[rd108]

I don't think it's fair to dismiss the OIT hypothesis as a Hindu nationalist thing. I wrote some more about it in a comment on this article, but it actually just seems like the most likely explanation given the evidence.

Arguing the archeology, genetics, and (grudgingly, it's barely a science) linguistics is totally fair but I'd like to see more open-minded debate on the data.

The Mitanni (1500 BCE) and the Tarim Basin mummies partially Indian DNA are important points to address, but the Sarasvati river dating has to force a dramatic re-think of how Indo-European languages spread.

[cmrdporcupine]

I am reluctant to engage with this, but... Sigh...

I have to ask... How do you explain the Anatolian languages?

Or what seems like a very clear progression: Sintashta -> Andronovo -> BMAC -> Indo-Aryan.

A long history in the subcontinent that predates BMAC does not imply linguistic continuity.

There's such an overwhelming set of evidence linguistic and materially... there's a reason why, yes, it's just Hindu nationalists/fundamentalists parroting this position.

[rd108]

Splitting the argument into two pieces is most concise:

1) Sarasvati river paleochannel radiocarbon dating of Rig Vedic Sanskrit (<=3000 BCE)

I talked about this here: https://news.ycombinator.com/item?id=42330343 This is by far the strongest piece of evidence that the Kurgan model for IE language diffusion, which has Indo-European languages entering India circa ~1500 BCE, is grossly incorrect.

We now know the Sarasvati, the most prominent river mentioned in archaic Rig Vedic Sanskrit texts, began to desiccate and disappear in 2600 BCE. That desiccation process is actually mentioned in later Sanskrit works like the Mahabharata, in Sanskrit that has significantly linguistically evolved.

There are hundreds of urban sites along the now-dried Sarasvati riverbed which have now been discovered, like Bhirrana, with cultural continuity back to 7500 BCE. But let's steelman the Kurgan model and assume there was a replacement of this advanced urban society's language by IE-speaking Steppe nomads, that left no archeological trace whatsoever. This still means that Indo-European languages were in India at least by 3000 BCE, making this the oldest attested Indo-European language in the world.

I would challenge you to explain how the Sarasvati paleochannel evidence doesn't completely break the Kurgan model. Even leaving aside the Sanskrit corpus which does not remember any migration or original homeland of the Aryans before India.

2) Anatolian languages (1800 BCE) The earliest hard chronology of Hittite is the Anitta text of the Kussara (1800 BCE). The linguistic analysis (such as it is, linguistics isn't the hardest science, much less so than radiocarbon dating) implies this IE branch diverged early. Like the dating of Rig Vedic Sanskrit to at least 3000 BCE, it's likely the Hittite language was evolving for hundreds of years before this.

Did IVC settlers directly migrate to Anatolia? Was there a cultural domino effect spilling out from the IVC to the BMAC, then either north through the Caucasus or south through Iran / Mesopotamia? We don't know.

We do know there were widespread cultural and economic ties between India and the Middle East at least as far back as 3300 BCE. IVC seals and finely-wrought carnelian beads were discovered in the royal Sumerian tombs in Ur, and many texts described the thriving economic trade between these two regions. Indian DNA has been found in Syria in 2500 BCE. The Mittani in ~1500 BCE are an even clearer example, with Rig Vedic deities, a military elite, and a horse-training text that are distinctly Indo-Aryan, not even Indo-Iranian.

In other words, cultural and human diffusion from India to the Middle East was already happening hundreds of years before the first attested Anatolian IE language is found in the Hittite.

And we know it was happening in the other direction (east) as well. The Tarim Basim mummies had Indian genetic markers, implying the Tocharian branch of IE may have also arisen out of India.

The IVC culture had a bigger population than Egypt and Mesopotamia combined. It was an advanced and urbanized culture. There were major environmental shocks that caused mass migrations. There has been migration out of India from at least 3300 BCE to the common era with the Gypies / Romani.

Is it so hard to believe IE languages could have come from such a place?

[cmrdporcupine]

> Did IVC settlers directly migrate to Anatolia? Was there a cultural domino effect spilling out from the IVC to the BMAC, then either north through the Caucasus or south through Iran / Mesopotamia? We don't know.

Except the directional flow of what you're describing here is exactly opposite of what the archeological evidence shows.. with clear material cultural continuity in an eastern/south-eastern direction from the Black Sea to south Asia.

It's likely that when the Indo-Iranian languages arrived in south Asia they were already in the process of heavy fusing with local cultures, esp in the BMAC. References in Sanskrit writings about events that may precede Sanskrit speaking doesn't prove anything. There are plenty of cases of elite language replacement, and it's likely an elite class spoke IE and potentially even translated and wrote down non-IE oral histories.

Cultures are not necessarily languages or "peoples." Language replacement is a common phenomenon. India can be India with an insanely long deep amazing history that the world recognizes as rich and powerful and ancient... without being the origin of one of the languages spoken there.

In any case, non-IE languages continue to prosper in the subcontinent, and probably did even more so among regular people back then. Which is very much not the case for the central European region, unless you count the crazy linguistic patchwork in the Caucasus (which is heavily divided up by mountains).

Anatolian divergence happened very early, so much so that it has an entirely different gender system than all other IE languages. And lacked common IE words related to horse riding, chariots, metal-working, etc. By the time of the written documents you describe, it had already diverged into several distinct languages, implying a long history in the region. Likely a founding population that made its way around the Black Sea on some path.

And a similar phenomenon to India: established itself as a local elite over a population that spoke a different language, and kept oral histories and eventually wrote in their own language about the gods and stories of the people they fused with.

Anyways, the strongest evidence is in the common IE vocabulary -- esp around plants and animals -- which is strongly biased towards the temperate climate around the steppe.

[rd108]

You're not explaining how the Sarasvati river evidence, which is radiocarbon dated, does not fully contradict your position.

Do you really believe, in good faith, that the most parsimonious explanation is that a small group of IE-speaking nomads entered India 500+ years after the Sarasvati was drying and population centers had been abandoned, then re-purposed the entire history of the "native" population, wrote themselves completely out of this re-purposing with no memory of where the nomads themselves came from, and then at least hundreds of years after that re-wrote the story of the desiccation of the river (in significantly evolved, non-Rig Vedic Sanskrit) to read as if it was occurring then, ~1000+ years after it actually occurred?

Occam's razor here is clear.

The Rig Vedic speakers were in India when the Sarasvati was a fully flowing river pre-3000 BCE, they documented the drying of the river in later Sanskrit texts, and the Rig Vedic IE corpus was created by at least 3000 BCE based on the radiocarbon dating of the Sarasvati's paleochannel.