Re: [tied] *pYerkW+

On 2007-05-23 03:26, Sean Whalen wrote:

> For this I did give a possible mot. (that rsn is a
> long uncommon chain of dentals). A dissimilation of
> rsn > rsm would seem fine to me; since it's followed
> by u there is a switch instead of straight
> dissimilation.

*-rsn- is not particularly uncommon in PIE: *kWr.sno- 'black' and
*persnah2 (~ *pe:rsni-) 'heel' are well-known examples of wide-ranging
words in which the clusters occurs (not to mention *tr.snah2 itself or
more complex structures like *-rh2sn-).

> Abstract changes based on mental classification are
> at least as important as those based on perceptual
> phonology.

Fine on paper, as I said, but where are all those tu > pi changes that
should have happened? If you insist on the importance of the dental
environment, let's imagine something comparable, like rstu > rspi. My
gut feeling as a phonologist is that there is something fundamentally
wrong with any model that predicts this as a plausible sound change.

Piotr