search this blog

Saturday, October 10, 2015

Eight thousand years of natural selection in Europe - take 2


Open access at bioRxiv [LINK]. Lots of new samples in this updated version. The Principal Component Analysis (PCA) below from the paper appears to be affected by projection bias or shrinkage, but it's more or less correct. Can't wait to get my hands on the genotype data.

Abstract: The arrival of farming in Europe around 8,500 years ago necessitated adaptation to new environments, pathogens, diets, and social organizations. While indirect evidence of adaptation can be detected in patterns of genetic variation in present-day people, ancient DNA makes it possible to witness selection directly by analyzing samples from populations before, during and after adaptation events. Here we report the first genome-wide scan for selection using ancient DNA, capitalizing on the largest genome-wide dataset yet assembled: 230 West Eurasians dating to between 6500 and 1000 BCE, including 163 with newly reported data. The new samples include the first genome-wide data from the Anatolian Neolithic culture, who we show were members of the population that was the source of Europe's first farmers, and whose genetic material we extracted by focusing on the DNA-rich petrous bone. We identify genome-wide significant signatures of selection at loci associated with diet, pigmentation and immunity, and two independent episodes of selection on height.



Mathieson et al., Eight thousand years of natural selection in Europe, bioRxiv revised preprint, posted October 10, 2015, doi: https://dx.doi.org/10.1101/016477

See also...

Lactase persistence and ancient DNA

204 comments:

1 – 200 of 204   Newer›   Newest»
Krefter said...

Y DNA and mtDNA results for new sample sis listed!! R1a-Z94 dating over 4500 years Poltavka!!


http://www.biorxiv.org/highwire/filestream/7735/field_highwire_adjunct_files/1/016477-2.xlsx

Krefter said...

R1b, R1a, Q1a in Samara_Eneolithic 5200-4000 BC. One had mtDNA H2a1 which first appears in Europe with Corded Ware.

Davidski said...

Yep, I knew Corded Ware and R1a-M417 were from around those parts.

Dude Manbro said...

Y-DNA J in an EHG individual from Karelia 5500 BC. J2a1 in Barcin in 6500-6200 BC. No R1b in the Chalcolithic samples from Atapuerca.

Dude Manbro said...

One R-Z93 in Poltavka and four in Srubnaya.

Krefter said...

Lots of Late Neolithic Y DNA from Germany. Almost all are R1a, are contemporary with Corded Ware, but aren't labelled as Corded Ware.

Rob said...

Wow
Gotta read

Krefter said...

Here's a spreadsheet with the new samples. Year, culture, location, mtDNA, Y DNA.

https://docs.google.com/spreadsheets/d/1Vjbp450AwI7R-Y9J1YGSm9FjJWu9s9lx1azjUJbS8hQ/edit?usp=sharing

Kurti said...

"Y-DNA J in an EHG individual from Karelia 5500 BC. J2a1 in Barcin in 6500-6200 BC. No R1b in the Chalcolithic samples from Atapuerca. "


Now take this and swallow it deep down your throat people who insisted on the "female Teal wives theory" :D

So much to that. I go so far and say a good chunk(Not all) of the R1b was also coming together with J.

Kurti said...

Maybe directly from South_Central Asia (Teal farmers) or through the Caucasus. At the end of the day this J sample proves that the Teal ancestry in EHG (and therefore in Yamna) is not linked to "horny H&G kidnapping Teal wives). I go that far and say the l23 in Yamna is also POSSIBLY of Teal origin.

Don't hate on me and no I am not on drugs :D

Rob said...

KRefter

Thanks for S/S

@ Kurti

I2c in Anatolia. Back -> from Europe :)

Was that J2 sample from Karelia "Teal" ?

Davidski said...

R1b still looks like an ANE/EHG marker native to North Eurasia, along with R1a and Q.

Dude Manbro said...

@Rob

It isn't J2 from Karelia. It is just labeled J and it is in an EHG context, as far as I can tell (5500 BC). I do not think the "teal" people had arrived.

@Davidski

Yeah, 8 male samples from Chalcolithic Iberia (2800-2600 BC) and no R-L51 or Steppe ancestry.

Kurti said...

@David we will see my friend :) We will see.

I say R1b was widespred some branches reached the Eastern farmers other EHG groups. Even some among the EHG groups might be Teal derived and founder effect. At least the J samples proves an influx from Teal like groups in EHG. so it was there. No need for Wives theory.

Where are all those "knights" now who defended this obvious bogus theory and made sarcastic comments?

Come out come out where ever you are

Having a little fun now :D

a said...

I have been saying it all along.

Story of R1b- 3000 years in the same location Samara and Steppe.
R1b-HG[Samara]. 7650BP-R1b-Khvalynsk[ near Samara] 6000BP+/- and R1b-YamnayaR1b-Z2105[Samara] 5000BP+/-/ With two of the Elite Kurgan and pre-Kurgan burial containing prestigious copper artifacts and separated by as much as 1000+/- years.

Davidski said...

Kurti,

It's well documented that steppe males were taking wives from near the Caucasus (see paragraph 4 on page 58).

http://www.researchgate.net/publication/272027328_The_Steppe_and_the_Caucasus_during_the_Bronze_Age_Mutual_Relationships_and_Mutual_Enrichments

It should be obvious that this is where most of the Near Eastern ancestry on the Eneolithic/Early Bronze Age steppe came from.

You'll have to accept this at some stage.

Lank said...

Exciting stuff. Looks like the majority of the Anatolian samples are from Barcin.

The EHG J male looks similar to the other EHG samples in the PCA and ADMIXTURE, so he presumably lacks Basal Eurasian. This is not consistent with ancestry from groups resembling "teal", neither is the fact that his J lineage wasn't assigned to a subclade.

Rob said...

@ Dude

"8 male samples from Chalcolithic Iberia (2800-2600 BC) and no R-L51 or Steppe ancestry."

It seems Iberia was swamped with I2a2 in Late Neolithic Copper Age, replacing earlier "Sardinian" -like M26.

Rob said...

Can someone make sure that Nirjahar doesn't read this:

" Previous work documented that such ancestry appeared east of the Urals beginning at least by the time of the Sintashta culture, and suggested that it reflected an eastward migration from the Corded Ware peoples of central Europe. However, the fact that the Srubnaya also harbored such ancestry indicates that the Anatolian Neolithic or EEF ancestry could have come into the steppe from a more eastern source. Further evidence that migrations originating as far west as central Europe may not have had an important impact on the Late
94 Bronze Age steppe comes from the fact that the Srubnaya possess exclusively (n=6) R1a Y-
95 chromosomes (Extended Data Table 1), and four of them (and one Poltavka male) belonged to haplogroup R1a-Z93 which is common in central/south Asians12 96 , very rare in present-day Europeans".
97 and absent in all ancient central Europeans studied to date."

a said...

Catastrophic failure grips the Anatolian & Iberian camp, as they try to figure out the reason as to why no R1b is showing up in the regions. Let the creative explanations begin.What do they have say now?

Krefter said...

@Kurti,

Stop taking the Teal wives theory emotionally. No one is saying EHG men were superior and super-macho. They're just trying to figure out where Yamnaya's Near Eastern ancestry came from.

The Y DNA J from Karelia probably had no detectable Near Eastern ancestry and was probably J*(xJ1, J2). My guess is that it came to Europe with its brother hg I, it's just that J was less successful than I.

Also, the Eneolithic guys from Samara probably didn't get their R1a and R1b from the Near East. This is because before Teal ancestry appeared in Russia, R1a and R1b was already there. They're more evidence of the association between Y DNA P(Q and R) with ANE. The R1b1 that existed in Russia in the Mesolithic may have mostly died out and R1b1a2 may be from the Near East, because it looks that way in modern Y DNA. We need more ancient Y DNA to know for sure.

Rob said...

So blue eyes and light complexion was a mixture of WHGs and EEFs, but not the "Kurgan warriors" ? Interesting..

a said...

Sorry Krefter your theories also suffer a huge setback tonight with your constant Middle Eastern origin of R1b1a2; the Khvalynsk R1b has just shown us the presence/burial of R1b in the region for over 3000 years.

Kurti said...

Yamna looks 50-60% "Armenian like" and 40-50% EHG like

Samara H&G(EHG) look 75% "EHG" and 25% "Armenian" like.

Sintashta looks 40% Anatolian(Western) farmer and 60% Yamna (50/50 teal/EHG)

Andronovo on the other hand 80% Yamna and 20% Anatolian farmer.

Afanasievo ~55% Armenian like and 45% EHG

Poltavka similar to Afanasievo.

page 21
http://www.biorxiv.org/content/biorxiv/early/2015/10/10/016477.full.pdf

Kurti said...

David said



"t should be obvious that this is where most of the Near Eastern ancestry on the Eneolithic/Early Bronze Age steppe came from."

Give up on it mate. The wives theory is dead. Not saying that EHG didn't take Teal wives. Just saying it was a simple merging of two populations (EHG and Teal) to one. And not kidnapping.


Dude Manbro said...

"Our paper presents a complete transect of the Samara region beginning with the Samara EHG
hunter-gatherer (~5,600BCE)5 and ending with the Srubnaya culture (~1,850-1,200 BCE) and
a singleton “Scythian” Iron Age individual (~300BCE). In eastern Europe outside the steppe,
a new individual from the Karelia region resembles the two previously published EHG
individuals5 autosomally, but surprisingly belongs to Y-chromosome haplogroup J usually
associated with Near Eastern populations (Supplementary Data Table 1)."


page 43

Kurti said...

Lank

"
The EHG J male looks similar to the other EHG samples in the PCA and ADMIXTURE, so he presumably lacks Basal Eurasian. This is not consistent with ancestry from groups resembling "teal", neither is the fact that his J lineage wasn't assigned to a subclade."

All the EHG samples had some weird ~6-8% Gedrosia type ancestroy. Some of them 25% Teal (Armenian like by the paper) ancestry. Look at the paper mate.

a said...

Kurti said...

"Yamna looks 50-60% "Armenian like" and 40-50% EHG like

Samara H&G(EHG) look 75% "EHG" and 25% "Armenian" like.

Sintashta looks 40% Anatolian(Western) farmer and 60% Yamna (50/50 teal/EHG)

Andronovo on the other hand 80% Yamna and 20% Anatolian farmer.

Afanasievo ~55% Armenian like and 45% EHG

Poltavka similar to Afanasievo.

page 21
http://www.biorxiv.org/content/biorxiv/early/2015/10/10/016477.full.pdf"

How many R1b samples come from Anatolia? 0%
How many J samples come from Steppe Kurgan 0%

What else would you like to show?

MOCKBA said...

It's well documented that steppe males were taking wives from near the Caucasus

and generally the PIE terms of kinship strongly suggest that the brides were taken from afar. There are lots of terms for the husband's relatives but nothing specific at all for the wife's blood relatives or for in-laws in a bride's new family. Moreover, the word for wife-of-a-brother-of-the-husband has the hallmark of an age-status group uniting wives of the "brothers" in a clan or band - think "immigrant community". It's as if the maiden's relatives ceased to exist after she was married off. And even where patrilocality is lost (Hittites), the very name of a son-in-law coming to live with the in-laws indicates that it is an unusual situation (the meaning of the word is essentially "accepted into").

On the other hand, words for wife-trading are as common as for wife-stealing, and there are penalties for violent wife-stealing in the early IE codices. Avuncular is widespread (including Hittite ruling family), and Omaha-like terms for in-laws are widespread. Perhaps it denotes specifically cross-cousin marriages but at the very least, it signifies a dual-society system where daughters are married off and then their daughters return as brides to the group from which their mothers hailed. In other words, the linguistics implies that trade-like bride-taking from related clans was the norm (even if specifically cross-cousin marriage perhaps wasn't).

What can we learn about it from the genetic data? Are there South-North clines of farmer ancestry (as expected if most of the brides were exchanged on dual-society exogamy basis)? Are there long tracts of farmer DNA (if, instead, most of it came from the full-bloodied farmer women stolen in raids)? What about homozygocity tracts (critical for deciding the cross-cousin question?)

Kurti said...

a said

"Catastrophic failure grips the Anatolian & Iberian camp, as they try to figure out the reason as to why no R1b is showing up in the regions. Let the creative explanations begin.What do they have say now?"
Mate you seem to have missed the 5500 BC Neolithic R1b1 sample from
Iberia.

Dude Manbro said...

J2a was present already in Barcin 6500-6200 BC. So that certainly makes this Y-DNA J EHG interesting. Perhaps some of the bloggers can give us more details in the near future.

a said...

@Kurti
Sorry mate.
Your R1b Teal explanation is dead. Your going to have to get real creative to come up with a good explanation as to why/and how R1b managed to stay in an elite- burial in the same region for 2000-3000+/- years.

Dude Manbro said...

Kurti, that R1b1 sample was previously published in the Haak paper was shown to be positive for a V88 equivalent by smal and others.

Lank said...

"All the EHG samples had some weird ~6-8% Gedrosia type ancestroy. Some of them 25% Teal (Armenian like by the paper) ancestry. Look at the paper mate."

Their lack of Basal Eurasian means the minor "teal" is probably indicating a steppe contribution to teal, rather than a teal contribution to EHG.

