News Article on Wang Paper - PIE is Anatolian again?

Wang et. al:

Reply: We’re afraid that this might be a misunderstanding. There is indeed very limited gene flow between the Caucasus and the steppe groups (apart from the examples highlighted). However, we have based our PIE-related speculations on the observation that the CHG/Iranian (green) ancestry component is increasing already during the Eneolithic north of the Caucasus. This led us to propose that this might be the actual ‘tracer dye’ of an early PIE spread, which could then also accommodate the spread of PIE south of the mountain range where this ancestry component also rises in frequency resulting in a relatively homogenised dual ancestry (Anatolian + Iranian farming-related ancestry) in Chalcolithic times (see also brown arrow in Figure 2).

Problems are, though, that: CHG was already in very high proportion in the steppes as early as 4300/4200 B.C, and it only increased marginally after that, slowly accumulating an increase (that might have been an internal affair of the steppes, anyways, with more CHG-rich people prevailing over less admixed ones); and the Y-DNA makeup of the Chalcolithic and Early Bronze Age steppe does not fit that of any Caucasian population of the same period, therefore we will have to assume the language transfer involved a minor migration with tiny impact in the male lineages. A language transfer without profound sociocultural shift/rupture, without much admixture (especially in the male lineage). Possible, but not likely, particularly in a time before organized empires, writing, sophisticated political and bureaucratic apparatus for far and wide territories, and so on. PIE is a Chalcolithic language, so if it as already being spoken and evolving in the steppes by the early Chalcolithic it is hard to believe it did not start to split there, and not elsewhere, even if the language family it belonged to had come from the Caucasus.
 
As autosomal analysis suggests that:
1. All IE-speaking populations best fit predominantly with the East Balkan Chalcolithic,
2. East Balkan Chalcolithic populations were predominantly of Anatolian origin with minority Suvorovo EHG admixture, and
3. Suvorovo were present in the East Balkans/Carpathians, having admixed with EEF for at least several hundred years before moving out,
- it would not be surprising if the ancestors of IE populations had adopted a language principally derived from Anatolia.

Although of course, we are unlikely to find out for sure which languages these people spoke, or indeed whether they were multi-lingual.

Well, if they were of Anatolian origin in the sense that they were mainly EEF mixed with some EHG, then their language was only distantly Anatolian in any way, they would have made their language evolve in Europe itself for some 2500 years by the time of the early Chalcolithic, more than enough time to become a totally different language (sub)family. Anatolian certainly split much later, in Chalcolithic times, not during the initial spread of farming. It shares too much common cultural/economic vocabulary with the other IE branches to be that early. So, even if your hypothesis is correct, that would make PIE a Balkanic language, not an Anatolian one.
 
If they establish that as a premise, well, then the most honest thing they should say is: "we have come to the conclusion that genetics has no place in this linguistic discussion". And they of course will have to avoid deducing anything touching on the subject of linguistics or even on ethnicity based on genetics, because they will have concluded that genetically untraceable, tiny migrations and cultural contacts might often have triggered language shift, even when there was clearly no acculturation as in the steppes, where despite heavy foreign influences there is a cultural evolution without severe ruptures from the Neolithic to the Bronze Age. Sorry, but this is pseudoscience. If they really think a language transfer happened without admixture, but they obviously do not what languages were spoken by Maykop or other Caucasian cultures back then, or even by the EEF farmers, then they should just quit from this discussion as eeal scientists, not "speculators".

This is pretty much what i think. But before reaching this conclusion, it would be better to understand how they are actually working. Jena/Max Planck, Reich Lab from Harvard and Copenhagen. Both 3 seems to ties up together in most of papers, but i would understand actually how they implement each hypothesis of their collaborators into the general hypothesis. Surely, every scientists would like to be published and his hypothesis put forward. So i'm wondering exactly how it's work, do they select for exemple linguists collaborators wich hypothesis are fitting their genetic conclusions? Are they selecting archeologists wich views corroborate the archeogenetic facts? Because obviously, in the big scheme, Genetic will never be " that " relevent for most scientists if Linguistic and Archeology is not implemented together. Is there in Jena or Reich lab even some collaborators that believe in Steppe Hypothesis? Are Mallory, Anthony consultant for those peoples? Why big names that actually have done work on terrain like Gimbutas, Dumézil, Anthony are relegated to just being mainstream vulgarizators. Why Ivanov and Gamkrelidze are more relevent that Dumézil who could speak fluently more than 40 languages?
 
