January 01, 2014

Happy New Year 2014

What's on your wish list for the new year in the world of anthropology and human genetics?

Here's my #1 item:

Any ancient African DNA.

The study of prehistoric Eurasians has revealed that modern populations are not simply descended from the people who lived in the same areas even a few thousand years ago.

The people best preserving the genetic legacy of central European Neolithic farmers can be found on the island of Sardinia; of west European hunter-gatherers in the shores of the Baltic; of Upper Paleolithic Siberians in the jungles of the Amazon; of Middle Paleolithic Siberians in Papua and Australia. 

And yet, the model for Africa largely remains one of continuity across two hundred thousand years, since the emergence of anatomically modern humans in eastern Africa.

There have been hints that this isn't the case; the study of modern populations has revealed evidence for both archaic African, and -more recently and surprisingly- even a little archaic Eurasian ancestry in virtually all Sub-Saharan Africans. Populations from one of the presumed cradles of H. sapiens (Eastern Africa) are now conclusively known to be recent mixtures of West Eurasians, and even the Bushmen of southern Africa, the subject of so many TV documentaries as an exemplum of the ur-Humans did not escape this admixture.

Paleoanthropology also hints that some of the people who lived in sub-Saharan Africa well into the Lower Stone Age may have been quite divergent, and so do modern human Y-chromosomes. T

So the €1,000,000 question is: who lived in Africa 5 or 10 or 50 or 100 thousand years ago?

57 comments:

Grognard said...

Based on the evidence the only reasonable conclusion I can come to is that the people living there have been largely absorbed by the constant migrations into africa.

Daniel Szelkey said...

THe sudanese university who published this went off line, but I saw the site before it went down, and the results were pronounced valid by the university, altough they are vague and I do not trust the authencity to much.
Haplogroups A-M13 was found at high frequencies among Neolithic samples. Haplogroup F-M89 and YAP appeared to be more frequent among Meroitic, Post-Meroitic and Christian periods. Haplogroup B-M60 was not observed in the sample analyzed.

INFINITO said...

Needless to always look for the same thing for years never found. If I say that my research leads me to prove, that the man has been widespread (not coincidentally) around the planet along with all species of animals and plants? The territory made ​​after a selection of adaptation of species?

mooreisbetter said...

Here's my 2014 wish list:

(1) People realizing that the letter lines of Y-DNA ancestries are just arbitrary splits assigned years ago that have little relevance to how closely related two distant lines may be. For example, the splits within the various clades of "I" and "R" are so deep, they might have been later (also arbitrarily) assigned their own letters, if there were more letters left. See the Haplogroup K to T transition.

(2) People realizing that the companies that advertise knowing your exact percentage of one nationality or another -- or your likelihood of getting a disease -- are fraudsters, and seeing more prosecutions of the same. None other than the New York Times just reported how one reporter took the same test at three different companies and got three different results. I've also heard of people with traceable ancestries within the Alps being told they were Native American, and when pressed, the testing company employee asked, "isn't that what you wanted?"

(3) Getting some more Ancient DNA from far Western Europe: the British Isles, France, Portugal, Ireland, etc.

(4) We've seen a lot of I Y DNA in Ancient Specimens. I'd like to see a workable model as to whether the modern hotspots like Bosnia and Croatia were the source -- or a modern refuge. That is, no one can really state whether I-bearing men expanded (into Mesolithic Central Europe) from said hotspots after the Paleolithic or whether the modern concentrations can be explained by I-bearing men starting elsewhere in Europe and then moving to seize territory in more southerly climes.

(5) Speaking of #4, another 2014 wish is for people to understand what I call the Goths/Huns model. There is too much chest thumping among people in DNA, based on their misguided pride and notions of ethnic superiority. Some of this, I suspect, is subconscious.

Everyone would be wise to know the Huns/Goths story, to see how the truth is always more complex.

Many people used to be tempted to depict the Goths as fierce warriors and conquerors, who victimized the more domesticated peoples in their way. This was both praise and the pejorative. Savage yet pure supermen, right? This was a concept popular among historians in the 1800s with Anglo-Germanic notions of superiority.

But the truth is much more nuanced. As scholarship has come to recognize in the last 100 years, the "mighty Goths" were really victims themselves. Far from conquering all, they were themselves pushed out of their territory by the Huns, and it was the Huns' might that necessitated the Goths, in turn, like dominoes, trying to pick on other tribes in their general area. Again, this is praise and pejorative. The Goths were not as savage and bloodthirsty as thought: but they were themselves forced out of their lands by a mightier force, which forced them to try to conquer.

I hope this parable can inject some subtlety into the oversimplified discourse that you see on half of these boards. You know what I'm talking about: wacky, oversimplified notions of R1b conquerors and R1b being pure Indo-European and other such nonsense.

