March 07, 2009

Genetic structure in northern Europe revisited

The results of the STRUCTURE analysis are quite interesting. When Finland is included (B), it is the first one to be separated from other Northern Europeans, confirming previous results. Sweden, and to a lesser degree Denmark seems to possess some admixture with the Finnish-related (red) element. The next split (blue vs. green) differentiates continental Germnics (esp. Scandinavians, and somewhat less Dutch) from Irish-British and Australians of largely British ancestry.

When Finland is not included (C) and for K=4, it becomes obvious that the UK population includes three components ("Dutch" green, "Scandinavian" red, and "British Isles" blue).
See also a previous study on the topic.

Genome Research doi:10.1101/gr.083394.108

Geographical structure and differential natural selection amongst North European populations

Brian P McEvoy et al.

Abstract

Population structure can provide novel insight into the human past and recognizing and correcting for such stratification is a practical concern in gene mapping by many association methodologies. We investigate these patterns, primarily through principal component (PC) analysis of whole genome SNP polymorphism, in 2099 individuals from populations of Northern European origin (Ireland, UK, Netherlands, Denmark, Sweden, Finland, Australia and HapMap European-American). The major trends (PC1 and PC2) demonstrate an ability to detect geographic substructure, even over a small area like the British Isles, and this information can then be applied to finely dissect the ancestry of the European-Australian and -American samples. They simultaneously point to the importance of considering population stratification in what might be considered a small homogenous region. There is evidence from FST based analysis of genic and non-genic SNPs that differential positive selection has operated across these populations despite their short divergence time and relatively similar geographic and environmental range. The pressure appears to have been focused on genes involved in immunity, perhaps reflecting response to infectious disease epidemic. Such an event may explain a striking selective sweep centered on the rs2508049-G allele, close to HLA-G gene on chromosome 6. Evidence of the sweep extends over 8Mb/3.5cM region. Overall the results illustrate the power of dense genotype and sample data to explore regional population variation, the events that have crafted it and their implications in both explaining disease prevalence and mapping these genes by association

Link

101 comments:

mikej2 said...

It is very difficult to see the truth behind pictures in the text "Genetic structure in northern Europe revisited", because I have no idea what exactly means Finns and other groups in this point of view. How large are the genetic finger prints used to indentify each group.

If it has been done by a small biggest common denominator for each group, so how can be said that it is NOT only one insignificant part of genome that in the past existed more widely than today and was somewhere reduced by new populations. So we cannot say anything about directions of genome flow, we can just list some similarities.

I cannot also understand how a little block in genome affects so much to Finns, but this kind of studies rarely show comparable identity blocks for example for Danes or Swedes and compare them to Finns and Estonians, although we know that much more Swedes have during past centuries moved to Finland than Finns to Sweden or Estonians to Denmark. This is quite contradictory.

Just

Maju said...

I cannot also understand how a little block in genome affects so much to Finns...

It's simple: Finns have a different origin from Scandinavians (for the most part), even if both are European. Finland was colonized originally from NE Europe, while Scandinavia was from Central Europe. Additionally Finns also display a smaller "Siberian" component, that is rare further west.

sardiniankid said...

what does the pink coloring in the graphs mean?

Polak said...

>>what does the pink coloring in the graphs mean?

Gay factor.

pconroy said...

Dienekes:

When Finland is not included (C) and for K=4, it becomes obvious that the UK population includes three components ("Dutch" green, "Scandinavian" red, and "British Isles" blue).

This should read:
...and "Irish" blue...
as that's where most blue is at K=4

Dienekes said...

This should read:
...and "Irish" blue...


The blue is almost non-existent in non-British origin populations, so "British Isles" is appropriate. Ireland probably has more of this element because it did not receive significant continental Germanic elements compared to Great Britain.

eurologist said...

So then, is there any consensus that the "British blue" component may relate to a people with more maritime sustenance, that worked their way up the coast from Northern Iberia/ Southwest France after glacial maximum? I would imagine there would have been enough cultural differences between these and and the upper Rhone/ pre-maximum Rhine/Danube river hunters to maintain some differentiation.

Also, the reforming bodies water East of (the then much farther West) mouth of the Rhine would have been unattractive to such people, who likely were used to the rich fishing and shell fish habitats of the rocky Atlantic coastline. Thus, they would have preferred crossing the "Rhine" north to the British isles, instead of trying to edge out a living along the shallow and sandy waterways of the slowly reestablishing North Sea.

And by the time Norway's coastline was inhabitable, the inland folks had already made it, there.

Maju said...

What blue component? Admittedly the colors are kinda messy (what is one color at one depth is another in the next row) but, anyhow, looking at both graphs, the blue component at K=5 is always "Finno-Scandian" (two diferent components in fact, depending on which graph).

The "Irish" dominant component at K=5 is red, and it's not too marked in Britain (though more than in the rest).

So then, is there any consensus that the "British blue" component may relate to a people with more maritime sustenance, that worked their way up the coast from Northern Iberia/ Southwest France after glacial maximum?

Impossible to say. We lack comparison with SW or Central Europe. From what's been found in other studies it could be from either origin or even the marker of islander relative isolation since Epipaleolithic.

The Yellow, Magenta and Green components (always at K=5, graph C) appear also to be of Western distribution (Scandinavians have much less). Instead the blue component is dominant in Northern Europe strictu sensu (Scandinavia, Netherlands and presumably Low Germany). The red marker seems to be especifically Irish. Though maybe is more generically ancestral British (hard to say when the UK sample is from London only), it could also represent the Irish legacy in urban England.

Maju said...

Erratum: Sorry, was mixing pears and oranges in the last sentence. Was thinking in another pan-European autosomal study commented by Dienekes recently. I have no idea if the UK sample is from London only though my doubt still stands.

Maju said...

Another thing.

Eurologist said:

I would imagine there would have been enough cultural differences between these and and the upper Rhone/ pre-maximum Rhine/Danube river hunters to maintain some differentiation.

Also, the reforming bodies water East of (the then much farther West) mouth of the Rhine would have been unattractive to such people, who likely were used to the rich fishing and shell fish habitats of the rocky Atlantic coastline. Thus, they would have preferred crossing the "Rhine" north to the British isles, instead of trying to edge out a living along the shallow and sandy waterways of the slowly reestablishing North Sea.


From other studies (Bauchet 2007) the presence of Iberian or Basque autosomal DNA (two different components) is detectable in Britain and other parts of Europe but always as minority component. The main component in the British islands is shared with Central Europe (i.e. Rhin-Danub area).

There's no particular archaological indication that Britain was populated from SW Europe in the Epipaleolithic. Rather Mid-Western Europe (Rhin-Seine area) is probably the main source, as well as that of Scandinavian Epipaleolithic colonization.

Nevertheless, SW European genes may have flowed along the atlantic facade at later periods, notably in the Megalithic Age (Chalcolithic mostly, though know in Britain as late Neolithic and early Bronze). There's been speculation that cod fishermen could have been vectors of this cultural phenomenon. Cod was also an element of linkage between the Basque Coutry, Brittany and Ireland (the main Basque fishery) in the late Middle and along the Modern Age. The flow therefore may have been going through the ages, not just limited to some specific period.

eurologist said...

Possibly later - but most of Europe, in particular North-East France, Belgium, The Netherlands, Scandinavia except Finland, and much of Austria and Germany don't show signatures of significant late colonizations - except very few, well-documented events and admixtures. Agriculturalist were very successful and easily dominated other people at the margins, due to their economic and reproductive success. There are documented interactions between alpine residents and Stonehenge, and even earlier, British settlements from the earlier mainland waves (3,000 to 5,000 BC) of agriculturalists. Also, there are later historically documented cases of South Rhine-area heads of Germanic peoples possessing large amounts of property of South England around 0 BC. All of these would have dominated residing tribes just by sheer numbers and economic strength.

Yet, an earlier signature remains - and I believe, that one must be before ~5,000 to 6000 BC, and must be tied to people who in addition to fishing later adopted mountain shepherding over pig/cow agriculture to edge out a living at the margins when mainstream (but flatland) agriculture arrived.

pconroy said...

Dienekes,

I've seen studies that Australians are from 25-40% Irish, and in this study seem almost exactly the same as the UK - that would lead me to think that the UK must have a similar percentage of Irish descent. Some UK cities have large Irish populations, offhand I'd estimate Liverpool 80%, Glasgow 50%, London 25%

So, at K=4, the UK sample seem to composed of Irish, Dutch and Scandinavian components.


Maju,

The "Irish" dominant component at K=5 is red, and it's not too marked in Britain (though more than in the rest).

Dienekes and I were talking about K=4, that's why the Irish component is blue, but at k=5 it's red.

But in either case your point holds.

Cheers,
Paul

pconroy said...

Also the CEU sample in C at K=5 seems like a UK sample with extra Scandinavian input

Dienekes said...

I've seen studies that Australians are from 25-40% Irish, and in this study seem almost exactly the same as the UK - that would lead me to think that the UK must have a similar percentage of Irish descent. Some UK cities have large Irish populations, offhand I'd estimate Liverpool 80%, Glasgow 50%, London 25%

If it represented "Irish" admixture, then we would expect the individual percentages to be quite variable. On the contrary, it is present in almost all UK individuals, but in (almost ?) none of them above 50%, which suggests that it represents an ancient component rather than the result of admixture with the Irish.

pconroy said...

Dienekes,

Well there were many Irish colonies in the UK, the most well known being Western Scotland - Dal Riada - but also in Galloway - Gall Gael - North Wales, South Wales and Cornwall. Neil of the Nine Hostages is supposed to have had a colony among the Saxons also.

In either case the Irish are known as having a propensity to admix easily with their host population, so that in Australia, after 200 years, they are so mixed, that the average Australian shows up similar to the UK.

Look at the Irish Soccer team - almost all players were born in England, and of these, almost none have 4 grandparents who are Irish.

Maju said...

