Re: [tied] PIE *h3 and PPIE **n

You wouldn't convince _me_ either, Miguel. A hypothetical *sah2wlr is bad enough (and you would need to explain why the syllabification should be what you propose rather than [sah2ul@...]), and a Gen. sh2wél-n-s instead of *sh2ul-én-s or *sh2w@... is odd, to say the least. Again, the syllabification of *-ln-s as [-ln@...] rather than [-l@...] seem unacceptable.

Even the analogic influence of -r/n- heteroclita is a more convincing solution.

Piotr

----- Original Message -----

From: Miguel Carrasquer Vidal

To: cybalist@egroups.com

Sent: Wednesday, December 13, 2000 1:41 AM

Subject: Re: [tied] PIE *h3 and PPIE **n

The -l/-n- stems (e.g. *sáh2wl, G. *sh2wén(o)s "sun") are a mystery to me. All I can think of, but I haven't been able to convince myself, is something like: N. *sáh2w[@]lr > *sáh2w[@]ll > *sáh2wl G. *s[@]h2wéln[@]s > *s[@]h2wénn[@]s > *sh2wéns with /lr/ > /ll/ > /l/ and /ln/ > /nn/ > /n/.