Up for public comment at bioRxiv this week is this paper on the population history of the Near East, with a special focus on Armenians. Here's the abstract:
The Armenians are a culturally isolated population who historically inhabited a region in the Near East bounded by the Mediterranean and Black seas and the Caucasus, but remain underrepresented in genetic studies and have a complex history including a major geographic displacement during World War One. Here, we analyse genome-wide variation in 173 Armenians and compare them to 78 other worldwide populations. We find that Armenians form a distinctive cluster linking the Near East, Europe, and the Caucasus. We show that Armenian diversity can be explained by several mixtures of Eurasian populations that occurred between ~3,000 and ~2,000 BCE, a period characterized by major population migrations after the domestication of the horse, appearance of chariots, and the rise of advanced civilizations in the Near East. However, genetic signals of population mixture cease after ~1,200 BCE when Bronze Age civilizations in the Eastern Mediterranean world suddenly and violently collapsed. Armenians have since remained isolated and genetic structure within the population developed ~500 years ago when Armenia was divided between the Ottomans and the Safavid Empire in Iran. Finally, we show that Armenians have higher genetic affinity to Neolithic Europeans than other present-day Near Easterners, and that 29% of the Armenian ancestry may originate from an ancestral population best represented by Neolithic Europeans.
Unfortunately, the authors failed to even mention the main cause of what they're seeing; the massive influx of Ancient North Eurasian (ANE) admixture into the Near East. They included ancient genomes Oetzi the Iceman and La Brana-1 in their analysis, but not MA-1 or Mal'ta boy, the main ANE proxy.
MA-1 is a low coverage genome, and not easy to work with, but until better ANE reference genomes are sequenced, it simply can't be ignored in studies on the population history of West Eurasia. Here's why:
Above is my Fateful Triangle PCA. Note the eastern shift of the Islamic Near Eastern groups relative to their non-Islamic neighbors. Here are the relevant ANE ancestry proportions:
Anatolian Turks ~16.54%The differences aren't very dramatic, but they're consistent and, as per the PCA, hard to overlook. Indeed, the contrast would be even more obvious if we were to add to the list other exotic admixtures, such as East Asian, South Asian and/or Sub-Saharan.
Armenians ~15.48%
Iranians ~19.61%
Iranian Jews ~14.01%
Lebanese Muslims ~9.82%
Lebanese Christians ~7.14%
If you're wondering why it is that Muslims generally carry more ANE than their non-Muslim neighbors, it's probably because the Islamic expansion had a homogenizing effect on the Near East, and it didn't have as much of an impact on the religious minorities in the region.
How and when ANE arrived in the Near East is still a mystery which can only be solved with ancient DNA. However, my bet is that most of it came after the Neolithic from the Eurasian steppe, the northeast Caucasus and the Altai, with the Indo-Europeans, Kura-Araxes people and Turks, respectively.
Citation...
Marc Haber et al., Genetic evidence for an origin of the Armenians from Bronze Age mixing of multiple populations, bioRxiv, Posted February 18, 2015. doi: https://dx.doi.org/10.1101/015396
See also...
First look at an ancient genome from Neolithic Anatolia
411 comments:
«Oldest ‹Older 201 – 400 of 411 Newer› Newest»rk,
Consider the implication of the higher EHG among the ancestors of the German CWC than among the Yamnaya in the Samara Valley.
Also consider that Yamnaya was a vast archeological horizon split into six different zones based on archeology.
Can you do that?
Krefter and Chad
You need to learn to disconnect historical authors' references to "Celts" (originally the people's North of the Greek colony of Marseille; then applied more generically to divergent groups) from proto-celtic (the reconstructed *proto-language spoken by vast and otherwise diverse communities during the pre-roman Iron Age temperate Europe who's initial spread was not intimately linked to any single group of people, "tribe" or volk; as well as the genetics of R1b, etc
So whilst your obviously both bright- esp when it comes to number crunching and pattern recognition; your overall syntheses still sound like baby talk.
So You need to grasp both aspects fully - social and sciences to not only cultivate your own understandings but to also formulate better models
I'm not sure where that came from. The great social changes we see are accompanied by genetic change, pre-globalism. That is obvious. Genes do match language. There was a paper out this month, that showed that much. Anyway, I'm not sure how that applies Racton man, the one that I was referring to. As far as bronze goes.
Chad
It comes out of good faith
Either accept it and improve (it'll take you a few years) or remain static.
And I'm not saying that language and culture simply waft through the air lol.
Not was I suggesting your understanding of R1b itself was wrong
@ryukendo kendow
regarding your brilliant post:
"We already know that both CW and East Euros have higher EHG than others, incl. possibly Yamnaya. The key question here, is why does Corded Ware--not yamnaya, Corded Ware--seem to share highest drift with NW Euros, and not E Euros, if this is the only thing that mattered?"
"WHY DOES CORDED WARE--NOT YAMNAYA, CORDED WARE--SEEM TO SHARE HIGHEST DRIFT with NW EUROS, AND NOT E EUROS??"
see HAMBURGIAN:
http://linearpopulationmodel.blogspot.com/2014/12/the-archaeological-record-in-northern.html
http://archaeologynewsnetwork.blogspot.com/2014/04/earliest-evidence-of-human-presence-in.html#.VOlunmc5CAN
NW Euros got Karelian thing during or before Mesolithic. (I think.)
"and not E Euros, if this is the only thing that mattered?""
ryu, I don't know, but there is something going on in Oetzi and Stuttgart that happened before the Bronze Age that probably did not come from the "Far East" or "Middle East" . . . and not from Northern Europe either.
What is your argument that Celtic can't date back to 1800-2000BCE, after the full-Beaker package reached its max extent? Why can't the formation of a Germanic and Celtic branch be at 2200BCE, when Bell Beakers had been split to different directions? Would these not be a social reasons to maintain communication over great distances, that eventually diffused to the locals between with genetic and cultural mixing.
I'm just looking to see how our views differ here.
Does a shared language in Western Europe during the Bronze Age not make sense?
Chad
Remember what I have said earlier
I admit the likelihood that IE entered the upper danube and central alpine region from the Carpathian during the bronze age c. 1800 BC).
These might have been two or three (ie not just one) early proto-western IE dialects, which gradually diverged and expanded locally.
But these were not (yet) celtic .
Celtic developed later through a process of secondary convergence of these related predecessor dialects . Likely the central innovating regions were Gaul and eastern Britain ( ie Brittanic and Gallic) cf more "peripheral" Goidelic, celt-Iberian, Pictish , etc
This convergence process likely occurred c. 1000BC; likely when the expansion of modern R1b groups was almost finished .
(Remember the Garret and Ramat articles I refered you to?)
So Bronze Age western Europe was still far more uniformly and completely IE.
Ie Celticization reached, both, its zenith and precipitous decline in the late pre-Roman Iron Age
February 21, 2015 at 10:54 PM Delete
Mike, do you say that because linguistics suggests proto-Celtic is not old enough to have spread in Beaker times?
Ryu
Your observations are interesting
If you wouldn't mind explicitly stating what you think this means on a more 'global' level
Yes - if you accept Garrets model of celtic "emergence".
If u accept that celtic was a phenomenon of secondary convergence ; then it's age gets pushed later. When exactly is hard to tell. But given the similarity of Brittanic and Gallic ; it can't have been much early than the Iron Age. henig Anderson suggested the very same thing for Balto- slavic. Doing away with ideas that baltic and slavic separated c. 1500 BC; he suggested that they still formed a dialect continuum at the turn of the common era (!)
Ramat argued the same for italic ; italic formed through local convergence processes within Italy c, 1000BC
Naturally one must nevertheless admit that predecessor IE lects had entered somewhat earlier. Ie Bronze Age . Which broadly fits with what we all envisage ; but we have refined our understanding of the process and the details .
See ?
Why do you push it back to 1800BCE? That's a several hundred years after the Bronze Age starts in the Carpathian Basin and even Germany. I'm not sure what's special about that timeframe. People were making Bronze daggers in Britain before 2200BCE.
Chad
But what have daggers to do with it ?
You can't treat daggers, or pots, as "index fossils" which allegedly track linguistic change and intrusion
While migration most certainly plays a role ; there are multiple other dimensions to language change .( Eg note the change from Pictish to Scots Gaelic involved *no* migration). The introduction of an item of production doesn't necessarily mean there was a mass migration of peoples and associated language intrusion.
My dating is relative only; don't take it as set .
But as for BB- given that it started in the Atlantic copper age; I think it is too early to be even PIE, let alone celtic. Granted ; the BB peoples would have needed a 'common tongue' to communicate the ideas behind the ritualism and objects they traded, but there's no reason to assume it was one single language - but possibly a few. At best; an early western offshoot of IE might have taken part . But given that Mycenaean (c 1700) was still a "new" language; celtic can have only been considerably later.
@ Davidski
This has nothing to do with the diversity of Yamnaya, and everything to do with Corded Ware, which, as you said, is the pre-eminent IE culture in Europe. Corded Ware is where NW Euro West Asian ancestry thing is most apparent in comparison to Karelian.
And CW seems to be the projenitor of all the later IE cultures in Europe, if the aDNA is to be believed.
Pops with high West-asian-like ancestry, which show high affinity with NW Euros among all Euros, appear in both the easternmost Yamnaya and the western end of Corded Ware. So I see no reason why we need to postulate IE pops with little or none of this ancestry somewhere cotemporaneously with Yamnaya and CW. Any IE pop that did not carry this West Asian Yamnaya signature prob emerged as a product of local dilution, which in fact seems to be the case for e.g. some Unetice indivviduals in ADMIXTURE in the paper.
@ Chad
I think the YDNA data is in fitting with EHG and Mal'ta forming a clade to the exclusion of the ANE in NAms, since that's Q while both Mal'ta and EHG are R.
@ Marnie
Where did you get that Stuttgart also has this eastern shift?
Q was in Motala. Besides, Scandinavian hunters are close to Native Americans than MA-1. So, I think that MA-1 is out, but as sister clades, they can share yhaps. It just so happens that Q won out on that one.
rk,
The higher EHG in Eastern Europe is best explained by higher EHG within the Indo-European core prior to expansions in all directions from this core, not some unaccounted survival of EHG.
You can believe what you like, but it's clear to me that the early Yamnaya or pre-Yamnaya in Eastern Europe - and not on the edge of Eastern Europe nor in Germany - will show a very high level of EHG.
RK
I think what you're saying is what some others have been pointing out in Dienekes blog ...
Or MA-1 is an early branch off that didn't get North in time, oooor, it has something in it that doesn't fit Native Americans.
Sorry marnie
Maybe I'm missing it
But oetzi sits west of Labschour and Sardinians on haaks plot (?!)
Lol david I'm such a thorn in your side. So we're going older and older, to see who's the more 'original' yamnaya, aren't we?
To more serious issues. Unaccounted survival? What do the authors say about hg survival in e europe? Whg figures in Yamnaya and corded ware vs those in e and ne Europe today tell us something already, and we would be quite blind to suggest that this affected whg figures without affecting ehg ones when ehgs were practically on the doorstep to Finland and the baltics.
Note that the yamnaya are modeled as 0% whg, and this precludes autosomal contrib from much west of karelia.
Unless one were to postulate that the earliest ie pop were 100% ehg and contained high whg as well, what happened with e europe and the baltics when the yamnaya and cw reached is pretty obvious to me.
Anyway, I'm just getting the point out that any idea that e euros, esp those around the baltics, are close to 100% cw is plainly ridiculous, esp as it turns out there are strong indications that they're not even the most cw pop, much less 100% cw.
@Chad
where did you get that karelian is closer to NAms than ma1 is?
They're equally far aren't they?
Let's see how those early, more EHG-like Corded Ware people from further to the east turn out, and also the mtDNA C Yamnaya people from around the Dnieper.
I do expect some butthurt from certain quarters when those results come in.
I'll have to find it in the morning, but the Kostenki paper had motala as a better fit into Siberians and Native Americans, than MA1. Modeling had native Americans as 58% EHG, 42% EEur. The opposite of MA1.
@David
That will bring those eastern cw closer in space and closer in time to the yamnaya samples...? Oops they turn out to have the west asian nw european thing too!
Ehg has nothing to do with it, the cw samples now have more than enough to make them score with ne euros, but they don't.
Let's see how they turn out indeed.
@Chad
where did you get that motala has q? Weren't all of them i?
Chad, all the Motalas had I2a1, except one who had I2c2.
rk,
"That will bring those eastern cw closer in space and closer in time to the yamnaya samples...?"
Only on the PCA.
The ancestral Corded Ware population had higher EHG than the Samara Yamnaya, not lower.
And again, just in case you missed it: higher, not lower.
Davidski: Consider the implication of the higher EHG among the ancestors of the German CWC than among the Yamnaya in the Samara Valley.
Possible implications:
- German CWC would have to be more admixed with Middle Neolithic Europeans than we estimated from the Yamnaya we have. Less demographic impact from the Yamnaya horizon side ancestors of German CWC.
- Or, if more MN European doesn't cut it, some other population that was more "West Asian" than the Samara Yamnaya would have to come in as well to explain German CWC
- Final implication - it seems that Eastern Europe is shifted further from EHG than Corded Ware and Samara Yamnaya on PCA, so whatever influence goes to Eastern Europe (probably from the Central Europe, Poland-Germany) to shift it further from EHG than Corded Ware and Samara Yamnaya would have to become larger, in a world where Samara Yamnaya and Corded Ware are less representative (less EHG) of the far East European population.
