March 27, 2014

Major new article on the deep origins of Y-haplogroup R1a (Underhill et al. 2014)

Five years ago, Underhill et al. (2009) presented a major advance in the study of haplogroup R1a. Much new knowledge was added in the interim by genetic genealogists and some scientists,  and now a major new paper by Peter Underhill comes to update our knowledge of this important and widely spread human lineage.

The shallow coalescence time within R1a will not surprise many genetic genealogists while its diversification in the vicinity of present-day Iran might. A ~5-7kyBP coalescence would make the expansion of R1a lineages presumably visible to future ancient DNA studies which will probably be the final arbiter of the veracity of the date estimate in this paper and its postulated place of origin.

I'll try to digest what the new information has to say about Eurasian prehistory, but in the meantime...



... I will, however, take some time to highlight the passing of the guard from Y-STRs to Y-SNPs which I had long ago anticipated. There is some lingering controversy about the substitution rate on the Y chromosome is, but it is hopeful that this will be resolved before not too long as the price of whole genome sequencing is always dropping and the samples sequenced in this study are probably the first of many to come.

In any case:
Our phylogeographic data lead us to conclude that the initial episodes of R1a-M420 diversification occurred in the vicinity of Iran and Eastern Turkey, and we estimate that diversification downstream of M417/Page7 occurred ~5800 years ago. This suggests the possibility that R1a lineages accompanied demic expansions initiated during the Copper, Bronze, and Iron ages, partially replacing previous Y-chromosome strata, an interpretation consistent with albeit limited ancient DNA evidence.54, 60 However, our data do not enable us to directly ascribe the patterns of R1a geographic spread to specific prehistoric cultures or more recent demographic events. High-throughput sequencing studies of more R1a lineages will lead to further insight into the structure of the underlying tree, and ancient DNA specimens will help adjudicate the molecular clock calibration. Together these advancements will yield more refined inferences about pre-historic dispersals of peoples, their material cultures, and languages.
It would of course be great to get some ancient DNA data from Iran and Eastern Turkey:
Among the 120 populations with sample sizes of at least 50 individuals and with at least 10% occurrence of R1a, just 6 met these criteria, and 5 of these 6 populations reside in modern-day Iran. Haplogroup diversities among the six populations ranged from 0.78 to 0.86 (Supplementary Table 4). Of the 24 R1a-M420*(xSRY10831.2) chromosomes in our data set, 18 were sampled in Iran and 3 were from eastern Turkey. Similarly, five of the six observed R1a1-SRY10831.2*(xM417/Page7) chromosomes were also from Iran, with the sixth occurring in a Kabardin individual from the Caucasus. Owing to the prevalence of basal lineages and the high levels of haplogroup diversities in the region, we find a compelling case for the Middle East, possibly near present-day Iran, as the geographic origin of hg R1a.
Also, the finding that...
The four subhaplogroups of Z93 (branches 9-M582, 10-M560, 12-Z2125, and 17-M780, L657) constitute a multifurcation unresolved by 10 Mb of sequencing; it is likely that no further resolution of this part of the tree will be possible with current technology. Similarly, the shared European branch has just three SNPs.
... seems to imply some Copper-to-Bronze Age guys did more than their fair share of fathering.

European Journal of Human Genetics , (26 March 2014) | doi:10.1038/ejhg.2014.50

The phylogenetic and geographic structure of Y-chromosome haplogroup R1a

Peter A Underhill et al.

R1a-M420 is one of the most widely spread Y-chromosome haplogroups; however, its substructure within Europe and Asia has remained poorly characterized. Using a panel of 16 244 male subjects from 126 populations sampled across Eurasia, we identified 2923 R1a-M420 Y-chromosomes and analyzed them to a highly granular phylogeographic resolution. Whole Y-chromosome sequence analysis of eight R1a and five R1b individuals suggests a divergence time of ~25 000 (95% CI: 21 300–29 000) years ago and a coalescence time within R1a-M417 of ~5800 (95% CI: 4800–6800) years. The spatial frequency distributions of R1a sub-haplogroups conclusively indicate two major groups, one found primarily in Europe and the other confined to Central and South Asia. Beyond the major European versus Asian dichotomy, we describe several younger sub-haplogroups. Based on spatial distributions and diversity patterns within the R1a-M420 clade, particularly rare basal branches detected primarily within Iran and eastern Turkey, we conclude that the initial episodes of haplogroup R1a diversification likely occurred in the vicinity of present-day Iran.

Link

112 comments:

Davidski said...

One of the few useful things about this study is that its data suggest there were major expansions of sister clades Z282 and Z93 at about the same time from the Eurasian steppe.

Apart from that, the phylogeography of European R1a is a mess, with Z280 shown as ancestral to Z282, M458 and Z284, which is nonsense.

Also, M420 is found in Europe. They just missed it there. So no, there was no Copper Age expansion of R1a to Europe from eastern Turkey, although I'm sure you'll keep thinking otherwise.

Kurti said...

As a person who has been recently tested for being yDNA R1a1a1*.
I am going to give some informations I know about the diversity of R1a1a found in this region, particularly among my people. Considering that Kurdish is part of the Iranic language family someone could come to the conclusion that the majority (if not all) of R1a* among them is R1a-z93. But interestingly this is not the case. It is highly diverse. from the only three Kurdish individuals who have taken deeper tests on their yDNA. One turned out as R1a +z93

http://i1127.photobucket.com/albums/l625/ft-d/maps/z93_zps1e5c12bf.jpg

The second turned out as R1a +z283
http://i1127.photobucket.com/albums/l625/ft-d/maps/z2832.jpg

And last but not least turned out to be simply R1a* (L62+, L63+, SRY10831.2-, M17-)! Which is ancestral to z93 and z283
http://kurdishdna.blogspot.de/search?updated-max=2013-10-15T00:56:00-07:00&max-results=7

There is obviously big diversity . As I said in the past I still believe there was a time when R1a, as well R1b was more common in the region, until it was reduced by the expansion of the Akkadian and later Assyrian empire.

mooreisbetter said...

Wow. Does everyone realize the implications of the TMRCA splits within the different Asian versus European clades of R1a?

It's huge.

The clades appear to have split too far in the past to indicate Indo-European origin.

This blows a huge hole in the oft-repeated maxim, that the presence of R1a in India shows it is the PIE marker.

Kurti said...

Note, I am not implying that Haplogroups like J1, T or other are not native to the region (this would be ridiculous) I am saying that with the advance of Akkadians the demographic frequency of Haplogroups was changed and new subclades were likely introduced to the region like E-M123, E-Z827, E-M8 and J1c3d.

Davidski said...

mooreisbetter,

R1a-M417 is estimated to be 4800-6800 YBP in this paper based on complete R1a sequences. The other estimates, based on a few STRs can be safely ignored.

So if R1a-M417 is only dated to the late Neolithic, then the split between the European Z282 and Asian Z93 happened even later (since they share the Z645 mutation under M417).

What this suggests is that there were almighty expansions of Z282 deep into Europe and Z93 deep into Asia sometime during and after the Copper Age, probably from between their current hotspots, which you can see on the maps above. That looks to be the southern Urals, or thereabouts.

So if that's not a signal of the Kurgan Proto-Indo-European expansions, then I don't know what is.

By the way, I should've made the point earlier that R1a-M420 isn't dated to the Copper Age, as Dienekes claimed. It's older than that because it's well above R1a-M417 on the phylo tree.

Mauri said...

They forgot again Finland, but did collect samples from tiny relative groups from Russia. Those groups, like Vepsas and Karelians, counting whole populations and all haplogroups, correspond only 20% of the Finnish R1a and 1% of the Finnish population. I would have been interested in finding migration routes to Finland. It is more likely that Vepsians and Karelians are today less or more mixed Russians and represent very young history, but this didn't happen in Finland and we could reveal something really interesting about North European ancient migrations by studying also Finns.

eurologist said...

Compared to the known, contemporary late Gravettian ancient Siberian Malta-boy R*, we here have an estimated ~25,000 ya R1a vs. R1b divergence. It therefore looks very probable that this estimate is wrong by at least 5,000 years.

More importantly, we know with much certainty that haplogroups: late P, Q, and R originated (or in the case of P moved) in the NW portion of the sub-continent. So, R1, and most likely R1a and R1b, originated in that neighborhood, and not in Iran nor in Turkey, and earlier than in the 25,000 - 30,000 time frame.

Hector said...

From what I know of Underhill I suppose he thinks that the entire MP clade has an origin in the Middle East.

He has a habit of doing this. A close parallel is DE which he thinks has an origin in Africa.

He favors less parsimonious models that fit his preconceived notion of what it ought to be with archaeological pseudo evidences as excuses.

But I am confident that he and Eurocentric readers of this blog will be disappointed. But if the hypothetical comical journey of M all the way from middle east to Papuan islands does not dissade them from their entrenched beliefs and wishful thinking I don't know what will.

Dienekes said...

So no, there was no Copper Age expansion of R1a to Europe from eastern Turkey, although I'm sure you'll keep thinking otherwise.

I said that there should be ancient DNA work in Eastern Turkey and Iran, not that it expanded from there.

War Lord said...

"Wow. Does everyone realize the implications of the TMRCA splits within the different Asian versus European clades of R1a? It's huge.
The clades appear to have split too far in the past to indicate Indo-European origin. This blows a huge hole in the oft-repeated maxim, that the presence of R1a in India shows it is the PIE marker."

But Indo-Iranian languages definitely originated from the steppes north of the Black Sea. Furthermore, R1a is typical for Aryan upper castes. Letting aside backward migrations of Indo-Iranians from Central Asia to Europe. There is something fishy about the results.

By the way, the maps don't agree with the tree. Z280, Z282 and Z284 are very probably in incorrect places.

War Lord said...

"Wow. Does everyone realize the implications of the TMRCA splits within the different Asian versus European clades of R1a?"

TMRCA mean nothing. Errouneous TMRCA also stand at the root of the infantile theory about the spread of R1b-M269 from Asia Minor to Western Europe during the Neolithic.

Aaron said...

I disagree with the assumption that Y-STR genotyping is somehow less informative than Y-SNP genotyping especially at this timescale.

If you look at supplemental figures 1 & 2, which contain graphical representations of Y-STR diversity among the populations, they tell a much richer story than the 3 SNPs that differentiate the Z282 Europeans.

3 data points are not sufficient to draw any conclusions about timing of differentiation.

Using the genealogical mutation rate for Y-STRs it puts the expansion of R1a at 2600 YBP, that perfectly coincides with an Indo-European expansion into Europe.

If you read the paper that is referenced for the genealogical mutation rate, there is no reason to doubt their conclusions. They were well powered and their experimental techniques were sound. The evolutionary rate on the other hand is riddled with assumptions about population sizes, growth rates and differentiation times. The evolutionary rate cannot be trusted when we have direct measurement of the rate without assumptions.

Dienekes said...

What this suggests is that there were almighty expansions of Z282 deep into Europe and Z93 deep into Asia sometime during and after the Copper Age, probably from between their current hotspots, which you can see on the maps above. That looks to be the southern Urals, or thereabouts.

I'm glad you now seem to agree with my theory about the early distribution in the flatlands north and east of the Caspian and you've abandoned the theory of "Polish Neolithic" R1a and expansion into the steppe:

"Anyway, Poland has everything it needs in terms of R1a1a to be considered an ancient expansion point, rather than a recent destination. It's the central point of several R1a1a subclades, not just one. Indeed, Southeastern Poland scores over 60% R1a1a, and that includes M458 and Z280. And to top it all off, we know that the Corded Ware culture expanded from Poland, and ancient DNA shows that this culture included individuals who carried R1a1a.

So R1a1a has been in Central and East Central Europe, and in Poland, since proto-Indo-European times, and the area probably had high frequencies of R1a1a since the Neolithic. That's why I see it as a plausible region from which light-pigmented Europeans, with extreme frequencies of R1a1a, expanded to the east during the Eneolithic."

Progress has been made.

Unknown said...

“The evolutionary rate cannot be trusted when we have direct measurement of the rate without assumptions.”

