search this blog

Wednesday, May 31, 2017

A homeland, but not the homeland


It seems increasingly likely that ancient DNA has identified a massive expansion, or a series of expansions, from Mesopotamia and/or surrounds in basically all directions dating to the Chalcolithic (ChL) and Bronze Age (BA). This phenomenon is mainly characterized by the simultaneous spread of:
- Iran_ChL-related genome-wide ancestry

- Y-haplogroup J

- South Caspian-specific mitochondrial haplogroups such as R2 and U7

At least two of these characteristics are shared by five groups that have appeared in the Near Eastern and African ancient DNA record as probable post-Neolithic newcomers, at least in part, at their respective sampling sites:

- Anatolia_BA, Western Turkey, 2836-1800 calBCE (Lazaridis et al. 2017)

- Egyptian mummies, Middle Egypt, 776-2 calBCE (Schuenemann et al. 2017)

- Iran_ChL, Western Iran, 4839-3796 calBCE (Lazaridis et al. 2016)

- Levant_BA, Northwestern Jordan, 2489-1966 calBCE (Lazaridis et al. 2016)

- Sidon_BA, Southern Lebanon, 1750-1600 BCE (Haber et al. 2017)

I'm confident that many more such groups will soon be added to the ancient DNA record, probably including Levant_ChL from the upcoming Harney et al. 2017 (a teaser of the paper can be seen here). Below, a map of Mesopotamia courtesy of Wikipedia.


It's an interesting and important question who these likely Mesopotamian migrants and their descendants were in terms of linguistic affinities. It seems that they left a massive genetic imprint on the Near East and much of North Africa, and perhaps also Central Asia and Southeastern Europe, so they probably also left some sort of linguistic legacy.

Obviously, it's highly improbable that most of them were Indo-European speakers. So if most of them weren't Indo-Europeans, then the phenomenon I'm describing here can't be related to the Proto-Indo-European (PIE) expansion. Forget the idea of an West Asian linguistic hot spot spewing out different, distantly related language families, including Indo-European, via the migrations of closely related Iran_ChL-like populations over a span of a few thousand years; it's plain stupid.

So who were they?

See also...

A homeland, but not the homeland #2

A homeland, but not the homeland #3

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...

Steppe admixture in Mycenaeans, lots of Caucasus admixture already in Minoans (Lazaridis et al. 2017)

154 comments:

Unknown said...

Did the original Mesopotamians have any Natufian/Bedouin admixture?

Davidski said...

I'd say no. My best guess is that this expanding pop was something like a mix between Iran_ChL and Levant_Neolithic.

Simon_W said...

Yes, good point. I recently also noticed that the Balkan_BA samples and many of the Hungary_BA samples are already on the "bridge" which, in West Eurasian PCAs, links the northern European/Steppe cline with the Westasian one. This might be coincidental, but my experience with using the Maros sample in nMonte to model my own DNA showed me that it includes a lot of Armenia- and Anatolia-related admixture. And we can also see an increase of y-haplogroup J2 in Bronze Age southeastern Europe. And in Mathieson et al. 2017, in the unsupervised ADMIXTURE analysis, at K=10, the dark green CHG/Iran component often shows up with little or without a corresponding medium green EHG component, at least in the Balkan Bronze Age and also in some earlier samples like some Trypillians and Varna individuals. Makes me wonder how representative our only Baden individual really is. She looks quite regular Neolithic European, but we've seen strong individual variation in some other samples of that time. Maybe the Baden-Ezero-Troy horizon already contributed to the spread of Armenia-related admixture.

Simon_W said...

To offer a speculative ethnic interpretation: In southeastern Europe this may have been Etruscan-related and taken the ancestor of Raetic to central Europe.

Unknown said...

I'd say no. My best guess is that this expanding pop was something like a mix between Iran_ChL and Levant_Neolithic.

But Levant N does contain Natufian-like components.

Synome said...

I believe this phenomenon was caused at least in part by an effect similar to the mechanics of the "Saharan pump" theory in Africa. Multiple periods of severe aridity caused dense populations of Iran Chalcolithic descended Mesopotamians to flee in all directions. Multiple waves of mass emigration from the Mesopotamian core.

I'm not sure if all of these peoples spoke the same languages, but one family of speakers that stands out in the historical record from the Bronze Age forward is the Semitic branch of AA. Semites may have been the main group of pastoralists in the Fertile Crescent and Levant from the Early Bronze Age on. They were of course highly mobile compared to their settled neighbors, and we've already seen from BA Europe what a highly mobile (and patrilineal) pastoralist population is capable of in terms of admixture into surrounding groups, even if late PIE is an extreme case.

Akkadians. Babylonians. Assyrians. Canaanites. Early Arab tribes. All spoke Semitic languages, which overtook Sumerian as the spoken languages of Mesopotamia.

This 2009 study suggests that Semitic originated in the Northeast Levant c. 3800 BCE and spread outward to Mesopotamia, Canaan, and Arabia.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2839953/

There are multiple historical opportunities for Semitic speaking peoples to have entered Egypt en masse as well, but apparently without a permanent language shift.

What I'd like to see is some early pastoralist DNA from the dry area between Mesopotamia and the Levant. This is the original haunt of the Semite herders. That's asking for a lot, but mountain caves in the area would be a good place to start I think.

Unknown said...

This is certainly a fascinating phenomenon and needs to be considered comprehensively. We should not be too obsessed with the question of the origin of Indo-European.
But why should expansion from Mesopotamia and its surroundings not have involved various very different languages? This is what we find in the earliest records. Apart from Semitic and Indo-European there are Sumerian, Elamite, Hurrian, Hattic and perhaps others. Y-haplogroup J is very old and deeply divided. Today we often find neighbouring, genetically similar people speaking very different languages.

Rob said...

I don't think these "Iran Chalcolithic" expansions occurred simultaneously either

Azarov Dmitry said...

It was not a migration from Mesopotamia but migration via Mesopotamia. The Euphrates and Tigris played a role of major water routes for pops living in the Fertile Crescent and/or migrating via the West Asia.

Gioiello said...

@ Simon_W
"To offer a speculative ethnic interpretation: In southeastern Europe this may have been Etruscan-related and taken the ancestor of Raetic to central Europe".

For what I know from the literature on the argument, the first to hypothesize that the EEF from Northern Anatolia and Aegean Sea were the "Etruscans" was I in a letter to the great Dutch linguist Schrjiver who wrote some paper thinking having demonstrated that they speak Hattic, but my theory is that they had nothing to do with "Natufians" neither Iran as we knew from the recent papers, but they were anyway linked to Europe and the European HG, believing that I/J was anyway an HG haplogroup with R1 and C1.
By a linguistic point of view Etruscan is linked to Indo-European (Giacomo Devoto spoke of "periindoeuropeo") and the link with Rhaetian is out of any doubt. The problem is if the language of Lemnos is a residual of that migration or recent, due to mercenaries of the Mycenaeans or merchants even after that as is thought from the great linguist De Simone.
Only deep tests in Italy will say if Y hg J (both 1 and 2) is recent as the most part of people thinks or old, very old, as I think having demonstrated many times also from the YFull tree.

Davidski said...

@Martin

But why should expansion from Mesopotamia and its surroundings not have involved various very different languages?

I think it's parsimonious to assume that massive expansions of closely related peoples from a genetically homogeneous core will also largely be expansions of closely related languages.

Y-haplogroup J is very old and deeply divided.

Yeah, but I listed three genetic characteristics, so even if these groups had a really high diversity in J it's not an issue. All it might mean is that they all came from a region where J was common and diverse.

Today we often find neighbouring, genetically similar people speaking very different languages.

Yes, but we're talking about expansions here. Usually it's just one language family expanding in a big way, eating up everything else, until some of the survivors rebound. Right?

Gioiello said...

Anyway Afro-Asiatic languagues were of Natufians' (and Semite languages formed only in the peryphery of that world where hg. J dominated and not the Hg. E (some subclades, not all, being too old and too widespread). I think that that about AA is demonstrated from the great Russian linguist Milaterev (read Militaryof).

Davidski said...

@Azarov

It was not a migration from Mesopotamia but migration via Mesopotamia. The Euphrates and Tigris played a role of major water routes for pops living in the Fertile Crescent and/or migrating via the West Asia.

No, it was a migration from Mesopotamia in all directions, including into the Iranian Plateau, because Iran_N was replaced by Iran_ChL there.

Samuel Andrews said...

Ancient DNA also demonstrate people from the west migrated east into Iran and Armenia. We don't really know what happened all I see is genes got really mixed up in the Chaloclithic Middle East and migrations didn't only go in one direction.

Davidski said...

Ancient DNA also demonstrate people from the west migrated east into Iran and Armenia.

That's what I'm saying.

From Mesopotamia in all directions, including into Iran, where there was clearly a genetic change.

Iran_ChL is not wholly native to Iran. That's part of my argument in the post.

Azarov Dmitry said...

@Davidski
No, it was a migration from Mesopotamia in all directions, including into the Iranian Plateau, because Iran_N was replaced by Iran_ChL there.


Or it was a migration from all directions (Levant, Anatolia, Caucasus and Iranian Plateau) in all directions (Levant, Anatolia, Caucasus and Iranian Plateau) via Mesopotamia.

Davidski said...

I think Synome is onto something there with his early Semites theory.

http://eurogenes.blogspot.com.au/2017/05/a-homeland-but-not-homeland.html?showComment=1496230784741#c8145790518198229239

But yeah, there's no reason why other language groups could not partake in this phenomenon at some stage from the edges of the Semitic expansion.

I just don't see the Indo-Europeans taking part. It makes no sense to put them there.

Unknown said...

@Davidski
Good summary of the genetic data. A possible expansion would be the Ubaid expansion circa 4500 BCE or so. The pottery of Seh Gabi--Chalcolithic Iran--has a mixture of Dalma ware from near Lake Urmia and Ubaid pottery. We know that Ubaid pottery is found all the way to the Southern Levant at that time. Linguistically, Ubaid could be the pre-Sumerian substrate language, the so called Banana language with names like Kubaba, Inana, etc...ie syllabic reduplication. These Ubaid people could have adopted NW Semitic languages, ultimately, perhaps just like they would have adopted Sumerian in Mesopotamia.

Gioiello said...

@ Roy King
"Linguistically, Ubaid could be the pre-Sumerian substrate language, the so called Banana language with names like Kubaba, Inana, etc...ie syllabic reduplication".

In fact they are the ancestors of la la land...

Synome said...

To be clear, I don't think the Semitic expansion can account for the following Iran Chalc related groups:

Caucasus/Anatolian peoples speaking Hurrian, Urartian, Hattic and carrying Ygh J2 in large numbers.

Farmer migrants to the Iranian Plateau and Central Asia

Sumerians were in Mesopotamia first, and before them the Ubaid culture which I believe may have spoken pre or proto Sumerian; that's total speculation.

MaxT said...

@Martin
"But why should expansion from Mesopotamia and its surroundings not have involved various very different languages? This is what we find in the earliest records. Apart from Semitic and Indo-European there are Sumerian, Elamite, Hurrian, Hattic and perhaps others."

Except, Indo-European material culture originated within steppes unlike other groups you mentioned.

Synome said...

I'm also not the first person to link Yhg J to the Semitic expansion, though I am additionally linking it to the Iran Chalcolithic autosomal component, which wasn't discussed in this paper because it had not yet been discovered. Note that the authors link the Semitic expansion specifically with J1 P58:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2987219/

Ultimately, I believe both J1 and J2 derived from CHG, who may have been found all along the uplands of the Caucasus, Zagros, and Taurus mountains since at least as far back as PPNB.

MaxT said...

@Synome

Ultimately basal J represents WHG-UHG lineage, UHG being unknown hunter-gatherers from West Asia who were apparently similar to WHG. WHG is good proxy for UHG ancestry there.

IJ split somewhere on borders of Europe and West Asia. Way before 'Basal Eurasian' showed up. CHG has Basal.

Synome said...

I agree that basal J originated in WHG like ancestors. I believe they were a reflux from South Asia.

I think the J1 J2 split occurred in CHG, since both forms are found there with high diversity, though some of the diversity definitely reflects back migrations from Mesopotamia and elsewhere.

Unknown said...

@Synome
Thanks for the plug of our paper! At the time, we didn't have any aDNA to back up our ideas, but now there is a flurry of support--at least for J1 participating in the Semitic expansion. I wanted to add that some speculate that the pre-Sumerian substrate language could be related to Elamite which also has syllabic reduplication. Elamite is in the right area for Seh Gabi--Chalcolithic Iran. Also Elamite is found later in areas rich in J diversity.