The main implication is that there appears to have been an ancestral Corded Ware population at a location in between the German Corded Ware and Samara Yamnaya with higher levels of EHG than either of them.
The German Corded Ware are obviously mixed, but whether they're part Yamnaya is an open question. They might be of pre-Yamnaya steppe origin.
Btw, Eastern Europe was shifted west by Baltic, Slavic and even to some degree Germanic expansions during the Iron Age.
This is fairly well documented and can be seen in the high levels of IBS/IBD sharing across a very large area of Eastern Europe. This is also what probably screws up the drift stats for most of these groups, except Lithuanians, who are kinda special in their own way.
"The German Corded Ware are obviously mixed, but whether they're part Yamnaya is an open question. They might be of pre-Yamnaya steppe origin."
Ah yes ; those Mesolithic supermen from the Ponto-Caspian steppe. Flew right round the world
@David
Huh?
I already said that high ehg will not change anything, because nw euros and ne euros have rather similar ehg levels, such that even the 100% ehg karelians differentiate them but little.
Cw and yamnaya are approx 40% and 50% ehg. This is more than twice as much ehg inferred for moderns in laz modelling, but they still fail to cluster with e euros.
Its obvious its the other half of the autosome of yamnaya that plays the decisive role for ie comparsons, because high ehg is a dime a dozen in Europe, but high retention of the specific combination of ehg and whatever it is other than that is not as common.
Unless the eastern cw are more than 70% ehg and the ne enf is less than 20% (its at 50 in Yamnaya and at least 30 in this cw) I highly doubt they will resemble the Karelians instead of the cw and yamnaya in terms of their Drift stats.
"The German Corded Ware are obviously mixed, but whether they're part Yamnaya is an open question. They might be of pre-Yamnaya steppe origin.
Btw, Eastern Europe was shifted west by Baltic, Slavic and even to some degree Germanic expansions during the Iron Age.
This is fairly well documented and can be seen in the high levels of IBS/IBD sharing across a very large area of Eastern Europe. This is also what probably screws up the drift stats for most of these groups, except Lithuanians, who are kinda special in their own way."
The German CW did look like Yamnaya + Neolithic Farmer + some extra HG. The difference to Yamnaya was small enough that if that part is pre-Yamnaya, the steppe was like that for a long time. The Behar Lithuanian sample which is used in academic studies shows very high internal IBS sharing too, which may explain the fits for them in Haak et al (high shared drift towards all "directions" means they won't get as big a CW proportion assigned to them as Norwegians).
On another note, on page 69 of Haak 2015, "Early Neolithic Europeans were descended from a common ancestral population", I thought this section might be of interest to some posters here (probably Maju mainly, so I may re-post it on his blog):
We study statistics of the form f4(LBK_EN, Test; Other EN, Chimp), varying Other EN to be
any Early Neolithic population from the set: Starcevo_EN, LBKT_EN, Spain_EN,
HungaryGamba_EN, Stuttgart, and Test to be any modern or ancient (but not Early Neolithic)
population. We use LBK_EN as a baseline for early farmers, as this is the population with the
highest sample size (N=12).
We find only two significantly negative statistics, f4(LBK_EN, Spain_MN; Spain_EN,
Chimp) = -0.00133 (Z=-4.2), and f4(LBK_EN, Esperstedt_MN; Spain_EN, Chimp) = -
0.00173 (Z=-3.6).
The first can be interpreted as Spain_MN having some drift in common
with Spain_EN (but not LBK_EN), indicating continuity of Neolithic populations in Spain.
The second is surprising, as it indicates some gene flow from Spanish related populations into
Germany. With these two exceptions, all other statistics are not significant (|Z|<3) or
significantly positive (Z>3), suggesting that all Early Neolithic Europeans had ancestry in
common with the LBK_EN and are thus descended from a common ancestral population.
Suggests that the Early Neolithic population of Europe was mainly characterized by spreading out from a common ancestor without admixing they all spread out as a clade from the EN ancestor, plus any admixture from WHG they picked up...
However, the exception is Spanish EN / MN populations seem to have expanded to the German Middle Neolithic, not the other way round.
Seems interesting in light of the unusual changes in the Spanish Neolithic mtdna pool documented by Skoglund, and Maju's observations that unique archaeological developments happened in the Southwest Atlantic sphere, and later what appears to be an expansion of Bell Beaker linked to Spain.
rk,
Why are you fixating on the f3 drift stats? For the Corded Ware the Lithuanians top the f3 list out of all the modern samples. So these drift stats are obviously influenced by more than just shared history.
The Yamnaya admixture estimates in this paper look OK to me, but would be better if the Scottish sample had more than four people from one small location IMO.
There's nothing unexpected and unusual about Eastern Europeans showing more Yamnaya ancestry than, say, the English or Orcadians. But if you have a problem with this, then pass on your grave concerns to the authors of the paper and let us know what they say. Good luck with that.
Matt,
So what are they talking about? The Bell Beaker expansion?
Davidski: So what are they talking about? The Bell Beaker expansion?
Esperstedt sample where they find the signal looks pre Beaker (3360-3086 cal BCE).
Looks only like a signal of population movement by Spanish Early Neolithic people or their Middle Neolithic descendants into Germany before the Beakers or Corded Ware. Spanish_EN / Spanish_MN mixing into Germany. Might have something to do with the Atlantic Megalithism phenomenon Maju talks about as coinciding with a pre-Corded Ware population boom in Europe and Megalithic construction.
I guess any conclusion there would be harder for them to quantify and more tentative, as drifts between Spain_EN and LBK_EN should be much lower than other populations they're looking for signals of migration or replacement for, so those drifts would give a weaker signal. Even if Spain's Early Neolithic groups had substantial impact on later Germans, it would seem harder to get a signal distinguishing that from other Early Neolithic (I'm not sure they got a signal from the other Early-Middle Neolithic Germans, for'ex).
Wow wow david, who's the one displaying 'grave concern' about the relationship between e euro and yamnaya over so many years? I'm not the one getting off my chair for this.
I'm as surprised by this as you are. And I'm east asian, I have no vested interest whatsoever.
The plotted statistics cancel out drift, because why would pops drift the the direction of ehg to the exclusion of cw, unless they had extra contrib from ehg in top of what they received from cw?
Do you realize that the plotted f4 stats with modern pops in the position of 'test' are a basic 'reverse' version of what the authors did with laz modelling, and thus everything the authors say regarding robustness, etc applies?
The thing about Lithuanians is whether their drift with cw is proportional to their drift with ehg, and no ne euros have it proportionate, and this is extremely consistent, so Lithuanians are not an exception, in fact they display this most strongly. Compare the e euros, who are displaced off-Cline, to the situation in the paper when early neolithic is used as an outgroup for cw and all euros are off-cline. What does this imply for e euros and how 'excess' ehg affects their eastwards drift with cw in the f4 plot?
And anyway, who constructed an argument out of Lithuanians alone? or for that matter drift Stats alone? good luck wishing that f4 plot away, but that's far from the only thing. And I focus on f Stats, because they behave so predictably (the linear correlation is striking, no?) Which makes things like laz modelling possible in the first place.
Dude, I've always respected you a great deal through the years, you always knew your stuff much better than dienekes did. But I think you've got your blinders on for this.
I'm expecting more stuff once we get the ADMIXTURE results, and the aDNA from the caucasus which in all likelihood will help us get a handle on the other part of the ancestry of the PIEs, and some of that will be unpalatable to you to be sure. Hope you react rationally. Sigh.
And actually it might be a good idea to email the professors about this pattern in their data. I'll ask them what they think.
Back to Yamnaya, but not on continuing the discussion on substructure for Yamnaya.
Really, for Russia and ancient dna, looking at the Corded Ware Culture in Russia is great, what I really think would be useful for these questions is *any* coverage of the period between 5600 BC, where the Samara Pottery Neolithic HG sample is found, up to 3100 BC when the Yamnaya samples are found. That's a pretty substantial chunk of time.
If there was any mixing between the local people and migrants, on the scene in or near Samara, it should show up in the sample range from this period. There are graves through the whole transitional period. Probably we'd want Khvalynsk culture, since from the dna, I'd expect Sredny-Stog to just be a transitional Old-Europe to EHG culture and not really what we're looking for.
To illustrate, check out Extended Data Figure 5 in the Haak 2015 paper - the Neolithic movements in Europe are OK, in Russia you have question mark from the Caucasus, which seems like a population movement with no archaeological correlate or currently any genetic evidence, followed by nothing, then Yamnaya.
They say One important future direction for genetic research into Indo-European origins is to obtain ancient DNA data from India, Iran, northwestern China, and intervening regions to test hypotheses about the spread of Indo-European languages to the east ... A second direction for future genetic research is to study additional ancient European populations ... We highlight Anatolia and the Balkans as particularly promising places to study to make further progress toward understanding these open questions ... By examining ancient samples from these regions it should become possible to determine if there are genetic discontinuities prior to the appearance of these languages that may be correlated with the migration of a new population.
And that's all great (sampling the Balkans means Marnie might get her wish for sampling of this region, whether the results will satisfy we have to wait and see).
But filling in that transition between Yamnaya and the Samara culture sounds like it should be important as well? If there were ever a (Yamnaya_minus_EHG_part) people living on the steppe, they should be able to find them there. Plus this area has known preservation conditions.
Ryu,
That is correct, on Motala. I misremembered. I still think that EHG is closer to Native Americans, and possibly even Motala, from the Kostenki paper.
From this paper..
Fig. S8.6 accounts for this “symmetry” between MA1 and Karelia_HG with respect to Native Americans by proposing a balancing of two processes: first, Karelia_HG share more alleles with Native Americans due to sharing additional common genetic drift (and thus Native Americans should share more alleles with Karelia_HG than with MA1), but, second, Karelia_HG has WHG-related ancestry which dilutes this affinity to Native Americans. However, this requires that the additional common drift shared by Karelia_HG and Native Americans to be nearly perfectly balanced with the dilution due to WHG ancestry, resulting in the observed symmetry, which is not very parsimonious.
Plus, their model of Karelia as a mix of EHG and MA-1 failed. They obviously aren't the same.
Combine this with the fact that Native Americans can be modeled as 58% EHG, versus 42% MA-1, makes a pretty strong case.
If EHG is a better fit with Native Americans, then it should also be with Scandinavian hunters, and the input should be more than the 20%ANE. It would be closer to 30% EHG, if the rate of Native Americans hold. It may not of course. If that is the case, then Scandinavian hunters are close to 78%EHG. That rate of input would put MA-1 at 72% EHG like, and explain how Motala and EHG are both closer to Native Americans, according to this and the Kostenki paper.
MA-1's position relative to Motala may just be because of less shared drift with Native Americans, as Motala clearly has other stuff from Western Europe.
I wonder if EHG could explain why Loschbour is closer to East Eurasians. EHG and Motala is apparently a better fit as a mix into Siberians than MA-1. The Kets may come back as more than 50% EHG, and just over 40% Motala, compared to 36-38% MA-1.
@ryu
"@ Marnie
Where did you get that Stuttgart also has this eastern shift?"
From looking at the above triangle.
Stuttgart is shifted "east" relative to Sardinians and Loschbour.
"Sorry marnie
Maybe I'm missing it
But oetzi sits west of Labschour and Sardinians on haaks plot (?!)"
Rotate the above "triangle" plot counter clockwise by 90 degrees.
Then Loschbour, Western European Hunter Gatherer (WHG), sits clearly in the west.
The Ancient North Eurasians (ANE) will then sit in the north.
The Early Neolithic Farmers (ENF) will sit east.
Other pops such as NW Europeans, S Europeans, etc will set more closely to their actual geographic orientation in space.
Anyway, its relative, depending on how you want to orient the graph.
But in any case, Stuttgart and Oetzi are closer to the "Near East" pops than Loschbour.
Near Eastern pops on the above plot are not well defined, by the way.
But in any case, Oetzi and Stuttgart nevertheless shift toward "Near Eastern" pops. If you look at where early Neolithic actually happened, Greece-Asia Minor-Zagros on the same latitude with Greece, it is clear that Oetzi and Stuttgart shift toward the area of the early Neolithic.
However, it is very unlikely that there is very much ANE in early Neolithic farmers.
Which is what I've been saying all along. There are at least two processes that shift Europeans away from Loschbour. It is the Neolithic process that dominates the shift, not the Bronze Age ANE process.
Furthermore, the ANE process in Northern Europe is probably multilayered, starting in the Mesolithic. (Based on the papers I posted on Hamburgian etc.)
Colin also noted in his recent comments that it is very unlikely that early Neolithic farmers had much "ANE".
@Kefter
That makes sense, Iberians on PCA plots are halfway between Sardinians and Central Europe.
rk,
Like I said, you should pass on your concerns to the authors.
But what they will probably tell you is that f3 drift stats aren't formal admixture proportions, and need not show exactly the same fine scale patterns.
Try them, see what happens.
@Matt
I totally agree - earlier aDNA from the region ought to be relatively easy to get and highly informative.
Migration across the Caucasus into the steppe isn't totally archaeologically unattested. There are North Caucasian sites beginning around 4500 BC or so that connect with earlier South Caucasian agricultural societies, and that clearly traded with Chalcolithic cultures in the steppe region - well before Maykop.
