by Jim English

Most teachers who are serious about their work do a lot of experimenting. They invent new courses, change up the authors, texts or media on their syllabi, test out different pedagogical methods and platforms, link up with diverse collaborators and co-teachers, or even shift over to a completely new field or discipline. Any notion of having already and forever mastered the mysteries of the classroom, so that one need only repeat oneself to succeed, is unlikely to survive the actual dynamics of higher education, which demand a lot of adjustment and adaptation. But while our work encourages experimentalism, any genuine experiment is by definition uncertain in its outcome and susceptible to failure. Fortunately, most of us find this condition more liberating than stressful. Experimenting releases us from the oppressive ideal of mastery and grants us license to fail.

My own experiments, and concomitant failures, go back a long way. When I first started at Penn 30 years ago I was responsible for teaching a general survey of 20th-century literature every autumn. Tackling this course for the second or third time, around 1990, it already felt stale. I redesigned it to incorporate music videos, a rapidly emergent and arguably “literary” form. What I hadn’t reckoned on was the difficulty of obtaining the specific videos I planned on using. I had to rig up a video cassette recorder with a long-playing VHS tape, leave it running all day to record six or eight hours of MTV programming, then rewind and spend my entire evening watching the tape on fast-forward, looking for the videos I needed and copying them onto fresh VHS tapes, one for classroom use and another for student reserve. It was a maddeningly slow and dubious process that yielded wretched video quality. I still recall the frustration of viewing 15-20 hours of MTV on fast forward in search of the “Cult of Personality” video by Living Colour, only to give up in the end and rush together a lecture on Billy Idol’s “Rock the Cradle of Love,” a slickly produced video that was running hourly in the primetime rotation but was only marginally relevant to the concerns of the course. The students were frustrated as well. A number of them wanted to write essays about videos other than the ones we studied in class, but were discouraged by all the technical rigmarole of recording, capturing and copying.

A failed experiment, then, at least in some ways—and one that anticipated my struggles two decades later with a class on the audiobook. Once again the course was a basic survey (The British Novel Since 1900) that I had taught before and wanted to shake up by introducing a new literary medium. Audiobooks had been helping me maintain sanity as an I-95 commuter for years, transforming the dull hours of driving between Princeton and Philadelphia into an extra novel per week. And with the rise of mp3 technology and the smartphone, the old “talking books” were suddenly the fastest-growing segment of the publishing industry, spreading more rapidly even than e-books.

The course I designed covered the same British novels I had taught just one year before, proceeding from Conrad’s Lord Jim (1903) to David Mitchell’s Cloud Atlas (2005). The experiment, as declared at the top of the syllabus, would be to do it with “no reading allowed.” We would not only study the novels by listening to unabridged recordings downloaded from Audible, but we would rely on the audio format for nearly every aspect of the class, with no recourse to printed text for help with preparation, discussion, or written work. Indeed, “written work” in this context meant students’ recordings of their own essay-performances, submitted, revised and resubmitted as digital sound files.

My students and I soon realized that, even though the mp3 is a digital format, for classroom purposes it can be just as clunky a technology as the VHS tape. In an English course, quotation is an essential tool; close scrutiny of specific words, phrases and short passages is our bedrock practice, providing the evidentiary basis for nearly all good arguments about literature. In an ordinary class my students and I all have a book in our hands, with key words or passages indicated by earmarks, sticky-notes, highlighting, marginal scribblings or other ready and familiar means. Page numbering lets us all quickly focus on the same passage. The shift to audio disabled all these standard features of the book. Our playback app contained no search function and no provision for bookmarking. Even if we kept track of the exact hours, minutes and seconds corresponding to a passage, it was no easy matter to manipulate a timeline scrubber to land on that precise moment. To cope with these disadvantages each of us had to extract and edit any sound clips we might want to discuss, label them according to agreed naming conventions and place them in a shared folder prior to class. Of course, after doing all that advance work, we would inevitably find our discussion leading to some passage in the novel that none of us had thought to upload.

Both of these classes were at times pretty shambolic, more like collective trouble-shooting and gripe sessions than proper teaching. Such has been the fate of many of my pedagogical experiments. In a recent class on Empirical Methods in Literary Studies we attempted as a kind of side project to learn the programming language R, an overambitious task that soon had me floundering to the point where students spent a good chunk of their learning time helping their professor get up to speed. In a class last semester on the National Book Awards, we devoted half our time each week to a “literary research lab” where student teams gathered and refined data about the system of status and reward in contemporary literature. Among other lab projects we attempted to predict the winner of the 2016 NBA by feeding our data through a computer algorithm. Developed at McGill, the algorithm appeared to work well for Canadian prizes, but it didn’t work for us. Worse, the data that emerged from our lab proved in the end too spotty to support the kind of empirical claims we had hoped to make.

And yet, messy and muddled though they were, I think these classes were productive. Confronting the limitations of our hard-won data about literary prizes and prestige helped us to think more clearly about forms of evidence and levels of argument in our discipline. The struggle to extract particular music videos from hours of tape raised for us an important argument made by the cultural theorist Raymond Williams, that when literary scholars study television they need to break their habit of close-reading individual texts and focus more attention on the general stream or “flow” of programming. Grappling with audiobooks taught us much about the unique affordances of the codex and the significant challenges of superseding it even in the digital age; as my colleague Peter Stallybrass has stressed, the audio or video recording is essentially a scroll, a primitive technology in comparison with the printed book.

It may even be the case that students learn more from participating in these kinds of failed experiments than they do from sure-fire standard classes. Learning is bound up with failure, after all. And equally with pleasure. For many of us, teachers and students alike, there is considerable pleasure to be had in trying out something new, embarking on a collective enterprise of discovery with no guarantee of success.

Jim English is the John Welsh Centennial Professor of English and
director of the Penn Humanities Forum and the Price Lab for Digital Humanities.
______________________________________________

This essay continues the series that began in the fall of 1994 as the joint creation of the
 College of Arts and Sciences, the Center for Teaching and Learning and the Lindback Society for Distinguished Teaching. See https://almanac.upenn.edu/talk-about-teaching-and-learning-archive for previous essays.