Bruce Hayes A Grid - based Theory of English Meter

In recent years, we have seen a real deepening of our understanding of the nature of poetic meter and the English metrical system. In part, this has resulted from the application of the formal methods oflinguistics to metrical study-the willingness to formulate metrical rules explicitly and to check out their consequences has naturally led to progress. However, the new results also depend on recent advances in notions of prosodic phonological structure, as an account of a metrical system can only be as adequate as the theory of prosodic structure on which it is based. In particular, the work of Kiparsky (1977) has shown that the metrical theory of stress, as developed in Liberman and Prince (1977) and other works, provides the basis for a more adequate and explanatory theory of English meter than the numerological stress representation of The Sound Pattern of English. Kiparsky's work has in turn repaid its debt to metrical stress theory-it is surely an argument in favor of Liberman and Prince's system that it can illuminate the facts of English meter so welL At the level ofdescriptive adequacy that has now been reached, it is fair to say that the study of meter can serve as a window into phonological structure. With this purpose in mind, I offer here what I believe to be an improvement on Kiparsky's analysis; one that simplifies the rules, accounts for the data more accurately.., and links up the English system with what is known about metrical systems in generaL These improvements are the result of basing the rules on a different representation of English stress, drawn again from the work of Liberman and Prince (1977). I argue that the aspects of stress that are relevant to meter are embodied not in metrical trees, as Kiparsky assumes, but rather in metrical grids, which Liberman and Prince develop at the end of their paper in a treatment of rhythmic stress clashes. My goals are to clarify the metrical system of English and to shed light on a current controversy concerning the proper representation of stress.