Probabilities improve stress-prediction in a CFG of Hawaiian phonology

Should probability play a role in linguistics? Whereas Chomsky (1957: 16–17) influentially rejected probability in syntactic theory, “[i]n phonological theory, probability has not so much been rejected as disregarded” (Coleman, 2003: 89). Disregard has, however, given way to a growing literature on the use of probability across the various linguistic sub-disciplines (see, e.g., Bod et al., 2003; Coleman, 2003). This paper is a case-study of probability in phonology, both as it applies to an improved description of Hawaiian stress-assignment, and as this description, in turn, reflects back on the probability question, above.