(6) I'd like people to similarly realize that admixture models can be more complex than we give them credit to be. More on this later.

apostateimpressions said...

I would like to see more research on the origins of the British. How much comes from the Germanic tribes and how much from the Celts? I would also like to see more about the origin of the Celts in Britain. It is said that they came to Britain from the Hallstatt area in southern Germany/ Austria that became the core of ancient Celtic civilization: can that be verified? And how much, if any, is Mesolithic or pre-Celtic? How much Meso or pre-C came with Celtic or Germanic waves (if you see what I mean)? Thanks

Enjoy 2014, all eyes on interest rates and oil production.

Grey said...

My wish list is:

Funnelbeakers
Wei Valley
Hyperboreans

Grey said...

and

4) dwarves

Fanty said...

@more is better

4)
Well, from those ancient DNA tests we kind of learned another thing:

Drift/founder effects seem stronger than expected because there does not seem to be a logical connection between uniparental markers and autosomal DNA.

Countries with tons of WHG Y-DNA and decent WHG mtDNA show quiet low WHG autosomal DNA, while countries with almost zero WHG uniparental markers show extreme high levels of WHG autosomal DNA.

So, the question wether the Balkans are a source or a destination of WHG Y-DNA could aswell be: By pure random (drift) WHG males had more male children than others in the balkans while elsewhere they randomly had less. ;-)

The neolithic is already too far back that Y-DNA can have a meaning. For example, the tested WHG and EEF people lived BEFORE the MRCA of all existing I1 people for example.

Single males who lived like 1-2 thousand years after the tested persons managed to be responsable for 40% of the males of whole modern countries. How could any uniparental distribution of that time have anything to do with modern ones if something like this is possible?

Brad Foley said...

@mooreisbetter: "None other than the New York Times just reported how one reporter took the same test at three different companies and got three different results." This is an inaccurate representation of what was (at best) indifferently accurate reporting.

Therefore my wishlist is as follows

1) people learn to interpret uncertainties, such as probabilities and proportional increases in risk

2) people learn to understand what science in progress means

I know 23andMe (at least) provides only minimally useful health info (some of their other problems, such as ancestry assignment, have been thoroughly dealt with on this blog). But this is the state of the art for gene-phenotype associations. The plain truth is, we have few robust associations, 23andMe is upfront about that, but the only way we're going to get there is through larger, more detailed samples.

Hector said...

Where is the evidence for "Middle Paleolithic Siberians in Papua and Australia"? If it is about haplogroup M and P, the assertion is rather comical because you need to explain all other descendents of M526+.

Dienekes said...

Where is the evidence for "Middle Paleolithic Siberians in Papua and Australia"? If it is about haplogroup M and P, the assertion is rather comical because you need to explain all other descendents of M526+.

M and P are not Middle Paleolithic.

Hector said...

Middle Paleolithic spans up to 300000 years ago so the split between M and P as well as M526+'s is included in there.

Dienekes said...

Middle Paleolithic spans up to 300000 years ago so the split between M and P as well as M526+'s is included in there.

The split between M and P happened in the Upper, not Middle, Paleolithic.

http://dienekes.blogspot.com/2012/10/calibrated-human-y-chromosome-phylogeny.html

Hector said...

Actually I meant "up to 30000" but 0 was added accidentally. It should be remembered that these terminologies are defined in terms of stone tool industry so regional variations exist.
Anyway 41000-52000 ago in Papua NG and Australia would be generally classified as middle paleolithic.

I think you actually were referring to Denisovans and in that case there is obviously no evidence that they were Siberians as their distribution is unknown.

DocG said...

As I see it, all these years of listening too attentively to the speculations of archaeologists and paleontologists have skewered our sense of what happened in history, especially as it pertains to those of us alive today.

If we take for example the Out of Africa model, what that suggests is that a very small part of the African population of that time migrated out of Africa into Asia and that, if the model is accurate, this population is ancestral for most people now living outside of Africa. However, it tells us literally nothing about the ancestry of all the many groups both in and out of Africa whose lineages went extinct, which could in fact represent the great majority of people who ever lived on Earth.

When we examine archaic DNA or archaic bones and stones, we first of all must rest content with a miniscule sample, which in itself creates problems, but more important we can never be sure if the lineages of any of these tiny samplings actually survived to the present day. Chances are most did not.

When we examine the DNA of living people, we CAN be sure of that. That to me is a huge difference, and I think it a mistake to lump the two very different groups into one pile and treat it simply as historical or evolutionary evidence.

Dr Rob said...

@Mooreisbeter

"That is, no one can really state whether I-bearing men expanded (into Mesolithic Central Europe) from said hotspots after the Paleolithic or whether the modern concentrations can be explained by I-bearing men starting elsewhere in Europe and then moving to seize territory in more southerly climes."