Guys the link works and there are way more gems than just that point:



"If correct, the evident linguistic conclusion would be that Anatolian languages were already in situ in Anatolia by the 6th millennium BCE.
By contrast, the PIE wine word is absent from Indo-Iranian, a culture which, according to Gamkrelidze and Ivanov (p. 562), lost it because it adopted soma (whatever that was). We find a word related to ‘vine’ in Avestan vaēiti, but it tells us nothing about when proto-Indo-Iranian migrated eastwards and whether it preserved the original word for wine.
Nor were there ever any lions on the Pontic Steppes but the term is present in all IE families including Anatolian (albeit with reduplication -*wolw-o-) other than Armenian and Indo-Iranian, which derive their forms from a root singho-. On this basis, a PIE term for ‘lion’ (ibidem, p. 430) *leu- can be reconstructed (although as stated, this by no means implies that there actually was a PIE word for lion). There is also a cognate in Kartvelian *lom-, but the term is probably an Afroasiatic borrowing (cf. Old Egyptian rw, Akkadian lābu)
.

"In my 2008 article (Sherman Morris, 2009), I highlighted the word for ‘broadband’ in Germanic: we have breiðband in Icelandic, breëband in Afrikaans, breedband in Dutch, breitband in German and bredbånd in Danish, etc., all
showing regular sound shifts from Proto-Germanic, which is reconstructible as proto-Germanic *braiđazbanđan, albeit without this implying that proto-Germanic tribes had the Internet.

While these examples are deliberately facetious, there is absolutely no qualitative difference between them and the kind of roots which Anthony proposes for wheels, which have a limited semantic range meaning “thing that turns/goes round/runs”. As such, they are intended to show that
calque formation can generate an extraordinary diversity of expressions for an item of technology invented on a single occasion within a very restricted time frame and hence that such expressions do not allow any inferences to be drawn as to whether there was a single invention of an item of technology or multiple inventions (e.g. the French mania for coining their own neologisms, in this case ‘high flow rate’, by no means entails an independent invention). Nor do they tell us anything about the point of origin of this item. They also violate the rule of thumb according to which, the degree of lexical differentiation is a function of the age of the item. All that is required for this process to occur is a network and if anything, it shows that the more extensive the network linking speakers, the more diversity it generates, since the number of routes for dissemination is multiplied."
 
Is it possible that the Yamnaya samples are later than those “early” Yamnaya folk that migrated down the Danube? Yamnaya samples seem to be Sicilian-like in pigmentation, but the Z2103 Hungarian Bell Beaker with tonnes of Yamnaya ancestry has light skin and red hair. Perhaps that would change affinities somewhat, idk
Suvorovo derivatives could be labelled early Yamnaya, although this is likely to mislead. All I see if that Central Russian, Southern Russian, Ukrainian and East Balkan Yamnaya all best fit with each other, but none with Bell Beaker.
 
That said, I think it should still be explained how and when precisely non-Anatolian IE came to be spoken as early as the Chalcolithic in the steppes probably even before Yamnaya, and therefore before any significant even if minor Anatolian Neolithic ancestry existed there).
How do we know non-Anatolian IE was spoken in the Steppe that early? Who do we know spoke it, when and where?
 
