Re[4]: [tied] Re: Nominative: A hybrid view

At 9:22:55 PM on Sunday, June 8, 2003, Miguel Carrasquer
wrote:

> On Sun, 08 Jun 2003 20:53:11 -0400, "Brian M. Scott"
> <BMScott@...> wrote:

>>At 6:13:18 PM on Sunday, June 8, 2003, Miguel Carrasquer
>>wrote:

>>> For OIr. tír < *ters-, I see no other explanation than
>>> lengthening caused by the *-s-. Do you?

>>For what it's worth, Thurneysen tentatively suggests an
>>original suffixed *ters-r-, yielding *te:sr and OIr. tír.
>>He compares this with OIr. mír 'morsel' < *me:msr-.

> Yes I saw that, but I didn't quite see why -rs- couldn't
> and -rsr- could have yielded -:(s)r-.

Examples like <barr> 'top, tip, end' and <err> 'arse' seem
to show -rs- > -rr- without lengthening, so the question
would presumably be why -rsr- could have yielded -:(s)r-.
In view of Sect. 216 I wonder if he's thinking of something
like -rsr- > -sr- > -:r-.

> The example of mír is not quite to the point, as the vowel
> in *me:msr- already *had* length of itself.

Yes, I did wonder about that, unless he was offering it as
evidence of what happens to such a consonant cluster.

Brian