Kurti said...

Krefter said
"
Also, the Eneolithic guys from Samara probably didn't get their R1a and R1b from the Near East. This is because before Teal ancestry appeared in Russia, R1a and R1b was already there. They're more evidence of the association between Y DNA P(Q and R) with ANE. The R1b1 that existed in Russia in the Mesolithic may have mostly died out and R1b1a2 may be from the Near East, because it looks that way in modern Y DNA. We need more ancient Y DNA to know for sure."

Mate I know it burns deep down in all of your Wive theory supporters hearts but you still failed to show me one freakn single Basal R1a or R1b in EHG groups. All I see are upstream clades. ANd I think you guys should improve your reading skills. I said SOME of the R1b in those EHG samples might be from further South coming with the J samples. And again you seem to have overseen that Samara_Eneolithic EHG samples (to which this J sample belongs) had 25% Armenian like ancestry.

page 21 in the pdf

a said...

@Kurti
"Mate you seem to have missed the 5500 BC Neolithic R1b1 sample from
Iberia. "

Yet not one sample from Anatolia.None -nada zero. Whats going on?
I tell you it's complete catastrophic collapse of your theory. I would be clutching with an Iberian sample too; if it wasn't so funny.

Lank said...

The Samara Eneolithic samples are not EHG...

Dude Manbro said...

Kurti, you are mistaken. This J individual is not from the Eneolithic Samara. It is from Mesolithic Karelia.

Kurti said...

a said

"How many R1b samples come from Anatolia? 0%
How many J samples come from Steppe Kurgan 0%

What else would you like to show?"

Mate you make no sense. Can you read or are you not following the data published in the past year.

There is a freakn R1b1 sample in 5500 BC EEF sample in Iberia.

Do you have any South_Central Asian or Iranian Plateau samples yet that you can exclude it.

Kurti said...

@ Dude ManBro

Even better. Another indication that there was a very early South to West movement So if J is in Mesolithic EHG which OBVIOUSLY came from South as it was found among other farmers. Why shouldn't some of the R1b also have arrived with it?

I know some people are on the edge of heart attack reading my comments :D

Dude Manbro said...

Kurti, once again, the 5500 BC sample from Iberia was proven to be positive for a V88 equivalent (R1b1c). It is not ancestral to R-M269.

Dude Manbro said...

I'm not saying that the J did not come from the South, but it does not appear to have anything to do with the "teal" people as it is similar autosomally to the R1b Samara HG and the R1a Karelia HG from the previous paper and from 5500 BC.

Kurti said...

"On the other hand, words for wife-trading are as common as for wife-stealing, and there are penalties for violent wife-stealing in the early IE codices. Avuncular is widespread (including Hittite ruling family), and Omaha-like terms for in-laws are widespread. Perhaps it denotes specifically cross-cousin marriages but at the very least, it signifies a dual-society system where daughters are married off and then their daughters return as brides to the group from which their mothers hailed. In other words, the linguistics implies that trade-like bride-taking from related clans was the norm (even if specifically cross-cousin marriage perhaps wasn't).
"

Since most people here know the Indo European culture only from books because it is not practized by them (not trying to start here who is the Indo European war here) let me explain how "wive traiding" works in Indo European cultures.

The wive becomes part of the family of the husband. The wives family gets something in revenge for losing their daughter to another clan (in ancient past probably domesticated animals and in recent past gold or money). Kidnapping was rather a rare thing. If it happened it was very dangerous It could lead to clan wars, therefore as revenge the clan who lost a girl by kidnapping from a different clan gets a girl from this clan as revenge. Wives were often taken from nearby communities. Thats at least how it worked just few decades ago among my people. And this is how it worked among INDO EUROPEANS.

No one freakn knows how the EHG groups handle it since obviously they are not the PIE.

a said...

Kurti said...

Mate you make no sense. Can you read or are you not following the data published in the past year.

There is a freakn R1b1 sample in 5500 BC EEF sample in Iberia.

Do you have any South_Central Asian or Iranian Plateau samples yet that you can exclude it."

Oh believe me it makes perfect sense. 0% percent R1b from Anatolia in this paper; versus 13-15% R1b range in modern Turkey, majority most likely R1b 2105+.

Anyway why so much copper in R1b Samara graves? Is that some Iranian tradition?

Kurti said...

a said

"Sorry mate.
Your R1b Teal explanation is dead. Your going to have to get real creative to come up with a good explanation as to why/and how R1b managed to stay in an elite- burial in the same region for 2000-3000+/- years. "

Why prance and skip the R1b in neolithic Iberia now? I thought there wasn't any R1b connected to EEF? What happened mate?

Owned I would say ;)

Chad said...

I think what we have with this teal, is what I said before. It is Anatolian Neolithic, plus native Caucasus hunters. It can't be Armenian or Georgian-like, without at least 30% Anatolian input. I will demonstrate this, with my new calculator. Here is why...

There is archaeological evidence of WHG admixture into both Karelia, and Samara. Both inputs came from the Swiderian folks. These groups never made it further south, into the Caucasus. So, the native hunters in the Caucasus will be closer to MA1, than EHG, plus they will probably have a little of this so called Basal Eurasian. I think this will end up being true, once we have a Mesolithic Caucasus sample.

What I am seeing, is Yamnaya being 20% EEF like, plus about 10% SW Asian, along with about 60% from a group closer to the non-ENA part of Native Americans, with minor WHG, thanks to input into EHG. Once I have those Anatolians, this will be much easier to show.

The Q1 in Khvalynsk also shows that geneflow from the east could've continued, making a hunter component that is closer to Native Americans. Which is also why I think this red Native American component shows up.

Kurti said...

Dude Man Bro said

"Kurti, once again, the 5500 BC sample from Iberia was proven to be positive for a V88 equivalent (R1b1c). It is not ancestral to R-M269."

Thats not important for the argument is it? I didn't said it wasn't R1b v88 (which is very close to basal anyways)

a said there is no R1b in Neolithic context, I disproved that.

However the J ended up in EHG is a mysterium but it is also found in EEF and Teal context therefore from South. So if J with no " evidence of Southern admixture" is there. Why should some R1b with no evidence of Southern admixture no?

If you know what I mean. I think there is still allot to learn and I am convinced SOuth_Central Asian and Iranian Plateau samples will shatter the world of some people here, when suddenly some very EHG/Teal like groups appear there :D

a said...

Kurti said
"Owned I would say ;)"

Oh yeah, no R1b in Iberia in this paper either,L51 or Z2103 v88. What else is left to say? Your going to have to accept the fact that only two of the main branches left the Steppe and went South L277 and L584, 9219 stayed behind with R1b m73, make sense, good.

Krefter said...

@Kurti,
"And again you seem to have overseen that Samara_Eneolithic EHG samples (to which this J sample belongs) had 25% Armenian like ancestry."

They aren't EHG. They are intermediate between EHG and Yamnaya. They had "25% Armenian-like" ancestry.

Chad said...

Kurti,
We get what you're saying, but it doesn't make sense. Latching on to one sample, is not going to work. You might as well go extreme and point to Africans that show little West Eurasian and have R1a and R1b, then say they're African markers. Come on now. You can't look at a long isolated lineage and make generalizations for the whole haplogroup.

Chad said...

We have full Anatolians and full WHGs with I, EHG and Anatolian with J... things happen over thousands of years. People weren't stationary. M269 didn't go from Iran to Germany in 3000BCE. That's fantasy land.

Grey said...

"Even better. Another indication that there was a very early South to West movement So if J is in Mesolithic EHG which OBVIOUSLY came from South"

There's an assumed association now but is there any actual reason J had to come from the south originally?

Nirjhar007 said...

Autosomal Structure is a subject of massive change, i think we give too much credit to it...

Nirjhar007 said...

@Rob
Can someone make sure that Nirjahar doesn't read this
I'm on it, don't you worry ;).

Nirjhar007 said...

Just need Some aDNA from SC Asia-Iran now to seal the Z-93 issue.

Aram said...

The position of Anatolian farmers on the PCA is amazing compared to the European Neolithics. It is not moving toward Bedouins but it is "moving" toward Cypriots.

Kurti

Congrats! Your theory about missing clusters on the PCA seems to be correct. The IA Scythian is on a "empty" place on modern populations PCA. Well it is right on the place where other Bronze Age Steppe samples are. It seems the Turkish invasions and later the Slavic expansion had a dramatic impact on the Steppe populations.



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

It will be interesting to see when the natural selection sweep that established SLC45A2 and Lactase persistence as (for the most part) fixed alleles occurred.

Rob said...

The Eneolithic mtDNA from Khvalysnk are U5a, U4, and H2a
These appear to be "east European"

(Der Sarkissian found H in Karelia )

Krefter said...

@Romulus,

Something interesting about the SLC45A2 SNP is that was a huge rise in frequency in all of Europe, even Caucasus. It's rise in frequency tied to a specific type of ancestry Everywhere from Greece to Ireland 80%+ of the people are homogeneous. But it's hard to say how great of an affect it has.

Rob said...

Krefter
You mean homozygous ?

Anonymous said...

So, no one's gonna comment on Bell Beakers clustering with today's French by one side, and Tuscans by another? Or on Bell Beakers already showing "Italo-Gaulish" R1b1a2a1a2b? Or Nordic bronze age sample having "Germanic" R1b1a2a1a1? Or regarding the R1b-v88, which entered Subsaharan Africa from Egypt, being present in neolithic Iberia, but not Anatolia neolithic?

All those subclades appeared quite recently. They most likely ultimately originate from the extinct North Eurasian population that had undifferentiated R, and already had many of the mutations (probabily as far as R1a1a and R1b1a) before moving southwards into a quite wide range. De to their dating, those R1b1b lineages of mesolithic and samara steppe are likely dead ends, and don't tell us much about the origin of the R1b1a2 present in Yamna culture; all they indicate is that those steppe mesolithic EHGs had early contact with ANEs (which does give them statistical advantage as the origin for Yamna male lineages). Kinda hard to speculate about the origins of R1b1a2 without genetics from adjacent areas.

Nirjhar007 said...

Eastern Side Anatolian Samples, when they will come up?? come on!!....

postneo said...

positive selection for light skin will be driven not just by latitude/sunlight but also on elevation. cold temperatures result in less exposure of skin.

Davidski said...

Hayarotle Hayarotle,

The Bell Beakers don't cluster with Tuscans. They're being pulled out of position by projection bias.

When I get the samples I'll show you where they really cluster.

Davidski said...

Aram,

The Neolithic Anaotlians aren't moving towards Cypriots. They're being pulled towards the middle of the plot like all of the ancient samples because of projection bias.

"Principal component analysis of modern West Eurasian samples (grey), with ancient samples projected onto the first two principal component axes and labeled by culture."

Ebizur said...

"In eastern Europe outside the steppe, a new individual from the Karelia region resembles the two previously published EHG individuals autosomally, but surprisingly belongs to Y-chromosome haplogroup J usually associated with Near Eastern populations (Supplementary Data Table 1)."

Please keep in mind the "topology" of the phylogenetic tree of extant human Y-DNA. Haplogroup J2 (the MRCA of J2a-M410 and J2b-M102) is not significantly different in age from haplogroup I (the MRCA of I1-Z2698 and I2-Z2667).

YFull (accessed Oct. 11, 2015)
TMRCA (J2a + J2b): 27800 [95% CI 25700 ~ 30000] ybp
TMRCA (I1 + I2): 27500 [95% CI 25200 ~ 29800] ybp

In other words, it is plausible that even a pre-J2a or a pre-J2b individual may have lived in Gravettian Europe alongside (or at least not too far from) Mr. proto-I. It is even more likely that Mr. proto-J2 and Mr. proto-I were roughly contemporaries, though their most recent common direct patrilineal ancestor had lived approximately 15,000 years before that time.

The popular association between Y-DNA haplogroup J and Southwest Asian Neolithic people seems to have blinded many people to the fact that haplogroup J is a relatively old haplogroup.

On the other hand, the basal phylogenetic position (within F-M89) of haplogroup G seems to have made many people fail to notice that G-M201 is one of the youngest of the major haplogroups:

TMRCA (G1 + G2) 26600 [95% CI 24500 ~ 28600] ybp

H (and H1-Z4176), IJ (and J), and K (and NO and P) are all significantly older (i.e. internally diverse) than G. Haplogroups G, J2, and I do not significantly differ from one another in age, and all date roughly from the era of the Gravettian industry in Europe (though none of them necessarily has any direct connection with the Gravettian or Europe).

Anyway, discovery of a pre-Neolithic example of Y-DNA haplogroup G-M201 would be really interesting.

Krefter said...

I updated my ancient mtDNA and Y DNA spreadsheets with the data in this study. BTW, the Poltavka samples are all from Samara.

Y DNA
https://docs.google.com/spreadsheets/d/12G2cfjG0wHWarsl5bB99ridFmvUWzqlZfZ6_e_R6oIA/edit#gid=1665098932

mtDNA
https://docs.google.com/document/d/1HRfrKlDa5Z0Gd1SdeC-y4qgZa7lOrFqBcuxptgZ8HSA/edit

Onur Dincer said...

