Sarasvati (texts & arch.III)

[Erik Seldeslachts]

Michael Witzel wrote:

> We do not now, it is likely Pre_Indo_Aryan, precisely because of
> aika(-vartana)  = Pre-Vedic aika > Ved. eka. In Pre-Old-Iranian it would
> have been *aiva as in the old Iran. languages. Seems to be an old dialect
> difference. But a slim one.
>
> The usual other arguments (Varuna, Nasatya) don't work that well (due to
> the reform of Zoroaster: Ahura Mazda, *one* Nanghaithya in Videvdad).
>
> At any rate, the form of the IIr language in Mitanni is pre-Vedic : IIr
> sounds are preserved, *zdh, in Priyamazda ::  Ved. priyamedha :: Avest.
> -mazda... But I suppose you are well aware of all of this...


This is a typical example of the kind of selective argumentation we are
swamped with on this list.
aika- : long e and o have always considered to be diphthongs by the
Sanskrit grammarians and it is likely that they still were real diphthongs
in Vedic and even later times.(On the other hand there is also Old-Iranian
*aivaka- 'one' preserved in Persian yek).
I think it is not only wrong to consider the language preserved in the
Near-Eastern documents as IIr in stead of IA, but also there is no
conclusive evidence to state it is pre-Vedic. One could as well argue that
this language conserves some archaic features but on the whole presents a
transitional phase between Old-Indo-Aryan and Middle-Indo-Aryan: e.g.
satta(-vartanna) : MIA satta- 'seven', with assimilation of the consonant
cluster in OIA sapta-; Indara, with typical MIA vowel-epenthesis to resolve
the consonant cluster in OIA Indra-; Bardashva : Skt VRdhAZva-, with
evolution of initial v- to b-; etc.

Erik Seldeslachts
Universiteit Gent
Gent, Belgium