Ibrahim Abdullahi said...

Where do Arabians fit in the theory?

K33 said...

Davidski:
I think Synome is onto something there with his early Semites theory.

http://eurogenes.blogspot.com.au/2017/05/a-homeland-but-not-homeland.html?showComment=1496230784741#c8145790518198229239

---

Major problem with this theory: Even the referenced study dates the massive explosion and diversification of South Arabian languages into the Horn of Africa around 1,000 BC. Coincidentally, this is the same time Lazaridis et al (2016) estimated West Eurasian ancestry to have entered into the Horn.

HOWEVER, this West Eurasian ancestry was almost entirely Natufian-like (and not Iran-N like) in nature. Modern East African South-Semitic speakers like Amhara and Tigrayans have 40-50% Natufian ancestry but basically ZERO (or in any case, <5%) Iran_N or CHG-related ancestry.

So the South Semitic languages seem to have been brought to East Africa by almost unadmixed, entirely Natufian-like people. It seems unlikely to me that more northerly Semitic speakers could have been so radically different...

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

Referring to Basal J as a WHG lineage is totally wrong

The official definition of WHG begins in Europe 14,000 B.C. with the Villabruna cluster which arose from admixture between Pre-Ice Age European Hunter Gatherers belonging to Haplogroups I & C mixing with an incoming population from Eastern Europe carrying R1b and bringing affinity to Near Easterners and East Asians.

Basal J can't even be similar to upper paleolithic Europe as after the I/J split I went on to mix with C1a/C1b individuals living in Europe such as the Vestonice Cluster. WHG is totally divorced from whatever basal relation the IJ lineage had before the split. It could very well be that the original Hap I lineages were more similar to CHG prior to arrival in Europe.

Palacista said...

There is a rough line near the present Turkish border that demarcates Semitic v. non Semitic, this language border is ancient, going back to the Bronze age at least, with records in Eblaite and Akkadian from at last 2500 BCE to the south but with virtually no Semitic intrusion above the line. This line has survived despite language replacement on both sides.
If there had been an expansion of one cultural/linguistic group it would have to have had a starting point either above or below that line.

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

I remember reading a stupid suggestion a long time ago that haplogroup J was connected to the distribution of Turkic languages. If J is connected to Afro Semitic then whose behind Turkic? Probably all that Central Asian R1b, being Aggulinative just like Sumerian, Basque, Uralic, Etruscan, and Turkic.

MaxT said...

@Romulus

Where do you believe IJ split occurred? I don't see basal J being CHG-like but something related to UHG-related most likely.

"It could very well be that the original Hap I lineages were more similar to CHG prior to arrival in Europe."

This sounds unlikely to me, care to explain why hap I lineages would be similar to CHG?

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

@MaxT

I entered Europe with the Gravettian, which from the archeological record entered Europe from the East via the Caucasus. I suppose I could have entered Europe from the Balkans but there are some very basal forms of I in the Caucasus that would lead one to believe I is very old there. The presence of J in CHG there also leads one to beleive that the IJ split occurred in the Caucasus and it was from here that I split off and spread north into Europe.

UHG is something that looks like Kotenski or Oase, who we can seee now belong to K2a so different from wherever IJ was.

Given that WHG is a composite of some UHG like Kotenski + some Eastern R1b group + the I group I'd say it's pretty mixed and distant from the original base I group. CHG on the other hand is CHG + Basual Eurasian, CHG without the basal Eurasian I would assume is pretty isolated and closer to the original IJ group.

Unknown said...


>So the South Semitic languages seem to have been brought to East Africa by almost unadmixed, entirely Natufian-like people. It seems unlikely to me that more northerly Semitic speakers could have been so radically different...

They were as ancient Yemenis introduced those languages to the Horn.
It seem South Arabia was the Natufian refugium. Modern Yemenis are ~26% SSA and they also have CHG, ASI and Iran N admixtures. Yet it seems they were nearly pure Natufian 3000 years ago. Modern Saudis however aren't that much different from modern Yemeni jews.

Cordialmente Adrián said...

The social position of the individuals analyzed should be taken into account, because according to the " Dynastic race Hypothesis " an expedition from Mesopotamia arrived in the Nile valley bordering the Arabian peninsula with boats and formed the elite of the beginnings of the Old Kingdom.

Unknown said...

Romulus UHG seems more like Vestonice16 than Kostenki14 or 12. However Natufians didn't seem to have ASE/Ust'-Ishim-like admixture displayed by UP Europeans and Iran N. Which means it was Vestonice after LGM drift or a Vestonice branch that didn't pick up ASI related admixture unlike Aurigncians/Gravettians.

However what's interesting according to Genetiker's test one Natufian sample did have the purple ASE/ASI component.

Also it looks like not all Neolithic Iranians were australoid admixed as only two Iran EN samples have this purple Papuan component. Same with modern Iranians-not all have ASI.

https://genetiker.files.wordpress.com/2016/08/all-13-14-1.png

BedouinB
"Vestonice16" 53.85
"Basal" 29.55
"AfontovaGora3" 11.3
"Esan_Nigeria" 5.25
"MA1" 0.05
"GoyetQ116-1" 0
"Kostenki14" 0
"ElMiron" 0
"Villabruna" 0

Could someone run a formal stats admixture test like the one above on Natufians? They would probably show over 60% Vestonice.

Samuel Andrews said...

I live relatively close to one of the best ancient Near Eastern museums in the world: Oriental Institute. Coincidentally the last time I visited was one day before all of the Mesolithic and Neolithic Near Eastern genomes were published last year.

The museum displays many ancient Near Eastern cultures in chronological order. It begins with the Natufians. Before any of these aDNA results were published I remember reading that Neolithic cultures in Iran migrated west into Mesoptamia and the Levant. At the time I suspected this migration replaced much EEF-like genes in the western half of the Middle East.

Unknown said...

>At the time I suspected this migration replaced much EEF-like genes in the western half of the Middle East.

More like Levant N than EEF.

Samuel Andrews said...

The field museum holds an Egyptian pyramid. Inside of it are Egyptian artifcats including mummies. One of the mummies is opened. You can see the body wrapped up in cloth. His face is exposed and I took a selfie with him.

He had a huge nose bridge and a thin nose. That's what the stero typical Jewish nose is, I think it could be a legacy of IranNeo.

capra internetensis said...

@Lenny Dykstra

That's assuming the admixture event being detected (South Semitic arrival) is actually the first or most important influx of West Eurasian ancestry.

However, the Natufian-like ancestry could have arrived with pastoralists from Egypt at an earlier period, with the LD signal obscured by later admixture. (Evidence is Cushitic languages and archaeological spread of Sudanese pastoralists into the Horn back in the 3rd M BC.)

Matt said...

Don't disagree with what you mention, but would say although you we do see convergence between ME groups, thinking about it in a way really wide time scales here make it harder to link to a unitary phenomenon (Iran Chl is about 2000 years before Sidon_BA!).

We could also add Armenia_EBA vs Armenia_Chl. Relatively less steppe like, relatively more CHG like (mainly, but also Levant and Iran).

Synome said...

@Lenny Dykstra

You're right. It is a pickle. The authors of Chiaroni 2009 state

"Studies have shown that Ethiosemitic-speaking populations are genetically similar to Cushitic-speaking populations within Eritrea and Ethiopia (Lovell et al 2005). Thus, we propose that the current distribution of Ethiosemitic reflects a process of language diffusion through existing African populations with little gene flow from the Arabian peninsula (i.e. a language shift)."

However, this is contradicted by the paper cited in Lazaridis 2016, which is Pickrell 2014, which finds the first west Eurasian contribution in east Africa to have taken place ~2,700-3,300 Kya during the same period.

Most measures of the west Eurasian ancestry in Horn Africans show that it is dominated by the Natufian related component like you said. However, see this post by Awale about CHG in Horn Africans. It may not be quite so nonexistent after all, especially among Ethiosemitic speakers: http://anthromadness.blogspot.com/2015/12/caucasus-hunter-gatherer-related.html?m=0

For the king said...

There are some methods that show significant Iran Neolithic(Rather than CHG, despite those 2 groups being almost identical) ancestry in Iran ChL. In some other methods, the farmer ancestry in Iran ChL looks mostly Anatolian rather than Levantine noelithic (Again, those groups are pretty much the same). The most likely origin of the Iran ChL population is somewhere around NW Iran or the Armenian homeland.

I1293 Iran Mesolithic J2a-CTS1085
AH2 Early Neolithic 8205–7756 J2b-M12*
I1662 Iran Copper Age J2a-PF5008(xL581)
+ the J1 and J2 Mesolithic Caucasian samples

I dunno why would anyone label a population that was mostly Caucasian + Iran Neolithic admixture and Y-DNA wise " Mesopotamian". Just because of 15-25% Levant neolithic? we should then label Yamnaya as Anatolian because of their 5-15% Anatolian farmer admixture.

Also, the Migration route of J into the proper Middle East doesn't have to be mediated through Mesopotamians. There are many rivers and lakes in Anatolia, that pretty much stretch from the Caucasus/Iran region to the proper Levant

I'm pretty sure more J will pop out in Iranian farmer samples, and more non J y-DNA will pop out in samples from the Bronze age Levant or Egypt.

Anthro Survey said...

@Davidski

Ok, here is that txt format file you requested.

I hope all the names are correct(esp the Roman Brit outlier). Let me know so I can correct any that are not.

https://pastebin.com/eN5BkRXB

Unknown said...

>There are some methods that show significant Iran Neolithic(Rather than CHG, despite those 2 groups being almost identical)

CHG seems to be Iran N-ASI+WHG or EHG. Unsampled intermediate population with higher Basal Eurasian but without WHG and ASI probably existed somewhere in the region at some point.

Anthro Survey said...

@Samuel Andrews:
"He had a huge nose bridge and a thin nose. That's what the stero typical Jewish nose is, I think it could be a legacy of IranNeo."

LOL! I agree. I think the mixing of Iran_Neo lineages with Levant_Neo lineages ultimately accounts for the stereotypical Near Eastern "gas station" look. It's not just in the nose,btw. Had this not taken place, I believe the people of Levant, Anatolia, Messopotamia and Egypt would have resembled Berbers more(well, those who don't show strong ssa traits).

Iran_Neo probably had most of the alleles to make it happen, I'm guessing. Check out the Sikh(who likely has lots of non-steppe Iran_Neo) in picture 3:

https://www.buzzfeed.com/tasneemnashrulla/11-super-stylish-photos-that-prove-sikh-men-rock-the-best-be?utm_term=.atd5B67K1#.ehd436kY9

Average man from Khurasan(probably not much Levant-related ancetry compared to non-steppe Iran/CHG).
https://commons.m.wikimedia.org/wiki/File:Young_Tajikistani_man_with_guitar.jpg#mw-jump-to-license

Unknown said...


>I believe the people of Levant, Anatolia, Mesopotamia and Egypt would have resembled Berbers more(well, those who don't show strong ssa traits).

They would probably look more similar to Yemenite Jews, Saudis and Bedouins without their modern negroid and australoid admixtures than to Berbers.
That hooked nose is certainly from Iran N/CHG admixtures as Natufian-rich populations have long, thin, straight noses. Same with unibrows and hairiness. Bedouins with low CHG/Iran/ANE can barely grow any facial hair. Sargon of Akkad also had that hooked nose.
Berbers in addition to SSA have some WHG and Iranian admixture.

Anonymous said...

@Davidski and others

Do you remember a poster on some preliminary Rakhgarhi results that mentioned descendants of Middle Eastern traders? Might that be part of the same expansion?

It would be interesting to see if you can detect this pulse in Indian populations

Ryan said...

Looking back at the Lazaradis paper a couple things that didn't sink in before...

On the PCA, Iran_CHL is closer to Neolithic Mesopotamia than Iran_Neolithic is. So I don't think Iran_CHL can be part of this same process, as if anything it looks like Iran_CHL is just Iran_Neolithic watered down by some Anatolia Neolithic. I think that makes more sense here too. First Iran becomes more like the Levant/Anatolia, then the Levant/Anatolia becomes more like Iran. Different layers.

Also, never noticed this before, but it shows Armenia_EBA having less steppe ancestry than both Armenia_CHL and Armenia_MBA. So Armenia got some steppe ancestry from Anatolian IE maybe, then got it watered down by some new migrants - Hurrians? - and then gained new steppe ancestry when the actual Armenian speakers arrived?

