Government Phonology
From Wikipedia, the free encyclopedia
 |
Please help improve this article or section by expanding it. Further information might be found on the talk page or at requests for expansion. |
This article is linked to the wikipedia section called phonology, in order to understand the ideas of phonology generally the reader is invited to pereuse the said page. This page is intended as a link from that phonology page.
Government Phonology's main aim is to present a principle and parameter based framework to phonology. Unlike the constraints of optimality theory, a principle inviolable, and the rest is parametrised. Government phonology is a representational framework, meaning unattested forms are equally warranting investigation. The main basis of Government Phonology is the syllabic structure. The only constituents in phonology are thought to be the familiar onset and nucleus, as a wealth of data was used to eliminate the 'coda' as a constituent. Therefore the /t/ in a word like /kat/ is considered to belong to the following onset with an empty word final nucleus (strong evidence for this comes from Italian and Icelandic 'Metrical Lengthening' (Kaye 1990, Harris and Gussmann 1998).
There are four tiers in Standard Government Phonology: the melodic tier (for the segments), the skeletal tier (which buffers the melodic and the constituent tier), and the autosegmental tier known as the nuclear projection.
In a lexical domain (word) all the positions (skeletal points, constituents) must be licensed save one, the head of the domain (KLV 1990). The head of the domain will receive the stress for that domain (Charette 1991, Harris various). Phonological processes are all seen in terms of government (where, simplistically speaking, phonotactic constraints hold) and licensing (where phonotactic constraints do not hold).
Contents |
[edit] Scope and Aims of Government Phonology
[edit] Select History of the Framework
In recent years there have been a number of off-shoots of GP the most important being Lowenstamm's strict-CV developed to popularity by in particular Tobias Scheer [1](more on strict-CV forthcoming).
Dienes and Szigetvári (1999) in 'Repartitioning the Skeleton: VC phonology' introduced the notion that the phonological syllable may not be CV underlyingly but VC and this idea has gained some popularity, it is available for download here: [2].
[edit] Perceived Problems of Phonology
The aim of GP is to reduce arbitrary, language-specific rules and attempt to engulf all phonological phenomena in a universal principles and parameters cognitive component, a phonological universal grammar.
[edit] SPE Rules
The The Sound Pattern of English was a major influential work. The problem with ‘SPE’, however, is that it created purely language-specific rules which were not falsifiable, parsimonious (see Occam's razor), nor predictive.
Given the rule {[d] ---> [h] /__[r]#}, one could not know if this was unattested in every language. In other words, there are no constraints on positing rules. It is un-parsimonious as one would require a second rule to show that {[s] ---> [y]/__[r]#} and these rules cannot predict what would happen to [t] in the same environment.