If they establish that as a premise, well, then the most honest thing they should say is: "we have come to the conclusion that genetics has no place in this linguistic discussion". And they of course will have to avoid deducing anything touching on the subject of linguistics or even on ethnicity based on genetics, because they will have concluded that genetically untraceable, tiny migrations and cultural contacts might often have triggered language shift, even when there was clearly no acculturation as in the steppes, where despite heavy foreign influences there is a cultural evolution without severe ruptures from the Neolithic to the Bronze Age. Sorry, but this is pseudoscience. If they really think a language transfer happened without admixture, but they obviously do not what languages were spoken by Maykop or other Caucasian cultures back then, or even by the EEF farmers, then they should just quit from this discussion as eeal scientists, not "speculators".

I think I've misinterpreted the authors and they do indeed trace IE to the CHG in the Caucasus steppe samples - see the tracer dye comment. I'm not sure if that's much more convincing, but let's see what they have.

It does make one wonder why they didn't test Caucasus Neolithic samples for this study if that's their hypothesis. It could have cleared so many things up, now we're left wondering :frown:
 
Well, if they were of Anatolian origin in the sense that they were mainly EEF mixed with some EHG, then their language was only distantly Anatolian in any way, they would have made their language evolve in Europe itself for some 2500 years by the time of the early Chalcolithic, more than enough time to become a totally different language (sub)family. Anatolian certainly split much later, in Chalcolithic times, not during the initial spread of farming. It shares too much common cultural/economic vocabulary with the other IE branches to be that early. So, even if your hypothesis is correct, that would make PIE a Balkanic language, not an Anatolian one.
Anatolian in the sense of Anatolian Neolithic, not EEF Neolithic. Yes, a Balkanic language of recent Anatolian descent, mixed with substantial Southern Steppe features. Only a possibility, not a theory.

By the way, the small amount of Anatolian in Yamnaya best fits with East Balkan (retreating Suvorovo, proto-CW) ancestry.
 
How do we know non-Anatolian IE was spoken in the Steppe that early? Who do we know spoke it, when and where?

That is if non-Anatolian IE is associated with some relevant demographic impact (there was no major change between steppe Eneolithic and Yamnaya, either cultural or genetic, only gradual changes along 1000 years), and if Afanasievo was indeed IE, because its star dates to the very birth of the Yamnaya horizon, implying that LPIE was already spoken there before the Yamnaya expansion. Of course these are all just speculative premises, but they make much more sense than claiming a language transfer due to a "tracer dye" without any other relevant evidence, especially when the data of the study itself do not suggest that the bulk of the Caucasian-like ancestry in the steppes was related at all with the contemporaneous Chalcolithic Caucasasian people, at least Maykop if that is what theycare thinking as a source (and that is particularly true for the male lineages, suggesting a very minor and to top that female-biased migration).
 
Anatolian in the sense of Anatolian Neolithic, not EEF Neolithic. Yes, a Balkanic language of recent Anatolian descent, mixed with substantial Southern Steppe features. Only a possibility, not a theory.

By the way, the small amount of Anatolian in Yamnaya best fits with East Balkan (retreating Suvorovo, proto-CW) ancestry.

So according to you Suvorovo-Novodanilovka, Ezero and so on were not a product of the Neolithic culture of Southeastern Europe that had been evolving since more than 2000 years earlier, but a recent arrival from Anatolia?

As for the EEF in Yamnaya I already thought it probably should have come from the East Balkans and/or Cucuteni-Tripolye. That is another issue I have with th claims of Wang. If Maykop and other cultures of the Caucasus from the Chalcolithic to the Bronze Age had indeed made occasional but not negligible migrations to the steppes and influenced their people decisively, since the Caucasus was heavy in ANF proper ancestry by then, we should expect the Anatolian in Yamnaya to be more southern (Caucasian, Anatolian) than western (Eastern European, mixed with relevant amounts of WHG). Claims about Yamnaya being in West Asia do not make things any better.
 
I think I've misinterpreted the authors and they do indeed trace IE to the CHG in the Caucasus steppe samples - see the tracer dye comment. I'm not sure if that's much more convincing, but let's see what they have.