Not really, the evolutionary rate can still be appropriate in some circumstances, for instance, I assembled 5793 R1 STR haplotypes (12 markers) from the FTDNA project pages (2793 R1a and 3000 R1b), their evolutionary rate TMRCA on average was 1292 generations (1060 if you use median repeats as the ancestral, 1525 if you use modal repeats) , @ 25 years / Generation → 32.3 KYA . Where as the Pedigree rates (from various sources) give a mean TMRCA of 675 Generations, @ 30 Years /Generation → 20.25 KYA. This publication announces 25 KYA for R1 using an SNP based system, all that means is that the mutation rates they employed using their SNP dating system is analogous to using an intermediate mutation rate between the pedigree and effective rates using an STR based TMRCA and not making the evolutionary rates 'untrusted', especially when we don't even know if the Mutation rates they used in their SNP based TMRCA system is entirely reliable.

References:

5,793 , - 12 marker R1 haplotypes from FTDNA

Y-TMRCA calculator for a group of STR haplotypes.

Rokus said...

So if R1a-M417 is only dated to the late Neolithic, then the split between the European Z282 and Asian Z93 happened even later (since they share the Z645 mutation under M417).

So if that's not a signal of the Kurgan Proto-Indo-European expansions, then I don't know what is.

I am sure you don't know, because it requires a certain blindness to deny that North-Western European R1a split off before this great Asian Z93 - European Z282 divide. The enormous expansion of R1a appears to be not unlike e.g. the impressive advent of European YDNA in Australian, Polynesian, Amerindian and African American populations. Such a quick advent also implies the observed reduction of ancestral lineages. Whatever the direct origen of the quick R1a expansion in Eurasia, it must have arrived there from an older source. Evidence accumulates this ultimate source may be found in Mesolithic North-Western Europe.

Davidski said...

Present-day Poland was indeed an ancient expansion point for R1a1a, probably from the late Neolithic to the middle ages. Have you heard of the Corded Ware culture? This PDF has some background on it...FYI Weichsel = Vistula.

http://www.sciencemag.org/content/suppl/2013/10/10/342.6155.257.DC1/Brandt.SM.pdf

The point I was making here was that a few M420 samples from the Near East, which appear to belong to the same young subclade anyway, don't say anything about the origins of M420, because we know for a fact that M420 is also present in Europe. The only major R1a hotspot where M420 is missing is South Asia.

How about I drag out some old posts of yours where you claim R1a originated in India?

In fact, your flatlands theory is still a bit out. What you need to do is move things a bit further west to archeological cultures commonly associated with the early Indo-Europeans, like Khvalynsk, Yamnaya, Abashevo, etc.

These people don't exist anymore, so testing a few Chuvashs from the same area today won't tell us what sort of R1a they carried.

Rokus said...

But I am confident that he and Eurocentric readers of this blog will be disappointed. But if the hypothetical comical journey of M all the way from middle east to Papuan islands does not dissade them from their entrenched beliefs and wishful thinking I don't know what will.

Not YDNA M but YDNA MP either originated or travelled to the respective places of origin for YDNA P (basically Q and R) and YDNA M. I don't see why a journey from A to B would be more "comical" than a journey from B to A. Since you do, obviously you are hampered by a share of wishful thinking all of your own.

I guess the geographic mystery of YDNA MP will be solved conclusively once it becomes clear that SE Asian and Australian K* and PNG YDNA S are all closely related. The belt of related (MP/K*) YDNA that stretches from Northern Eurasia to Oceania already follows a similar pattern as YDNA C1. This pattern is in line with the pattern of genetic correlation between Northern Eurasia and Oceania as revealed by Malta Boy. I already warned against too much confidence in "ancestral" subclades in the Southeast Asian "Terra Incognita."

andrew said...

@eurologist. The 5000 years BP estimate, give or take, isn't for the most basal R1a lineages, it is for the time period at which R1a experienced a star-like expansion.

Given the solidity of their data, the Middle East seems suspect as a center of that expansion, and while the presence of diversity in Iran today could be from the most typically PIE population genetically migrating to Iran from the plains, I am not inclined to dismiss the possibility of a PIE expansion out of Iran possibility out of hand either.

PIE expansion appears to have involved superior metallurgy seemingly inconsistent with steppe pastoralists, and PIE vocabulary seems awfully rich in agricultural vocabulary for a population of steppe pastoralists. Mountain farmers and metal workers might be a better fit.

And, while an "Out of India" theory seems disfavored by the evidence, I don't think that we should prematurely rule out the possibility that PIE was part of a linguistic family that included the more civilized copper age centers of Persian, Harappa, and Bactria. Politics has severely limited our access to good modern archaeological evidence from Iran for the last 35 years, so we shouldn't assume that absence of evidence is evidence of absence of PIE cultural traces perhaps even older than Kurgan cultures in these regions.

Davidski said...

Rokus,

The vast majority of Northwestern European R1a is Z284, which is under Z282.

The other Northwestern European subclade is CTS4385. This marker isn't part of Z282, but there's no reason why it couldn't have arrived from the east during the Copper Age as M417. In fact, the Corded Ware R1a from Eulau appears to be CTS4385.

Just in case you're not aware, Underhill et al. didn't test for CTS4385, so their M417 results from Northwestern Europe probably mostly belong to this subclade. In fact, that's what the STR values suggest for all but one of the samples.

Andrew,

Their data isn't as solid as it looks at first glance. The M420 from the Near East appears to belong to a single young lineage.

terryt said...

"if that's not a signal of the Kurgan Proto-Indo-European expansions, then I don't know what is".

I have long suspected the expansion i
of the Indo-European languages is somehow associated with the expansion of R1a. Of course langauges often expand far beyond any original haplogroup association.

"But if the hypothetical comical journey of M all the way from middle east to Papuan islands does not dissade them from their entrenched beliefs and wishful thinking I don't know what will".

P is a long way upstream of R1a and so to me it is more than likely that P is the haplogroup that moved. Especially when we consider that all the haplogroups most closely related to P are specifically eastern. That includes M, S, NO, K1, K2 and K3.

"I guess the geographic mystery of YDNA MP will be solved conclusively once it becomes clear that SE Asian and Australian K* and PNG YDNA S are all closely related".

The relationship is uncertain at present. The main problem as I see it is: what is the relationship between the K haplogroups and MNOPS? Does the clade split into K1, K2, K3, MP, NO and S or do the Ks belong to a separate branch?

"The belt of related (MP/K*) YDNA that stretches from Northern Eurasia to Oceania already follows a similar pattern as YDNA C1. This pattern is in line with the pattern of genetic correlation between Northern Eurasia and Oceania as revealed by Malta Boy".

But it doesn't tell us the direction of correlation. Remeber that the only members of Y-DNA found in northern and western Eurasia are all P-derived. That suggests it is the odd one out.

"I already warned against too much confidence in "ancestral" subclades in the Southeast Asian 'Terra Incognita'".

And yey many insist Y-DNA O is ancient in Southeast Asia, and even South China. All the evidence indicates some movement into SE Asia/South China of a Mongoloid phenotype early in the Neolithic. The only haplogroup that fits the scenario is Y-DNA O along with a few (very few) mt-DNA haplogroups. O has replaced (or supplemented) the earlier haplogroups in SE Asia. Interestingly it failed to reach (or survive in) Australia and is present in New Guinea at a very low frequency. On the othe rhand it did make it out into the wider Pacific.

Kapak said...

A Transcaucasian (east Turkey + west Iran) origin is consistent with Dienekes rolloff experiments with north-east Europeans:

http://dienekes.blogspot.com/2012/10/the-indo-european-invasion-of-baltic.html

Basically I been saying for years, that Uruk expansions resulted in Maykop culture. Maykop itself is pre-proto-Indo-European, it has everything, including earliest Kurgans and advanced metallurgy. Those dark pigmented Kurganids recently tested may have been Caucasus people who descended their mountains and hit the open steppe.

Unknown said...

In the green coloured maps, apart from the norwegian map, the rest show that the origins of those markers where all around the borders where belarus, ukraine and russia meet.
The contour/isobar lines clearly show this.

The colour is only a way of showing the density of marker and not it's origin

eurologist said...

I still think the Afghanistan/ Kyrgyzstan region is a much more likely home for R1a.

Interesting to see Germans to the right of most E Europeans on PC1 in Fig. 4 - but the authors should have been wiser and not included Jews (because of their heritage that sheds little light on the region < 2,000 ya) also from this analysis.

As we have long known, also from ancient DNA, the expansion of R1a in eastern (R1a-Z282*), north-central (much of R1a-M458), pan-central-eastern (much of R1a-M558), and northwest (R1a-Z284) Europe predates Slavic expansion by millennia. Had the authors dug a bit deeper below these sub-haplogroups, it would have made it relatively easy to calculate the percentage of male population change due to Slavic expansion - which I previously estimated to about 10-15% in the Baltics and N and E Poland.

The Z93 distribution maps don't exclude either herding cultures nor PIE expansion, to me.

The R1a-R1b splintering estimate of ~25,000 ya seems a bit recent, given that we have the eastern Siberian Mal'ta example of R in a Gravettian context, at that time. I would put that split closer to the beginning of the Gravettian, 30,000 to 28,000 ya.

The R1a-M417 splintering (before the split between all of Europe and Asia!) estimate of ~5,800 ya would fit better with the beginning of the split between Pontic agriculture and Eastern step herding, some time before either Corded Ware or PIE origins.

Va_Highlander said...

andrew:

"I am not inclined to dismiss the possibility of a PIE expansion out of Iran possibility out of hand either."

Agreed.

"PIE expansion appears to have involved superior metallurgy seemingly inconsistent with steppe pastoralists, and PIE vocabulary seems awfully rich in agricultural vocabulary for a population of steppe pastoralists. Mountain farmers and metal workers might be a better fit."

Again, I agree and superior metallurgy may in fact point to Iran or its greater cultural sphere.

"And, while an "Out of India" theory seems disfavored by the evidence, I don't think that we should prematurely rule out the possibility that PIE was part of a linguistic family that included the more civilized copper age centers of Persian, Harappa, and Bactria."

I think that stands to reason and agrees with the material record we have, scant on the ground though that is. I might suggest adding Margiana and the Zarafshan Valley to that list, as well.

Rokus said...

@Davidski,
'The other Northwestern European subclade is CTS4385. This marker isn't part of Z282, but there's no reason why it couldn't have arrived from the east during the Copper Age as M417. In fact, the Corded Ware R1a from Eulau appears to be CTS4385.'

How you can be so sure? I can find plenty of similar haplotypes to Eulau in all major M417+ subclades, at least for dys439=10 and dys385a=11 that are most typical also for most of the 10 presented R1a-M417* in table S3. This suggests the Eulau profile rather presents an ancestral pattern.

Unfortunately, Table S3 only represents 1355 haplotypes (< half) out of Underhill's 2923 R1a-M420 Y-chromosomes. One Eulau look-alike is found in India, what comes as a surprise given that India was IE-ized more recently - suggesting M417* was still involved in early IE expansions. Star-like branching occurred especially near the root of the Asian subtree, though even this Indian M417* is not unlike a couple of M558 haplotypes of Slovakia.

Half of the presented R1a-M417* haplotypes were found in the Northsea region. This can't be exactly due to sampling bias: although the table doesn't give a clue about 95% the Dutch mere 87 participating haplotypes (<3%), this rare ancestral haplogroup would correspond to over 2% of Dutch R1a haplotypes!

Population dynamics typically render the replacement rate of YDNA haplogroups quite impredictable, and old haplotypes may be preserved most in isolated and decentral regions. Was Northwestern Europe such a region? At least in the sense that foreign admixture was rare, according to a recent study of Hellenthal et al. For sure the Caucasus was an isolated region, what might explain why all "Iranian" M420* and SRY10831.2* where actually Azeris! But SRY10831.2* elsewhere didn't disappear from the face of the earth. The Brabant project revealed a quite less divergent SRY10831.2* individual in eastern Netherlands, and although his background and name strongly suggest Polish/Prussian roots I'd rather implore caution about the origin question. The MRCA for this individual with M417 definitely exceeds the estimated diversification downstream M417 at 5800 years ago! In other words, it is quite unlikely that R1a was a foreign marker for Corded Ware. Most probably it was already around at the time of their local forebears. Eulau was just an early or intermediate stage, and can't be dismissed as an introgression.

andrew said...

Margiana was on the tip of my typing fingers but the word wouldn't come to me.

I'm not familiar with the archaeology or Holocene pre-history of the Zarfsahn Valley. What was going on there?

Anonymous said...

Eurologist: "The Z93 distribution maps don't exclude either herding cultures nor PIE expansion, to me."

Indeed. Actually, this study shows the Altai/south Siberian R1a to be Z93*, if I made no mistakes.

