On 2006-07-04 00:19, Abdullah Konushevci wrote:
> Really, but *bh > Greek ph: *bhH-n-yo > phainein 'to shine', Alb. <bënj>
> 'to do, to bring to light'; *bha-k'o > Greek phagos 'lentil', Alb.
> <bathë> 'broad bean'; *bher- > Greek pherein, Alb. bie, Lat. ferre.
What do you take me for, Abdullah? :) *bH does give /pH/ in this root,
cf. the derivatives I quoted, such as <óropHos>. However, in the present
tense of the verb 'to cover' (*h1rebH-je/o-) the *bH was followed by *j.
The pre-Greek sequence *-pHj- developed regularly into /-pt-/ (via a
cluster with a prehistoric affricate), merging with the reflex of *-pj-.
As an independent illustration, consider <tápHos> 'grave' vs. <tHapto:>
'bury'. Both these words derive from preforms with *dH..bH- > *tHapH-.
In the former, Grassmann's Law applies; in the latter (another *-je/o-
present stem) it doesn't, since the aspiration of the second stop
disappeared in the process of palatalisation, which was earlier than
Grassmann's Law in Greek.
Piotr