This was utterly fascinating, thank you. Can’t imagine the amount of time and research that must go into writing your pieces.
I hadn’t heard of the Southern Arc theory before, only the Steppe Theory. If I’m not wrong, Razib Khan seems to advocate only for the latter? Though I haven’t read all of his pieces so perhaps I’m misreading him.
Genetic archaeology is such a fascinating new frontier, can’t wait to see what we learn next.
Since *PIE is a linguistic construct the homeland problem should really revolve around its reconstructions. Thus it's always baffled me that any linguist would back the Anatolian theory.
But I think you're wrong (or not completely right) that linguists are its main proponents; in my experience it's archaeologists who like the Anatolian theory best. The reasons are manifold: non-violent expansion (allegedly but implausibly; oh how they love a 'non-violent expansion'), their ignorance of the geography and technology implied by *PIE, similarity to Austronesian expansion (i.e. spread of farming), even the scope it permitted for fitting prehistory into Gould's 'punctuated equilibrium' model. In criticising archaeologists I'm by no means trying to mount a defence of linguists, who are quite as bad in their own ways.
Incidentally, Renfrew largely repudiated the Anatolian theory some years ago, which I thought was very magnanimous of him.
Of the remaining two theories, I'm not sure which is more likely. It's worth bearing in mind that the homeland problem is one thing; the location from where the historically important expansions occurred is quite another. For example, I think there can be little doubt that the Corded Ware/Battle Axe people expanded from the Forest-Steppe zone and not the Caucasus--and we know well how consequential *they* were for European and world history. But then there is the problem of the Armenians, Greeks etc. (see Robert Drews, who favours a much later expansion than the Steppe hypothesis suggests, with--I think--the Caucasus as the urheimat).
There is linguistic evidence of deep structural similarities between Indo-European and Kartvelian languages, but it's also there for Indo-European and Uralic (mainly in the form of tell-tale loanwords, as I understand it). Anyway I will stop now because I'm a dilettante and might get shot down if I continue.
Uralic and IE contacts are probably from later - mid-2nd millennium when the Indo-Iranians were wandering around South Siberia. The Kartvelian similarities are interesting - perhaps a legacy of those south Caucasian migrants who made it onto the steppe in the 5th and 4th millennia.
I'm uncertain about dates for contact between IE and Uralic, but the idea is that there are several temporal layers and that some borrowings are very early indeed (going back to proto-Uralic - *PIE). They're also said to be more secure as borrowings than those proposed for IE - Kartvelian.
Anyway, the most interesting of them is that Finnish preserves a word for slave, 'orja', which--and you don't have be a linguist to see this--comes directly from the word Aryan. But this is probably among the later ones you mention.
I think it is a Western Hunter-Gatherer language that was imposed on the Early European Farmers in Iberia about 4400 BC. There was a big WHG refuge in Cantabria.
What are the closest samples (in time and space) to the Suvorovo and Ezero cultures? How much steppe ancestry do they have?
I think they’re going to end up having a bunch of steppe ancestry that simply got diluted by the time they entered into Anatolia (and then diluted further in Anatolia itself). No wonder how little steppe all the Bronze and Iron Age Anatolians have - they are mostly dated to 1000 years or later than the Anatolian-steppe populations arrival.
there are some Suvorovo samples that should be published shortly and are supposed to have a noticeable amount of steppe ancestry. For Ezero, I believe there is a post-3000 BC sample published that has steppe ancestry, but what we really need is a pre-3000 BC sample. There are some late 4th millennium BC samples from northwestern Anatolia, but unfortunately they are low resolution so hard to draw any conclusions from.
Any insights into the Tartaria tablets, somewhat controversial but certified by some competent researchers to be the oldest written tablet ever found,the Turdaș - Vinça culture and the Cucuteni culture which seem to have many common elements?
The Anatolian theory was also proposed by the Soviet linguist, Gamqrelidze and Ivanov. They thought it's modern eastern Anatolian where it spread out from.
Looking at the date of divergence in the Heggarty paper the thing that’s most striking to me is that 8120 years before the present is right after the so-called “8.2-kiloyear event” that you’ve written about previously. That would be a very natural time for PIE language to fragment as climatic change prompted significant migration.
Have you read Lazaridis paper in Science from last year? “The genetic history of the Southern Arc: A bridge between West Asia and Europe”. Even has David Reich’s imprimatur. Pretty much seals the case for Anatolian being a sister language to Indo-European.
This was utterly fascinating, thank you. Can’t imagine the amount of time and research that must go into writing your pieces.
I hadn’t heard of the Southern Arc theory before, only the Steppe Theory. If I’m not wrong, Razib Khan seems to advocate only for the latter? Though I haven’t read all of his pieces so perhaps I’m misreading him.
Genetic archaeology is such a fascinating new frontier, can’t wait to see what we learn next.