Kurti:

The wive becomes part of the family of the husband. The wives family gets something in revenge for losing their daughter to another clan (in ancient past probably domesticated animals and in recent past gold or money). Kidnapping was rather a rare thing. If it happened it was very dangerous It could lead to clan wars, therefore as revenge the clan who lost a girl by kidnapping from a different clan gets a girl from this clan as revenge. Wives were often taken from nearby communities. Thats at least how it worked just few decades ago among my people. And this is how it worked among INDO EUROPEANS.

I do not think the marriage customs of PIE, who are assumed to be a Late Neolithic nomadic pastoralist polytheist people from the Pontic-Caspian steppe, would be much similar to those of Kurds, a largely agriculturalist and largely Muslim people from the Near East.

jackson_montgomery_devoni said...

lol Y-DNA J in a EHG sample from Karelia is huge!!!!!!!

Aram said...

Davidski

Ok I inderstand. I learned a lot off things from Your blog. Thank You very much.

Katharós said...

@Onur

Well, other hominids exhibit the same behaviour.

Gombe Chimpanzee War
https://en.wikipedia.org/wiki/Gombe_Chimpanzee_War

Anonymous said...

@Chad

"I think what we have with this teal, is what I said before. It is Anatolian Neolithic, plus native Caucasus hunters. It can't be Armenian or Georgian-like, without at least 30% Anatolian input."

Archaeologists consider Sredny Stog a predecessor to Yamnaya. They are supposed to be heavily influenced by Cucuteni-Trypillian culture. There is a whole area, north of the Black Sea, of the Yamnaya horizon not sampled yet. Maybe samples from there will show different admixture.

http://s155239215.onlinehome.us/turkic/btn_GeographyMaps/BC4500-3500EneolithSrednyStog.gif

Onur Dincer said...

Creative, your example is supportive of David's hypothesis, not of Kurti's hypothesis, but an example chosen from chimpanzees has the problem of representativity of humans.

Krefter said...

Some useful links. I posted a few earlier. You guys should safe all of this somewhere incase you haven't made something similar yourself.

List of Mathieson samples.
https://drive.google.com/open?id=1Vjbp450AwI7R-Y9J1YGSm9FjJWu9s9lx1azjUJbS8hQ

Extended Data Mathieson useful information
https://docs.google.com/spreadsheets/d/14xg-GQX6HwzMTtw4mcnvkStLT8CA87z5qvcCLWCcCcE/edit#gid=2094157768

Updated Ancient West Eurasian Y DNA.
https://docs.google.com/spreadsheets/d/12G2cfjG0wHWarsl5bB99ridFmvUWzqlZfZ6_e_R6oIA/edit#gid=479090567

Updated Ancient West Eurasian mtDNA.
https://docs.google.com/spreadsheets/d/12G2cfjG0wHWarsl5bB99ridFmvUWzqlZfZ6_e_R6oIA/edit#gid=479090567

Krefter said...

In terms of Selection: These SNPs had the strongest signal. Just as in ancestry, LN/BA Central Europeans/Asians are a big step forward to being like modern Europeans.

Lactose Persistance(rs4988235) first appears with LN/BA Central Europeans and Asians but was unpopular. In *some* parts of Europe today it is much more popular.

Pale skin color(rs16891982) was decently popular in Mesolithic Sweden and Neolithic Anatolia. Rose in frequency with LNBA Central Europeans Asians. Is at near fixation in most of Europe today.

rs4833103(Immunity) was non-existent in Mesolithic Europe and Neolithic Anatolia. At similar low frequency in all other ancient Europeans. At very high frequency in all modern Europeans.

Kurti said...

Arame said

"
Kurti

Congrats! Your theory about missing clusters on the PCA seems to be correct. The IA Scythian is on a "empty" place on modern populations PCA. Well it is right on the place where other Bronze Age Steppe samples are. It seems the Turkish invasions and later the Slavic expansion had a dramatic impact on the Steppe populations."


Yeah, thanks mate. But you know my theory are always bogus until proven to be right. Some heros here as well in Eupedia said I was nuts and crazy

But I used logic and said Northeast Iranic tribes would be the missing link between Europe and West_SouthCentral Asia and the Turkic and later Slavic expansion into the region.

But it's always the same I am the bad guy until proven unguilty :D



It's always the same

Kurti said...

For those who don't know what I am talking about.

http://www.eupedia.com/forum/threads/30706-Europe-West-and-South_Central-Asia-and-the-unnatural-gap

Davidski said...

That Scythian will end up very close to Erzyas when not projected IMO.

By the way, none of these non-Neolithic samples look particularly West Asian or even Southern European. Heck, even the Neolithic samples don't.

You guys will have to wait a little bit for your main ancestors to show up in ancient DNA. At the moment it's hard to say where they were exactly. Maybe Mesopotamia and further east?

Kurti said...

Onur

"I do not think the marriage customs of PIE, who are assumed to be a Late Neolithic nomadic pastoralist polytheist people from the Pontic-Caspian steppe, would be much similar to those of Kurds, a largely agriculturalist and largely Muslim people from the Near East"

I don't think you have much knowledge about Indo European rituals do you?

Starting at 4:04
It not only resembles Kurdish culture it's like 1:1 adaption.

https://www.youtube.com/watch?v=ErXa5PyHj4I

Kurti said...

Take in mind Onur

Muslim societies are patriachal and similar to PIE cultures.

And could you please explain me the difference between late Neolithic pastoralist (agricultural system) nomads and early neolithic farmers?

I am not aware of any, at least no study found which states much differences.

Krefter said...

It's weird that R1a-Z94 exists in Yamnaya-like Poltavka in 2925-2536 BC. That's 500-900 years before Sintashta and Potapovka who had significant EEF ancestry. Does this mean R1a-Z93 is very old and went through a founder effect in Sintashta/Potapovka/Srubnaya ancestors somewhere further West?

Burgos Spain Y DNA that is contemporary to Corded Ware is all I, G2a, and H2. R1b-P312 obviously hadn't expanded yet, at least not in the cave they were from. Also, in the paper they proved with F4 stats Basque and Spanish have Steppe-related ancestry. R1b-DF27 is a post 2800 BC founder effect and we know Steppe ancestry was in Burgos by 1500 BC. IMO, all that is needed at this point is Y DNA from Ukraine and western Russia dating between 3500 and 2500 BC.

Kurti said...

Chad said

"
"I think what we have with this teal, is what I said before. It is Anatolian Neolithic, plus native Caucasus hunters. It can't be Armenian or Georgian-like, without at least 30% Anatolian input.""


Chad I don't think anyone claimed something different. At least I always said

Teal is simply ~60% EEF merging with 40% ANE from the Iranian Plateau and Southeast Caucasus. I expect the North Caucasus to be even more EHG admixed. Something on line with ~70% Teal and 30% EHG.

So yes I kinda agree with your theory.

Rob said...

Yes
I think R1b in Iberia is a post -BB phenomenon!
ANE and R1b arrived after the copper age, during the true Bronze Age.

Nirjhar007 said...

I think its impossible that Z-94 Came from West, Its East.

Davidski said...

Very funny. Now go and have a look where Z94 Poltavka outlier clusters.

Kurti said...

David said

"
By the way, none of these non-Neolithic samples look particularly West Asian or even Southern European. Heck, even the Neolithic samples don't."
No one claimed any of the samples being typical modern West Asians?

What I and Arame said and what the ancient data seems to confirm is that ancient Scythians and North Iranic tribes were a link between modern Europeans and West_SouthCentral Asians so halfway modern West Asian and halfway East European like. That Turkic and later Slavic expansion created a gap of two opposing groups. Thas how it looks All cultures from Yamna to Sarmatian period end up somewhere between North Caucasians, Central Asians and East Europeans.

Davidski said...

That Scythian doesn't look half West Asian. He looks Eastern European.

The reason he's being pulled down is because of projection bias.

Kurti said...

Oh and do not worry about me or our ancestors. I know perfectly where I am from. And it would lead to plain absurdity to believe any modern population could be 1:1 genetically derived from one ancient group.

Nirjhar007 said...

H2a1 is also interesting, see here: http://www.forumbiodiversity.com/showthread.php/13338-H2a1
Assyrian.
http://www.eupedia.com/forum/threads/29415-Were-mtDNA-H2a1-I-R1a-and-W-the-haplogroups-of-the-Maykop-culture
Interesting quote, its similar to what i'm trying to suggest.
H2a1, I and W then all suddenly show up in both the Corded Ware and the Unetice culture in Central Europe during the Bronze Age. This makes me wonder whether they were original R1a lineages from the forest-steppe, or if they represent an ingression of Northwest Caucasian people (Maykop culture) into the R1a steppe population.

Kurti said...

"That Scythian doesn't look half West Asian. He looks Eastern European.

The reason he's being pulled down is because of projection bias."

Of course it's projection bias, the Iron Age Scythian is actually a Polish-Lithuanian guy in disguise :)


I am out for now.

Nirjhar007 said...

But he does not understand apparently that H2a1 and I and W come from Iran or nearby :(...

Aram said...

Davidski

My comment was a remark on the hot discussion on Eupedia. Actually I don't think it can be used as an argument for the overall PIE umerheit question, because it is a very late sample ( 300BC), with very different social dynamics. I just wait some more southern aDNA to have an definitive opinion. Till now the Steppe theory is quite strong.
It was just amazing to see where was that Scythian. Btw Afanasiev claims that his Alanian samples also look very European. And this Alanians ( the ancestors of modern Ossetians who fled to mountains after the Mongolian invasion ) also could be somewhere near that Scythian.

a said...

Kurti said...

"Oh and do not worry about me or our ancestors. I know perfectly where I am from. And it would lead to plain absurdity to believe any modern population could be 1:1 genetically derived from one ancient group."

I think it is great that members of R1a community can debate about their ancestral origins. I agree with Eurogenes.
I rarely, if ever get involved in R1a origin debates other than when it is connected with R1b split.
However, I too know where I am from. That's why, when people/hobbiysits, who are not R1b and in particular not even remotely related to 5000B.P. +/- Kurgan R1b-Z2105..... try and lecture/pontificate about my ancestral origins explained in the above scientific results, by trained professional's; crack me up :)

Hopefully Eurogenes can run the samples through the same set of runs using K6-K10 and K15 :)

Va_Highlander said...

Rob:

"Can someone make sure that Nirjahar doesn't read this"

I think that cat slipped out of the bag with Razib's tweet.

For my part, I'm almost surprised that genetic analysis seems to be supporting a westward spread like this.

Nirjhar007:

"Just need Some aDNA from SC Asia-Iran now to seal the Z-93 issue."

How likely is that in the near future?

Nirjhar007 said...

Very.

Coldmountains said...

Seriously everyone who think Srubnaya got EEF from Central Asia is on drugs. An origin of Z93 in Bavaria is more likely now than in Central Asia. But Ukraine or Southwest Russia are my guesses and that is not far away from CT farmers from which Proto-Indo-Iranians took some wifes from to get their EEF admixture.

Nirjhar007 said...

Relax, just wait for the aDNA, Z-93 came from East.

Va_Highlander said...

Coldmountains:

"Seriously everyone who think Srubnaya got EEF from Central Asia is on drugs."

I'm sure there are other, less pejorative explanations. An east-to-west movement was unlikely from the outset, given the historical behavior of populations on the steppe, environmental considerations, and so on. This new evidence is what I should expect to see.

Still early days, though, and sometimes even unlikely outcomes nonetheless come to pass.

Va_Highlander said...

Va_Highlander:

"An east-to-west movement was unlikely from the outset..."

Whoops. Better make that, "west-to-east".

Unknown said...

The J2a and the G-L91s, G-PF3146 and G-P03s and G-CTS342 in NW Anatolian Barcin is exciting! 1) It confirms how widespread L91 was during the Neolithic. 2) J2a and not J2b is in NW Anatolia from the Early Neolithic. Perhaps it will be J2a-M67 which is frequent there now. What is useful about NW Anatolia is the presence of painted pottery (and J2a) in the region. LBK and Cardial pottery is incised/impressed. It will be interesting to see where the Anatolian Neolithic is shifted towards--Cyprus/Levantine/Bedouin.

Kurti said...

a said

"

I think it is great that members of R1a community can debate about their ancestral origins. I agree with Eurogenes.
I rarely, if ever get involved in R1a origin debates other than when it is connected with R1b split.
However, I too know where I am from. That's why, when people/hobbiysits, who are not R1b and in particular not even remotely related to 5000B.P. +/- Kurgan R1b-Z2105..... try and lecture/pontificate about my ancestral origins explained in the above scientific results, by trained professional's; crack me up :)

"

So according to yourself someone who has no idea about Haplogroups and DNA at all has more to say about "his own Haplogroup" than some schooler?
I as an R1a individual am perfectly fine with non R1a's commenting on it, IF what they say makes sense. Also I comment on R1b because My people are among the few worldwide popultion to have anything from the Basal form m343 (yet to be seen among EHG just for the account, not saying it didn't exist) to Yamna l23 and everything in between (m269) .

So anyone with knowledge is welcome to debate about R1a even though they don't belong to this Haplogroup themselves.

Kurti said...