Gutians I think are an ok candidate for these migrants. That doesn't really help too much since we don't know who the hell the Gutians were except that they weren't closely related to their Semitic, Sumerian, Hurrian and Elamite neighbours - and probably not IE either, but it would be a starting place. The Sumerians describe them as a collections of tribes and clans that came down from the Zagros mountains and overwhelmed the Sumerian and Akkadian city-states. They overwhelmed the Akkadians in 2154 BCE, so I think it's about the right time, and if they managed to defeat an organized regional hegemon, I think it's likely they had a huge impact on less organized pastoralists in the hinterland.

This would be after the break-up of Semitic, but before the break-up of Central Semitic? With the Gutians getting absorbed into Semitic rather than the reverse?



Davidski said...

@Ryan

So I don't think Iran_CHL can be part of this same process, as if anything it looks like Iran_CHL is just Iran_Neolithic watered down by some Anatolia Neolithic.

Anatolia_Neolithic was largely replaced by a more eastern population at about the same time as Iran_Neolithic was largely replaced by a more western population.

So you reckon that instead of these expansions being part of one expansion from somewhere between Anatolia and Iran both to the east and west, it was actually an expansion of Anatolia_Neolithic to the east?

Are you sure, because Iran_ChL doesn't model well as Anatolia_N/Iran_N; it's better modeled as Anatolia_ChL/Iran_N, and of course Anatolia_ChL was largely a new population in Anatolia that arrived from the east post-Neolithic.

Rob said...

@ Davidski

"Are you sure, because Iran_ChL doesn't model well as Anatolia_N/Iran_N; it's better modeled as Anatolia_ChL/Iran_N, and of course Anatolia_ChL was largely a new population in Anatolia that arrived from the east post-Neolithic"

But that's a problematic "better model". Iran Chalcolithic should not be modelled on Anatolia Chalcolithic, as they are contemporary.
Indeed, the changes toward Iran Chalcolithic are already evident in the Iran _lateNeolithic individual from Lazaridis (c. 5000BC). On the other hand, we have c., 5000BC individuals from Anatolia which still look similar to earlier individuals.

I therefore think that we are looking at different events. The Iran Chalcolithic IMO takes some Levant_Neol. too, and probably represents a gradual miscegenation process associated with the "pulling in' of populations toward Mesopotamia.

On the other hand, in Anatolia we are seeing a frank population turnover occurring between 4000 - 3000 BC

Ryan said...

@David - "So you reckon that instead of these expansions being part of one expansion from somewhere between Anatolia and Iran both to the east and west, it was actually an expansion of Anatolia_Neolithic to the east?"

I'm reckoning there were different waves in different directions. So Anatolia to the East and then Iran to the West or something.

Take a look at the PCA.

https://drive.google.com/file/d/0B9o3EYTdM8lQSnhVZlEwUktfQUk/view

Chalcolithic Anatolia and Chalcolithic Iran are midway between Neolithic Anatolia and Neolithic Iran. I don't see how 1-way mixture could account for that.

Davidski said...

@Rob

Iran Chalcolithic should not be modelled on Anatolia Chalcolithic, as they are contemporary.

I'm only modeling Iran_ChL as part Anatolia_ChL to show that they share recent ancestry, probably from between Anatolia and Iran, which would mean Mesopotamia.

@Ryan

I'm reckoning there were different waves in different directions. So Anatolia to the East and then Iran to the West or something.

I don't find this "highway" model very convincing.

Why not an expansion both to Anatolia and Iran from Mesopotamia of a population similar to Iran_ChL except more westerly?

a said...

@Mister Bobby,

Whatever became of all your teal scenario's.

Mister Sammy,
"He had a huge nose bridge and a thin nose. That's what the stero typical Jewish nose is, I think it could be a legacy of IranNeo."


Mesolithic profile of wooden carving from Russia;aka "Shigir" idol.


http://siberiantimes.com/PICTURES/SCIENCE/Big-Shigir-Idol/inside%20profile.jpg

" The Shigir Idol (Russian: Шигирский идол), is the most ancient wooden sculpture in the world,[1][2] made during the Mesolithic period and was carved around 11,000 years ago"

Rob said...

@ "a"

Are you asking me ?

Matt said...

Yes, the simple position on Lazaridis PCA probably would fit Iran_N+Anatolia_N much better in PC1 and PC2 than the mix they actually give - http://i.imgur.com/jxXZiLu.png

But there may be reasons for that. These seem like somewhat imperfect models.

(Although by the same token Anatolia_Chal probably would fit better as Anatolia_N + CHG / Anatolia_N + Armenia_Chal than Anatolia_N+Iran_N)

Also, re: the topic of whether Mesopotamia was more of a sink or source throughout this period, really very hard to have enough data to check...

Unknown said...

This is most likely a silly, lurker question, but how long does it take a regional accent become a dialect and then a semi- distinct language outsiders can understand with difficulty, and then finally a "What did he say?" sort of thing? How much distance is required between the variants to keep them evolving/changing separately?

I am visualizing this simplistically as collection of expanding "bubble-fronts" (with the maximum variation at the leading edges) linked by pidgins.

Thanks! You folks are fascinating to read...

Anthro Survey said...

Wait, can Chalcolithic Anatolia be modeled as Neo Anatolia+Neo Levant+Iran/CHG? Or formal stats only accord w/Neolithic Anatolian DNA w/out Levant?

By extension, does Chalcolithic Iran just have Anatolia or also Levant?

Seinundzeit said...

Sam,

Lol bro, I don't think we can make such assumptions, or try to postulate such casual connections.

For what it's worth, I've read old anthropological analyses concerning the Natufians, and also a few analyses on ancient remains found in the Hotu and Belt caves (although, I have yet to find anything on the ancient Zagros pastoralists).

Based on those analyses, I really wouldn't be willing to jump to such conclusions, even though they are tempting.

For example, Yemeni Bedouins (Bedouin_B) have the most Natufian-related ancestry among all living populations. Like all Bedouins, these people often have "chiseled" facial features, and often have narrow/prominent noses.

Yet, that isn't how most of the Natufians looked. Instead, most had broad and blunt noses. Some even had prognathous jaws.

Many scholars noted that quite a few of these individuals had a "African" vibe to their cranio-facial features.

Honestly, the only physical characteristics they shared with modern people rich in their sort of ancestry would be extreme gracility, elongated limbs, and small hands/feet relative to body-size.

In addition, the Iran_Hotu people looked very distinct from the stereotype many have concerning contemporary Iranian plateau populations.

In fact, one scholar noted that these people were very "Cro-Magnon"-like.

Contrary to the narrow jaw + hooked nose stereotype, one specimen from these caves had a very concave nose, very broad/deep jaws, a broad face, and heavy brow-ridges + deep-set eyes.

Interestingly, unlike Natufians, they were very robust people.

So personally, I really shy away from even attempting to correlate deep streams of genetic ancestry with physical variation across contemporary populations, no matter how great the temptation.

Seinundzeit said...

That being said, it is interesting to compare the most Natufian/Levant_Neo-rich living people with the most Iran_Hotu/Iran_Neo/Iran_Chal-rich living people.

Yemeni Bedouin (Mostly Jordan_EBA-related, who were mostly Levant_N)

https://assets.vice.com/content-images/contentimage/132458/DSC06768.jpg

https://thumbs.dreamstime.com/z/yemeni-team-group-bedouins-military-archaeological-site-baraqish-yemen-38142691.jpg

https://assets.vice.com/content-images/contentimage/132460/DSC06797.jpg


Baloch/Brahui (Mostly Iran_Neo and Iran_Chalc, with some moderate steppe admixture)

http://c8.alamy.com/comp/FNRDFA/pakistani-army-chief-general-ashfaq-kayani-l-shakes-hands-with-baloch-FNRDFA.jpg

http://3.bp.blogspot.com/-88CzRgzeL8E/Ufi434Gg7FI/AAAAAAAAHlo/InJl_f2us8M/s1600/salman+rashid.jpg

https://newint.org/features/web-exclusive/2016/03/24/24.03.16-balochi-590.jpg

I guess the Iran_Neo/Iran_Chalc-rich population is more facially robust, and perhaps physically "larger", than the Levant_Neo-rich population, despite living in a similar environment, despite living in similar economic conditions, and despite having a similar lifestyle.

So, perhaps that was the primary differentiating factor?

Who knows.

Anonymous said...

Better sampling would be Sardinians and Georgians. I don't know why you chose those two heavily mixed populations.
Also, you simply cannot infer similarity, as you don't know how much drastic their appearece changed due to admixture, you don't know if the admixture became phenotypically dominant and you don't know that in those thousand of years natural and sexual selections changed their appearence too.
For instance, the African admixture of the Bedouins and the Dravidian of the Balochis made them much more dark skinned.

Arza said...

@ Chalcolithic Iran

I did a little exercise. I've modelled Iran Chalcolithic samples in this way:

https://drive.google.com/file/d/0BzuotSparaF-QndjNUFCdFgtSkk/view

Then using nMonte output I've recreated the mix without Iran Neolithic for each sample (coordinates for DoHA are in the file above). This is how the ghosts plot:

https://4.bp.blogspot.com/-CoB-IWQaNJo/WS-C9j_C99I/AAAAAAAAAMo/pcuzIn9T-DIX7pSTfm0tMu0ECY85xrENwCLcB/s1600/IranChalcolithic.png

Additionally at the end of the file are Iran Late Neolithic modelled as ghost + Iran Neolithic and Armenia EBA samples modelled as ghost (with and without Anatolia Chalcolithic) + CHG.

For the king said...

@Sein

The BedouinB samples are from the Negev desert in Israel.

Mark Moore (Moderator) said...

So who were they? I am struck how the time (taking the oldest way to compute the era) and the place and the questions about language has connotations like those described in Genesis 10 and 11, at least if you assume the flood was local and Noah's descendants came out of the mountains and "found a plain in the land of Shinar." Maybe a language hotspot is a dumb idea, but the tower of Babel account is basically about just that. They came down out of the mountains and brought a lot to the table civilization-wise, attracting large households of locals to themselves which became the core group of various nations once they scattered. This would be an awe-inspiring connection that would give data a lot like you show here.

Anthro Survey said...

Look, folks, I don't want to be Captain Obvious, but Neolithic, Chalcolithic and BA genomes from Messopotamia are the elephant in the room here and we've yet to analyze them. We have a pretty good idea of how the cake is gonna bake, but the findings will end a lot of debates and speculations.

To name a couple:
At what point did Messopotamia start to feature CHG/Iran DNA? Or perhaps this package was there "first" and Levant-like stuff came later?

Krefter said...

First look at ancient Egyptian mtDNA

http://mtdnaatlas.blogspot.com/2017/06/first-look-at-ancient-egyptian-mtdna.html

R0a, HV1, T1a, J2a2, N1, M1a might all ultimately derive from Natufian-like people. N1, T1a, J2a link Neolithic Anatolians with Natufian-rich people.

Check this out!!

Andronovo, Bronze age British, and these ancient Egyptians were roughly contemporary to each other. Each carries roughly the same frequency of T1a and I.

That demonstrates the widespread distribution of Neolithic West Asian ancestry.

Gioiello said...

@ Krefter (or everyone you are)
I wrote that to a compatriot of yours on YFull page of FB

Of course only the aDNA will decide, but are you able to read the mt haplogroups? From what I see in your spreadsheet, the Italian samples are the oldest and the Middle Easterner ones derived in G709A. Look at the Ian Logan spreadsheet and perhaps you'll understand where hg HV and subclades derived from. If you are a rich American, put 100000 dollars on the plate and bet with me.

Matt said...

@Arza, how do Jordan_EBA sample(s) model in the same analysis?

Anonymous said...

@David

"- Anatolia_BA, Western Turkey, 2836-1800 calBCE (Lazaridis et al. 2017)"

What paper was that?

Davidski said...

It hasn't been published yet, but will be very soon.

Samuel Andrews said...

@Gioiello,

HV could have originated in many different places. But when discussing the mtDNA affinity of a population there's no reason to discuss where a haplogroup originated 40,000 years ago. People liked to discuss the distant origins of haplogroups 5 years ago but nowadays all people don't.

HV1, the type of HV many ancient Egyptians had, today peaks in the Near East(inclu. Egypt). There's no point to discussing where HV originated.

For example, when discussing Y DNA R1b in modern Mexicans there's no point to mentioned R1b1a probably originated in Mesolithic Europe. What matters is their R1b is from Spain.

R1b expanded from Spain to Mexico starting 500 years ago, R1b L151 expanded from the Steppe to Western Europe starting 4,700 years ago.

Gioiello said...

@ Samuel Andrews who answers for Krefter and Onur Dincer says you aren't Sam...

