When Jonathan Larson, author of the musical RENT, died suddenly of an aortic aneurysm at age 35, he left about 180 floppy disks in his apartment that were gathered by his friend Jonathan Burkhardt and eventually donated to the Library of Congress. The disks contain many drafts of Larson's lyrics and libretti written using various word processors and digital music files mostly composed using Mark of the Unicorn's (MOTU) software Performer (now called Digital Performer). When considered alongside the analog items in the collection, the scholar can use these digital artifacts to reconstruct the development of RENT in remarkable detail.
The broad narrative of the early history of RENT is well known. Playwright Billy Aronson and Jonathan Larson decided to collaborate on the musical, Larson initially only as composer, and together they wrote three songs before going their separate ways: “Santa Fe,” “Rent,” and “I Should Tell You.” In the late autumn of 1989, Aronson and Larson agreed to produce a demo tape of these songs to send out to producers. Aronson told me in a phone interview that the plan was originally to gather a few musicians to work for very low cost and rent a studio to record the songs for a total of around $80. Aronson remembers, however, that Larson became enamored with the work of sound engineer Steve Skinner and the possibilities of digitally produced sound. Aronson agreed to pay for half the costs but was shocked to find that the total for the demo was not $80 but nearly $900.Footnote 2
On 31 January 1990, Larson used his computer to compose an invoice requesting payment from Aronson for his share of the costs. The invoice lists three studio visits: one for six hours on 21 December to create “Music Trax for santa fe & rent,” one for 4.5 hours on 16 January to create “Music Trax for i should tell you,” and one for 7.5 hours on 30 January 30th to “Record Vocals for all songs” and create “Mix Trax for all songs.”Footnote 3 Using the timestamps on the music files, it is possible to verify that Larson's accounting was correct. That Larson sought out a digital studio so early in the creative process (and was willing to pay about $300 per session at a time when his primary source of income was part-time work at the Moondance Diner) suggests how important he felt digital technology was for his creative process. Aronson remembers being initially skeptical about the cost but acknowledges that he was impressed by the result.Footnote 4
After recording the demo, however, it seems that work on RENT was shelved for almost two years. On 6 February 1990, Larson created a directory on a disk he titled “RENT” and filed all of the work to date in it; he doesn't seem to have touched the project again until January 1992. The reasons for shelving the project appear to have been complex and varied. Producer Ira Weitzman has said of the collaboration between Larson and Aronson, “Their lifestyles were so different. Billy was married with kids. Jonathan was living the Bohemian life. Clearly they were at odds.”Footnote 5 In December 1989, Larson wrote to his idol, Stephen Sondheim, asking for advice about artistic collaborations. In the letter, preserved electronically and dated 28 December 1989, Larson wrote:
I'm writing because I'm in the midst of discussing possible collaborations with playwrites [sic], and I have a few questions that I thought you might [expound] on.
Lately, it seems that playwrites [sic] get peeved when I try to nail them down as to what their intentions are for a certain piece. For example, Ira has introduced me to someone who wants to take a famous story and do a contemporary version of it. Although I think it's a good idea, he can't explain why he wanted to rework it—beyond, “it's a neat idea.”
Aronson himself remembers that the partnership was always somewhat difficult because, he says, at that stage in their careers neither of them was especially good at responding to the other's work.Footnote 6
And so RENT languished for about a year. Jonathan Burkhart remembers that Larson expressed interest in continuing work in January 1991 while the two were on a road trip in the southern United States, but it does not seem Larson considered it an active project during that year. In August 1991, Larson created a “catalog” of his works, but he took credit only for the music of RENT. On 5 December 1991, Larson created a year-end retrospective for a holiday party for his friends detailing, as the filename puts it, “What we did.” Larson's own summary of the year lists Tick, Tick . . . Boom! and Billy Bishop Goes to War but not RENT. However, Larson does seem to have decided to work on RENT in earnest by the end of the year. On 19 December 1991, Larson wrote to Sondheim, “I'm working (passionately) on my next piece, a rock opera.”
Although there is no evidence of musical or textual work on RENT in 1991, Larson does appear to have been researching the musical in the latter part of the year. A folder labeled “RENT STUFF” created in March 1992 contains a subfolder labeled “MISC/NOTES” that contains two files generated in October 1991 that record Larson's summaries of Susan Sontag's book AIDS and Its Metaphors as well as a prepublication version (probably a lecture) of the ideas antinuclear activist Helen Caldicott presented in her 1992 book If You Love This Planet: A Plan to Heal the Earth. Larson's interest in these works suggests his own uneasy relationship with modern technology. Caldicott was and is famously suspicious of it (claiming, for instance, that space shuttles are destroying the ozone layer),Footnote 7 and Sontag observes, with not a little anxiety, the effect of technology on AIDS-era sexual practices. She writes, “Machines supply new, popular ways of inspiring desire and keeping it safe, as mental as possible,” citing as an example “the commercially organized lechery by telephone.”Footnote 8 Larson makes special note of this paragraph in his “notes” file, summarizing: “Machines replace sexual contact. Phone sex, for example.” It is perhaps worth noting that while both works examine ideas that are thematically important in RENT (present in the song “Contact,” for instance), it is only the file structure on Larson's disk that suggests that Larson himself saw the connection.