Arame said
"My comment was a remark on the hot discussion on Eupedia. Actually I don't think it can be used as an argument for the overall PIE umerheit question, because it is a very late sample ( 300BC), with very different social dynamics. I just wait some more southern aDNA to have an definitive opinion. Till now the Steppe theory is quite strong.
It was just amazing to see where was that Scythian. Btw Afanasiev claims that his Alanian samples also look very European. And this Alanians ( the ancestors of modern Ossetians who fled to mountains after the Mongolian invasion ) also could be somewhere near that Scythian."

What some people don't seem to understand, is that North Caucasus is technically East Europe and Teal genes since at least around the Neolithic are European genes.

If Scythian individuals end up somewhere in between Caucasians (East Europe/West Asia), Central Asians and East Europeans. Than this makes them certanly "East Europeans also. Even if they are genetically half way Near Eastern/West Asian like, heck going by this paper even Sintashta was ~70% in total EEF/teal farmer/herder like.
You would assume by now, seeing all that drastic genetic shift in all Eurasia some people would understand that there is no European no West Asian or Central Asian genes what so ever. If population A is halfway Neolithic European and halfway Mesolithic European, that just makes them Neolithic/mesolithic Euros, because they are located in Europe. If all of Europe was today replaced by new waves of Africans, what would we call them, if not European, since they are living in Europe.

However what some people yet seem to miss out as often. No matter if Neolithic, Mesolithic or Bronze Age at the end of the day all of them are different waves of migrats from the Near East. Graviterian probably arrived directly through the Caucasus.

But it seems like some people have problems in accepting even this and want to portrait the North Iranic groups as "Ultra-East Europeans" identical to modern Balto_Slavs, instead of seeing how it is.

Katharós said...

Onur,I was trying to be sarcastic both ways. Either way you put it, "both ways" exist within hominid society.

Wild Chimpanzees Exchange Meat for Sex on a Long-Term Basis
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2663035/

Our findings also shed light on our current knowledge on meat sharing in humans, suggesting that the increased reproductive success of accomplished hunters compared to unsuccessful hunters in forager societies could be driven by female choice and be linked to direct exchanges of meat for sex between men and women. Future studies on human foragers, which adopt methods used by field ecologists, will allow the testing of this hypothesis.

Kurti said...

Some people seem to not understand that Scythians ending up somewhere in the hap region between Russians, Caucasians and Central Asians perfectly fits their physical appearance based on the data we have (the samples so far show 40% Blonde, 20% Brown and 40% dark hair).

They have this wrong image in their head that this physical appearance must be explained with EHG genes (despite EHG being dark skinned).

If you want to know how 50/50 West Asian/North European groups would look like. Imagine something Georgian or Tajik like mixing with something Russian or Ukrainian like.

Something like Sergei Lawrow or Stalins daughter Svetlana. So now tell me that those individuals don't look like how Scythians would have been described. So for me if scientist say they were European genetically, I imagine something in that direction.

pnuadha said...

@creative

could be driven by female choice and be linked to direct exchanges of meat for sex between men and women

Why is there so much emphasis on female choice? Women have breasts along with other sexual features that are the result of male choice. Behaviorally, modern women spend large amounts of money to make themselves more attractive to men. But somehow its always ignored that women spend money to attract men as if its meant to fit some sort of strange idealization of a 1950's male provider.

As far as I know, the women in hunter gatherer societies are productive in terms of the food they gather; so they aren't a 1950's stay at home mom.

Given that women do spend large amounts of money on attracting men, why wouldn't they also use meat and usable resources to attract men?

pnuadha said...

when i say women spend large amounts of money to attract men, I'm referring to beauty products.

Nirjhar007 said...

good god!.

Coldmountains said...

@Kurti

Scythians are not relevant here. They existed much later and in the case of late Scythians they were also extremely mixed. Nevertheless Sintashta had typical North European pigmentation and was genetically closer to modern Europeans and much more "western" than Yamnaya. They carried a lot of WHG and don't tell me they got it from Central Asia

George Okromchedlishvili said...

Been drinking vodka and signing Russian 90s pop-songs. Pretty sure that's totally and exactly how protokartvelians spent their time 4-5 thousand years ago cause I'm their direct descendant (tm).
Let's get real, Lithuanians and Latvians have way more in common with PIEs than any West Asian group in the world.

As for the whole "PIE from Central Asia!!11" - how are we supposed to reconcile this with the fact that PIE has very little in common with Elamite, Sumerian, Hurrian and many other languages that were around that area for at least several millennia but magically shares a lot with Uralic?

Nirjhar007 said...

Sumerian, Hurrian
A lot actually, especially the first, with Uralic not so much, though it also exist...

Katharós said...

@Colin Welling
Your point is valid, but the outcome is the same; intercourse for a service in return. At least out of the male’s perspective.

Alberto said...

So the Samara_Eneolithic samples are just 3, and the figure of 25% "Armenian" seems to be just a compromise. The 3 individuals seem to be different.

One is almost an EHG. We'll have to wait the get the samples, but it would seem that this one would be the one carrying Y-DNA Q1a and mtDNA U4. His burial was not of someone belonging to the "tribe", but rather someone killed.

Another one might be 25% "Armenian" (maybe some 30% when using Georgian instead). Again, guessing, this one could be R1a and U5a1 respectively. He was part of the group, by the burial.

The last one seems to cluster with Yamnaya, so probably 50%-60% "Armenian/Georgian". This one might be the R1b1 and H2a1. A high status burial containing 80% of the copper objects found in the combined Khvalynsk I and II cemeteries.

None of the samples are C14 dated, only archaeologically. At least he 2 R1a/b ones should be post 4700 BC.

Alberto said...

Re: Srubna samples, it would seem that they might derive from Sintashta?

The West to East migration of R1a seems to have happened through the forest steppe, while the open steppe was R1b dominated. Somehow, the Sintashta people replaced the Afanasievo ones to the east, in the Altai region, and the Poltavka ones to the west, in the Volga region.

We don't really know what happened in the western steppe.

Va_Highlander said...

George Okromchedlishvili:

"As for the whole "PIE from Central Asia!!11" - how are we supposed to reconcile this with the fact that PIE has very little in common with Elamite, Sumerian, Hurrian and many other languages that were around that area for at least several millennia"

Why should they have anything in common? Language doesn't spread like ink on paper. Of those three, only Proto-Elamite might have come into direct contact with core IE and I see no particular reason to think it should have had a significant linguistic impact. The numbers of Proto-Elamites would have been relatively small and they wanted what the tribes of Central Asia had, not the other way around. Unless you can show that it's probable they would have had some impact, and not just possible, there is nothing to reconcile.

"but magically shares a lot with Uralic?"

Geography, I imagine.

Creative:

"Your point is valid, but the outcome is the same; intercourse for a service in return. At least out of the male’s perspective."

It's been suggested that female hominini may have been polyandrous. Human females are rather vocal during intercourse, for instance. Polyandry might fit well with trading meat for sex.

Alberto said...

@George

"As for the whole "PIE from Central Asia!!11" - how are we supposed to reconcile this with the fact that PIE has very little in common with Elamite, Sumerian, Hurrian and many other languages that were around that area for at least several millennia but magically shares a lot with Uralic?"

The questions related to long extinct languages are complicated. Experts cannot agree about these things, and while there are hypothesis linking those languages, there is no consensus. I don't think you really have an informed opinion about such complicated matter, or do you?

Regarding the living languages, there is some more consensus. For example, regarding the Uralic question that comes up often, the most accepted hypothesis is that proto-Uralic has some influence from PIE. But PIE has no influence from Uralic. This probably suggests that PIE developed away from a contact zone with proto-Uralic, and that later it arrived to an area where proto-Uralic developed in that contact zone with IE.

The situation is different with Kartvelian and other Caucasian languages. In this case, it seems that these languages did have an influence in PIE (substrate/superstrate...). This suggests that PIE could have developed in a contact zone with these languages.

Nirjhar007 said...

Alberto,
I have said before, IE-Uralic connections exist but they are a bit exaggerated, than they what really are.
OTOH IE with Sumerian,Hurro-Urartian and Semetic etc are mostly neglected or not studied with much effort...

Grey said...

Va_Highlander

"An east-to-west [west-to-east] movement was unlikely from the outset, given the historical behavior of populations on the steppe"

Does this historical behavior include the Russian eastward expansion from the 1700s?

If it did then the historical record would show the direction of expansion on the steppe depended on who had a numbers / technology advantage at the time.

And the light that would shine on pre-historic expansions might be to ask who had a numbers / technology advantage at any particular time.

Arch Hades said...

Is there not a pigmentation data table in this study like the earlier edition? I'm curious what the pigmentation of the Anatolian farmers is, as well as some of the EHGs.

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

If we have Near Eastern admixture in EHG as early as 5000-4000 B.C. then it pre-dates the Maykop by a fair but, finding J in EHG kind of implies that J might be related to pre-Maykop and that J is the source of the Near Eastern Admix. Also finding the Near Eastern admixture this early means that Steppe people could have brought it in to Europe much earlier than the Yamnaya period, possibly with CT.

Matt said...

Comments:

- Seems really weird to revise the paper in part to correct Haak 2015, but I guess Haak 2015 is published and this is not, so there you go?

- Would have been cool if they could've estimated proportions in modern populations as well.

- On proportions again, it's odd that the estimated proportions of WHG here between MN and EN European farmers only seem to differ by around 10% WHG, while in Haak et al they differ by 20% (http://tinyurl.com/oyg9bok).
On this note, I'll be extremely interested to see what Davidski finds on the D(Ju_hoan_North,WHG;BedouinB,Pop) and D(Ju_hoan_North,EHG;BedouinB,Pop) and D(Ju_hoan_North,MA1;BedouinB,Pop) stats for these new populations. I have some skepticism the qpAdm method is finding the right differences in relatedness.

- For the height selective pattern, there's clearly a big difference between the papers in the "genetic height" effect sizes
http://i.imgur.com/iXf0b7z.png

The effect sizes in Mathieson 1 were clear, in that a size of 1, was 1 SD in height, which is 2.5 inches. It's not clear what the effect sizes in 2 translate to.

If they're set to the same scale, the differences with steppe-farmers are cut to about a third... but then if that were true, it would translate to modern day IBS and CEU being totally indistinguishable in height, while the best analysis of modern populations seems to point to a difference of around 0.5 cm (http://www.nature.com/ng/journal/vaop/ncurrent/full/ng.3401.html) between Spanish and British. Not sure I agree with them that modern heights in Europe (which per the above paper seem to max out at 1cm or 0.5 inches genetic height between Dutch and Italians, which is believable) are reflective of Yamnaya ancestry proportion rather than ongoing selection, though.

- I think the lack of a timeline for alleles is a bit of a shame here, as e.g. the Steppe sample overall has higher frequency of derived SLC45A2 than the Yamnaya did in the previous paper, so it would be interesting to see if this gene had climbed in frequency over time. (Funnily enough, the Anatolian Neolithic sample seems to have a higher derived frequency here than their descendents in Europe).

- The Samara Eneolithic looks like one overlaps Yamnaya, while the other two are intermediate with Yamnaya and EHG. It'll be interesting if the overlapper is the "high status" and high wealth SVP35 (293 copper artifacts), who also had the unique H2a1 mtdna, the lower wealth SVP46 (U5a) or the violent death, no grave goods SVP47 (ydna Q1a, U4a2). Google suggests H2a1 may be related to the Caucasus.

Krefter said...

@Matt,

Many "Steppe" samples included for frequencies were Andronovo, Sintashta, and close relatives. They rose frequency of derived alleles in SLC45A2 and LCT for Steppe. I don't understand why the authors grouped Sintashta, etc. with Yamnaya when we've already seen they have more modern-like results for selected SNPs than Yamnaya.

@Alberto,

Makes sense the high-status Eneolithic guy who clusters with Yamnaya had mtDNA H2a1. This lineage is widespread in Europe today, absent in Neolithic Europe, and first appears with Corded Ware. Looks like a Teal-lineage. All Eneolithic Y DNA is dead today. Looks like we're seeing R1a1* and R1b1* again, and I'm sure R1b-L23 and R1a-M417 were already around by 7,000-6,000 years ago. It's weird. There were massive founder effects in Y DNA after 6,000 years ago in Russia and Ukraine.

George Okromchedlishvili said...

@ VA_highlander , Alberto

I've read several different papers written by Nostratic scholars and other linguists. None imply that Elamite or any other MENA language is closer to PIE than Uralic. In fact they all group IE, Uralic and Altaic into one tight subgroup. Both from the point of morphology and glottochronology IE and Uralic are very close. And it's not simply about borrowed words we have the correspondence between very basic elements of language that implies much deeper relation.
I'm afraid that's not the case with say Kartvelian. There are identified borrowing into Proto-Kartvelian from PIE but almost none in the other direction.

Rob said...

pIE a "tight cluster" with Altaic and Uralic ? ?
Ridiculous
People just don't understand language change and how language families form.

Matt said...

@ Krefter, yeah, I do agree on their grouping strategy. Presumably this is to get higher sample sizes to try and blur out differences from chance, but I hope it doesn't lead to too much confusion where people argue that there are differences between steppe and European farmer populations at the same time (e.g. in pigmentation, lactase), which either may not actually exist or be as pronounced as they might seem.