Of course I largely agree with you about that, but what I am fighting against from ten years and more is that your friends wanted to impose us that everything came from Middle East, and that isn't true.
1) Not only I said that R1b1 came from the "Italian Refugium", at least since 14000 years ago (Villabruna), but we'll see how much before or not as many write here.
2) WHG hadn't only mt hgs. U4 and U5, but many others. The last paper of Mathieson says also mt K1, as we are saying from so long, thus your K1a1b1a and my K1a1b1e.
3) I wrote a lot about all the subclades from R0a'b to HV to H. The refugium was very likely in Southern Europe (Sicily and nearby) and so far there isn't only my analyses but also a peer reviewd paper.
4) Of course you know that I don't agree with you and the kurganists about R-L51 and subclades, and also about part of the R-L23, comprised my R-L23-Z2110. We are studying that. We'll see. Let's wait that aDNA from Tyrrhenian Italy is tested and we'll speak again then.

Unknown said...

Genetiker posted the Y SNPs for the 3 Egyptians:

Sample Period Date BC Haplogroup
JK2134 Pre-Ptolemaic 776–569 J1a2a2-Z2329
JK2911 Pre-Ptolemaic 769–560 J2b1-PF7314
JK2888 Ptolemaic 97–2 E1b1b1a1b2-V22

Unknown said...

The JK2134 J1a2a2 is at a level equivalent to YSC234.
THe JK2911 J2b1 is M205.

Very similar to the EBA Jordanian samples and the MBA Sidonian samples!

Unknown said...


>1) Not only I said that R1b1 came from the "Italian Refugium", at least since 14000 years ago (Villabruna), but we'll see how much before or not as many write here.

R1b1 may have come from Siberia as Villabruna is known to have ANE or some kind of East Asian admixture most similar to Ami. Anyway Mesolithic R1 clades are extinct. All clades in modern Europe are from the Bronze Age steppe.

Gioiello said...

@ Unknown
"R1b1 may have come from Siberia as Villabruna is known to have ANE or some kind of East Asian admixture most similar to Ami. Anyway Mesolithic R1 clades are extinct. All clades in modern Europe are from the Bronze Age steppe".

All extinct except the one R-P297* who generated all the descendants.
That R-V88 was in Italy, migrated to Iberia, the sample from 7100 years ago, and it is dominant in its oldest subclades in Italy is largely demonstrated. The rest will come.

Unknown said...

I'd say they spoke Alarodic (this includes Nakh-Daghestani, Hurrian, and urartian languages, and all of these speakers are high in the Iranian component and in J).
These people may have been what some people called the Armenoid / Iranian Afghan Race.

Unknown said...

Maybe Semitic is a mixture of Alarodic and Afro Asiatic.

Ric Hern said...

What happened to the Maykop Culture or descendants of Maykop ? Were they pushed South into Anatolia or Armenia via the Western Caucasus ?

Arza said...

@ Matt

nMonte & coordinates:
https://drive.google.com/file/d/0BzuotSparaF-LTg4Qzd6cDJGZU0/view

PCA123
https://2.bp.blogspot.com/-nDZtJYhID2E/WTAtkt8AQYI/AAAAAAAAAM8/BiUncz-C4zshhEfc7DWhBudq-aIExDfFwCLcB/s1600/JordanEBA_PC123.png

PCA126
https://1.bp.blogspot.com/-NO7jb6WCzvk/WTAtkk2QdyI/AAAAAAAAAM4/YAEZvJx8KhoTcjtVuh5ijKFN_sfpEA73gCLcB/s1600/JordanEBA_PC126.png

hollow - ghost
hollow circles - previous Iran Chalcolithic ghosts
hollow squares:
exNAT - Natufians excluded
exIRAN - Iran Neolithic excluded
smaller hollow squares:
exNatIran - Natufians & Iran Neolithic excluded

PC123 C - two samples on both sides of Jordan_EBA are Tep003 and Tep006.

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

JK2911 Pre-Ptolemaic 769–560 J2b1-PF7314

This is the same Y DNA Haplogroup that was found in the Middle Eastern Gladiator from Roman Era Briton. I personally believe this guy was a Phoenician and possibly this underscores a connection between Pre-Ptolemaic Egypt and the Phoenicians.

Gioiello said...

@ Unknown
Read www.eng.molgen.org sometime

Re: Z156*
Posted: Thu Jun 01, 2017 4:01 pm
by Gioiello
dartraighe wrote:
"1. Several Y‑chromosome re-sequencing studies have concurred in finding bursts of expansion within specific lineages within the past few thousand years."

"2. The expansion in western Europe of lineages within haplogroup R1b‑L11 ~4.8–5.9 kya, which was possibly associated with technological advances in the Bronze Age."


In fact they put the sepration from A00 at 275000 years ago (as also Poznik did) thus the YFull dates should be multiplied for an 1.17 factor. But Shi Huang and colleagues put that at more than 300000 years ago, thus it should be multiplied for an 1.26 factor. Anywway we are right and all the kurganists wrong.

Samuel Andrews said...

@Gioiello,

The only other username I use is Krefter. I am not Onur Dicer.

"Of course I largely agree with you about that, but what I am fighting against from ten years and more is that your friends wanted to impose us that everything came from Middle East, and that isn't true."

I'm not trying to impose anything on anyone. Ancient DNA papers speak volumes with their data. All you need to do is read these papers.

10 years ago people cared about mtDNA/Y DNA. Nowadays people don't really care. What people care about is autosomal DNA because autosomal DNA reveals where all of one's ancestry is from not just one's maternal and paternal lineage.

Autosomal DNA from ancient Europeans shows that people from the region we today call the Middle East migrated en masse into Europe in at leats three separate waves one occuring 8,000 years ago, one 6,000 years ago, and one maybe as recent as 4,000-2,000 years ago.

-The first wave brought Anatolian farmers. They're the most important ancestors of modern Europeans. Lithuanians have about 30%, other Northern Europeans 40%, Iberia and Northern Italy 50%, most of Southern Europe maybe 40-50%.

-The second wave brought CHG to Russia/Steppe. "Steppe" people were 50% or so Near Eastern(most of it CHG). Later CHG was brought from the Steppe to the rest of Europe. All Europeans, except Sardinians, are at least 15% CHG.

-The third wave brought a mix of Anatolian farmer, Natufian, and CHG/IranNeo. This wave mostly only affected Southern Europe. It lightly affected France, Western Germany, and England. Most of Southern Italian's ancestors arrived in this migration.

Samuel Andrews said...

@Gioiello,

None of what I just posted is opinion. It's all fact.

European's mother(mtDNA)/father(Y DNA) markers show the same pattern as autosomal DNA. Their Y DNA is dominated by relatively recent founder effects; R1b L151, R1a Z282, I1, I2a-DIN, N1c.

All of those Y DNA founder effects, except for N1c, ultimately trace back to Mesolithic Europeans. And yes R1b L151 and R1a Z282 are from the Steppe not Italy.

European mtDNA isn't as affected by founder effects. Only roughly 10-20% of European mtDNA belongs to U5, U4, and U2e. Besides the Iron Gates HGs, 95% of over 100 mtDNA samples from Mesolithic Europeans belong to U5 or U4 or U2e. That probably means roughly 10-20% of European maternal lineages trace back to Mesolithic Europeans. The other 80-90% of European mtDNA was in the Middle East before 8,000 years ago.

Yes, the Iron Gates HGs had a lot of mtDNA K1 but so did Anatolian farmers. Anatolian farmers had a lot of WHG-like ancestry, K1 is apart of haplogroup U, and so it is no surprise that K1 ultimatly traces back to WHG-like people. But there's a very low chance that haplogroups like JT and X and W and N1(inlu. I) orignated in WHG-like people. R0 might have but there are only two examples of R0 in Mesolithic Europe. R0 in modern Europe, including all H, ultimatly comes from the Middle East.

AlexDeLarge said...

@Samuel Andrews

Nay, even your Polish boss agrees with me that South Eastern Europeans are the result of mixing between Bronze Age Central Europeans and Copper Age Anatolians.

https://j2-m172.info/wp-content/uploads/sites/3/2016/06/PCAtest2_Eurogenes_2016-06-23_detail-Levant.png

Maciamo's opinion:

"It is very likely that J2a, J1 and G2a were the three dominant male lineages the Early Bronze Age Kura-Araxes culture (3,400-2,000 BCE), which expanded from the South Caucasus to eastern Anatolia, northern Mesopotamia and the western Iran. From then on, J2 men would have definitely have represented a sizeable portion of the population of Bronze and Iron Age civilizations such as the Hurrians, the Assyrians or the Hittites. It is very possible that bronze technology spread from the South Caucasus across the Iranian plateau until the Indus Valley, giving rise to the Harappan Civilisation (see below).

The Minoan civilisation emerged from 2,700 BCE and could have been founded by colonists from the Kura-Araxes culture who would have brought bronze working with them. Modern Cretans have the highest percentage of G2a (11%), J1 (8.5%) and J2a (32%) in Greece (and the highest percentage of J1 and J2a in all Europe for that matter), the three haplogroups associated with the Kura-Araxes culture. Although little data is available at present about deep clades in Crete or Aegean Greece, the parts of Italy that were colonised by Ionic and Doric Greeks, notably Sicily, Calabria and Basilicata, possess substantial percentages of typically Caucasian haplogroups, such as G2a-L297, J1-Z1828 and J2a-L581, as well as considerable levels of Middle Eastern and Caucasian autosomal admixture by European standards. In fact, it seems that many branches of J2a (e.g. M319, Z7671, F3133, Z6046, L581) may have expanded from the South Caucasus from the Chalcolithic onwards. The presence of these haplogroups and admixtures in southern Italy almost certainly represent Kura-Araxes ancestry inherited from Minoan Greeks from the Aegean islands."

http://www.eupedia.com/europe/Haplogroup_J2_Y-DNA.shtml#J2a_Mediterranean

Ryan said...

@David - Why not an expansion both to Anatolia and Iran from Mesopotamia of a population similar to Iran_ChL except more westerly?

Agreed that it's consistent with the data too, but wouldn't it require huge population turnover? And how did that original mixed population form? The boundaries between populations in the Neolithic seemed to be rather sharp rather than soft.

Remind me though - do Dravidians model better as part Iran_Chl or part Iran Neolithic? Though I guess we don't precisely know the timing of Dravidian's expansion either.

Ryan said...

@italicroots - G2a was widespread in Europe long before the Kura-Araxes culture formed. The Iberian neolithic and LBK both had G2a at least 6,000 years ago.

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

We already have y DNA from Kura Araxes and it's R1b

AlexDeLarge said...

One single sample had R1b. Almost half of modern Armenians have R1b as well.

Arza said...

@ Matt

My amateurish interpretation:
https://1.bp.blogspot.com/-ZWpu5YkUjXM/WTBbzaLRFGI/AAAAAAAAANM/Xpde5M2C894MoWyib1ZN2IWpXawxe79PgCLcB/s1600/IranChalcoJordanEBA.png

Probably we are lacking some Anatolian population in the place where three clines meet (orange circle).

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

1/1 samples. 100% 😂

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

Anybody who quotes eupedia is a total moron as they are obviously unable to differentiate some deluded Belgian's ancient DNA fan fiction from reality.

Chad said...

An irrelevant L389.

Gioiello said...

@ Samuel Andrews

"@Gioiello,
The only other username I use is Krefter. I am not Onur Dicer".

I didn't say that you were Onur Dincer, I know him very well. Was he who said that you weren't Sam [Huelsenrath].
Of course I don't agree with your concept of Middle East: There were at least three very different populations: Northern Anatolia linked to Europe, Iran, and Natufians. It is to-day clear that Semites, who spoke a Natufian language of the Afro-Asiatic stock, received their hg. J from Iran or Caucasus, not being hg. J amongst Natufians.
Al these people received introgressions from the Italian Refugia: the Alpine one with surely R1b1 (I think also R1a and J, we'll see) and from the Southern one with all the mt from R0, HV, H etc.
That they test Italy and we'll see next.
About the R1b1-L389 in the Caucasus very recent the highest ignorance: Caucasus has only one haplotype with YCAII=23-23, and Italy all the 4: 18-22, 18-23, 22-23, 23-23, and 23-23 is due to a RecLOH from 18-23.

Unknown said...


>For example, Yemeni Bedouins (Bedouin_B) have the most Natufian-related ancestry among all living populations. Like all Bedouins, these people often have "chiseled" facial features, and often have narrow/prominent noses.

Yet, that isn't how most of the Natufians looked. Instead, most had broad and blunt noses. Some even had prognathous jaws.

Many scholars noted that quite a few of these individuals had a "African" vibe to their cranio-facial features.