Eurologist: I mean Epipalolithic colonists from Middle Europe (Rhin-Danube province). That area later did suffer other demic impacts (Neolithic, Indo-Europeans...) that the islands were largely spared of. When these arrived, they did after long periods of filtering in the mainland.

There are documented interactions between alpine residents and Stonehenge...

While not denying them, I must say these are hyped mediatically (and certain style of spectacular "prehistory" favored in Britain doesn't help at all). The Amesbury archer is just one of the early carriers of the Bell Beaker cultural (and probably mercantile) phenomenon, which is always minoritary in Atlantic Europe (inserted in mainstream older traditions, normally Megalithic) and whose followers were in most cases of local type. The phenomenon did begin in just Indoeuropeanized Central Europe (Bohemia most likely) but soon became a cosmopolitan one. In the second phase the main center was in Portugal, which was the only really urbanized area in all Atlantic Europe and also the likely center of typical Megalithism (Dolmenism) itself.

...and even earlier, British settlements from the earlier mainland waves (3,000 to 5,000 BC) of agriculturalists.

I have never read anywhere that British Neolithic (Windmill Hill culture) relates specifically to Danubian Neolithic. Of course there was an influence but in general Atlantic Neolithic appears as a series of local evolutions - and maybe climatic constraints (high humidity) were the reason that the typical Neolithic cultural waves (Cardial and Danubian) did never really arrive as such to these areas.

Also, there are later historically documented cases of South Rhine-area heads of Germanic peoples possessing large amounts of property of South England around 0 BC. All of these would have dominated residing tribes just by sheer numbers and economic strength.

Not sure if you're right in this but remember that Roman intervention in the Celtic areas was mostly driven by their interest in containing the then ongoing Germanic expansion, which was succesfully stopped for centuries. So I doubt that whatever influence the Germanic warlords could have in Britain c. 1 CE (there's no "year zero") would survive by c. 50 CE.

Yet, an earlier signature remains - and I believe, that one must be before ~5,000 to 6000 BC, and must be tied to people who in addition to fishing later adopted mountain shepherding over pig/cow agriculture to edge out a living at the margins when mainstream (but flatland) agriculture arrived.

IMO the earlier signature remains not just in the islands but in the continent as well. In the islands it is surely stronger though because of relative isolation.

...

PConroy

Also the CEU sample in C at K=5 seems like a UK sample with extra Scandinavian input

What is exactly what one would expect from Utahns according to self-reported ancestry in US census: mostly English but also Danish (and other British).

Average Joe said...

PConroy:

By "Irish blue" I think you mean "Celtic blue" since most people in Britian are probably not of Irish ancestry. The results seem to indicate that there was a significant survivial of the Celtic population after the Anglo-Saxon invasion though this may have been a largely female component.

Idler said...

Have there been any studies correlating facial types or other anthopometric measures to the genetic material? I ask as a layman interested in the ground covered in popular essays and books by the likes of Stephen Oppenheimer and Brian Sykes.

I was interested originally by Oppenheimer's claim that no immigrant group had added more than 5% of the genetic material to the base population (which he said was related to the Basques). This struck a chord with me because it seemed from my personal observations that, while there were different facial types concentrated in various parts of Britain, one can find certain very similar types all over the island, such that you really couldn't tell whether someone had English, Welsh, Scottish or Irish ancestors.

I've come across a web page that looks at different "nordish" facial types but it seems to be somewhat fanciful and aimed at glorifying one race over another.

Clearly, one can recognize certain faces to be characteristic of, say, Eastern England, and the population west of Glasgow, for example, is more homogeneous than that of the area around Edinburgh. There are certainly recognizable stereotypically "Irish" faces, but they don't exhaust the full menu of faces one will find there. I know I don't have one, despite the fact that seven of my great grandparents had Irish surnames and at least six were born in Ireland.

Maju said...

The results seem to indicate that there was a significant survivial of the Celtic population after the Anglo-Saxon invasion though this may have been a largely female component.

Really? And how do you explain the hyper-high levels of R1b (pre-Celtic rather than Celtic for the most part but anyhow) in all Britain? C'mon!

Average Joe said...

Maju:

"Celtic" is the term generally used for the pre-Anglo-Saxon population of Britain and Ireland. If you would rather use "pre-Celtic" then that is your business.

Maju said...

Yes, before the Anglosaxon British and Irish were Celts, at least for the most part. But before the Celts (arrived c. 300 BCE from France and Belgium, and to these areas from Central Europe in earlier times) they were something else.

Obviously there was people living and thriving in the islands before 300 BCE and they probably constitute the vast majority of the ancestors of modern British and Irish. The proper name is pre-Celtic, for lack of better ones.

So it's not "my business". It's a matter of correct description of prehistorical facts.

Onur said...

The proper name is pre-Celtic, for lack of better ones.

So it's not "my business". It's a matter of correct description of prehistorical facts.


Then we should apply the same rule to the other populations: pre-Dutch, pre-Scandinavian, etc. What a nonsense!

eurologist said...

I agree with Maju here, for clarity.

On the other hand, the Celtic culture IMO included many different peoples and languages.

When agriculture first arrived on the isles, the language these people spoke likely was just an early form of Indo-European - they predated even Proto-Celts by more than two millenia. Yet, these people with their agricultural success may have wiped out most earlier language(s), even if they did not eventually dominate the population. Subsequent waves of settlers may have both spoken Celtic and Germanic dialects. The existence of earlier Indo-European may have given the insular Celt its particular flavor - not the fact that it was separated from e.g. the Celtic of cisalpine Gallia.

Also, quite a number of people who study old English come to the conclusion that there was a native Germanic substrate before Anglo-Saxon invasion.

Similarly, I believe that much of Central Europe was Germanic-speaking during "Celtic" times. Of course, there is little evidence either way - to me it is just the simpler explanation of historical facts, and does not require the mysterious disappearance of Celts in Middle and Southern Germany, and in Austria.

In Central Europe, place and river names have both origins, but if they are truly old, what now appears to be Celtic-related may just be very old Germanic. There are many such words in Germanic languages that you could either view as Celtic loan words, or for which you could postulate a common Indo-European origin, instead. I distrust such studies; for many of the place names with so-called Celtic origin, you can find analog Greek words. Is Celtic more closely related to Greek than to Germanic? Of course not, the answer is, they are all Indo-European.

This includes the prime example of the purportedly Celtic "Hal" of Hallstadt. The Greek ἅλς gave us halogen and halides. Perhaps the trading Alpine residents simply kept a form that was more easily mutually understandable between them and the Greeks, and other tribes. That doesn't mean everyone spoke a Celtic language.

Maju said...

Then we should apply the same rule to the other populations: pre-Dutch, pre-Scandinavian, etc. What a nonsense!

Scandinavian is a geographic denomination, not ethnic. Never heard of pre-Dutch before (after all they are a branch of Middle Germans, we know the name of heir predecessors). But terms like pre-Indoeuropean, pre-Semitic, pre-Celtic, pre-Greek do exist and are applied.

Celtic implies speaking Celtic languages, the people living in the islands in the times of Stonehenge did with all likehood not spoke Celtic nor had Celtic culture, even if we don't know which was their acual languages or ethnonymes.

I agree with Maju here, for clarity.

Thanks.

When agriculture first arrived on the isles, the language these people spoke likely was just an early form of Indo-European

Most probably not. Most probably agriculturalists never really arrived in meaningful numbers, rather locals were seduced by such economical practices, gradually, possibly with some minor demic input initially.

Also with all likehood Central European Neolithics did not speak Indoeuropean. Some have suggested Semitic or other Afroasiatic languages, others Basque-related languages, or "Pelasgian", etc. But the really wise ones suggest nothing: the evidence is lost.

And Indoeuropean is quite clearly derived from an steppary core (Samara valley, Russia) and expanded at a later date (Chalcolithic for Europe east of the Rhin, even later in Western Europe and South Asia). I have discussed this issue a zillion times and the Kurgan model always wins, while Renfrew's fantasies are shown as what they are: baseless romanticism.

...and does not require the mysterious disappearance of Celts in Middle and Southern Germany, and in Austria.

They never disappeared: they became Germans and Romans. There is no "mystery". What the heck?!

There is a well known process of expansion of Germanic tribes from Northern Europe "sensu stricto" southwards. This is well attested both archaeologically and historically. The reason for this push, parallelled by the Roman expansion from the south and the Dacian one from the East, into Celtic territory was that Celts had gone semicivilized in the La Tène period, having too much concentration of the economy and power structures in their trading towns (often known as "oppidae"). The destruction of such cities simply meant that the whole economic and social network of Celts could be destroyed with them, something that Germanics apparently enjoyed doing (lots to plunder).

When Caesar intervened in Transalpine Gaul, he did moved by the troubles that Germanics were causing. Germanics had conquered the Helvetians, which in turn had decided to mass migrate and attack Roman allies in the region. Romans could watch from the fence or intervene: they decided the latter and put a limit to Germanic expansion for the next several centuries. They also assimilated the conquered Celts gradually. Similar geostategical reasoing surely brought them to Rhaetia, Noricum, Panonnia and Britannia soon after: securing their borders and interests against the unruly tribes of the north, very especially the expansive Germanics. They even attempted to conquer Germania Magna itself.

Is Celtic more closely related to Greek than to Germanic? Of course not...

Probably yes (even if I share your escepticism re. lexical comparison). When Idoeuropeans migrated out of Eastern Europe in the late 4th milennium BCE, they did it in two distinct "waves": one went to the Balcans and another to Eastern Germany (later expanded to Poland, which would in fact become the main center). Greek may well derive from the Balcanic branch, while the rest of European IE languages (excepting Albanian surely, as well as Iranian languages like Ossetian) seem to derive from the Central European branch.

Greek is Indoeuropean, arguably Western Indoeuropean too but certainly from a different sub-origin than most other European languages, which derive from a common Central European core (Baalberge-Lubon-Globular Amphorae-Corded Ware). The whole genesis is not fuly understood but this difference between the Central European and Balcanic groups seems most likely IMO.