If one were to speculate, I'd say the latter: from nearby "Carpathia" toward the Balkans. But there was nothing to 'seize'. they entered, or were allowed to enter, an empty landscape (it seems).

Dienekes said...

I think you actually were referring to Denisovans and in that case there is obviously no evidence that they were Siberians as their distribution is unknown.

Ok, there is no evidence that a population whose only existing sample is from Siberia was Siberian.

eurologist said...

We've seen a lot of I Y DNA in Ancient Specimens. I'd like to see a workable model as to whether the modern hotspots like Bosnia and Croatia were the source -- or a modern refuge. That is, no one can really state whether I-bearing men expanded (into Mesolithic Central Europe) from said hotspots after the Paleolithic or whether the modern concentrations can be explained by I-bearing men starting elsewhere in Europe and then moving to seize territory in more southerly climes.

moreisbetter,

It only takes a few minutes of internet search that the highly specific I-subgroups in Croatia cannot be much older than ~2,000 years. My best bet is and has been that they are part of the regional, combined East-Germanic and Slavic expansion after the fall of the Roman Empire.

AdygheChabadi said...

@Daniel Szelkey

"THe sudanese university who published this went off line, but I saw the site before it went down, and the results were pronounced valid by the university, altough they are vague and I do not trust the authencity to much.
Haplogroups A-M13 was found at high frequencies among Neolithic samples. Haplogroup F-M89 and YAP appeared to be more frequent among Meroitic, Post-Meroitic and Christian periods. Haplogroup B-M60 was not observed in the sample analyzed."


I would also love to know if those results you reported are, indeed, valid as the Meroitic language is the subject of my research right now. If those results are valid, then that would indicate that, by the Y-chromosome, the Kushites/ Meroites were not very different from the people of southern Egypt and northern Sudan today.

I also have to agree with Dienekes. Genetically, Africa is a treasure trove. There are probably genetic lineages yet to be discovered there. I think Africa has some surprises yet to be revealed. Africa can be a beast to assay.

I would, also, love to see some Admixture results on Afroasiatic Chadic speakers. I am fascinated by what it may reveal about those who carry Y-DNA R1b among them. Afroasiatic speakers from East Africa have been well assayed, many of them anyway.

Hector said...

Denisovan-like hominids admixed into Papua NG and Australians are of unknown provenance of course.
Denisovans that left remains were in Siberia. These two are not equivalent.

Dienekes said...

Denisovan-like hominids admixed into Papua NG and Australians are of unknown provenance of course.
Denisovans that left remains were in Siberia. These two are not equivalent.


This in no way conflicts with "The people best preserving the genetic legacy of ... Middle Paleolithic Siberians in Papua and Australia." which is a factually correct statement.

Locrian said...

I haven’t seen this commented upon so it may have been missed. Researchers at the University of Adelaide have claimed that Australian Aborigines left Africa some 20,000 years before the waves that became Asians and Europeans. I read about this some time ago, but couldn’t post. I’ve found this press release with the following (though this article doesn’t contain all of the claims that I’d seen in the earlier version):

““It definitely strongly supports the idea that Aborigines were an early and separate wave of human expansion Out of Africa, before the subsequent wave that established Europeans and Asians. This has long been thought to be the case, due to the very early archaeological signs of Aboriginal presence in Australia (~50 kyr) and existing genetic data, but the highly-resolved view available from a genomic sequence is a really valuable contribution. While the information is only from a single individual, it provides a powerful view of the common, shared heritage of the movement of the ancestors of modern Aboriginal populations from Africa around half the world to Australia – which is one of the most important and poorly understood stories of human history.
 However, while this is a major step forward, the key unresolved question remains the unique story of Aboriginal history within Australia, ie what has happened in those 50,000 years of life in the harsh Australian environment? Unfortunately, the information from a single individual tells us very little about this fascinating, and critically important part of human history. Aborigines are one of the oldest continuous human populations outside Africa, as they note in the paper, and due to the geographic isolation and limited archaeological records – remain one of the most mysterious chapters in human history.”

the press release is here :
http://www.smc.org.au/2011/09/rapid-reaction-aboriginal-genome-reveals-new-insights-into-early-humans-science-–-experts-respond/

Grey said...

"“It definitely strongly supports the idea that Aborigines were an early and separate wave of human expansion Out of Africa"

Wild speculation but...

I think Out of the Tropics was a much more fundamental step than Out of Africa so how could you get Out of Africa before Out of the Tropics?

Answer, if the tropical environment ever expanded far enough north due to climate for early humans to spread around the coastal bottlenecks without changing environment and then retreated south again this could leave humans in the remaining tropical zones of south and south-east asia without ever needing to adapt to a non-tropical environment.

