From: Rob
Message: 34770
Date: 2004-10-19
> As far as I'm concerned, *pi-ph3-e/o- didn't count as decisiveFirst I have a couple questions :-P
> evidence as long as it was an isolated known instance of the
> voicing. I saw no way of ruling out other possible explanations,
> such as sporadic voicing dissimilation (*p...p > *p...b). But
> Hamp's elegant analysis of the "river" word *abon- as *h2ap-h3on-
> has strengthened the case for voiced *h3 considerably, and Olsen's
> demonstration that the Hoffmann suffix regularly induced
> assimilatory voicing in preceding stops practically proves that *h3
> was _distinctively_ [+ voice]. The only proviso is that we agree to
> reconstruct the suffix as *-h3on- (rather than, say, *-h1on-,
> which, if my memory serves me well, was Hoffmann's own
> reconstruction). I used to have doubts about it, but now, having
> done more reading on nasal suffixes and some more thinking on my
> own I have overcome my initial objections. I even think Olsen is
> probably right in identifying *-h3on- with the root *//h3en(h2)-
> // 'carry (a burden)', as in <onus>.