- Foreign Study
- News & Events
Back to Top Nav
Back to Top Nav
"Prosodic correspondence in Tgdaya Seediq: insights from corpus and experimental evidence"
Recent Dartmouth Linguistics major Jennifer Kuo '18, now an Assistant Professor of Linguistics/Phonology at Cornell University, joins us for the Department of Linguistics Homecoming Lecture to talk about her work in a talk titled, "Prosodic correspondence in Tgdaya Seediq: insights from corpus and experimental evidence."
She explains the work she will discuss as follows: "Related words tend to be similar; in English for example, singular-plural pairs like `goat/goat-s', where the pronunciation of the root word stays the same, are more common than pairs like ‘goose/geese’, where a part of the root's pronunciation changes. This similarity effect is typically formalized as a linear correspondence relationship, where sounds in the same ‘position’ match each other. In this talk, I present new evidence for a different kind of non-linear relationship, citing corpus and experimental evidence from Tgdaya Seediq (Austronesian, Taiwan).
Seediq has extensive vowel alternations between unsuffixed and suffixed forms, resulting in paradigms like [ˈbaro]~buˈragan]. Typically, we would expect the underlined vowels to be similar to each other, since they are in a linear correspondence relationship (both being the second sound in their respective words). For Seediq, however, I find that prosodic correspondence, which compares the stressed vowels of related words, overrides this linear relationship. Specifically, the boldface vowels in [ˈbaro]~[buˈrag-an] tend to match, even though they are not linearly in the same position.
In an experiment aimed at probing Seediq speakers' innate knowledge of prosodic correspondence, speakers were asked to produce novel suffixed forms. From this experiment, I find that speakers have learned and internalized the vowel matching pattern; they changed vowels in a way that would satisfy prosodic correspondence, and even over-extended these changes beyond environments observed in the language. The approach I take, which combines corpus evidence and experiments, could prove useful in finding more cases of prosodic correspondence."
Sponsored by the Department of Linguistics.
Events are free and open to the public unless otherwise noted.