| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 9.25 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
At the interface of morphosyntax and phonology, some phonological behaviors in Chinese
languages are sensitive to word domain (stress assignment/stress resolution and tone sandhi).
In this thesis, we focus on how morphosyntactic structures can contribute to some
phonological behaviors that remain to be puzzles in the Chinese languages. Additionally, a
highly-functional morphosyntax-based framework is shown to be realistic to construct a
simplified and consistent model in domain construction of T3 tone sandhi in Chinese
Mandarin, which has been considered challenging in the literature.
Following “Little x heads” theory (Marantz 1995; Marantz 2001) and syntactic incorporated
compounding structures (Harley 2009), we use a syntactic multiple-root incorporated
structure for Chinese compounding structures to account for the stress assignment and stress
resolution (stress clash avoidance) in Shanghai Chinese with revised Phase Impenetrability
for Phonology (rPIP) (Embick 2013). Meanwhile, a tentative Concatenation rule (Pak 2008;
Chen 2018) after Linearization of Morphological words is proposed to account for the
domain construction in T3 tone sandhi in Mandarin Chinese, which refers to specific
morphosyntactic information (morphosyntactic locality characteristics and c-command
relations). Different from the literature, we add the syntactic multiple-root incorporated
structure of Chinese compounding structures into the algorithm of Concatenation rule. This is
proved to be essential to successfully construct a unified framework of T3 tone sandhi in
Mandarin Chinese both above and below the classical word domain, showing a noteworthy
ability to deal with the exceptional situations in Chen (2009), e.g., syntactic words,
phonological words and complex predicates. This project supports that morphosyntax-based
analysis under syntactic word formation, e.g., Concatenation rules in Distributed
Morphology, is a powerful weapon to reveal the processing logic of some controversial
phonological rules vaguely floating between the classical lexical and postlexical rules in the
literature, e.g., sandhi behaviours.
Under the current framework, differently from multimorphemic structures, the
monomorphemic structures seem to be opaque in the application process of specific
non-cyclic phonological rules. Such opaque monomorphemic structures can be postulated to
be a product or outcome of certain phonological rules’ processing economy and efficiency,
instead of a true grammatical identity.