What If Andronovo Was Never Indo-Iranian?

Added: 2026-05-24

[00:00:00]What if the famous Andronovo and Sintashta cultures were not Indo-Iranian at all?

[00:00:05]What if instead they were the ancestors of the Tocharian speakers? Let me briefly introduce you to the Tocharian languages.

[00:00:13]Tocharian languages were ancient Indo-European languages once spoken in the Tarim Basin in what is now Western China.

[00:00:21]There were actually two main Tocharian languages. Tocharian A is the older one and seems to have been used mainly for religious texts. Tocharian B is younger and it was the language that people used in their daily life.

[00:00:33]So, what if the Tocharian languages were actually brought to the region by groups that are generally labeled Indo-Iranian?

[00:00:41]I want to be clear from the start. This is not a claim. It's a hypothesis, a discussion.

[00:00:47]For years, the Tarim Basin mummies with their European-like features were thought to be the earliest Indo-European speakers in the region.

[00:00:54]But, genetics tells a different story.

[00:00:56]These mummies carry mostly Siberian ancestry linked to the pre-step Botai culture, which makes them unlikely candidates for Indo-European languages.

[00:01:07]That led scholars to point to the Afanasievo culture as the likely source of the Tocharians.

[00:01:14]Afanasievo, flourishing around 3300 BCE, was a Bronze Age society famous for horses and kurgan burials. But, there is a problem. The Tocharian languages recorded between 400 and 800 CE can only be reconstructed back to 1000 BCE at most.

[00:01:32]That leaves a huge time gap that Afanasievo alone cannot explain. While most scholars like Mallory, Mair, and Koryan have assumed from the start that Andronovo was strictly Indo-Iranian, this assumption may be blocking a different possibility. If Andronovo was actually connected to the Tocharians, it would explain some patterns we see in Xinjiang that don't make sense under the Indo-Iranian model. For example, starting around 1500 BCE, Xinjiang experiences a huge genetic impact from the step. The Andronovo component is even larger than what we see in India or among Iranian Plateau populations. Out of 125 male samples published in this paper, 39 of them, which makes about 31% belong to R1a subclades directly matching the Sintashta and Andronovo cultures R1a subclades.

[00:02:20]Many scholars assume that the Andronovo culture spoke Indo-Iranian languages.

[00:02:24]Based on that assumption, they that when step people reached Xinjiang region, they brought early Saka or Iranic languages with them.

[00:02:31]But there's a problem here. Xinjiang later becomes the home of the Tocharian languages, which are Indo-European, yes, but not Indo-Iranian. For that to make sense, we have to believe that a major step influence entered Xinjiang, yet somehow failed to replace the local language, allowing Tocharian to survive there for nearly 2,000 years. Now, compare this with the Indian subcontinent. In South Asia, the same model claims that a step migration around 1500 BCE was strong enough to replace languages across northern India.

[00:03:00]But genetic evidence tells a very different story. In regions like Swat Valley, between 1200 and 800 BCE, step ancestry makes up only about 20% and much of that ancestry came through women. It was a female-mediated migration, which is not how languages usually spread, at least not for the patriarchal Indo-European languages. On top of that, northern India at that time was one of the most densely populated regions on Earth. So, we are asked to believe two contradictory things at once, that a large step presence in sparsely populated Xinjiang somehow failed to replace languages, while a much smaller step presence in massively populated northern India succeeded in doing exactly that.

[00:03:40]That inconsistency is the real issue.

[00:03:42]The Andronovo expansion into Xinjiang looks exactly like what Aryan migration theorists wished they had for Indian prehistory, a clear step movement visible archaeologically and genetically detectable. Now, when we look at contacts between Iranic and Tocharian speakers, something interesting shows up in the data. In Xinjiang, Andronovo ancestry is dominant, while the brown component here, associated with the Bactria-Margiana Archaeological Complex, appears only in much smaller amounts.

[00:04:09]This matters because the BMAC-related ancestry shows up consistently in Scythian groups. Wherever the Scythians go, this component goes with them. That tells us something important. The spread of this ancestry across Central Asia lines up with Scythian movements, not with the earlier Andronovo horizon. So, when we look at where East Iranic languages later appear, they follow the same paths as Scythians and their BMAC-associated ancestry. That makes the Scythian expansions a much better explanation for the spread of East Iranic languages than Andronovo. And this fits well with the broader evidence, which strongly links the BMAC cultural world itself with Iranic-speaking populations. So, once we separate these two movements, a new picture emerges. Andronovo no longer looks like the source of Iranic languages in Xinjiang.

[00:04:53]And this leads us to one of the strongest clues connecting the Andronovo horizon to Tocharian, well, at least for some of its groups.

[00:05:01]Old Chinese preserves several Tocharian loanwords connected to chariots, wheels, and chariot equipment.