There might be a much earlier agricultural connection across the Caucasus (~700-6500 BC), but that depends on some apparently very controversial identifications of animal remains and pottery types from Lower Don sites.
The source of Neolithic in Ukraine/SW Russia still seems to be up in the air - west, south, east, all of the above? There is also the intriguing possibility that the Black Sea coast was colonized by Cardial Ware people.
@Davidski,
In order to clear up some of the confusion over the awful nomenclature from both this blog and aDNA studies, confusion that Marnie also felt, please clarify the following:
1) Is Basal Eurasian=ENF? If not what is the difference?
2) It seems that EEF was based on Stuttgart girl. In that case is EEF an admixture between WHG and ENF/Basal European in some approximately equal proportions?
3) If the Greek/Crete/Cypriot Mesolithic people were ENF or Basal Eurasian, how could you distinguish the Neolithic farmers from the Near East from southern European Mesolithic people who may have intermixed with the Neolithic farmers on their way to Italy/France/Spain?
Why do lithuanians top modern europeans in the f3 stats for WHG, EHG, CW, BB, and Unitice samples? With those samples, we see each of the 3 main components vary a lot. So how can lithuanians top the list for all of them.
Does it have do with that fact that lithuanians are an isolate populations with a lot of drift?
@ryu
The thing about Lithuanians is whether their drift with cw is proportional to their drift with ehg, and no ne euros have it proportionate, and this is extremely consistent, so Lithuanians are not an exception, in fact they display this most strongly. Compare the e euros, who are displaced off-Cline, to the situation in the paper when early neolithic is used as an outgroup for cw and all euros are off-cline. What does this imply for e euros and how 'excess' ehg affects their eastwards drift with cw in the f4 plot?
Sorry, I haven't been following your posts closely but what is your conclusion? What do you think is missing from picture and how would it effect the relations we see now?
Colin,
rk would like to see the farmer/Near Eastern component in Yamnaya and Corded Ware associated with the Proto-Indo-Europeans. Hence, he's arguing that the higher EHG among Eastern Europeans is the result of a hunter-gatherer survival, and unrelated to the Indo-European expansion. Not long ago, just before the Haak et al. paper came out, he made a comment that the early Indo-Europeans mostly belonged to Y-HGs J and G.
But as I've just pointed out here several times, the unadmixed ancestors of the German Corded Ware carried a higher level of EHG than them and the Samara Yamnaya.
Thus, a reasonable conclusion is that the R1a/R1b EHG were the Indo-Uralic precursors of the Indo-Europeans, and both the German Corded Ware and Samara Yamnaya were their descendants, but heavily admixed with farmer/Near Eastern ancestry.
Also, I think the situation with the f3 drift stats and the Lithuanians is opposite of what you describe. They're actually probably less drifted than the Belorussians and Ukrainians, who are the result of a rapid and rather recent Slavic expansion, which I think is screwing up their drift stats.
See that's why it's important to look at all of the analyses which the authors have kindly provided for us: the formal stats, admixture proportions, PCA, and Fst distances.
But like I say, if you have a problem with the study, then there's only so much we can argue here. I think at this point it's better to e-mail the authors and ask them what the drift stats and admixture proportions mean, and how they relate to each other.
Roy,
1) The K8 ENF component attempts to approximate the main Neolithic Near Eastern component that was probably part Basal Eurasian and part-WHG-like (possibly proto-WHG) from the Near East.
2) EEF is indeed the product of admixture between ancient ENF Near Easterners and WHG-like foragers, probably in the Balkans and/or the Danube region. The K8 ancestry proportions for Stuttgart are 72% ENF (which includes around 44% of Basal Eurasian) and 28% WHG-related ancestry from somewhere in Europe.
3) I expect Balkan hunter-gatherers to be a lot like KO1 from early Neolithic Hungary, with 0% Near Eastern and Basal Eurasian ancestry. It is possible that there were mixed European/Near Eastern forager populations in parts of the Mediterranean. But even if that was the case, it's still difficult to argue that they contributed most of the Near Eastern ancestry to modern Europeans, considering that the Neolithic transition started in the Fertile Crescent.
Anyway, the genetic divide between Mesolithic Europeans and Neolithic Europeans/modern Near Easterners is almost on a continental scale, roughly comparable to that between Europe and India today. So it's hard to imagine that there was a smooth overlap between the two before the Neolithic transition.
David & Ryu
Have you considered that your linear equation of certain Y haplogroups and autosomes components ("west asian", "EHG") with alleged IE expansion to be simplistic ?
I certainly do; But it is apparent most other bloggers here don't
Whatever the case ; the west asian, and possibly Balkan, intrusion catalysed yamnaya . This is what transformed the steppe foragers to pastoral nomads , and what elicited the "yamnaya culture" (an invented pseudo-category), via Majkop & Cucuteni.
You need an agropastoral base to become nomads. It can't develop straight from foraging.
Nomad studies 101
@Davidski
" It is possible that there were mixed European/Near Eastern forager populations in parts of the Mediterranean. But even if that was the case, it's still difficult to argue that they contributed most of the Near Eastern ancestry to modern Europeans, considering that the Neolithic transition started in the Fertile Crescent."
The Neolithic transition did not start in the fertile cresent.
Goat domestication started in the Zagros at the northern limit of the fertile crescent. It was the first animal to be domesticated, according to Melinda A. Zeder of the Smithsonian and many others.
From there, domesticated goats travel both south into the levant and west, to Greece, almost simultaneously.
"Mesolithic Europeans and Neolithic Europeans/modern Near Easterners is almost on a continental scale"
you can't assert this, as you do not have any ancient DNA from the Balkans.
By the way, the component that you are calling "EHG" is a bogus component. It's simply a combination of "WHG" and "ANE".
Neither "ANE" or "EHG" are in any way representative of southern European hunter gatherers.
Anyway, I hardly care at this point. The descendants of some of those southern European hunter gatherers are about to tank the entire economy of Europe.
I can't say that I think that is a good thing, but when Germany spurges on a study like this, completely ignoring the history of the Southern Balkans, Italy, and Asia Minor, it's very representative of the over all bad attitude of Germany toward the Southern Balkans.
@ Capra, interesting. I was trying to find more detail on the early archaeology of the region through The Horse, The Wheel and Language (which is, by the way for all, available to read online at https://archive.org/details/TheHorseTheWheelAndLanguage). Found it quite dense and got the impression that there was an argument for a hard cultural frontier between "Old Europe" and the Russian Neolithic, with arguments for transfer of domesticates from Old Europe, more strongly than the Caucasus, which of course is unusual given that the dna seems to indicate that the Near East related portion of Yamnaya does not seem too similar to Early Neolithic Farmers.
Also, for all not to fixate too much on the f3 stats again, the table S7.1, some more unusual combinations maximizing f3(Dinka; X,Y) - most related population Y to population X via f3 methodology:
- Bell_Beaker_LN with HungaryGamba_HG
- BenzigerodeHeimburg_LN with Motala_HG
- HungaryGamba_IA with SwedenSkoglund_NHG (this is the third lowest stat in the set, after K14 with Loschbour and Ust Ishim with Chane Indians from Bolivia)
- Karsdorf_LN with SwedenSkoglund_MHG
- Unetice_EBA with Motala_HG
- HungaryGamba_BA with HungaryGamba_HG
It seems like the Swedish hunter-gatherers are a popular choice for the most matching populations for a number of Late Neolithic and metal ages populations, which seems odd as I'd have expected other LN / metal age populations (Yamnaya, Beaker, Corded Ware, etc.) to take that slot for them.
The f3 stats, which are the maximums, between early neolithic populations seem quite a bit lower than the mesolithic hunter gatherer populations, e.g. Stuttgart has a lower stat with LBK_EN than MA1 with Samara.
Is that because of genetic diversity? There's less sharing between the individual neolithic farmers because they come from diverse populations, and likewise when it comes to hunter gatherers sharing with descendant farmers their low diversity means they'll match closer their descendant populations.
Re: high drift in the East Slavs in the sample affecting their f3 stats, that's hard to judge, however the Belarusian FST from Yoruba looks pretty typical for Europeans at 0.150, which would maybe contraindicate any particular drift, at least for Belarusians? Lithuanians do look from their FST maybe a little more drifted / bottlenecked compared to other Europeans based on their FST but not very much on a global scale.
More FSTs from the table:
EHG-WHG: 0.078, WHG-SwedenSkoglund_NHG: 0.050, WHG-LBK_EN: 0.091
The hunter gatherers in Europe were really drifted compared to modern populations (differentation from Africans is at higher levels than Han Chinese) and it inflates their differentiation both from one another and the farmers a lot (but especially from one another because the extreme drift is adding from two directions). LBK seem more normal in their level of drift / diversity compared to modern Europeans (e.g. LBK_EN has FST 0.163 from Yoruba, which is higher than all present day Europeans but lower than the HGs).
Matt,
Maybe Dinkas are a poor choice of outgroup? Apparently, they were chosen because they're close to the Eurasian bottleneck. I'll try some f3 stats with these ancient samples using Mbuti Pygmies and Yoruba.
Mike,
No, it's not too simplistic to correlate EHG with R1. It makes really good sense actually.
In fact, I'd say the Indo-Uralic theory is looking pretty damn good based on these results. So it might be a good time to update that pastoralism 101 guide.
The Fst stats look better IMO.
David
"No, it's not too simplistic to correlate EHG with R1. It makes really good sense actually".
Yeah sure, at face value, which is how you obviously like to analyse things. But i wasn't really arguing against that, despite the fact that what makes 'really good sense" to you, given that we still don;t have aDNA from vast parts of the world, and how the modern phylogenic tree of R sits, nevertheless is ambitious / premature at best, or possibly Eurocentric / Nordocentric at worst.
"In fact, I'd say the Indo-Uralic theory is looking pretty damn good based on these results. So it might be a good time to update that pastoralism 101 guide."
David, try to make some sense and thread your comments into a coherent narrative.
a) What does R1 and ANE directly have to do with the Indo-Uralic theory, which has little if any support from actual linguists.
b) Whatever the case, i suppose you think that the earliest 'northern men' which colonized north Eurasia spoke an ancestral form of Indo-Uralic and spread R1. This makes little sense. The Palaeolithic north Eurasians were small bands, living often in isolation, where drift greatly enhanced linguistic differentiation - as seen similarly in the great variety of native American languages (complicated admittedly by later ad-migration). Whatever the Paleo-Siberian languages which existed in western Eurasia, most of these are now extinct. FU and PIE - if indeed PIE first existed north of the Caspian- are but two lucky survivors which expanded at the expense of others, and otherwise do not have any real 'genetic' relationship to each other.
c) this still changes nothing about pastoralism . Pastoralism could not just develop out of foraging. Im not even going to argue the last point with you, because it suggests to me , alarmingly, that you have not read anything on anything, which must betray a severe chauvanism in your thinking. Am I wrong ?
@david
rk would like to see the farmer/Near Eastern component in Yamnaya and Corded Ware associated with the Proto-Indo-Europeans. Hence, he's arguing that the higher EHG among Eastern Europeans is the result of a hunter-gatherer survival, and unrelated to the Indo-European expansion
The Yamnaya ARE the Proto-Indo-Europeans; they are a mix of EHG and some Armenian-like component. So Proto-Indo-European is neither an EHG group nor an Armenian-like group. Sure, there will be variation in the yamnaya horizon but I'm pretty sure most of the horizon will be a sizable mix of the two components. Remember that PIE is primarily a linguistic and archeological question; this paper just supports the dermic expansion necessary.
The fact that CW may have higher EHG than Yamnaya (PIE) shows that there was survival of EHG in Europe apart from the steppe. I actually think part of the reason Norwegians top the list for EHG is because of survival of motala like people in Scandinavia.
But, these are more like small details for the bigger picture. We know that Yamnaya and CW groups spread IE and large amounts of EHG throughout. But of course places like Lithuania or Norway probably already had some EHG prior to IE.
I think the real point your getting at is which "component" is more responsible for PIE. I don't think the question makes a whole lot of sense. But related to that question, I agree that the IE types of r1a and r1b came from EHG. Obviously I have already said that its pretty much a fact that PIE started in the steppe (yamnaya and possibly including very closely related precursors to yamnaya). I don't know what else you want?
As for the Lithuanians and their f3 stats how do they relate so closely to WHG, EHG, and the heavily MN Unetice/BB. The EN in MN should turn Lithuanians away, it just doesn't match up with admixture.
My final point is - whatever its real 'origin' the spread of R1 - at least earlier clades- have little directly to do with PIE and your beloved Yamnaya culture.
Not all basal R1a and R1b peoples in north Eurasia were pre-proto-Indo-Europeans, Yamnaya-going -to-be's. In any case the Yamnaya cultural horizon is nothing but a construct, an heuristic category invented to explain language expansion because the scholars who created it in the 1930s did not have greater methdological arsenal at their disposal.
The Kurgan culture "which Gimbutas (1970:155-97) insists really existed, is only an artificial and speculative
construction which unites under one heading many archaeological cultures
which themselves are very different and are from different periods" (Anatoly Khazanov- the worlds foremost authority on nomadism).
Mike,
You're wrong.
It's OK to be wrong when speculating, but it's not OK to be wrong time and again when interpreting very strong and obvious data, which is what you, Marnie and others have been guilty of here.
Colin,
My bet is that EHG is the Indo-Uralic or pre-proto-Indo-European component.
