The origin and spread of Indo-European languages should be relevant to the origin of Uralic languages too, so in order to asses how the proposal of the former in the main article affects the latter, we’ll look at it in this appendix.
The origin of Uralic languages has been traditionally thought to be either around the Ural Mountains or Siberia. With the advent of population genetics, those traditional theories started to be linked to a certain amount of Siberian ancestry present at varying degrees in most Uralic speakers and/or to a paternal lineage (formerly known as N1c, now as N3a) that is present also at variable frequencies in most Uralic speakers. When ancient DNA started to become available, studies related to the origin of Uralic languages tried to follow those previous steps, trying to find the link between that Siberian ancestry and/or that paternal lineage to Uralic speakers in the ancient DNA record. But just like with the studies dealing with Indo-European languages, that approach has proven to not be very fruitful. Ancient DNA has changed our views on populations prehistory in such radical ways that trying to fit the new findings into those old theories has been very detrimental for an objective assessment of the actual finds. I wrote a post back when the first study related to Uralic was published where I already highlighted some of the discrepancies between what the study said and what the data presented actually showed. The study was about Estonia’s genetic history, and what it showed by just looking at Estonia alone was that the most likely origin of the Uralic language they spoke was in the population that arrived with the Corded Ware Culture from the North Pontic Steppe, and not any other from the Urals or Siberia (in fact, the second best source by that study’s data would be the local hunter-gatherers). In any case, looking at Estonia in isolation is not very helpful to find the origin of Uralic languages, so those remarks are just to highlight the discrepancy between the data presented and conclusions mentioned above.
More recent studies (link, link) have insisted in trying to link the Siberian ancestry and/or the N1c Y Chromosome to the spread of Uralic languages, but in doing so they have done the opposite by bringing us the actual samples from where those genetic traits come from and proving that they can’t be the origin of Uralic languages. This is because, as it was already indirectly known, they have been able to track the origin of that population to Yakutia (North East Siberia) at around 4500 BC, which makes it impossible on linguistic grounds to be the Proto-Uralic homeland. Additionally, they also showed that the spread of this ancestry and lineage was unrelated to the Seima-Turbino phenomenon (another popular belief), which in itself is unrelated to the spread of Uralic too.
The first constraint that one must face when looking for the Proto-Uralic homeland should be the fact that Uralic is clearly influenced by (or somehow related to) Indo-European. The similarity between both language families is something widely accepted and hardly controversial. The second constraint would be the necessity of that Indo-European language that first influenced Proto-Uralic to be pre-Indo-Iranian. We do know that Indo-Iranian was the language spoken throughout the steppe for over 1500 years, and therefor it’s perfectly normal that Uralic languages must have influence from it given their geographical distribution. But most Uralic linguists point out that the older layer of IE similarities predates the Indo-Iranian layer. If this is correct, then it limits greatly the options for a putative Proto-Uralic homeland, since Indo-Iranian was the only IE language spoken in the steppe that could have influenced Uralic.
With the mainstream current studies about the origin and spread of IE languages now being at a phase where they’re considering that the CWH people would have spoken a language that was ancestral to both Indo-Iranian and Balto-Slavic (what they refer to as Indo-Slavic, which would already contain the common features to them) it would be pretty much impossible to find an appropriate Proto-Uralic homeland (but then again it would be impossible to find how Celtic, Italic or Germanic languages came to be too). The alternative view that I’ve explained in the main article, however, allows for a Proto-Uralic homeland to be located in contact with a pre-Indo-Iranian IE language in a geographical area that is plausible for the distribution of Uralic languages.
The Proto-Uralic homeland, according to this latter view, would then have to be in Eastern Europe, in contact with the IE speaking people who would later become Balto-Slavic under Scythian (Indo-Iranian) influence. That would place it somewhere around Eastern Poland or Belarus in the period around 1700-1200 BC (possibly some culture that was neighbouring the Trzciniec Culture to the north). The language would be one descended from the CWH and therefor what I’ve referred to in the main article as a North Eurasian Bronze Age (NEBA) language, but strongly influenced by an early form of the IE languages that reached Europe around 2400-2000 BC.
If this is correct, then being able to have Uralic languages as part of the NEBA language family would be very useful for studying both ancient Uralic as well as the rest of the NEBA language family.
When it comes to predictions for future research (in the way of ancient DNA) that could help verify or falsify this hypothesis, I think the most obvious one would be to find out if people from that putative Proto-Uralic area (who would have the typical paternal lineage of the CWC, i.e, R1a-Z283) moved to the east reaching the Urals (where local populations had the sister clade of it, R1a-Z93, which would allow us to distinguish them) during the subsequent period (1000-500 BC or slightly later).
5 thoughts on “Appendix I: Uralic languages”
I have come across something I was absolutely not aware of, some apparently convincing evidence that an old form of Hungarian was already spoken in the Carpathian Basin since at least the 3rd century AD.
A Tale of Two Sphinxes: Proof that the Potaissa Sphinx is Authentic and Other Aegean Influences on Early Hungarian Inscriptions
The Potaissa Sphinx had been suspect of being a 19th century forgery, but this seems quite unlikely given the evidence provided.
