Published online by Cambridge University Press: 07 December 2020
This article examines a turning point in the history of the theory of phonological distinctive features. In Morris Halle’s (1959) The Sound Pattern of Russian, features are organized into a contrastive hierarchy designed to minimize the number of specified features. Redundancy rules, however, ensure that the resulting underspecification has no real phonological consequences and, in subsequent generative approaches to phonology, contrastive hierarchies were largely abandoned. We explore how Halle’s hierarchy would have been different if it had been based on phonological patterns such as voicing assimilation, and show that this reorganization makes plausible predictions about other aspects of Russian phonology. We conclude by pointing to recent work in which the concept of a contrastive hierarchy has been revived, illustrating the range of phenomena that this theoretical device can account for if minimizing specifications is not the primary concern.
We are grateful for the helpful comments and questions of three anonymous reviewers, and for those of participants in various conferences where we have presented work leading up to this article, including the Manchester Phonology Meeting, the North American Phonology Conference, the North East Linguistic Society, the Atlantic Provinces Linguistic Association, and the Slavic Linguistics Society.
Abbreviations for morphosyntactic properties follow the Leipzig Glossing Rules. Additional abbreviations used in this article: [c.g.] = [constricted glottis]; IPA = International Phonetic Alphabet; MS = morpheme structure (rule); [RTR] = [retracted tongue root]; RVA = regressive voicing assimilation; SPE = The Sound Pattern of English (Chomsky & Halle 1968); SPR = The Sound Pattern of Russian (Halle 1959).