@Alberto: The West to East migration of R1a seems to have happened through the forest steppe, while the open steppe was R1b dominated.
To my very, very amateur perceptions, Srubna is known as the "timber grave" culture, suggests a parallel there. Mallory's IE culture encyclopedia notes that presence of the pig indicates a mixed agricultural-stockbreeding culture at Srubna settlements, as a contrast to Catacomb / Yamnaya / Poltavka.

Maybe Srubna genesis then includes some mixed agricultural culture from a forest zone, where people log more and build more with wood (maybe more of a minimal practice among cultures more focused on stockbreeding, where this happens less except as to build and maintain wagons which were essentially steppe tents mounted on a wagon base). Possibly they had the right "mix" of mobility and settled farming skills for the time and place, better at holding territory, better at surviving (then their descendants may have had to shift again as the technological wavefront and climate changes again).

capra internetensis said...

Unfortunately there are only 3 Khvalynsk men and again we don't know what subclades they are. That R1b guy could be the L23 patriarch himself or just another R1b-M73*. The R1a could be ancestral to Corded Ware R1a or some extinct clade. There are not enough samples to know what minority lineages may have existed.

It does confirm (what was already obvious) that there was a huge founder effect and that it came during or after the Khvalynsk period. These guys are probably dated 4700-4000 BC which falls right in the highest likelihood interval of M269>L23>Z2013 from Y-Full (middle of the 5300-3500 BC 95% CI). So the wealthy R1b Khvalynsk guy really could be one of (or close kin to) the founders of the incredibly successful L23 clan.

Krefter said...

Can't wait till geneticker gets his hands on the Khvalynsk Y DNA. I suspect we're looking at three dead ends. The R1b1 I think will be R1b1a2 or R1b1a1. But nonetheless a dead end.

Tobus said...

@Kurti:despite EHG being dark skinned

EHG were light-skinned with both SLC24A5 and SLC45A2, lighter than EEF who didn't get SL42A5 until it came in from the steppe. It's WHG who were dark-skinned.

@Colin:women spend large amounts of money to attract men

I think this is really a modern phenomenon. My (admittedly limited) understanding is that in hunter-gather societies generally it's the male who dances, dresses well, presents gifts, jumps highest etc. etc. to attract the female. It's only relatively recently in urbanised societies that the roles have been reversed somewhat.

Tobus said...

Grrr: "...who didn't get SLC45A2 until..."

Matt said...

Looking at the D-stats, it seems really striking how the stats for the set D(Anatolia_Neolithic,AncientPop,WHG,Chimp) are fairly substantial, but don't appear to translate into that much admixture. They're like
D(Anatolia_Neolithic,LBK_EN,WHG,Chimp)= -0.00114, translates into around 7% admixture
D(Anatolia_Neolithic,Hungary_EN,WHG,Chimp)= -0.00212, translates into around 11% admixture
D(Anatolia_Neolithic,Iberia_EN,WHG,Chimp)= -0.00244, translates into around 7% admixture (note this seems abberant to the other two).
At the same time
D(EHG,Yamnaya_Samara,Armenian,Chimp)= -0.00191
which is supposed to translate into around 50% ancestry from an Armenian like source.
Similarly D(EHG,Yamnaya_Samara,Armenian,LBK_EN)=-0.00080, statistically significant, but pretty weak.
The stat D(EHG,Yamnaya_Samara,MA1,Chimp)= -0.00513 is really quite strong, in that Yamnaya_Samara really is further from MA1 than EHG by quite a bit. But Yamnaya_Samara does not seem to be that much closer to Armenian.

@ Tobus, re: SLC45A2 in Figure 3 of the revised Mathieson paper, derived SLC45A2(rs16891982) looks around 0.35 in Early Farmer CEM (Early to Mid Neolithic Central European), vs around 0.4 in Early Farmer AN (Anatolian Early Neolithic), then vs 0.65 in the whole steppe sample and 0.7 in LNBA Europeans.

While Mathieson et al Version 1 gave the Samara Yamnaya as having a frequency of derived SLC45A2(rs16891982) of 0.4. Allentoft et al gave the Bronze Age Steppe sample as around 0.3 for SLC45A2, less than the 0.4 frequency they gave Neolithic Europe. Mathieson 1 also gave SLC45A2 in Swedish HG as 0.7.

So of course they all had derived SLC45A2 before it came in off the steppe (and it seems in some populations in higher frequencies than some steppe populations at the same time). All these Mesolithic and Neolithic populations should all have been darker pigmented compared to modern day Spanish or Italians, of course, from what we know.

pnuadha said...

@tobus

dude, all the makeup women buy looks pretty deep seeded. But here are other examples of less modern societies modifying the body of women. http://jasminewillis-wildkingdom.blogspot.com/2012/10/body-modification-within-tribes.html

The most obvious tell tale sign of sexual selection on women are B cup breast and upwards which seem to serve no purpose other than attracting men. The fact that women have redder lips and lighter skin than men might suggest the same.

Tobus said...

@Matt:

Yes, I was simplifying - SLC45A2 was at low frequency in EEF before the Steppe incursion, not completely zero. It rose to it's current 80-100% levels after (because of!?) the Steppe influx - which is why the paper identifies it, but not SLC24A5 (which was already basically fixed), as a selection target. It's interesting that this new version shows a clear East-West cline in the farmer samples, cementing the idea that it developed outside or at eastern edge of Europe. It's a shame they combined the SHG/WHG samples and left out EHG, as 50% for HG is quite misleading (WHG=0, SHG/EHG=~75).

In the Haak data EHG are fixed for SLC24A5 and have SLC45A2 at 75%, similar levels to Motala but n=2, so the new Karelia sample could change that by a fair amount. I think Motala and EHG skin colour would be roughly similar to modern Spanish and Italians, from what we know (both SLC's but no TYR).


@Colin:

Yes there are examples of women changing their appearance, and undeniable sexual selection in women, but my point is that there are even more in men, and hence the answer to your question "Why is there so much emphasis on female choice?" - *most* of the time it's males competing for female attention, so that's the most logical place to focus on first, especially in pre-farming contexts.

Krefter said...

@Tobus,

Many of the "Steppe" samples are actually immigrants who had a lot of EEF. Sintashata and Andronovo from Allentoft had a lot of SLC45A2 like their relatives in Europe did. Wilde. 2015 showed Yamnaya did not have close to modern-like frequencies of SLC45A2.

Also, the SNPs in this paper have *potential affect*. My dad lacks both lactose persistence mutations but has drank milk his whole life with no problems. There are several Europeans on 23andme who lack SLC24A5 but have normal skin tone. There are West Asians with both mutations but with Dark skin.

Unknown said...

One thing interesting about the J2a from Neolithic NW Anatolia that Ted Kandell and I noticed is that this J2a also has a mtDNA that is N1b1a. N1b is common in Georgia 6.9%; NE Caspian in Iran 9.5% and among my n=88 Assyrians 4.5%. He could very well be an assimilated forager if J2a has a more northern and eastern distribution at this time. Most of the burials in the paper are infants/neonates. His is recorded as disturbed. Perhaps he was a trader or an assimilated forager. The autosomal PC plot will help if he is one of the 3 NW Anatolian Neolithic samples that seems to drift toward ANE.

Onur Dincer said...

I don't think you have much knowledge about Indo European rituals do you?

Kurti, like all other Eurasian steppe pastoral nomadic societies, PIE were exogamous, so not like Kurds, among whom cousin marriage is widespread.

Tobus said...

@Krefter:

Many of the "Steppe" samples are actually immigrants who had a lot of EEF

Yes, but I don't think the increase of SLC45A2 in those Steppe pops is *due to* the increased EEF-like ancestry. The Mathieson paper shows SLC45A2 levels in Europe only started rising after the Yamanaya incursion. The finding of higher levels in Anatolian EEF compared to Central/Iberian EEF does make them a potential source for the Steppe, but David has shown that the EEF in later Steppe pops like Andronova/Sintashta comes more from the West than the South, so I still hold that EHG (via Yamanaya) into Europe is the most likely trigger of today's frequencies.

There are several Europeans on 23andme who lack SLC24A5 but have normal skin tone. There are West Asians with both mutations but with Dark skin.

It's estimated that the two SLC's+TYR account for about 50% of the African->European spectrum, so in your first example an individual missing one but having all of the 50% unknowns probably wouldn't stand out in a West European crowd. The second one sounds completely wrong though and I'd like to see the evidence - I suspect "Dark skin" here is relative and these individuals are still significantly lighter than the darkest non-Africans.

Nirjhar007 said...

Yes Krefter, I expect them to be dead like the previous Karelian dude, though we never know...

Nirjhar007 said...

oops i mean the mutation as a dead end :P...

pnuadha said...

Why is there so much emphasis on female choice?" - *most* of the time it's males competing for female attention, so that's the most logical place to focus on first, especially in pre-farming contexts.

I don't think that has been validated and certainly not the one sided attention thats currently given.

I showed that women's bodies have evolved to be feminine, proof of male choice. Comparing their level sexual dimorphism to men, I don't see much of a difference actually.

As far as I can tell women in hunter gatherer societies provide about as much resources as the men, so why the focus on men giving gifts when the general trend is that men and women are both producing usable resources.

Im still not seeing the evidence that men do so much more of the attraction rituals compared to women in tribal societies or that the cosmetic industry isn't simply exploiting and evolutionary behavior in women that seeks to attract.

Matt said...

@ Tobus, the way I'd tell it is we can tell that the Steppe sample, with an average date of 4.5 kya, vs the Central European Early and Middle Neolithic, with an average date of 7.1 kya, have around a 0.25 difference in their frequency of SLA45A2.

The Central European Bronze Age sample (CLB), also with an average date of 4.5 kya has a slightly higher frequency than the steppe at the same time, by around what looks like 0.05.

This suggests that an influx of around 0.4-0.5 MN Central European ancestry into the CLB (averaged out) did not lower the frequency of SLC45A2 (unless accompanied by a rapid burst of high selection).

Dates are based on the relative numbers of each contributing to the "selection population" - the MN samples are very few in number compared to earlier groups.

Whether there is an impetus by steppe population movements to the higher frequencies of around 1 today, is fine as an idea, not really something I know we can say for sure one way or the other, particularly as we don't know why it was selected (pigmentation or something else?).

In any case, on this pigment SNP that is at around 1.0 in Italians and around 0.25 in the Levant, the largely Bronze Age Steppe and largerly Early Neolithic Central European farmers seem closer together than either are to later Europeans (who themselves have low level variation in skin colour that I'm not sure is explained by the main pigment locii we've discovered).

(By comparison the Iberian farmer sample would have an average date of 5.3 kya, so it's more comparable, so actually stronger evidence that there was not an inevitable rise of this variant over time the same in all regions, but selection may have differed in Iberia than Central Europe).
EHG's sample size is quite low at the moment, but fully possible they could be like SHG.

Karl_K said...

Wow. The paper is great. But is it really necessary to argue about if it was men or women who selected their mates in pre-historic societies?

By this time, things were like today. Different societies had cultural differences in how mates were selected. These were fully modern people with complex languages, history, and societies. They were not chimpanzees.

We can speculate how it occurred, based on genetics. But to say it must have been one way or another based on evolution is obviously crap. We are basically the same people today, and people select mates in many different ways, which are culturally diverse, even among genetically similar populations.

Rob said...

Colin

I thought the thing amongst kids these days is to be androgynous and confused in gender roles ?
:)

Alberto said...

@George Okromchedlishvili

"I've read several different papers written by Nostratic scholars..."

Nostratic is not anything mainstream. You could also read Gordon Whittaker's "The Case for Euphratic" and conclude that IE and Sumerian are closely related.

How close proto-Uralic is to IE is not the point. It only tells us something about the location of proto-Uralic (that it was in a contact zone with IE). It could well be a branch of IE and it wouldn't change anything. What you have to look at is at languages that influenced PIE (not languages that where influenced by IE), because those are the ones that can give us clues about possible locations of PIE (though the assumption that Kartvelian or other Caucasian languages were spoken in the Caucasus 6000 years ago might be wrong in itself, so all these clues have to be taken cautiously).

Anyway, more interesting: The abstracts from the conference about IE origins are now complete (some of them were missing a week ago when it was first posted here):

http://www.shh.mpg.de/105110/lag_conference

Here's a new one related to PIE and proto-Altaic:

Indo-European and Altaic landscapes: reconstructed lexics
Anna Dybo, Russian State University for the Humanities, Moscow

It is the first part of a larger work that represents an attempt to systematize our ideas on the natural environment and material culture of the Proto-Indo-Europeans. It is based on a more or less complete selection of reconstructed words from the appropriate semantic areas and on their comparison with a similar selection performed for a protolanguage of similar time depth, whose speakers evidently inhabited a territory that was not in contact with the Proto-Indo-European one — Proto-Altaic. In this part, only the words that belong to the semantic field of landscape terms are analyzed. The main conclusion is that the hypothesis of a steppe environment is more applicable for the Proto-Altaic population, whereas for Proto-Indo-Europeans
a mountainous region seems more appropriate.

Tobus said...