It does make one wonder why they didn't test Caucasus Neolithic samples for this study if that's their hypothesis. It could have cleared so many things up, now we're left wondering :frown:

From the link I posted :

"Interestingly, the autosomal genetic component in Europeans considered to derive from the Steppe is almost fixed in two pre-Neolithic ancient genomes from the South Caucasus
."

That does seem to me that there was 50%EHG and 50%CHG South of the Caucasus in Neolithic times, doesn´t it ?
 
Guys the link works and there are way more gems than just that point:



"If correct, the evident linguistic conclusion would be that Anatolian languages were already in situ in Anatolia by the 6th millennium BCE.
By contrast, the PIE wine word is absent from Indo-Iranian, a culture which, according to Gamkrelidze and Ivanov (p. 562), lost it because it adopted soma (whatever that was). We find a word related to ‘vine’ in Avestan vaēiti, but it tells us nothing about when proto-Indo-Iranian migrated eastwards and whether it preserved the original word for wine.
Nor were there ever any lions on the Pontic Steppes but the term is present in all IE families including Anatolian (albeit with reduplication -*wolw-o-) other than Armenian and Indo-Iranian, which derive their forms from a root singho-. On this basis, a PIE term for ‘lion’ (ibidem, p. 430) *leu- can be reconstructed (although as stated, this by no means implies that there actually was a PIE word for lion). There is also a cognate in Kartvelian *lom-, but the term is probably an Afroasiatic borrowing (cf. Old Egyptian rw, Akkadian lābu)
.

"In my 2008 article (Sherman Morris, 2009), I highlighted the word for ‘broadband’ in Germanic: we have breiðband in Icelandic, breëband in Afrikaans, breedband in Dutch, breitband in German and bredbånd in Danish, etc., all
showing regular sound shifts from Proto-Germanic, which is reconstructible as proto-Germanic *braiđazbanđan, albeit without this implying that proto-Germanic tribes had the Internet.

While these examples are deliberately facetious, there is absolutely no qualitative difference between them and the kind of roots which Anthony proposes for wheels, which have a limited semantic range meaning “thing that turns/goes round/runs”. As such, they are intended to show that
calque formation can generate an extraordinary diversity of expressions for an item of technology invented on a single occasion within a very restricted time frame and hence that such expressions do not allow any inferences to be drawn as to whether there was a single invention of an item of technology or multiple inventions (e.g. the French mania for coining their own neologisms, in this case ‘high flow rate’, by no means entails an independent invention). Nor do they tell us anything about the point of origin of this item. They also violate the rule of thumb according to which, the degree of lexical differentiation is a function of the age of the item. All that is required for this process to occur is a network and if anything, it shows that the more extensive the network linking speakers, the more diversity it generates, since the number of routes for dissemination is multiplied."

Well, I think the example is really facetious, because they are comparing a very obvious calque word (broad + band) used in a field of technology of literate, modern societies. Things certainly worked differently in the late Copper Age when wheels were invented. But, anyways, that kind of argument could be used to virtually any word, even "wine" (why couldn't it have been borrowed from Kartvelian with slightly different forms . Also, the Anthony proposal for the dating of PIE does not rest solely on the word for "wheel", but on words for agricultural implements (which did not exist much before the Chalcolithic in the steppe cultures), in the word for "wool" (which was not developed in sheep much before that either), in words for some metals, and so on. It is a cultural package. It is also confirmed by some assumotions of linguists about the rate of linguistic difergence, It is kind of facetious to just claim that, if PIE was much older than thought, all the IE would still have maintained a strong network and exchanged many borrowings without many changes, all of them adapting the borrowed words to their own phonetics consciously. Not even in modern integrated Europe you can see that consistently.

As for wine, it may be that, as many have speculated in the past, Indo-Iranian was one of the northernmost dialects of the PIE dialect continuum. They may not have lost the word, they may just have never used it regularly, but the southerners had mire use to it after they borrowed it from Kartvelian or a third source that gave both language families the root "wine".