Everyone is free to disagree with that. But I'd say that linguistics and ancient DNA will be able to prove it eventually.
@ COlin
@ Colin
"I think the real point your getting at is which "component" is more responsible for PIE. I don't think the question makes a whole lot of sense. But related to that question, I agree that the IE types of r1a and r1b came from EHG. "
I agree with that small part of Colin's contribution.
"Obviously I have already said that its pretty much a fact that PIE started in the steppe (yamnaya and possibly including very closely related precursors to yamnaya).'
COlin, why are you hiding your time machine from the rest of the world? :)
David
You CLAIM I'm wrong. But how can your speculations on R1 prove that PIE originated on the steppe? But that's not even my point. I've always stated that the steppe hypothesis is possible.
But i choose not to take your pontifications at face value, becuase you have such **major deficits** in your knowledge, that it is worrisome. You dont even know anything about pastoralism, or forager social networks. Yet you claim to have 'discovered' the prehistory of Eurasia. You impress other bloggers here becuase you're adept at number crunching, but to other readers who have developed all other aspects of anthropoligy, your arguements are comical ! I mean you really think Siberian foragers moved into north africa, and come up with some BS map of the spread of pottery, which has nothing to do with anything. !
So its not denialism, but realism.
Sorry.
Ah, looking forward to the papers from this. The abstracts should be out soonish.
The precursors of Proto-Indo-European: the Indo-Hittite and Indo-Uralic hypotheses
http://www.hum.leiden.edu/lucl/research/conferences/upcoming-conferences/precursors-of-proto-indo-european.html
Sumerian Share more words to PIE than Uralic does and FYI David Indo-Uralic is not that hot if you compare with other language groups....
@ Dave
Thanks for the link , but so what
One non mainstream paper doesn't mean much
And that's not even my main point - which is your arguementation is severely undevelopeded. My critique is intended as an amicable advice - for your sake and the power of your blog
U seem to be taking my critique as personal; How dare I challegne the kurgan hypothesis ??!
But you appear to have a bias which shines through in the way u only shop around for pro-Kurganist studies
If it is right ; then it'll be proven in good time
But as a (? Reputed) blogger u should be neutral and present all aspects of the problem .
Indo-Uralic could help explain why Finno-Urgics are genetically so similar to Balts. But if Uralics originated in Siberia and there's no east Asian in EHG and Yamna, that's a big problem for the Indo-Uralic idea.
There's a Finnish blogger who has a theory that Finno-Urgics used to be Siberian-like genetically and after 1,000s of years of exchanging women with their neighbors became 80-95% east European and Scandinavian genetically.
She says because near by pops had larger population sizes and so 10 new Baltic women reproducing every generation in a small Siberian-like tribe would make a big effect.
@david
Pre-proto-indo-european? Sounds a little pseudo science like. There may be a deep relation between Proto Euralic and PIE and maybe Nostratic exists but they way your imagining pre-PIE just looks silly.
I don't think you can go back that far and have things be meaningful. What is the pre pre pre pre pre... pre proto indo european found in Africa when many of our ancestry left there?
David, I think you a treating language as if it were genetics. I don't think that quite right.
@mike
COlin, why are you hiding your time machine from the rest of the world? :)
Don't you think I have good reason? Maybe, this is the past (for me) and I've seen what time machines do to the world.
Don't be jealous ;)
I agree Colin
Langauge trees aren't like gene trees
They behave in far more complex way
And my original point today was as you highlighted, Colin, is whatever the yamnaya groups spoke ; try are clearly a "novel" population which underwent significant cultural , genetic and possibly linguistic evolution
You can't simply trace it back to the forager component or west Asian component without firm supporting evidence ; none of which really exists - although Dave is free to clutch at straws . But that only undermines his credibility as a reputable blogger
@Matt
"Even if Spain's Early Neolithic groups had substantial impact on later Germans, it would seem harder to get a signal distinguishing that from other Early Neolithic"
This relates to my point about initially two sets of farmers. For various path of least resistance reasons I think any Levantine strand would initially be more likely to be maritime island hoppers who could spread directly west relatively rapidly.
So even if they merged with an overland set of farmers somewhere around the Aegean to become Cardial in the eastern and central med, in the far western med they might have been the people responsible for the Atlantic Megalith culture.
If the out of Iberia signal you mention exists I think it would be there: Atlantic Megalith or island hopper settlements in the med. that don't fit Cardial.
Colin,
I'm not treating languages like genes. Basically, what I'm saying is that linguists have already put forward a number of theories, and we can test these theories with some confidence using aDNA.
For instance, what will you say if in the near future EHG genomes are found to belong to all of the R clades leading to R-M269 and R-M417? That will be very strong evidence of the development of pre-proto-Indo-European in an EHG population, eventually leading to Proto-Indo-European in mixed EHG/Neolithic steppe groups.
I think this is very plausible and like I say, now possible to test.
Krefter,
It's unlikely that the proto-Uralics were fully East Asian. They were probably mixed, and my guess is that they were more than a little EHG.
The Indo-Uralic theory is very attractive now considering the northern provenance of ancient genomes belonging to R. But if it doesn't work out for whatever reason, then we have the alternative of proto-Indo-European and proto-Uralic developing as neighbors and borrowing from each other.
@Mike Thomas
"Pastoralism could not just develop out of foraging."
If mounted hunters rustled herds and took captives to herd them then they'd effectively import pastoralism rather than develop it.
A society that developed like that would almost automatically be hierarchical with a big cheese at the top wanting a nice big burial mound.
Grey,
A huge and rather unlikely cultural shift would be required in order for your hypothetical scenario to pan out.
@ David
1) It would still be debatable: I could make the arguement that PIE still arrived with the detectable central- west Asian influence; seen not only genetically; but culturally, economically; etc
2) what if the Mycenaeans turn out to have nothing directly to do with EHG or yamnaya ?
Then whatever your reconstructions of R and the prehistory of Northern foragers; they'd have nothing directly to do with indo -european language expansion
3) notwithstanding the previous two points; even if PIE is northern; it still doesn't supprt fanciful accounts of indo-Uralic or Nostratic. Those theories are advanced fringist scholars who merely want to stand out for their controversial theories, and find no support from mainstream scholarship . Moreover they wholly miss the point as to how forager communities interact
Grey
Are you a troll ? Or are you just seriously slow ?
We've discussed this already ad nauseum; and with modern technology today there's no excuse to be so uninformed!
Pick up Anatole Khazanov''s "Nomads and the outside world" Ch -"origins" p83; then Michael Frachetti "non-uniform complexity & the multi-regional emergence pastoralism ".
Both avail on Google
February 22, 2015 at 8:56 PM Delete
@ VR
"A huge and rather unlikely cultural shift would be required in order for your hypothetical scenario to pan out."
Subtract guns and rail roads from this and you're almost there.
http://en.wikipedia.org/wiki/Plains_Indians#The_Horse
I agree you'd need just the right conditions: horses and plains, steppe or desert.
Thanks for that enlightening Wikipedia link to a wholly unrelated topic .
you're welcome
I'm serious
I have now discovered the meaning of life
It's because the plains Indians used the horse to hunt buffalo !
Apart from being a lot easier how is rustling sheep and cattle culturally different from hunting?
On an unrelated note, new work on the human mutation rate. Lots of info!
http://biorxiv.org/content/early/2015/02/20/015560
Where is this stuff about foragers developing pastoralism ab nihilo coming from anyway? The societies in question had agriculture, and half their ancestry probably came from Near Eastern farmers. What mystery is there to explain? Foragers adopted farming from farmer migrants with whom they intermarried. The details could vary considerably, but why do we need foragers to steal livestock or whatever? Why can't they just be regular pastoralists?
Capra,
How would you explain the dominance of R-M417 and R-M269, very likely markers of Eastern European hunter-gatherer origin, in Kurgan remains and among present-day Indo-European speakers?
Both of these R clades had to have first expanded from roughly the same place, at about the same time, and via the same social mechanism(s).
@capra
"why do we need foragers to steal livestock or whatever? Why can't they just be regular pastoralists?"
Maybe they are. That's what I thought was most likely before the Haak paper came out.
Even now maybe they were R1b farmers/herders from the south who moved onto the territory of R1b foragers or the haplogroups were different originally and one set disappeared for some reason.
On the other hand it seems to me there is another possibility which is the farmer intrusion created a violent backlash from the steppe foragers and the IE culture was created in the tussle.
@ Capra
"Where is this stuff about foragers developing pastoralism ab nihilo coming from anyway"
Its not. That pastoral nomadism developed from agro-pastoral societies has been known for over a century. That it cannot develop ab nihilo in hunter-gatherers is known to all but ignorami and trolls.
@ David
"How would you explain the dominance of R-M417 and R-M269, very likely markers of Eastern European hunter-gatherer origin, in Kurgan remains and among present-day Indo-European speakers?
Both of these R clades had to have first expanded from roughly the same place, at about the same time, and via the same social mechanism(s)."
Apples and Oranges, David. R1 was more widespread than you like to think, and even if R1 originated in eastern European foragers, there is no denying the fact that the EE foragers received major cultural and demographic impetus from Neolithic world of Central-West Asia and Southeastern Europe.
Moreover, the earliest expansion of R1 in EE was likely related to post-LGM movements which have nothing proximately to do with Indo-Europeans and Kurgans.
@capra
"Where is this stuff about foragers developing pastoralism ab nihilo coming from anyway?"
It's a different argument. The argument here isn't about whether foragers can develop pastoralism it's about whether foragers can - in the right conditions - kidnap pastoralism from pastoralists who wandered onto their turf.
@MT
''Apples and Oranges, David. R1 was more widespread than you like to think, and even if R1 originated in eastern European foragers, there is no denying the fact that the EE foragers received major cultural and demographic impetus from Neolithic world of Central-West Asia and Southeastern Europe.
Moreover, the earliest expansion of R1 in EE was likely related to post-LGM movements which have nothing proximately to do with Indo-Europeans and Kurgans.''
You nailed it Mike!.....
Mike,
It doesn't matter how common and widespread you think R1 was during the late Neolithic. Firstly, I can see from the data that it wasn't all that common outside of Russia, and secondly, I'm referring to very specific young R1 clades that exploded at around the same time. It's not like R-M417 and R-M198 came from India and somehow wound up in Andronovo Kurgans and Corded Ware burials, and R-M269 from Iberia and landed in Samara Kurgans. These markers came from the same place.
What's the point of even denying/ignoring that R1 has already been dug up in double figures from Kurgans at the expense of other haplogroups?
You're never going to sound very convincing if all you're ever going to do is argue that up is down and down is up.
@Davidski
R1a-M417 and R1b-L23 are not very old. Probably they date to very very approximately 3500 and 4500 BC respectively.
These two clades were the ones that hit it big, the patrilineages of the most successful clans. That means they did not enter the mixed gene pool from foragers in general. They came from two patriarchs who both happened to have R1 ancestry (and whose last R1 ancestor probably lived 15-20 000 years earlier).
Assuming they are in fact both forager haplogroups, we do not need any special mechanism to explain their dominance. They got lucky. If forager and farmer haplogroups were originally in equal numbers, the chance of the two lucky ones both being from foragers is one in four.
Or by the usual threshold of statistical significance, we can say that significantly less than 78% of Y haplogroups were of farmer origin.
Even supposing there was a predominance of forager Y haplogroups, there are many ways such a thing could have happened. For instance of period of hiatus in the spread of farming with foragers on one side of a major river and farmers on another, bride exchange between them, the foragers adopt farming, and then the new mixed population expands on the far side of the river, with forager Y-DNA predominating. Or the farmers could have been matrilocal, as appears to have been the case with the Austronesians, those headhunting tooth-sharpening tattooed pirates, who spread predominantly *maternal* ancestry along with their languages.
@David
''It doesn't matter how common and widespread you think R1 was during the late Neolithic.''
??
'' Firstly, I can see from the data that it wasn't all that common outside of Russia''
???
'' I'm referring to very specific young R1 clades that exploded at around the same time. It's not like R-M417 and R-M198 came from India and somehow wound up in Andronovo Kurgans and Corded Ware burials, and R-M269 from Iberia and landed in Samara Kurgans. These markers came from the same place.''
One 2500 BC R1a1a1 from CWC has made you convinced that nothing else is Possible???
''What's the point of even denying/ignoring that R1 has already been dug up in double figures from Kurgans at the expense of other haplogroups?''
No One Is Doing That.
''You're never going to sound very convincing if all you're ever going to do is argue that up is down and down is up.''
Yes quite right.
David
For instance, what will you say if in the near future EHG genomes are found to belong to all of the R clades leading to R-M269 and R-M417? That will be very strong evidence of the development of pre-proto-Indo-European in an EHG population, eventually leading to Proto-Indo-European in mixed EHG/Neolithic steppe groups.
I already agree thats very likely to be the case. The r1a and r1b found in CW and yamnaya almost surely came from the EHG side rather than the EN side, if such components truly represent populations.
But that means squat when it comes to determining language tree. Besides, the PIE Yamnaya already have genetic and linguistic ties to the south. The linguistic ties to Uralic and Caucasian is one big reason the steppes are such an appealing location.
The major debate has always been the question of where PIE was located and to some extent who they were. I guess before this paper, "everything from the near east" supporters could have tried to argue that even if yamnaya were PIE, they were really just recent near eastern migrants. But now we have proof thet yamnaya could play the role of IE progenitors, that they were a mix of European HG and neolithic types (like most modern europeans), and that they had paternal lineages going back to EHG, like many IE people from India to Ireland.