Honestly, the only physical characteristics they shared with modern people rich in their sort of ancestry would be extreme gracility, elongated limbs, and small hands/feet relative to body-size.

In addition, the Iran_Hotu people looked very distinct from the stereotype many have concerning contemporary Iranian plateau populations.

In fact, one scholar noted that these people were very "Cro-Magnon"-like.

Contrary to the narrow jaw + hooked nose stereotype, one specimen from these caves had a very concave nose, very broad/deep jaws, a broad face, and heavy brow-ridges + deep-set eyes.

Interestingly, unlike Natufians, they were very robust people.

So personally, I really shy away from even attempting to correlate deep streams of genetic ancestry with physical variation across contemporary populations, no matter how great the temptation.


Well, Natufians probably weren't 100% uniform. Some may have had negroid and/or australoid admixtures whereas others were just west_basal+eurasian. Geniteker detected purple ASE component in one Natufian sample for example.

https://genetiker.files.wordpress.com/2016/08/all-13-14-1.png

However while one Natufian skull from Jericho looked weird the clay masks they put on skulls look quite bedouin-like dolichocephalic with long noses.

http://img.4plebs.org/boards/pol/image/1496/34/1496340863474.jpg

http://img.4plebs.org/boards/pol/image/1496/34/1496340984293.jpg

Are you sure BedouinB is from Yemen and not Saudi Arabia? According to Genetiker and I think that Ancient Egyptian study BedouinBs don't have any negroid admixture unlike BedouinAs from Levant. Yemeni Bedouins lacking SSA admixture would be very strange as even Yemeni Jews have around 2%-5% of it.

MaxT said...

Whats with all the stupid outdated typologies and oids'? These don't mean much when it comes to genetic admixtures. Geneticker blog isn't the best source of information.

Unknown said...


>The first wave brought Anatolian farmers. They're the most important ancestors of modern Europeans. Lithuanians have about 30%, other Northern Europeans 40%, Iberia and Northern Italy 50%, most of Southern Europe maybe 40-50%.

This seems incorrect. Lithuanians have around 30% CHG not EEF. EEF admixture is actually very low in modern North Europe and especially the Baltic region and Russia. I remember seeing a test on some forum that showed Poles from some particular region of Poland have only have ~5% EEF but ~30% CHG. Steppe CHG admixtures drastically decreased EEF ancestry everywhere in Europe except for Sardinia and maybe Iberia and Sicilia who got additional EEF+Levant N+SSA from Moors/North Africans and Arabs.

More Natufian skull masks

http://img.4plebs.org/boards/pol/image/1496/34/1496340955575.jpg

http://img.4plebs.org/boards/pol/image/1496/34/1496341008069.jpg

http://img.4plebs.org/boards/pol/image/1496/34/1496341033853.png

Unknown said...


>Whats with all the stupid outdated typologies and oids'? These don't mean much when it comes to genetic admixtures. Geneticker blog isn't the best source of information.

In my opinion known genetic components correlate surprasingly well to old racial taxonomy.

Unknown said...



>In fact, one scholar noted that these people were very "Cro-Magnon"-like.
Contrary to the narrow jaw + hooked nose stereotype, one specimen from these caves had a very concave nose, very broad/deep jaws, a broad face, and heavy brow-ridges + deep-set eyes.
Interestingly, unlike Natufians, they were very robust people.


Some modern Chechens with high CHG also have such robust features. Anyone tested if they have Iran Neo admixture?

MaxT said...

@Unknown
All those so-called scholars who were suggesting Natufians had African-like skull and therefore probably had African admixture turned out to be bullshit. Ancient DNA of Natufians pretty much shows they were Near Easterners without any SSA admixture.

Anthro Survey said...

@Seinundzeit:

You're absolutely right that we should refrain from rigidly ascribing a certain set of facial traits as being highly representative of ancestral components. Nevertheless, we can make some parsimonious guesses as to which components certain trait changes are more associated with.

The thing is, the "711" look spectrum Sam and many others have in mind does seem to be associated with the spread of IranChalco-related ancestry. Why do we often see recurring phenotypic themes from Calabria to Armenia to Herat and Bukara* these days that are largely rare in Morocco, Iberia, non-Balkan Europe and other regions not affected by IranChalco ancestry?

We have to keep in mind that those Iran_Neo samples you mentioned significantly predate Iran_Chalco phenomena and even the Neolithization of India. Such temporal separation kinda precludes them from being good proxies. Moreover, a cranial structure gives us rather limited info about pilosity and finer elements facial structure. We have to also remember that even back then, CHG/IranNeo would have had subpopulations with their own unique allele frequiencies---no reason to think they weren't phenotypically diverse.

*let me clarify smth: although Punjab and outer Khorasan(samarkand, bukhara) probably doesn't have much IranChalco ancestry per se(mainly just CHG/IranNeo) it's likely that those "strains" of IranNeo were temporally and cladistically close to those from which IranChalco package formed.

Chad said...

I have Copts. Someone needing some stats?

Samuel Andrews said...

@Unknown,
"This seems incorrect. Lithuanians have around 30% CHG not EEF. EEF admixture is actually very low in modern North Europe and especially the Baltic region and Russia"

EEF levels are lower in Northern than in Southern Europe but they're not low anywhere in Europe.

Check out pg. 85 in the supplemantary info of the Bell Beaker behemoth...

http://biorxiv.org/content/biorxiv/suppl/2017/05/09/135962.DC1/135962-1.pdf

Most Northern Beaker folk get at least 30% Anatolia Neolithic. The ones from Germany get 40%. That's about what modern Northern Europeans would get if put through the same test.

The Baltic stats definitely have significantly less Anatolia Neolithic, maybe 25-30%. But 25-30% is still a lot.

Seinundzeit said...

JohnP,

Sorry for the wait, finally have some downtime.

You've actually switched things around, as Sardinians and Georgians are extremely distant from Levant_Neo and Iran_Neo, and are also exceedingly mixed in relation to them.

By contrast, Bedouins and Baloch have huge amounts of direct ancestry from ancient populations which were closely related to Levant_Neo and Iran_Neo.

Also, the Baloch are like 5%-10% ASI (assuming ASI is ENA), so surely they have no phenotypic imprint from ASI.

Furthermore, the Bedouin_B are often 0% African in different analyses, and I've been told by multiple academic researchers that the Bedouin_B samples hail from Yemen (although, one did tell me that the Negev was also a possibility).

In addition, we already know that these populations were very dark; we have their genomes.

If I'm not mistaken, Natufians had ancestral alleles at all sites implicated in lighter West Eurasian skin pigmentation. So, essentially, they were "black"!

(I'm using that sort of language for shock-value. But even we were to be more cautious, there is no doubt that the Natufians were very dark)

Finally, the Iran_Neo samples were actually similar to most North Indians and some South Indians, when it comes to pigmentation genetics. So again, probably even darker than the Baloch, who are actually quite pale (the individuals in the photos have lived all their lives under constant, and very intense, sunlight; they have been tanned to the fullest maximum.

I know educated, urban Baloch with indoor jobs, and they are pretty similar to Iranians and Armenians, in terms of skin pigmentation)

Sidenote: I hope you realize the fact that Iran_Neo admixture is probably equivalent to "Dravidian admixture", since the spread of the Elamo-Dravidian language family was likely to involve the diffusion of Iran_Neo-related peoples, and Iran_Neo is the best proxy for the West Eurasian ancestry seen in South India.

Also, Georgians have CHG, which is easily distinguishable from Iran_Neo.

Anyway, Georgians are actually more Anatolia_Neolithic than they are CHG! Not to mention their steppe admixture.

Sardinians are Anatolia_Neo + WHG, with perhaps minor CHG and steppe admixture.

And Anatolia_Neo was already WHG-shifted and Iranian-shifted in relation to Levant_Neo.

So, again, you kinda switched things up.

Anthro Survey said...

Wow, thanks for the stats, Dave. Appreciate it. Really confirmed my prior suspicions, but it seems to have been more of a continual ebb across the board than a spike at any particular stretch of time.

The couple of places where it isn't negative for Mota can mean any number of things----that it was a drifted population, genetic dead end, featured a diff population micro-structure than SSA source populations, etc.

Anthro Survey said...

@Chad:

Yeah, can you please run some Copt stats? I can write them up and post txt file link here for you. Would be great.

Seinundzeit said...

Anthro Survey,

"You're absolutely right that we should refrain from rigidly ascribing a certain set of facial traits as being highly representative of ancestral components."

I certainly agree with that.

"The thing is, the "711" look spectrum Sam and many others have in mind does seem to be associated with the spread of IranChalco-related ancestry. Why do we often see recurring phenotypic themes from Calabria to Armenia to Herat and Bukara* these days that are largely rare in Morocco, Iberia, non-Balkan Europe and other regions not affected by IranChalco ancestry?"

I think it's probably somewhat more complicated than that; natural selection, drift, phenotypic plasticity, changes associated with the adoption of agriculture and pastoralism, etc, all should be taken into consideration.

Once we know more about the genetics and epigenetics underlying phenotype, we'll be on much better footing.

Anthro Survey said...

Seinindzeit---

I wanna hear your take on something I'm considering regarding Dravidians but first---

Of course. I agree it's more nuanced.
I wish I was a plastic surgeon to be able to explain precisely what I mean but there is definitely a strong recurring pattern of phenotypes stretching from Calabria to the Punjab that is somewhat weaker in the Levant and absent in north africa. Not without its bifurcations, of course. Take Riz Ahmed. Man, if I had a dollar for every Armenian, Caucausus person, Messopotamian, Iranian or Turk I saw resembling him I'd be a millionaire.
John Turturro is another great example of someone with traits in that area, albeit he is found more in its western sectors----in the true Iran_Chalc proper areas. He's southern Italian and a common one at that.

Anyway regarding Dravidians--- I'm also of the opinion that Iran_Neo ancestry in India corresponds to the spread of the language family there. Evidence is too damning. lol The Brahui were always a breadcrump in that trail.

One of my Tamil friends once trolled me by saying that Cretans were Dravidian. lol Now, it's hillarious but at the same time raises a serious question.

Do you believe that an Alarodo-Dravidian super family is a legit concept? This may be something to ask Agamemnon about.

Unknown said...

http://www.razib.com/wordpress/?cat=2670

Population movement within and at the destruction of a pre-literate Uruk hegemony/empire?

Unknown said...


>Most Northern Beaker folk get at least 30% Anatolia Neolithic. The ones from Germany get 40%. That's about what modern Northern Europeans would get if put through the same test.

No they wouldn't as modern Europeans aren't Bell Beakers. Gedmatch calculators I've seen showed very low EEF in Baltic countries in contrast to much higher CHG. This makes perfect sense as HGs in Baltic started farming without farmer admixture and later mixed with Steppe invaders without intial EEF admixture other Europeans had.


>By contrast, Bedouins and Baloch have huge amounts of direct ancestry from ancient populations which were closely related to Levant_Neo and Iran_Neo.
Also, the Baloch are like 5%-10% ASI (assuming ASI is ENA), so surely they have no phenotypic imprint from ASI.

Baloch are more like 10-15% ASI and Iran N like 10-11% ASI so yeah it wouldn't influence the phenotype much since both had slightly ASI influenced phenotype.

>Furthermore, the Bedouin_B are often 0% African in different analyses, and I've been told by multiple academic researchers that the Bedouin_B samples hail from Yemen (although, one did tell me that the Negev was also a possibility).

If they are from Yemen then how is it possible that they have 0% SSA when even Yemenite Jews are slightly SSA admixed and regular Yemenis over 20%. Do they have endogamous culture?

>In addition, we already know that these populations were very dark; we have their genomes.
If I'm not mistaken, Natufians had ancestral alleles at all sites implicated in lighter West Eurasian skin pigmentation. So, essentially, they were "black"!

Just like WHG but I think many unsampled Natufians must have had derived light skin alleles because many Natufian admixed Horners have quite light skin.

>Finally, the Iran_Neo samples were actually similar to most North Indians and some South Indians, when it comes to pigmentation genetics. So again, probably even darker than the Baloch, who are actually quite pale (the individuals in the photos have lived all their lives under constant, and very intense, sunlight; they have been tanned to the fullest maximum.

Iran N samples had derived depigmentation alleles and one sample(probably WC1) had blue eyes allele despite lacking WHG admixture which means Neolithic Iranians were lighter than other contemporary populations. They may have been slightly lighter than Baloch.

a said...