Onur said...
This comment has been removed by the author.
onur said...

Scandinavian is a geographic denomination, not ethnic. Never heard of pre-Dutch before (after all they are a branch of Middle Germans, we know the name of heir predecessors). But terms like pre-Indoeuropean, pre-Semitic, pre-Celtic, pre-Greek do exist and are applied.

Celtic implies speaking Celtic languages, the people living in the islands in the times of Stonehenge did with all likehood not spoke Celtic nor had Celtic culture, even if we don't know which was their acual languages or ethnonymes.


There is no consistency in your definitons. There are very few places in the world that have been dominated in time by only one language group. The Netherlands was once a non-Indo-European speaking territory (at least the parts above sea level :D), then why not call its population pre-Dutch according to your logic?

eurologist said...

Maju,

I truly appreciate your comments and insights, but I happen to disagree strongly on the origins of Indoeuropean. I am posited about half-way between the paleolithic continuation folks and the idea of agricultural spread of the language(s). Meaning, that I think the then modern version of the language spread about ~6,000BCE from an Eastern Danubian origin, because
(i) it was the then lingua franca of central Europe
(ii) it still resonated with many post glacial maximum folks

From there, the language spread in all directions, at first easily West, because within a thousand miles, it really was not all that different from what was spoken, already. Agriculture quickly made it even more of a uniform language, due to trade, knowledge exchange, and sheer number domination. All long before there was such a thing as a steppe culture - which slowly developed out of former agriculturalists, and because of that, shared the same origin, but developed sufficient differences when moving East to become a second branch.

Western IE is so similar to each other because it was earlier and in close proximity.

That is, early IE was spoken in some form even just before the arrival of agriculture, and its mainstream form was well-established and predominant by about 4000BCE in central Europe. Which is the only way, IMO, to explain its wide success, and its West-East linguistic structure.

Maju said...

The Netherlands was once a non-Indo-European speaking territory (at least the parts above sea level :D), then why not call its population pre-Dutch according to your logic?

"Pre-Indoeuropean" is more probably the term you will find for what you say. Pre-Dutch sounds like whatever existed before the Netherlands was founded or at least before the Dutch ethnic identity arose.

But whatever. If you think that pre-Dutch is a convenient term for whatever pre-/historical reality, I am not going to make an isue of it.

What I dispute is to call "Celtic" to peoples who were with all likehood non-Celtic and non-Indoeuropean. Britain was not Celtic before the Celts, with Celtic language(s) and other cultural elements, arrived.

...

I truly appreciate your comments and insights, but I happen to disagree strongly on the origins of Indoeuropean. I am posited about half-way between the paleolithic continuation folks and the idea of agricultural spread of the language(s).

Virtually all linguistic analysis of PIE (the best studied linguistic family ever) makes it about 6000 years old. PIE has no common words for agriculture but it does have some common roots for horse, apple, cold climate vegetation, etc. all of which are coincident with the Kurgan homeland of the Eurasian steppes and don't fit with Anatolia or any other Mediterranean area of origin. Paleolithic origins would be even more far-fetched: we would not be able to recognize such a clear family after such a long divergence. Additionally you have the closest family in Uralic (Indo-Uralic having been proposed as superfamily, while others argue for just sprachbund at their neighbouring origins between the Volga and Ural mountains).

For details on the linguistic reasons for PIE being steppary, check this post at Mathilda's blog. She was also seduced by Renfrew's ideas but she eventually realized it could not be right.

Additionally the Kurgan model is very solid arcaheologically, at least in Europe. Anyone with decent knowledge of European post-Neolithic archaeology can trace the IE migations, influences, counter-influences... with great precission.

Western IE is so similar to each other because it was earlier and in close proximity.

Not earlier: the evolution of all IE branches is necesarily equally old. Additionally, most trees of IE show western and eastern IE branches in relative proximity, with other older branches like Tocharian or Hittite, and sometimes also Greek, Armenian and Albanian, spawning much closer to the ancestral root.

Western IE (if we include Greek, which sometimes is instead derived from close to the IE root, out of the west-east main branches) may have been formed in Seredny-Stog II, when IE entered in contact with Dniepr-Don peoples, just before the first migrations out of Eastern Europe. If we exclude Greek, it was without almost any doubt formed in Central Europe, in the series of cultures that link Baalberge to Corded Ware, through a whole milennium, at times under heavy Danubian influence.

I have some doubts on wether Germanic began to form already in the Single Burials culture (Nordic offshot of Corded Ware) or is rather a later arrival from Central Europe, maybe with the strong influences from Urnfields culture that radiated in all directions. In any case, it is directly related to the Central European branch of IE, both by origin and by neighbourhood influences (sprachbund). Hence intense connections with Celtic, Italic or Balto-Slavic should surprise no one.

onur said...

What I dispute is to call "Celtic" to peoples who were with all likehood non-Celtic and non-Indoeuropean. Britain was not Celtic before the Celts, with Celtic language(s) and other cultural elements, arrived.

Then why don't you also dispute calling Dutch to people who weren't obviously Dutch before the arrival of the Germanics? This is utter double standard! You are making injustice to the Celtic people and heritage of Britain. If all of Britain spoke Celtic languages now, would you make the same statement? I don't think so.

I advise you to be fairer in such sensitive issues.

terryt said...

"Then why don't you also dispute calling Dutch to people who weren't obviously Dutch before the arrival of the Germanics?"

I think you're confusing the people with the language. There were probably people living in Holland before any Indo-European language came in, but, as Maju pointed out before, the language they spoke is a mystery. Similarly in Britain. The people may have been at least partly genetically connected to those in Holland but again we have no way of knowing what language they spoke. The term "Celtic" is really just a name given to western Indo-European languages. Individually they were probably no more closely related to each other than any of them are related to the German group. What we call Celtic and German were simply part of a dialect chain stretching across most of Europe. Latin was also part of the chain.

onur said...

I think you're confusing the people with the language.

No, I have taken great care not to confuse the two.

The term "Celtic" is really just a name given to western Indo-European languages. Individually they were probably no more closely related to each other than any of them are related to the German group.

Celtic is an established branch of the IE family, well separated from Italic and Germanic. And your definition of Celtic languages according to just geography is a misconception; Celtic languages were once spoken in a far greater area (as far east as Anatolia).

Maju said...

Then why don't you also dispute calling Dutch to people who weren't obviously Dutch before the arrival of the Germanics?

I would dispute that too. Nobody has yet made such claim though, at least not in this thread.

Celtic is an established branch of the IE family, well separated from Italic and Germanic. And your definition of Celtic languages according to just geography is a misconception; Celtic languages were once spoken in a far greater area (as far east as Anatolia).

I agree with this. There are some cases that are dubious (Lusitanian, Pictish) but in general Celtic is a clear subfamily of Indoeuropean, just like Germanic, Slavic or Indo-Aryan.

But this is just even more reason not to call Celtic to a people who did not speak Celtic nor Indeouropean and who did not participate of the Celtic culture.

Exactly the same you do not call Latin to historical Iberians, or you do not call Germanic or Anglosaxon to the Welsh, or you do not call Slavic to the Tatars, or you do not call Semitic to ancient Egyptians. It would be just wrong.

onur said...

But this is just even more reason not to call Celtic to a people who did not speak Celtic nor Indeouropean and who did not participate of the Celtic culture.

Exactly the same you do not call Latin to historical Iberians, or you do not call Germanic or Anglosaxon to the Welsh, or you do not call Slavic to the Tatars, or you do not call Semitic to ancient Egyptians. It would be just wrong.


The problem with your pre-x model of definition is that we don't know and can never know (especially in western Europe) how much of the lineages of an x population is descended from the pre-x and the "original" x population.

Returning to our Celtic example, calling contemporary Celtic populations pre-Celtic implies that they have virtually no "original" Celtic blood in them. Even if this is true, we can never know that in this genetic chaos. So it is most wise to define populations as they are defined today.

McG said...

some random thoughts. the Finss have a strong eastern(hun) contribution. certainly from the east. My wife is 100% Finn, blood is Bminus and she has a rare form ofanemia. Two of the scottish tribes, scottis and laiginn, are of Dutch ancestry, driven from france by the Romans. I also believe the Irish gaelsic is an Iberian latecomer. Dienekes comments on facial appearance are appropriate. The scottis have the typical dutch square face. the older race is oval faced. JMHO:

eurologist said...

Maju,

Thanks for the links, I will give this some more thought. Again, nothing personal, I just need something more convincing than what I have read, to date.

The Kurgan hypothesis just sounds so improbably, to me. Most people think there is sufficient evidence to conclude that the (pre-Kurgan) central-European agricultural revolution did not replace the original population. Same is then even more true about any semi-nomadic steppe people entering Central Europe (CE). Why would these people, who by sustenance and also by today's genetic evidence clearly could not outnumber local agriculturalists, and whose methods of transportation initially were horribly ill-adapted to the muddy, soft, wet and densely forested CE, provide these people with a new language? Why would such people have a superior language to that of a very entrepreneurial new local culture, in the first place?

I don't deny Kurgan expansion - I just think that these same people originally derived from or were heavily influenced by the adjacent (west and southwestern) agriculturalists who already spoke an early form of IE. Kurgans then just would be (partially) responsible for spreading the Satem shift.

Replacing almost all prior languages in Europe is a tough thought even if it took place during the advent of agriculture (thus my thought that it was helped by being seeded on a receptive stratum - in a well-populated region that already happened to speak similar dialects) - but without that? Unimaginable.

Giving just Kurgans an already existing IE is a much simpler explanation, IMO.

Again, I agree I need to read more!

Maju said...

The problem with your pre-x model of definition is that we don't know and can never know (especially in western Europe) how much of the lineages of an x population is descended from the pre-x and the "original" x population.

It doesn't matter at all, we define them as it's always been done naturally: by cultural identity and language.

Returning to our Celtic example, calling contemporary Celtic populations pre-Celtic implies that they have virtually no "original" Celtic blood in them.