You already won most conceivable moral victories for the origin of PIE. I don't get the brand new question of who the pre proto IE people were.
I honestly think that Nostratic might hold some influence but PIE probably rapidly evolved, meaning that it is not the descendant of a previous single language. The Yamnaya horizon sprang up fast and became huge. That doesn't spell linguistic continuity to me.
capra,
Yes, I know that these markers aren't very old. That's the point.
Their rather recent massive expansions weren't accidental. They didn't just both explode by chance from a vastness of R1a and R1b across Eurasia. They did it from a specific cultural horizon well after hunter-gatherers ceased to exist on the steppe, so there had to be a mechanism by which they survived the transition from foraging to pastoralism in Eastern Europe, and went ahead with their expansions out of the steppe.
Natural selection could be an explanation, because it seems that European eye and hair pigmentation lightened up just as rapidly as R1a and R1b moved across Eurasia. But I'm not aware of any advantages that these markers give.
''now we have proof thet yamnaya could play the role of IE progenitors, that they were a mix of European HG and neolithic types (like most modern europeans), and that they had paternal lineages going back to EHG, like many IE people from India to Ireland. ''
Only to Europe not Asia specially considering the absence of Asian Sides aDNA from decisive archaeological sites, there is no way one will say that R1a-M417 will not be found in Central and SC Asian sites in tons taken that all other studies indicate it will be!
I'm not arguing that R1a1 of Karelia can't be native it can! but too old probably and where is R2a?? West Asia-Central Asia is the hot spot for the origin of R1 and R2! we also know people following the East Of Caspian route to Urals were Entering from Zagros area from Mesolithic times! Mesolithic itself again appeared early in Asia than Europe anyway.
Like Postneo says again and again its the Horse Evidence that is the strongest point for the steppe hypothesis...
Sure, the foragers weren't just replaced by farmers - we already know that they made a more-or-less equal genetic contribution. My point is that we don't need any exotic mechanism to cause a massive bias toward forager Y haplogroups, because we are only really concerned with 2 of them.
@ David
"R1 clades that exploded at around the same time. It's not like R-M417 and R-M198 came from India and somehow wound up in Andronovo Kurgans and Corded Ware burials, and R-M269 from Iberia and landed in Samara Kurgans. These markers came from the same place"
Hhhmm. Im not sure they do , Dave. If 'by the same place' you mean from the Vistula to south-central Asia, then yes, at a cosmological level.
"Natural selection could be an explanation,"
So finally you're considering it, after you repeatedly said "forget it" to me. That's good, 2 years behind, but you;re catching up.
"They didn't just both explode by chance from a vastness of R1a and R1b across Eurasia. They did it from a specific cultural horizon well after hunter-gatherers ceased to exist on the steppe, so there had to be a mechanism by which they survived the transition from foraging to pastoralism in Eastern Europe"
Nobody claims EE foragers were wiped out by the southern arriving farmers- they fused with them and became pastoral nomads. But this 'explosion' of R1a and R1b from the steppe remains to be proven. Ive stated repeatedly:
- central European ANE came from a more proximate source : local central-eastern European groups, and not only Yamnaya steppes.
- the situation of western Europe needs definite proof. We have no Bronze Age samnples outside Germany ! Hello !
- The Samaran Yamnayans seem to have been a (near) dead end. If they were so succesful, why couldn;t they even survive in their own 'homeland' ??
@ Capra
You're an intelligent person. Ive been saying the same thing. There was nothign special about Yamnaya. Right place and right time perhaps, but they rode the pastoral train. Im not sure they were even Indo-European. ! The steppe nomads gropus which emerged in the BA steppe each had their own unique origins, and not from one "pre-proto-IE" Yamnaya precursors liek Repin and S.S. Ths is clear in studies, but David chooses to ignore the evidence. You can't just lump R1 - derived groups and all probably pastoral groups together and say voila ! There are your indo-europeans.
@ Nirj
"Like Postneo says again and again its the Horse Evidence that is the strongest point for the steppe hypothesis"
yes, but they ate them. They weren;t mounted blitzkreig warriors until 800 BC.
Forget it Nirjhar, there wasn't any R1a-M417 in South Asia before the Bronze Age.
All Indian R1a comes from the Bronze Age steppe one way or another.
@David
''Forget it Nirjhar, there wasn't any R1a-M417 in South Asia before the Bronze Age.
All Indian R1a comes from the Bronze Age steppe one way or another.''
Forget it David It will be there in the Central and SC Asian,West Asian Sites dating from ~4000-2500BC.
Hi guys I'm back!
@ Davidski
Stop misconstruing me. Since when did I say that the Armenian-like pops were the PIEs. The only thing I'm saying is that its obvious from CW and Yamnaya that E Euros, esp Lithuanians, are nowhere near the most Yamnaya-like groups, or even the most CW-like ones, alive today.
And Huh? Why in the world would the authors say that?
The entire basis of Laz modelling for nothing less than actual admix proportions is based off of f4 stats, precisely because they behave in such a predictable manner and are so transparent and easy to manipulate, unlike ADMIXTURE or other such methods with such opaque results. There is nothing 'funky' or 'cloudy' about plotted f4 stats, in Lithuanians or otherwise, and the authors' opinion on this is abundantly clear.
The authors did not do a laz modelling with both EHG and Yamnaya alongside EEF + WHG, but you do realise that I'm just taking the baby steps to such an analysis? A complete analysis will replicate my plots with WHG and EEF as well, and use the 15 outgroups instead of comparing directly, but this is a baby step, and anyway the WHG and EEF figures can be gotten off the analyses they already did because neither EHG and Yamnaya pull in the direction of WHG and EEF that strongly. Its EHG and Yamnaya we have to break down, because the 55% Yamnaya 0% EEF figure for the Finnish tells us that an analysis with only EHG or only yamnaya is obviously incomplete and attributes spurious Yamnaya all over the place, which is confirmed in the f4 plot.
Quite frankly, its beginning to be pretty obvious that you're becoming ever more evasive of the points I make because you just don't have a good reply, and you don't like the implications that E Euros and poles are not the 'purest' IE group. Whatever man, different strokes for different folks.
Anyway I've sent the email.
@ Collin Welling
If I were to repost everything that points to this it would take far too long. You can read the rest of the posts in this thread if you like, its worth it :)
But to make things short, I'll only post the strongest evidence. Look at this plot:
http://imgur.com/kpNrdcX,k73e5qN#1
All Euros are black, except for NW Euros in orange and NE Euros in green. Similarity with EHG shifts south, similarity with Corded Ware shifts east. It is immdiately apparent that most Europeans have as much similarity with EHG as their Corded Ware contribution would allow; thats why there's such a linear relationship between similarity with EHG and similarity with Corded Ware. Most Euros only have EHG from Corded Ware. This is only broken for NE Europeans, who have much more similarity with EHG than what their similarity with CW will make us predict.
Because no pop drifts in the direction of another, E Euros obviously received EHG contrib on top of their CW contrib.
More importantly, EHG and CW contrib are not completely uncorrelated. EHG contrib pulls spuriously in the direction of Corded Ware, because CW havve such high EHG; thus the E Euros are pulled 'southeast', not just south. This kind of thing is abundantly clear if you look at supp fig 8.17 in the paper, where off-cline pops are displaced in both the horizontal and vertical dimensions, if the pops on the x and y axis are not outgroups to each other. If this southeast pull is subtracted, E Euros will turn out less CW than NW Euros, because they are already as east as NW Euros as is and any reduction in this will make them less so than NW Euros.
In fact, because f4 stats are so robust, this plot allows us to predict very precisely what will happen to E Euros in Laz modelling were both EHG and CW to be included as sources of eastern ancestry. First imagine a space where points representing pops are arranged, and closer points have gene flow betwween them. Laz modelling gives enough contrib to a 'test' pop from some ancient references A, B, C etc, so that the 'test' pop reaches the correct 'position', as far away from each of the references A, B, C as their f-stats properly show. Now imagine two points in space that are somewhat close, EHG and CW. They are cllose because EHG contrib to CW, but they are not in the same place becasue CW carries West Asian ancestry, which pulls CW away. Now there are another lot of points representing Europeans, arranged in a line towards CW, with Sardinians at the rear and Norwegian and Lithuanian closest to CW. Most euros in the line are as far from EHG as their CW contrib would allow, so most Euros will be modeled with only CW contrib and no EHG contrib, because CW contrib (or closeness to the CW point, in other words) alone is suffficient to account for their distance from EHG. But E Euros will have to have EHG contrib on top of that, because they are much closer to EHG than their distance to CW will allow--they are off the line. In fact, because EHG and CW are close, pops like Lithuanian and Estonian will be brought pretty close to CW because of their closeness to EHG, a.k.a EHG contribution, and this will reduce the amount of CW contrib needed to bring them to the correct distance from CW in Laz modelling, in the case of finnish by a vvery large amount, moderately in Estonian and Lithuanian, less in Belarussian and Russian, very little in Ukranian and Czech. By the way, this is how much the f4 plot displaces each of these pops from the line, which is why I make this prediction.
In-non mathematical terms, the above scenarios are as good as saying that high levels of EHG ancestry will inflate the inferred Yamnaya/CW percentage, whatever the actual degree of Yamnaya/CW ancestry. Thats why in a model with only Yamnaya, finns are modeled in laz modelling as the most Yamnaya pop of all in Europe, which is patently ridiculous. The reason why CW/yamnaya ancestry is distinguishable from EHG ancestry is because CW and Yamnaya carry high levels of West Asian ancestry. Thats why the two points, EHG and CW, are close but not in the same place, and thats why NW Europeans end up being closer to CW and Yamnaya than NE Euros. The distance created by West Asian ancestry between CW and EHG is enough to discriminate between Euros too, as Euros that are pulled purely toward CW will be pulled in the West Asian direction as well.
The stats show us that NW Euros are very close to CW while being a bit further from EHG, while NE Euros are as close to CW as NW Euros, but very close to EHG as well. This means that any amount of 'unmasking' of CW/Yamnaya ancestry as actually EHG in a laz model with both EHG and Yamnaya/CW will decrease the amount of Yamnaya/CW ancestry to a level below that in NW Euros.
These are simple, concrete mathematical deductions; there is nothing 'fuzzy' about them. Drawing them out will make it easier to explain, but thats a PITA. I hope mental visualisations will do. Stats of this kind in f4s are the most 'real', because these deal with distances in genetic variation, which makes up quite literally the 'stuff' of pop gen in the first place, which is why Laz modelling and ADMIXTUREGRAPH and D stats are the beating heart of the best papers on pop gen.
This in facct brings NE Euros into line with the results of ADMIXTURE. It was always an open question whether or not IEs introduced West Asian-type ancestry into Europe, which was impt because this had big implications for the distr of yamnaya ancestry in Eur. Now that we know that Yamnaya was half West Asian, while Corded Ware were at least a third West Asian (or if you like a more restrictive defn of West Asian that does not incl SC Asian pops, 'Caucasus'-type ancestry, for which the numer is 30% and 20% from ADMIXTURE at K=20) the fact that the baltics score so low in West Asian is an important issue to address, wherever the West Asian component originally derived, because there is obviously an important dimension of autosomal variation which baltics and CW do not share, or shared at a minimal level.
PS As the component is more and more localised to the Caucasus and less and less to the Kalash, and thus and richer and richer in ENF and poorer in ANE, the baltics score less and less, which is precisely what we would expect.
rk,
You seemed to be talking about the f3 drift stats before, which actually look a bit dodgy, as was pointed out by Matt. The Fst distances look more reliable as far as I can see.
In any case, the Yamnaya admixture proportions were not based on f3 stats, so they need not show exactly the same patterns. The reason being that f3 stats are based on affinity, and do not necessarily reflect the level of direct admixture. You probably know this anyway, but Laz will likely reiterate it to you, and I'm sure explain the fact that the Yamnaya horizon needs to be sampled more extensively to work out more precise Yamnaya admixture levels in modern populations.
Btw, obviously Poles weren't sampled. Although it's interesting to note that one of the ancient German samples, probably from the Urnfield-Lusatian Culture, belonged to R1a-Z280, which is a major Polish and indeed Balto-Slavic lineage.
Then, RK, how much EHG/ ANE in CWC would you place from pre-existing FOrager groups c.f. Yamnaya input directly ?
David
"belonged to R1a-Z280, which is a major Polish and indeed Balto-Slavic lineage."
In which case its not a Balto-Slavic lineage, because it pre-dates Balto-Slavic.
@colin
"The Yamnaya ARE the Proto-Indo-Europeans; they are a mix of EHG and some Armenian-like component. So Proto-Indo-European is neither an EHG group nor an Armenian-like group."
Is this your conclusion or are you stating the Haak papers conclusion. All it seems to say is that evidence supports a "yamnaya like" source for northern and central european IE languages.
So it supports a steppe hypothesis in this context only not PIE. It specifically mentions that it does not cover south east Europe or Asia.
David, funny enough that most of the slavic populations fit in lines between Poles and something else in that PCA I requested a while ago.
https://drive.google.com/file/d/0B8XSV9HEoqpFMVNnd2RkSXNiTUE/view?pli=1
Kargopol between Poles and Komi, other Russians between Poles and Erzya, South Slavs between Poles and Greeks. The overlap of Lithuanians and Kargopol Russians is probably coincidental and would change with more dimensions even without adding samples to the PCA.