Even wilder...

I = dwarves?

terryt said...

Thanks for the link Locrian, although it is not part of this Dienekes post. Several extracts from the link:

"Aboriginal Australians and New Guineans have long been under-represented in global surveys of human DNA"

Very, very true. In fact in admixture estimates both are grouped together while we know they have almost totally different haplogroups and so are very likely to be different autosomally. The two are also usually lumped in with Melanesians and even Polynesians who have a substantial East Asian component.

"this man was part of population that had split perhaps 75,000 to 62,000 years ago from the other groups of Homo sapiens that had also exited Africa in the initial, single dispersal event. But the authors show that the subsequent expansion into Asia most likely occurred in multiple waves, with the Aboriginal Australian being descended from the first wave to wash on to Australia’s shores. The timing of the genetic split, before 62,000 years ago, fits neatly with the initial colonisation of Australia and Papua New Guinea some time between 60,000 and 45,000 years ago, based on archaeological discoveries such as stone tools and human fossils. The existence of Denisovan DNA in the Aboriginal Australian’s genome indicates that original dispersing population of Homo sapiens must have encountered resident Denisovans en route to Australia, possibly in New Guinea".

I don't see how Denisovans could have reached New Guinea before reaching Australia as the contact with SE Asia seems to have been Timor. Although I agree that the 'expansion into Asia most likely occurred in multiple waves' I see no reason why the Denisova element may not be exactly what it seems to be: the contribution of a Central Asian ancestral population. If the ancestors of the Australian Aboriginals passed through Central Asia in the first wave of expansion from Africa they would have mixed with Denisovans there. I have long argued that mt-DNA N and Y-DNA C look far more likely to have moved east via Central Asia rather than through South Asia, and these are the two most common haplogroup branches in Australia. Subsequent climate cooling would have eliminated humans through much of Central Asia leading to the apparent complete separation of Australian Aboriginals and other Eurasians. That would explain exactly why 'It definitely strongly supports the idea that Aborigines were an early and separate wave of human expansion Out of Africa, before the subsequent wave that established Europeans and Asians'.

Simon_W said...

I used to be a stern critic of Maciamo's and Colin Welling's theories too, but now that it has become apparent that an ANE component is linked to the R-haplogroups and that it spread at a late date, and moreover that it's relatively weak in non-IE or late indoeuropeanized populations, I can only take my hat off to them and congratulate that their concept appears to be basically right. Of course it would be nice if we had some really old R1b from eastern Europe, but my guess is that it will be found. The uncertainty whether R1b was earlier in IE Celts or in non-IE Basques is resolved by the observation that Basques have much less autosomal ANE-admixture. I would suggest that Basque R1b is from IEs who became assimilated to the Basque culture. There was a time, when such an event would have been quite natural, and that was the time of the spread of the Bell Beaker culture from the Iberian peninsula to central Europe. Certainly I don't agree with Maciamo in every detail; for instance the Unetice culture seems much more linked with the Corded Ware / R1a complex, it's unlikely that it was dominated by R1b. And we can also discuss that R1b began to migrate quite early to the Carpathian basin, possibly already at the time of the Tiszapolgar culture. There is the alternative view that IE spread with Early European Farmers. It has been noted that this is hard to reconcile with the reconstructed PIE vocabulary and culture, and also with the relationships between IE subgroups who are seperated by large geographical distance, like Celtic and Italic, for instance. But the strongest counter-argument is still the large number of non-IE languages in southern Europe and the near and middle East. Alone in the area around the origin of farming we've got Hattic, Hurrian, Sumerian, Kassite, Elamite, and the three Caucasus language families, plus Sumerian which belongs to the successful families. Most of these are isolated languages; Hurrian may be related with Northeast Caucasian, and Elamite perhaps with Dravidian. Turning to southern Europe we've got Eteocypriot, Eteocretan, the language of Linear A which appears to have sounded very different from Eteocretan, we've got North Picene, the Tyrsenian family with Lemnian, Etruscan and Raetic, and finaly in Iberia there was Basque-Aquitanian, Iberian and Tartessian. So it's madness to ascribe the Indoeuropeanization of southern Europe to the spread of farming. There may be a psychological fallacy in thinking the pre-IE languages don't matter just because they are poorly attested and understood.

Simon_W said...

the three Caucasus language families, plus Sumerian which belongs to the successful families.

Oops I meant Semitic, not Sumerian.

Simon_W said...