I'm not calling anyone alive anything. What I say is that the substrate of the islands and the main share of the ancestry is pre-Celtic.

It is clear that there was people in the islands before the Celts and that these cannot be called "Celtic" with any consistence: it would just confuse all. The identity of modern islanders is rather independent from their biological ancestry: it's cultural.

Anyhow, there's no particular reason to think that Celtic immigrants were more than the Roman, Anglosaxon or Norman or had a greater influence in the overall genetic pool of the Brits.

The Celts were flowing to the island for like 300 years, the Germanics instead for like 600 years, yet there is consensus that the Germanic paternal (Y-DNA) genetic influence is not much more than 40% in the most affected areas of Eastern England (and virtually zero in the West). The overall paternal-only influnce could be like 15-20%, and the overall genetic influence naturally much smaller.

Now estimate the Celtic influence, considering they flowed into the islands for half that time, please.

It is well known that Celtic culture suffered important transformations in the islands, obviously because of the influence of the native substrate. Druidism was the main one probably. According to Roman geographers and historians, Druidism arrived to the continent from the islands and it is well documented, against popular belief, that Celts outside of transalpine Gaul did not practice druidism. Iberian Celts who had been cut from the continent centuries before La Tène (in Hallstattic times) were certainly not druidistic, and in some cases are even described as "atheist". But I can't recall mention of druids in contexts other than Transalpine Gaul and Britannia, so it's likely that the religious phenomenon never really penetrated far into the east either.

Even if this is true, we can never know that in this genetic chaos. So it is most wise to define populations as they are defined today.

Well, we can find out, I believe. And the "genetic chaos" does give some clues, though the issue of Celtic influence as distinct from other continental influence (Epipaleolithic, etc.) in the islands does have some problematic. The West European heartland (i.e. France, West Germany and the Low Countries) has not been researched well (this is annoyingly true regarding France) and a systematic study to determine the real Celtic and pre-Celtic layers in the islands has not been done yet either (at least AFAIK). But it can be done, exactly the same that it has been done with Nordic ancestry.

I will call Irish Irish and Scotts Scotts but claiming that the only background of these peoples or the English is "Celtic" is misleading, confusing and, in brief, false.

So I won't fall in that terminological trap. Give a thought to the peoples of the islands prior to Le Tène for what they actually were, not for whatever category your subjectivity wants. We don't know their language nor the name they gave to themselves. There are no written records. But we know they were not Celts yet, the same they were not Roman nor Germanic.

Maju said...

I also believe the Irish gaelsic is an Iberian latecomer.

Outmost unlikely. Iberian Celts were isolated from the main Celtic areas of the continent and were not aprticularly prone to sailing. The cultural (and genetic) connections between Iberia and the islands are with all likehood pre-Celtic. There was a lot of Atlantic interaction in the Chalcolithic and Bronze Age but not really in the Celtic Iron Age.

Maju said...

Why would these people, who by sustenance and also by today's genetic evidence clearly could not outnumber local agriculturalists, and whose methods of transportation initially were horribly ill-adapted to the muddy, soft, wet and densely forested CE, provide these people with a new language?

Because they became the aristocracy. Elite domination.

Central European IEs' history begins at Baalberge, which is a very odd Kurgan offshot in southern East Germany. There they took over a minor Danubian group, maybe in the context of intra-Danubian strife (Michelsberg vs. epi-Rössen) and certainly we find clear signs of a rural aristocracy and increaed Kurganization of the burial practices (what surely implies Indoeuropeization of the natives). Baalberge expanded to the Vistula and (for a time) Northern Moravia and colonized the Brandemburgian forest.

Later the resulting cultures shrunk somewhat and became strongly influenced by Danubians south of them (Boleraz-Baden), who experimented a Chalcolithic renaissance. But in the end the Danubians collapsed and the resulting culture (Vucedol) shows again clear signs of IE aristocracy rule. Simultaneously the Central European IEs expanded (Corded Ware) massively through all Central and Northern Europe and a good deal of Eastern Europe as well.

The whole process took about 1000 years and was followed by another milennium of quite apparent peace at the western border of the Rhin. The lands west of it had also been taken from the Danubians by Atlantic peoples (Artenac culture, surely ancestors of historical Aquitanians). We understand poorly the details of the process but Danubian culture was over by c. 2400 BCE in any case, though obviously provided a substrate for later developements like Bell Beaker.

Replacing almost all prior languages in Europe is a tough thought even if it took place during the advent of agriculture (thus my thought that it was helped by being seeded on a receptive stratum - in a well-populated region that already happened to speak similar dialects) - but without that? Unimaginable.

That's just your prejudice. Iberians or Etruscans became Romans quite easily in historical times and neither of them were Indoeuropean-speakers. Same for Sumerians becoming Semites. Elite domination works miracles: people want to climb the social escale and for that speaking the language of the elites is essential. Few languages have resisted such pressure and in most cases they had the advantage of being even more elitistic than the aristocratic elite themselves (Latin and Chinese, for a time Sumerian too). Elamites became Persians, Hellenized Anatolians became Turks, Egyptians became Arabs, Hurrians became Medes (Kurds), and a long etc.

pconroy said...

Eurologist,

You said:

Replacing almost all prior languages in Europe is a tough thought even if it took place during the advent of agriculture (thus my thought that it was helped by being seeded on a receptive stratum - in a well-populated region that already happened to speak similar dialects) - but without that? Unimaginable.


But of course as I pointed out to YOU BEFORE, in the Eulau thread back in November, in regard to the R1a1 Indo-Aryans:

http://dienekes.blogspot.com/2008/11/y-chromosomes-and-mtdna-from-eulau.html
pconroy said...
eurologist said,
... and (ii) a very successful military and agricultural invasion/expansion.

My response:
This need not be the case. R1a and particularly R1a1 were probably spread into South Asia by 2 driving forces:

1. Elite dominance - whereby powerful men had more wives - from the native population - and more offspring, over many generations. Like the Genshis Khan, Niall Naoighiallach phenomenon.

2. R1a1 Steppe herders may very well have carried the Lactose Persistance gene (LCT), that gave them a huge nutritional advantage in a dairying economy, like that of South Asia. Here the real spread is of the LCT gene into the population, and R1a1 is just carried along for the ride!

In other words a relatively small percentage of R1a1 invaders, could result in a large percentage of descendants a few thousand years later.

Tuesday, November 25, 2008 7:01:00 PM



Now I see that Greg Cochran and Henry Harpening offer this EXACT SAME explanation for the Indo-European expansion in general, not just the Indo-Aryans into South Asia.


Why do you think that the area around Denmark and Southern Sweden, was called the, "Womb of Nations" by Jordanes. It so happens that this area of Europe today has almost 100% LCT - Lactase Persistence gene, and probably also had high concentrations in times past. As Cochran and Harpening pointed out, with similar amounts of land, the calorific intake from a herd of cows kept for dairying, as opposed to keeping them for beef, is about 5 times as great - FIVE TIMES!!! So it's no wonder that these area were producing a multitude of people, and were able to expand into nearby agricultural areas, by sheer force of numbers - not to mention the fact that milk drinking makes a warrior much taller and possibly stronger.

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

Here's more on the Germanic Expansions, which illustrate the general Indo-European expansion.

It includes a nice map of Germanic Expansions too!

Maju said...

My response:
This need not be the case. R1a and particularly R1a1 were probably spread into South Asia by 2 driving forces:

1. Elite dominance - whereby powerful men had more wives - from the native population - and more offspring, over many generations. Like the Genshis Khan, Niall Naoighiallach phenomenon.

2. R1a1 Steppe herders may very well have carried the Lactose Persistance gene (LCT), that gave them a huge nutritional advantage in a dairying economy, like that of South Asia. Here the real spread is of the LCT gene into the population, and R1a1 is just carried along for the ride!


These are both very speculative explanations. AFAIK nortehrn South Asians with R1a1 apportions similar to Russia or Poland are not strong in lactase tolerance. In fact the highest concerntrations of lactase tolerance are especially in high R1b regions like the British Islands and the Basque Country, as well as in the more mixed of Northern Europe. I don't make much sense of associating lactase tolerance with IEs or R1a1, sincerely.

The "Chingis Khan phenomenon" has been seriously cricized too. I really hate this kind of hypersimplistic, populist pseudo-explanations of genetic phenomenons. Obviously all descendants of any single well described Y-DNA haplogroup have a single common male ancestor somewhere in the distant past but knowing who he was, when he lived and why his lineage happened to be so succesful is another story. You emphasize patriarchal polygyny, I rather believe more in founder effects and drift.

Why do you think that the area around Denmark and Southern Sweden, was called the, "Womb of Nations" by Jordanes.

Wasn't he writing about the Germanic expansion? What's your point? Restoring the long dead hypothesis that IEs sprung from Northern Europe as Hitler wanted to believe? Sorry but that theory is dead and buried for good. Scandinavia only became source of demic/cultural expansions of significance precisely in the Nordic Iron Age and its continuation through the Middle Ages.

...not to mention the fact that milk drinking makes a warrior much taller and possibly stronger.

As the Japanese case demonstrated, including milk or other dairy products in the diet has a very limited effect in size. Anyhow, most agricultural populations around the world have included milk in their diets, though often in processed forms (cheese, yogurth, curdled milk, etc.) You explain me now what's the nutritional difference between drinking raw milk and eating cheese or yogurth, ok?

Here's more on the Germanic Expansions, which in many way is similar to the general Indo-European expansion in general.

Well, the Scandinavian timeline charter is kind of bad regarding the period they call Neolithic: they make many distinctions in the Bronze and Iron Age but the big periodization interest, IMO, belongs to what they call Neolithic (including Chalcolithic), in which Scandinavia was destination for two succesive waves of migration from Eastern Europe: one from Dniepr-Don, affecting more strongly to Sweden than Denmark, and then the Corded Ware IE one that liquidated the Dolmenic culture altogether.

