23rd Manchester Phonology Meeting, University of Manchester, Manchester, UK
The file is open to annotations via Hypothes.is. Comments are always welcome!
I present the results of a phonetic study of the relationship between vowel quality and vowel quantity. The results confirm that the descriptions are correct in setting up a categorical distinction between two varieties of long mid vowels (tense [eː oː] and lax [ɛː ɔː]) in South-West Welsh. This system plausibly arose as the enhancement of a pattern where there is no quality distinctions between short and long vowels, with continuous variation in height in both prosodic contexts. The rise of the south-western pattern is connected to the phonologization of a trading relation in duration between the stressed penultimate and the final syllable, which is still attested in the relevant varieties. This presents an interesting example of ‘rule scattering’ (Bermúdez-Otero 2014).
The handout is much more detailed than the slides. However, there is a rather embarrassing misinterpretation of the durational effect for the ‘non-enhanced’ system. I’m leaving this unchanged in the interest of historical veracity, but see my Gregynog presentation, once it’s available, for the corrected version.
Also, I plan on releasing the data behind this presentation fairly soon — check back later for a link.