[00:05:07]Linguist Alexander Lubotsky documented many of these in 1998.

[00:05:11]Now, pause and think about what that means.

[00:05:13]If the Andronovans were Indo-Iranian and they were the first major users of horses and spoked-wheel vehicles in the region, why would Old Chinese borrow Tocharian terms for this technology instead of Indo-Iranian ones?

[00:05:26]The earlier Afanasievo population, dated to around 3000 BCE, cannot explain this.

[00:05:31]The spoked wheel had not been invented yet, so they wouldn't have had this vocabulary.

[00:05:36]That leaves only one realistic source, the Andronovo horizon itself.

[00:05:40]And that strongly suggests that at least some Andronovo groups were Tocharian-speaking.

[00:05:44]There are also other lines of evidence that support a connection between Tocharian and the Andronovo horizon.

[00:05:50]First, there is the Uralic substrate.

[00:05:52]Tocharian shows clear traces of contact with Uralic languages, while Indo-Iranian does not.

[00:05:58]That makes sense geographically, because before the steppe populations arrived, much of Southern Siberia and Kazakhstan was likely inhabited by Uralic-speaking groups.

[00:06:07]Second, when we look at Tocharian structure, it often aligns more closely with European Indo-European branches like Balto-Slavic, Germanic, than with Indo-Iranian.

[00:06:16]This is a key point. Andronovo would essentially be the eastern extension of the Corded Ware horizon, which is directly ancestral to many European language families. So, a Tocharian-Andronovo connection would make sense from both the linguistic and archaeological perspective.

[00:06:31]Overall, there is no reason to dismiss Andronovo as a potential source of the Tocharian languages in the Tarim Basin.

[00:06:38]The main obstacle is the widespread assumption that Andronovo equals Indo-Iranian, which blocks the discussion before it even starts.

[00:06:46]Chariot-related terminology entered Old Chinese through Tocharian at around 1200 BCE, which lines up perfectly with the Andronovo horizon, not Afanasievo.

[00:06:56]Yet, some still try to trace the Tocharians back to Afanasievo. That doesn't make sense, because Afanasievo people only used cattle-drawn wagons and carts, not the horse-drawn spoked-wheel wagons.

[00:07:08]So, the evidence keeps pointing back to Andronovo as the most plausible source for the Tocharians.

[00:07:13]Take a look at this tree.

[00:07:15]It shows all the ancient R1a samples from Russia, Kazakhstan, and China mapped onto the YFull topology. What stands out is the impact of Andronovo, which is a culture from Russia and Kazakhstan, on China.

[00:07:27]You can see it clearly. Andronovo lineages reach into Chinese samples, and the effect is quite significant. The colors at the ends of each branch indicate the date of each individual, so you can track how these lineages spread over time.

[00:07:41]Haplogroup U2e1h, found in Andronovo also shows up in this sample from the Tocharian region. And the autosomal DNA in this sample is clearly Andronovo related. That's another strong clue.

[00:07:53]Now, a few points that may seem really strange if we assume Andronovo equals Indo-Iranian.

[00:07:59]First, Andronovo had a huge impact on Xinjiang starting around 1500 BCE, much bigger than its impact on India or Iran.

[00:08:07]Yet, in Tocharian, we find only a few loanwords from Old Iranian. Most Iranian loanwords appear much later from Middle Iranian after Proto-Tocharian had already split.

[00:08:18]Meanwhile, in India and Iran, a much smaller steppe genetic contribution is often claimed to have completely replaced local languages. Also, there are no loanwords from the undifferentiated Proto-Iranic stage, which you would expect if Andronovo were Proto-Iranic, let alone Proto-Indo-Iranian.

[00:08:35]All this makes sense if we consider Andronovo as the source of Tocharian instead.

[00:08:41]As Andronovo groups moved through the steppes into Xinjiang, they could incorporate an Uralic substratum and later borrow Iranian words. The Iranian loanwords we see likely arrived later with Iron Age Scythian or Saka migrations, while the Sanskritic words probably came even later through Buddhist influence.

[00:08:59]When we look at the Andronovo horizon and its impact on Xinjiang, the picture that emerges is clear and consistent.

[00:09:06]Genetics, archaeology, and linguistic evidence all seem to point toward Andronovo being closely related or connected to the Tocharians rather than Indo-Iranians. The survival of Tocharian, the specific Uralic and Iranian interactions, and the timing of technological and linguistic transfers all make sense only if we separate Andronovo from the Indo-Iranian framework.

[00:09:28]This discussion has focused on exploring what Andronovo could represent if it isn't Indo-Iranian.

[00:09:34]That being said, I would love to hear your thoughts on this perspective and whether you find this interpretation of Andronovo as a source for Tocharian languages convincing enough. That was it for this video. Thanks for watching, and I will see you in the next one.