I think Razib is agnostic on Southern Arc vs Steppe - I'd have to ask him though.
Since *PIE is a linguistic construct the homeland problem should really revolve around its reconstructions. Thus it's always baffled me that any linguist would back the Anatolian theory.
But I think you're wrong (or not completely right) that linguists are its main proponents; in my experience it's archaeologists who like the Anatolian theory best. The reasons are manifold: non-violent expansion (allegedly but implausibly; oh how they love a 'non-violent expansion'), their ignorance of the geography and technology implied by *PIE, similarity to Austronesian expansion (i.e. spread of farming), even the scope it permitted for fitting prehistory into Gould's 'punctuated equilibrium' model. In criticising archaeologists I'm by no means trying to mount a defence of linguists, who are quite as bad in their own ways.
Incidentally, Renfrew largely repudiated the Anatolian theory some years ago, which I thought was very magnanimous of him.
Of the remaining two theories, I'm not sure which is more likely. It's worth bearing in mind that the homeland problem is one thing; the location from where the historically important expansions occurred is quite another. For example, I think there can be little doubt that the Corded Ware/Battle Axe people expanded from the Forest-Steppe zone and not the Caucasus--and we know well how consequential *they* were for European and world history. But then there is the problem of the Armenians, Greeks etc. (see Robert Drews, who favours a much later expansion than the Steppe hypothesis suggests, with--I think--the Caucasus as the urheimat).
There is linguistic evidence of deep structural similarities between Indo-European and Kartvelian languages, but it's also there for Indo-European and Uralic (mainly in the form of tell-tale loanwords, as I understand it). Anyway I will stop now because I'm a dilettante and might get shot down if I continue.
Gud article!
Uralic and IE contacts are probably from later - mid-2nd millennium when the Indo-Iranians were wandering around South Siberia. The Kartvelian similarities are interesting - perhaps a legacy of those south Caucasian migrants who made it onto the steppe in the 5th and 4th millennia.
I'm uncertain about dates for contact between IE and Uralic, but the idea is that there are several temporal layers and that some borrowings are very early indeed (going back to proto-Uralic - *PIE). They're also said to be more secure as borrowings than those proposed for IE - Kartvelian.
Anyway, the most interesting of them is that Finnish preserves a word for slave, 'orja', which--and you don't have be a linguist to see this--comes directly from the word Aryan. But this is probably among the later ones you mention.
I’d be curious to know your thoughts on where the Basque language came from.
I think it is a Western Hunter-Gatherer language that was imposed on the Early European Farmers in Iberia about 4400 BC. There was a big WHG refuge in Cantabria.
They were also unique in retaining signifcant GoyetQ2 ancestry, so I wonder if the origins of Basque in Western Europe are deep indeed.
What are the closest samples (in time and space) to the Suvorovo and Ezero cultures? How much steppe ancestry do they have?
I think they’re going to end up having a bunch of steppe ancestry that simply got diluted by the time they entered into Anatolia (and then diluted further in Anatolia itself). No wonder how little steppe all the Bronze and Iron Age Anatolians have - they are mostly dated to 1000 years or later than the Anatolian-steppe populations arrival.
there are some Suvorovo samples that should be published shortly and are supposed to have a noticeable amount of steppe ancestry. For Ezero, I believe there is a post-3000 BC sample published that has steppe ancestry, but what we really need is a pre-3000 BC sample. There are some late 4th millennium BC samples from northwestern Anatolia, but unfortunately they are low resolution so hard to draw any conclusions from.
Any insights into the Tartaria tablets, somewhat controversial but certified by some competent researchers to be the oldest written tablet ever found,the Turdaș - Vinça culture and the Cucuteni culture which seem to have many common elements?
I thought we won that debate 5300 years ago?
Don’t these EEFs ever give up?
The Anatolian theory was also proposed by the Soviet linguist, Gamqrelidze and Ivanov. They thought it's modern eastern Anatolian where it spread out from.
Looking at the date of divergence in the Heggarty paper the thing that’s most striking to me is that 8120 years before the present is right after the so-called “8.2-kiloyear event” that you’ve written about previously. That would be a very natural time for PIE language to fragment as climatic change prompted significant migration.
Absolutely brilliant piece! Late Prehistory is a great interest of mine, and you got a subscriber.
Interesting take. Thanks Peter.
Have you read Lazaridis paper in Science from last year? “The genetic history of the Southern Arc: A bridge between West Asia and Europe”. Even has David Reich’s imprimatur. Pretty much seals the case for Anatolian being a sister language to Indo-European.
Of course he has. That’s where he’s getting his explanation of the Southern Arc hypothesis
Ya, Lazaridis, et al.'s preprint has overturned or clarified a lot of the Southern Arc paper which was very convoluted, even for experts.
In what era do you think Tocharian diverged from other Indo-European languages?
it was probably an early split off from Yamnaya - maybe 3300 BC - and became Afanasievo culture.