This article investigates several questions of harmonic function using aggressively data-driven approaches. We apply Hidden Markov Modeling-a technique used to identify contextual regularities within streams of data-to the Kostka-Payne, McGill Billboard, and Bach chorale corpora. The resulting models question the generalizability of the traditional three-function model, illustrating the syntactic uniqueness of various corpora while also highlighting recurrent characteristics of tonal repertories. Finally, this article offers some general observations, including questioning the role that tonal hierarchy plays in theories of function and discussing the cultural politics inherent in assuming the universality of one functional system.
Mendeley helps you to discover research relevant for your work.