Where did proto-IE language start?

Source of proto-Indo-European language

  • R1a

    Votes: 23 31.9%
  • R1b

    Votes: 22 30.6%
  • Cucuteni-Tripolye

    Votes: 10 13.9%
  • Caucasus-Mykop

    Votes: 17 23.6%

  • Total voters
    72
My hypothesis :

I think R1b came from somewhere in Turkmenistan east of the Caspian. They were nomadic HGs. They spoke a language more "primitive" than PIE (as reconstructed), a language that was heavy with laryngeals, and without some of the tenses that appeared later (aorist, etc.). Also a language that retained traces of Uralic elements from their prolonged ancient wanderings across western Siberia.

They moved round the southern tip of the Caspian into NW Iran, and stayed a while around Lake Urmia or Lake Van. There, they subdued and/or mixed with J people. They got their Gedrosia/Caucasus admixture. They learnt cattle-breeding from the J people, and possibly saw the first wheels, made further south in the fertile crescent. R-V88 went its way down the Levant and into Africa.

Others stayed in Anatolia, perhaps imposing their language on J people. They moved west along the Caucasus. Some of them continued further west into Anatolia - the Hittites et alia. Their old version of PIE retained the laryngeals and unevolved verb forms.

Eventually, those who were south of the Caucasus crossed over into the Steppe, possibly along two routes, one along the Caspian, another one east of the Black sea. The Caspian route would have brought assimilated J2b2 into the steppe (or pushed them ahead and assimilated them later on). The Black Sea route would have brought G2a-L140 into the steppe, along with some farming vocabulary, and ended up around Maykop. In the steppe, the language would have evolved into "reconstructed PIE".

If some neighboring J tribes were indeed indo-europeanized in NW Iran, and later moved west across Anatolia and into the Balkans, it might explain why the Myceneans spoke their ancient Greek, but didn't much alter the autosomal makeup of the Minoans. They just added some more J and Anatolian Farmer where there was some already.

What really baffles me is how R1a fits into the picture. They didn't mix much with R1b. There were contact zones, of course, but otherwise they were rather far away in their forested steppes. They do not appear to have been conquered (unless it was by R1b women!). When, where, and how did they start speaking PIE before they "turned satem" ? Did they come into the steppe from the north of the Caspian? But then, if they did, their language should have diverged much more than it did. R1a's PIE, to my eyes, is the real mystery.
 
The western R1b is of anatolia, almost without any doubt, and arrived at the western extreme of Europe by the south of Europe, came with the Neolithic, R1b is the predominant haplogroup in Western Europe.


I think that certain Indo-European languages ​​like Ligur, Lusitano, reached the western end with this emigration from Anatolia (western end I mean Portugal, Spain, France, Italy), autosomally non-steppe farmers, very related to IE italic, Greek and Anatolian languages , which were mixed with WHG in southern Europe.


This makes me wonder if Ibero - Basque was a WHG language?


Or if Ibero - Basque was a Neolithic language with western R1b and Ligurian, Lusitanian was the ebb and flow of the subsequent mixture that occurred in central Europe?


It is clear that the languages ​​of southern Europe, French, Italian, Spanish, Portuguese, Greek, have a facility to pass them to writing that do not have those of the north, I do not know if this is so or is my impression and also are centum languages along with those of anatolia.


This emigration of Neolithic anatolia to the western end of Europe originated the bell-shaped culture, an expansive culture that was mixed in central Europe with, among other things, what has been called steppe.
 
They learnt cattle-breeding from the J people, and possibly saw the first wheels, made further south in the fertile crescent. R-V88 went its way down the Levant and into Africa.

Wasn't a very ancient R-V88 sample found somewhere in or around the Balkans? Considering the very early clades of R1b in the Balkans, how could we explain the same clades also being found in people that came from Turkmenistan or even, before that, further eastward in Siberia, many thousands of km away? I really don't know... I think that by far the most probable scenario is that R1b-M269 (or its immediate ancestor) and R1b-V88 parted their ways much earlier, even before the adoption of primitive Neolithic package, and in my opinion they had totally distinct and parallel histories of migration and expansion.
 