If this is true, it may change the possible origin of Uralic languages. There are two scenarios that I can think of right now:
– Sarmatians brought the language to the Carpathian Basin: Sarmatians were closely related to Scythians, but it’s probably that their origin was not as far east in Central Asia. As Scythians moved west around 1200 BC they probably incorporated populations from the former Srubnaya Culture. Some may have switched to Indo-Iranian, and some may have retained their original non-Indo-European language but being influenced by Indo-Iranian. If this language was Proto-Uralic, then it would imply that the earliest IE influence did not predate Indo-Iranian as it usually maintained. The spread of the language itself would seem plausible given the areas where the Sarmatians moved around, but it seems a bit strange that if they were part of the larger Scythian confederations all the populations from north of the steppe switched to Uralic and not to Indo-Iranian.
– A more likely option would be that the Yamnaya people would have already switched to an IE language around 3500-3200 BC (I mention this alternative scenario in the main article) and therefor IE languages would have spread through the Balkans since 3100 BC or so, some in close proximity with the CWC. Proto-Uralic could then have emerged close to an older form of IE, as linguists usually agree, in any of the contact zone areas between the north Balkans and central-eastern Europe. This wouldn’t be very different to the scenario proposed in this appendix, but it would allow for an earlier date and more possibilities (for example, the Únětice culture being Uralic and help to spread the languages to NE Europe as well as the Carpathian Basin. This still requires the west to east spread of Uralic languages but it adds the constraint to needing to prove that Hurrians didn’t come from the steppe (something not clear at this point), since this would imply that Yamnaya was non-IE and the scenario wouldn’t work.
As for the origin in NE Siberia, this makes it even more difficult. Yes, Scythians originated in the NE part of Central Asia and had some Siberian influences, but their main influence was from Turan (SC Asia) and they adopted their language (Indo-Iranian). It’s not very plausible that they adopted both Indo-Iranian and some form of Uralic and brought both together to the west. Besides, the Sarmatians that were in or near the Carpathian Basin were quite western, mostly a mix of Sintashta and local North Balkans populations. Not much in the way of East Asian/Siberian admixture. And we can’t even be sure that the Hungarian found there has anything to do with Sarmatians themselves. It could be from local populations.
So based on this model when and how did Uralic reach northern Asia?
“It’s not very plausible that they adopted both Indo-Iranian and some form of Uralic and brought both together to the west.”
Not saying that it happened but I think it’s possible that one minor clan that is a part of the larger confederation could have ended up bilingual with a “native” language. Finding the exact linguistic affiliation of some of these nomadic groups can be pretty difficult (see the Xiongnu or the “Iranian Huns”).
Speaking of Scythians, can you model UKR036, UKR013, UKR014?
@Vara
It’s hard to say too much about Uralic at this point. But if it came from Eastern Europe, which is what I propose as more likely, we should be able to see some sort of migration from that place towards the Urals somewhere in the 1st mill. BC. Such migration could be detected by the R1a-Z283 marker (modern Uralic speakers have it, but not R1a-Z93, so we should find out why). Once you have Uralic through the Forest steppe to the Urals it shouldn’t be difficult for it to spread to more northern populations that were largely hunters or incipient reindeer herders.
Yes, the confederations from the steppe were often multilingual, but my point about how the possible evidence of some kind of Proto-Hungarian in the Carpathian Basin in the early centuries of our era is that it would be very unlikely for it to have been picked from NE Siberia (the proposed homeland or PU). First, because the easternmost Scythians were precisely the ones who adopted Indo-Iranian, not Uralic, and they were the ones in the position to have picked PU instead if it did come from NE Siberia. And then, the putative speakers of that Proto-Hungarian would be either a local population or the Sarmatians who moved around the area at that time, but even if it was the Sarmatians, those were mostly of European genetic ancestry (meaning like the Sintashta/Srubnaya people rather than the more admixed Scythians from eastern origin). This is in start contrast with what we see some centuries later with Turkic, a language that did come from Siberia (South Siberia, not NE Siberia), where it’s very clear from the genetics of its carriers.
Here’s a model of those Scythians you requested. Since they’re from Ukraine, I added a South Caucasus and a Balkans source too:
https://adnaera.com/wp-content/uploads/2025/01/vah_ukr0.png
In this case the original language of Andronovo before the adoption of Indo-Iranian 1800-1500 BCE is pre-Uralic (Uralic related) but still not Uralic?
For the Scythians, is there an iron age west Asian source that isn’t Armenia_IA that would improve the fit?
Yes, if this finding about Proto-Hungarian in the Carpathian Basin c. 3rd century BC or earlier is confirmed, that would be a possibility. Early Andronovo would have been a sort of pre-Uralic, same as early Srubnaya. But if Uralic has an older layer of IE influence predating Indo-Iranian, then Srubnaya around 1500-1200 BC having contacts with an IE language from around Romania would be the candidate for Proto-Uralic. If instead Indo-Iranian is enough to explain the IE influence in Proto-Uralic, then it could be that Andronovo under the influence of Indo-Iranian speakers became a mix of those who switched to it and those who kept their original language but with I-Ir influence (which would be Proto-Uralic).
All very conjectural, of course. I’m still more inclined to think that Proto-Uralic formed in Eastern Europe rather than the steppe itself, but who knows.
As for the Scythians, not much change with IA West Asian sources. There’s quite a bit of overlap at that point, but here’s a run using Vahaduo:
https://adnaera.com/wp-content/uploads/2025/01/vah_ukr0_2.png
For reference, with the previous sources:
https://adnaera.com/wp-content/uploads/2025/01/vah_ukr0_2b.png