Oh I agree! If we assume the proto-zagrosian hypothesis is true and then assume that the munda language family existed in India along with the later dravidian l

Oh I agree! If we assume the proto-zagrosian hypothesis is true and then assume that the munda language family existed in India along with the later dravidian languages or maybe even prior to that - there is no reason to believe that these account for all languages spoken in India around 8000BCE - 4000BCE. Afaik, it’s still very possible that entire language families that were even older have now gone extinct. My point was to show how old 4,500BCE was. As you point out, that alone doesn’t rule out a city at Varanasi around 4500BCE.