Re: [tied] Perfect of Vedic i-

05-03-2004 18:54, tjodrikperson wrote:

> This is partly a test to see if one can use this group for these kind
> of questions.
>
> My questions is: the Vedic perf. of i "go" is iyâya (3.sg.) and îyúr
> (3.pl), how have these forms developed? If we follow LIV's
> reconstruction *h2ej-, I would expect a development *h2e-h2oi- > *Ha-
> ai- > *ai-/âi- > Vedic *e-/ai-/ây (2.sg. *e-tha or *aitha, 3.sg. *ây-
> a)
>
> Does anybody have the explanation?

Shouldn't it be *h1ei-?

One gets the correct forms if the reduplication vowel is *i, not *e (as
normally with *Cei/u(C)- roots _in Indic_, cf. juha:va [not *jaha:va]
from *g^Heu- 'pour'):

sg. *h1i-h1oj- > iya:ya (the first *y filling the hiatus)
pl. *h1i-h1j- > i:yuh.

You can have the same "coloured" reduplication even in the Sanskrit
perfect of some roots of the form *weC-/*jeC-:

vac- --> 3sg. uva:ca, 3pl. u:cuh.
yaj- --> 3sg. iya:ja, 3pl. i:juh.

Piotr


Piotr