Re: [tied] Vedic Rta... one last time

Sean Whalen wrote:

> What about the Old Persian forms? I don't think
> there's any possibility of hrt > s^ there. It looks
> like free variation to me (with OP having a greater
> tendency to analogic recreation of -ta-, etc.
>
> Av. OP
>
> m@...^a-/m@...@ta- marta- dead
> m@...@Tyu- mars^iyu- death
> mas^ya- martiya- mortal
> p@...^u-/p@...@tu- bridge
> as^a- ars^ta:- true, truth
>
> I don't see any way to use accent to account for
> "bridge" variation; OP has no devoicing + coalescence
> to account for other forms.

Whatever happened in OP seems to be independent of the Avestan
development, which is quite clearly correlated with the location of
accent. As for *pr.tu-, an accentual doublet wouldn't surprise me, since
the original paradigm before the levelling-out of the root vocalism must
have been proterokinetic, i.e. something like *pért-u-s, gen. *pr.t-éu-s.

>>>What about k@...; is the accent on r there too?
>>
>>Yes. I assume you mean <k&hrp-> 'shape'; it's a root
>>noun, after all, so
>>the accent has nowhere else to fall on in the strong
>>cases!
>
>
> Do you mean this was original PIE or analogy?

Can't it be both? "Originally", the vocalism of the nom.sg. was
certainly full, but how do you know that the analogical nil grade does
not date back to the protolanguage?

> But it's marz^-di-ka- with no schwa; both variations
> are explained by variation in syllabification. I used
> this as an example because you said that Av. ar came
> from er or or while I said that syllabic r > ar in
> closed syllables or word-final in Av. and r > ar in OP
> except after m, etc.

Ah, well, no disagreement with that. I only objected to your strong
statement that the /a/ in <as^.a-> MUST be due to *h2-metathesis.

Piotr