The Morphophonology and Morphosyntax of Kerinci Word-Shape Alternations

Abstract:

THIS DISSERTATION examines the grammar of a dialect of Kerinci spoken
in the village of Tanjung Pauh Mudik. Kerinci is a Malayic language
spoken primarily in the Kerinci Regency, in the mountainous western
part of Jambi Province, Sumatra, Indonesia.

In many Kerinci dialects, lexical items exhibit two distinct forms which
differ from one another in the phonological realization of their final
syllable rime. These two morphological forms developed historically
from a phrasal-level phonological alternation, and have been referred
to in the literature as the `absolute' (historically phrase-final) form and
`oblique' (historically phrase-medial) form (Steinhauer and Usman
(1978) inter alia).

The aim of this dissertation is to understand the grammatical factors
which determine the phonological shape and syntactic distribution of
absolute and oblique forms. From a phonological perspective, I
investigate whether the relationship between the phonological shape of
absolute and oblique forms is predictable. From a syntactic
perspective, I attempt to determine whether the properties of the
alternation within the nominal and verbal domain can be explained via
the same grammatical mechanism. In answering this question, I
consider whether the choice between absolute and oblique is purely
syntactic (i.e. structure dependent), or whether it involves a
combination of syntactic and phonological factors (e.g. linear word
order).

In the verbal domain, I argue that the O-form is an incipient marker of
nominal agreement which functions to license null pronouns in certain
contexts. This proposal explains the observation that TPM appears to
permit extraction of non-subject arguments, a fact which runs contrary
to the robust typological generalization that languages with `Philippine-
type' voice systems do not permit extraction of non-subject arguments.

Within the nominal domain, there is evidence to suggest that the
alternation is conditioned by linear word order. Despite this, I argue
that the same underlying mechanism (nominal agreement) accounts for
the alternation within both domains. This unified analysis is possible on
the assumption that both nominal and verbal domains exhibit
Larsonian-type structures in which adjuncts are treated as
complements. Thus, I conclude that the alternation is determined by
purely syntactic factors in both nominal and verbal domains.[]

_____________

Timothy McKinnon, “The Morphophonology and Morphosyntax of Kerinci Word-Shape Alternations”, Ph.D-thesis, University of Delaware, 2011.