@Matt:

Supplementary Data Table 3 lists average frequencies of .32 for EEF (including Anatolians) and .64 for Steppe, which is .32 difference.

It sounds small but in terms of visible phenotype I think it could be a significant differences - at 32% only 10% of the population would be homozygous and 44% heterozygous derived, the remaining 46% would be double ancestral. The invading people at 64% would be 46% homozygous and 41% heterozygous, with only 13% fully ancestral - most of the invaders would be lighter skinned than the defenders. If one assumes the surviving victors would have a preference for woman similar to themselves there is an inbuilt selection event here.

Anyway as you say, we can't know for certain, but it does seem that timing of the seemingly sudden SLC45A2 increase coincides with Haak's "massive migration from the Steppe" of a population with ANE/EHG ancestry.

Rob said...

Marc van der Linden's paper looks interesting

Old stories, new failures? Genetics, migration and mobility during third millennium
European archaeology

Marc Vander Linden, Institute of Archaeology, University College London, London

Over the past few years, several publications based upon the application of scientific
techniques (e.g. Sr isotopes, aDNA) have revived the interest for migration, mobility
and demography during the 3rd mill. cal. BC in Europe. These contributions should
be welcomed by archaeologists for forcing us to revisit our data from another perspective,
as well as for bringing back the spotlight on a period sometimes forgotten
between the Neolithic and the Bronze Age. Yet, many archaeologists have expressed
either doubt or a relative lack of interest for these papers, largely because some of
their results worryingly echo interpretations which have been lurking throughout
the history of the discipline for more than a century (e.g. steppe influences, Iberian
homeland,...)
This paper will briefly discuss the scientific and archaeological evidence for the
3rd mill. cal. BC, with a focus on the Corded Ware and Bell Beaker complexes. This
review aims at showing the complexity inherent to this period, which cannot be read
as a mere suite of migratory events which would have distributed artefacts, genes
and languages across Europe. On the contrary, distinct facets of the archaeological
record, genetics and linguistics all seem to tell different stories. Rather than proposing
a mere cautionary tale rejecting cross-disciplinary dialogue, this paper will explore
alternative ways aiming at retuning together these apparently discordant voices.

Nirjhar007 said...

one thing which is remarkable is that finding Z-94 from ~2600 BC (if i'm not mistaken), once it was concluded that its a post 1500 BC mutation, now we have its presence from over a thousand year old sample, i think this makes a reasonable possibility that Z-93 mutation is of ~3000 BC or even before!
It was suggested by me that there is good chance of Afanaseivo being R1a, David also agrees with that, i wonder if they come out as Z-93!!! (there is also an equal possibility that it will be just M-417 or even R1b), then what we have to see if CWC shows any R1a-Z93 or not, things will get very interesting if Afanaseivo turns out to be Z-93.

Va_Highlander said...

Grey:

"Does this historical behavior include the Russian eastward expansion from the 1700s?"

Yes, it does. That expansion was part of a broader European expansion that began in the Middle Ages.

"If it did then the historical record would show the direction of expansion on the steppe depended on who had a numbers / technology advantage at the time."

Indeed. It was also a state power, with a written language, and so on, and therefore it should be clear that Russian is in a very different reference class than Turkish or Mongolian or IE at such an early date. So, arguing a fortiori, we can say that the likelihood of any language expanding east across the step is 1-in-3 and be confident that the actual probability is much lower. Any model entailing an expansion east across the steppe must overcome those odds and, so far, I don't see the steppe hypothesis doing so.

George Okromchedlishvili:

"None imply that Elamite or any other MENA language is closer to PIE than Uralic."

I think you've lost the thread of this discussion. You have yet to present a cogent argument demonstrating that PIE should be "closer" to these other languages. Just pretending that you have is insufficient.

I also suspect that you're arguing against an expansion of PIE out of southern Central Asia without knowing the actual details of that model. There might be some problems with basic geography, too.

Alberto, quoting, "Indo-European and Altaic landscapes: reconstructed lexics":

"The main conclusion is that the hypothesis of a steppe environment is more applicable for the Proto-Altaic population, whereas for Proto-Indo-Europeans
a mountainous region seems more appropriate."


Which, one might add, agrees with what we find in the Avesta, for instance.

Coldmountains said...

@Nirjhar.

Z93 will be not found in Afanasievo. Explain why all Z93 samples so far carry Central European EEF admixture unlike Yamnaya and how can Z93 originate east of Yamnaya when they are genetically much western? Don't forget that Afanasievo lacked EEF and that Andronovo had much less of it than Sintashta

Nirjhar007 said...

VAH,Alberto
Which, one might add, agrees with what we find in the Avesta, for instance..
Yes ,the Significance of Mountains is also suggested here.
Not only, in the hymns of the Avesta (e.g. Yt. 5) the ancient Iranian heroes are often associated with mountains, including the progenitor Yima, who is described as offering a sacrifice on the Hukairya mountain, which is probably in Pamir. Whence came these traditions if they came from the northern flatlands?
http://new-indology.blogspot.in/2013/02/indo-iranians-new-perspectives.html
ColdMountains,
Samplings are not done from important Asian Sites, i think the Upcoming Hotu-Belt caves aDNA will carry huge significance and also the Upcoming Rakhigarhi one.
The Current Authors agree that Z-93 didn't come from C Europe/CW, you can argue against it but at the end the decisive results will come from SC Asia and Iran etc.
Since we lack the required aDNA from Asia, its unfair to compare, the sampling is hugely biased so far....

Va_Highlander said...

Nirjhar007:

"Yes ,the Significance of Mountains is also suggested here."

Indeed. It's too often overlooked that pastoralism arrived in southern Central Asia millennia before its advent on the steppe. There is no need to posit an invasion of pastoralists. They were already there. What empirical evidence we have is thin on the ground, but it suggests the pastoral adaptation is as old in Bactria as it is at Mehrgarh, in Arachosia.

Alberto said...

I've been reading this paragraph (quoted already above by Rob) several times, but I can't seem to understand what it is they are suggesting:

"After the Poltavka period, population change occurred in Samara: the Late Bronze Age Srubnaya have ~17% Anatolian Neolithic or EEF ancestry (Extended Data Fig. 2). Previous work documented that such ancestry appeared east of the Urals beginning at least by the time of the Sintashta culture, and suggested that it reflected an eastward migration from the Corded Ware peoples of central Europe [Allentoft et al]. However, the fact that the Srubnaya also harbored such ancestry indicates that the Anatolian Neolithic or EEF ancestry could have come into the steppe from a more eastern source. Further evidence that migrations originating as far west as central Europe may not have had an important impact on the Late Bronze Age steppe comes from the fact that the Srubnaya possess exclusively (n=6) R1a Y-chromosomes (Extended Data Table 1), and four of them (and one Poltavka male) belonged to haplogroup R1a-Z93 which is common in central/south Asians, very rare in present-day Europeans, and absent in all ancient central Europeans studied to date."

So what's with all this? Which is that possible "more eastern source" for EEF-like ancestry in the steppe? And they clearly seem to disconnect CW with Sintashta/Srubnaya on the base of Z93, which indeed is absent in CW (but still derived from M417->Z645).

So are they saying something like:
- Anatolian-like ancestry in Sintashta/Srubnaya didn't come from Europe, but from a more eastern source (which one?)
- R1a-Z93 didn't come from CW, but from somewhere else (and they seem to imply a southern origin?)

With the data available, I'm surprised of these suggestions. I would think that they would go for the more obvious one: CW moving east and bringing EEF-like and Z93 with them from Central Europe. But it seems they suggest other origins (which ones?).

Any ideas?

Krefter said...

@Alberto,
"I would think that they would go for the more obvious one: CW moving east and bringing EEF-like and Z93 with them from Central Europe. But it seems they suggest other origins (which ones?)."

That's the most common opinon. But the authors are seeing the possibility that people related to Neolithic Anatolians who lived East of Samara carried R1a-Z93 and brought it along with EEF-ancestry to Samara.

Alberto said...

@Krefter

Yes, something like that is what they are implying, it seems. But I don't see their logic. To have Sintashta types, you would need not just Anatolian-like farmers (in S-C Asia?) carrying R1a and mixing with Afanasievo types. Those Anatolian-like farmers should be more like MN European Farmers (i.e, with a lot of WHG ancestry).

Do the authors know something what we don't?

capra internetensis said...

A later passage says:

"It is unclear how the Srubnaya acquired farmer ancestry. One possibility is that contact between early farmer and steppe populations produced populations of mixed ancestry that migrated eastward to the Samara district and further east to form the Sintashta/Andronovo populations. A different possibility suggested by Ref 1 [Allentoft] is that the Corded Ware population of central/northern Europe migrated into the steppe. Our new data document the existence of farmer-admixed steppe populations in the European steppe and provide a plausible source for the more eastward migrations of such populations."

So I think what they mean by a "more eastern source" is the people of the western steppe who were partially descended from Tripolye.


Coldmountains said...

@Krefter, Alberto
Sintashta got his EEF ancestry not from people east of Samara. That is impossible. They scored in ANE K7 just 20 ANE so they mixed with farmers lacking or having low amounts of ANE. Such kind of farmers did not exist in Central Asia during this timeframe. It is impossible that farmers moved through Central Asia without picking local ANE ancestry or they had to teleport themselves from Anatolia to northern Central Asia. Davidski already showed various times that their additional farmer ancestry was from the west and very EEF-like. Afanasievo much more in the east lacked also EEF. If the authors really conclude they got EEF ancestry east of Samara than they are wrong here.

Anonymous said...

@Alberto

Dienekes had an article on that theory. This is a link to it:

http://jolr.ru/files/%28108%29jlr2013-9%2869-92%29.pdf

I think her conclusion is too strong. For one thing, while the existence of certain words may prove certain things on the origin of languages - for instance a word for plough proves the language came from an agricutural culture - the frequency of words only indicates things. We know nothing of all the words that are lost, to say something.

So she does have a point that mountainous areas seem part of the PIE landscape. It does neatly fit an Indo-Uralic language on a Northwest Caucasian substrate scenario.

Krefter said...

@Tobus,

Rise in rs16891982 G is due to selection or an unsampled people who had a lot. Besides a few Poltavka there are no new Steppe-samples. They didn't show much more of the G allele than EEF. The reason Steppe scored so high in G allele is because most are Sintashta and close relatives of Sintashta.

We see a rise in rs16891982 G with LN/BA North Europeans. The source is mysterious and the best guess is it is because of selection. Same goes for lactose tolerance.

Matt said...

@ Tobus, 34% vs 64% is lower than I eyeballed it for sure. Counting pixels the full CEM sample should be around 29% frequency (lower than the aggregated Anatolia+Spain+Central European) vs the whole steppe 64%.
There was a fusion of some kind between these communities and it's possible the steppe like people, might select women like themselves. Or maybe women different to themselves.

Even if there was any noticeable skin colour difference I doubt any form of skin colour sexual selection was going on though. It would have to have a really huge selective advantage in a very short timeframe to equilibriate, which is hard to believe.

Plus, there's not enough information on the frequency of SLC45A2 at the moment.

I'll stick with the idea of the steppe having groups at the time of Yamnaya having the previously reported Yamnaya frequency of around 0.4, unless there's evidence it's greatly different.

Note that low frequency in Yamnaya, around 0.4 or lower of the derived SLC45A2 is found independently in Allentoft (http://www.nature.com/nature/journal/v522/n7555/images/nature14507-f4.jpg), Wilde (http://www.pnas.org/content/111/13/4832/T1.expansion.html or http://tinyurl.com/nwb5ten) and Mathieson 1 (http://i.imgur.com/GYjyrLv.png). It seems quite doubtful that the steppe Yamnaya who were near contemporaries of the MN_Central Europeans had significantly higher frequency of light skinned individuals than the overall EEF population.

Although that said, the Corded Ware that were actually intrusive to Europe, maybe, though. Blogger Genetiker had compiled genotypes on his blog and gave German Bell Beaker at 70% and German Corded Ware 83%, while none of the 3 German MN Baalberge and Esperstedt samples had any derived SLC45A2 variants. Whether he's got it right, I don't know, but he did find the Yamnaya samples at 40% so at least that's consistent.

Alberto said...

@Capra

Thanks, that seems to at least partially explain what they mean. Though Srubnaya post-dates Sintashta, so it should be derived from it, not ancestral to it.

What their data shows is that the open steppe was occupied by Yamnaya types (and R1b, except that loner Poltavka R1a-Z94), so an estwards migration of people with EEF ancestry could only happen north of the steppe. And a CWC origin looks like the most plausible (rather than speculation about how the western steppe was, which we don't know, and even if it was R1a and CW-like, there is no evidence of eastward migration of them through the (open) steppe.

But yes, it seems clear by that second passage that they don't mean anything very different, they just expressed it in rather confusing terms.

@Coldmountains

Yes, I don't disagree with you. I was just surprised at that paragraph and its possible meaning.

@epoch

Thanks for the link. I'll read it now.

Anonymous said...

@Va_Highlander

"It was also a state power, with a written language, and so on, and therefore it should be clear that Russian is in a very different reference class than Turkish or Mongolian or IE at such an early date."

Does cultural spreading count as well? Because among those Turkish peoples that swept central Asia in the slipstream of the Mongolians Nestorian Christianity was quite popular.

