Our department is delighted to welcome back Avery Ozburn (MA 2014), currently a Ph.D. candidate at the University of British Columbia. She is a phonologist whose work straddles theoretical and experimental approaches and is centered on harmony of vowels or of consonants. Her talk, "Why low vowels are special: Contrastive neutral vowels in Mayak ATR harmony," will be taking place in SS 560A at 3:00 PM on Friday, March 15. A reception will follow in the department lounge.
In the literature on vowel harmony, there is a long-standing tradition of equating neutral vowels, which do not undergo harmony, with vowels that are non-contrastive for the harmonic feature. Theoretical approaches to neutrality, then, often depend crucially on constraints against the harmonic counterpart of the neutral vowel. However, this generalization is based on a small number of widely studied languages; in fact, in the broader typology, a range of behaviours is attested: a vowel may or may not be neutral, regardless of whether it has a counterpart for the harmonic feature. In this talk, I examine a situation that is particularly problematic for many theoretical frameworks: in Mayak (Nilotic; Andersen 1999), low vowels are contrastively paired for ATR/RTR (advanced/retracted tongue root), yet generally neutral to ATR harmony. In Mayak, non-low RTR vowels alternate in the presence of an ATR vowel (e.g. [kɔc] 'take-PRES' vs. [koj-u] 'take-PAST'), while the low RTR vowel [a] does not (e.g. [ʔam] 'eat-PRES' vs. [ʔam-u] 'eat-PAST'), even though there is a contrastive low ATR vowel [ʌ] that is possible in the contexts in which [a] fails to undergo harmony. This case has been described in the literature but not analyzed, yet it is crucial to our theoretical understanding. Specifically, while it is common for low vowels to be neutral to ATR harmony, Mayak shows that this tendency is at least partially independent of harmonic pairing. I argue that this case is illustrative of a broader special status of low vowels in ATR harmony systems, and that this unique behaviour is phonetically motivated. The analysis I propose allows for ATR harmony to be relativized to non-low vowels, and this approach can capture the range of typologically attested neutrality behaviour in ATR harmony systems.
No comments:
Post a Comment