In relation to several ideas in some of the discussion threads of the last few days, I wonder if anyone else has had occasion to try out an idea I've long had. This would be to teach a classical Hollywood survey course, whether on a specific period or genre, but *not* choosing films on the typical basis of prototypes, art, or other significance as discerned from our current perspective. Instead, screening selections would be limited to only those films which box-office figures reveal to have been the most popular of their time with their original audience. While obviously very constraining, with results that would certainly be top-heavy with stars, I suspect the results might be most interesting. For instance, it would give a more accurate perspective on what audiences actually saw, as opposed to the canon or what is interesting to us today, and might force students to move beyond their own reactions to (hopefully) a fuller understanding of context. Has anyone tried this? With what results? Brian Taves Motion Picture/Broadcasting/Recorded Sound Division Library of Congress Washington, D.C. 20540 202-707-9930 202-707-2371 (fax) Internet: [log in to unmask] ---- To signoff SCREEN-L, e-mail [log in to unmask] and put SIGNOFF SCREEN-L in the message. Problems? Contact [log in to unmask]