How did the Sanskrit gen./abl. singular of pitr-/pitā́ ("father") came to be pitur (and the genitive of the entire noun class as well, of course)? The evolution of all other forms (even pitā́, which seems to be a regular result of pitā́r < pitár-s) seems to be fairly transparent but this one has always eluded me...

AFAIK, the source should be ph₂tr-és so I would expect it to yield something like pitr-ás.


The classic explanation is that *ṛš becomes ur in Old Indian. The ending of the gen. sing. of Indo-Iranian -r stems is sometimes *-as and sometimes *-s. pitur is like hotur (Avestan zaotarš, with full-grade stem).

  • Interesting - isn't *ph₂tr- hysterokinetic declension with accent typically on the ending? So I find it surprising that this would get reduced to *ph₂tŕ-s/pitŕš... – Eleshar Jan 2 '17 at 12:17
  • Is it more surprising than acc. pl. pitṝn? – fdb Jan 2 '17 at 12:36
  • Actually yes. The whole r-stem plural paradigm seems to mirror semi-vowel declensions (i-/u-stems - which does not seem that strange since in sanskrit, [i]&[u] are not difficult to reinterpret as syllabic /j/&/w/ - atypical as it may be), so while it is irregular with regards to the original PIE reconstruct, it sounds like a direct analogy to agnī́n/śátrūn. I see no such analogy for the gen./abl. singular. – Eleshar Jan 2 '17 at 13:36
  • Maybe I am derailing it a little bit - -ṛš > -ur by itself does not seem like implausible at all, so do we have some further evidence that genitive of r-stems in Indian evolve from C-r-ás to C-ŕš? I take your Avestan example as and indication that in closely related Aryan languages this is the case. – Eleshar Jan 2 '17 at 14:27
  • 1
    Please correct it then! – Colin Fine Feb 2 '17 at 0:02

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.