My colleague Nancy Niedzielski and some of her students have started work on the Houston Urban English Survey and some preliminary results were presented at our grad workshop last weekend. Andrew Pantos and Elizabeth Gentry gave presentations ontheir summer work with long-term (40+y.o) and teenagers respectively. They’d earlier presented their work at NWAV.
One very interesting thing that came out of the work was in relation to /ai/ monophthongisation. It’s one of the sibboleths of “Southern” speech but the environments for monophthongisation are different in different area. InTexas it usually occurs in open monosyllables (“my” is [maː] or [mæː]) and before voiced stops (“ride” is [raːd]), but not before voiceless stops (so “night” is [najt], not *[naːt]). However, the teenagers in Elizabeth’s study didn’t monophthongise, and explicitly associated it with “Southern” and/or “older” speech. This was the case even when their parents do monophthongise.
From the point of view of characterising sound change, this is really interesting. Would we want to characterise this as a reversal in sound change? Or is it a change in rule application? That is, speakers have a rule that monophthongises /ai/ to [a:] or [æː] in some environments, but in the kids they don’t apply that rule on the model of more prestigious “northern” or “western” dialects?