So in that sense it's a shame Poles weren't sampled for the Human Origins set. The same is true for Irish, Karelians and Vepsians.
No, that is patently false.
The plots are plotted f4 stats. Plotted f4 stats, david. They are the same thing the authors use for Laz modelling; I have already done a stripped down version of the first step of a Laz model with both EHG and Yamnaya, and the behaviour of east euros in that model is extremely clear.
Everything the authors say about comparative f4 stats : ".... has an advantage
over the use of the f3(Test; Ref1, Ref2) statistic in that these statistics are not affected by post-admixture
drift in the admixed population, but rather rely on allele frequency correlations deep in the phylogeny" applies, and there is not a single mention of f3s in my entire post.
The geometric implications of the above f4 plot in laz modelling are so obvious it would be surprising if the authors didn't say something like "this is consistent with a scenario of ehg survival" or "increased affinity with ehg might affect Yamnaya estimates, like what happened for finns".
As for the substructure argument, I've already said what the consistency of IE samples across space and time from both the extreme east and west implies, though of course the scientists are gonna hedge on this one, everyone does that.
Let's see what they say.
@ MT
CW prob got all their EHG from Yamnaya and Yamnaya-likes, but many Euros in E Europe prob got this diluted to a certain degree more than NW Euros, because the 30% or 20% W Asian ancestry in CW reaches a trough there. This implies that some EHG in E Euros comes from EHG survival, which is not surprising considering the high levels of WHG survival (the Yamnaya and CW are fitted as having quite a bit less WHG in Laz modelling than estonians and lithuanians and even norwegians) and ADMIXTURE results, not to say the f4 stats.
@Davidski,
Thanks for your explication! Can you tell me how you define Basal Eurasian and how it is different from your K8 ENF? Are there frequency distributions of Basal Eurasian from modern populations in your analyses? I do think you should pay attention to the island hopping/maritime colonization of the impressed/cardial horizon. The seafarers were likely Mesolithic fishers who knew the currents of the Mediterranean.
ENF = Basal Eurasian + X
What is X? My understanding is that X is a component in the Near East that is similar to the Native European WHG component.
In terms of mtDNA I suspect its related to K and U3.
ENF is possibly close to 60% Basal Eurasian and 40% of X.
@VOX
"What is X? My understanding is that X is a component in the Near East that is similar to the Native European WHG component."
Do none in your propaganda machine actually read the papers of other research groups?
The Haber paper, the subject of this thread, clearly states that populations in Anatolia, Armenia and the Levant were more early Neolithic European like than they are today.
Furthermore, on any PCA plot, Italians plot almost identically to Greeks.
Which means that the early Neolithic population continuum probably extended from Italy, through the Southern Balkans, into Asia Minor, and then extended to Armenia and separately south through the Levant.
So, in other words, this ENF component is not specific to the Near East.
Moreover, archaeology supports the idea of a late Gravettian in places like Italy, Austria, Slovenia, Bosnia and Albania.
It is not difficult at all to deduce the identity of "ENF" and its ancestros, who are certainly European-Anatolian in origin.
Anyone can see on a PCA that the Southern Balkans does not have very much "ANE" or "EHG" ancestry. We don't need piles of ancient DNA to tell us that.
So . . . the Gravettian . . . spans a vast region, from Iberia to Siberia.
You will just have to deal with the fact that "Basal Eurasian" is pretty much a meaningless concept and any models using it are bogus.
@Colin Welling
"I think the real point your getting at is which "component" is more responsible for PIE. I don't think the question makes a whole lot of sense. But related to that question, I agree that the IE types of r1a and r1b came from EHG."
This contradicts your earlier statement that you don't think that EEF (early European farmers) had much ANE.
The British Museum would be pretty empty without the Elgin Marbles, I have to say.
Thanks RK
Marnie - have you noticed something on Haaks plot - see which region lies smack bang in the middle of everything
@RK
Yeah, thanks.
@Mike Thomas
"Marnie - have you noticed something on Haaks plot - see which region lies smack bang in the middle of everything "
You mean, the Balkans, Southern France, Italy?
rk,
It's very likely that the population ancestral to the German Corded Ware had higher EHG than anyone else in the study.
And it's this EHG that was diluted by mixing with non-Kurgan groups.
Roy,
Lazaridis et al. explains the concept of Basal Eurasian better than I can. But I will say that I think it's very unlikely that there was any Basal Eurasian in Europe, or even very close to Europe, before the start of the Neolithic transition, because of the almost continental wide genetic divide between Mesolithic European foragers sequenced to date and Neolithic farmers.
http://biorxiv.org/content/early/2013/12/23/001552
All:
There is no evidence at all that the authors of the Neolithic were none other than Southern European, Anatolian and Zagros Hunters who successfully made the transistion to pastoralism and small scale farming.
See, for instance, this recent 2014 paper:
"A forager-herder trade-off, from broad-spectrum hunting to sheep management at Asikli Hoyuk, Turkey"
Stiner, et al:
http://www.pnas.org/content/111/23/8404.short
The paper is edited by Ofer Bar-Yosef.
"Scandinavian team looking for Indo-Europeans in Kazakhstan"
http://dienekes.blogspot.com/2015/02/scandinavian-team-looking-for-indo.html
Important quote.
"The researchers collected about 120 Bronze and early Iron Age bone samples in total from Pavlodar, Kostanai and Karaganda during their week-long trip to Kazakhstan"
@Davidski
"But I will say that I think it's very unlikely that there was any Basal Eurasian in Europe, or even very close to Europe, before the start of the Neolithic transition, because of the almost continental wide genetic divide between Mesolithic European foragers sequenced to date and Neolithic farmers."
The paper that you cite to prove there is a wide genetic variation between Mesolithic European foragers and Neolithic farmers is the 2013 Lazarides paper.
Yet, if you look closely at this paper, you can see clearly that Greeks, Turks, Turkish Jews and Armenians sit adjacent to each other.
So the suggestion that populations of Southern European Mesolithic foragers and Early Neolithic farmers are separated by a "continental wide" genetic variation is a complete figment of your imagination.
In any case, I think I would trust Ofer Bar-Yosef, before I would trust you.
@Krefter
""The researchers collected about 120 Bronze and early Iron Age bone samples in total from Pavlodar, Kostanai and Karaganda during their week-long trip to Kazakhstan"
That's nice, they will get a good look at the interaction on the Steppe and its relation to CW and R1a.
However, that will tell us very little to nothing about R1b hgs and the Early Anatolian and Balkan Neolithic, which dominates the "eastward shift" in most Western and Southern Europeans.
@Marnie,
Iberians are intermediate between north Europeans and EEFs, not EEFs and west Asians or Balkans. This much I know.
There's obvious non-EEF/Yamna-derived west Asian ancestry in Europe but it hasn't been seriously investigated by any academic studies or posters online.
Considering German Bell beaker looks like a mix of EEF, WHG/SHG, and Yamna I tend to think it's R1b came via east Europe.
I don't know though that's just where the most recent evidence seems to be pointing. David Reich also thinks this is the case.
@ marnie
"ere is no evidence at all that the authors of the Neolithic were none other than Southern European, Anatolian and Zagros Hunters who successfully made the transistion to pastoralism and small scale farming."
Yes I agree; but we don't know what their genetic affinities were cf other european foragers .
@ Krefter
Yes I sae that title. Wish they also samples neokithic Kazakhstan
@Krefter - "Iberians are intermediate between north Europeans and EEFs, not EEFs and west Asians or Balkans. This much I know."
Didn't EEFs themselves have West Asian and Balkan heritage?
@Mike Thomas
"Yes I agree; but we don't know what their genetic affinities were cf other european foragers . "
We don't have ancient DNA, but we have a very, very good idea, based on this Haber paper, and the position of Greeks, Turks, Turkish Jews, Italians and Sardinians as to the genetic signature of the foragers that became pasturalists in the Early Neolithic.
Furthermore, if you read recent papers on early Neolithic ceramics in the Southern Balkans, the are rejecting Childe and suggesting that ceramic technology was known for over 10,000 years before the Neolithic in the Balkans. They just didn't use it to make pots.
So, I don't buy it at all that the authors of the early Neolithic were dramatically different from people in the Southern Balkans today.
That's just BS that people have been saying in Western Europe for centuries.
@Ryan
"Didn't EEFs themselves have West Asian and Balkan heritage?"
The Balkans and Asia Minor were not separated 10,000 years ago. The Bosphorus only formed about about 8000 years ago, after the Early Neolithic.
Genetically, on a PCA, even today, people from Western Turkey and Northern Greece sit adjacently.
You're right Marnie
Recent insights into the Gravettian see it as something forming over a wider area: Balkans - Near East rather than arriving de novo into the Balkans from the near East.
However today , perhaps except for non-northern Greeks ; there is a clear 'gap' between modern Balkan groups cf near eastern ones- undoubtedly the result of more recent events
@ All
Is there a chance we're focussing too much on eastern intrusions in BA Central Europe ; and not seeing it holistically as a process of "homogenization" of previously separate elements
@Krefer
"Considering German Bell beaker looks like a mix of EEF, WHG/SHG, and Yamna I tend to think it's R1b came via east Europe."
Why are all these studies focused mostly on Germany?
"I don't know though that's just where the most recent evidence seems to be pointing. David Reich also thinks this is the case."
Which most recent evidence? The Samara samples? Yet, no attempt to find ancient DNA in the areas where greatest R1b diversity shows up, which is in the Southern Balkans. And you can't argue that it's because of lack of preservation, because some alpine area in the Southern Balkans would have very good preservation.
You just haven't bothered to look.
@Mike Thomas
"However today , perhaps except for non-northern Greeks ; there is a clear 'gap' between modern Balkan groups cf near eastern ones- undoubtedly the result of more recent events "
Yes, the formation of the Bosphorus and other events, which Haber nicely summarizes in this Armenian paper.
Re: continent wide divergences, going on a slight tangent, if you go by FSTs
Basque-Han = 0.120, Basque-Mala (from India) = 0.061, Czech-Mala = 0.051, Basque-Yoruba = 0.155, Czech-Yoruba = 0.149
WHG-LBK_EN = 0.091, WHG-EHG = 0.078, WHG-SHG = 0.05, WHG-Yoruba = 0.195, EHG-Yoruba = 0.179
I don't think its lacks legitimacy to say WHG and LBK had an almost continent wide FST, as comparing Basque and Han divergence to them.
However this is also true of a lot of fairly recently divergent groups in the Americas of only Native American ancestry. The divergences between the European hunter gatherers are also very high.
Admixture has pretty non-linear effects on between group FSTs -
For instance, the Kargopol Russians and Mordovians in the FST table in Haak probably model OK as a combination like Yamnaya, EHG and Han like Siberians, all of whom have higher differentiation from Yoruba than present day Europeans.
Yet Russians and Mordovians have a slightly lower FST (population differentiation) from Yoruba than all the other North Europeans in the sample (0.148 vs 0.150 and north of that for other North Europeans). Probably, I think, the enriching effect of combining these different sources cancels out the fact that all these sources have a relatively high FST from Africans. The FST is not a linear combination of the presumptive ancestral sources (so I think you can't really estimate ancestral combinations from testing linear combinations of FST very well, particularly for complex scenarios).
Estimating the time depth of separation between the WHG and Early Neolithic group seems difficult. It seems apparent that there was heavy drift in the European hunter gatherer groups which strongly separates them from relatives, how much time this took to build up is harder to say, and where the limits of each population (or meta-population fall).
Could be that there was a genetic cline from the Near East->Balkans then a harder barrier with the rest of Europe (although KO1 the Hungarian HG has some drift stats that suggest a slightly greater relationship with the Near East IRC). We'll know more when we have the samples. It's not like a cline there is as a proposition any crazier than a hard genetic boundary between the Levant / Anatolia and Europe.
@Matt
I appreciate your analysis.
However, it's kind of useless when ancient DNA is missing for virtually all foragers and early pasturalists in the areas where it occurred.
And in this case, I will lean on recent archaeology, and not analysis.
Recent archaeology suggests that people in Anatolia and the Southern Balkan who made the transition to early pasturalism did not come from far away in their recent past, but made the transition from foraging to pastoralism in place.
I've posted one of the papers that suggests this. There are many others that have been published in the last five years that suggest the same thing.
Marnie
I agree that R1b will pop up in the pre-Bronze Age balkans, as I've mentioned earlier. I'm just not sure as to its overall sitting in the R1b family, and it's autosomal behaviour. But I suspect that it'll be ancestral to Western Yamnaya rather than derived !
I know Dave will think this is stupid and unfounded :)
As for lack of research into balkans; I don't think we can blame the Germans . It's only natural they'd dig up their own backyard first
@ Davidski
It is at least as true that 'Yamnaya/CW introduced something near-east-like into Europe which was not present before' as it is that 'Yamnaya/CW introduced EHG into Europe which was not present before'.
The group that contrib to CW in Germany--the 'more EHG yamnaya group'--are at least 42% West Asian going by the 60% yamnaya 10% Samara figure, and thats the lowest figure, a floor, as fitting CW as 75% yamnaya or 80% yamnaya will drive up the West Asian estimates. And the CW was in germany, from the western end of Yamnaya presumably. So good luck wishing that half of the ancestry away.
