On 2006-03-30 20:31, Patrick Ryan wrote:
> This is no proof and not even any persuasive indication that PIE
> -*ró _ever_ was "dissimilated" to -*ó, under any circumstances.
One can actually see this process at work in Gk. argós 'shining, bright,
swift' < *h2r.g^-(r)ó-, cf. Ved. r.jrá- 'swift'. The compositional form
is always *h2r.g^-i-.
> PIE -*u/*ú (from -*w) is an affix that is unrelated to -*ó
> (plurality) or -*ró (high degree); and has a totally different
> significance: it provides the idea of definite but limited
> repetition and successfully completed activity so that *kr.t-ú would
> primarily mean 'empowered, provided with power'.
The distribution of *-ú- and *-ró- is more or less complementary. They
are only exceptionally formed from the same root, and there's certainly
a strong tendency to use *-ú- after roots containing *r (or, more
generally, a liquid). When they do occur as variants, as in Gk. elapHrós
'light (in weight and movement), nimble' ~ elakHús 'small, little'
(where's "the idea of definite but limited repetition and successfully
completed activity"?), the difference in meaning is minimal. <elapHrós>
means practically the same as RV ragHú-. All these words reflect
*h1ln.gWH-ú/ró-; both variants share the same comparative,
*h1léngWH-(i)jos-.
Piotr