We _KNOW_ that these people arrived during chalcolithic from south Russia (common morphology, light pigmentation, copper (and even silver) metallurgy, kurgans, early Yamnaya potteries and cultual objects and axes too, pastoralism (all this also goes well with west cattle DNA in Mongolia and beyond), almost fully R1a male lineages and west female lineages in the oldest aDNA in the region).

Z93* in chalcolithic south Russia (apparent origin of the Afanasevo population of south Siberia, as seen above) really yells Kurgan theory validation.

Rob said...
This comment has been removed by the author.
postneo said...

the reason archaic r1a survived in western europe and the middle east is because population sizes were small for various reasons. It does not indicate origin.

Having said that, maybe kurdish areas are the best candidates. In western Iran and eastern turkey r1a was gradually replaced by other lineages. This actually helped preserve older lineages. In a place like england archaic r1a was preserved because the founding populations were small. They were migrants from the east in an archaic age when populations were small.

Va_Highlander said...

Kapak:

"Basically I been saying for years, that Uruk expansions resulted in Maykop culture."

I think you'll find that the Uruk expansion is too late to have triggered Maykop.

"Maykop itself is pre-proto-Indo-European, it has everything, including earliest Kurgans and advanced metallurgy."

Modern archeology suggests that the practice of kurgan burial actually developed in the Leylatepe Culture, south of the Caucasus in Azerbaijan. Leylatepe, in turn, may have resulted from the Northern Ubaid expansion.

Hector said...

*ed haplogroups(C1* K* etc) are possibly paraphyletic groups. They are not "ancestral". Each of them is as derived as any other when the whole Y-genome or at least a significant portion of it is fully sequenced.

To anyone who does not understand the above basics, I suggest he not even bother reading my post.

According to the best estimate of our time the rate of SNP occurrence is about once per 1-3 generations depending on the mutation rate you choose. This means genealogical MRCA is roughly the same as MRCA in terms of SNPs.

Pick anyone from R and another from Q, if you go back far enough there was a man who is the MRCA of these two. He had two sons, one leading to Q and the other leading to R eventually. When Q and R are sufficiently expanded to include all P*s that are descended from each of these two sons, these two sons represent Q and R respectively in the genealogical sense.

That is what I meant they are brothers in the literal sense.

Now consider MRCA of MP. Also consider MRCA of all M526+. If MRCA of MP was in West Eurasia and his ancestors never had been to east of South Asia previously, it stands to reason that there were at least 2 (M and another M526+) that migrated from Middle East all the way to Southeast Asia.

But unless M526+ has some kind of overwhelming adaptive advantage it is highly unlikely that they,not I not J, not F,not G, not H are the only ones who made such successful trips.

This is bad enough as it is but I assumed that all East Eurasian non MP M526+ descended from a single line separate from MP. If the tree topology of M526+ is more complicated than this, you need to assume not 2 but multiple lineages all making independent trips from middle east to east eurasia as if they were zombies whose foreheads were stamped with a seal "Go East and prosper".

That is hilarious.

Kapak said...

Va_Highlander:

Uruk Expansions are just around the same time Maykop begins to flourish, so I don't see how chronology is too much of a problem here.

By the way, Trialeti culture of Georgia precedes the Leylatepe culture. Here is the wiki summary of it:

""The Trialeti culture was a second culture to appear in Georgia, after the Shulaveri-Shomu culture which existed from 6000 to 4000 BC. The Trialeti culture was known for its particular form of burial.[7] The elite were interred in large, very rich burials under earth and stone mounds, which sometimes contained four-wheeled carts.[7] Also there were many gold objects found in the graves.[4] These gold objects were similar to those found in Iran and Iraq.[3] They also worked tin and arsenic.[8] This form of burial in a tumulus or "kurgan", along with wheeled vehicles, is the same as that of the Kurgan culture.""

As far as having the first Kurgans it goes Trialeti>Maykop>Yamna

@ Vooruit:

How can those light Altayans came from south Russia when the Yamna culture of south Russia was overwhelmingly dark pigmented? It's clear Andronovo and maybe Afanasevo has different origins, rather than emerging from south Russia they likely came from the forests of the upper Volga (now inhabited by various Finnic and Turkic speakers).

Rokus said...

If the tree topology of M526+ is more complicated than this, you need to assume not 2 but multiple lineages all making independent trips from middle east to east eurasia as if they were zombies whose foreheads were stamped with a seal "Go East and prosper".

I know a "comical" female teacher of architecture that insists on something similar for R1b-U106, R1b-P312 and R1a, of Indo European horseriders that each brought their share through a different route to Northwestern Europe. The zombies are the ones that subscribed to her blogs and commit cannibalism on those having a living brain. However, there is one difference. The time depths of YDNA MP/K* migrations are vastly superior and allow for the migration of single ancestral strains, while this is insanity in the case of Kurganism. I am reasonable sure the K* branches of Oceania/SE Asia will soon be recognized as close relatives of YDNA MP, that spread in a single expansion. What of course doesn't mean there only was a single expansion, this being another matter.

terryt said...

"I assumed that all East Eurasian non MP M526+ descended from a single line separate from MP".

That is far from what the currently accepted phylogeny shows.

"If MRCA of MP was in West Eurasia and his ancestors never had been to east of South Asia previously"

You're ignoring the fatc that MP is part of MNOPS from which NO is definitely eastern and S is definitely SE Asian at least, and probably Wallacean. Another factor to take into consideration is that MNOPS itself is part of the K1-P60, K2-P79, K3-P261 complex, all of which are SE Asian or Austalia/New Guinea.

"you need to assume not 2 but multiple lineages all making independent trips from middle east to east eurasia as if they were zombies whose foreheads were stamped with a seal 'Go East and prosper'".

As you point out that is extremely unlikely. Because P is the odd one out of the whole clade it is surely far more likely to be the branch that moved rather than requiring all the others to have individually moved all the way 'from middle east to east eurasia'. I think you are obsessed with a Middle Eastern origin for the whole of humanity. Influenced by biblical sources?

eurologist said...

(@Davidski:)
"In your colourful theory that, both IE and Slavic originated in Poland, how do you account for the sharp genetic discontinuity between extant Germans and Poles..."

Rob,
Actually, such a sharp boundary does not exist. Along the Rhine river, the vast majority of males are R1b, around the Vistula R1a. There's a very wide gradient both ways, with "I" making up around 25-30% in the middle (nowadays ~ Eastern Germany). While the exact location of the gradients and proportion and e.g. areal size of "I" have changed over the millennia, we know from ancient DNA that this picture was pretty much the same at least 3,500 ya, and perhaps also during Corded Ware.

Haplogroup I likely used to be more prevalent and distributed more broadly, but got eaten up due to Frankish expansion from the west, Slavic expansion from the east, and earlier Roman and Celtic excursions, among others. So, the wide gradient of R1b and R1a surely is not only due to diffusion, but also due to expansion waves - while some peoples like East Germanic tribes were almost completely dissolved, with few descendants left in their former region.

Davidski said...

Dr Rob,

Ever heard of the Corded Ware Culture and Bell Beakers? Doesn't sound like it, but here's a little clue that might help you to figure out that riddle you posed: it's likely that one of them was mostly R1a and the other mostly R1b.

By the way, how's that theory of yours coming along that Polish R1a is of Avar origin? Does the data in this paper support it? If you actually say yes, then I'll finally know that you're posting here from a psychiatric ward, but not as a doctor.

Rokus,

You're not listening to what I'm saying. The Northwestern European R1a you're talking about is almost all L664 (the STRs in the supp data from this paper prove it). This marker is under M417, so as per the dating of M417 based on complete R1a sequences, it's no older than the late Neolithic. So it probably got to Western Europe with the Kurgan expansions.

Apart from that, some of the posters here are very likely high on the same barbiturates.

Do you people realize that we have an ancient genome from Siberia that belonged to Y-HG R? The close relatives of this sample, known as Mal'ta boy or MA-1, contributed DNA to modern Europeans in a couple of waves. One of these waves came before the Neolithic, and influenced Northern Europe, and another wave came after the Neolithic, and influenced most of Europe except places like Basque country and Sardinia.

That second wave of ANE very likely brought R1a from what is now Russia to Europe. In fact, ANE and R1a probably also spread at about the same time to places like Iran and India, including the ancestor of that Iranian M420* haplotype.

So no, R1a doesn't come from Iran. It got there with ANE and very likely the Proto-Indo-Iranians from the steppe to the north.

Grey said...

@andrew

"PIE expansion appears to have involved superior metallurgy seemingly inconsistent with steppe pastoralists, and PIE vocabulary seems awfully rich in agricultural vocabulary for a population of steppe pastoralists."

There are the Vinca copper-working culture and Cucuteni agricultural culture west of the Black sea that could have provided the source for those thing.

As those two cultures seem to have been the earliest victims of PIE expansion I'd say they were the likeliest candidates.

.

@Dr Rob
"theory that, both IE and Slavic originated in Poland, how do you account for the sharp genetic discontinuity between extant Germans and Poles, in Y -DNA terms. Ie sharp inversity of R1b/ R1a"

If the expansion of PIE displaced people to their west and pushed them further west that would explain the sharp discontinuity.

.

"And keep in mind, the lack of any geographic barriers between Germany and Poland to account for this, situated as they are on the north European plain."

Maybe not. If you look at it in terms of ecozones there may have been - and I think will have been - a sharp discontinuity in ecozones on the north European plain marked by the reach of the warming Gulf Stream effect.

.

I think the (currently) most likely option is the PIE existed for a long time alongside Cucteni and then when they expanded they displaced them west. One branch went the northern route and another branch took the southern route going
->Vinca
->unknown
->northern Italy
->southern France
->Iberia
->Atlantic coast
with the two branches rejoining around Holland / Denmark.

terryt said...

"I assumed that all East Eurasian non MP M526+ descended from a single line separate from MP".

You're beginning to sound like German: completely dismissing the currently accepted phylogeny because it doesn't fit what you want to believe.

"If the tree topology of M526+ is more complicated than this"

Which the current phylogeny shows to be the case.

"you need to assume not 2 but multiple lineages all making independent trips from middle east to east eurasia as if they were zombies whose foreheads were stamped with a seal 'Go East and prosper'".

On the contrary, you need to assume just one independent trip from East Eurasia to the Iranian Plateau. You mentioned elsewhere that haplogroups usually form subgroups in pairs rather than in multiples. I agree. P ultimately had two sons: Q and R. M can be most likely be said to have had two sons also: M1 and M3. M2 almost certainly has a connection with one or other of those haplogroups as it suddenly appears in western Polynesia and seems associated with the 'Polynesian Outliers'. Whether MP's brother is NO os S is a mystery at this stage. I'm inclined to thing it will turn out to be NO by reason of the different distributions of M and S in New Guinea/Melanesia. I accept the K1, K2, K3 complex may be a brother to MNOPS. K(xLT) is obviously brother to LT, which also had two sons: L and T. And KLT was brother to IJ. With their father IJK we find ourselves in the Middle East somewhere. Hope that provides a framework for haplogroup pairs.

Hector said...

Extending the principle that I mentioned in the previous post it seems odd that all 3 prevalent Northern Eurasian haplogroups R1a R1b Q are all P-descended.

Thus it is more parsimonious to propose that their dispersal originated from some kind of frontier region where their ancestral P gained a measure of dominance if not fixture.

So Central Asia is more likely than Iran or Turkey. That region was repeatedly glaciated and it is natural that the inhabitants migrated to somewhere else or were wiped out.

The Malta boy with his R(xR1 xR1)corroborates this view.

Also the paper on Malta was heavily flawed. His West Eurasian affinity is not uniform but tends to 0 toward the western end of the continent. You should expect something else if the affinity was deep and cognate.

Also Han Chinese were the only representatives of East Asians. I am sure they were scared that if Mongols or Northeast Asians were assayed they would not get the desired result(or they would be forced to explain it away as some kind of admixture).

As David Reich and others pointed out it bespeaks of a population that contributed to the later West Eurasians and almost none to East Asians. The population was not born as West Eurasians but their descents became one.

Simon_W said...

I would associate the „Norwegian“ Z284 with Unetice-related influence. The Nordic bronze culture had strong ties with the Unetice culture. The other branches in Z282 (including Z280, which according to the R1a project at FTDNA isn't ancestral to Z282, but the other way round) appear to be primarily related with Balto-Slavic populations. (NB: Even though the Slavic expansion took place late, and from a relatively small area of origin, Balto-Slavic populations in general had been occupying a vast region for very long times.)
I'm currently thinking that Celtic populations are also ultimately derived from an R1a-dominated group, and that must have been in CTS4385. In central Europe, they mixed with I2 + R1b mixed groups and moved on to the already R1b dominated west of Europe.