Apart of that I see only a limited parallel with IE expansions in general. Certainly no other IE expansion sprung from Scandinavia before in any case.

onur said...

I'm not calling anyone alive anything. What I say is that the substrate of the islands and the main share of the ancestry is pre-Celtic.

It is clear that there was people in the islands before the Celts and that these cannot be called "Celtic" with any consistence: it would just confuse all. The identity of modern islanders is rather independent from their biological ancestry: it's cultural.

Anyhow, there's no particular reason to think that Celtic immigrants were more than the Roman, Anglosaxon or Norman or had a greater influence in the overall genetic pool of the Brits.


I haven't said anything contrary to what is written above. I think you have quite misunderstood what I have been saying all along. I use the terms Celtic, Dutch, Scandinavian, etc. to define the existing populations.

My use of Celtic has nothing to do with the historical Celtic populations (e.g., Gauls). Only Celts for me are the existing Celtic ethnicities. Because the amount of the "original" Celtic contribution in the existing Celtic ethnicities is irrelevant for this particular STRUCTURE analysis, but, for instance, the Celtic (I mean the pre-Anglo-Saxon Celtic, not the "original" one) contribution in the people of the UK is.

Well, we can find out, I believe. And the "genetic chaos" does give some clues, though the issue of Celtic influence as distinct from other continental influence (Epipaleolithic, etc.) in the islands does have some problematic. The West European heartland (i.e. France, West Germany and the Low Countries) has not been researched well (this is annoyingly true regarding France) and a systematic study to determine the real Celtic and pre-Celtic layers in the islands has not been done yet either (at least AFAIK). But it can be done, exactly the same that it has been done with Nordic ancestry.

The genetic tests only tell us something about the source regions, not ethnicities. How will you distinguish between Celtic continental and non-Celtic continental?

pconroy said...

Maju,

You seem to be missing the point?!

To paraphrase Eurologist:
If two groups of people - IE's and non-IE's - both have agriculture, then how can one group - IE's - completely dominate the other - non-IE's - over time, given they are at the same or similar levels of technology. That's the 6 million dollar question!

The answer is as I stated above is:
LCT + Elite Dominance, which produces male geneflow into the native or host - non-IE - population(s).

There's nothing simplistic about this - many people fail to understand the compounding effect of a small advantage over time. So they look at archaeology and see no massive invasions and can't understand how male population replacement happened.

Across much of Europe, especially Northern Europe, mtDNA, is fairly consistent across populations, and has relatively old TMRCA's. Yet Y-DNA is not like this at all, it shows evidence of huge sweeps in different parts of Europe. So for example Basque Y-DNA is mostly R-M153 (aka R1b1c4) and R-M167 (aka R1b1c6 or R-SRY2627), which IIRC have a TMRCA of at most 4,000 years ago. So male mediated geneflow, carrying LCT, into the Basque population, has over time largely wiped out the original Basque male genetic signature.

Maju said...

Only Celts for me are the existing Celtic ethnicities. (...) but, for instance, the Celtic (I mean the pre-Anglo-Saxon Celtic, not the "original" one) contribution in the people of the UK is.

That sentence is utterly contradictory. Hope you realize that.

The genetic tests only tell us something about the source regions, not ethnicities. How will you distinguish between Celtic continental and non-Celtic continental?

You have a point here.

Still, the difference between what you see in Britain (excluded what is clearly Nordic) and what you see in the mainland (northern France, Belgium, Rhineland, etc.) is pre-Celtic with all likehood and probably even pre-Neolithic. So for instance when you see more R1b in the islands than in the mainland, it means that it's pre-Celtic. Inversely, as you see more R1a in the mainland, it means it's Celtic (or other IE). I doubt you'll find much difference in mtDNA because IE migrations were mostly male-mediated.

In any case the continent was earlier and more intensely affected by the IE waves that eventually also reached the islands. So any cline increasing to the sea is pre and any cline growing to Central or Eastern europe is most likely post.

If two groups of people - IE's and non-IE's - both have agriculture, then how can one group - IE's - completely dominate the other - non-IE's - over time, given they are at the same or similar levels of technology. That's the 6 million dollar question!

Give me those 6 million dollars. Or better euros, the dollar is not anymore what it used to be.

You have many well documented historical examples of militarist minorities dominating peasant majorities and absorbing them ethnically. Well known examples:

- Akkadians absorbed Sumerians
- Persians absorbed Elamites
- Germanics absorbed Celts
- Romans absorbed Etruscans and Iberians
- Turks absorbed diverse Greek-speaking Anatolian peoples

All these peoples were comparable in technology and culture but there were winners and losers, rulers and subjects.

So for example Basque Y-DNA is mostly R-M153 (aka R1b1c4) and R-M167 (aka R1b1c6 or R-SRY2627), which IIRC have a TMRCA of at most 4,000 years ago.

I strongly contest those estimates. They just make no sense. TRMCA estimates is not rocket science, otherwise we would not be able even to fly.

And seriously I make no sense of the at all. I have discussed in other places: Y-DNA is like surnames and we know that most Basque surnames have been existing since the Middle Ages and none has apparently become dominant in any way since then. Sure that there's some minor erratic drift but that's it.

And I am positive that the same lack of rapid swifts in surname dominance can be detected in other places where these have been used since long ago and there are registries.

For such drastic sweeps, you need much lower populations where drift can effectively act in the way you claim.

pconroy said...

Maju,

You say:
You have many well documented historical examples of militarist minorities dominating peasant majorities and absorbing them ethnically

But the examples you give are Cultural domination for the most part, not Genetic domination, which is what I'm talking about.

The reason the Basque language and culture exist today, but the Basques are genetically almost identical to other Spaniards is because the Basque have been historically matrilocal and not patrilocal - simple as that!

The Basque language is probably Neolithic from the Southern Caucasian area, or Kartvelian in other words, and a relatively late arrival to Europe.

Maju said...

But the examples you give are Cultural domination for the most part, not Genetic domination, which is what I'm talking about.

My bad then, I misunderstood your question.

But still elite dominance can cause genetic dominance as well.

In my opinion and for what I see for instance in regard to R1a, it's more likely that a lineage switch happens where the natives are not too dense anyhow. So the relatively marginal NE Danubians or the rather lowly populated Chalcolithic Sweden were more easily swallowed (up to apoint) by the IE tide than more densely populated areas like Hungary or Denmark. Similarly in South Asia, IE genetic dominance was more succesful in Uttar Pradesh than in IVC areas, that probably hosted a much more dense population back then. Or in other temporal context, Turco-Mongol Y-DNA took over the semidesertic Khazakstan but is much more moderate further south, even among Turkic speakers.

I think that dense well estabilished agrarian populations are much less likely to be genetically displaced by any newcomers, even if very dominant.

Instead there are people like you who seem to believe that a lucky person can virtually take over the genetics of a whole nation almost alone, at least the Y-DNA lineages. I think that such ideas are way closer to the realm of fantasy than to realism. Read what I wrote before about surnames.

eurologist said...

pconroy,

Milk use is documented 8,000 years ago in Southeast Europe - it pretty closely followed the advance of agriculture, and pre-dates the Kurgan expansion or any other by many millenia. For Central Europe, it is not far off to say that milk usage was part of the agricultural package almost from the get-go.

onur said...

I say:
Only Celts for me are the existing Celtic ethnicities. (...) but, for instance, the Celtic (I mean the pre-Anglo-Saxon Celtic, not the "original" one) contribution in the people of the UK is.

Your response:
That sentence is utterly contradictory. Hope you realize that.

The premise of my argument is that little change has occured in the genetic structure of the Celtic populations that survived the Anglo-Saxon invasions since pre-AS times (at least in the regions unaffected by the Viking invasions).

Still, the difference between what you see in Britain (excluded what is clearly Nordic) and what you see in the mainland (northern France, Belgium, Rhineland, etc.) is pre-Celtic with all likehood and probably even pre-Neolithic. So for instance when you see more R1b in the islands than in the mainland, it means that it's pre-Celtic. Inversely, as you see more R1a in the mainland, it means it's Celtic (or other IE). I doubt you'll find much difference in mtDNA because IE migrations were mostly male-mediated.

In any case the continent was earlier and more intensely affected by the IE waves that eventually also reached the islands. So any cline increasing to the sea is pre and any cline growing to Central or Eastern europe is most likely post.


I think there is little evidence to substantiate your claims above. The genetic structure of Europe was already messed up before the IE invasions. That is why it is near-impossible to detect the IE invasions genetically.

Maju said...

The premise of my argument is that little change has occured in the genetic structure of the Celtic populations that survived the Anglo-Saxon invasions since pre-AS times (at least in the regions unaffected by the Viking invasions).

I agree with that. But I'd also think that the peoples you call Celtic now are also probably the ones that recieved less genetic input from Celts.

The genetic structure of Europe was already messed up before the IE invasions. That is why it is near-impossible to detect the IE invasions genetically.

I think we can safely say that R1a expanded with Indoeuropeans. Other clades surely also benefitted from IE expansion here and there but R1a is a very clear marker. Where R1a is weak we can safely say that the genetic influence of IEs was tenuous at best, regardless of their cultural and linguistic impact.

I would not agree that the genetic structure of Europe is "messed up" (i.e. overly complicated), in fact it's one of the most simple of all Eurasia. It is the simplicity which is actually striking, not its complexity.

pconroy said...

Maju said:

I would not agree that the genetic structure of Europe is "messed up" (i.e. overly complicated), in fact it's one of the most simple of all Eurasia. It is the simplicity which is actually striking, not its complexity.


I wholeheartedly agree - especially on the Y-DNA side of things - as how can there be so many invasions and settlements and colonizations, and today we find a relatively simple pattern of associated Y-DNA lineages account for the vast bulk of male lineages in Europe - I'm talking about R1a1 and R1b1c and their downstream markers. Likewise on the mtDNA side H and its downstream clades are over 50% almost everywhere. So it looks like, among other things, selective sweeps. In the first case LCT, in the second case anti-sepsis.

McG said...

