I had a lovely time at PHREND (PHonology Research weekEND) 2019, which was held at UC Berkeley this year. I presented the briefest of snapshots of my dissertation work and enjoyed hearing more about the development of Q Theory in Panãra (Myriam Lapierre, Martha Schwarz, Karee Garvin, Sharon Inkelas) and catching up with Emily Grabowski and learning about tone in Coatlán-Loxicha Zapotec. Cherry on the top of this conference was catching a beautiful sunset over the city and both bay bridges from Larry Hyman's gorgeous home in the Berkeley hills.
Ghost elements in Ende phonology
Kate L. Lindsey (Stanford University)
Ende phonology exhibits several phenomena where partially underspecified segments seem to appear and disappear at the service of phonotactics, much like yers in Slavic. Following Zoll (1996) and Kiparsky (2003), I call such elements ghosts. I will present two types of ghosts in Ende and show how the interaction of these two patterns informs formal theories on the representation of underspecification in the input.
Ende floating nasals demonstrate alignment of an underspecified nasal segment to the leftmost non-initial obstruent in the word, much like how stress and affixes may be aligned to left or right edges of stems or feet (McCarthy & Prince 1993) or how tone patterns may spread to adjacent tone-bearing units. A phonotactic analysis of the Ende dictionary and corpus reveals that prenasalization is a contrastive feature of morphemes, much like nasalization in Máíhɨ̃ki (Sylak-Glassman 2013). Ende phonotactic reduplication displays semantically vacuous copying of segmental structure to repair verb roots that violate a phonotactic constraint on word minimality. Monosyllabic verb roots reduplicate in isolated forms, but multisyllabic verb roots do not. Curiously, morphological structure also seems to play a role.
Representing both ghost elements as subsegments in the input allows for straightforward constraint-based analyses of the phenomena independently. However, when the two ghost patterns co-occur in the same word, a ranking paradox arises. This puzzle is solved if the two types of ghost elements are represented distinctly in the input.
Kate L. Lindsey (Stanford University)
Ende phonology exhibits several phenomena where partially underspecified segments seem to appear and disappear at the service of phonotactics, much like yers in Slavic. Following Zoll (1996) and Kiparsky (2003), I call such elements ghosts. I will present two types of ghosts in Ende and show how the interaction of these two patterns informs formal theories on the representation of underspecification in the input.
Ende floating nasals demonstrate alignment of an underspecified nasal segment to the leftmost non-initial obstruent in the word, much like how stress and affixes may be aligned to left or right edges of stems or feet (McCarthy & Prince 1993) or how tone patterns may spread to adjacent tone-bearing units. A phonotactic analysis of the Ende dictionary and corpus reveals that prenasalization is a contrastive feature of morphemes, much like nasalization in Máíhɨ̃ki (Sylak-Glassman 2013). Ende phonotactic reduplication displays semantically vacuous copying of segmental structure to repair verb roots that violate a phonotactic constraint on word minimality. Monosyllabic verb roots reduplicate in isolated forms, but multisyllabic verb roots do not. Curiously, morphological structure also seems to play a role.
Representing both ghost elements as subsegments in the input allows for straightforward constraint-based analyses of the phenomena independently. However, when the two ghost patterns co-occur in the same word, a ranking paradox arises. This puzzle is solved if the two types of ghost elements are represented distinctly in the input.
RSS Feed