What really baffles me is how R1a fits into the picture. They didn't mix much with R1b. There were contact zones, of course, but otherwise they were rather far away in their forested steppes. They do not appear to have been conquered (unless it was by R1b women!). When, where, and how did they start speaking PIE before they "turned satem" ? Did they come into the steppe from the north of the Caspian? But then, if they did, their language should have diverged much more than it did. R1a's PIE, to my eyes, is the real mystery.

R1a-heavy areas of the steppe and even later forest zones like CWC had too much CHG in them for this hypothesis that they did not mix much with other more "southern" people to be true. We don't know how that happened, but maybe they mixed especially through exogamy but also through male lines (R1b and so on), but eventually when they really expanded much to the north and other directions they underwent a severe selection of Y-DNA haplogroups due to founder events and the particularly high success of just a few groups of males, who happened to be mostly R1a. Anyway, it is certainly imposible that they had little genetic interaction with the people from Caucasus or mixed with Caucasians. The northern Yamna and CWC would've been basically Neolithic EHG if they had remained totally shielded from those migrations, and they weren't.
 
doubt

R1a-heavy areas of the steppe and even later forest zones like CWC had too much CHG in them for this hypothesis that they did not mix much with other more "southern" people to be true. We don't know how that happened, but maybe they mixed especially through exogamy but also through male lines (R1b and so on), but eventually when they really expanded much to the north and other directions they underwent a severe selection of Y-DNA haplogroups due to founder events and the particularly high success of just a few groups of males, who happened to be mostly R1a. Anyway, it is certainly imposible that they had little genetic interaction with the people from Caucasus or mixed with Caucasians. The northern Yamna and CWC would've been basically Neolithic EHG if they had remained totally shielded from those migrations, and they weren't.

I agree with your objections, rather less so with your suggestions. What's mysterious is : if R1a was indo-europeanized by R1b, why is there so little R1b among them, and vice-versa ? If R1a stayed a while down south with R1b, admixing with CHG, how could the two communities take to the same language but never swap haplos more than they did? Exogamy ? Maybe, but this could work mostly for fringe areas. R1a seem to have occupied vast expanses far to the north. I doubt people west of the Urals would have "mail ordered" brides from an area several hundred kms to the south. And in such numbers that their language was changed. Founder effect and selection ? If a mixed group went north, it's hard to imagine why R1b, after running behind wild beasts across central Asia for millenia, would have been less adapted to the northerly climes.
 
See:
https://www.eupedia.com/forum/threads/35776-Central-and-South-Asian-DNA-Paper?p=537049#post537049
As I said, it was unlikely Reich would go out on a limb like that without some pretty heavy duty evidence.

It was the simplest explanation after all, given what we know from Anatolia Chalcolithic and Bronze Age, an increase in CHG but no EHG ancestry, Anatolian languages like Hittite are a movement west from Caucasus while the rest descend from Yamnaya.

It wasn't just an R1b-M269 in Iran but also J2b, there is no R1 clade that satisfies the pattern for Anatolian languages, maybe J2 subclade then ?
 
See:
https://www.eupedia.com/forum/threads/35776-Central-and-South-Asian-DNA-Paper?p=537049#post537049


As I said, it was unlikely Reich would go out on a limb like that without some pretty heavy duty evidence.

I had thought R1b had something to do with migration from outside the steppes and a decisive role in the early formation of PIE (maybe a mixed language), but not that truly transformative scenario posited by the latest evidences. If I had to bet before I'd think R1b came from either the Balkans or from the Black Sea coast near Transcaucasia, but anyway West Iran is just to the east of that region, so not that far off, and thinking again it does seem much more likely that R1b came with heavily CHG/Iranian_Neo populations that certainly did not exist that early in the Balkans, still in the Neolithic (the evidences point to growing CHG or CHG-like admixture in Southern Europe probably only after the Copper Age). So, I think it makes perfect sense to me, but that is just maybe because I personally have no horse on this race. Whatever comes out of the results is very welcome.
 
It was the simplest explanation after all, given what we know from Anatolia Chalcolithic and Bronze Age, an increase in CHG but no EHG ancestry, Anatolian languages like Hittite are a movement west from Caucasus while the rest descend from Yamnaya.