As for the PIEs and their basic splits: I think it would appear realistic that they originated where Gamkrelidze and Ivanov suggested. Then the Anatolian branch moved west to Anatolia and the Aegean; the Tocharo-Celto-Italic branch moved east along the Silk Road to Central Asia and southern Siberia; and the Remainder moved north across the Caucasus and spread more or less along the usual Kurgan theory lines. The original PIEs may well have been dominated by y-haplogroup J2, but if so, they soon and strongly influenced R1a and R1b people.

By the way, here the link to the R1a1a tree of the FTDNA project, it differs from the tree suggested by Underhill et al.:
https://www.familytreedna.com/public/R1a

Va_Highlander said...

andrew:

"I'm not familiar with the archaeology or Holocene pre-history of the Zarfsahn Valley. What was going on there?"

I was thinking of Sarazm, an early center of relatively advanced metallurgy. Perhaps especially relevant in this context, in the late 4th and early 3rd millennia we find at Sarazm an imported ceramic assemblage with clear parallels in northeast Iran, the Helmand basin of Afghanistan, and Baluchistan.

terryt said...

"I am reasonable sure the K* branches of Oceania/SE Asia will soon be recognized as close relatives of YDNA MP"

They are already 'recognized as close relatives of YDNA MP'. MP is part of MNOPS, united with the Ks but M526. Called K(xLT) in the ISOGG tree. Even 'Phylotree', where the MP grouping was first mentioned, has that connection:

"that spread in a single expansion".

Yes. And most likely from the region where the majority of them a found. Otherwise you're postulating multiple separate migrations in the same direction from the same place. While I admit that is possible it seems unlikely to me.

"Thus it is more parsimonious to propose that their [all 3 P-descended prevalent Northern Eurasian haplogroups R1a R1b Q] dispersal originated from some kind of frontier region where their ancestral P gained a measure of dominance if not fixture".

I agree, but their history before arriving at that 'frontier region' is more difficult to discern.

"The population was not born as West Eurasians but their descents became one".

I'm sure that is correct.

Grey said...

@Hector

"Extending the principle that I mentioned in the previous post it seems odd that all 3 prevalent Northern Eurasian haplogroups R1a R1b Q are all P-descended.

Thus it is more parsimonious to propose that their dispersal originated from some kind of frontier region where their ancestral P gained a measure of dominance if not fixture."

In and out of America chasing mammoth.

P bottleneck Q
Q bottleneck R

Rokus said...

Davidski,

'You're not listening to what I'm saying. The Northwestern European R1a you're talking about is almost all L664 (the STRs in the supp data from this paper prove it). This marker is under M417, so as per the dating of M417 based on complete R1a sequences, it's no older than the late Neolithic. So it probably got to Western Europe with the Kurgan expansions.'

I just reject your conjectures. You try to derive the Eulau haplogroup from its STRs, while these are demonstably ancestral: similar STR patterns are abundantly available in ALL OTHER MAJOR M417+ SUBCLADES. This is the logical consequence of a star like pattern and recent expansion, I wonder who's the one that doesn't listen!
Maybe L664 originated in the late Neolithic, about the time R1a already existed in Eulau. Less reason to consider L664 part of some Kurganist scheme. Moreover, since Eulau was clearly part of the Corded Ware culture, not of some ill defined "Kurgan" horse tribe, there is no reason to resort to Kurgan expansions at all.

BTW, Mal'ta Boy predated Corded Ware for over 20,000 years. He never saw a Kurgan.

eurologist said...

As I have mentioned earlier, I find the extreme NW of the subcontinent the most likely place harboring P that eventually resulted in a Q and then R* wave East, and an R1 wave west. Firstly, it is easy to get there from western SE Asia where K(xLT) likely split, perhaps after arrival at the S Brahmaputra/ Padma rivers area. Secondly, the Upper Indus River Valley is a climatic and geographically confined isolate with mild whether, above-average cloud cover, and plenty of water resources. Thirdly, whenever winters and springs are milder and rainier, huge resource-rich grasslands open up in the north from the European plains all the way to S Siberia and C Asia. Remember, such a climatic anomaly was also what provided the Mongols with so much richness and enabled their expansion in the first place.

Since Mal'ta boy has Gravettian association, and Gravettian is ~32,000 years old, this makes it likely that the major R* descendants such as also R1 are roughly 30 - 35 ky old, with the R1a/R1b split following shortly thereafter, almost certainly before LGM, with apparently R1a settling a more northerly region.

It is a bit weird that the Gravettian apparently spread into C and W Europe without haplogroup R, but pending surprise ancient DNA finds it is currently hard to argue against that. However, NE Europeans likely were R1a from then on, and perhaps there was an R1b enclave on the western Pontic region, or R1b got there from e.g. the Transcaucasus during the early Metal Ages. My hypothesis is that all European PIE originated from there, although the language might have gotten there from the S Pontic (Anatolia) or E Pontic (Transcaucasus) regions, or simply was a Copper Age/ early Bronze Age pan-Pontic Lingua Franca.

Generally, I prefer an early PIE expansion into C and NC Europe with the climate change to much more arid conditions, i.e., with the Globular Amphora Culture. I wouldn't even be surprised if LBK had a sister language to pre-PIE, which would explain how PIE could have been so easily adopted over such a wide area.

R1b analysis shows that the European and Anatolian subgroups are separate, with their predecessor now found in the Balkans. This actually fits a PIE expansion model from the W Pontic W and S and then E, while the N, NE, and E expansion was carried by the N Pontic R1a people (who were in strong trade contact with the culturally more "advanced" W Pontic people).

Rob said...
This comment has been removed by the author.
eurologist said...

"By the way, Trialeti culture of Georgia precedes the Leylatepe culture"

Kapak,

Wikipedia also says:

"The Trialeti culture, named after Trialeti region of Georgia, is attributed to the first part of the 2nd millennium BC."

Instead we have:

Maykop: ~3,700 - 3,000 BC
Yamna: ~3,600 - 2,300 BC
G. Amphora ~3,400 - 2,800 BC
Kura–Araxes ~3,400 - 2,000 BC

Since Maykop is contemporaneous with Yamna and Globular Amphora and Kura–Araxes, I surmise that neither one coincides with the origin of PIE, but all are signs of its spread after severe climate change. While the "Kurgan" burial practice may have arisen in the Transcaucasus, all evidence shows that this did not enter C and NC Europe via the Yamna culture or "Kurgans" - and with great certainty definitely not so east of the Vistula.

We must look for a PIE origin before these cultures for many reasons, and one that was well-established on the prosperous W Pontic region before ~ 4,000 BC, but with likely connections to Anatolia and the Transcaucasus.

Va_Highlander said...

Kapak:

"Uruk Expansions are just around the same time Maykop begins to flourish, so I don't see how chronology is too much of a problem here."

Maykop begins as early as 4,000 BCE, some centuries before the Uruk expansion.

"By the way, Trialeti culture of Georgia precedes the Leylatepe culture. Here is the wiki summary of it"

Apparently, the Trialeti culture does not precede Leylatepe, which might have been clear had you quoted the first two sentences of that Wiki summary:

"The Trialeti culture (Georgian: თრიალეთის კულტურა), named after Trialeti region of Georgia, is attributed to the first part of the 2nd millennium BC. In the late 3rd millennium BC, settlements of the Kura-Araxes culture began to be replaced by early Trialeti culture sites."

Thus, if Wiki may be believed, Leylatepe and Maykop precede Trialeti by as much as a thousand years.

Hector said...

The model that I discussed first - MP and another M526+ that led to all other M526+s, NO S K1 K2 K3 etc. - was the one that was designed to be most favorable to Rokus. And I showed even in that case a Middle Eastern origin of MP is not likely.(just barely possible, but not the most parsimonious)

Odds are very high though that this is not the case and in such cases the middle eastern origin of MP becomes even less likely. Actually comical.

Terry does not seem to understand my point so I felt the need to clarify.

Grey said...

As they seem to be the first victims of the earliest PIE expansions it seems to me the obvious question should be where did Cucuteni (agriculturalists) and Vinca (copper smiths) go?

I wonder if small groups of wandering (R1b?) refugee copper smiths from the Balkans might be relevant to later events somehow?

terryt said...

"Terry does not seem to understand my point so I felt the need to clarify".

I'm fairly sure I understand it but even if some undiscovered mutation serves to unite haplogroups previously considered separate that mutation must have occurred in a single individual. You would still have haplogroups spreading from a single region.

"I showed even in that case a Middle Eastern origin of MP is not likely.(just barely possible, but not the most parsimonious)"

I agree.

"Since Mal'ta boy has Gravettian association, and Gravettian is ~32,000 years old, this makes it likely that the major R* descendants such as also R1 are roughly 30 - 35 ky old"

That's what this post says (well, P actually):

http://dienekes.blogspot.co.nz/2012/08/dates-of-major-clades-of-y-chromosome.html

Kapak said...

@ VA highlander

Yeah that wiki article was a bit misleading, my mistake I got a bit excited (it mentioned two different dates for Trialeti). But at least Maykop Kurgans came before Yamna ones, that goes to show who started what first.

eurologist said...

Grey,

IMO Vinca's decline is too early to be an effect of steppe people expansion. Cucuteni fits the climate change argument well, in many respects --- Vinca perhaps.

However, both could have contributed to PIE, since the W Pontic region was extremely populous and culturally advanced. As I said before, we must look for populous and advanced peoples and regions around the Pontic before 4,000 ya to better understand PIE origin.

Va_Highlander said...

Kapak:

"Yeah that wiki article was a bit misleading, my mistake I got a bit excited..."

It is. I read it twice, before noticing. Thanks for the mention. I hadn't heard of that culture, before.

"But at least Maykop Kurgans came before Yamna ones, that goes to show who started what first."

I don't think there is any doubt about the direction in which the custom spread. It is interesting, though, that it began in Transcaucasia, and not closer to the steppe.

matt said...

The "automation" of finding SNPs is evolving painfully with the BigY false positive, false negative, no calls and so forth. Someone else can tell me which will have the higher standard deviation on estimating time, a one in a million per generation SNP or 10 one in a 10,000 STRs. For bushy yDNA trees like R1 of course SNPs are a big gain.

Grey said...

@eurologist

"IMO Vinca's decline is too early to be an effect of steppe people expansion"

I agree but disagree in that I think the earliest "expansion" was unlikely to be actual expansion in the usual sense. I think when they first got their horses it made their **raiding** range much wider and more destructive.

I don't think the Cucuteni / Vinca were expanded over or conquered in the normal sense I think they may have been driven away by raiding.

https://www.youtube.com/watch?v=d864bwyCAoA#t=1910

500-600 abandoned settlements

no settlements for 500-700 years

invasion for conquest or displacement through raiding?

eurologist said...

Grey,

Just remember that there is absolutely no indication of "Kurgan" expansion west of the Vistula, and that horses in general were useless and expensive and completely unimportant west of the Vistula until Roman times.

Horses require high-quality and high-protein fodder (like oats) that compete with humans, and need ~5 months of warm shelter and high-quality fodder during the winter in regions with significant snowfall - which at the time was not available nor affordable in northern central and NW central nor N Europe. Also, the wetlands and dense forests and lack of "roads" in Europe prevented any type of horse- or wagon-driven invasion. Even much later, Romans had a hard time moving about in what is now Germany, and more or less successfully used water ways and their banks to transport soldiers and goods, for the most part.

Anonymous said...

Kapak : "How can those light Altayans came from south Russia when the Yamna culture of south Russia was overwhelmingly dark pigmented?"

You're thinking of the bronze age samples from Ukraine. The ancestors of Afanasevo would have been from the eastern tribes east of the Volga.
The admixture between the population from south Russia (up to Samara) and Ukraine might have been rather weak.


"It's clear Andronovo and maybe Afanasevo has different origins, rather than emerging from south Russia they likely came from the forests of the upper Volga (now inhabited by various Finnic and Turkic speakers)"

I see no reason to think that since they had the complete panoply of an early eastern Yamnaya population (thus a Russian Kurgan culture).

Rob said...
This comment has been removed by the author.
Va_Highlander said...

eurologist:

"Just remember that there is absolutely no indication of "Kurgan" expansion west of the Vistula..."

There are no kurgans south of the Oxus, either.