@ apostateimpressions

I think it's instructive to play around with the Dodecad K12b Oracle in the mixed mode. But first we have to exclude the CEU30 sample. It resembles the English too much and just obscures the relationships. Then we see: The sample English_D is similar to:

"52.7% British_D + 47.3% Dutch_D" "0.3846"

The Dutch_D sample is probably the best approximation to the Anglo-Saxons that is available in K12b, the German_D sample being too southern. The British_D sample appears to have quite some Celtic ancestry, maybe from Wales?

Next we find the blend:

"49.3% Dutch_D + 50.7% Cornwall_1KG" "0.4723"

Very similar to the above finding, and Cornwall is known to be strongly Brythonic ancestry-wise.

In my experience, the British_Isles_D sample seems to be the most Germanic one from Britain, I suspect it contains some people from Yorkshire or surroundings.

In the end it might interest you what blend of Dutch_D with British_D or Cornwall_1KG approximates yourself best, and this can be calculated when you've got your autosomal raw data.

Locrian said...
This comment has been removed by the author.
Locrian said...

terry said: “Thanks for the link Locrian, although it is not part of this Dienekes post. Several extracts from the link:”

Yes I should have been clearer as to what I saw as the connection to the ancient African dna issue. My thought was that if the claim is correct that the Australian aborigines left Africa first then the best picture that we may have of what ancient Africans were like may come from them — minus the Denison component they picked up in their Volkswanderung (I recall an earlier article that suggested this took them up to Siberia and then down through China — taking them through the heart of Denisovan territory).

terryt said...

"My thought was that if the claim is correct that the Australian aborigines left Africa first"

I doubt we could say that Australian Aborigines 'left Africa first'. It appears that all we can say is that they expanded from somewhere outside Africa 'first'. Y-DNA evidence shows that F remained for quite some time in some isolated region before expanding through South Asia whereas C expanded quite quickly after the C/F split. From that I think we can gather that South Asia was not the immediate main highway usually claimed for it.

"the best picture that we may have of what ancient Africans were like may come from them — minus the Denison component they picked up in their Volkswanderung"

Possible, except that Australian Aborigines look even less 'African' than do Papuans. Papuans are mainly Y-DNA K(xLT) and mt-DNA M. Besides which it is easy to differentiate between most Papuans and most Africans. Much of the difference between any 'original' OoA population and modern Africans can probably be explained by the expansion of the West African phenotype over that last few thousand years though.

"I recall an earlier article that suggested this took them up to Siberia and then down through China — taking them through the heart of Denisovan territory".

Where did you read that? Although I have long suspected that to be the case I have never seen any paper suggesting it to be so. It certainly makes sense to me in many different ways.

apostateimpressions said...

Thank you Simon, interesting stuff. I would guess that Holland is a reasonable approximation to the Anglo-Saxons because they came here via the Frisian coast and they also settled there. Frisia would likely be the best approximation if we could get data for that region. And as you say, the AS were NW continental Germanics from Denmark and NW Germany so Holland would be a better fit than more southerly Germany.

I would guess that Wales and Cornwall are 50% Germanic. I am not sure what to suggest as a better approximation for the Celtic British. Perhaps southern Ireland would function in that role? I would guess that NI is 50% Germanic and SI about 5%. I heard that Ireland was settled through Britain, so Ireland might make sense in that respect as an approximation for the Celtic British.

Is it possible that you could calcucate the Engish as (Irish + Holland)? It would not give a definate figure for admixture proportions but it would be interesting.

Simon_W said...

@ apostateimpressions

No, the Dodecad K12b Oracle doesn't want to suggest this. But anyway, I cannot believe that Wales would be 50% Germanic, never. They are predominantly English speaking now, true, but afaik the language shift was mostly effectuated by social forces in the past few centuries, not by mass immigration of Anglo-Saxons. And this is very evident on the y-chromosomes: Judging by the maps at Eupedia the Welsh have less than 1% of R1b-U106! That's the R1b-subvariant that's typical for Germanic tribes. For comparison: In most parts of England this marker is had by 20-30% of the male population. Which is of the same magnitude as in the old Anglo-Saxon homelands in NW Germany and W Denmark. Therefore I think the Welsh are hardly Germanic and a good approximation of the Celtic Britons, and the English may be predominantly Germanic on the paternal side, and about equally Celtic one the maternal one...

apostateimpressions said...

Simon, yet most Welsh and Cornish men have the Frisian Y chromosome segment, which implies that most of their male lines are AS. Southern Ireland has much lower levels. So I doubt that the Welsh function well as an approximation to the Celtic British.

http://www.spiegel.de/international/europe/the-anglo-saxon-invasion-britain-is-more-germanic-than-it-thinks-a-768706-2.html

The AS was not a male invasion, the entire AS relocated to Britain and left their homeland empty of people. We would expect heavy AS input also on the female side.

quote:

Entire family clans set out to sea, usually in the spring and summer when the water was calm. Their ships were bulging with household goods, cows and horses. According to an old chronicle, the land of the Angles was soon "abandoned."

apostateimpressions said...

