This question is similer to my previous question.

I came across a person who makes the following claims:

The names of Previous Kings of Mitannis that are mentioned on inscriptions belongs to period of Late vedic era we don't find Names and such words used in Rigveda and Its a Undisputed fact that whole rigveda was Composed in Sapta Sindhu or Panjab-Doab Region. Plus the Numbers and Color names that were found on the Inscriptions are also Late Indo Aryan Prakriti words and cannot be Prior to Rigvedic terms and words. Plus it don't even have Iranian Feature In Which Sa changes to Ha For e.g the term Sapta is Same as Prakriti and Pali "Satta" and not Iranian Hapta Or Hafta. Also they Celebrated seasonal Solistice as their main festival and Term for that is also Belongs to Prakirt Late Indo Aryan voccabulary......(Vishuva) (still found pahari languages of North West as Vishu/Visova or Basoa or also called Baisakhi). How ever these features are absent in Early Vedic Times Indicating the mitanni ancestors were immigrants from Indus valley-(1700B.C) or post Vedic times.

Now, as we know, the Indo-Aryan loanwords in Mitanni inscriptions are said not to be derived from Vedic Sanskrit, but from Proto-Indo-Aryan. But how do linguists determine this?

  • 2
    I don’t know very much about the Mitanni inscriptions or forms, but presumably this conclusion was reached because the attested forms match the reconstructed PIA values better than their Sanskrit counterparts – e.g., the numeral ‘one’ is attested as aika, whereas in Indic the diphthong *ai had merged into e and the Vedic form is eka; and Iranian uses *aiwa- for the numeral, so it can’t be an Avestan form either. – Janus Bahs Jacquet Jun 5 at 14:40
  • 1
    @JanusBahsJacquet. I think aika/eka is purely an orthographic matter and does not prove the greater archaism of the Mitanni form. – fdb Jun 5 at 16:29
  • 3
    @fdb Do you mean in Sanskrit or in the Mitanni forms? A quick Wikipedia glance gives a few examples of retained sibilant before voiced dental as well (priiamazdaPriyamedha, mišta-nnu ‘payment’ ≈ mīd̥ha), which can’t be purely orthographic, unless one believes the Sanskrit retroflex actually represents /zd/ like Gk. ζ (but does anyone believe that?). – Janus Bahs Jacquet Jun 5 at 16:40
  • @JanusBahsJacquet believe what, that ζ represented /zd/? I think I've read there was some debate about the values and not everyone seems to be confident with the results. – vectory Jun 8 at 13:54
  • @vectory It’s not entirely certain whether ζ actually represented only /dz/, only /zd/ or both, and to what extent they were distinct at what periods; but it is generally agreed that it is at least likely that ζ was at some point used to represent what would likely have been /zd/ at the time in question. But the ‘who believes that?’ bit was about Sanskrit – I’ve never come across the notion that the retroflex stops टठड should (also) represent /st, sth, zd/ in Sanskrit, and it seems unlikely. – Janus Bahs Jacquet Jun 8 at 14:17

Because there are certain sound changes that happened between Proto-Indo-Iranian* and Vedic, and we don't see those sound changes in the Mitanni texts.

For example, the second half of the name Bi-ir-ya-ma-aš-da is almost certainly from the Indo-Iranian word for "wisdom", *mazdháH. (Compare Avestan mazdā, as in Ahuramazda "lord of wisdom". In Hittite/Hurrian cuneiform, š is the only sibilant fricative, so we can't draw more detailed conclusions about its value from this.)

But in Vedic and its ilk, sibilants disappeared before voiced dental stops: the Vedic cognate is medhā. If the word had come into Mitanni from Vedic, where did the sibilant come from?

(How do we know it's not Iranian either? As your source mentions, we don't see a change of /s/ to /h/ that's characteristic of Iranian: Mitanni ša-at-ta "seven", Sanskrit saptán, Avestan hapta.)

In the comments, Janus Bahs Jacquet also mentions a-i-ka- "one" next to Vedic eka, but as fdb points out, it's not clear whether e in Vedic was a monophthong or a diphthong. Same issue with Mitanni au next to Vedic o. The sibilants before stops seem more convincing.

* I'm using "Proto-Indo-Iranian" here to refer to the last common ancestor of Sanskrit, Avestan, Persian, etc. Confusingly, I've seen some sources who use "Indo-Aryan" for a superset of "Indo-Iranian", and some who use it for a subset, and I'm not sure which is more standard/correct. So swap out the terms if appropriate.

  • 1
    Normally, Indo-Aryan is a subset of Indo-Iranian, which also includes the Iranian and Mittani-Aryan branches. There is no obvious reasons to believe that Mittani-Aryan derives from Indo-Aryan and comes from India. – Arnaud Fournet Jun 5 at 19:33
  • 2
    @ArnaudFournet. "derives from Indo-Aryan" does not equate with "comes from India." – fdb Jun 5 at 20:35
  • 1
    @fdb There’s also <z> which was probably some sort of sibilant affricate, like in other languages in the area (panzapánca). – Janus Bahs Jacquet Jun 5 at 21:18
  • 2
    @fdb Yep, as Janus said, <z> also existed and seems to have been a sibilant affricate. For example, when the nominative ending -š is applied to a root ending in -t in Hittite, the result is written -z. – Draconis Jun 5 at 21:57
  • 1
    "In Hittite/Hurrian cuneiform, š is the only sibilant fricative, ..." Wow, did Hittite use the Latin script? So cool. – vectory Jun 8 at 13:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.