@Anthro Survey said...
"
Of course. I agree it's more nuanced.
I wish I was a plastic surgeon to be able to explain precisely what I mean but there is definitely a strong recurring pattern of phenotypes stretching from Calabria to the Punjab that is somewhat weaker in the Levant and absent in north africa."
Good points of interest, keeping in mind that nasal phenotype can drastically change within one generation, depending one the couples features. I've also wondered about this, defining Caucasus/Hittite feature. Pashtuns and Brahmins, Caucasians, Iranians, physical anthropology of the nose is a defining feature.
When one compares Shigir Idol with the groups above; does it infer an evolutionary advantage to an environment that was at one time very cold/glacial border[ dry air ]+ extremely dusty ? For example comparing [endangered saiga antelope ] + Neaderthal's natural Northern Hemisphere range. Natufian/Iranian samples virtually 0% Neaderthal admix? The explanation of humid/warm climate in Neaderthal/human hybrid development does not quite add up. To compare;

https://www.scientificamerican.com/article/the-shape-of-a-nose/

https://www.newscientist.com/article/dn15042-why-did-neanderthals-have-such-big-noses/

Tesmos said...

''
R0a, HV1, T1a, J2a2, N1, M1a might all ultimately derive from Natufian-like people. N1, T1a, J2a link Neolithic Anatolians with Natufian-rich people.
''

No, HV1 is too old to be ''Natufian or Natufian-like''. It's possible some subclades of HV1 derive from Natufian-like people but that's based on speculations.

Samuel Andrews said...

@Unknown,

Formal stats like qpADM and D-stats give Northern Europeans 40% Anatolia Neolithic on average. Davidski ran the same test on modern Europeans as the Bell Beaker behmouth did on Bell Beaker.

Most Northern Europeans get 40-45% Barcin Neolithic.

https://docs.google.com/spreadsheets/d/1g-cEJmS9cZ6r-_9p0qj1f2qszxiJbactk89DOd0TlHg/edit#gid=1320636554

Estonian 32.8%
Belrusian 40.1%
Norweigan 42.2%
English 45.8%
Bell Beaker Germany 45.9%

When using formal stats Yamanya scores 10-15% Anatolia Neolithic. So some isn't from Neolithic farmers. Of course formal stats aren't flawless. Maybe EF ancestry is lower.

European mtDNA though looks very EEF. Even Baltic and Finnish mtDNA includes lots of EEF stuff.

For example mtDNA haplogroups H1 and T2b are as high in the Baltic states as anywhere in Europe. Both those lineages are quintessential Neolithic farmer haplogroups.

Infact there are already significant matches between Northern-Eastern European farmers and modern Balts. H1c, H1b1, and H1b2 have been found in Swedish Funnel Beaker and Polish Globular_Amphora cultures. All have also been found in Bronze age Latvia and are quite common in modern Balts.


Alexandros said...

Whatever this migration(s) was, it seems to have 'flooded' the Eastern Mediterranean islands (e.g. Cyprus, Crete), as well as some parts of Italy, with haplogroup J2, which still predominates there until today..

AWood said...

@Alexandros

R1b is overwhelmingly more common in Italy than J2 at least in the north-central parts. In the south, R1b still has a very narrow edge over J2 from the data I have seen.

Perhaps you are referring to the orientalizing period of the east Mediterranean. I think you are probably correct, but it seems contact with the Middle East existed already, as did the entrance of J2 in places like the Balkans and likely Italy if we ever get data from that region. While the first European farmers may have been from western Anatolia, I don't believe it was long after that IranNeo/IranChl followed in the same direction.

Synome said...

@Alexandros

My current hypothesis regarding Iran Chalc related ancestry and J2 in Eastern Mediterranean and Italy is as follows:

Iran Chalc and J2 was carried by members of the Kura Araxes culture that expanded rapidly in the bronze age. It reached into Anatolia and seems to have made a large demographic impact, pushing out or replacing Anatolia_Neo related peoples. In time, these migrants spread farther west into the Aegean and Crete, and may have been the founders of the Minoan civilization and at least some of the Greek "Pelasgians".

During the historical period, multiple migrations from the Aegean sea area introduced the bulk of the Iran Chalc like ancestry into Italy, mostly in the south. The density of J2 and recent West Asian ancestry in southern Italy seems to closely correspond to the colonized areas during the "Magna Graecia" era of Aegean Greek colonization of Italy. The migration of Tyrsenian(Etruscan related) peoples from the Aegean area may have brought more, especially to northern Italy where it would have been scarce or nonexistent before.

There were many other mostly later contacts that could have brought this type of ancestry to Italy e.g. Phoenicians, Roman trade/migration, Arab/North African occupation etc. But I think they would be responsible for only a relatively small amount of the Iran Chalc and J2.

Unknown said...


Formal stats like qpADM and D-stats give Northern Europeans 40% Anatolia Neolithic on average. Davidski ran the same test on modern Europeans as the Bell Beaker behmouth did on Bell Beaker.

Most Northern Europeans get 40-45% Barcin Neolithic.

https://docs.google.com/spreadsheets/d/1g-cEJmS9cZ6r-_9p0qj1f2qszxiJbactk89DOd0TlHg/edit#gid=1320636554

Estonian 32.8%
Belrusian 40.1%
Norweigan 42.2%
English 45.8%
Bell Beaker Germany 45.9%

When using formal stats Yamanya scores 10-15% Anatolia Neolithic. So some isn't from Neolithic farmers. Of course formal stats aren't flawless. Maybe EF ancestry is lower.

European mtDNA though looks very EEF. Even Baltic and Finnish mtDNA includes lots of EEF stuff.

Thanks for the table Samuel Andrews. Eh different methods, calculators and tests produce wildly different results about WHG, EHG, EEF, CHG and ANE ancestry in Europe but D-stats too were shown to be often fallible.
40% of EEF seem to be unrealistic as Sardinians plot far from other Europeans and a majority of modern Europeans seem to have much closer affinity to CHG/Iran Chalcolithic dominant populations such as modern Greeks, Turks, Iranians and Caucasians than to more EEF/Levant N-rich Levantines, Bedouins and North Africans (including tribes with low SSA and ~20 WHG). This points to CHG being much higher in modern Europe than EEF.

Chad said...

No one that is full European is more CHG than EEF. This is not debatable.

Seinundzeit said...

Anthro Survey,

Lol, I totally see where you're coming from.

Personally, I'll be much more willing to draw these kinds of connections, once we have a deeper understanding of how phenotype is affected by different genetic and epigenetic processes, and once we see some serious studies on phenotypic plasticity.

But yeah, I completely agree on the Dravidian angle.

And, I do think that this might be a question for Agamemnon; he's the expert on linguistics (I think linguistics might even be his actual field of research?).

Unknown,

"Baloch are more like 10-15% ASI and Iran N like 10-11% ASI so yeah it wouldn't influence the phenotype much since both had slightly ASI influenced phenotype."

If ASI is some form of ENA, 10% is the maximum for Baloch/Brahui; they're more between 5% and 10%. And in that sort of scenario, Iran_Neo is 0% ASI.

Of course, if ASI isn't merely ENA, things could be different.

We'll find out soon enough.

"If they are from Yemen then how is it possible that they have 0% SSA when even Yemenite Jews are slightly SSA admixed and regular Yemenis over 20%. Do they have endogamous culture?"

Who knows.

For what it's worth, nearly everyone has told me that Yemen is where these samples were collected. Supposedly, Bedouin_A are from a totally different country. Which makes sense, considering how different they are from Bedouin_B.

"Just like WHG but I think many unsampled Natufians must have had derived light skin alleles because many Natufian admixed Horners have quite light skin."

For East African populations, natural selection seems to be operative, and is probably a much more recent dynamic; so far, we see the ancestral alleles in all of the Natufian samples that we have.

"Iran N samples had derived depigmentation alleles and one sample(probably WC1) had blue eyes allele despite lacking WHG admixture which means Neolithic Iranians were lighter than other contemporary populations. They may have been slightly lighter than Baloch."

They had the same combination of alleles (at sites implicated in skin depigmentation) as darker North Indians, and many South Indians.

So, Neolithic Iranians were probably dark brown/medium brown, like most modern Indians.

The Baloch/Brahui are more "genomically depigmented" in comparison to Iran_Neo (that barely makes sense, but you probably know what I mean), and most have pale unexposed skin (the dark skin one sees in photos is essentially the product of extreme tanning. Paler Europeans who travel in those parts without skin protection often end up looking like walking pieces of raw-flesh), so they probably are paler than Iran_Neo.

Ryan said...

This talk of skin colour is pretty silly. It's under selection - heavy selection. Just looking at % of ancestry and trying to figure out skin tone or other superficial traits that are heavily influenced by the environment is pointless.

Anthro Survey said...

@Sein
Btw, there are three studies to date that have utilized 3D imaging to connect certain facial features to a few loci in the genome. David made a post about this and some guy left 2 additional links in the comments. That type of stuff I find really neat.

The thing is, as swarthy as the Iran_Neo crowd would have been, they were still likely lighter than the resident ASI population there prior to their arrival. I imagine the ASIs to have been darker than Sub-Saharan africans.

Just look at how dark tribals like the Paniya and some Adivasi are in relation to average Dravidians from Hyderabad. There is clearly a rough but statistically significant gradient in play.

Outside of India, on average, eastern basal-rich populations tend to be somewhat darker than the western ones when we account for steppe and ssa. Dagestanis tend to be darker or at least no less dark than Syrians. Syrians and even Berbers tend to be lighter than untanned Pashtuns(who together with Balochs are lighter than upper caste Punjabis).

Of course, this set of observations is quite crude and ignores potential drift and selection dynamics over time, but still worth noting.

My intuition tells me that selection for and proliferation of lighter alleles likely began with Western Basal-rich and/or Villabruna-related peoples.

@a It's not just in the nasal size or convexity. It's also about the finer structure, the proportions of the midface face in relation to the nose, other tell-tale facial landmarks, and so on.

batman said...

High or huge noses with long, narrow nostrils is an arctic trait - like paled skin and depigmented hair. All of them known as "neccesary adaptions" to survive in the dim and cold winters of the arctic hemisphere.

Chad said...

Anthro Survey,

Post what you want run.

Seinundzeit said...

@Anthro Survey

Lol, I made a few comments in that same thread.

For what it's worth, I think that one can probably get a very broad/rough picture of how ancient people looked, just by creating an IBS list using SNPs that are supposedly implicated in human phenotype.

In fact, Kurd (of Anthrogenica) actually ran an IBS for me using all the SNPs utilized in those three papers, plus SNPs involving pigmentation.

My closest matches were a bunch of Kurds, Afghan Pashtuns, Tajiks, Dagestanis, an Upper Caste North Indian, etc, which actually does make sense in terms of how I look IRL (I can pass anywhere from Istanbul to Peshawar; no one would notice me as a foreigner in West Asia, the Caucasus, the Tajik and Uzbek portions of Central Asia, and the Pashtun and Dardic parts of Pakistan. Not to mention Latin America).

So, an interesting exercise would be to run an IBS list for a single ancient genome of interest (in comparison to modern populations, using those SNPs implicated in facial morphology and pigmentation), to see which modern population they are closest to.

Anyway, I think we have to recognize that ancient West Eurasian populations were very diverse in terms of pigmentation, and many were likely to be far outside the skin color variation seen in modern Europe, West Asia, and South Central Asia.

For some reason, lighter pigmentation has been under heavy selection across West Eurasia.

But, we really shouldn't project current differences onto ancient populations, precisely because of these ongoing processes of selection.

Also, almost every example of the Villabruna cluster has had ancestral alleles, so I would really shy away from that sort of inference.

Again, the differences could very well boil down to different selective dynamics, and could even be chalked up to the vagaries of genetic drift.

At the end of day, we just have to keep in mind that none of this is "inherent" to any ancient stream of populations. "Evolution" (which is a reification of multiple mindless/absolutely contingent processes operating at different biological scales) is all about divergence.

Personally, I feel like essentialism and Platonism are always to be avoided.

Alexandros said...

@Synome

Very interesting hypothesis about entry of J2 and Iran_Chalc admixture in the Eastern Mediterranean. I am thinking along the same lines. For Cyprus, there is very clear archaeological evidence of a substantial EBA migration from Anatolia. Probably, as you rightly say, this may have also affected Crete and is very likely to have played a major role in the initiation of the Minoan civilization. It is a mystery however why such a civilization (or similar) did not also appear in Cyprus, since the same stock of people were involved. Also, during the Neolithic, Cyprus has been much more developed and 'active' than Crete, so what could have sparked such an unprecedented development in BA Crete?

Samuel Andrews said...

The Ubaid culture could have played a part in spreading IranNeolithic-like ancestry to many parts of the Middle East.

