Origin of Indo-European Languages Traced Back to 8000 Years Ago
WeepingElf wrote:
A New IE Origin Theory
A New IE Origin Theory
I came across an article on some new work on the geographical origins of Indo-European speakers and thought it might be of interest to the community. Also, I am hoping that Joerg might offer some comment on this.
Origin of Indo-European Languages Traced Back to 8000 Years Ago
Origin of Indo-European Languages Traced Back to 8000 Years Ago
Re: A New IE Origin Theory
Coincidentally, WeepingElf started asking about this over on the ZBB's PIE thread (linked through a Language Log post. I think the running thought over there at the moment is that while the lexical dataset (and its sources) is quite solid, the methodology for determining age probably isn't the bestelemtilas wrote: ↑08 Aug 2023 02:32 I came across an article on some new work on the geographical origins of Indo-European speakers and thought it might be of interest to the community. Also, I am hoping that Joerg might offer some comment on this.
Origin of Indo-European Languages Traced Back to 8000 Years Ago
WeepingElf wrote:
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
Re: A New IE Origin Theory
It's bollocks.
- glottochronology is extremely suspect
- you cannot learn whether a language was spoken north or south of the caucasus by studying which languages split off first - not in an extremely complicated family that evolved over thousands of years, in which all known branches are known to have travelled large distances
- they think Indo-Iranian branched off thousands of years before the invasion of Europe. But, factually, we know that Indo-Iranian speakers were involved in the invasion of Europe. They did not move southeast from the steppe to India and Iran; they moved northwest from the steppe into northeastern Europe, and THEN moved back across the steppe to India and Iran. We know this because they obtained genes from Europe along the way; they're also generally more related to other modern IE groups than they are to the genes of their original steppe ancestors. This is also strongly reinforced by the archaeology and the history: Indo-Iranian identity is generally considered to descend from the Sintashta Culture (around 4k ago), which expanded (as the Andronovo culture) into what became II territory, and the material and ritual culture of which closely match the Vedic descriptions. Sintashta individuals are genetically and largely culturally an eastward extension of the baltic Corded Ware Culture. Now, language doesn't always follow genes, and it's conceivably possible that Sintashta borrowed its language from the people they replaced on the steppe (who would have been their distant cousins), and thus that II as a language split earlier than Sintashta as a culture did. However, that's a less neat and probable story, and, combined with the extemely suspect nature of glottochronology as a method, should make us very skeptical of the results. Dodgy method + unlikely results = skepticism.
- there is no genetic evidence of ancestry in the middle east. There are some cultural suggestions of it - the building of kurgans has long been considered suspiciously similar to the building of mounds south of the caucasus (perhaps inspired by pyramids and ziggurats) and may have spread across the caucasus.It is also known that steppe cultures did have some genetic ancestry south of the caucasus. However, similar ancestry had been on the steppe for thousands of years prior to the PIE expansions. There's no smoking gun suggesting any migration from the middle east in a timespan that would be relevant for the formation or spread of PIE. [in fact, PIE speakers probably had two sorts of middle-eastern ancestry - one from across the caucasus, and one from European farmers who had spread there from the Levant].
- what is known is that all IE speakers we have good genetic evidence for are descended from an expansion out of the steppe. It's not certain that this is true of Anatolian speakers because there's very little evidence, but there's also little reason to think that it isn't.
- glottochronology is extremely suspect
- you cannot learn whether a language was spoken north or south of the caucasus by studying which languages split off first - not in an extremely complicated family that evolved over thousands of years, in which all known branches are known to have travelled large distances
- they think Indo-Iranian branched off thousands of years before the invasion of Europe. But, factually, we know that Indo-Iranian speakers were involved in the invasion of Europe. They did not move southeast from the steppe to India and Iran; they moved northwest from the steppe into northeastern Europe, and THEN moved back across the steppe to India and Iran. We know this because they obtained genes from Europe along the way; they're also generally more related to other modern IE groups than they are to the genes of their original steppe ancestors. This is also strongly reinforced by the archaeology and the history: Indo-Iranian identity is generally considered to descend from the Sintashta Culture (around 4k ago), which expanded (as the Andronovo culture) into what became II territory, and the material and ritual culture of which closely match the Vedic descriptions. Sintashta individuals are genetically and largely culturally an eastward extension of the baltic Corded Ware Culture. Now, language doesn't always follow genes, and it's conceivably possible that Sintashta borrowed its language from the people they replaced on the steppe (who would have been their distant cousins), and thus that II as a language split earlier than Sintashta as a culture did. However, that's a less neat and probable story, and, combined with the extemely suspect nature of glottochronology as a method, should make us very skeptical of the results. Dodgy method + unlikely results = skepticism.