The 100% EHG Karelians distinguish NW and NE Europeans by little. unless you find some group with >15% West Asian its not gonna work. Laz modelling with both EHG and CW make this clear.
@Mike
"As for lack of research into balkans; I don't think we can blame the Germans . It's only natural they'd dig up their own backyard first "
Yes, sure, but then only make inferences and press announcements about your own back yard, and not grandiose press grabbing statements for all of Europe that specifically reach beyond the geographic coverage of your sampling.
Don't get me wrong. I drive a Volkswagen, and as an engineer, even regularly interact with Germans on a professional level. My favorite beer and favorite sausage are German. I admire many things about Germany and even feel bad for the average German worker that has to tow the boat for the PIIGS.
However, ignoring the impact of the Neolithic in a major population study on Europeans and the origin of PIE seems kind of crazy. Especially given the diversity of R1b in the Southern Balkans, as well as the fact that Greek and Albanian are near the PIE root.
"as well as the fact that Greek and Albanian are near the PIE root."
Not sure I agree that we can posit what original PIE was like based on modern languages. Esp Albanian given that is has been radically affected by Slavic , Greek, Roman, etc - whatever it originally descended from.
But I agree that the oldest IE languages are in the Balkans and anatolia
Western and Northern European IE are a younger subset
rk,
I'm pretty sure I'll be able to show you the process of EHG dilution in the early Kurgan groups sooner or later, perhaps even with the Haak et al. samples, because it appears to me that there is substructure in this direction in their Corded Ware and even Yamnaya genomes.
Also, I think you're focusing on a red herring when you talk about the West Asian component.
What I'm seeing in the data are not two different Near Eastern or West Asian components, but two different types of ANE; one mostly found across northern Europe, Siberia and the Americas, and now basically known as EHG, and another mostly limited to West Asia, Central Asia and South Asia. My guess for now is that MA-1 comes from a time before they fully separated.
It looks like Yamnaya had mostly the European type of ANE, but also some of the Asian ANE, which isn't surprising since the Kurgan cultures straddled the Europe/Asia border.
I think this is what's causing the assumed difference between the Near Eastern admixture in the EEF and Yamnaya. I'd say their Near Eastern admixture was basically identical, and what differentiates Yamnaya from EEF is not only a high level of EHG admixture in the former, but also Asian ANE admixture.
@Marnie,
"Yet, no attempt to find ancient DNA in the areas where greatest R1b diversity shows up, which is in the Southern Balkans."
A heavy ENF pop can't be the source of R1b in west Europe. That is unless a heavy EHG and non-R1b pop made a much bigger effect around the same time. The evidence is pointing towards the steppe.
If R1b came from a pop originating in the Balkans, it made little genetic impact, and a another pop who didn't have R1b and had alot of EHG had to of made a huge impact at the same time.
Unless R1b-L23 was introduced from the Balkans to the steppe long before it arrived in west Europe, then arrived with heavy EHG pops, could this work IMO.
better late than never Dave !
Well done
Slowly you're seeing what I stated long ago.
Finally, you can admit that the eastern shift has more than one sources.
So you're admitting that ANE was already present in west Asia before the Bronze Age ?
Soon, you'll see that ANE was also present in Europe, apart from Scandinavian Mesolithic and Karelia, but also more centrally.
Now, just brush up on what you think about archaeology of Kurgans and Mesolithic foragers and we'll make half a scholar out of you yet ! :)
Mike,
No one here said steppe is the only source of ANE. Many of us said it's the main source.
No Mike, I don't think ANE was in West Asia before the Bronze Age. The substructures created by ANE in the Near East and Caucasus are far too choppy for it to be native to any part of the Near East. Just look at the PCA I posted above.
I think we can point to a number of sources of ANE in the Near East, including the people ancestral to the Hurrians, the early and late Indo-Europeans, etc...
Here's a paper I'm reading now.
http://www.google.com.au/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&cad=rja&uact=8&ved=0CDMQFjAG&url=http%3A%2F%2Fkura-arax.tau.ac.il%2Fsystem%2Ffiles%2FKohl.pdf&ei=odbrVMmdM-fimAWZsoLADQ&usg=AFQjCNEy213W_cOURJO3TyNMEyXq896ulw&sig2=5e787jpibhZLDEILfZL5XA&bvm=bv.86475890,d.dGY
Thanks for that paper ; I'll have a gander
But then; isn't that against what Ryu and u have discussed - the west Asian element in yamnaya also had ANE. Ie, If true, given that the west asian signal preceded the Bronze Age,then there was already ANE in central- west Asia ?!
There's no such thing as an ancient West Asian component. There's just ancient Near Eastern with and without ANE, plus recent drift in the Caucasus and Hindu Kush doing weird stuff to the results of the ancient genomes.
I'll show you what I mean when I get the Haak et al. ancient genomes.
YEs that paper is good , Dave
BTW Kohl is a Kurgan criticist :)
the KA phenomenon is also mysterious. A touch later than Majkop, it nevertheless does not derive from it. In fact, it has a clear difference to it. That it spills out onto Anatolia and northern Iran, but also as far as Syria is interesting.
But I take it you're suggesting that is a source of ANE in west Asia ? Perhaps, but then, as above, it doesnot appear to derived from Majkop or anything north of the Caucasus.
Im sure itll all be cleared up soon, as western scholars take more heed of this region and refine their chronology.
@David
''
There's just ancient Near Eastern with and without ANE, plus recent drift in the Caucasus and Hindu Kush doing weird stuff to the results of the ancient genomes.
I'll show you what I mean when I get the Haak et al. ancient genomes.''
But does the Academics like Haak etc Understand that?
It's easy to be led astray by modern DNA and unsupervised ADMIXTURE runs, and I think that's what happened in this case to some extent.
For instance, I very much doubt that the Near Eastern population mixing with the EHG hunter-gatherers on the steppe was Armenian-like or Iraqi Jewish-like. That's just what it looked like when they tried to model the Yamnaya based on all of the samples they had.
'' I very much doubt that the Near Eastern population mixing with the EHG hunter-gatherers on the steppe was Armenian-like or Iraqi Jewish-like. That's just what it looked like when they tried to model the Yamnaya based on all of the samples they had.''
Archaeology,Anthropology agrees David anyway for logical sake tell then Who were those ''Near Eastern'' like populations actually were (Genetically)?
It's hard to say; let's wait for aDNA from Neolithic Syria, Armenia and Iran.
But it will take time! Give a Guess i think you already have it:)
@Krefter
"A heavy ENF pop can't be the source of R1b in west Europe. That is unless a heavy EHG and non-R1b pop made a much bigger effect around the same time. The evidence is pointing towards the steppe."
How do you know?
"If R1b came from a pop originating in the Balkans, it made little genetic impact, "
Yes, R1b obviously didn't scale up in the Balkans the way it did in Western Europe. But that doesn't mean it didn't pass through the Balkans.
"and a another pop who didn't have R1b and had alot of EHG had to of made a huge impact at the same time. "
Depends what part of Europe you're talking about.
"Unless R1b-L23 was introduced from the Balkans to the steppe long before it arrived in west Europe, then arrived with heavy EHG pops, could this work IMO."
I think its more complicated, as per:
http://linearpopulationmodel.blogspot.com/2015/01/r1b-l23xm412.html
and
http://linearpopulationmodel.blogspot.com/2015/01/r1b1a2-or-m269xl23.html
M269(xL23) possibly in Italy and Balkans before or at same time as L23 on the Steppe and Anatolia.
Also, Haak R1b sample in Iberia. That's another hint that R1b may have been west of Balkans prior or along with Steppe expansion of R1b.
Also, given Oetzi, early Neolithic expansion was associated with other lineages besides R1b . . . certainly J2 and G.
I don't necessarily think this had to be all cloak and dagger. It could have been something like pasturalists exploiting a niche that I hunters weren't expert at. Probably, pasturalists had more children than hunters because of more predictable food source. So in the long term, pasturalists could increase their pop above hunters (without horses or continuous warfare.)
@Marnie,
"How do you know?"
A high ENF pop as the source of western R1b(aka R1b-L11) doesn't work because after the Neolithic west Europe became less ENF and more EHG/ANE.
I know this because of PCAs, admixtures, and F-statistics from Haak 2015, Laz 2014, and Eurogenes.
"Depends what part of Europe you're talking about."
Northwest Euros and Iberians have the same signal of post-Neolithic admixture. Iberians are apart of the same trend as northwest Europeans, just more on the EEF side. West Asian intro. of R1b in Iberia doesn't work.
Iberians can fit as ~25% Samara Yamna and ~50% German Bell Beaker(info from Haak 2015) who were both rich in R1b.
You don't see such obvious signs of admixture from west Asian-like pops in Europe like you do with steppe pops. We lack ancient DNA from the Balkans and west Asia, but I doubt they'll change anything when it comes to post-neolithic gene flow into west Europe.
"M269(xL23) possibly in Italy and Balkans before or at same time as L232 on the Steppe and Anatolia."
I'm specifically talking about west European R1b-L11. Yes I know R1b there may be very old, but autosomal DNA suggests L11 is derived of the steppe.
I'm just stating what seems obvious to me. Ancient DNA has thrown out plenty of surprises so I not 100% confident in this.
I'm not quite sure what happened but I made a post yesterday and I don't think it went through properly. It was basically giving a linguistic proof why proto-IE couldn't have come from India because proto-IE had words for washing according to James Mallory.
If that is not enough proof for people like Nirjhar, heres a video from David Anthony:
https://www.youtube.com/watch?v=HliaR2Ep24s#t=3679
David Anthony points out that the lactase gene that is found in Indo-Europeans, is the European version of the gene. More undeniable proof that IE language was carried from Europe to the rest of the world.
OK, sorry, I thought it was an attempt to insult Indians.
But how is that relevant? The word for washing I mean?
"OK, sorry, I thought it was an attempt to insult Indians.
But how is that relevant? The word for washing I mean?"
I'm not sure if it was meant as an insult by James Mallory, just a humorous but reasoned take on why proto Indo-European coud not have originated from India as:
1. proto Indo-European contained words for washing
2. Indians aren't particularly known for their personal hygiene.
At least that is my take on it. Here is the video if you want to see for yourself.
https://www.youtube.com/watch?v=oERPeM1A_GM#t=675
Anyways, I love reading this blog and am looking forward to future posts.
@Krefter
Let's just say there was probably a lot of fine scale shifting back and forth, as well as in place population expansions and crashes.
I'll leave it at that for now.
Proto-indo-european had words for washing but they also had a word meaning to "fart in a very loud and ostentatious manner" which is *Perd.
In russian there is a word today, пердеть, which as a fluent russian speaker, I can tell you means to fart, but not just fart, but unleash a nuclear explosion upon your enemies, :P
This is one of those curiosities of linguistics that are really funny.
So, paleolinguistics tells us that the PIE-ians were fond of farting, but what they ate to give them indigestion escapes me. Dried horse meat?
http://en.wiktionary.org/wiki/Appendix:Proto-Indo-European/perd-
@David
DELETE that kids comments!! he is sounding like a retard.
Well, he's quoting Mallory, so it's not a random ethnic slur. I don't want to censor anyone here unless I really have to.
I think that's an ardently pseudoscientific comment , Dave, by some troll cowering behind anonymity.
@Davidski
"Natural selection could be an explanation, because it seems that European eye and hair pigmentation lightened up just as rapidly as R1a and R1b moved across Eurasia. But I'm not aware of any advantages that these markers give."
There's a pretty mega irony brewing there if the phenotype is the product of multiple de-pigmentation genes from different adjacent populations.
The same process could have merged advantageous alleles from each of the contributing populations creating an advantage.
@spaghetti
"So, paleolinguistics tells us that the PIE-ians were fond of farting, but what they ate to give them indigestion escapes me."
Milk before fully LP maybe?
@Matt
Nice catch, I hadn't noticed that! The Esperstedt MN guy was Funnelbeaker, and they are megalithic. Looks like there was some gene flow involved there.
Better yet, one of the other Funnelbeaker guys carried Y haplogroup R... though he was *not* L11.
The MN sample belonging to R* is from a Baalberge group burial site.
According to Gimbutas, this culture represented an early Indo-European intrusion from the steppe into Central Europe.
http://en.wikipedia.org/wiki/Baalberge_group#History_and_scholarship
But yeah, generally Baalberge is just seen as part of the Funnel Beaker culture.
Dave, the R* dude in Baalberg was virtually all EEF, with a touch of WHG. Still think R originated in the forests of far eastern Europe ?
Yes I do. The classification was R* because it was a bad sequence, not because it was ancestral to modern R lineages.
Also, uniparental markers don't always correlate with genome-wide genetic structure in individuals, but they usually do in populations, especially homogeneous ones. So it's already very difficult to decouple early R1a/R1b clades from EHG. There's simply no way that Karelia_HG and Samara_HG could've got their R1a/R1b Y-chromosomes from recent admixture.
Im not saying that the R1a and R1b in EE arrived from anywhere, at least, not recently, and not of those earlier clades from Mesolithic samples.
But nevertheless, it is patently obvious, even at this early stage of aDNA field work, that R* , R1, and even R1b and R1a represent a mixture of demographic histories in Eurasia, whose span crosses various different ages.
IMO, the mistake you're performing in your reconstruction is that your envisioning a 1-and-a half dimensional process.
R-M198 and R-M269 just look like late Neolithic markers that expanded from the steppe with the early Indo-Europeans, who were in large part of EHG stock.
