People who write about movies have traditionally referred to the conventions of cinematic representation—such things as low-angle shots, fade-outs, or flashbacks—as the “language” of film. Does the ability to understand this language require previous experience? Or, to put this question differently, would a “naive” viewer, someone who had never seen a movie before, be able to make any sense of his or her first encounter with this medium? The term visual literacy, popular among media scholars, reflects the widely held belief that the comprehension of cinematic conventions is indeed an acquired skill, comparable to fluency in reading or writing. In contemporary film scholarship, this belief is based largely on an extrapolation from the work of such writers as E. H. Gombrich regarding the cross-cultural variability of pictorial conventions. This body of literature is commonly assumed to have shown that any perceived similarity between pictures and the things they represent is simply the result of viewers’ unwitting assimilation of the representational standards of a particular culture or historical period. Consequently, it is argued, the ability to connect a picture to its intended referent must depend on prior familiarity with the conventions employed in that picture. As far as film is concerned, this argument has occasionally been supported by stories about misinterpretations reportedly experienced by early-twentieth-century filmgoers or other inexperienced viewers.