hawk/hock

[Alice Faber]

Mark A Mandel said:
>
>ISTR a finding, prob. by Labov, that people can maintain distinctions in
>their own production that they fail to recognize in other people's
>speech or their own.

Absolutely. Labov has several papers on the topic, and there's
extensive discussion in his 1994 book. In addition, Marianna Di Paolo
and I had a paper on this phenomenon in Language Variation and Change
in 1995. Marianna has some additional work specifically on the
cot/caught near-merger in Utah. Essentially, many folks produce words
in these classes more distinctly when they're not focussing on the
contrast. That is, if the lists of words they're reading into a tape
recorder just contain words in the COT and CAUGHT classes, there's
less distinction than if the words are embedded in a list in which
all English vowels are represented. In addition, people's accuracy at
labeling the two classes in a classic speech perception experiment is
only somewhat above chance, but they can make use of the presence or
absence of a distinction in making the social judgements implicit in
a matched guise experiment.

--
 =============================================================================
Alice Faber                                             faber at haskins.yale.edu
Haskins Laboratories                                  tel: (203) 865-6163 x258
New Haven, CT 06511 USA                                     fax (203) 865-8963