It's really not any more complicated than that.
OK Dave. Your position is clear. It might be right, and I guess only time can tell.
But can you explain one thing to me:
How do you know that CWC descends from Yamnaya, and not that they both share more or less the same composition- ie a similar mix of EHG, WA, and ENF mix ?
Afterall, Reich et al simply chose to model other groups on Yamnaya based on their already formed hypothesis, and the still limited sampling of data.
Moreover, you claim that "WA" is a heuristic category, whilst maintaing unequivocally that "EHG" is a pure, long-lasting R1-distinct collective ?
@Davidski
"What I'm seeing in the data are not two different Near Eastern or West Asian components, but two different types of ANE; one mostly found across northern Europe, Siberia and the Americas, and now basically known as EHG, and another mostly limited to West Asia, Central Asia and South Asia."
David, if once you get the genomes you can pinpoint this difference it will be awesome.
I remember many months ago a short discussion we had about ANE vs. West Asian being a Red Herring. My argument was that ANE was too ancient (taken from MA-1) to track the Neolithic movements of people. "West Asian" is far from perfect, but I argued that ANE was a step back, not forward.
Now, as you say, this ANE might have split in 2 subgroups, one north (modern Russia, Kazakhstan, Ukraine..) and one south (Turkmenistan, Iran,..).
With the Samara samples you might be able to sort this out, though maybe until we get some old sample from around early (7th millennium BCE) Kopet Dag settlements (or somewhere else, who knows) we won't know for sure.
Also I agree about the not so "Armenian-like" expectation from the population that entered the steppe. We'll see what they were exactly, but I think they will be more "eastern" (Pashtun might be the best fit, we'll see).
The good news is that more and more aDNA is coming (Maykop on the way, apparently also Kazakhstan Kurgans, Greece,... and in max.2-3 years we'll have samples from all around the relevant places).
Mike, the model is a resounding success; the Corded Ware people can be fitted without any trouble as mostly Yamnaya with some Central European farmer admixture, which makes perfect sense, and the stats also show that their ancestors were basically like Yamnaya, albeit with a higher level of EHG ancestry.
Whether this means Corded Ware was an offshoot of Yamnaya, or they both shared recent ancestors, is an open question, but one that doesn't complicate anything for the PIE steppe hypothesis.
You'd have an argument if Corded Ware arrived in Central Europe a couple of thousand years before or after Yamnaya, but that's obviously not the case. Both of these cultures were expanding at about the same time, and on top of that, many archeologists and linguists saw them as close relatives simply based on archeological and linguistic data, which aDNA has now confirmed.
Alberto,
My guess at the moment is that the Asian ANE among the Yamnaya came from contacts with the Botai people, who might have been overwhelmingly ANE.
"Whether this means Corded Ware was an offshoot of Yamnaya, or they both shared recent ancestors, is an open question, but one that doesn't complicate anything for the PIE steppe hypothesis. "
Yes it does. It totally changes it.
"You'd have an argument if Corded Ware arrived in Central Europe a couple of thousand years before or after Yamnaya, but that's obviously not the case. Both of these cultures were expanding at about the same time"
Im not saying that CWC culture "arrived " earlier than Yamnaya. They both formed at the same time, yes, and they expanded at similar if not same times; but this doesn't mean one derived from the other. It's a simple premise.
" many archeologists and linguists saw them as close relatives simply based on archeological and linguistic data,"
Dave, that doesn't mean much, especially when the current archaeological concensus does not see CWC derived from Yamnaya. So what are you talking about ?
Whatever the case, David, you're soon going to face reality and realize that what set everything in motion was events in the Caucasus, Anatolia and NW Iran in late M5, early M4.
"My guess at the moment is that the Asian ANE among the Yamnaya came from contacts with the Botai people, who might have been overwhelmingly ANE"
But you mean that kind of ANE that is not the same as the one in the steppe?
That would be strange. If there is a split in ANE (one north, other south), Botai should belong to the north one (they are speculated to be Uralic speakers, like the Kelteminar people south from them in Uzbekistan).
The "southern ANE" should start just below, in Turkmenistan, and it's the one that spread to the Near East and South Asia.
Europe seems to have received a mix of both, weather from Yamnaya (which was in itself a mix of both) or from more sources/routes.
Botai were steppe people, while early Uralics are generally thought of as forest people.
I think the northern ANE was restricted to Europe and Siberia, while the southern ANE was all over Kazakhstan, and indeed most of the ~stans, up to the Altai and down into the Hindu Kush.
"Botai were steppe people"
Precisely, that's why I think they should be more related to the northern ANE than to the Southern ANE. People from Turkmenistan/Iran were definitely not steppe people, they had the Neolithic package since very early, and cultural -and likely genetic- contact with the Near East. So it's easier to put Botai in the same group as Pontic-Caspian steppe than in the group with Turkmenistan/Iran.
In any case, I do like much more your current ideas about ANE split and Botai influence than your previous speculations. Now things start to make a lot more sense.
Botai could be either , or more likely - in between. This means that "ANE" is a far older pan- North eurasian signature which partly underlies later EHG, LN Central asian groups, etc.
Nah, I think these ancient populations had their own very specific niches and only came beck into regular contact when some of them started expanding in a major way, like the Yamnaya.
European and Siberian EHG/ANE weren't running around Kazakhstan because there's a lack of the European forager signal in most of Asia, except Siberia and Tajikistan.
Anyway, this looks interesting...
http://www.shh.mpg.de/32426/settlement_of_europe
So David Where did R* originate?
In South Siberia, same as Q.
I see, then Where the Split of R1 and R2 happened?
David
"I think these ancient populations had their own very specific niches and only came beck into regular contact when some of them started expanding in a major way, like the Yamnaya."
Yes, that's right. Glad youre picking up.
Except the expansion was in multiple directions across the steppe from c. 2500 BC, and not solely from Yamnaya
And it was a secondary phenomenon, following precedent events in central Asia, Caucasus, Balkans, etc.
Mike, is there expansions from Yamnaya to Asia?
Depends what you mean by expansions and asia ?
Yes people moved around; but in many directions . There was nothing central about yamnaya specifically; and ive outlined about what the major impetus was.
IMO- we need a clearer genetic picture of the vast regions between Central Europe and Central Asia . That'll take years yet
R* must have developed in central Asia ; if not further southeast
Mike You are the coolest comment and conclusion maker here!:)
that is hypothetical, admittedly based on modern phylogeny ; which is problematic admittedly. But I can't imagine it being far off
The more important question is specifixally what happened with R1-M269, Z-93, Z-280 etc.
Far more samples needed
@Mike
I have one further thing to say about Greek.
Greek is a very hard language to learn. I actually took some classes in both Modern and Ancient Greek at UC Berkeley.
In addition to being highly declined, it has many ways to construct sentences that don't exist in French, German or English.
It has one of the largest vocabularies of any Indo-European language.
I have spoken to some Czech speakers and Czech my be as complicated as Greek in its declination scheme and genitive.
In addition to this, the Mountain Greek spoken in Epiros and Western Macedonia could not be a recently learned language. Based on how unusual it is, with verb formation retained, but with also grammatical structure sharing with Macedonian and Albanian, I suspect it has been spoken in place for at least 2,500 years.
You speak of Slavic invasions, and these did occur in some parts, but there is not much evidence of this in the mountains of Greece.
So I'm not sure what to make of this, but I suspect that the Greek language did not arrive in Greece with Steppe invaders.
@Marnie
"I have one further thing to say about Greek.
Greek is a very hard language to learn."
This statement obviously needs to be qualified. Greek is a very hard language for YOU to learn. My cousin moved to Greece with her four year old daughter. The child had no problem picking up the language in just a few months! And as a matter of fact, most of the local kids her age also spoke quite good Greek without any schooling whatsoever.
Also, I'm very sorry to say that your suspicions about how long the language has been spoken in a particular location, and your conversations with international friends and colleagues doesn't add any weight to your arguments.
@Davidski
"European and Siberian EHG/ANE weren't running around Kazakhstan because there's a lack of the European forager signal in most of Asia, except Siberia and Tajikistan."
Yes, this is true. So you may be right about Kazakhstan being part of the same population as the rest of Central Asia, with the "southern" ANE component (maybe we should tentatively call it Ancient Central Asian (ACA), till you find it and name it yourself :)
So this is interesting. No idea if this is what you're thinking too, but the way I see it, this would leave a map like:
- ANE spread throughout Eastern Europe (Russia, Ukraine,... mixed with WHG to make the EHG) and Siberia (on its own or mixed with East Asian as you move east)
- ACA in Kazakhstan, Uzbekistan, Turkmenistan, Afghanistan (?) and good part of Iran (north east, at least).
Let's say that Pakistan was mostly South Asian, and that the rest of Iran was mixed with the neighbour populations (NE in the west, SA in the south).
From this situation, people from Kazakhstan move west to Yamnaya. People from Turkmenistan, Iran, Afghanistan moving also west (Near East, Armenia, Anatolia) and south (Pakistan, North India,...)
This would solve the problem with the strange ANE/West Asian duality, and the problem with many ANE-rich populations lacking HG ancestry (the EHG to India hypothesis never made a whole lot of sense, IMO. And now we even know that EHG as such mostly went away with Yamnaya).
David, are the volga-ural people like chuvash and maris direct descendants of some of the PIE people? As far as I know none of them speak indo-european languages?
Will you be able to test yamnaya proportions among various eurasians groups when you get the samples?
The R marker in Baalberge could be spurious, but he did have P1, so he was either Q, R, or P*, and R is the most likely possibility. In any case if Q and R are both from Siberia (or wherever you like), then he must have got his Y from there one way or the other.
@Karl_K
"This statement obviously needs to be qualified. Greek is a very hard language for YOU to learn. My cousin moved to Greece with her four year old daughter. The child had no problem picking up the language in just a few months! And as a matter of fact, most of the local kids her age also spoke quite good Greek without any schooling whatsoever."
Very petty. If you knew anything about language acquisition, you would know that children can quickly acquire any language.
However, for adult speakers, it is more difficult. For adult speakers of English, here is the rating system that qualifies difficulty:
http://www.effectivelanguagelearning.com/language-guide/language-difficulty
For adult speakers learning as a non-native speaker, Greek is rated as a category 4, which is difficult.
My point is that it is unlikely that the people in Greece today could have passed their language on for 2,500 years, with not very much change, even retaining some dialects, without mostly passing it directly from parent to child.
"Also, I'm very sorry to say that your suspicions about how long the language has been spoken in a particular location, and your conversations with international friends and colleagues doesn't add any weight to your arguments."
My view of the longevity of Greek is shared by many language experts, including those in the linguistics department at UC Berkeley and the Aristotle University of Thessaloniki.
But, don't mind me. Take the opinion of some of these pseudo linguists, working at second tier universities, who study a few words about horse, wheel and fart vocabulary, over the opinions of an actual native speaker expert on Greek at the UC Berkeley and also native Greek speakers at the Aristotle University of Thessaloniki.
sM,
Yeah, I should be able to learn a lot with those Yamnaya genomes.
Apparently the dataset is coming out very soon, maybe within days.
"My point is that it is unlikely that the people in Greece today could have passed their language on for 2,500 years, with not very much change, even retaining some dialects, without mostly passing it directly from parent to child."
What was your point exactly? That the genetics of the Greek people couldn't have changed for over 2000 years because the language is hard for adults to learn?
As you just stated, children can easily learn languages that neither of their parents can speak. These children then occasionally grow up and teach their own children how to speak these other languages.
Also, some families have two parents that natively speak different languages, and the children learn either one, or both, or even a third bridge language. I know it sounds crazy, but it's true. You can ask your linguistic friends at UC Berkeley about it.
@Karl_K
The point I'm making isn't a very difficult one to test genetically, and I'm sure that someone will eventually get around to running the appropriate ancient DNA tests.
Phony balonie youtube videos and books about [fantasy blonde haired] Steppe warrior kings probably won't cut it in the long run. (Although I'm sure it will receive disproportionate funding, media attention, and even preferred consideration by Science Magazine, for as long as possible.)
@Davidski,
"Apparently the dataset is coming out very soon, maybe within days."
Who told you this?
A reliable source.
@ Marnie
"You speak of Slavic invasions, and these did occur in some parts, but there is not much evidence of this in the mountains of Greece."
Well, there is no evidence of continuity in Greece between 600 - 800 AD, apart from a handful of centres like Thessaloniki, Philippi, around Athens, Cornith, and eastern Peloponessus. Literally, large swathes of the country were depopulated and later repopulated by mixed populations - initlally Slavs, then "Greeks", Armenians, etc from Asian Minor and Southern Italy.
THus the conservacy of Macedonian GReek highlands is still a recent phenomenon.
", but I suspect that the Greek language did not arrive in Greece with Steppe invaders"
Possibly. Might have come directly from Anatolia.
Alberto
"From this situation, people from Kazakhstan move west to Yamnaya. People from Turkmenistan, Iran, Afghanistan moving also west (Near East, Armenia, Anatolia) and south (Pakistan, North India,...)"
Quite possibly. Quesiton is what route ? Via south of Caspian thence through Caucasus, or directly via the north. Both ?
Either way, it is clear that this severely undermines, nay, disproves the centrality of Yamnaya, thus requiring a radical requalification of the Kurgan hypothesis at best, or wholly discarding it at worst.
Post a Comment