This question is an extension of this one: How did Chinese recursion evolve? . In the comments, Mark Beadles helpfully pointed out that center-embedding is absent from Sanskrit in the early Vedic period, but you see some embedding in the later Vedic period.

The interesting thing about this is that the later Vedic period is past Homer's time by an amount too short for reasonable spreading of the technique. Further, Alexander the Great would not have brought Greek recursion to India until a few centuries later. So this might be a separate independent invention of recursion.

  1. What is the date of the earliest center embedding in Vedic Sanskrit?
  2. Is there any evidence of Greek/Indian contact which pre-dated this production?

This would be another nice data point in the fascinating history of the spread of recursive speaking and writing throughout the ancient world.

  • I wouldn't know about Sanskrit, but surely recursion was invented independently more than once because it is there in the Turkic family where Greek influence has to be excluded. I think a construction very similar to Turkic is also present in Eskimo-Aleut, but I might be very wrong on this one.
    – kamil-s
    Mar 10, 2012 at 12:23
  • If you're interested in history of recursion world-wide, I just remembered Z. Krążyńska deals with the topic in a couple of her papers, e.g. in kwartjez.amu.edu.pl/Krazynska_2010_3-4.pdf (from Kwartalnik Językoznawczy 2010/3-4).
    – kamil-s
    Mar 10, 2012 at 12:30
  • There is no way Greek or Portuguese could have so deeply influenced Turkic in the 8th C. in Northern Mongolia. The paper deals with the Middle Ages which is when I believe recursion started in Polish, and I can't remember ever seeing a sign that it could have been under a foreign influence. I might be wrong with Polish, though, because I don't specialize in it.
    – kamil-s
    Mar 10, 2012 at 19:43
  • 1
    It is also possible that the Indians and the Greeks invented recursive grammar independently, based on some recursive structure common to some lost indo-European language. I don't know. As far as Polish, you can check for double embedding, it isn't so hard to identify, but a foreign influence in this case is just a population of bilingual speakers of a recursive language, it doesn't have to be an invasion.
    – Ron Maimon
    Mar 10, 2012 at 20:10
  • 1
    I just can't imagine center embedding occurring at all in Sanskrit, because its relative clauses aren't just relative clauses--they also have a corresponding correlative clause, which limits center embedding. So in a sentence, effectively only one relative clause is possible. I requested for a source/example from Mark on the other question. I hate to make this comment without getting confirmation but my opinion (perhaps formed too early) is that it simply is impossible. I'll delete this comment when I'm able to find attestation at all of center embedding in late Vedic or Classical Sanskrit.
    – user67444
    Aug 10, 2015 at 22:08


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.