Sometime between 4,900 and 4,000 BC this culture expanded out of Southern Iraq all the way to Syria, Eastern Turkey, and Israel. A museum I went to argues it spread with people not just ideas.

It became the dominate culture in Mesoptamia. The Sumerians were people of the Ubaid culture that learned how to write.

The Ubaid cullture may have laid down the foundations for later Mesoptamian civilzations: Sumerians, Akkadiaans, Assyrians, Persians.

Anonymous said...

Unfortunately, it just doesn't seem possible to conclude anything here - the picture is far too muddled and the evidence lost in the haze of time.

We can talk with relative confidence about the IE linguistic and genetic expansions because the pattern of IE linguistic expansion is so overwhelming that it cannot be denied. The Middle East, unfortunately, seems far more complex.

It seems likely that there were at least five (or six) major linguistic stocks in the area in the third millenium BC: Sumerian, Elamite, Hattic, Hurrian, Semitic, Anatolian and probably Egyptian (while outside formal Egyptian rule, large parts were under Egyptian influence at the very least). None of these languages appear to be identifiably related to any of the others, except that Semitic and Egyptian are related. By the 2nd millenium, at least two more linguistic stocks are attested: Indo-Aryan (related to Anatolian, but they had split millennia before) and Kassite (some people have suggested a link between Kassite and Hurrian, but this is speculative).

Of those eight language families, only one, Semitic, has survived. [Well, Indo-Aryan has survived but the branch that invaded the Mesopotamian area has died out; it's not closely related to Iranian and it's unclear what relation it has to modern Indo-Aryan languages].

We know that there was considerably movement of tribes and language - Sumeria is known to have been ruled by speakers of Sumerian, Kassite, Semitic, Elamite, and also Gutian (of unknown affiliation). We also know that these societies were often multilingual - Mitanni, for instance, had a Hurrian-speaking populace but an Aryan ruling class. The Hittites spoke Anatolian, but also recorded rituals in Hattic.

And then there's the other big problem: all these groups, all this moving around... but the time of the expansions you're talking about, and the time of the early civilisations, was thousands of years before the languages we know about. We don't know what Ubaidic Sumer spoke - Sumerian? Elamite? Something else?

So there's really no way to know which of the many known language families of the area are the 'survivors' of the initial expansions, or of the initial civilisation - or even if any of them are.


Well, it's not Indo-Aryan. And it's not Anatolian - at least, the initial wave certainly wasn't.

I don't think it can be Semitic, either. The dates you give for Iran and Western Turkey seem too old and to widespread to match up with a Semitic expansion. But there's nothing stopping it being an earlier expansion of "pre-Proto-Semitic", of which Proto-Semitic was one of the daughters. Or of being some other, unattested Afroasiatic family.

There may also have been many waves of expansion. In particular, the findings in the Levant and Egypt may represent a later Semitic expansion out of a Mesopotamia.


--------

I wouldn't discount the idea of there being multiple language families involved in the expansion, either. Many language families could have been living side-by-side in the Zagros (etc), without close relations - just look at the Caucasus. Or at New Guinea!

Anthro Survey said...

@Sein

Oh, Gui compiled the list of the loci implicated in facial phenotype from all three(?) papers, right?

It was a good start, but I think we will defintely have to wait longer for a more comprehensive list to accumulate. After that, aprcity and antroscape will go out of business. lol Or, they will have to adapt.

Personally, I would be more interested in running something like that on ancient populations not exceeding 7000 ybp mark. Well, let's say 7K to 2K BP. You're right about selection dynamics and in all likelihood, the bulk of it accounting for much of present-day phenotype(craniofacial) distribution took place in the last 7K years, if not within 5. Of course, running Natufians and their contemporaries is not a bad idea. I'm just afraid we'll find few examples of "interesting" mutations and if we do, we won't be able to make meaningful conclusions as to how they spread or whether that geographical population was even reaponsible for their dessimination. Similarly, looking at modern populations is tricky unless you've got a keen eye and can pick up breadcrumb trails(as opposed to assuming smth like Brahui came from S.India).

So I think looking at some intermediate period will ensure that what we're looking for is, uh, caught in the act!

Anthro Survey said...

@Chad

Sure thing! Somehow the link with the initial results here got de-activated so just waiting until it's restored. I remember the general results but want to see them again so that my selection of stats you'll run in the next round is efficient.

If you're on anthrogenica somewhere I can send them there, or, if you check this thread for the next 3-4 days, I can post here. Let me know.

Chad said...

Here, or a message at Anthrogenica is fine.

Unknown said...


https://docs.google.com/spreadsheets/d/1g-cEJmS9cZ6r-_9p0qj1f2qszxiJbactk89DOd0TlHg/edit#gid=1320636554

Could someone knowledgeable try to make a similar table to the one above but with ANE, East Asian based on Lake Baikal/Devil's Gate HG, Natufian, Iran Neo, Ust'-Ishim and without Nganasan.(Keep CHG, WHG, EHG, Onge)

Yeah it looks like indeed most Europeans have more EEF than CHG but if so why Europeans generaly plot closer to Iranians than Levantines on PCAs? Is it because of SSA in Levantines?
Also how is it possible that North and East Europeans have such low CHG admixture when Yamnaya/PIE was 50% CHG and the steppe ancestry is supposed to be high in this part of Europe?

https://docs.google.com/spreadsheets/d/1kPtrpIFqUYUPxym-VBcFrtcrKP3yzIny6r1PRDej6cc/edit#gid=849660723

Samuel Andrews said...

@Unknown,
"Yeah it looks like indeed most Europeans have more EEF than CHG but if so why Europeans generaly plot closer to Iranians than Levantines on PCAs?"

In the classic West Eurasian PCA WHG, Natufian, and EEF are in the western edge and EHG, CHG, IranNeo are in the eastern edge.

Northern Europeans are about 50% far-western(WHG, EEF). The other 50% pulls them really far east. Iranians have about as much CHG/IranNeo stuff as Northern Europeans have EHG+CHG.

So Iranians plot about as far east as Northern Europeans. But the eastern ancestry of Northern Europeans and Iranians is very different because Northern Europeans have a big chunk of EHG but Iranians don't.

Populations can plot where they do on the classic West Eurasian PCA for different reasons. The PCA has two clines a west vs east cline and north vs south cline.

West=WHG, Natufian, EEF. East=EHG, CHG, IranNeo. North=WHG, EHG. South=Natufian, EEF, IranNeo, CHG.

Seinundzeit said...

@Anthro Survey

"So I think looking at some intermediate period will ensure that what we're looking for is, uh, caught in the act!"

One idea I've been chewing over: a supervised ADMIXTURE test, using only SNPs implicated in cranio-facial morphology + pigmentation.

Basically, the idea would be to use ancient populations as fixed references, and see if modern populations can come out as mixtures of them (looking at phenotype).

If that makes any sense.

Unknown said...



>So I think looking at some intermediate period will ensure that what we're looking for is, uh, caught in the act!"
One idea I've been chewing over: a supervised ADMIXTURE test, using only SNPs implicated in cranio-facial morphology + pigmentation.
Basically, the idea would be to use ancient populations as fixed references, and see if modern populations can come out as mixtures of them (looking at phenotype).


I found this http://i.imgur.com/qeunXW0.png

You can try fineStructure+Chromopainter. It's supposed to be more accurate than ADMIXTURE.

I wonder if admixture dating programs such as ALDER and G-PHOCS could be run on just facial SNPs. There are supposed to be some programs that can create images of faces based on DNA.

>Northern Europeans are about 50% far-western(WHG, EEF). The other 50% pulls them really far east. Iranians have about as much CHG/IranNeo stuff as Northern Europeans have EHG+CHG.
So Iranians plot about as far east as Northern Europeans. But the eastern ancestry of Northern Europeans and Iranians is very different because Northern Europeans have a big chunk of EHG but Iranians don't.

It's not just North Europeans but Europeans in general which is puzzling to me. Modern Greeks cluster particularly close to Iranians despite the fact that they should have more EEF and less ENA and CHG compared to Iranians. Iranians also have ASI and Mongolian admixtures. Perhaps Europeans plot like that due to EHG and recent East Asian admixture which makes them closer to significantly ENA admixed Iranians than African admixed Levantines. Although modern Levatines also have high CHG/Iran Chl admixture and therefore ENA.
Iranians don't have much EHG or WHG but they seem to have pretty high ANE and EHG is ~75% ANE.

Vara said...

@Samuel Andrews
"He had a huge nose bridge and a thin nose. That's what the stero typical Jewish nose is, I think it could be a legacy of IranNeo."

Actually history tells us the hooked nose was a big thing in the Iranian Plateau.

"The Persians affect such as are hawk-nosed and think them most beautiful, because Cyrus, the most beloved of their kings, had a nose of that shape." - Plutarch

"The Ethiopian preferring the blackest and most snub-nosed, and the Persian approving the whitest and most hook-nosed" - Sextus Empiricus

However, I don't think we can compare modern day populations with prehistoric ones. Persians for example were described as tall and pale, by the Greco-Romans, even in the Sassanid mosaics and textiles they are, yet modern day Iranians aren't tall or pale.

@Anthro Survey
"Dagestanis tend to be darker or at least no less dark than Syrians."

I'm not sure if you've seen the Russian Wrestling team. North Caucasians are literally the palest people on Earth.

Unknown said...


>However, I don't think we can compare modern day populations with prehistoric ones. Persians for example were described as tall and pale, by the Greco-Romans, even in the Sassanid mosaics and textiles they are, yet modern day Iranians aren't tall or pale.

Modern Iranians may have more ASI than during the Sassanid era. They mixed with other middle eastern, south and central asian populations and light skin is recessive so their dark skin frequencies could have increased since ancient times.

Unknown said...


I'm not sure if you've seen the Russian Wrestling team. North Caucasians are literally the palest people on Earth.

North Caucasians have dark and lighter types. I remember seeing some study that showed Poles being the lightest in Europe and the World on average.

Vara said...

@Unknown

"Modern Iranians may have more ASI than during the Sassanid era. They mixed with other middle eastern, south and central asian populations and light skin is recessive so their dark skin frequencies could have increased since ancient times."

We still don't know how much the Iranian Plateau was changed by the Arab and the Turko-Mongol invasions, but autosome wise they have been almost the same since the Copper Age.

"North Caucasians have dark and lighter types. I remember seeing some study that showed Poles being the lightest in Europe and the World on average."

I was being hyperbolic, but the dark types are usually South Caucasians. Also, do you know where can I find that study?

@Salden

"The Ethiopian preferring the blackest and most snub-nosed, and the Persian approving the whitest and most hook-nosed" - Sextus Empiricus

"Agesilaus gave orders to his heralds that the barbarians who were captured by the Greek raiding parties should be exposed for sale naked. Thus the soldiers, seeing that these men were white-skinned because they never were without their clothing, and soft and unused to toil because they always rode in carriages, came to the conclusion that the war would be in no way different from having to fight with women." - Xenophon

And quotes about Darius and Artaxerxes being the tallest and most handsome men, Alexander passing by the tall Persian beauties, Shapur II being the tallest in the battle...etc. I don't have my Arab history books with me but you'll find the same thing about Persians in Masudi's works as well, or you can read the description of any king in the Shahnameh.

All I'm saying is that finding pale people in Iran that can creep the Greeks out is pretty rare nowadays.

As for art:
http://www.angelfire.com/rnb/bashiri/Sassanian/Sassan8.gif
http://www.karakalpak.com/images/kimhis807.jpg
http://www.karakalpak.com/images/kimhis811.jpg
https://www.zum.de/whkmla/sp/1011/ignoramus/figure6.jpg
https://en.wikipedia.org/wiki/Abyssinian%E2%80%93Persian_wars#/media/File:Khosrau_I_Textile.jpg (comparison with Ethiopians)

Vara said...

From the Bundahishn: "First of the human species Gayomard was produced, brilliant and white, with eyes which looked out for the great one"

Also, one more thing, in the Shahnameh evil men are described as dusky with hooked noses and that contradicts what the Greeks have said.

Vara said...

You ignored the quotes I posted and I can't find you a Renaissance artist who lived during that time. What matters is how they depicted themselves. And the Geishas wear makeup which gives them that look.

Parsis aren't Sassanid Era Persians and many of them even descend from Safavid Persians. We only have one mention of the earliest Zoroastrian migrants who were from the Ghaznavid Era, and that poem was from the 16th century. They aren't wholly endogamous because they marry into North Indians for health reasons.

Vara said...

"Realistic artwork was done long before the Renaissance."
I'm not sure what you consider realistic. What was not realistic about the artwork? They didn't have ghost skin for sure.