Sorry, page one of the link:

http://www.spiegel.de/international/europe/the-anglo-saxon-invasion-britain-is-more-germanic-than-it-thinks-a-768706.html

Onur said...

Simon, yet most Welsh and Cornish men have the Frisian Y chromosome segment, which implies that most of their male lines are AS. Southern Ireland has much lower levels. So I doubt that the Welsh function well as an approximation to the Celtic British.

Apostate, it is very unlikely for most Welsh or Cornish to possess Anglo-Saxon admixture. That is because their native languages were in inferior status all through their existence under the English rule. So a Welsh, Cornish, Gaelic Scot or Irish person admixing with an Anglic-speaking person would most probably switch to the Anglic language. Thus the genetic segment in Britain you associate with Anglo-Saxons should be something native to Britain and, to a lesser extent, Ireland and thus not necessarily connected to the Anglo-Saxon invasions. Maybe Anglo-Saxons increased its levels in the historically Anglic-speaking parts of Britain, but that is all.

Grey said...

There's a distinction between North and South Wales as the North is much more mountainous.

Personally i'd expect DNA from North Wales to contain lots of interesting secrets but whether those secrets were Celtic or pre-Celtic I wouldn't want to bet. I think those mountains may have been a refuge from both Celtic and Germanic invasions.

(The latter based on nothing more than climbing in north wales and noticing (or imagining) different phenotypes.)

Locrian said...

terry said "Where did you read that? Although I have long suspected that to be the case I have never seen any paper suggesting it to be so. It certainly makes sense to me in many different ways."

I have been racking my memory trying to think where I read this (in the last 6 months I’m sure) but I can’t remember. If I come across it again I will let you know.

Simon_W said...

Apostateimpressions, that Spiegel article is written in a very unscientific language, presumably to make it more readable for complete laymen. What do they mean by „Frisian y-chromosome segment“?? But I believe I've heard about this research before, they compared the y-DNA of England, Wales and Friesland and found the English to be much closer to the Friesians than to the Welsh. This is actually what I'm saying too, as far as y-DNA is concerned. The map is a bit odd, with Cornwall apparently being more „Frisian“ than most of the remaining south of England. However, if the Eupedia map of R1b-U106, aka R1b-S21 is correct, and I assume it's correct, then the Welsh can only have negligible Germanic admixture, because from that map it's obvious that R1b-U106 is strongly correlated with Germanic influence, and there are no truly Germanic admixed populations who have little or none of it. So it would be a mystery how the Welsh could have acquired Germanic admixture without acquiring R1b-U106 as well. But that said we can also see a little bit of Germanic admixture in Cornwall there (5-10% R1b-U106), and it's not clear to me what that British_D sample was composed of – clearly it cannot be purely Welsh, otherwise it would be called Welsh_D. But, it seems to be similarly Celtic as the Cornish sample. Therefore the Celtic Britons were probably a bit more Celtic than either samples. But on the other hand, the Dutch_D sample isn't identical to the Anglo-Saxons either. The Angles and Saxons were from the area where in the iron age the Jastorf culture was flourishing. The Dutch in contrast are from the area of the Harpstedt-Nienburger group and from farther southwest of it. It's not even undisputed if that latter culture was already Germanic and when it had become so. And there were similar cultural differences in more ancient periods, like in the late bronze age: the Netherlands rather belonged to the Urnfield groups than to the Nordic bronze culture, and already in the Chalcolithic the two areas belonged to different groups of the TRB. And this long history of differences can also be seen physically: The Dutch are on average very leptoprosopic (narrow faced), the Saxons (and Angles) had and have more Cromagnoid admixture. And after all, Dutch, the official language of the Netherlands, comes from the Frankish, not from the Saxons. But of course, there are also areas of Saxon and Frisian dialect in the Netherlands. It's true that the homeland of the Angles and Jutes became quite depopulated when they emigrated, that's why the remainders became assimilated by the incoming Danes. The Saxon area however remained Saxon, and the Low German dialects are still derived from Old Saxon. In short, while the Celtic Britons were a little more Celtic than the Cornwall_1KG sample, the Anglo-Saxons were a bit more eastern genetically than the Dutch_D Sample.

Simon_W said...

Btw, of interest, the progression of the English-Cornish language border:

http://commons.wikimedia.org/wiki/File:Cornish_language_shift.svg

Too late for Anglo-Saxons marching in, they merely admixed with already mixed English neighbours.

Chad Rohlfsen said...