My studies indicate that a 12 at 393 for R1b may indicate a preceltic? origin. Jackson says that early scotland spoke 3 languages, pict, gaelic and an older language understood by the Pict but not by the gael. where these older pre picts came from is a TBD in my opinion.

pconroy said...

McG,

Many of the descendants of the Border Reivers think that they are ultimately of Central Asian, especially Alano-Sarmatian descent,
Border Reiver

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

But I'd also think that the peoples you call Celtic now are also probably the ones that recieved less genetic input from Celts.

I agree with that, but as I said before, that isn't my concern, at least for this particular STRUCTURE analysis.

I think we can safely say that R1a expanded with Indoeuropeans. Other clades surely also benefitted from IE expansion here and there but R1a is a very clear marker. Where R1a is weak we can safely say that the genetic influence of IEs was tenuous at best, regardless of their cultural and linguistic impact.

It isn't clear how much of R1a was spread by the Kurgan people, Ukrainain LGM refuge, Indo-Aryans, Slavs or some other populations. As I said, genes say almost nothing about the language (this is especially true for Europe).

I would not agree that the genetic structure of Europe is "messed up" (i.e. overly complicated), in fact it's one of the most simple of all Eurasia. It is the simplicity which is actually striking, not its complexity.

It is this simplicity, the relative homogenity across populations, which I call as messed up. It makes the detection of migratory events almost impossible.

Maju said...

how can there be so many invasions and settlements and colonizations, and today we find a relatively simple pattern of associated Y-DNA lineages account for the vast bulk of male lineages in Europe - I'm talking about R1a1 and R1b1c and their downstream markers. Likewise on the mtDNA side H and its downstream clades are over 50% almost everywhere. So it looks like, among other things, selective sweeps. In the first case LCT, in the second case anti-sepsis.

Uh!?

There have not been "so many migrations", IMO. Since Paleolithic or Epipaleolithic times you can think of basically two, both quite limited in their impact: Neolithic waves from West Asia and the Balcans and Indoeuropean/Uralic waves from extreme Eastern Europe (maybe confused with the Dniepr-Don "aperitive" wave but not substantially distinct).

Both are very easily detectable in the Y-DNA landscape: J, E, T, G and I2a for Neolithic and R1a and maybe I1/I2b for Indoeuropeans/Dniepr-Don.

And both decay towards the west, as we would naturally expect.

The only doubt I could have is wether R1b could be a Neolithic marker, instead of late Paleolithic but it's really impossible to find a plausible Neolithic source for it.

Regarding mtDNA, H is evrywhere as you say and is estimated normally (TRMCA estimates) to be of Gravettian origin. Gravettian culture was the first and maybe the only pan-European culture ever and had offshots in West Asia (Zarzian) and probably North Africa as well (Iberomaurusian/Oranian). It's only logical that if we are looking for the origins of genetics that are pan-European, we must look at Gravettian. No other culture ever since then has covered the continent so homogeneously.

where these older pre picts came from is a TBD in my opinion.

Ok, I have been able to infer that LCT means "lactase" but what the heck does TBD mean?

Many of the descendants of the Border Reivers think that they are ultimately of Central Asian, especially Alano-Sarmatian descent,

If they are so, it would be irrelevant for our discussion but for what I've seen re. their haplotypes, they fit perfectly within the Western European context. That page is way too old for me to make sense of it in any case.

It isn't clear how much of R1a was spread by the Kurgan people, Ukrainain LGM refuge, Indo-Aryans, Slavs or some other populations. As I said, genes say almost nothing about the language (this is especially true for Europe).


You are talking about basically the same phenomenon, hope you notice. Still R1a is hardly a sub-IE (Slavic, Indoaryan) marker. You could not explain its presence in Sweden or Turkey using those populations as vectors. All those you mention (but Paleolithic Ukranians probably) were IE speakers. The influence of the Dniepr-Don refuge outside of Eastern or Balcanic Europe anyhow appears to be null, except for the Dniepr-Don migrations in the Chaclolithic era that did affect the Balctic. In general I tend to associate these rather with haplogroup I and exotic variants of R1b but I may be wrong in this.

onur said...

You are talking about basically the same phenomenon, hope you notice. Still R1a is hardly a sub-IE (Slavic, Indoaryan) marker. You could not explain its presence in Sweden or Turkey using those populations as vectors. All those you mention (but Paleolithic Ukranians probably) were IE speakers. The influence of the Dniepr-Don refuge outside of Eastern or Balcanic Europe anyhow appears to be null, except for the Dniepr-Don migrations in the Chaclolithic era that did affect the Balctic. In general I tend to associate these rather with haplogroup I and exotic variants of R1b but I may be wrong in this.

I think you are confusing genetics with language. We don't know much about pre-historical European cultures and their migratory patterns. Genetics is newly beginning to give some clues. But these clues themselves tell us almost nothing about specific migrations; they mostly demonstrate only some broad patterns of human dispersion.

McG said...

TBD is to be determined. The border reivers were intermingled with the Roman fighters early on and preserve that heritage. That blood didn´t make it much into the highlands. I agree with your irrelvance statement.

Gioiello said...

"My studies indicate that a 12 at 393 for R1b may indicate a preceltic? origin. Jackson says that early scotland spoke 3 languages, pict, gaelic and an older language understood by the Pict but not by the gael. where these older pre picts came from is a TBD in my opinion"

Hold out with these ideas, Mac Gregor. We shall overcome.

Maju said...

I think you are confusing genetics with language. We don't know much about pre-historical European cultures and their migratory patterns. Genetics is newly beginning to give some clues.

Before taking a good dip into genetics, I took also a good one into European archaeology - and genetics has not deterred me from continuing learning in this field. I understand that both fields are complementary but in general archaeology seems to offer more direct and solid clues for understanding the past.

And while the archaeology of other regions may have immense blanks, European one is very well studied. We know pretty well about the cultural flows of the past in this little corner of the World only based on archaelogy.

So when I started to learn about genetics in the last decade, I soon correlated it with prehistory as known from artifacts, settlements and tombs. Most often it just enriches the fine detail of the knowledge gained from arcaheology, in some cases poses little mysteries but in general it is very much in agreement.

When they strongly seem to disagree, I tend to suspect that the failure is in population genetics, after all a very young science. Most of he aparent disagreements, if not all, anyhow come from the TRMCA estimates, so I have developed a deep lack of respect for the molecular clock hypothesis.

But these clues themselves tell us almost nothing about specific migrations; they mostly demonstrate only some broad patterns of human dispersion.

Human dispersion = migrations.

I think (and I'm not alone in this) that those broad patterns are best understood in relation with well documented prehistorical migrations and cultural flows.

Maju said...

My studies indicate that a 12 at 393 for R1b may indicate a preceltic? origin

Hardly so.

The three most common European haplotypes (I'm following Alonso 2005 here) in Europe have all DYS-393=13. Instead DYS-393=12 appears to indicate the Anatolian branch, which may have expanded in Neolithic times.

For reference check my reconstructed tree of R1b(xR1bf) haplotypes based on that study. Only the SE group has DYS-393=12.

onur said...

I think (and I'm not alone in this) that those broad patterns are best understood in relation with well documented prehistorical migrations and cultural flows.

Which prehistorical migrations and cultural flows do you mean?

Maju said...

Which prehistorical migrations and cultural flows do you mean?

I'm too tired to explain them again, buy a good prehistory manual.

Gioiello said...

"For reference check my reconstructed tree of R1b(xR1bf) haplotypes based on that study. Only the SE group has DYS-393=12"

What are you saying, Maju? The "Anatolian haplotype" (and we shall see if it is "Anatolian" or not) is the most ancient, the ancestor, and the "Western" is derived. My interpretation, and my bet, is that it arrived to Italy during the "Younger Dryas" and from that derived all European subclades. Anyway you hypothesis of a Western Refugium is full falsified. Don't you know the pathway of the SNPs? And that we find the most ancient in Italy and not in Spain? What we find in the British Isle has come from South (I think from Italy), anyway from South, South-East or East, being the origin in West Asia. Which are you speaking about?

Ponto said...

I could have told you without the results that Australians would be indistinguishable from the UK, mostly English people.
The Irish element is 25% of the population but they are wholely Roman Catholic, and they choose to segregate themselves preferring the Catholic school system.
I have lived in Australia my whole life and I don't know any Irish Australians as friends and know no Catholics, and would not choose one as a spouse.

Maju said...

@Gioello:

1. My recosntruction is based only in the assumption that the most common haplotype is in fact the root. It can be wrong but makes some sense (and is extremely convenient if you don't want to end with a confuse labyrinth)

2. I just say that the DYS-393=12 is typically Anatolian (and Balcanic) and not typically Western, whichever the direction of the mutation.

3. Anyway you hypothesis of a Western Refugium is full falsified.

Since when? Archaeologically speaking we know that the Franco-Cantabrian province was the most densely populated of all Europe in all the UP (by large), we know there was an expansion into Central Europe and into southern Iberia in Magdalenian times and we know that later on there was an expansion from more northern areas (Tardenoisian complex), for those who are inclined to younger age estimates.

Don't you know the pathway of the SNPs? And that we find the most ancient in Italy and not in Spain?

No, I don't.

But even if that is certain it would not matter much, especially because Spain is not the postulated "western refuge" but Southern France (with a small fraction of modern Spain). And also because the original direction of the migration must have been from the East in any case.

Dwelling on misconceptions and confuse information doesn't seem to help at all.

What we find in the British Isle has come from South (I think from Italy), anyway from South, South-East or East, being the origin in West Asia.

I have not found yet any clear evidence of such claims. At most alleged higher diversity in Anatolia but this is unclear and may have been caused by other reasons (back migrations, reduction of diversity in the west in the LGM...) if correct. Alonso 2005 gives the impression anyhow that Anatolia does not harbour higher diversity within R1b but that the highest haplotype diversity is in Western Europe. We also find higher subclade diversity in the West (nearly all known R1b subclades are there).

