wryly recommended "intellectual property law" as an ideal career for the next generation of academics, given the ramifications of "the Internet, Google Scholar, and Google Books." She also said that as an instructor herself, she found it more and more difficult to persuade her students of the "continuing importance of canons of responsible research." She also gently mocked the pessimism of Peter Sloterdijk, who in his Elmauer Rede argued that "thanks to the influence of television and video games, succeeding generations would transcend (or degenerate away from) print, from literacy – would lose the sense of themselves as participants in what Richard Rorty once described as the continuing conversation of mankind." For her, electronic access would nourish print publication and her husband's legacy not supplant it.

Next, Head of Special Collections at the UC Irvine Libraries Michelle Light began to explain some of the features of the Rorty archive's design. She described how librarians realized that the 50,000 pages from word processing documents captured from Rorty's dozens of floppy disks did not duplicate his paper records and were thus extremely valuable for the collection as a whole. She also told audience members how they soon realize that such a "born-digital archive," which was "created and only maintained in digital form," might be valuable for launching the library’s digital scholarship service. Light argued that it was critical for Rorty scholars to understand the collection as a "hybrid archive" in which both parts are integral.

Light also detailed how this collection might pose many challenges to archivists, not only in the present but of the future: "Will we have the right hardware and software? Will the bits still be there for us to read?" She also asked if UCI archivists were right to preserve "content only" and not the "context of author’s computing environment" and asked "How can we be sure that the files have not been tampered with?"

She also discussed a number of issues around "respect for privacy." After all, not only do scholars like Rorty write confidential letters of recommendation, but those letters can be considered to be student records that are further protected by FERPA. Moreover, authors might be saving many of these files unintentionally, since even a "deleted" file is not truly removed from a hard drive. She mentioned how Stanford University had been developing "digital forensics" that would recover lost traces in the digital record. In contrast, she noted that Harvard University was willing to admit that "we don’t have any methodology," since "we just store" materials in climate-controlled rooms and hope for "universal Harvard guidelines" to be written.

Light explained that "we didn’t wait for clear best practices," because archivists had a legitimate concern that "bit rot" would cause loss of data, particularly as the physical material comprising the disk decayed. Furthermore, they wanted to take advantage of the fact that readers of 3.5 inch disks were still readily available on campus, unlike zip drives or the eight-inch drives that once graced the computer of my own youth.

Then Light showed an "Emulation vs. Migration" diagram and told how UCI had chosen to migrate "the files from one obsolete platform to another" and thus lose the "original functionality of programs that Rorty used." Nonetheless, they "opted to preserve content but lost context." To demonstrate a radically different approach, she described the most attention-getting current example of an emulation archive, the Salman Rushdie Archive at Emory University. Information technologists actually created a Mac emulator, which is available in the rare books reading room, where one can open the writer's calendar, read his e-mail, and relive his experience of writing a novel. Emory archivists have described this approach as preserving the "ecosystem" or "biostructure" of Rushdie's digital files.

Light suggested that such whiz-bang digital sleight-of-hand might actually cover over certain issues about "authenticity and integrity" that might be revealed just by looking at the observable difference in the hexadecimal code of documents that are migrated from their original word-processing document format to archival PDFs. "When we transformed the file, we tampered with the digital content." Furthermore, as Light noted, with paper manuscripts, it is much easier to "notice that parts are missing or have been defaced." As she asked, "What is the record we are preserving anyway?"

Although she expected that many would be irritated by the fact that the digital archives are online but not completely "open," she justified the library's decision to put documents behind a password-protected wall on the grounds of concern for copyright and privacy, since archivists didn’t want to open up the Rorty files to Google indexing.

(Ironically, as the archivists were addressing the Rorty audience, participants at the Computer Forensics and Cultural Heritage were meeting on the opposite coast at the University of Maryland. Simultaneously there was a Symposium on the Digital Humanities at Dartmouth College taking place in yet another parallel Twitterverse to the Rorty conference.)

Light was followed by Dawn Schmitz, who emphasized the fact that these were "born-digital not digitized" records. ("Digitized" content would have been created on paper and duplicated electronically after the original had been scanned.)

To understand the theoretical significance of this particular kind of "cultural form," Schmitz referrred audience members to Mark Poster's The Mode of Information. (Poster had been scheduled to be a participant in the conference, but he had been unable to attend because of illness.) She read from a section in Poster's book that theorized that there was a "mirror effect of the human" in which "the human recognizes itself in the uncanny of machine," which "doubles the subject of writing." As an example, she pointed out how the file back-up program on Rorty's computer "automatically generated drafts," unlike the intentionally created iterations that were formed in a conventional archive. She described her own difficulty working with these pseud0' "drafts." She also argued that this digital proliferation makes the fact that these files aren't printed out more advantageous practically, since printing out digital files could create many problems for a library with limited space.

Then Schmitz showed some examples of data corruption, such as the image above, as Microsoft Word imperfectly read the file created in an earlier word processing program of Rorty's that may have treated the ASCII characters he typed somewhat differently from the word processing programs of today. She also argued that one of the virtues of a digital archive was that it provided more affordances in searchability, since one could easily look through the whole correspondence with a search term like "Clinton" to find his discussions of the former president. (See below.) However, in closing, she reminded the audience that creating metadata continued to be the most time consuming part of the archival process and that keyword searchability was never enough.


