/Vlk/ --> /Vhk/ --> /Vk/

[Donald M. Lance]

Rudy, no disagreement.  I knew they weren't parallel.  I just couldn't think of a truly
parallel example.  There may be some articulatory feature in the /ae/ of Sonja's
population that does not allow a vocalized /l/ before /k/.  Have the srimp-speakers *lost*
a contrast?  Did this dialect have such a contrast historically?  How would an
"unsophisticated" srimp-speaker write this word in a dictation exercise?  I'm essentially
questioning the practice of assigning the same underlying forms in all dialects of the
same language and then applying the same phonological rules to derive all surface forms.
In the case in question, does the /l/ HAVE to go through a vocalization stage?  Generally,
vocalization affects the preceding vowel, in which case we would need an additional
fronting rule to take Sonja's speakers' vowels back to the articulatory position of /ae/
in other 'ak' words like axe.
DMLance

Rudolph C Troike wrote:

> Disagreeing with my good friend Don Lance, Southern /SrImp/ becoming
> /srImp/ (loss of contrast of /S/:/s/ before /r/) is in no way parallel to
> the vocalization and loss of /l/ in the context V__k (or more generally
> V__C). The closest thing to an inverse parallel is Southern loss of /r/
> following a (voiceless) dental fricative and preceding a back vowel, as in
> "from", "throw", "through".
>
>         Rudy