As for the word for lion, honestly if the authors themselves claim that a very similar word is found in Kartvelian and looks like a borrowing from Semitic (or, they should also have thought this, maybe a common borrowing in Kartvelian and Semitic from a third language?), why are they assuming that lions should be present in the steppes for PIE to be a steppe language? Lions were present in the Balkans and in the Caucasus neighboring the steppes on both sides, so it is no wonder they could have had a word for it, especially since it is certain that steppe people migrated to the Balkans as early as Suvorovo or even earlier Vinca period. And it would then make sense that they used a foreign word and not a native term. they really want us to believe that Kartvelian lom- is a loanword related to Egyptian rw- and Akkadian labu, but the even more similar PIE leu- is a native word for an animal that supposedly existed in their own environment? Hmm, not very convincing, but it fits their preferred narrative.
 
As for the word for lion, honestly if the authors themselves claim that a very similar word is found in Kartvelian and looks like a borrowing from Semitic (or, they should also have thought this, maybe a common borrowing in Kartvelian and Semitic from a third language?), why are they assuming that lions should be present in the steppes for PIE to be a steppe language? Lions were present in the Balkans and in the Caucasus neighboring the steppes on both sides, so it is no wonder they could have had a word for it, especially since it is certain that steppe people migrated to the Balkans as early as Suvorovo or even earlier Vinca period. And it would then make sense that they used a foreign word and not a native term. they really want us to believe that Kartvelian lom- is a loanword related to Egyptian rw- and Akkadian labu, but the even more similar PIE leu- is a native word for an animal that supposedly existed in their own environment? Hmm, not very convincing, but it fits their preferred narrative.

Not only that, but think also how animals' names tend to shift rather freely from one species to the next, from the general to the particular, or vice versa.

The puma is known in French as "lion des montagnes", the mountain lion. A "Tier" in German is just an/any animal, while an English "Deer" is specific. "Felis" in Latin could designate a cat or... a marten. "Bear" is speculatively assumed to descend from PIE *bheros (brown animal), which ended up as *bhebhros (beaver) with reduplication.

Who can tell for sure what kind of felines there were north of the Caucasus that could be callled lions without too much semantic laxity?
 
From the link I posted :

"Interestingly, the autosomal genetic component in Europeans considered to derive from the Steppe is almost fixed in two pre-Neolithic ancient genomes from the South Caucasus
."

That does seem to me that there was 50%EHG and 50%CHG South of the Caucasus in Neolithic times, doesn´t it ?

I think they are merely referring to Kotias and Satsurbalia. The problem with those is of course that people like them likely weren't around anymore by the time of the IE dispersals. Shuvaleri Shomu might be the remaining option for the hybrid hypothesis, but if I had to bet I'd probably say that this isn't where the southern component in the North Caucasus steppe comes from. An origin in SS doesn't fit with that:

eneolithic_steppe.png
 
Well, I think the example is really facetious, because they are comparing a very obvious calque word (broad + band) used in a field of technology of literate, modern societies. Things certainly worked differently in the late Copper Age when wheels were invented. But, anyways, that kind of argument could be used to virtually any word, even "wine" (why couldn't it have been borrowed from Kartvelian with slightly different forms . Also, the Anthony proposal for the dating of PIE does not rest solely on the word for "wheel", but on words for agricultural implements (which did not exist much before the Chalcolithic in the steppe cultures), in the word for "wool" (which was not developed in sheep much before that either), in words for some metals, and so on. It is a cultural package. It is also confirmed by some assumotions of linguists about the rate of linguistic difergence, It is kind of facetious to just claim that, if PIE was much older than thought, all the IE would still have maintained a strong network and exchanged many borrowings without many changes, all of them adapting the borrowed words to their own phonetics consciously. Not even in modern integrated Europe you can see that consistently.

As for wine, it may be that, as many have speculated in the past, Indo-Iranian was one of the northernmost dialects of the PIE dialect continuum. They may not have lost the word, they may just have never used it regularly, but the southerners had mire use to it after they borrowed it from Kartvelian or a third source that gave both language families the root "wine".

