“Every teacher knows the excitement, and chaos, in learning about a subject by undertaking to teach it.” So writes Stanley Cavell of his decision in 1963 to teach a seminar on film. The idea struck him as pedagogically promising, in that all present would have memories of their experiences with movies, and no established body of criticism yet existed to distract them from the nature of those experiences. As Cavell recounts in The World Viewed, though, the seminar was a failure. “Or rather,” he writes, “what was learned was important enough, but it came from our failures.”
Each week I assigned one or two students the responsibility of opening the discussion by reading a two- or three-page description—nothing but description—of the film we all had seen. It turned out that the descriptions were never quite accurate, not always because some gross turn in the plot was out of order or an event had been forgotten, but often because more was described than had been shown. (For example, “The car followed her to the hotel.” But in viewing the film, we had not known until later that the structure was a hotel.) After that, I noticed that almost every summary statement of a movie, whether in newspaper “criticism” or in brochures for a projected series, contains one or more descriptive inaccuracies. Is that because summaries don't really matter? Or because it is unclear what one wants from them? Only about operas, certainly not about novels or stories or poems or plays, would we accept so casual and sometimes hilariously remote an account as we will about movies.
[…]
Another failure in the seminar’s work was no less pervasive, and far more disheartening. The willingness to forgo theory and criticism was too proud a vow, particularly in view of our continuing inability to discover categories we had confidence in or to make comparisons (e.g., with the novel, plays, and painting) that really carried the weight we wished upon them. A frequent reaction to these dead ends was to start getting technical; words flowed about everything from low-angle shots to filters to timings and numbers of set-ups to deep focus and fast cutting, etc., etc. But all this in turn lost its sense. On the one hand, the amount and kind of technical information that could be regarded as relevant is more than any of us knew; on the other hand, the only technical matters we found ourselves invoking, so far as they were relevant to the experience of particular films, which was our only business, are in front of your eyes. You can see when a shot begins and ends and whether it’s long, middle, or close; you know whether the camera is moving back or forth or sideways, whether a figure brings himself into the field of the camera or the camera turns to get him; you may not know how Hitchcock gets the stairwell to distort that particular way in Vertigo, but you can see that he got it. Then what is the reality behind the idea that there is always a technical something you don't know that would provide the key to the experience?
The questions raised by the “failure” of the seminar stuck with Cavell over the next several years, and he describes their lingering effect as the genesis of The World Viewed, which was first published in 1971. The significance of that book, and of Cavell’s work on film more generally, is noted in two recent appreciations from the folks at n+1. In a piece earlier this month, Charles Peterson deems Cavell “one of the greatest American philosophers of the past half-century” but one who’d have become “little more than a curious footnote in American intellectual history” had he not turned to “the more practical concern of how to ‘reconstruct the everyday.’ ”