I think that we should not forget the Belgae, when speaking of England and the Saxon input. The East Midlands, down to 2/3 across the south coast was Belgic people. At least the ruling class and culture How much the genes moved towards the continent prior to the Saxon invasion is the thing that should be sorted out first. Then we can measure the Saxon influence. We need autosomal data from those Saxon graves to know what the true effect was.

http://www.ancestraljourneys.org/celtictribes.shtml#Belgae

Chad Rohlfsen said...

Perhaps, a whole round of samples from Wales, Midlands, SE and NE England; and testing from samples of Belgae and Saxon burials on both sides of the Channel and Sea will be necessary. Probably not coming in the near future. Does anyone know if they have started autosomal testing on any of the old graves, including Sutton Hoo?

Chad Rohlfsen said...

Here's a couple links on British testing. Interesting reading.

http://www.oxfordtoday.ox.ac.uk/features/what-makes-british

http://www.wired.co.uk/news/archive/2012-07/04/genetic-mapping-britain/viewgallery/285522

Now to pick apart what is native, Belgae, Saxon and Dane.

Simon_W said...

I would bet the R1b variant typical of the Belgae was U152 aka S28, and that they didn't have much U106/S21. It would make sense if their autosomal makeup was best preserved in Belgium, especially in the Walloons - and in the Dodecad K12b analysis, the Belgian sample isn't very similar to the Dutch sample, and presumably even less to the Anglo-Saxons. In the end everything came from the continent sooner or later, and there may have been pre-Germanic links to northern Europe too...

Chad Rohlfsen said...

@ Simon_W

I'm not so sure about the Walloons. It is possible. We need old samples. The Belgae should have had a decent amount of L-21 and possibly similar I varieties found in the isles. Modern Belgium is only part of the old Belgae realm. The Belgic tribes covered everything West of the Rhine, to the present day Eastern border of Normandy. I am curious if the Belgic tribes of Britain had something to do with the 'invitation' of West Germanic tribes (old neighbors, and partial cousins if the Roman accounts are true), all grouped and labeled as Saxons, as S-21's strongest signals are in those old Belgic areas.
Yes, you're right that everything came in at some point from the continent. I think with all of this pots are people stuff coming up now, it may be safe to assume these people with hill-forts, coins, and wheeled pottery were certainly all Belgic, at least in part. Who knows how long they had been living there, maybe forgetting that they had crossed the channel 2-300 years ago or so. Hopefully some samples will be published this year and this debate about replacement vs. integration will end.

Simon_W said...

@ Chad Rohlfsen

Interesting links, thank you!
I agree that the similarity of the southeastern cluster with Belgium must be due to the Belgae, not to the Anglo-Saxons and Vikings as the texts suggest - the latter influences are rather reflected in the observed similarity with Germany and Denmark. Moreover the Danish influence proper was strongest in northeastern England, so this may have caused the northeast to form its own cluster.

Simon_W said...

On the other hand it doesn't overlap well with the distribution of Danish place names, so presumably it's more a matter of relative isolation.

apostateimpressions said...

Bodmer has surmised the same as everyone else. He should give some figures if he has analysed the data. How much of British ancestry is Germanic? Give me figures for each region. When is he going to open up the data to all other researchers? Or is it a secret? Likely the British establishment does not consider it PC.

Why has no researcher examined the autosomal data from AS burials. Is science quite dead in Britain?

Interesting, were the Belgae Celtic or Germanic?

http://en.wikipedia.org/wiki/Belgae

Chad Rohlfsen said...

Did anyone else hear the rumor that La Brana 1's yDNA is C-V20? Very interesting, if true. Supposedly the paper comes out in 10 days.

Simon_W said...

But while we're at it, it would be equally interesting to have some autosomal a-DNA from Elbe Germanic tribes and from Celts and Gallo-Romans from southwestern Germany and Switzerland. Then we could calculate how Germanic the Alemannic speaking dialect group in southwestern Germany, Switzerland and the Alsace region really is. The problem is very similar to the one in England: A Romanized Celtic population experiences Germanic influx after the Romans left and shifts the language to a Germanic one. From my trials with the Dodecad oracle I can say that if we take the German_D sample as an approximation to the Germanics, then, surprisingly, the Gallo-Romans in southwestern Germany and Switzerland appear to have been North Italian-like. Which makes me think of a thick Raetic substrate under the dominant Celtic layer.

But leaving such ethnocentric issues asside, I would love to see more ancient y- and autosomal DNA from all of Western Eurasia and northern Africa, and from the Mesolithic to the Iron age. Especially we need more from eastern and southeastern Europe and from the northwest. And while the hunter-gatherer – early farmer encounter has got quite some treatment by now, we still need to learn much more about the dramatic changes during the chalcolithic and bronze age in particular.

Chad Rohlfsen said...