"Horses require high-quality and high-protein fodder (like oats) that compete with humans..."

My father, who grew up on a farm in the 1930s, claimed that you need about 10 acres (4 hectares) of farmland for every horse. They are indeed very resource-intensive.

In fact, there are multiple problems with the proposed horse-and-rider or horse-and-chariot models for I-E expansion. Chariots were apparently quite rare on the Eurasian steppe and there is little if any evidence that they served in some military capacity. Controlled horsemanship seems to have first appeared in Iran and on the steppe near the end of the 2nd millennium BCE and riders were not adapted to warfare in the Near East until the 8th century BCE.

Kapak said...

@ vooruit
So essentially you are in agreement with the conclusion of that paper, that "extreme rapid positive selection for light pigmentation"?

Yamna precedes Sintashta by a millennium, yet Yamna = dark pigmented and Sintashta = light pigmented. Means either 1. within 1,000 years Sintashta changed from dark to light rapidly or 2. Yamna and Sintashta were different people, genetically.

Grey said...

@eurologist

1) I never said anything about west of the Vistula or Germany.

I am talking about the extent of the Cucuteni culture which was **adjacent** to the PIE range near the Black Sea.

A culture that **disappeared** around the same time PIE were developing a horse culture.

http://en.wikipedia.org/wiki/Cucuteni-Trypillian_culture


2) Read again what you said about the amount of grazing land needed to support large herds of animals in the context of the quotes from the David Anthony lecture in the youtube video describing how very large areas along the west of the Black Sea were depopulated for hundreds of years.

https://www.youtube.com/watch?v=d864bwyCAoA#t=1910

Grey said...

@Rob

1. Was the Cucuteni culture directly adjacent to PIE?

2. Did Cucuteni and PIE interact for a very long time?

3. Did Cucuteni disappear around the time PIE developed riding horses?

http://en.wikipedia.org/wiki/Cucuteni-Trypillian_culture

The answers my primitive understanding gives me are yes, yes and yes.

Perhaps you can refute any of those points?

4. David Anthony's lecture on youtube presents evidence that there was large scale **depopulation** - nb not expansion but depopulation - of settled populations in regions adjacent to PIE?

https://www.youtube.com/watch?v=d864bwyCAoA#t=1910

Perhaps you can refute that?

@VA-Highlander

"you need about 10 acres (4 hectares) of farmland for every horse. They are indeed very resource-intensive."

That would be a good reason for expanding grazing land then.

"In fact, there are multiple problems with the proposed horse-and-rider or horse-and-chariot models for I-E expansion."

I'm not talking about cavalry armies or chariots or expansion in the sense of mass population movements. I'm talking about horse-borne raiding creating de-population in the same way Moorish sea-borne raiding depopulated the Mediterranean coast in the middle ages.

Long before cavalry or chariots, ride-able horses would have given greater range and an improved getaway vehicle for **raiding**.

eurologist said...

Grey,

Nothing personal, here - just trying to dispel myths that have been around for way too long, especially in the Anglo-American swamp of lack of context and education.

I never said anything about west of the Vistula or Germany.

The problem is that "Kurgan expansion" is projected into not only eastern or central and NW central Europe, but pretty much all of Europe, without a shred of evidence. So you just suffered in the crossfire against that alchemy.

I am talking about the extent of the Cucuteni culture which was **adjacent** to the PIE range near the Black Sea.

A culture that **disappeared** around the same time PIE were developing a horse culture.

http://en.wikipedia.org/wiki/Cucuteni-Trypillian_culture


There is a weak disagreement whether W Pontic cultures declined due to harsh climate change, or due to invasions. IMO, pretty much the entire evidence favors climate change, and there is zero evidence of invasions. Were left areas re-populated by peoples more adapted to dry steppe conditions? Surely, but most likely not immediately, and most likely not immediately successfully, so.

Simon_W said...

@ Davidski

The close relatives of this sample, known as Mal'ta boy or MA-1, contributed DNA to modern Europeans in a couple of waves. One of these waves came before the Neolithic, and influenced Northern Europe, and another wave came after the Neolithic, and influenced most of Europe except places like Basque country and Sardinia. 

In fact, the Basques are also somewhat MA-1 shifted, relative to the early farmers and western hunter-gatherers. Obviously that must be related to them being predominantly R1b.
Interestingly though, they are not West Asian admixed, which speaks against a West Asian origin of their R1b.

@ Eurologist

the W Pontic region was extremely populous and culturally advanced. As I said before, we must look for populous and advanced peoples and regions around the Pontic before 4,000 ya to better understand PIE origin.

Sometimes even the languages of culturally advanced, populous regions vanish into thin air, as e.g. happened to Sumerian.

horses in general were useless and expensive and completely unimportant west of the Vistula until Roman times.

But the Gaulish and Germanic cavalries had quite some reputation in the eyes of the Romans, and the Britons still made use of war chariots. Sure, all these armies were predominantly infantry, and the shift to the predominance of knights happened only in Carolingian times – but horse riders were not completely unimportant before that. I also remind you of the cultural significance of horses in pre-Christian times, as evidenced e.g. by the Irish horse sacrifice, the Celtic horse goddess Epona, or the stallion idols of the Terramare culture.

Of course aeneolithic and bronze age warfare was primitive by absolute standards, but all you need for one group to dominate over another group was some relative advantage.

As for Kurgans west of the Vistula, it's clear that the term „Kurgan culture“ is about as valid (or rather not valid) as the term „Megalithic culture“. It's more appropriate to speak of the Yamnaya culture and other cultures. But since the Corded Ware's typical burial type was a kurgan, it belonged to the „Kurgan cultures“ in the wide sense. Do you want to suggest that this custom reached C Europe via SE Europe?

Simon_W said...

There are a few lines of evidence now, suggestive of some seizable eastern influx in late Neolithic to Bronze age Europe:

1. The appearance of the rather eastern y-HG R1a in the Corded Ware people.
2. The appearance of eastern European mt-DNA U2 in the Corded Ware people. I would also add K2b which was found in Bronze age South Siberia, but which is now most common in central to western Europe.
3. The striking increase in Ancient North Eurasian autosomal admixture at around the same time.

4. I would bet that the dramatic increase of the North European autosomal component and the proportional decrease of the Mediterranean component, that must have occured since the days of the early farmers, was also related to these events.
Because we have to be careful not to overestimate the significance of the observed reappearance of „hunter-gatherer“ mt-DNA in the Upper Neolithic cultures. After all, the Basques also have lots of hunter-gatherer mt-DNA, yet they are autosomally one of the most Mediterranean peoples of the world! Also remember that both Ötzi and Gok4 are actually not early farmers at all. Ötzi falls within the late Neolithic. And Gok 4 belongs to the Nordic middle Neolithic, but is in absolute terms even 300 years later than Ötzi. So even though both are treated as „early farmers“ by the geneticist community, they actually aren't. The only autosomal sample so far that's really early neolithic is Stuttgart, the LBK farmer. I think it's likely that previous to the eastern influx I'm arguing for, much of Europe was autosomally Basque- and Sardinian-like.

I think while we cannot be sure about the effect this eastern admixture had on language, it certainly seems possible that it induced language change.

In the case of Southeastern Europe it's also noteworthy that during most of the Neolithic, with its advanced, flourishing, populous cultures, the people were rather small and gracile; the observed late Neolithic increase of robust Cromagnoid people was presumably related to Steppe influence.

I have toyed with the idea that only the predecessor of Balto-Slavic and Germanic was introduced from the east, while the old Balkanic languages arrived from Anatolia to SE Europe. But it's difficult to imagine, since there is quite some continuum there: Albanian isn't that far from Slavic, and Armenian isn't that far from Albanian, and then Greek is related to Armenian.

Rokus said...

'But since the Corded Ware's typical burial type was a kurgan, it belonged to the „Kurgan cultures“ in the wide sense.'

Completely ridiculous, you can't lump all graves similar to a barrow together as if it involves one single culture, or group of cultures. Have you ever seen an Atlantic Megalithic grave? Definitely not flat, instead those people responible put a lot of effort in raising impressive hills on top of their graves. "Kurgans" too? This kind of simplifications are exactly the reason why I despise Kurganism so much. Actually, the Beaker folk in the Netherlands still used and were continuous to Funnelbeaker gravehills. Nothing to do with Kurgan pseudoscience, especially since their beakers had a Mesolithic continuity (Swifterbant).

Grey said...

@eurologist

"Nothing personal, here - just trying to dispel myths that have been around for way too long, especially in the Anglo-American swamp of lack of context and education."

Apologies, I was testy over something else.

I appreciate the Kurgan, Hunnic horde type model doesn't apply. I'm wondering about a raiding no man's land developing around the border of the PIE range after the domestication of horses i.e. a few hundred Black Sea Cheyenne on ponies raiding along the west shore of the Black Sea every year after the spring lambing.

My understanding is Cucuteni / Vinca had early agriculture, early (earliest?) pottery, early (earliest?) copper working, early (earliest?) script etc and all directly adjacent to PIE before PIE domesticated horses.

That's kinda interesting.

For example if Cucuteni / Vinca were displaced (not replaced) by PIE then you might expect to see evidence of groups of refugee copper smiths wandering about.

Grey said...

Apart from the recorded instances of sea-borne raiding leading to depopulated coastal zones another historical analogy might be the expansion of the Plains Indian tribes after they got horses.

Simon_W said...

Rokus, to me it's a fact that the Corded Ware burials were essentially kurgans. They were definitely not megalithic graves. You may question the causal connection to the Yamnaya kurgans – perhaps the similarity was merely a coincidence. But even if that was true, they were rather similar. It's not just the barrow, also the principle: A single grave under a tumulus, with the dead in crouched position and with weapons as grave furniture. The transition from collective to single graves was revolutionary. (Well, it didn't happen over night, but still...) Even the construction of the wooden chamber was often similar. The main difference, to my knowledege, was the orientation of the dead, and the lack of sex dependend orientation in the Yamnaya graves.

Va_Highlander said...

Grey:

"Long before cavalry or chariots, ride-able horses would have given greater range and an improved getaway vehicle for **raiding**."

Would have, certainly, but did anyone actually ride horses like that at such an early date? The evidence we have suggests that practical horse riding, such as you describe, did not appear on the Eurasian steppe until the end of the 2nd millennium BCE.

Simon_W said...

A small addendum, my apologies:
It's true that the TRB, right from the start, occasionally made use of single graves (flat, with inhumation, in the final phase with cremation) besides the dominant collective burials. And it's also true that their megalithic graves were usually under large barrows. Yet the principle of the single non-megalithic grave under a tumulus was a new combination, and is found also in the Yamnaya, Maikop and Leilatepe cultures.

Rob said...
This comment has been removed by the author.
Simon_W said...

By the way, these maps are from the R1a project at FTDNA. They show the project members verified to be R1a*, R1a1* and R1a1a*. It's striking, you can almost see how it spread! It's perfectly in line with Underhill's conclusions:

http://i57.photobucket.com/albums/g240/simon_w1/jj_zps1dd64c1e.png
http://i57.photobucket.com/albums/g240/simon_w1/sgdf_zps99214534.png
http://i57.photobucket.com/albums/g240/simon_w1/hjvhj_zps835396c6.png

eurologist said...

But the Gaulish and Germanic cavalries had quite some reputation in the eyes of the Romans, and the Britons still made use of war chariots

Simon,

So had some Native American tribes, more than ten millennia after the horse had become extinct in the Americas. They learned everything and got their horses from the Spaniards.

But since the Corded Ware's typical burial type was a kurgan, it belonged to the „Kurgan cultures“ in the wide sense. Do you want to suggest that this custom reached C Europe via SE Europe?

If by SE Europe you mean the W and NW Pontic, yes, certainly - although partially similar practices had been around before then. I still see Globular Amphora as one of the main vectors into C Europe and what now is C-N to C-E Germany (SE Lower Saxony to the Ukraine).

My understanding is Cucuteni / Vinca had early agriculture, early (earliest?) pottery, early (earliest?) copper working, early (earliest?) script etc and all directly adjacent to PIE before PIE domesticated horses.

Grey, I totally agree with you on this point, except for the horses and PIE thing. Horses were domesticated on the steppes to the east, at a time when the W Pontic region was way too wet for simple herding or horse domestication, and much better suited to traditional agriculture, and much more densely populated, and much richer, and much more "culturally advanced" - whatever that means...

So, any language transfer is more likely to have happened in the opposite direction.


THEY DID NOT RIDE HORSES.
THEY ATE HORSES.


