Morphophonological & Morphosyntactic Domains

In this dissertation, I explore domains in the grammar in the context of Distributed Morphology (Halle & Marantz 1993; 1994). I propose that the morphological word, or topmost complex X0-node, defines a special linguistic domain for processes in: i) morphosyntax and ii) phonology. In the first part of the dissertation, I respond to syntactically-based analyses for suppletion, and in the second part of the dissertation, I analyze mismatches between phonological processes and ‘words.’ First, I show that the complex X0 forms a maximal morphological domain for suppletion. Recent analyses for word-external suppletive triggers propose that Vocabulary Insertion (VI) of a suppletive root is directly triggered by a word-external argument (Bobaljik & Harley 2017; Toosarvadani 2016; a.o.). However, cross-linguistically, participant number suppletion overlaps in its distribution with reduplication and affixation and is part of a broader phenomenon, verbal number. I present a unified structural account for participant number suppletion and affixation and argue that affixation suggests there is a node in the verbal word that expresses number. I propose that the number node mediates an agree relationship between the closest c-commanding argument and the verb. After licensing, number is realized as an affix or directly triggers VI of the suppletive root. The closest c-commanding argument, therefore, controls the form of the verb through agreement, but the trigger for suppletion is strictly local. In the second part of the dissertation, I explore how the morphological domain bears on phonological processes, such as Vowel Harmony and Tone. First, I present evidence from three Niger Congo languages which suggests that Vowel Harmony and Tone are not restricted by strict syntactic phases. Then, I investigate cases where there is one word at the surface level of the utterance but the harmonic and/or tonal behavior suggests that there is more than one phonological domain in the word. Phonologically, these ‘words’ behave similarly to compounds as compounds have been widely observed to contain more than one domain for vowel harmony. With supporting evidence from previous morphosyntacic analyses, I propose that harmonic and tonal asymmetries result from ‘words’ composed of more than one complex X0.