Headmost Accent Wins: Head Dominance and Ideal Prosodic Form in Lexical Accent Systems

Abstract

'Headmost Accent Wins' investigates the accentuation of lexical accent systems within the framework of Optimality Theory. The central claims of the book are: first, words with a lexical accent have unpredictable stress but predictable prosodic shape, and second, prosodic structure is built on the basis of morphological structure. A lexical accent is an autosegmental feature which is phonetically realized as stress or pitch according to language-specific constraints. Even though the specification of accents is free and unrestricted, independent prosodic constraints on word form limit their distribution. As a result, accented words have a strictly binary prosodic structure. Freedom of the input, on the one hand, and constraint ranking on the other derive a confined set of 'ideal' prosodic forms for words with lexical accents. Conflicts among lexical accents for prominence are resolved by morphology. The prosody-morphology interface centers around the principle of prosodic compositionality. It is articulated in terms of a 'theory of head dominance', which states that the accent of the morphological head of the word prevails over other accents. The theory of head dominance is tested in a number of morphological constructions in languages with different types of morphology (i.e. fusional, polysynthetic). In addition, it is shown that head dominance voids the need for the complex derivational machinery of cyclic and non-cyclic levels. Moreover, it directly derives the effects of the metaconstraint ROOTFAITH >> SUFFIXFAITH (McCarthy & Prince 1995) and, more importantly, it accounts for the counter examples to this metaconstraint. This book is of interest to metrical phonologists, linguists working on the prosody-morphology interface and researchers interested in Greek, Russian and the Salish languages.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 204 citations exist upstream.