Rob,

As I have joked in the past, on their mystical journey west into the swamps and dense old-growth-forests west, without roads nor grazing areas nor fields of oats, the "Kurgans" ate all the oats they carried with them the first winter, then all their horses the second one, by which time they were ill-adapted dry steppe-herders, without a heard, in a wet and and hostile country well-equipped to run them down.

Rokus said...

'It's not just the barrow, also the principle: A single grave under a tumulus, with the dead in crouched position and with weapons as grave furniture.'

Most Neanderthal skeletons were already in a crouched position. And also the use of red ochre, current in many IE burials, played a prominent role in Neanderthal burials.

Mallory already mentioned the differences in "In Search of the Indo-Europeans":
'Lothar Kilian isolated twenty-three diagnostic features. He argued that the Corded Ware burials possessed a series of traits not found in the Pontic-Caspian - amphorae, cord-decorated beakers, battle-axes - which are the essential markers of the Corded Ware culture. In contrast, the steppe burials utilized egg-shaped pottery, hammer-head pins, ochre and a variety of burial postures unknown in the Corded Ware horizon. While there may be some generic similarities, Kilian concluded that the specific differences do not support an historical connection between the two regions.'

One solution also mentioned by Mallory is a shared Mesolithic origin. I don't even think Yamnaya was fully ancestral to the steppe cultures that without doubt were ancestral to Indo-Iranian and some Balkanic IE groups, and so far paleogenetic evidence has not supported such claims.

Simon_W said...

Dr Rob,

Horse-riding in all probability, as VA Highlander has already mentioned to you, only really began some 3000 years later, on the outskirts of Central Asia, c. 6-500 BC. 

No need to exaggerate in the opposite extreme. Va_Highlander spoke of the end of the 2nd millennium BC. In fact the Assyrians already had a real cavalry in the 9th century BC.

And in fact, the "blitzkrieg" type of nomadic equestrianism which most people picture, really took off another thousand years after that, - with Huns, Avars, etc - when spurs were first 'invented' probably somewhere near Japan and Korea. 

Are you sure you mean spurs and not stirrups? Spurs were already used by the La Tene culture in the 5th century BC. Stirrups came roughly a thousand years later. One of the first European groups to use them were the Goths who presumably had it from more eastern populations. Indeed the stirrup provided the rider with a better footing and made more powerful attacks possible. But nonetheless cavalries throughout antiquity were able to do without it – as long as no one had it, it was no disadvantage.

As for the general question of when horse riding first appeared: There's the problem that it need not produce much physical evidence. As Marsha Levine put it:
„Organic materials such as leather and wood are only very rarely recoverable from the archaeological record. In unfavourable soil conditions even bone is eventually destroyed. Moreover, not only is it possible to ride a horse without the use of a saddle or bridle, but also, during the early stages of horse domestication, it is likely that they were usually ridden that way.”
So absence of evidence is not necessarily evidence of absence of at least occasional horse riding.

But since there is evidence for the expansion of Yamnaya groups into SE Europe, the question if they rode horses is of secondary importance. No matter if they migrated by foot, by ox cart or by horse, there is evidence that they did migrate.

Also there is unambiguous archaeological evidence for chariots in the Sintashta Petrovka culture around 2000 BC. Again Levine: “By the middle of the 2nd millennium BC horses were being used to pull chariots – from as far afield as Greece, Egypt, Mesopotamia, Anatolia, the Eurasian steppe; and in China by the 14th century BC.”

Simon_W said...

@ Eurologist

So had some Native American tribes, more than ten millennia after the horse had become extinct in the Americas. They learned everything and got their horses from the Spaniards.

My point was to counter your assertion that horses were completely unimportant until Roman times – seemingly implying that the Romans propagated the military use of horses. But obviously the Celts didn't learn the use of cavalry and chariots from the Romans. It was the Romans who neglected the use of cavalry, that was always one of their biggest weak points.
And domesticated horses had been present in Europe for a much longer time, possibly longer than there had been riding skills.

As I've said before, the typical burial rite of the Corded Ware culture developped some centuries prior to the existence of the Corded Ware artifacts (2900 BC). The Baalberge group (3950 – 3400 BC) already had Kurgan type burials. And in central Germany there are a few Corded Ware burials without pottery which were contemporaneous to Salzmünde (3400 – 3100 BC). Also in other countries there are Corded Ware burials without pottery that date well before 2900 BC.

So quite obviously that custom wasn't simply imported from Yamnaya, which started around 3100 BC according to recent datings.

However, Leylatepe dates from 4350 BC.
Maikop from 4000 BC.
I'm thinking that the Kurgan custom diffused from there into various directions, influencing hunter-gatherer descended groups (autosomally rich in ANE ancestry and the North European component) to its north. So Corded Ware and Yamnaya would rather go back to a common source than that the former is descended from the latter.

Something important appears to have spread from the Transcaucasus, and now the new genetic evidence seems to suggest that this was strikingly paralleled by a rapid spread of R1a.
Remember also Dienekes' Rolloff dating of the West Asian admixture into the Lithuanians and Ukrainians, it's 3800 BC and 3530 BC, respectively.

Eurologist mentions the Globular Amphora culture as a main vector into central Europe, and indeed, the Globular Amphora folks did occasionally erect Kurgans. And they arrived quite early in central Germany, around 3400 BC. They were not as early as Baalberge, but may still be an important additional vector, or even the main vector. In fact, Marija Gimbutas once wrote that the Globular Amphora culture was the result of Maikop influence upon part of the TRB culture. So this line of reasoning is not at all at odds with a Kurgan theory in the wider sense (leaving a simple fixation on horse riding and Yamnaya aside).

I have yet to see how the Kurgan custom may have reached the Globular Amphora culture from West Asia via SE Europe. Aren't the Kurgans in SE Europe Yamnaya derived?

Simon_W said...

Rokus,

Most Neanderthal skeletons were already in a crouched position. And also the use of red ochre, current in many IE burials, played a prominent role in Neanderthal burials.

I agree that the use of red ochre isn't particularly telling, it was common among too many cultures.
Afaik the typical posture of the dead in the TRB however was stretched, on the back.

Mallory already mentioned the differences in "In Search of the Indo-Europeans": etc

The use of cord decoration was also present in the Yamnaya culture, although not on beakers. But as I wrote above, I don't believe that the Corded Ware can be derived from Yamnaya either.

The genetic evidence however, in particular the recently discovered facts about the Ancient North Eurasian ancestry of modern Europeans (the Lazaridis et al study from last December) are very clear about some substantial eastern admixture in modern Europeans that arrived after the early farmers and that cannot be derived from the western nor the northern hunter-gatherers.

I think many people have a psychological problem with such a notion, because they don't want to be eastern hordes, no Turks or Hunns, nor even Slavs, they want to be pure, old Europeans, western Europeans moreover, and Caucasoid on top of that. The wish to be as autochthonous as possible is an old and wide-spread one. See also German Dziebel's postings for that. However, it seems we have to change our view of the ancient eastern populations quite a little. They may have been the main source of the modern north European component and even of blondism, possibly. They may have been more like ourselves than many appear to imagine. The old neolithic Europeans on the other hand look more and more like Basques and Sardinians. Europeans too, for sure, but not quite conforming to the identity some more northern commenters regard as their own.

terryt said...

"The wish to be as autochthonous as possible is an old and wide-spread one. See also German Dziebel's postings for that".

You might be onto something there. It is certainly a mystery to me what motivates his denial of the obvious.

"Indeed the stirrup provided the rider with a better footing and made more powerful attacks possible".

It certainly allowed for more accuracy with bow and arrow. Just as an aside on the subject. I am sure that horses were used as draught animals before anyone plucked up the courage to leap onto a horse's back. Oxen and donkeys had been used as draught animals long before horses appeared in the Middle East.

Grey said...

@Dr Rob

"you have bought into the "Kurgan" theory without critically evaluating it"

Straw man. I fully accept that hordes of proto-huns is not plausible at this time.

I am saying you might need a hunnic horde to *conquer* a large settled population like Cucuteni but you don't need a hunnic horde to make them **move away** from being raided.

So not horse-riding hunnic hordes rampaging west but displaced Cucuteni/Vinca moving west.

.

You don't refute any of the points I made.

(*replacing the word PIE with "the people north of the Black Sea" to avoid a different argument.)

1. The Cucuteni culture was directly adjacent to the people north of the Black Sea.

2. Cucuteni and the people north of the Black Sea traded and interacted for c. 2000 years.

3. Cucuteni disappeared around the time the people north of the Black Sea domesticated horses.

4. Somewhat later the region of the Vinca culture became depopulated - not replaced - depopulated.

.

It seems to me an entirely plausible explanation is Cucuteni / Vinca were the first people in history to be forced to move away from steppe raiders (not conquerors, raiders.)

(And by a strange coincidence near this time scattered groups of copper smiths start showing up in central Europe and around the Med.)

.

@eurologist

"So, any language transfer is more likely to have happened in the opposite direction."

Maybe so. I don't have a strong view on the language issue although PIE is a lot easier to type than "the people north of the Black Sea."

Rokus said...

'The genetic evidence however, in particular the recently discovered facts about the Ancient North Eurasian ancestry of modern Europeans (the Lazaridis et al study from last December) are very clear about some substantial eastern admixture in modern Europeans that arrived after the early farmers and that cannot be derived from the western nor the northern hunter-gatherers.'

Europeans today are genetically intermediate in a three-way mixture model. The third source is an Ancient North Eurasian (ANE) population, that also contributed ancestry to Native Americans. ANE ancestors admixed with the ancestors of most present-day Europeans, what based on discontinuity in mtDNA haplogroup frequencies in Central Europe it was indeed suggested this may have occurred during the Late Neolithic or early Bronze Age ~5,500-4,000 years ago. But why ANE should have come from the east at that time, if it was already available in northern Europe?

Its not true that Lazaridis refers to "some substantial eastern admixture" other than a MA1-derived component already in place. Neither is it true that this element was not already available in northern hunter-gatherers. MA1 is more similar to Motala12 than to other European hunter-gatherers and least related with Stuttgart, while Motala12 is not more similar to Stuttgart than to MA1. The dwindling contribution of EEF and expansion of northern ANE areas already suffice in explaining the expansion of ANE. Indeed, all points at a massive mid-neolithic expansion of the northern element in Europe. This element clearly has Mesolithic roots.

'many people have a psychological problem'

The most obvious problem I can see is in accepting parsimony and some desire to rely on answers from terra incognito. Without a shred of evidence some tend to embrace unverifiable nonsense, and then lecture about psychology or politics.

Rob said...
This comment has been removed by the author.
Simon_W said...


I've given some more thought to the issue of North European + ANE admixture:

As the Brandt-Haak paper last year showed, there is some mt-DNA evidence suggesting that already the Corded people of central Germany had more of this admixture than previous cultures. I surmise that they had this from the Globular Amphora people who in turn had it probably from the steppe cultures. I'm not necessarily thinking of Yamnaya here, but more generally the hunter-gatherer descended cultures immediately to the east of the Tripolye culture. Tripolye and steppe people had interacted for a long time, it wouldn't surprise me if they also had mixed genetically. Then Tripolye collapsed around 3400 BC. Maybe there was more influx of steppe people after that or maybe not. In any case from there the vector goes via Globular Amphorae to central Germany.

However, this needn't be the whole explanation for the North European and ANE admixture in modern Celtic, Germanic, Baltic and Slavic populations, as history didn't stop here. The Corded Ware culture subsequently expanded far eastwards and assimilated and absorbed groups that had been living as hunter-gatherers till then. The proto-Balto-Slavs may presumably be derived from one of these more admixed groups than from the original Corded Ware. At least Stanislav Grigoriev suggests that a lot of population movements also took place in the Bronze Age, in particular he suggests that the ancestors of the above mentioned populations moved in from further east. Personally I suppose that the Unetice culture played an important role in the formation of the proto-Germanics, but the modern north Germanics are rather Gedrosia-admixed like the Celts, they show only very slight Caucasus admixture, and thus there appears to have been an even more eastern influence that links them with the Italo-Celts.

The Albanians, Armenians and Greeks (particularly the old ones) however don't have a lot of North European admixture, yet they are not notably less ANE shifted than the above mentioned populations. IMO the reason is that ANE isn't only linked to eastern European hunter-gatherers, but is equally strong in the Caucasus and nearby West Asian areas, and that's where the Albanians, Armenians and Greeks seem to have part of their ANE admixture from.

