4

I am new here and to linguistics. Recently I have developed a passion and an interest for linguistics, but I am not familiar with it.

So I got into debate with a person from India. He was claiming Avestan is a daughter language of Vedic Sanskrit.

I showed him Wikipedia which says Avestan is sister to Vedic Sanskrit and not its daughter and that all linguistics have confirmed this, but he asked me how we can be sure about that, so I told him that a good historical linguist should be able to disprove the hypothesis easily.

The normal way of showing these things is to look for irreversible sound changes, when sounds have merged together. It shows that Vedic as attested cannot be derived from Avestan as attested and vice versa, which is why the consensus amongst linguists is that the two languages share a recent common ancestor, called Proto-Indo-Iranian.

To which he replied, “What are some of those irreversible sound changes?”

Now due to my insufficient knowledge of linguistics I am stuck here.

Can anyone help me with this?

What are some of the irreversible sound changes and other linguistic evidences (other than the lack of retroflex sounds in Avestan) which prove that Avestan as attested cannot be derived from Vedic and vice versa

5
  • 2
    Here's one: PIE’s palatal K satemized to S in Avestan, indistinguishable from regular S, whereas S and Ś (also written Ç) are distinct in Sanskrit. May 22 at 10:31
  • Bert Barrois@ i didnt get it. Can you please elaborate ?
    – Nikos
    May 22 at 22:47
  • 1
    Avestan merged S < PIE *Kʹ and S < PIE *S, whereas Sanskrit maintains the distinction Ś < *Kʹ versus S (or Ṣ) < *S. The RUKI rule governs the choice between Ṣ and S. If you want examples, just look at the words for horse and dog: EKʹWOS and KʹWON in PIE. May 23 at 10:54
  • 1
    @BertBarrois. Avestan does not “merge” IE *s and *ḱ. *s usually becomes h (š after ruki; s in the cluster st). *ḱ > s is not affected by ruki.
    – fdb
    May 23 at 17:59
  • 2
    @BertBarrois that wouldn't proclude Avestan being a child of Sanskrit, it could just have merged the two Sanskrit phonemes as s. To show that Avestan is not a child of Sanskrit you need to show that Avestan lacks a merger that Sanskrit has (e.g. if Sanskrit merges proto-Indo-Iranian phonemes A & B, but Avestan has distinct reflexes for each). This can be distinguished from Avestan having a split if the distinction in Avestan aligns with that of other Indo-European languages (i.e. it does represent the original distribution)
    – Tristan
    May 24 at 9:05
4

It is easy to prove that Vedic cannot be derived from Avestan. Vedic preserves voiced aspirates as such, Avestan (indeed all of Iranian) merges voiced plain and voiced aspirates into voiced plain. Now going the other way, the palatal voiced aspirate *ĵʰ of PIA becomes h in Vedic, but z in Avestan, just like *ĵ becomes z. Vedic has eliminated a property that was preserved in Avestan: exx Avestan dažaiti 'he burns", Vedic dahati. Hence linguists do not derive one language from the other, they derive each language from a distinct path from a common language. Also, earlier voiced palatal plus dental stop e.g. PIE *mr̥ǵt 'forgive' preserves the expected fricative in Avestan mərəžd, but it is lost in Vedic mr̥ḷ-/mr̥ḍ (z as lost in Proto Indo-Aryan).

9
  • 2
    What is *ĵ(ʰ) meant to be exactly? PIE has *g̑ (also written *ĝ or *ɡ́; PIA has *j́ and *ǰ (can’t type that properly on my phone, but háček should replace dot, not be above it). Neither has *ĵ in any transcription system I’m aware of. May 22 at 18:05
  • 1
    Whose reconstruction of PIA has this ĵ ? Typically it’s ǰ (with a háček), see e.g. Lubotsky 2018 degruyter.com/document/doi/10.1515/9783110542431-031/html
    – Alex B.
    May 22 at 23:31
  • 2
    @AlexB. Whether you write j or ǰ is purely a matter of convention. The standard Avestan grammar by Hoffmann/Forssman uses the former.
    – fdb
    May 23 at 17:45
  • 2
    @fdb well, of course, it is a matter of convention. That being said, I have seen both j or ǰ with a háček, but I have never come across ĵ with a circumflex, so I was curious who exactly uses/used this rather unconventional ĵ with a circumflex. But I'm not going to argue with you since Indo-Iranian is your forte, not mine (and it's good to see you back btw!)
    – Alex B.
    May 23 at 19:33
  • 1
    isn't the question asking about proving that Avestan isn't derived from Vedic, not vice versa? This answer seems to be backwards
    – Tristan
    May 24 at 9:03
3

Another example of Avestan retaining a more archaic form than Vedic is in the declension of the word for “path”. Avestan retains the distinction of full grade *pantaH- (Av. Nom. Sing. paṇtå) and the double-zero grade *patH- (Av. Gen. Sing. paϑō). Vedic generalised forms with -th- in all cases (nom. panthāḥ, gen. pathaḥ, etc).

6
  • Thanks for your answering. Could you tell me please What are full grade and double-zero grade ?
    – Nikos
    May 24 at 14:53
  • 1
    @NikosPavlopoulos. There is a reasonably correct account here: en.wikipedia.org/wiki/Indo-European_ablaut
    – fdb
    May 24 at 15:05
  • 2
    To put it very simply: πα-τέρ-α has full-grade ablaut in the second syllable; πα-τρ-ός has zero-grade ablaut. “Double-zero grade” means that a vowel has been eliminated in two syllables.
    – fdb
    May 24 at 15:49
  • thank you so much
    – Nikos
    May 25 at 0:14
  • 1
    @NikosPavlopoulos. I have it from Hoffmann/Forssman, Avestische Laut- und Flexionslehre, p. 125.
    – fdb
    Jun 17 at 17:13

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.