Hannah Sande, Georgetown University
Phase-bounded phonology in Cophonologies by Phase
This talk showcases three very different interactions between morphology and phonology, all of which show that phonological alternations, specifically morphologically conditioned phonological alternations, are sensitive to syntactic phase boundaries. I introduce the Cophonologies by Phase model (Sande and Jenks, 2018; Sande, 2019), which relies on two main assumptions: 1) an expanded notion of vocabulary items in a Distributed Morphology approach, where morphosyntactic feature bundles can be associated with partial phonological constraint rerankings (or re-weightings), in addition to being associated with a (supra)segmental phonological form, and 2) phase-based spell-out and application of phonology. This framework is shown to account for a wide range of phenomena, including morpheme-specific phonological effects that can affect words, multiple words, sub-word domains, and cumulative morpheme-specific phonological effects within a phase domain. The model makes specific predictions about boundaries that we do not expect morpheme-specific phonological alternations to cross. The three primary case studies considered come from Kuria (Bantu), Guébie (Kru), and Amuzgo (Oto-Manguean), and each type of process is shown to be quite common across languages. I argue that with morpheme-specific phonology and phonological evaluation at phase boundaries, Cophonologies by Phase can account for these processes in a unified but restricted way.