Grey said...

@VA-Highlander

"The evidence we have suggests that practical horse riding, such as you describe, did not appear on the Eurasian steppe until the end of the 2nd millennium BCE."

There are youtube videos showing people galloping around on horses with no saddle, stirrups or bridle - just using the horse's hair or with a rope around the horse's chest.

https://www.youtube.com/watch?v=j2sYlGVB1_w

http://www.lessismorehorse.com/Desi%20Bridle-less%201.jpg

Grey said...

"I've given some more thought to the issue of North European + ANE admixture:"

If ANE = mammoth steppe then some ANE were already in Europe (and a lot of other places).

http://img571.imageshack.us/img571/6525/7dt9.png

Anonymous said...

Would anyone explain to me why the chariot and horse is given such importance for the dispersal of indo-europeans?

Is it not possible that the words that were used to refer to chariots and horses, were initially used to refer to wagons and asses/onagers? It doesn't seem in my mind a big stretch to have these words that referred to wagons, wagon parts (axle, wheel etc.), and asses later easily refer to chariots, chariot parts and maybe a harder switch but quite possible horses. Even the root word for horses in PIE is supposedly derived from a word that is thought to implied 'that which has swiftness.' The descendants of which at least in Armenian was used to name asses.

If that is possible, it would push the homeland for Indo-Europeans temporally much further back and who knows where in terms of geography. It would free us to look elsewhere instead of where the chariot and horse were first used together.

Va_Highlander said...

Grey:

"There are youtube videos showing people galloping around on horses with no saddle, stirrups or bridle - just using the horse's hair or with a rope around the horse's chest."

Alas, I am a make-believe scientist, and must rely on the data we have from the period in question. No Youtube videos have survived from ancient times, so far as I am aware.

Today, we already know that horses may be ridden. It is not a tremendous leap, conceptually, from riding with proper equipment to riding bareback, since the desired result is clearly understood. We know that some Amerindians preferred to ride bareback, though they also had saddles. However, there is no compelling evidence that anyone was riding horses in any tactically useful way until the end of the 2nd millennium BCE, at the earliest.

Reconstructing the past must be more than imagining plausible scenarios. If they cannot be anchored empirically in the real world, they remain just so many entertaining fancies.

Simon_W said...

@ Rokus
The Scandinavian hunter-gatherers don't suffice to explain the ANE admixture. They did have ANE admixture, but obviously not enough: The PCA in fig. 1B shows that most extant Europeans are closer to MA1 and AG2 (the Ancient North Eurasians) than a straight line drawn between the Scandinavian Hunter-Gatherers and the Early European Farmers would be. So the additional ANE admixture came from the east, where else should it have come from? But obviously this includes eastern Europe as well. We haven't got yet autosomal DNA from the hunter-gatherers there.

@ Dr Rob
Afaik, the view that there was some Yamnaya migration into SE Europe is pretty much accepted. Even Alexander Häusler, one of the fiercest critics of the Kurgan theory, admitted that there was some migration of Yamnaya groups into SE Europe, albeit limited, late and soon assimilated. This view isn't merely based on the appearance of Kurgan burials, but of the entire cultural inventory. See also the small map in fig. 3 in this article by Volker Heyd: http://www.aegeobalkanprehistory.net/article.php?id_art=10
It shows the Yamnaya culture expanded along the Danube and in the Carpathian basin east of the Tisza. Mallory used a very similar map in one of his recent lectures. See also this blog post, it deals with a Kurgan in Hungary that contained Yamnaya grave goods: http://dienekes.blogspot.ch/2013/01/diverse-occupants-of-hungarian-kurgan.html
Of course there were lots of contacts and influences between the SE European copper cultures and Anatolia/the Aegean, and also influence of Cucuteni-Tripolye onto the steppe, it would be stupid to deny this.

Note that I didn't take the position that the SE European cultures declined because of invasions. The decline of Varna for instance was far too early to have been caused by Yamnaya – which didn't exist at that time.

And I already knew that the Sintashta chariots were technically imperfect and deficient, but they are the earliest ones attested archeologically, so that's only logical. The first cars weren't as good as the latest ones either. At least it's proven that they had invented an early chariot prototype which was improved quickly in the following centuries (particularly in the Near East) and that they used domesticated horses to draw their vehicle.

@ Grey

If ANE = mammoth steppe then some ANE were already in Europe (and a lot of other places).

Yes, the presence of ANE admixture was proven for Scandinavian hunter-gatherers. And it seems very plausible that eastern European hunter-gatherers had even more of it. However in the case of western European hunter-gatherers, both in Luxemburg and Spain, no such admixture was detected, instead they were inferred to have had common descent. Note that since we're speaking of autosomal admixture it's impossible that it differed a lot between individuals of the same region.

Simon_W said...

@ Atlantis

I think it's because the proponents of the classical Kurgan theory thought that horse riding and/or chariotry provided a plausible mechanism for the PIE expansion. Afaik there is no reconstructed word for chariot, just for wagon. And the horse, but a wild horse is also a horse, isn't it.

@ all

Regarding the advanced, early cultures of the Balkan Copper Age, it's beyond doubt that they exerted an important influence upon central and northern European Neolithic cultures. Their influence in central Europe has been called the „Lengyelisation“. And the copper axes of the Balkans became the model for the prestige stone battle axes of the TRB.

However, both Ötzi and Gok4 lived roughly 1000 years after the climax of these Balkan Copper Age cultures, evidenced in Varna. And yet both are genetically still very similar to the earliest Neolithic LBK farmers. That means either the Balkan cultures didn't contribute much genetic influence, or they were similar too. And there is good evidence that after Gok4 and Ötzi there were still major genetic changes to come, in particular the increase of the North European and the West Asian autosomal components and the decrease of the Mediterranean component. So there are doubts indicated about the theory that the early Balkan Copper Cultures spoke Indo-European.

After all, just because they were advanced and flourishing they were not necessarily IE. The Minoans were very advanced too, yet they were not IE. Speaking of Minoans, a symbol that was typical for them was the ritual double axe – which reminds me strangely of the early copper axes of the Balkans and of the TRB axes...

Rokus said...

@Simon_W
'The Scandinavian hunter-gatherers don't suffice to explain the ANE admixture. They did have ANE admixture, but obviously not enough: The PCA in fig. 1B '

We don't have so many samples for a dense grid to have the gene flow processes involved already chewed out, but apparently the Motala individuals cluster with ~5,000 year old hunter-gatherers from the Pitted Ware Culture (PWC) in Sweden. This culture had strong influences from the Comb Ceramic culture/Pit-Comb Ware of Finland and other parts of north-eastern Europe, ~6,000 - 5,000 BC. On fig. 1B this neatly corresponds to modern Finnish, Russians and Mordovians, that are still amazingly close to AG2. Here you have sufficient indication that ANE influences were old and that ANE sources were nearby.

Ever since the Mesolithic gene flow accelerated, and just a little more ANE does not have any comparison for instance with the advent of mtDNA H, that was confirmed as far west as Mesolithic Iberia.

Another point of interest is a new method of Hochreiter/Povysil that locate a maximum of Denisova-matching IBD segments in CEU, rooted in NW Europe. Such Siberian tie has no derivation from nearby populations (including Finnish). I can see too many inconsistencies for running to conclusions.

Rob said...
This comment has been removed by the author.
Simon_W said...

Some further thoughts:
Both Gamkrelidze/Ivanov and Ringe agree that Graeco-Armenian is one branch of IE and Balto-Slavic-Germanic is another one. So it's still possible that they both entered Europe along different paths. Where they disagree is the position of Indo-Iranian. The former authors see it related with Graeco-Armenian, the latter with Balto-Slavic. Depending on the theory one prefers, the Indo-Iranians must have migrated eastwards either through the steppe or through Iran. In favour of the steppe route speak the early contacts with Uralic and the high incidence of haplogroup R1a which links the Indo-Iranians rather with Balto-Slavs than with Greeks and Armenians who are not known to have lots of R1a.

But how and when did the Graeco-Armenian branch enter Europe, if it didn't come from the steppe? I think the late Chalcolithic/early Bronze Age Baden-Ezero complex of cultures is a good candidate, since it has strong relations with eastern Anatolia, according to Alexander Häusler. So it seems plausible that from the eastern Anatolian/Transcaucasian area one branch crossed the Caucasus with R1a and Kurgans, while another branch waited a little longer and then jumped directly to SE Europe, with J2 and some (although not the first) R1b. The rather continuous chain of languages one observes today from Slavic via Albanian to Armenian may be due to later contacts then.

Simon_W said...

@ Rokus

I would rather say: in spite of their cultural ties with the Comb Ceramic, the Skoglund HGs were not sufficiently ANE to explain that admixture in modern Europeans. This is what the evidence tells us, and I prefer to go where the evidence leads me.
But as you allude to Russians and Mordovians, these are in eastern Europe, and as I said, IMO eastern European HGs are partly (!) the source for the ANE admixture in modern Europeans.

As for mt-DNA H, the Basques have lots of this haplogroup, yet they are autosomally very Mediterranean. IMO the Mediterranean component at higher Ks, like in Dodecad K12b or in Globe13, is a mixed component of earliest farmer and hunter-gatherer parentage. So, yes there was a reappearance of hunter-gatherer mt-DNA in upper Neolithic farmers, which indicates that they had absorbed the last remainders of hunter-gatherers by then. But this didn't prevent chronologically later Neolithic people like Ötzi or Gok4 from being strongly Mediterranean and Sardinian-like. Yet modern non-South European populations are much more North European, autosomally. This means that they must have admixed later with populations whith a strong North European component. But there were no western European hunter-gatherers left by then. There were however pastoralists in the Pontic-Caspian steppe who were largely descended from hunter-gatherers, and there were hunter-gatherers in the forests north of them. These provide a plausible source for later North European admixture. So while the overall increase of ANE admixture was small, as you correctly point out, the increase in the North European component was dramatic – because the eastern European HGs were no Ancient North Siberians either, they just had somewhat more of this admixture.

But you may still regard the Megalithic folks as your ancestors. After all, the Mediterranean component is still strong in the entire west of Europe, it didn't vanish into thin air.

Another point of interest is a new method of Hochreiter/Povysil that locate a maximum of Denisova-matching IBD segments in CEU, rooted in NW Europe. Such Siberian tie has no derivation from nearby populations (including Finnish). 

Interesting indeed. This is another piece of evidence in favour of my latest theory outlined in the last comment here:
http://dienekes.blogspot.com/2014/02/recent-radiation-of-r-m269-males-in.html

Rokus said...

@Simon_W
'But as you allude to Russians and Mordovians, these are in eastern Europe, and as I said, IMO eastern European HGs are partly (!) the source for the ANE admixture in modern Europeans.'

No, I referred to the areal influence of the Pit–Comb Ware culture. This culture came to be well within the boundaries of the Corded Ware expansion. Genetic cohesion and gene flow within cultural boundaries are a common feature, IMO hence the origin of the small ANE contribution that happened especially in northern Europe.
In a subsequent stage, northern expansions rooted in a northern Beaker horizon affected the rest of Europe, contributing to substantial levels of the northern genetic component, and a small ANE component lifing on top of it.

Your generalization "Eastern Europe" does not hold either, since Pit–Comb Ware cultural influences were never attested in kurgan territory.

BTW. Of all megalithic cultures I only regard Funnelbeaker ancestral to Beaker cultures.

Simon_W said...

Rokus,

I referred to the areal influence of the Pit–Comb Ware culture. This culture came to be well within the boundaries of the Corded Ware expansion. Genetic cohesion and gene flow within cultural boundaries are a common feature, IMO hence the origin of the small ANE contribution that happened especially in northern Europe.

Certainly a possible interpretation, though I don't favour it personally, since the eastern Corded Ware was derived from the western one, so the main direction of gene flow in that case was west-east. In the case of the preceding Globular Amphora culture, the movement was east-west.

Your generalization "Eastern Europe" does not hold either, since Pit–Comb Ware cultural influences were never attested in kurgan territory.

Granted, they were not the same. But I guess the North European component and the ANE admixture was strong in both. Definitely though, I don't want to ascribe everything to Yamnaya admixture.

Grey said...

@Simon_W
"But there were no western European hunter-gatherers left by then."

There is a major bottleneck (aka dramatic population explosion) in y dna I1 in northern Europe around the time of Funnelbeaker.

So you're half correct in that the Western European HGs turned into slash and burn farmers in the northern forest zone.