Perhaps the R1b group migrated, with most of them going north, and some west to form the Anatolian languages? Their original areas could then have been filled in by other y lines and other languages.
 
It was the simplest explanation after all, given what we know from Anatolia Chalcolithic and Bronze Age, an increase in CHG but no EHG ancestry, Anatolian languages like Hittite are a movement west from Caucasus while the rest descend from Yamnaya.

It wasn't just an R1b-M269 in Iran but also J2b, there is no R1 clade that satisfies the pattern for Anatolian languages, maybe J2 subclade then ?

But then IMHO we'll have to probably consider the possibility of Indo-Hittite as was once assumed. That is, though it is actually, I think, amazing how the Anatolian languages were still recognizably IE (though very distinctive especially in grammar), the evidences now point to the high probability that they came from a distinct branch of an Early PIE that had separated from the rest of IE branches some 3,000-3,500 years before the first attestation of Hittite.

That is a hell of a lot of time for phonetic and syntactic evolution. Culturally, economically and genetically they also seem to have been an entirely different people, much like Spaniards who stayed put in Spain as opposed to indigenous people that absorbed Spanish from a sizeable minority of Spaniard people in Peru or Ecuador.

The Late PIE, the Steppe PIE, was possibly always another language separate from Transcaucasian PIE (though closely related, kind of like Portuguese and Romanian). The Early PIE from which Hittite and Luwian came hadn't even such basic things of grammar like masculine and feminine gender, some of the fundamental verb tenses and some of the basic formation of cardinal and ordinal numerals of non-Anatolian PIE.
 
To think that I believed in the Steppe cowboys abducting Caucasus girls theory for some time, because of the abundance of R1b in the EHG. This is now the simplest explanation, and it is time for the author at Eurogenes to realize that.
 
Perhaps the R1b group migrated, with most of them going north, and some west to form the Anatolian languages? Their original areas could then have been filled in by other y lines and other languages.


The Indoeuropean expansion west did not stop in Anatolia, Southeast Bronze Age Europe(Including Italy) is modeled as "Iran Farmer" + "Local Farmer" by Reich, not "Steppe" + "Local farmer", I'm not sure why you struggle to understand that most of the Indoeuropean ancestry in Italy is not Steppe derived.
7AsGxTg.png
 
To think that I believed in the Steppe cowboys abducting Caucasus girls theory for some time, because of the abundance of R1b in the EHG. This is now the simplest explanation, and it is time for the author at Eurogenes to realize that.

Correct, but we have to admit that it is still quite an issue how to explain the high percentage of CHG and the Indo-Europeanization of R1a that, according to the results of samples of earlier Neolithic and Mesolithic, were once mostly EHG. I don't think the "abduction of Caucasian women" holds water, but a very high frequency of exogamy as inter-tribal alliances or something of the sort still looks plausible to me (if we can see a lot of EEF in R1b-M269 males among Basques, why not a relevant % of CHG in R1a males among Northern Yamna and their CWC offshoot?). There is too little R1b among the R1a-majority Yamna and CWC regions, compared to their chunk of CHG, for it to be explained exclusively by migration of entire peoples, displacement and intermixing.
 
The Indoeuropean expansion west did not stop in Anatolia, Southeast Bronze Age Europe(Including Italy) is modeled as "Iran Farmer" + "Local Farmer" by Reich, not "Steppe" + "Local farmer", I'm not sure why you struggle to understand that most of the Indoeuropean ancestry in Italy is not Steppe derived.
7AsGxTg.png
Not only the Italian Indo-European languages, I believe that all the Centum have nothing to do with the steppes, apart from the mixtures that could be produced in the contact areas as the center of Europe.
 
The Indoeuropean expansion west did not stop in Anatolia, Southeast Bronze Age Europe(Including Italy) is modeled as "Iran Farmer" + "Local Farmer" by Reich, not "Steppe" + "Local farmer", I'm not sure why you struggle to understand that most of the Indoeuropean ancestry in Italy is not Steppe derived.
7AsGxTg.png

Well, that is maybe because linguistically the IE branches found in Southern Europe (Italic, Albanian and even more Hellenic) are way too similar and too closely related to the other steppe-related branches of IE for this scenario to be plausible.