At the moment I have yet to read anything conclusive in favor of an overall Anatolian origin for R1b.

Gioiello said...

I don't believe to an Anatolian origin, but from West Asia through Russia (if R1b carried an IE language, certainly they lived North of Caucasus, being IE closest to Ugro-Finnish than to any other Languages Group). The opened problem is the time: I "hope" on a more ancient one (Zhivotovsky calculation, for supporting the possibility of an Italian Refugium during the Younger Dryas), Pinzochero to a recent one, for supporting an origin from Middle East and to save all the R1b Jews, I think of European origin. But if you watch the recent R1b-ht35 project, the most ancient R1b are in Italy, South Germany, British Isles. This I think is the pathway, and the more recent SNP are above all in Central-North Europe. I think this are findings very hardly deniable.

Maju said...

But if you watch the recent R1b-ht35 project...

I don't seem to have access to that. But I did find some related info when searching for it and "ht35" (i.e. R1b with DYS-393=12) clearly a SE distribution in Europe and can by no means be considered "native western" as you pretend, at least not over DYS-393=13, infinitely much more common.

I don't believe to an Anatolian origin, but from West Asia through Russia (if R1b carried an IE language...

All of which makes no sense whatsoever because R1b is as rare in Russia as rivers in the Sahara and the only area of West Asia where R1b is rather frequent is precisely Anatolia.

Additionally believing in R1b to be associated with IE expansion makes no apparent sense.

...the most ancient R1b are in Italy, South Germany, British Isles.

How do you know it actually is "the most ancient"? In principle all modern clades should be equally derived (young).

Gioiello said...

I can watch the "R1b-ht35 Project" and I am banned from all forums, then I think you can do too, if you want.You can reach it also by a link of Pinzochero (alias Vizachero) on "Genealogy-DNA".
If this is comfortable for you, Pinzochero is announcing by "twitted" that in a conference he is present Dr Wells supported yet the "Cantabrian Refugium" for R1b, with astonishment of Nordtvedt and others. I don't replay to the rest: that haplotype 14, 24, 10, 13, 12 is ancestral to all R1b subclades I think it is out of any possible doubt, then the way of your map must go on the contrary.

McG said...

Again, I think the truth is still unknown re R1b origins. The present destribution doesn´t support a east to west migration in my opinion. I do believe the clock works and I´ve refined my TMRCA estimates this winter and willpost on my blog this spring. I think the biggest problem of interpreting time is the prior occurrence of catastrophes which erase the records. There is too much common myth about the past.

Maju said...

@Gioello:

I can watch the "R1b-ht35 Project" and I am banned from all forums, then I think you can do too, if you want.

Maybe if you provided a link...

I found a Family Tree site administered by that Vizachero guy but the information provided is most limited and doesn't help me the least to make up my mind.

I don't replay to the rest: that haplotype 14, 24, 10, 13, 12 is ancestral to all R1b subclades I think it is out of any possible doubt...

Why?

Just repeating the same sentence once and again does not give you the reason. Please provide evidence backing your claim.

@McG

Again, I think the truth is still unknown re R1b origins. The present destribution doesn´t support a east to west migration in my opinion. I do believe the clock works and I´ve refined my TMRCA estimates this winter and willpost on my blog this spring.

I'd like to take a look at that, yet your blogger profile provides no link. :-(

I agree that with the currect evidence it is very hard to support an east to west migration for R1b1b2a but there's also a lot of non-European data to be analyzed (in particular African R1b and also where do Sardinian R1b1a and African R1b1c fit in the tree).

There must have been an East to West migration including R1b as a whole though (partly because of where the closest relatives are found), and that is the only thing that can explain the presence of rare branches out of West Eurasia (R1b1c, R1b1b1) or restricted to the Mediterranean (R1b1a). But it's very hard to explain an R1b1b2a explosion from Anatolia, much less from Russia or Italy.

Gioiello said...

For Maju

https://www.familytreedna.com/public/ht35new/default.aspx?section=yresults

Gioiello said...

For Maju

aspx?section=yresults

Maju said...

Got it, thanks.

Not sure what to make with it anyhow, it'd seem like most true HT35 (i.e. those that actually have DYS-393=12) would belong to R1b1b2a*, while all R1b1b2a1 and R1b1b2a1a belong to DYS-393=13, with two exceptions, one with value of 12 and the other of 14.

The big explosion of R1b is at the R1b1b2a1a (L11) level, so that basically suggests a "thin thread" flowing from the East and the big expansion already in the West with DYS-393=13. Or alternatively (less likely) a "thin thread" flowing towards the SE prior to that western R1b "explosion".

Whatever the case, the R1b1b2a1a (L11) explosion happened in Western Europe it seems. And following derived haplogroups may have got two subcenters: one in northern Europe (R1b1b2a1a1) and another near the Pyrenees (R1b1b2a1a2). Nevertheless most known R1b is R1b1b2a1a*, what suggests that the main explosion, with loads of shrapnel, took place at that level.

Gioiello said...

Yes, this was also my hypothesis: an Italian Refugium during the Younger Dryas of R1b1b2 and R1b1b2-a and after the migration to North, North West.
Last year a wrote a post on "dna-forums" on a Spanish paper and I think having demonstrated that also the "Basque" haplogroups (R-M153 and R-M167) probably was born in South France, during the migration from Italy. But for supporting my theory I need Zhivotovsky and not Pinzochero or Nordtvedt (who is a clever boy and remains a friend).

Maju said...

Well, the Italian refugium is a really unlikely origin for Western European Y-DNA, because we know that Magdalenian expanded from the Franco-Cantabrian region, but Italy remained epi-Gravettian all the way to Neolithic.

You have to realize that there is a Prehistory, which may not be fully clear in all its details but that it is well known anyhow - at least in Europe. And Italy, AFAIK, remained marginal from Gravettian to Neolithic, when it served as "bridge" between the Adriatic and teh Western Mediterranean - and, after Neolithic, till the Romans more or less. Overall Italy looks more like a reciever than any origin, unless you want to "blame" the Romans for R1b - what I would not even consider.

Also the presence of Western and Eastern haplogtypes in Italy is well explained by the existence of two different sources. And the eastern source is also found in other Italian haplogroups anyhow.

I stick to the Paleolithic model, even if I am now more inclined to consider Epipaleolithic flows as well.

Gioiello said...

R1b probably is recent in Europe, not Paleolithic at all. Pinzochero and many others are thinking to the late neolithic. I think to the Epigravettian, where I put my Italian Refugium.
I have many proofs also from mtDNA. I wrote something on a previous posting of Dienekes on R1b in Italy. That Italy was the Refugium of N1b (and that those Jewish N1bs come from Italy), I think it is demonstrated by the last "Phylo Tree.org - mtDNA tree Build3". I am collecting many samples from Italians who live in other countries: H5, T2f, J2 etc. etc.
And, last but not least, my K1a1b1 and my 23% Ashkenazi autosomal with many millions of documented Tuscan ancestors.

Maju said...

R1b probably is recent in Europe, not Paleolithic at all. Pinzochero and many others are thinking to the late neolithic.

But those conclussions are only based on TRMCA estimates, what is otally speculative. Sloppy pseudoscience. You need something more than just that to get a solid estimate and that's why I always demand a plausible pre-/historical scenario.

I have many proofs also from mtDNA. I wrote something on a previous posting of Dienekes on R1b in Italy. That Italy was the Refugium of N1b (and that those Jewish N1bs come from Italy), I think it is demonstrated by the last "Phylo Tree.org - mtDNA tree Build3". I am collecting many samples from Italians who live in other countries: H5, T2f, J2 etc. etc.

Not sure what's your evidence but one thing is Italy being the source of some small lineage, something that would be perfectly plausible, and another the source of the vast majority of paternal ancestors in Western Europe, what demands a solid explanation.

I can only think of the following founding situations (in chronological order):

1. Aurignacian - probably too old.
2. Gravettian - a good candidate that nevertheless clashes with the usual age estimates. It's probably at the origin of mtDNA HV, H and V spread, which are very much parallel in distribution (and starlike structure in the case of H) to R1b.
3. Magdalenian - a classical. The final Magdalenian area fits terribly well with modern R1b1b2a spread and the LGM before it offers a nearly perfect scenario for massive drift and fixation.
4. Tardenoisian and related Epipaleolithic cultures (geometric microlithism, Ertebölle, etc.) - fits well with the alleged "Neolithic" chronolgy (just 1000 years older or so) and spans for the whole R1b1b2a area as well.
5. Dolmenic Megalithism - it is "late Neolithic" (Chalcolithic for the most part) and its spread again covers pretty well the modern R1b1b2a area. The main problems are that Portugal (likely origin) is not particularly high nor diverse in R1b and how to explain massive replacement after Neolithic was consolidated in most affected areas.

So I'm very strongly inclined for the 2, 3 and/or 4 timeline. In dates: from c. 28,000 to c. 8000 BP. If we exclude Gravettian: 17,000-8000 BP. If we only consider post-LGM events: 13,000-8000 BP.

Gioiello said...

You are very well-read in Archeology, better than me, and I have no doubt about your knowledge.
I am not able to discuss your assertions.

"Not sure what's your evidence but one thing is Italy being the source of some small lineage, something that would be perfectly plausible, and another the source of the vast majority of paternal ancestors in Western Europe, what demands a solid explanation".

Remembering a recent paper on the peopling of North America, as large group are diffused overall and aren't able to demonstrate anything about their origin, I think that the very small group can demonstrate this. There are findings about mtDNA H in Italy 28,000 years ago and also of N1, etc. Ancient Italy wasn't that "waste land" many peoples think about and "April isn't the cruellest month".

McG said...

my blogspot is www.kerchner.com/mcgregor/table.htm
don´t pay any attention to ASD comments, I no longer use that approach. the flood study is appropriate to this discussion and suggests that a fairly large northern hemispherical effect was felt. I am suspicious of catastrophism but there is a lot of technical references that somewhat support the biblical reference.

Maju said...