"So, arguing a fortiori, we can say that the likelihood of any language expanding east across the step is 1-in-3 and be confident that the actual probability is much lower. Any model entailing an expansion east across the steppe must overcome those odds and, so far, I don't see the steppe hypothesis doing so."

I'll be gentle and call this poor mans statistics ;)

No, you haven't enough to warrant such a strong opinion.

Anonymous said...

@capra internetensis

Mallory considers Sredny Stog the first PIE culture. According to the wiki page it is considered (one of the) predecessor(s) to Yamnaya, and it is considered to have been heavily influenced by the Cucuteni-Trypillian culture. We have eastern Yamnaya DNA, not western. There could be EEF there.

Speculation, mind you.

Rob said...

Alberto

With regard to your confusion about the ultimate (?eastern) source of the EEF and Z 93 in timber grave, i very early made the cheeky remark about Nirj seeing that confusing sentence. But he didn't bite :)

Sonic Reducer said...

Since the subject of pigmentation has come up. Has anything shown data on red hair gene frequency? Origin?

Kurti said...

Onur said

"Kurti, like all other Eurasian steppe pastoral nomadic societies, PIE were exogamous, so not like Kurds, among whom cousin marriage is widespread. "

Not a single evidence presented that day just the typical turkish personal attacks. Have you even seen the video? Why are you even commenting on this stuff

1. neither have you enough knowledge about this stuff
2. Nor are you even Indo European yourself

You are basically a confused Bulgarian dude who thinks he is descend of Central Asians.

I have shown you a video which clearly illustrates that PIE had Bride prize (as compensation), The female became member of the males family because they were patriachal.

And you pick up something completely random (Just for the record almost every German village is based on cousin marriages thats at least in the villages here around) and throw it (obviously as insult) into the room because you don't seem to have any arguments. Simply the "Turkish" way?

Since the day I know you, you must be the tuffest guy on the whole Internet. So what the f. are you actually trying to say? I don't understand any of your comments because they don't make sense. You should keep to theApricity forum.

Kurti said...

Oh an the Tajiks, Pashtuns and all other Indo_Iranian groups in Central Asia are and were always " exogamous"?

Of course Turkic tribes might have been "exogamous" since they have such a diverse ancestry. But they are neither Indo European nor native to Central Asia

Please show me a single evidence that PIE were " exogamous". And even if it never happened that a society became isolated from the outside, like the Kalash or Druze?

You make absolutey no sense, but I am used to this from your site.

Rob said...

Kurti & Onur

Calm down you two ;)

VA -highlander -> George

"I think you've lost the thread of this discussion. You have yet to present a cogent argument demonstrating that PIE should be "closer" to these other languages. Just pretending that you have is insufficient.

I also suspect that you're arguing against an expansion of PIE out of southern Central Asia without knowing the actual details of that model. There might be some problems with basic geography, too."

Maybe apart from the lack of basic knowledge of facts, I'm detecting a small amount of personal bias. I'm guessing that to George nothing Indo-European let alone Slavic should pass through the caucuses.


Anonymous said...

@Rob

"Maybe apart from the lack of basic knowledge of facts, I'm detecting a small amount of personal bias. I'm guessing that to George nothing Indo-European let alone Slavic should pass through the caucuses."

But then again, that may serve as bias too. In order to prove George and his hideous nationalistic agenda [1] wrong one wishes the southern route to be true. Before you know it you are advocating ;)

The solution, Rob, is to have as many agenda's as is humanly possible. Preferably paradoxical.

[1] George, that was merely a thought experiment to present to Rob. I have no idea if you have an agenda, and if so which. So, no offence meant.

Rob said...

Don't get cute epoch
I'm Just calling it like it is

Alberto said...

@epoch

Yes, the conclusion of that paper is quite strong, if you want to use that word:

The peculiarities of the landscape-related lexicon in both families are as follows. First of all, the steppe must be excluded from the regions potentially inhabited by Proto-Indo-Europeans. Some relatively high mountains with many kinds of rocks and sharp or big stones are present. Some of these mountains are covered by forests. There are words for narrow passages, canyons, precipices, mines and caves, foothills, valleys and dells, meadows in forests and on the river-banks. The rivers have fords and are definitely smaller than their Proto-Altaic counterparts (there is no semantic variation between “river” and “sea”; nota bene that the only trace of the name of flood is GA (Grrek-Aryan); the lower Danube?); cf. here the noticeably weaker function of fish in the Indo-European economy (expressed in a substantially smaller number of terms for fishing tools, fish body parts and fish species — see the example below). But they could have lived near a sea or a big lake with sandy banks.

But yes, it agrees with the idea that however big or small the influence of IE on Uralic is, the genesis of PIE happened away from an Uralic contact zone, and potentially in a Caucasian contact zone.

Alberto said...

@Rob

Yes, I noticed your quote of that text above and your remark about Nirjhar, but I was still getting my head around all the details to get it myself. And Nirjhar is becoming smarter every day :)

Interesting quote, indeed, but probably just a strange way of explaining what they really wanted to say.

Rob said...

Yes they all like to put the odd spin on wording to keep us guessing :)

Taymas said...

Regarding: "it reflected an eastward migration from the Corded Ware peoples of central Europe. However, the fact that the Srubnaya also harbored such ancestry indicates that the Anatolian Neolithic or EEF ancestry could have come into the steppe from a more eastern source"

"More eastern" is relative to "Corded Ware peoples of central Europe". I don't think anyone here would be all that surprised if the EEF in Srubnaya came from further East than Central Europe.

This interpretation is further supported by additional language later in the paper:

"It is unclear how the Srubnaya acquired farmer ancestry. One possibility is that contact between early farmer and steppe populations produced populations of mixed ancestry that migrated eastward to the Samara district and further east to form the Sintashta/Andronovo populations. A different possibility suggested in Ref. 1 is that the Corded Ware population of central/northern Europe migrated into the steppe. Our new data document the existence of farmer-admixed steppe populations in the European steppe and provide a plausible source for the more eastward migrations of such populations. Further evidence for a connection between the Srubnaya and populations of central/south Asia—which is absent in ancient central Europeans including people of the Corded Ware culture and is nearly absent in present-day Europeans23—is provided by the occurrence in four Srubnaya and one Poltavka males of haplogroup R1a-Z93 which is common in present-day central/south Asians and Bronze Age people from the Altai24 (Supplementary Data Table 1). This represents a direct link between the European steppe and central/south Asia, an intriguing observation that may be related to the spread of Indo-European languages in that direction."

They meant EEF could've followed a path of [somewhere east of Central Europe] -> Srubnaya -> C/S Asia

I have no dog in this fight, I just think people are misinterpreting the authors' language.

Tobus said...

@Krefter:

rs16891982 G in the Steppe is present at about double the frequency of EEF, whether you take the Mathieson v2 figures (64%/32% including later Steppe and Anatolian EEF) or the Mathieson v1 figures (~40%/~20% for just Yamnaya and Central European EEF). The low frequency in Central EEF seems to have been stable for thousands of years before suddenly rising in the LN/BA. You are correct that the source for this sudden rise is mysterious, but it does coincide with a massive migration from the Steppe, who are known to have double the frequency, so it seems logical to consider this a possible trigger for the selection process...pending more definitive evidence of course.

@Matt:
Yes, Yamnaya are at around 40% - I was originally going to use the roughly 40%/22% ratio from Mathieson 1 in my example (which I think it probably more relevant) but I'd have had to guestimate the figures from the graph whereas v2 had the actual figures in SDT3. Also at 40/22 the "double ancestral" ratio would be ~35/62 which I think illustrates my point better - and I didn't want it to look like I'm cherry picking my figures to suit my argument. :)

This is a just a working theory trying to explain the sudden rise in European frequencies in the BA after being steady for a few thousand years - after the Mathieson v1 paper it seemed that EHG was the probable source for higher frequencies and Yamnaya the likely vessel. Mathieson v2 makes it a bit muddier, as we now have Anatolian EEF samples with higher frequencies than Yamnaya some 3,000 years earlier.

Krefter said...

@Sonic,
"Since the subject of pigmentation has come up. Has anything shown data on red hair gene frequency? Origin?"

Map of Pre-Historic Carriers of Red hair.

https://www.google.com/maps/d/edit?mid=zpAcd_2YNln8.ksMlbmYc9Pus

All have been found via Genome Bloggers. If you gathered up a room of 100 random Europeans, over 20 would be carriers of Red hair. It's very common to be a carrier. Plus, almost 10% of most West Asians are carriers.

In summary ancient DNA tells us Red hair is a rare trait that has been around in Eurasia since the early Upper Palaeolithic. My guess would be the first Redheads weren't distinctly Western genetically, that they lived before there was significant genetic diversity in Eurasia.

Onur Dincer said...

Kurti:

Not a single evidence presented that day just the typical turkish personal attacks. Have you even seen the video? Why are you even commenting on this stuff

What personal attacks are you talking about? I have just stated two facts about marriage customs. Yes, I have watched that video and I see nothing in it that contradicts what I have written. Also, what I have written has nothing to do with my ethnicity and/or your ethnicity, I have no prejudice against your ethnicity, to the contrary, I consider your ethnicity as close to my ethnicity and have many friends from your ethnicity (Kurdish).

1. neither have you enough knowledge about this stuff

How do you know that? You do not even know who I am. See below for the details.

2. Nor are you even Indo European yourself

Indo-Europeans are not an ethnic group frozen in time for the past 5000 years, they are genetically, culturally, linguistically, geographically and historically very diverse. So the characteristics of one IE culture do not say much about the characteristics of another IE culture, especially if they are remote spatially and/or temporally and, needless to say, there is no single IE culture.

You are basically a confused Bulgarian dude who thinks he is descend of Central Asians.

Here is where your confusion comes from. You are confusing me with another Onur, who writes in genetic forums. I previously told you in Dienekes' blog that I am not that Onur, but you are still saying I am that guy. My known ancestry is ½ from Anatolia, ¼ from Bulgaria and ¼ from Greece, not wholly from Bulgaria like the Onur you mention. You can ask about me to Turkish forumers such as icebreaker, Pecheneg and Ashina, they will tell you that I am not the Onur you talk about. I have only one forum membership, I joined Anthrogenica two months ago and have not written a single post there yet.

I have shown you a video which clearly illustrates that PIE had Bride prize (as compensation), The female became member of the males family because they were patriachal.

Have I written anything against these?

And you pick up something completely random (Just for the record almost every German village is based on cousin marriages thats at least in the villages here around) and throw it (obviously as insult) into the room because you don't seem to have any arguments. Simply the "Turkish" way?

I have written nothing about German culture, I have written about PIE culture.

Since the day I know you, you must be the tuffest guy on the whole Internet. So what the f. are you actually trying to say? I don't understand any of your comments because they don't make sense. You should keep to theApricity forum.

The thing is, you do not know me, you are confusing me with another Onur. There are probably tens of thousands, if not hundreds of thousands, of people with the first name Onur in Turkey, so is it logical to consider everyone you see in genetic blogs or forums with the name Onur as the same person? BTW, I do not have Apricity forum membership and, not to disrespect its members, but I consider it as a forum not worthy of being a member.

Oh an the Tajiks, Pashtuns and all other Indo_Iranian groups in Central Asia are and were always " exogamous"?

Have I written anything about Tajik or Pashtun culture?

Of course Turkic tribes might have been "exogamous" since they have such a diverse ancestry. But they are neither Indo European nor native to Central Asia

Not all Turkic peoples are traditionally exogamous, the ones who still live in the Steppe and/or preserve Steppe customs better are exogamous.

-- continues below --

Onur Dincer said...

Please show me a single evidence that PIE were " exogamous". And even if it never happened that a society became isolated from the outside, like the Kalash or Druze?

Sure!
https://books.google.com.tr/books?id=4DR-AgAAQBAJ&pg=PA48&lpg=PA48&dq=proto-indo-europeans+exogamous&source=bl&ots=aiuVzAl0GP&sig=6jCANa47jsAYJ6Mn2bMUWinT1eU&hl=tr&sa=X&ved=0CC8Q6AEwBDgKahUKEwjY1_yJub7IAhXFWRoKHXowANk#v=onepage&q=proto-indo-europeans%20exogamous&f=false

https://books.google.com.tr/books?id=bSxHgej4tKMC&pg=PA21&lpg=PA21&dq=proto-indo-europeans+exogamous&source=bl&ots=v9QB6gBd2M&sig=jExNMpJ-1TMAelqePQ58qBrU5ek&hl=tr&sa=X&sqi=2&ved=0CEYQ6AEwBmoVChMIy862ybi-yAIVhL0aCh3FdgvH#v=onepage&q=proto-indo-europeans%20exogamous&f=false

https://books.google.com.tr/books?id=03KrAAAAIAAJ&pg=PA247&lpg=PA247&dq=proto-indo-europeans+exogamous&source=bl&ots=V5BObzyDiu&sig=LdTqRS5EXnULCt3FkMTuvnKk0L0&hl=tr&sa=X&ved=0CDwQ6AEwBzgKahUKEwjSmdaNu77IAhUILhoKHZ9YBU4#v=onepage&q=proto-indo-europeans%20exogamous&f=false

You make absolutey no sense, but I am used to this from your site.