It is extremely hard to explain such high degree of similarity with the "northern" steppe-derived branches and instead a very distant relationship to the westward migration from Anatolia, like the speakers of the extremely diverged (and extremely unlike Italic or Hellenic) Anatolian IE, when your proposal seems to be that those branches are in fact related to the Anatolian IE speakers and had split from the people that went to the steppes and formed their own IE branches several milennia earlier.

This scenario does not fit the linguistics at all. We can even assume that some IE branches were once spoken in Italy and were later replaced by incoming IE branches from the north or northeast...

But it's really improbable that Italic, Ligurian or Hellenic would've somehow converged to be much more closely related to "steppe" branches, from which they had diverged in the middle Neolithic (before 4,000 BC), than to their Anatolian "sister" languages. The only clear "outlier" that looks only distantly connected to the others are the Anatolian languages. Italic and others are clearly derived from the same source as Balto-Slavic, Germanic or Indo-Iranian.
 
Not only the Italian Indo-European languages, I believe that all the Centum have nothing to do with the steppes, apart from the mixtures that could be produced in the contact areas as the center of Europe.

Unlikely. Apart from the centum vs. satem phonetic change, there is not much in the way of really fundamental bricks of the language, like morphology and syntax, that differs significantly among all centum languages vis a vis all satem languages as a whole. Centum languages share features with satem languages much more often than they do not. It is extremely improbable that they would end up so closely related to each other if they had diverged so early before the migration of R1b and CHG to the steppes. Also, the centum languages are extremely closer to satem languages than to the centum (in the sense that it had no fricativization of palatalized consonants) Anatolian IE branch. Centum languages' grammar in particular comes from exactly the same source as that of satem languages, and the sharing of some archaisms or instead of some syntactic innovations crosses across the centum vs. satem distinction, which has no strict boundary between them. One very famous case is that of Hellenic (centum) and Indo-Iranian (satem), which share several grammatical innovations even though they belong to different categories. Remember that centum vs. satem is just a useful guide based on distinct trends of phonetic evolution. It is not something much deeper and transformative than that. It does not even mean that those languages have very different lexical and syntactic structures.
 
​It is based on the hypothesis that in the steppes an Indo-European language was spoken, nowadays in the steppes a satem language is spoken of the balto slava family but in the initial age of bronze, what was spoken? Was it from the centum satem?
 
My "delenda Carthago" again : where were the R1a while the R1b were mixing with CHG and crossing the Caucasus ?
 
Last edited:
Hypothesis


1) Iran -> Anatolia (Western R1b) -> Southern and Western Europe.
2) Iran -> Steppes (R1a) -> Eastern Europe.
3) Iran -> (R1a) -> India.
 
It was the simplest explanation after all, given what we know from Anatolia Chalcolithic and Bronze Age, an increase in CHG but no EHG ancestry, Anatolian languages like Hittite are a movement west from Caucasus while the rest descend from Yamnaya.

It wasn't just an R1b-M269 in Iran but also J2b, there is no R1 clade that satisfies the pattern for Anatolian languages, maybe J2 subclade then ?

If that should turn out to be 2103, there might be might there not? Isn't there R1b Z2103 in Anatolia?

See Maciamo's map of what used to be called R1b-ht35, which includes Z2103.

7XjPCjG.png
[/IMG]

Does anyone have a map of just Z2103?

This is one of the reasons we were willing to entertain the idea, along with the fact that if you followed the culture, it was clear that lots of it, and lots of artifacts, including kurgans , if you can call them that, are attested frist south of the Caucasus not north of it.

We were arguing here about the findings relied upon by Ivanov, for example, years ago. One of our Kurdish members was passionate about it. I'm not going to pretend I was positive about it, but I was certainly willing to entertain it, based on Maciamo's work, the archaeology, and the culture etc.

I said then and I'll repeat now that Anthony benefited from the fact he wrote in English and that he's got a great, very approachable writing style, wonderful for hobbyists, whereas Ivanov and his partner's prose is very turgid and opaque in some places. Not that they were right about everything either. Their map was also wrong.

What was the R1b found in Kura Araxes? Anyone have the specific clade at their fingertips?

maps_Y-DNA_haplogroups.shtml
 

This thread has been viewed 181621 times.

Back
Top