@Simon_W

I would really be interested in Alsatian and Pomeranian German samples, as that is about 1/4 of my ancestry. I wonder if they are ever going to test some of these people. Didn't they test a tooth of the Armsbury Archer, or someone like that, and find that his enamel showed he grew up around central Europe (Austria)?

@apostateimpressions

I believe the Belgae may be an intermediary type. The Romans called them Gallo-Germanic and it seems a few of the tribes may have spoken more of a Germanic language. However, Germanic and Celtic R1b have the same parent, so how different can they really be? Romans couldn't tell Northern Europeans apart except for clothes and language.

Simon_W said...

Chad Rohlfsen, yeah you're right, northeastern France almost up to Paris (and in fact including the Upper Normandy) and parts of the Netherlands and Germany left of the Rhine were Belgic too, so at least we should add the northeastern French to the candidates of the most Belgic descendants, since the Germanic speakers are likely to be somewhat more Germanic admixed.

And true, they must have had some L21. Even in Bologna there is 10,3% L21, possibly the legacy of the Boii, who probably originated in southern Germany. But the majority of the British L21 appears to have been on the Isles for longer, I surmise it's from the earliest q-Celtic wave which may well have arrived in the middle Bronze age.

An important question, to which also Apostatesimpressions alludes to, is whether the Belgae were in any way Germanic or rather straightforwardly Celtic. The link you provided suggests the latter to be true, on the basis of recorded place names and personal names. The association with the Germanics is explained as merely refering to the Belgae's roots in the country east of the Rhine, what Caesar called Germania.

Judging from the Eupedia maps, S21 peaks in the Netherlands and eastern Friesland, both Germanic speaking areas. It correlates with known Germanic influence very well, except for the relatively high incidence in Austria – I guess the ancient Norici, whose language isn't known well, may originally have been closer to Germanics than to Celts, an idea the Eurologist often expressed too. In contrast, the Romance speaking parts of the old Belgic realm are more U152/S28 than S21, although S21 is quite strong towards its eastern fringe, but we must not forget that there was also a real Frankish invasion and settlement. After all the Franks even gave their name to France.

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

Already Cavalli-Sforza showed with the „classical“ serological markers, that the English are close to the Dutch and Danes, the Belgians close to the Germans, Swiss and Austrians, and the Scottish close to the Irish, but quite remote from the others.

http://imagizer.imageshack.us/v2/800x600q90/534/rc5k.png

Simon_W said...

Chad Rohlfsen, I'm not sure how your wish relates to ancient DNA. The Alsatians are still alive and well, although their German dialect is strongly endangered by French speech. I agree that it would be interesting to compare their DNA to ancient Elbe Germanic tribes and to Gallo-Romans from the Alsace region. The German Pomeranians who are still with us, however are an old generation. I doubt that there is going to be any study on them as long as some are still alive. But at least in the case of the uniparental markers we're in the lucky situation that it can be studied even in mixed descendants – as long as they correctly know the origin of the respective lineages. Of course it would be interesting to compare their DNA to northwestern Germans (northern Low Saxons and Westphalians being the main source of their German ancestry) and to Pomeranian Slavs who no longer exist as an ethnic group, except for the Kashubians on the lower Vistula. As for the tooth enamel sample you alluded to, this was about the isotopes, not the DNA; yes IIRC they showed he was from central Europe.

Chad Rohlfsen said...

In reference to the tooth of the archer; yes, I know it was isotopes. Another interesting deal with isotopes that I saw recently involved the Indo Europeans changing from heavy fish and wild game eaters, to finally eating their cattle and sheep regularly instead of just at special or religious occasions. This happened about the time they travelled down the Black Sea coast into the Balkans. Very interesting.

Grey said...

I think the Belgae were Celts from the Celtic / Germanic frontier who migrated to a more peaceful region for a quieter life hence being rougher and tougher than the people around them.

Chad Rohlfsen said...

Just a few things again with the links to Scandinavia among the British. In the Mesolithic some of the Motala people settled in Britain. This is attested for archeologically. I am trying to find the link regarding the dig sites. Also in Kent, there was a dig recently in Bronze and Iron Age graves. Of the 25 samples, 9 were local (probably a Loschbourg/Motala mix), 8 from Southern Norway or Sweden (2 Bronze, 6 iron), and 5 were from the western Mediterranean(4 Bronze and 1 Iron). This is all according to the isotopes. I am trying to find the best link for this as well. So obviously Britain has always been connected with the future Germanic people. It is no wonder in the new plot with ANE,WHG,EEF that Britain is so close to Scandinavia. Lots of shared ancestry to the Mesolithic-the Iron Age. All pre-Saxon, likely.

http://www.dailymail.co.uk/sciencetech/article-2336094/Bronze-Age-burial-site-3-000-years-ago-contain-Scandinavian-western-European-migrants.html