"Men are also depicted with the snow skin look. See Miyamoto Musashi:"
Fair enough.

"Still the Zoroastrians."
Doesn't matter. I was speaking of ancient Persians.

"The comment about skin is just talking about tanning. The rest is just hyperbole babble."
Sorry, I forgot to reply to this comment. It's past midnight here. Sextus was trying to prove a point, i.e to some beauty is that which resembles them. Ethiopians are black so they prefer blackest, Persians are white so they prefer the whitest.

Since I am currently in the No.1 wrestling state, I'm gonna use wrestling as proof against the tanning thing.

Some Iranians with less clothing:
https://www.google.com/search?rlz=1C1CHBF_enUS718US718&biw=1920&bih=974&tbm=isch&sa=1&q=Iranian+wrestling&oq=Iranian+wrestling&gs_l=psy-ab.3..0l2j0i24k1l2.205048.206087.0.206247.10.10.0.0.0.0.125.888.8j2.10.0....0...1.1.64.psy-ab..0.10.885...0i67k1j0i30k1.Td8slPopg2o

Do they match the descriptions? I'd say the Russian wrestling team matches the description.

Vara said...

Don't get too personal.

"However, I don't think we can compare modern day populations with prehistoric ones."

^ that's what I'm trying to show.

Anthro Survey said...

@Sein:

Yeah, that supervised admix scheme could actually work. It would be best to do this for populations w/moderate sample sizes like bell beakers and LBK. Also, it might be necessary to fiddle with individual alleles somehow----so as to put more weight on some than on others based on how dramatic the resulting change in look is. This would be somewhat subjective but necessary for realistic results, imo.

Anthro Survey said...

Vara,

We shouldn't look to Sassanian royalty(let alone preceding Achamenian nobility) to get an idea of how the average inhabitant would have looked like. They were relatively endogamous and likely had above average steppe ancestry. That's like looking to Jatt Sikhs to get an idea of how Punjabi dalits might look like.

Many ancient accounts were notoriously erroneous. The Greeks described Thracians as being ruddy. Well? Turns out they were as swarthy(by euro standards) as any Balkaners of their day. Things haven't changed much and, no, the "turks" didn't "taint" anything. And how could those observers possibly have known the origin of these "pale Persians"? Maybe they were Sakas? Maybe Greek mercenaries from Anatolia? Maybe Sogdians? Perhaps they were emphasizing a rare phenotype that jumped out at them? The Persepolis reliefs tell us something else and probably that sort of pigmentation was more in line with the climatic conditions of the Iranian plateau----which will quickly and mercilessly fry any pale skin. Even Bactria/Sogdia area isn't that hospitable.

The Arab demographic impact(and in general) on Greater Iran has been vastly overestimated. Many of the Arabs there were mere mowlas(like clients) and not actual migrants from the peninsula. North Africa was a different story.

North caucasians tend to be as cherrypicked almost as much as Berbers. Despite their steppe admixture, they aren't considerbaly lighter than Lebanese and many Syrians by any standard.

Vara said...

@Salden

"You seem really invested in this imaginary Eastern European Persian population that was mysteriously dissapeared. You wouldn't say you Wuz Kangz, would you?"

More strawmen, please. I never said Persians were European looking because Europeans are ruddy and blonde. Persians were never described as blonde.

" I'd say the Russian wrestling team matches the description."
^ let me spell that out for you: North Caucasians match that description.

@Anthro Survey

"They were relatively endogamous"

They weren't. They married Assyrians, Jews...etc.


"likely had above average steppe ancestry"

No evidence of that for the Sassanians.


"Many ancient accounts were notoriously erroneous. The Greeks described Thracians as being ruddy."

Fair enough.

"Things haven't changed much and, no, the "turks" didn't "taint" anything. And how could those observers possibly have known the origin of these "pale Persians"? Maybe they were Sakas? Maybe Greek mercenaries from Anatolia? Maybe Sogdians? Perhaps they were emphasizing a rare phenotype that jumped out at them?"

Xenophon was a first hand witness and a friend of Agesilaus, he could definitely tell the difference between them since he took part in the war.


"The Persepolis reliefs tell us something else and probably that sort of pigmentation was more in line with the climatic conditions of the Iranian plateau----which will quickly and mercilessly fry any pale skin. Even Bactria/Sogdia area isn't that hospitable. "
Not sure of this but wasn't the area colder back then?


"The Arab demographic impact(and in general) on Greater Iran has been vastly overestimated."

It is. However, Turko-Mongol is underestimated. They depopulated entire cities. Who knows what that changed.

Unknown said...

>North caucasians tend to be as cherrypicked almost as much as Berbers. Despite their steppe admixture, they aren't considerbaly lighter than Lebanese and many Syrians by any standard.

Depends on North Caucasians in question. Circassian slaves were considered white like Europeans by Turks. So Circassians might be ligter than some darker South Europeans such as Greeks and Iberians.

By the way anyone noticed that Hungarian from Iron Age IR1 plots near upper paleolithic Europeans? Why is that?

https://postimg.org/image/smwpvqvkb/
http://postimg.org/image/6f34xpm0x/
http://postimg.org/image/4xcmp1ayp/

Unknown said...

Does anyone know if Anatolia Neolithic has IranNeo or CHG admixture? According Lazaridis it's IranN but I think the existence of CHG wasn't discovered yet then.

http://i.imgur.com/JhmtF4J.jpg

For the king said...

@vara, you should generally ignore historians when it comes to phenotypes. Artwork is much more representatives of actual phenotypes. There are many historians that described Iranic nomads such as the Scythians/Sarmatians as swarthy and dark-haired, while other said that they were light. Light vs dark is a very subjective matter.


Darius III, the last Achaemenid Persian Shah according to the Romans(100 BC, Sassanids ruled Persia at that time)
https://upload.wikimedia.org/wikipedia/commons/f/f5/Darius_III_of_Persia.jpg


Sassanid Woman, Pretty much looks like Modern West Asians.

https://upload.wikimedia.org/wikipedia/commons/thumb/e/e2/Bishapur_zan.jpg/335px-Bishapur_zan.jpg

Sogdian Ambassadors, pretty similar to modern most east Iranics.

http://1.bp.blogspot.com/-mW7-xvz6Xyk/VBx8YZSu_BI/AAAAAAAAHcM/Bu2XGrYedBg/w1200-h630-p-k-no-nu/640px-Afrasiab_-_details_from_The_Ambassadors'_Painting.JPG

About the Turko-Mongol tangent, you said yourself. Modern Iranians are extremely similar to copper(Iran chl) and to a lesser degree neolithic Iranians.

The Mongols are the ones that "wiped out" cities, while Turkic rarely did that. The Turkic and Mongolian migration routes usually bypassed most of Iran and were concentrated in NE and NW Iran(Which explains why those areas have the highest east Eurasian in Iran, but it usually never goes higher than 5%). There are many central, Caspian, Lurs, Kurds(NW) and southern Iranians that don't score any east Asian/Siberian. The brutality of the Mongols is highly exaggerated throughout history. Most of the fighting done by the Mongols in Persia was against the Turko-Persian mamluk Khwarazmian dynasty.

Vara said...

@For the king

I already posted a lot of artwork above. Here some more:
http://www.iranreview.org/file/cms/images/stories/14128.jpg (Sassanid Noblewoman)
http://www.iranreview.org/file/cms/images/stories/14119.jpg (Sassanid Noblewoman and some old dude)
http://m7.i.pbase.com/o6/93/329493/1/141565877.yWj7z2QG.UZTM112501.jpg (Noble)


"Darius III, the last Achaemenid Persian Shah according to the Romans(100 BC, Sassanids ruled Persia at that time)"

The Sassanids did not even exist during that time and Romans did not have contacts with the Iranians during that time. In fact both Greeks and Persians look the same in that artwork but with different clothing because they were modeled after the Romans themselves.

https://en.wikipedia.org/wiki/Alexander_Mosaic#/media/File:Battle_of_Issus_mosaic_-_Museo_Archeologico_Nazionale_-_Naples_BW.jpg

The Persian King: http://2.bp.blogspot.com/-_m6zOrXEUeU/UW5X3ShszhI/AAAAAAAAKPU/G17QcN31y80/s1600/26.jpg


"Sogdian Ambassadors, pretty similar to modern most east Iranics."

This is a good one. However, it's not known which of them is Sogdian since the artwork depicts envoys from all over the world, including India and East Asia, and all of them wear the same clothing.

https://en.wikipedia.org/wiki/Afrasiab_painting#/media/File:Afrasiab_-_details_from_The_Ambassadors%27_Painting_2.JPG
https://en.wikipedia.org/wiki/Afrasiab_painting#/media/File:Afrasiab_-_details_from_The_Ambassadors%27_Painting_1_-_Northern_wall.-_Chinese_boat.JPG
^ East Asians.


"About the Turko-Mongol tangent, you said yourself. Modern Iranians are extremely similar to copper(Iran chl) and to a lesser degree neolithic Iranians. "

Indeed. That's to prove my original point: "I don't think we can compare modern day populations with prehistoric ones."
In other words, just because Yamnaya were dark doesn't mean their East European descendants are dark.


" There are many central, Caspian, Lurs, Kurds(NW) and southern Iranians that don't score any east Asian/Siberian."

Just googled some Lurs, some fit the description and supposedly their language is the closest to Middle Persian. Thanks!

"The Mongols are the ones that "wiped out" cities, while Turkic rarely did that."

They did actually. They wiped many cities and fortresses in the Caucasus and Chorasmia during the third war with the Sassanids.

Unknown said...



>For what I know from the literature on the argument, the first to hypothesize that the EEF from Northern Anatolia and Aegean Sea were the "Etruscans" was I in a letter to the great Dutch linguist Schrjiver who wrote some paper thinking having demonstrated that they speak Hattic, but my theory is that they had nothing to do with "Natufians" neither Iran as we knew from the recent papers, but they were anyway linked to Europe and the European HG, believing that I/J was anyway an HG haplogroup with R1 and C1.


EEF is ~65% Natufian-like.
http://i.imgur.com/JhmtF4J.jpg

http://desu-usergeneratedcontent.xyz/his/image/1475/20/1475209564091.jpg

Gioiello said...

@ Unknown

">For what I know from the literature on the argument, the first to hypothesize that the EEF from Northern Anatolia and Aegean Sea were the "Etruscans" was I in a letter to the great Dutch linguist Schrjiver who wrote some paper thinking having demonstrated that they speak Hattic, but my theory is that they had nothing to do with "Natufians" neither Iran as we knew from the recent papers, but they were anyway linked to Europe and the European HG, believing that I/J was anyway an HG haplogroup with R1 and C1.


EEF is ~65% Natufian-like.
http://i.imgur.com/JhmtF4J.jpg

http://desu-usergeneratedcontent.xyz/his/image/1475/20/1475209564091.jpg "

Not only Unknown is unknown but makes unknown also his/her/its opponent. Ridiculous! Which Natufians? Those hg E or those with the massive injection of hg J from Caucasus or perhaps Europe?

Unknown said...

>Which Natufians? Those hg E or those with the massive injection of hg J from Caucasus or perhaps Europe?

Those with E of course. Lazaridis showed EEF had a lot of neolithic levantine-like ancestry. EEF and Natufians were closely related populations and modern Sardinians share more genetic drift with Natufians than modern Saudis and Bedouins do or any other modern population for that matter.


http://www.anthrogenica.com/archive/index.php/t-8262.html

https://gnxp.nofe.me/2017/05/30/ancient-egyptians-black-or-white/

Anonymous said...

We know what spread from there eventually, semitic languages, but I doubt it originated there. But sumerian itself was very similar to IE languages and actually could well have been an IE language. The earliest writings there may actually have been in another language that was IE 'euphratic'.

So it may have been IE originally and may have been a language isolate, or several language isolates.

Davidski said...

But sumerian itself was very similar to IE languages and actually could well have been an IE language. The earliest writings there may actually have been in another language that was IE 'euphratic'.

So it may have been IE originally and may have been a language isolate, or several language isolates.


Nonsense.

Gioiello said...

@ Davidski

"But sumerian itself was very similar to IE languages and actually could well have been an IE language. The earliest writings there may actually have been in another language that was IE 'euphratic'.

So it may have been IE originally and may have been a language isolate, or several language isolates.

Nonsense".

Davidski, to say that Sumerian was an IE language, basing upon some questioning roots (one may make roots say all what he desires) and not knowing that Sumerian is an agglutinative language very likely of the Sino-Tibetan stock (I am thinking to Ladakhi and other Himalayaian languages) is out of any likelyhood.