Next, Erin Obodiac took theoretical reading of the Rorty Archive a few steps further with a talk on "digital immunity," which began with the example of two Rorty typing drills in the digital archive that are composed of the same 1867 sentence: "Now is the time for all good men to come to the aid of the party.” One, however, terminates with a period, while one does not. Her presentation, which she summarized as being about "questions concerning technical reproduction, copyright, publication, authorship, political action, and the significance of the banal and random" also included an anecdote about digital doubling of another kind. Members of the audience chuckled at the fact that Rorty's august new digital archive also included a letter in which Rorty complained that "I have two advantage numbers: K166506, as 'Mr. Richard M. Rorty' and FVJ8956 as 'Dr. Richard Rorty'." She then drew her discussion back in the realm of more serious theoretical reflections by discussing the archival character of Derrida's reading of Nietzsche's fragment about "I have forgotten my umbrella" as being like an archival fragment in a library that exists without its original context.

Perhaps the most clever part of her paper was the part in which she pointed to a 1995 Rorty letter about some tax matters that ended with the phrase "Burn this!" As Obodiac noted, "In the old days 'burn this' might mean, 'incinerate this piece of paper,' i.e., erase this inscription; now, 'burn this' might mean 'copy this file on a disc,' i.e., mechanically reproduce this inscription." For Obodiac, the "born digital" is the "born iterable."

As a champion for digitizing and making open the entire Critical Theory Archive at UC Irvine, there was also a serious critique in her paper about current copyright law and systems of valuation that place sixteen hundred dollar pricetags on images of postcards from one person represented in the archive (Jacques Derrida) to another (Paul de Man) and prohibit replication of the image for educational and critical use. Furthermore, Obodiac protested that the library participated in regimes of privatization in everything from its basic electronic archival design to its library user agreement. Obodiac put her central objections as follows:

Backtracking a second, we will recall the copyright statement on UCIspace, the website that hosts the Rorty Born-Digital Papers: "This material is provided for private study, scholarship, or research. Transmission or reproduction of any material protected by copyright beyond that allowed by fair use requires the written permission of the copyright owners. The author(s) of the material or their heirs retain copyright to this material." The point to emphasize here is that the material is for private study, private scholarship, and private research. In theory, there could be an infinite number of individuals engaged in private study, scholarship, and research, and yet no public study, public scholarship, or public research at all.

Much of her paper, which could not be read in its entirety, was devoted to a reading of Roberto Esposito, the author of Bios and Communitas. Of course, many Internet researchers have argued that "digital community" is impossible and that researchers shouldn't even use the term. (Now "networked public" is becoming a frequently substituted phrase.") Obodiac asserted that "digital immunity" received far less attention than "digital community" and that this form of "immunity" denied the "munus," which was both gift and obligation in Esposito's terms, that was at the heart of any archive given by an author or his heirs to an institution.

Tom Hyry, who has recently become the new Head of Special Collections at UCLA spoke last as an "outsider." As a respondent to the panel, he praised the conference's attention to the "relationship between researchers and archivists" and to the problem of the born digital on which he said "many archivists have punted." He described what he called "the disk in a box problem." As he joked, since he used to be at Yale, it could be called "the Harvard problem" as well. Like Light, he also criticized the Emory approach, in which researchers were placed in the "absurd situation of coming to physical reading rooms," which he described as a "conservative but safe interpretation of copyright law" that profoundly missed the point of research efficiency. He also appreciated UCI's straightforward admissions of anxiety about "authenticity and integrity" and the loss of "context," since archivists "live" by that central "tenet of archival theory."

Although patrons might appreciate the ability to search materials easily, Hyry cautioned about potential hidden costs. First, he notes the "concept of appraisal" is changed when material in an archive isn't "thrown out" if it is "considered banal." He also said that "volume" was transforming archival work and that "the Rorty papers" were "a pretty modest collection," because these were word processing "files meant to be printed out on paper." Now archivists must manage "whole different types of expression" and "massive numbers of formats."

Ian Bogost tried to introduce a platform studies perspective more explicitly into the discussion by asking if the archivists might be falling into a predictable patter of digital "idealism" that was perpetuated by "treating everything as documents" or "records" rather than executable code created in response to particular technical constraints. They did seem to understand his question about bit-level preservation, however. As Hyry noted, realms of corporate, finance, and law enforcement already were working with digital forensics, but in archival work such forensics could pose a real ethical difficulty to the profession. He gestured to the audience and reminded them that "your hard drives all have stuff that you think that you deleted."

Fellow philosopher and Rorty family friend Margaret Gilbert raised the question of Rorty's e-mail, which archivists explained that they did not keep in Rorty's case. However, Light explained that predicaments about electronic correspondence would not be avoidable with other digital collections. As Light put it, "e-mail is a beast that we have to wrap our heads around." Donor Mark Poster apparently exported e-mail folders that he thought would be of interest to others in the future. And Stanley Fish's hard drive with a skuzzy port may have other kinds of files as well.

In closing, archivist Jackie Dooley posed a question about cost to which the UCI librarians answered that it took "multiple people multiple months" to work with Rorty's born digital collection. They estimated that it took "20 times as long to process electronic papers." Hyry ended this line of discussion by saying that such cataloging would need to be automated in some way and that they would have to think about using future "technical means" to keep workloads manageable.

Update: Obodiac's paper has been posted here.