What site of me are you talking about? If you mean a site run by the other Onur, then I do not need to reiterate that I am not that Onur.

Nirjhar007 said...

lol,
Epoch,
Mallory considers Sredny Stog the first PIE culture.
Actually Mallory has become more practical than that-

http://jolr.ru/files/(112)jlr2013-9(145-154).pdf

Karl_K said...
This comment has been removed by the author.
Sonic Reducer said...

@krefter Thank you

Anonymous said...

@ Nirjhar007

Nice article. It doesn't mention Sredny Stog, however. It refers that no cereal cultivation east of the Dnepr in the PIE time frame has been uncovered as a big problem for identifying the IE Urheimat. However, there have been substantial finds in the Ukraine of all the mentioned cereals at Dnepr-Don culture sites.

http://www.archaeology.su.se/polopoly_fs/1.123143.1360243985!/menu/standard/file/Kashuba_Natalia_The_Neolithization_of_Ukraine.pdf

Perhaps that new archaeology might change the image. The whole area is loess grounds, which is very suitable for wheat and barley tillage. LBK culture in Europa can almost be superimposed on loess finds. The whole Yamnaya area is very fertile ground.

https://www.ufz.de/export/data/1/28154_European_Loess_Map_hires.jpg
http://what-when-how.com/wp-content/uploads/2011/03/tmp3517_thumb1.jpg

Va_Highlander said...

epoch2013:

"I'll be gentle and call this poor mans statistics ;)"

How very disappointing. In answer to our fervent prayers, a rich man happens by and all he can offer is condescension.

What was so special about PIE? What compels you to exclude it from the reference class of all languages known to have expanded across the Eurasian steppe? How exactly do the rich arrive at a defensible prior probability for an eastward expansion?

"According to the wiki page..."

Hmmm....

Rob:

"Maybe apart from the lack of basic knowledge of facts, I'm detecting a small amount of personal bias. I'm guessing that to George nothing Indo-European let alone Slavic should pass through the caucuses."

Quite possible. After a while, one tires of looking for motives.

epoch2013, again:

"In order to prove George and his hideous nationalistic agenda [1] wrong one wishes the southern route to be true."

I'm beginning to see a pattern, here.

Nirjhar007:

"Actually Mallory have become more practical than that-"

How refreshing to see a proponent of the steppe hypothesis honestly discuss the weaknesses of his preferred narrative.

Anonymous said...

@VA_Highlander

So, arguing a fortiori, we can say that the likelihood of any language expanding east across the step is 1-in-3 and be confident that the actual probability is much lower. Any model entailing an expansion east across the steppe must overcome those odds and, so far, I don't see the steppe hypothesis doing so.

The remark was answer to this. However, it is bogus statistics. Even if we have a number of historical events going east to west it doesn't really imply anything for the odds of IE to go eastward. If you have a son and two daughters and your wife is pregnant of your fourth child, then this child doesn't need to overcome a 1-in-3 probability to be a boy.

Krefter said...

The Natural-selection part of the paper.

http://www.anthrogenica.com/showthread.php?5602-Traits-that-were-Selected-for-in-the-last-8-000-years

Lactose Tolerance, Pale skin, and reduced body fat have the strongest signal of selection. Lactose Tolerance and reduced body fat have been risen a lot in frequency even in the last 4,000 years.

Va_Highlander said...

epoch2013:

"However, it is bogus statistics. Even if we have a number of historical events going east to west it doesn't really imply anything for the odds of IE to go eastward."

So, essentially, the rich just toss empiricism to the wind and assume observed behavior tells us nothing informative about the world. Thanks. That explains quite a bit.

"If you have a son and two daughters and your wife is pregnant of your fourth child, then this child doesn't need to overcome a 1-in-3 probability to be a boy."

I'm sorry, but I honestly expected a rich man to do better than this. Among the poor, at least, you can't just toss out an example of bogus statistics and claim it's just like that. There has to be some sort of discernible analogy, else it's just so much hand-waving.

Now, even a poor man can afford to offer a second chance, so I shall ask you again. How, exactly, do the rich arrive at a defensible prior probability for an eastward expansion of PIE? Or have you even bothered to consider the likelihood of your preferred narrative?

capra internetensis said...

@Va_Highlander

Iron Age and later nomads who populated the whole of the Eurasian steppe, subsisting by a fully-developed pastoral regime and using advanced cavalry tactics and equipment with the ability to organize large armies, tended to expand from east to west.

Therefore, pioneer steppe pastoralists with a different subsistence regime, prior to even the invention of the chariot let alone the kind of cavalry that would render it obsolete, should operate in the same way? When, as far as we know, they *started* in the west?

Va_Highlander said...

capra internetensis:

"Therefore, pioneer steppe pastoralists with a different subsistence regime, prior to even the invention of the chariot let alone the kind of cavalry that would render it obsolete, should operate in the same way?"

Okay. You're saying that what was special about PIE is that it did not have the admittedly overwhelming advantages of these Iron-Age steppe peoples. How do you think these advantages prevented a language spreading from west to east?

Since the climate worsens, as one migrates eastward across the steppe, why was it always advantageous for nomadic steppe-herders to continue pushing eastward into drier and colder conditions?

I understand why the Turks and Mongols should have brought their language west. I understand why Russian expanded across the steppe. I don't understand why, given what you say, we don't have empirical evidence of a viable language expanding east during the Iron Age. You might be right, but given the evidence we have what you propose does not seem predictive.

capra internetensis said...

@Va_Highlander

Your statistical argument is quite literally that Iron Age steppe nomads drove each other westward more times than Early Modern Russia colonized Siberia. Does this really strike you as mathematically sound?

Now you can say that Copper-Bronze Age steppe pastoralists were more similar to Iron Age steppe nomads than they were to Early Modern Russians, which is fine. But it is a pretty weak argument, unless you can identify the mechanism responsible and show that it applied in the earlier period.

Certainly the archaeology of Central Asia is not sufficiently well understood that we can rule out an early east to west spread of pastoralists. Maybe future work will find evidence of it. But the fact that Iron Age nomads tended to expand westward cannot tell us where the innovations of Copper Age steppe pastoralists first appeared.

(It was advantageous to move into colder and drier conditions for the same reasons it always is - because there are economic opportunities there, or because you are running away from trouble at home. Somewhere I have seen an ecological explanation for the Iron Age east-west tendency, but I don't recall what it was, or whether it even made sense.)

Anonymous said...

"Since the climate worsens, as one migrates eastward across the steppe, why was it always advantageous for nomadic steppe-herders to continue pushing eastward into drier and colder conditions?"

https://upload.wikimedia.org/wikipedia/commons/b/bb/Koppen_World_Map_%28retouched_version%29.png

D is humid continental, B Cold semi-arid continental.

I'd say that in the area the supposed Indo-European east expansion did not enter worse climatic conditions. Perhaps drier. But that wouldn't be a first in history.

In the past things weren't all that different, from what I see.

http://www.esd.ornl.gov/projects/qen/nercEURASIA.html

German Dziebel said...

"the statistic f4(Basque, Iberia_Chalcolithic; Yamnaya_Samara,Chimp)=0.00168 is
significantly positive (Z=8.1), as is the statistic f4(Spanish, Iberia_Chalcolithic;
Yamnaya_Samara, Chimp)= 0.00092 (Z=4.6). This indicates that steppe ancestry occurs in
present-day southwestern European populations, and that even the Basques cannot be
considered as mixtures of early farmers and hunter-gatherers without it."

This is significant in the light of my earlier discussion with Davidski. ANE is indeed present in Basques.

Anonymous said...

@Nirjhar007

The issue Mallory points to is this: No cereal finds before 2000 bC but the words do spread to indo-arian languages. That would be explainable if a secondary IE settlement happened, one which brought (more) agriculture.

Look at this:

"Previous work documented that such ancestry appeared east of the Urals beginning at least by the time of the Sintashta culture, and suggested that it reflected an eastward migration from the Corded Ware peoples of central Europe. However, the fact that the Srubnaya also harbored such ancestry indicates that the Anatolian Neolithic or EEF ancestry could have come into the steppe from a more eastern source. "

Va_Highlander said...

capra internetensis:

"Your statistical argument is quite literally that Iron Age steppe nomads drove each other westward more times than Early Modern Russia colonized Siberia. Does this really strike you as mathematically sound?"

Probability, not statistics. I am aware of the limitations of the sample size, but, yes, it does. It's all the empirical evidence we have relevant to how languages behaved on the Eurasian steppe. If your objection is that Russian should be in a different reference class than Turkish and Mongolian, I agree and said so, but including it raises the prior probability of an I-E expansion east, making for a stronger argument.

If your argument is that an I-E expansion should be in a reference class with Russian, one excluding Turkish and Mongolian, I suspect that might prove difficult. If you're arguing that PIE should be in a reference class of its own, I think that is going to prove more difficult still.

"But it is a pretty weak argument, unless you can identify the mechanism responsible and show that it applied in the earlier period."

If you really believe it's weak, then propose a number of your own that all rational and objective observers should accept. Based just on what we know and agree about how languages behaved on the Eurasian steppe, how probable is it that a language expanded east?

"It was advantageous to move into colder and drier conditions for the same reasons it always is - because there are economic opportunities there, or because you are running away from trouble at home."

I was following Nichols and she may have been wrong about the climate. I'm not an expert on the intersection of sheep herding, Eurasian geography, and climatology. That doesn't change the fact we have little empirical evidence of viable languages expanding east.

epoch2013:

Thanks for the links, but the central question still remains. Based solely on what we observe about how languages behave on the Eurasian steppe, how much of our prior-probability space does the steppe hypothesis represent? If we can't agree on that before we start looking at the relevant evidence for expansion, then we're just telling stories.

Davidski said...

People expanded east on the steppe during the Early Bronze Age, and they didn't mix with anyone for a very long time as they pushed east. We have their genomes so it's now settled.

And unless you think they turned into mutes as soon as they crossed the Europe/Asia border in the south Urals, then you don't have an argument.

Karl_K said...

@Davidski

"People expanded east on the steppe during the Early Bronze Age, and they didn't mix with anyone for a very long time as they pushed east. We have their genomes so it's now settled."

How could anyone believe that? You didn't use any invented probability to make your case, just facts. What are people supposed to argue about if you just present solid facts?

Anonymous said...

@Va_Highlander

"If you really believe it's weak, then propose a number of your own that all rational and objective observers should accept. Based just on what we know and agree about how languages behaved on the Eurasian steppe, how probable is it that a language expanded east?"

You are begging the question. By asking us to state what the odds are you are assuming that it is sure that some sort of mechanism drives folks east or west and thus that the odds can actually be calculated or at least estimated. But it can't: In Eurasia there are only so many places to go and an eastern cultures that expands has only one place to go. West.

You aren't sure how many languages went east or west. We don't even know how many extinct languages went east or west. You have to tell me if you'd count Mongolian or not: Nobody in the west speaks a Mongolian language nowadays and even in the times the Mongols went west hardly anybody spoke it in the west. Should we then count Mongolian as a language that went west or not? And PIE, in the Mallory scenario: It went east, south AND west. How to count that? For that matter, the Mongols went south as well and started the Yuan dynasty.

There is no sensible answer to your question.

"Probability, not statistics."

Probability IS statistics.

Krefter said...

@Karl_K,
"How could anyone believe that? You didn't use any invented probability to make your case, just facts. What are people supposed to argue about if you just present solid facts?"

Are you agreeing or dis agreeing? Anyways what he says is true.

Iranocentrist said...

Looking at the PCA, its obvious to tell that all of Central and Northern Europe are in between EHG and Caucasus/West Asia. Look at the big picture all of you Eurocentrists, you are a mixture of EHG and West Asia (Teal). Case closed

Krefter said...

@Iranocentrist,

Wondering where ANE-signal in Europe came from is so 2014. It comes from "Steppe" shorthand for Pre-Historic Russia and Ukraine. Most blood in 90% of Europe is not Teal or EHG.

Davidski said...

The PCA above suffers from projection bias.

Northern and Central Europeans are mixtures of Early Bronze Age steppe nomads and Middle Neolithic North-Central Europeans, with some extra WHG admixture in the north. See here...

https://drive.google.com/file/d/0B9o3EYTdM8lQM2FqQ1k2MUJCNDg/view?usp=sharing

And as you can also see in this PCA, West Asians are basically mixtures of steppe nomads and Southwest Asians.

Be proud of your Eurasian steppe ancestry Iranocentrist. That's where your language and culture are from too.

Iranocentrist said...

The steppe was a melting pot, not a source.

Davidski said...

It was a source of massive population movements during the Early and Middle Bronze Age.

And the populations streaming out of the steppe at this time had a very specific genetic structure with lots of EHG admixture. Quite unique in West Eurasia.

Your somewhat misguided interpretations of PCA gloss over these very important facts.

Iranocentrist said...

Your right, I am really a big amateur compared to you and many of the regulars here, you guys are amazing, and intelligent. But it would be really interesting to get more ancient Middle Eastern DNA, do you think we will get any soon?

«Oldest ‹Older   1 – 200 of 204   Newer› Newest»