Variable length vowels

[Sharon L Hargus]

You could take a look at John Alderete's article 'On Tone and Length in Tahltan (Northern Athabaskan)' in Hargus and Rice (eds.) Athabaskan Prosody.  The vowel length issue he discusses does not involve word-final vowels, however.
____________________________________________________________________________
Sharon Hargus                          http://faculty.washington.edu/sharon
(206) 685-4263                         FAX (206) 685-7978
Department of Linguistics
University of Washington
Box 354340                    or       A210 Padelford Hall, Stevens Way
Seattle WA 98195-4340



On Wed, 16 Jun 2010, James Crippen wrote:

> In Tlingit there are a number of word-final vowels that seem to have
> variable length, meaning that they are pronounced short by some
> speakers and long by others, or even variably for the same speaker.
> This is despite regular distinctions in vowel length elsewhere. I’ve
> taken to transcribing them in IPA using a “half colon” /ˑ/, with the
> usual triangular colon /ː/ for true long vowels.
>
> Here are a couple of examples:
>
> /ʔaːníˑ/ |ʔaːn-í| ‘land-pss’ = [ʔɑːnɪ́] ~ [ʔɑːní] ~ [ʔɑːníː]
> /χuˑ/ |–χuˑ| ‘among (inalienable)’ = [χʊ] ~ [χu] ~ [χuː]
>
> There are however minimal pairs of word-final vowel length:
>
> /tʼá/ ‘king salmon (O. tschawytscha)’ = [tʼʌ́] ~ [tʼá] ~ [tʼɐ́] ~ [tʼə́]
> /tʼáː/ ‘board’ = [tʼáː] ~ [tʼɑ́ː]
>
> My question is whether or not such a phenomenon occurs in any
> Athabaskan languages that list members are familiar with. Specifically
> in word-final position, but perhaps anywhere. Is there any good
> literature I can read on this phenomenon?
>
> Thanks,
> James
>