- there is no genetic evidence of ancestry in the middle east. There are some cultural suggestions of it - the building of kurgans has long been considered suspiciously similar to the building of mounds south of the caucasus (perhaps inspired by pyramids and ziggurats) and may have spread across the caucasus.It is also known that steppe cultures did have some genetic ancestry south of the caucasus. However, similar ancestry had been on the steppe for thousands of years prior to the PIE expansions. There's no smoking gun suggesting any migration from the middle east in a timespan that would be relevant for the formation or spread of PIE. [in fact, PIE speakers probably had two sorts of middle-eastern ancestry - one from across the caucasus, and one from European farmers who had spread there from the Levant].
- what is known is that all IE speakers we have good genetic evidence for are descended from an expansion out of the steppe. It's not certain that this is true of Anatolian speakers because there's very little evidence, but there's also little reason to think that it isn't.
-
WeepingElf
- greek
- Posts: 538
- Joined: 23 Feb 2016 18:42
- Location: Braunschweig, Germany
- Contact:
Re: A New IE Origin Theory
I fully concur with Salmoneus here. This highlights the difficulties of obtaining linguistic information from genetic evidence. I am not a geneticist, so I may have gotten those things wrong, but apparently, the Yamnaya people formed from the mixture of a population native to Eastern Europe (the "Eastern European Hunter-Gatherers", EHG) and one originating south of the Caucasus (the "Caucasian Hunter-Gatherers", CHG). The EHG component was closely related to the likely speakers of Proto-Uralic, and in turn was a mixture of native Europeans ("Western European Hunter-Gatherers", WHG) with a Siberian group (the "Ancient North Eurasians", ANE).
There are two different scenarios one can attempt to reconstruct here:
Either PIE was a language spoken by the EHG lineage influenced by that of the CHG lineage contributing to the Yamnaya.
Or PIE was a language spoken by the CHG lineage influenced by that of the EHG lineage contributing to the Yamnaya.
The scientists whose work we are discussing here apparently support the latter scenario, assuming that Anatolian was the language of an unmixed CHG lineage. But it could just as easily have been the former scenario - and the morphological resemblances between IE and Uralic suggest that this is the right answer (the much-discussed lexical resemblances look rather like loanwords from IE into Uralic, though).
Also, I don't think that the divergence of Anatolian and (as I call it) Northern IE was as early as 8,100 years ago. There are at least two reasons to assume that it was not very long before the breakup of Northern IE: 1. The phonologies of "pre-Anatolian" and "post-Anatolian" PIE were very similar, as the phonology of "handbook PIE" accounts for Anatolian reasonably well - it seems as if only the phonetic values of some of the phonemes may have changed, but no major restructuring of the phonology had happened. 2. Anatolian shares terminology pertaining to animal traction (e.g. *yugom 'yoke') with Northern IE, so their common ancestor was spoken by a society that used animal traction, at least for ploughs or sleighs if not wagons, which means that this common ancestor was spoken not long before 4000 BC. Nobody had animals draw anything in 6100 BC!
I don't know much about the archaeogenetics of Anatolia, but according to this blog post, the Hittites and other Anatolian IE speakers entered Anatolia only in the Bronze Age between 3000 and 2000 BC - and that probably from the west, as the area east of Anatolia was full of non-IE languages, and the Caucasus of course still is so today.
There are two different scenarios one can attempt to reconstruct here:
Either PIE was a language spoken by the EHG lineage influenced by that of the CHG lineage contributing to the Yamnaya.
Or PIE was a language spoken by the CHG lineage influenced by that of the EHG lineage contributing to the Yamnaya.
The scientists whose work we are discussing here apparently support the latter scenario, assuming that Anatolian was the language of an unmixed CHG lineage. But it could just as easily have been the former scenario - and the morphological resemblances between IE and Uralic suggest that this is the right answer (the much-discussed lexical resemblances look rather like loanwords from IE into Uralic, though).
Also, I don't think that the divergence of Anatolian and (as I call it) Northern IE was as early as 8,100 years ago. There are at least two reasons to assume that it was not very long before the breakup of Northern IE: 1. The phonologies of "pre-Anatolian" and "post-Anatolian" PIE were very similar, as the phonology of "handbook PIE" accounts for Anatolian reasonably well - it seems as if only the phonetic values of some of the phonemes may have changed, but no major restructuring of the phonology had happened. 2. Anatolian shares terminology pertaining to animal traction (e.g. *yugom 'yoke') with Northern IE, so their common ancestor was spoken by a society that used animal traction, at least for ploughs or sleighs if not wagons, which means that this common ancestor was spoken not long before 4000 BC. Nobody had animals draw anything in 6100 BC!
I don't know much about the archaeogenetics of Anatolia, but according to this blog post, the Hittites and other Anatolian IE speakers entered Anatolia only in the Bronze Age between 3000 and 2000 BC - and that probably from the west, as the area east of Anatolia was full of non-IE languages, and the Caucasus of course still is so today.