A file last modified on 24 January named “!Outline 1:23” appears to mark Larson's return to RENT after a two-year hiatus. By July a file named “*text” contained a draft of act 1 that is recognizably similar to the version eventually performed on Broadway. Sometime between 10:34 a.m. and 12:38 p.m. on 16 January 1996, just ten days before he died, Jonathan Larson made the last change to the text of RENT that is recorded in the digital archive in the Library of Congress. The change was arguably inconsequential—Joanne's line in “We're Okay” was changed from “Steve—Go Home/No you cut the Styrofoam” to “Steve—you're great/No you cut the paper plate,” but the fact that Larson made the change digitally two hours after he had saved the previous draft and saved the draft as a new file suggests how he used his computer in his creative process. Larson seems to have understood, in a way many authors even today do not, that maintaining a revision history of one's digital work (instead of simply saving over old drafts) can be very useful if ever one wishes to revert to an earlier version. One result is that the history of RENT's compositional process is exceptionally well documented in the digital archive.
This history might have been lost, however, had the Library of Congress not been exceptionally flexible in serving the collection. When I first encountered the disks, they were accompanied by a few informal human-generated descriptions of their contents, but in the twelve years since Larson's death no attempt had been made to migrate the data on the floppies to a more stable storage medium. The disks were 800K double-density 3.5″ floppies which, due to differences in the strength of the magnetic head and the speed at which the disks turn, cannot be read by the high-density 3.5″ floppy drives commercially available today. In order to read these disks, I used a PowerMac G3 WallStreet edition laptop (which came with the appropriate drive and could also accept a PC card to enable a USB port). Using this machine, I created bit-for-bit copies of all of the data on the disks (including, if present, deleted files).
This process required specialized technical expertise that was not present in house at the Library of Congress's Music Division. I do not in any way fault the institution for this; it would be unreasonable to expect that in addition to their unassailable content expertise, the subject librarians should have intimate knowledge of the eccentricities of every piece of obsolete digital media the library acquires. In the future, such artifacts may be handled in much the same way as obsolete video and audio media formats are now—media specialists at labs may preserve and migrate the material, and copies will be curated and served by the content specialists.
Until such digital preservation labs are common, however, an interim approach is needed, and needed immediately. In the past century, digital media have changed far more quickly than other storage technologies. Twenty years ago, for instance, although vinyl records and magnetic tape were still sold in many stores, the compact disk was an increasingly common medium for the distribution of recorded sound. At the same time, 5.25″ floppy disks were common and could be purchased for Commodore 64 and Apple IIe computers, which were still commercially available. The difference between these formats is striking; even if MP3 players and digital downloads have displaced the compact disc as the dominant sound technology, I doubt many would consider a CD as definitively anachronistic as a Commodore 64 floppy disk. At the very least, it is far easier today to purchase all of the necessary technologies to read a compact disc than it would be to locate everything necessary to play a music file from a floppy disk generated by some now-forgotten Commodore 64 music-composing software.
Library staffing and procedures, which tend to be defined by necessarily methodical and slow bureaucratic processes, are ill equipped to respond to rapidly changing media formats, computer hardware, and software in a way that will ensure that they can adequately care for and migrate the born-digital data in their collections. Scholars, however, can be much more agile. While it may seem unreasonable to expect librarians responsible for very large collections to have detailed knowledge of every media format they hold, it is not unreasonable to expect a scholar studying a single author or work to have or to acquire the necessary technical expertise to unlock those works. A scholar who wants to examine the earliest extant witnesses to the texts of Euripides or Aeschylus should know Greek. A scholar who wants to study RENT should know about 1990 Macintosh computers. We do not assume that the keepers of classical texts are the only ones who should attempt to translate them, nor should we assume that archivists have the sole responsibility for transferring our born-digital heritage.
Larson in front of the New York Theater Workshop in 1994. Photograph from the Jonathan Larson collection at the Library of Congress, Washington, D.C. Used with permission.
There are challenges, to be sure. Concerns that providing open access to unprocessed and fragile materials risks irreversible damage are legitimate, as are fears that such access might expose sensitive and private information in ways that violate donor agreements. There are, of course, ways to mitigate these risks. Libraries can provide access to unprocessed, born-digital materials only to scholars who have demonstrated their trustworthiness and qualifications and who are further approved by the donor or his or her representatives. Well-designed contractual agreements can limit the likelihood that a scholar will make sensitive information public should he or she stumble upon something unexpected in the archive. However, it is unquestionably easier and safer to take the path of caution and simply deny all outside access to unprocessed digital collections. Such an approach may prevent lawsuits and guard against certain kinds of damage, but it also almost certainly ensures that bit rot will destroy a good deal of our born-digital heritage.
There are presently about five disks in the Larson collection that I have, as yet, been unable to read. Digital media fail; and even when they do not, constructing the technical dependencies necessary to read them becomes more and more difficult as the time between the “last modified date” on a file and the present increases. I hope that I have demonstrated that the born-digital record is a uniquely rich one for theater historians, but it is also uniquely endangered. Theatre libraries and archives must, therefore, strive to develop policies and expertise to safeguard this record. For now, the best policy is to invite those with the greatest expertise—often representatives of those for whom the collection is being preserved—to join the library community, not just as users but as fellow stewards of theatre history. For a moment we have the choice. There is, as Jonathan Larson might say, “no day but today.”
