出版社：Department of Linguistics, University of Toronto
摘要：In this paper we present an analysis of the relationship between sonorancy and voicing based on a theory of segmental structure which recognizes the possibility that voice may not be a unitary phenomenon. We propose that voicing has two distinct realizations. One is through the activation of laryngeal features and the other is spontaneous voicing. We propose that sonorants involve a node which represents spontaneous voicing and that this may also be present in obstruents. We adopt the term SV for this node (see also Piggott 1989). We thus view spontaneous voice as distinct from laryngeal features, sharing the general position of Stevens & Keyser (1989) who state that "voice might be classed as a manner feature" separate from the laryngeal features which deal with laryngeal configurations. An advantage of the feature geometry that we propose in this paper is that it allows an account of sonorant-sonorant interactions, as sonorant features such as [nasal] and [lateral] are both dominated by the SV node. A further advantage is that we are able to shift the burden of explanation from the rule component to the representational component. By recognizing two types of voicing, we eliminate the use of redundancy rules in the specification of voice for sonorants as a method of accounting for phonological processes.
