|
|
 E-mail this message to a friend
|
|
Title:
|
Syllabification and Rule Application in Harmonic Phonology
|
|
Author:
|
Caroline Wiltshire
|
|
Email:
|
click here to access email
|
|
Homepage:
|
http://web.clas.ufl.edu/users/wiltshir/
|
|
Degree Awarded:
|
University of Chicago
, Department of Linguistics
|
|
Degree Date:
|
1992
|
|
Linguistic Subfield(s):
|
Phonology
|
|
Subject Language(s):
|
Arabic, Egyptian Spoken
Yupik, Pacific Gulf
Inuktitut, Greenlandic
Ganda
Malayalam
Selayar
Tamil
|
|
Director(s):
|
John Goldsmith
Karen Landahl
Bill Darden
|
|
|
Abstract:
|
|
A constraining statement of phonotactics at the Word and Phonetic levels forms a necessary phonological component of grammar, and licensing structures (the syllable, coda, and appendix) play a crucial role in capturing these phonotactics at both levels. The harmonic approach pursued here takes phonotactics as the central component for a theory of phonology; that is, rules apply towards the satisfaction of phonotactics, representations capture phonotactics, and levels are defined by phonotactics.
In the dissertation, I show how the constraint-based approach differs from the generative phonologies of SPE and its descendant, Lexical Phonology. Originally in generative phonology, rules applied mechanically whenever their environmental specification was met; only the underlying form was constrained by Morpheme Structure Conditions or Rules (Stanley 1967, Chomsky and Hale 1968). It was quickly noted, fo example by Kisseberth (1970), Sommerstein (1974), and Singh (1984), that the generalizations and conspiracies about what a language permits phonologically in terms of segments, clusters, and syllables did not necessarily coincide with single morphemes. Derivational theories were modified to account for such generalizations by constraining the rules which applied to the well-formed underlying representations, using principles like Structure Preservation, the Continuous/Strong Domain Hypothesis, invisibility, and the interpretation of rules by strong satisfaction and the Elsewhere Condition.
Harmonic Phonology (Goldsmith 1989), on the other hand, aims to directly capture phonological generalizations by focussing on the phonotactics in language. This dissertation argues for the syllable, coda, and appendix as phonotactics. Using licensing theory, I examine the role of these phonotactics at the W and P-levels of a harmonic phonology, and show that constraints may play a role at each of the two levels. This examination of the role of syllabification shows that the model of lexical phonology not only fails to capture generalizations, but also that it is too powerful in providing for multiple applications of syllable structure rules or templates.
Chapter One provides an overview of the different types of 'resyllabification' in the literature. It also lays out the theoretical machinery to be employed and the points of contrast with existing theories. Chapter Two examines languages which have different coda constraints at W and P levels; these include Malayalam, Luganda, West Greenlandic, and Alutiiq. In these languages, I argue that a more restrictive coda licenser at the P-level accounts for harmonic rule applications, thereby supporting the role of constraints at that level. Furthermore, the reanalysis of Malayalam shows that more than two levels of syllabification (as proposed in T. Mohanan 1989) are unwarranted.
Chapter Three investigates the special properties of word edges, focussing on Cairene Arabic, Tamil, and Selayarese. These examples illustrate that invisibility to rule applications cannot capture generalizations on well-formed words, while the appendix licenser can; crucially, segments at the ends of words need to be special with respect to specific properties rather than completely invisible to the phonology. Chapter Four shows that, contrary to the expectations in a system where Structure Preservation can turn off post-lexically, languages tend to require obedience to syllable constraints at the P-level, although the constraints may be looser. Chapter Five examines the theoretical implications of the preceding chapters for the proposed limitations on rules (Structure Preservation, the Strong Domain Hypothesis, etc.) and argues that these, as previously formulation, have no place in a constraint-based model. In this model, the constraints do not 'turn off', and the rules are thereby simplified in terms of environment and application.
|
|
|
|
|
|
Page Updated: 27-Nov-2009
Please report any bad links or misclassified data
LINGUIST Homepage | Read
LINGUIST | Contact us
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.
|
|
