Japanese exhibits two patterns involving palatality: palatalisation, which causes two adjacent segments to share palatality, and de-palatalisation, which renders one of those two adjacent segments unable to sustain the shared palatal property. These patterns are traditionally analysed by referring to the notions of adjacency and/or precedence. By contrast, in the context of Precedencefree Phonology (Nasukawa 2014, 2015ab) this paper re-analyses these phenomena by referring to the head-dependency relations that are necessary for building structure, rather than by appealing to precedence relations. In this model, precedence is merely a natural result of interpreting the dependency relations that hold between units in hierarchical phonological structure.
In the pursuit of a strictly monostratal model of phonology, syllable/ prosodic structure is fully specified in lexical representations. Accordingly, information relating to the linear order of segments is redundant in representations: dependency relations holding between syllabic categories are sufficient to account for phonological phenomena. This paper therefore investigates the possibility of omitting from phonological representations all precedence relations between units, which would allow positional precedence to be viewed merely as a by-product of phonetic interpretation relevant to the sensorimotor systems. As such, the division between phonology and its external systems would parallel the division between syntax and performance systems.*
Based on the cross-linguistic tendency that weak vowels are realized with a central quality such as əә, ɨ, or ɯ, this paper attempts to account for this choice by proposing that the nucleus itself is one of the three monovalent vowel elements |A|, |I| and |U| which function as the building blocks of melodic structure. I claim that individual languages make a parametric choice to determine which of the three elements functions as the head of a nuclear expression. In addition, I show that elements can be freely concatenated to create melodic compounds. The resulting phonetic value of an element compound is determined by the specific elements it contains and by the head-dependency relations between those elements. This concatenation-based recursive mechanism of melodic structure can also be extended to levels above the segment, thus ultimately eliminating the need for syllabic constituents. This approach reinterprets the notion of minimalism in phonology by opposing the string-based flat structure.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.