"But you may still regard the Megalithic folks as your ancestors."

Because they are.

Grey said...

@Simon_W

"But how and when did the Graeco-Armenian branch enter Europe, if it didn't come from the steppe?"

Displaced from the Western shores of the Black Sea by the horse dudes.

Grey said...

@Va_Highlander

"However, there is no compelling evidence that anyone was riding horses in any tactically useful way until the end of the 2nd millennium BCE, at the earliest.

Reconstructing the past must be more than imagining plausible scenarios. If they cannot be anchored empirically in the real world, they remain just so many entertaining fancies."

Entirely fair enough however there is a difference between saying x is impossible and saying there is no evidence for x.

There are videos of five year old children quite comfortably riding horses with just a rope tied around the horse's neck and others of people galloping quite comfortably with just the same.

So it is evidently *possible* that the early horse domesticating peoples could have developed a major raiding advantage over their neighbors.

I still think anything you could call "cavalry" was a long way off but I think those videos show that mounted infantry was *possible* almost as soon as horses were domesticated. Obviously, this is only proof of the possibility, not proof itself.

Rokus said...

@Simon_W
'Certainly a possible interpretation, though I don't favour it personally, since the eastern Corded Ware was derived from the western one, so the main direction of gene flow in that case was west-east.'

Very likely so, though cultural ties within the cultural boundaries must have been strong enough to explain much of the current homogeneity in northern Europe. Especially mitochondrial DNA, where mtDNA H came to dominate, seems to reflect the movement of marriage partners. This means cultural integration in both directions.

'In the case of the preceding Globular Amphora culture, the movement was east-west.'

I'm not so sure. There was a link between the Lower Mikhalayovka group, that preceded Yamnaya, and Globular Amphora, that preceded Corded Ware in some places. However, the direction has never been convincingly established. Some (eg. Sveshnikov, read Mallory 1989 p251) think the movement was into Ukraine and don't even consider a steppe origin. What seems logical, since this culture had a clear agricultural tradition. I think this Globular Amphora culture has its roots in the Mid-Neolithic revival of pre-Neolithic European elements, not unlike Corded Ware that finally incorporated this culture. It's ultimate domographic influences probably remained quite limited.

Va_Highlander said...

Grey:

"I still think anything you could call "cavalry" was a long way off but I think those videos show that mounted infantry was *possible* almost as soon as horses were domesticated. Obviously, this is only proof of the possibility, not proof itself."

The problem with such speculation is that it ignores what we know about the exploitation of horses based on the evidence we have. First, they were hunted, then later domesticated as a food source. Then horses were used as draft animals and lastly ridden.

How do you explain the fact that there is no evidence whatsoever supporting the existence of "mounted infantry", or even effective equestrian skills, at such an early date? Do you really think it probable that people were riding horses for thousands of years and no one thought to depict a horse and rider, anywhere, in any artistic medium?

Simon_W said...

@ Grey

I agree that the North European component in modern Scandinavians is too strong to be fully explained with the modest presence of R1 haplogroups there. So presumably that component is linked with I1, too.

"But you may still regard the Megalithic folks as your ancestors."

Because they are.


Of course, I didn't intend to say otherwise. The Dodecad K12b Atlantic_Med component and the Globe13 Mediterranean component are still strong in northwestern Europe.

@ Rokus

According to Ilse Schwidetzky there is a gradient regarding the facial breadth in the Globular Amphora people. While they are most similar to TRB people in general, they have gradually broader faces towards the east, thus approaching the measurements of the Yamnaya people quite closely. I don't know much about the cultural entanglements of the early Globular Amphora people there in the east, and how they relate with the late Tripolye people. But some (direct or indirect) steppe admixture doesn't seem impossible, which could account for the occasional kurgans, the introduction of R1a and of some West Asian admixture, both autosomally and in the mt-DNA. And the Globular Amphora culture is so closely linked with the earliest Corded Ware in Poland, that I tend to think they were the same people. So rather than having been incorporated by the latter, the latter grew out of the former.

@ all

I had the impression that Rokus' strategy of deriving the North European autosomal component in modern Europeans from „Beaker cultures“, rooted in an upper Neolithic revival of hunter-gatherer DNA, doesn't work chronologically. But let's recapitulate what we know:

The Neolithic in central Europe started around 5500 BC.
The Neolithic of northern Europe started with the TRB around 4400 BC.
The Neolithic in Britain started soon after 4000 BC.
Small amounts of hunter-gatherer haplogroups appear in Schöningen and subsequent Baalberge 4100 – 3400 BC.
Meanwhile there are still fisher-gatherers in Blätterhöhle 3922 – 3449 BC.
These appear to have been finally absorbed in the Bernburg culture 3100 – 2650 BC, where hunter-gatherer haplogroups are had by a stately fraction of the population.
The Sardinian-like Ötzi lived around 3300 BC, i.e. slightly prior to this and moreover in a secluded area in the Alps.
The equally Sardinian-like Gok4 lived around 3000 BC, so contemporaneous with the Bernburg culture. It's unclear however, if he/she can be regarded as autosomally similar to the Bernburg people. Possibly at that time groups still varied a lot in their hunter-gatherer ancestry. After all, the Salzmünde culture 3400 – 3025 BC had no hunter-gatherer haplogroups at all (if we neglect H for the moment).

Conclusion: The North European autosomal component doesn't have to be derived from eastern hunter-gatherers and pastoralists.

Simon_W said...

(continued)

But still, there is the slight, but appreciable increase of the ANE admixture for which only two sources come into consideration: The West Asian highlands on the one hand, and the eastern European and/or Siberian hunter-gatherers and pastoralists on the other. That West Asian admixture isn't the whole explanation can be best seen in the Scottish: They are among the most ANE admixed populations of Europe, yet their West Asian admixture is only modest, and in the PCA they are not drawn towards West Asian populations.

In the Brandt, Haak et al. study, there was an mt-DNA component called „late Neolithic / early bronze age“ by the authors. It was composed of haplogroups I, U2, T1 and R. This component is strongest in the Corded Ware and Unetice cultures, but strikingly it was also present in small quantities in the Baalberge group. So clearly, there is a connection with „Kurganism“ here (I'm not saying Yamnaya). Interestingly though, Baalberge had T1, but no U2, i.e. there is a connection to the Caucasus and Trans-Caucasus, but none to eastern hunter-gatherers. The Corded People tested so far had only one U2, but two T1 and one I. The main thing about this late Neolithic/EBA – Kurgan admixture thus seems to be the West Asian connection, and this may be paralleled by R1a on the y-DNA side. I mean, even the recent study by S. Wilde et al. on pigmentation in Yamnaya-Kurgan groups found only one U2 in them, but three T1 and one I, plus seven individuals with early farmer-haplogroups, so they were not simply hunter-gatherers that had turned into pastoralists.

Grey said...

@Simon_W

"so they were not simply hunter-gatherers that had turned into pastoralists."

What follows is taking later evidence and reaching back in time but

1) Forest Finns were still engaged in slash and burn farming up to a century ago

and

2) One of the classical writers - I forget which - talked about the Germans burning the forest around their villages as a defensive measure. I think he was seeing slash & burn farming but because he didn't know what that was he assumed it was for defensive reasons.

3) Lack of settlement evidence for Globular Amphora and Corded Ware hint at nomadic or semi-nomadic living.

So I wonder if the area covered by TRB, Globular Amphora, Corded Ware represents an ecozone that only supported slash & burn farming and it's not obvious to us now because that ecozone has been shrinking northwards since the LGM as it got warmer.

If slash & burn farming was the only viable option in that particular ecozone then that might explain the rare transition of HG to farmer i.e. they were still mostly HG but with some burned patches for feeding animals on the new growth.

And hence the surprising survival of Y DNA I in the far north.

#

If so the farming influences on that ecozone could possibly have come from all three directions at once, west (megalith), south (LBK) and east (Cucuteni) but those practical influences could only be marginal culturally because it was a totally different way of life: semi-nomadic slash & burn with no long-term settlements versus settled farming.

#

So I think those maps of TRB / Globular Amphora / Corded Ware are maps of a specific ecozone (which gradually moved north over time but which allowed the northern HGs to survive the advent of farming.

Slumbery said...

Grey:

"So I wonder if the area covered by TRB, Globular Amphora, Corded Ware represents an ecozone that only supported slash & burn farming and it's not obvious to us now because that ecozone has been shrinking northwards since the LGM as it got warmer."

This is a rather confusing claim. The start of any kind of agriculture in the region is about halfway between the present and the end of the LGM, so I really do not understand why you even mention LGM. (BTW, most of the mentioned region was actually iced-in during the LGM...)

The shrinking of the ecozone due to warming does not really make sense, because in the mentioned region agriculture started during the Holocene Climatic Optimum and the climate actually got gradually colder since then.

Simon_W said...

Indeed, and the TRB farmers of Sweden ( = Funnelbeaker culture) were still strongly Mediterranean, in spite of some hunter-gatherer admixture. In the PCA of the new Skoglund paper they are closer to Basques and Spaniards than to modern Swedes.

Nathan said...

Quote from Andrew
" And, while an "Out of India" theory seems disfavored by the evidence, I don't think that we should prematurely rule out the possibility that PIE was part of a linguistic family that included the more civilized copper age centers of Persian, Harappa, and Bactria."

That is dangerously close to Hindutva claims. PIE could not have emerged in and around Harappa because the other IE languages do not share the Dravidian and Munda words in Indo-Aryan and retroflexion.

There is also no mention in the Rig-Veda about building cities , ports,mercantilism and seafaring; all of which are aspects of IVC cities.

The RV speaks of a rural culture, not an urban culture like IVC. There is also no evidence for horse domestication prior to the 2nd millenium BC, the period that Aryans made it into South Asia.

Anonymous said...

No, I known as the areal impact of the Pit–Comb Ware lifestyle. This lifestyle came to be well within the limitations of the Attached Ware development. Inherited communication and gene circulation within social limitations are a typical function, IMO hence the source of the little ANE participation that occurred especially in north European countries.

Spybubble Gratis


blogmaster said...

It's hard to believe that even with such a detailed study as this, there is a persistent fascination with the Steepe theory. Underhill 2014, clearly supports R1a origins on the Iranian plateau (and quite likely IE out of NW IRan, in line with Ivanov), and movement outward, into India, the Steepes. There is really no viable alternative.... only the irrational clinging to one's self-deluded version of history.

Unknown said...

Davidski said "So if that's not a signal of the Kurgan Proto-Indo-European expansions, then I don't know what is".... AND.... "So no, R1a doesn't come from Iran. It got there with ANE and very likely the Proto-Indo-Iranians from the steppe to the north."


You are quite wrong, and in denial. The immediate implication of this study is, in fact, just the opposite. That R1a clade bearing Iranians made movements both North East into Central Asia and South West into India. What you consider to be "Kurgan expansion", based on the present day distribution of M17, is simply late Scythian (Iranian) migrations.


This study is a huge blow to the steepe theory, revealing haplotypes which have conveniently been labeled as 'central asian', actually have there immediate origins in Iran. For over a decade, I have argued that R1a (as well as early R1b, which seems to have originated in a nexus between NW Iran, E. Turkey, and the Lower Caucauses) reflects outward migrations from the Iranian Plateau, between the Neolithic and through the historical period. This study lends a lot of weight to that hypothesis.

amrik singh said...

I think it is important to make a distinction between genetic markers and location : Tribes can move en masse as a group and sometimes settle far away then they are diluted at variable proportion by the previous occupants . Often the periphery is more conservative than the center of expansion. Tribes in the center can be displaced by newer groups . The separation in the two main branches of R1a1a1b1 in Europe, and R1a1a1b2 in Asia about 6000 years ago seems to me to mark a separation of Khvalynsk and Repin cultures into Yamnaya in Europe and Afanasievo to the east . The remnants of oldest branches are found at the periphery of R1a but the center have been displaced and largely absorbed by the Turco-Mongol invasions.

Arun said...

Per the Vedic tradition, the Veda was composed by Rishis who lived in forest ashramas. Nevertheless, the Rg Veda mentions thousand-pillared halls, and a hundred-oared ships, which suggest urban and coastal. The Rg Veda uses the word that during historical times has always meant ocean. The Harvard Sanskritist Michael Witzel dismisses all such as poetic exaggeration and boasting. In order to keep the standard story of Indo-Aryan invasions or migrations going, this is the device that is used.

One must contra to Nathan, go where the evidence leads and not worry that "it is dangerously close to the Hindutva position".