As for the word for lion, honestly if the authors themselves claim that a very similar word is found in Kartvelian and looks like a borrowing from Semitic (or, they should also have thought this, maybe a common borrowing in Kartvelian and Semitic from a third language?), why are they assuming that lions should be present in the steppes for PIE to be a steppe language? Lions were present in the Balkans and in the Caucasus neighboring the steppes on both sides, so it is no wonder they could have had a word for it, especially since it is certain that steppe people migrated to the Balkans as early as Suvorovo or even earlier Vinca period. And it would then make sense that they used a foreign word and not a native term. they really want us to believe that Kartvelian lom- is a loanword related to Egyptian rw- and Akkadian labu, but the even more similar PIE leu- is a native word for an animal that supposedly existed in their own environment? Hmm, not very convincing, but it fits their preferred narrative.

I think the point is to show how weak the linguistic arguments employed by Anthony and co. are :grin:

Morris prefers an Anatolian homeland, but that's not the meat of the paper.

The likely Semitic loans pointed out by I&G include 'oxe', 'goat', 'axe', 'star', 'seven' etc. A list can be found in their 'Indo-Europeans', p. 768.
 
I think they are merely referring to Kotias and Satsurbalia. The problem with those is of course that people like them likely weren't around anymore by the time of the IE dispersals. Shuvaleri Shomu might be the remaining option for the hybrid hypothesis, but if I had to bet I'd probably say that this isn't where the southern component in the North Caucasus steppe comes from. An origin in SS doesn't fit with that:
eneolithic_steppe.png
Neolithic times in Northern Iran/Armenia starts at 8000BC. Late Neolithic/Early Bronze Age(5000BC) these guys went north and stayed there. David Reich said something like this area is really isolated IIRC. Also the Viking Royal lady, Late Hallstatt Princes have mtDNA U7, sounds like "Elite" to me. Which is also in the steppe group in Wang et al.(2018). U7 peaks in Northern Iran and Azerbaijan, area of former Leyla Teppe culture where the first Kurgans were found. The dates would fit like in the model proposed by Krauses Team. Wine was also found there about the same time like in Shuvaleri.
I bet that I have found those guys :) Coming papers will show anyways , I just wanted to share with you guys something I think is really fitting in like a puzzle.
 
Last edited:
I think the point is to show how weak the linguistic arguments employed by Anthony and co. are :grin:

Morris prefers an Anatolian homeland, but that's not the meat of the paper.

The likely Semitic loans pointed out by I&G include 'oxe', 'goat', 'axe', 'star', 'seven' etc. A list can be found in their 'Indo-Europeans', p. 768.

I wonder what methodology they used to establish words as likely Semitic loans. I hope not mere sound similarity, and that they in fact derived some regular sound correspondences that are consistently found in the Semitic vs. IE apparent cognates.
 
Neolithic times in Northern Iran/Armenia starts at 8000BC. Early Neolithic/Late Bronze Age(5000BC) these guys went north and stayed there. David Reich said something like this area is really isolated IIRC. Also the Viking Royal lady, Late Hallstatt Princes have mtDNA U7, sounds like "Elite" to me. Which is also in the steppe group in Wang et al.(2018). U7 peaks in Northern Iran and Azerbaijan, area of former Leyla Teppe culture where the first Kurgans were found. The dates would fit like in the model proposed by Krauses Team. Wine was also found there about the same time like in Shuvaleri.

I bet that I have found those guys :) Coming papers will show anyways , I just wanted to share with you guys something I think is really fitting in like a puzzle.

I also speculated that the CHG/Iranian part of the steppe expansion could have come partly from Leyla-Tepe, but given the Eneolithic steppe results I started to think that culture was maybe too late, for I had assumed a really significant demographic impact considering their economic and social/class superiority, not just an incremental increase. I think we really need more high coverage samples from the Neolithic steppe and Caucasus, roughly 5500-4500 B.C.
 

This thread has been viewed 83750 times.

Back
Top