... brought to you by the Weeping Elf
My conlang pages
My conlang pages
Re: A New IE Origin Theory
Regarding glottochronology, I thought it was amusing to hear someone recently point out: why was it taken seriously by anyone when languages as exotic as Latin > French vs. Italian disprove it? (Even if we limit "Italian" to the "natural" dialects of Tuscan, the point stands.) It is also obvious to everyone that 1 AD Latin > 1000 AD Old Spanish involved many more changes than 1000 AD Old Spanish > 2000 AD Spanish. If languages can have greatly faster or slower rates of change for whatever reasons it may be, then glottochronology falls quite flat on its face.
Another related and interesting data point is Semitic... Akkadian shows up around 2500 BC, and Proto-Semitic may be about as old as Proto-Indo-European. While this is quite subjective it's kind of remarkable how similar all four of Akkadian, Biblical Hebrew, Ge'ez and Classical Arabic are. As I heard someone once say, "there aren't any Tocharians or Armenians or Old Irishes in older Semitic, relatively old descendents that have just gone completely crazy"...
Over in Algonquian, Meskwaki/Fox is far more conservative in relation to Proto-Algonquian than Blackfoot or Arapaho.
Another related and interesting data point is Semitic... Akkadian shows up around 2500 BC, and Proto-Semitic may be about as old as Proto-Indo-European. While this is quite subjective it's kind of remarkable how similar all four of Akkadian, Biblical Hebrew, Ge'ez and Classical Arabic are. As I heard someone once say, "there aren't any Tocharians or Armenians or Old Irishes in older Semitic, relatively old descendents that have just gone completely crazy"...
Over in Algonquian, Meskwaki/Fox is far more conservative in relation to Proto-Algonquian than Blackfoot or Arapaho.
hīc sunt linguificēs. hēr bēoþ tungemakeras.
Re: A New IE Origin Theory
I have a feeling that Semitic stayed so similar because the family was and is quite close geographically, and because it was easy to borrow from one language to another due to similar grammar and phonology across the family. So in a sense, Semitic influenced itself quite a bit, enough to keep the languages similar to each other. I think.
Meanwhile, Tocharian and Irish and whoever never stood a chance to be like each other, and the same for Fox, Blackfoot, and Arapaho.
Meanwhile, Tocharian and Irish and whoever never stood a chance to be like each other, and the same for Fox, Blackfoot, and Arapaho.
-
WeepingElf
- greek
- Posts: 538
- Joined: 23 Feb 2016 18:42
- Location: Braunschweig, Germany
- Contact:
Re: A New IE Origin Theory
As much as this "hybrid theory" is misguided, one should not be too harsh about the authors. They are not the usual kind of crackpots one meets so often on the 'Net, but good scientists. The problem is merely that they ventured into a realm where they aren't experts, and made mistakes. They had not realized that glottochronology has been falsified long ago, and the conclusion they made about the origin of PIE seems plausible enough to anyone who doesn't know the relevant linguistic details. I have made similar mistakes in the past (though only in my conlanging), in a sense in the other direction (getting the genetics wrong), so I think I know what kind of mistake has happened here.
(The harsh criticism on the Eurogenes blog is inappropriate, not only in suggesting that the authors were idiots which they aren't, but also because he rejects the theory for the wrong reason: the author of that blog post is a dissenter in archaeogenetics who rejects the widely accepted theory that the Yamnaya descend in part from a population originating south of the Caucasus, as is evident from various earlier posts on his blog, and has used inappropriately strong language in the past.)
(The harsh criticism on the Eurogenes blog is inappropriate, not only in suggesting that the authors were idiots which they aren't, but also because he rejects the theory for the wrong reason: the author of that blog post is a dissenter in archaeogenetics who rejects the widely accepted theory that the Yamnaya descend in part from a population originating south of the Caucasus, as is evident from various earlier posts on his blog, and has used inappropriately strong language in the past.)
... brought to you by the Weeping Elf
My conlang pages
My conlang pages
-
WeepingElf
- greek
- Posts: 538
- Joined: 23 Feb 2016 18:42
- Location: Braunschweig, Germany
- Contact:
Re: A New IE Origin Theory
I have thought again about the "hybrid theory" papers we discussed a few weeks ago, and I have an idea why they have Anatolian and Tocharian diverge so impossibly early. Well, what those people did was to feed a distance matrix based on lexical cognates into an algorithm designed to compute family trees from genetic distance matrices. Now, Anatolian and Tocharian are extinct branches whose lexica are incompletely known - Tocharian because the corpus is small and doesn't cover all subject matters, and Hittite because many words are hidden behind Sumerograms. Now, unknown words count as "not cognate", and thus the distance appears exaggerated.
... brought to you by the Weeping Elf
My conlang pages
My conlang pages