Odp: boskO

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

From: Christos Galanis

To: cybalist@eGroups.com

Sent: Monday, January 31, 2000 10:37 PM

Subject: [cybalist] boskO

Piotr Gasiorowski wrote: > > No, and it could't be because Gk b cannot correspond to PIE *p. The > Greek word is sometimes associated with Lithuanian guotas 'herd' < > *gWoH-to- or *gWaH-to-; the verb could be *gW@... It could have > something to do with the cattle word *gWo:us if one reconstructed it as > a u-stem *gWoH-u- (as some people do!). P changes to b, in Gr. (cf. bateO instead of pateO). A PIE *p?sc- could rather well be transformed to Gr. boskO. - Christos I concede that irregular development may sometimes be admitted, but one can't assume it whenever there seems to be some degree of tempting phonetic similarity or other impressionistic justification. That's why the equation *pokW- = Arm. hacH is of no value (while *paxsk- = hacH is a good match), or why *pekW- is not equated with *bha(:)g- 'bake' despite the place-of-articulation match and a similar meaning. By admitting *paxsk- > bosk- (what about the Greek vowel quality??) we would be relaxing the formal constraints of the comparative method beyond what's normally acceptable.   Piotr