Even among critics not particularly concerned with detective fiction, Dashiell Hammett’s fourth novel, The Glass Key , is famous for carrying the so-called objective method to almost obsessive lengths: we are never told what the characters are thinking, only what they do and look like. Anyone’s decisions about anyone else’s intentions are interpretive decisions, dependent on correct presuppositions—on having the right interpretive key. The novel’s title, in part, refers to this kind of key. Ned Beaumont, the protagonist, has to decide (...) how to govern his relationship with Janet Henry; one of his major clues to her mind is a dream that she tells him, a dream that climaxes in an attempt to lock a door against an onslaught of snakes. Dream interpretation is difficult enough to begin with, and Janet Henry compounds that difficulty by telling the dream twice. In the first version, the attempt to lock the door succeeds; in the second, the key turns out to be made of glass and it shatters. Ned Beaumont, in deciding which dream to us as his key, chooses the second —but it is a choice based on an intuitive mix of experience and faith, knowledge and hunch.A reader often faces the same difficulties that Ned Beaumont does. Reading a book, too, requires us to make a choice about what key to use to unlock it, and that choice must often be based on an intuitive mix of experience and faith, knowledge and hunch. For example, as I shall show, the experience of reading certain texts—not all, but a significant number of them—is problematic because it depends in part on whether the reader has chosen, before picking them up, to approach them as popular or serious. My argument hinges on two prior claims. First, I contend that one way of defining genres is to consider them as bundles of operations which readers perform in order to recover the meanings of texts rather than as sets of features found in the texts themselves. To put this crudely but more modishly, genres can be viewed as strategies that readers use to process texts. Second, I argue that popular literature and serious literature can be viewed as broad genre categories.Peter J. Rabinowitz is associate professor of comparative literature at Hamilton College. He is currently working on a book about literary conventions and is also active as a music critic for such publications as Fanfare and Ovation. His previous contributions to Critical Inquiry are “Truth in Fiction: A Reexamination of Audiences” and “Who Was That Lady? Pluralism and Critical Method”. (shrink)