There are findings about mtDNA H in Italy 28,000 years ago...

That basically seems to demonstrate that H expanded with Gravettian, in Italy and elsewhere (that we lack of aDNA data for elsewhere does not imply that Italy was the source: Gravettian known origins are in Central Europe).


Ancient Italy wasn't that "waste land" many peoples think about...

Italy was not any wasteland. But doesn't seem central for any major cultural/demic process before the Roman Empire. Over the time it became receptor of many different peoples - and it's not the only European region in this situaton anyhow.

Maju said...

@McG:

Thanks for the link. The first document is the one relevant, right? You claim to find, following Zhivotovski mutation rate, that West European R1b has dates ranging from 12,000 to 8,000 BP. I find its most interesting and surpsisingly accurate for a TRMCA estimate. As I said above, if we count from the LGM to Tardenoisian/geometric microlithism, the window is 13,500 to 8000 BP. I am gladly surprised that you seem to reach to similar conclusions by totally different methods. Chapeau!

You also mention that you estimate the R1a-R1b split c. 23,000 BP. This I have some difficulty to fit within the archaeological frame instead. If the dates could be stretched some 20-25%, they would fit perfectly with R1b arriving to Europe with Gravettian and R1b1b2a "exploding" with Magdalenian.

This would be the most convenient explanation, the one that presents no difficulties other than TRMCA hunches. It would also restore the parallel between R1b and mtDNA H, which we know now it's Gravettian by origin and shows a starlike structure in Europe that resembles way too much the one of R1b1b2a.

McG said...

the work I have done this winter is better. I have used and analyzed a rootsweb database that that has about 22K entries for R1b´s first twelve dys loci. The R1a to R1b split is tough to predict. to few survivors. high SD.

Gioiello said...

"Italy was not any wasteland. But doesn't seem central for any major cultural/demic process before the Roman Empire. Over the time it became receptor of many different peoples - and it's not the only European region in this situaton anyhow"

We all come from elsewhere. The problem in "when". J2a1k1 came recently from Balkans (or Greece) or arrived to Italy many thousands of years ago? R1b1b2 has wintered in Italy during the Younger Dryas and after peopled all Western Europe or arrived to Italy from North (West or East)? But why we don't find in Italy R-L21? And why we don't find in Spain R1b1b2 (L23-,L51-), except some of Roman descent?

McG said...

re nomenclature. I call R1b with a 13 at393 Celtic. With a 12 I use the term Alban(alba, alpes,albania etc.). These people occupied the mountainous areas from the caucasus to the british isles. They suffered from severe climate change and in general didn´t migrate from their highland heritage. Alba is a very old term, used in latin to mean white.
When I return to NY I plan to recompute all my estimates and update my site.

pconroy said...

Ponto said...
I could have told you without the results that Australians would be indistinguishable from the UK, mostly English people.
The Irish element is 25% of the population but they are wholely Roman Catholic, and they choose to segregate themselves preferring the Catholic school system.
I have lived in Australia my whole life and I don't know any Irish Australians as friends and know no Catholics, and would not choose one as a spouse.

Ponto,

I think you need to hone your reading skills and get out of Alice a bit more!

Of course people of Irish descent are not 100% Catholic, I have a bunch of close relatives in Australia of Irish descent, who are both Catholic and non-Catholic, Irish married to English, etc.
Outside of Ireland, the most Irish culture in the world is Australian, your national sport Aussie Rules, is derived from a game Gaelic Football, played by Irish convicts. Aussie national icons like "Paul Hogan" are of Irish descent, along with many of you premiers. I know from my cousins that Aussies have a distaste for Pommies (or Poms) which are recent English settlers, as Aussies view themselves as anything but English.
I also have a bunch of cousins and family in New Zealand, and that country is much more English than Australia.

In regards to the structure chart, if the Aussies have 25 to 40% known admixture with Irish, AND they match the UK same, ERGO the UK sample is probably 25-40% Irish...

pconroy said...

typo
... and they match the UK sample...

Maju said...

@Gioello:

IF R1b1b2 arrived from West Asia BEFORE the LGM, evolved into R1b1b2a in (possibly) the Franco-Cantabrian Region, became fixated in this clade there at the LGM population minimum and THEN expanded with Magdalenian (and derived cultures), you have the THEORETICAL answer to your query.

The origin of R1b or R1b1b2 does not need to be the same as that of R1b1b2a, which happens to be main haplogroup in Western Europe. R1b can perfectly be West or Central Asian, R1b1b2 "Balcanic" (or from somewhere nearby like Asia Minor... or even Italy) and still R1b1b2a can be Western European by origin.

McG said...

re maju. R1a to R1b is a tough call. very few samples left in the population. VV has some older 12´s which I will look at. to get full TMRCA to archaeological correspondence we have to have descendants all the way back to the SNP!

pconroy said...

Maju, McGregor,

I see much of the recent evidence converging on the idea that R1b and its subclades in Western Europe are largely of Neolithic and later Indo-European origin.

The idea that R1b and subclades represent the Paleolithic population of Europe seem dead in the water.

For example there is a fine article Debunking the Basque myth, by Jean Manco.

McG said...

this is a very argumentative subject. my analysis says R1b = 13 occurred somewhere about 6K to 8K years ago, possibly after the flood? Maju has a better feel for the archaeology, I just report the numbers. In my judgment very few of the old population left bloodlines and thus the emphasis on a more recent time frame. Europe wasn´t a very nice place to be at the end of the ice age, JMHO

Maju said...

I see much of the recent evidence converging on the idea that R1b and its subclades in Western Europe are largely of Neolithic and later Indo-European origin.

I see a lot of people saying the same, yet where is the "evidence"? Some fancy TRMCA hunch? Not better than McGregor's surely. Has anybody tested any aDNA yet in European fossil humans? Not as far as I know. Where is the "evidence", then?

Ma che! (this is Italian for "get real!")

this is a very argumentative subject. my analysis says R1b = 13 occurred somewhere about 6K to 8K years ago, possibly after the flood? Maju has a better feel for the archaeology, I just report the numbers. In my judgment very few of the old population left bloodlines and thus the emphasis on a more recent time frame.

In my opinion, there was never any flood and you can trace the populations archaeologically step by step from Aurignacian to the present. Please: a few kilometers from my great-great-gandfather's farmhouse there is a cave that has the whole sequence from Chatelperronian to the Iron Age (Santimamiñe). It's maybe exceptional but it's not the only case certainly.

Anyhow, at no moment you get the feeling of discontinuity. Change for sure (but normally gradual) but never a widespread esterile layer or a techno-culture that you cannot trace to the previous one quite nicely. And certainly no indications of any "flood" at all.

And (this goes for the "Indoeuropeists") when the Celts arrive, you see conflict at the Ebro border. And then written history arrives and the Celts are south of it (and north of the Garonne) and then linguists come and see no trace of Celtic influence in Basque (in contrast with a very strong Latin one, for instance). Etc.

I could even take that Basque language is of Neolithic origin because languages can be replaced much more easily than genes. But I can't find any good explanation for a genetic explosion following the Atlantic coasts that can be attributed to Neolithic or post-Neolithic events.

It's just that the data don't fit well, or even poorly. I really couldn't care less but when you look at it with some true interest you see continuity. Furthermore, you see a very strong population growth following Neolithic (more in the Chalcolithic period and thereafter, as Neolithic proper was short-lived and tenuous here). Even anthropometrically you see continuity, except partly in the Ebro area, where gracile Mediterranean types are relatively common.

You have to present very solid evidence, much more than a MCH hunch, to persuade me that there was a demographic replacement in the Basque Country - or anywhere else in Atlantic Europe. In fact, I don't see clear evidence even for any signifcative colonization in the Western Mediterranean - just a gradual, or more probably epysodical, flow of Eastern Mediterranean and, in Iberia, North African elements.

Maju said...

Caveat:

I said: a techno-culture that you cannot trace to the previous one quite nicely

Graettian may be the only exception. It's sometimes considered "intrusive" in the Franco-Cantabrian area and arrived quite late to SW Europe, just before the onset of the Solutrean "reaction".

McG said...

Maju I´ll bet that cave isn´t at sea level? north america is just one huge glacial remanant. gorges shaped by glaciers etc. and not all that long ago. Look up doggerland and the 200 M increase in sea level. The ark on Mt Araraat wasn´t there on holiday? Again, this is argumentative and catastrophes are a huge eraser, removing earlier traces. I´ll try to explain more later.

Maju said...

Maju I´ll bet that cave isn´t at sea level?

Not at sea level like Gorham's Cave but just above it, overseeing a very rich swampy area that now is a natural reserve.

north america is just one huge glacial remanant. gorges shaped by glaciers etc. and not all that long ago.

An image is worth a thousand words: aprox. extension of the LGM ice sheet, for more detailed maps (only for Eurasia), check here.

Half or more of North America was ice free, including most of mainland USA and all Mexico. In Europe the ice sheet covered the Baltic area but only half on Britain and just small portions of Germany. Of course there was tundra just south of it...

Look up doggerland and the 200 M increase in sea level.

Sure there was an increase in sea level everywhere, and many areas then exposed are now under the sea (the extent varies depending of local orography) but most of the land inhabited in the UP remained above sea level all the time and, anyhow, the change was not catastrophic but gradual.

The ark on Mt Araraat wasn´t there on holiday?

Uh, what ark? That's just a tale for children, like Santa Claus. In fact it's a deformation of a Sumerian legend, which some believe refers not to any specific flood (not located archaeologically, at least not any one of even remotely comparable dimensions) but to a flood of people: the Semites.

Again, this is argumentative and catastrophes are a huge eraser, removing earlier traces.

They leave traces of their own. Toba explosion left a thick layer of ashes, several meters wide. A flood like the one you speculate with would leave massive sedimentary deposits of silt. I am totally undaware of any such evidence. I know of more localized catastrophes and their clear signature, like the silting of the marine branch that linked Zambujal (major Chalcolithic city of Portugal) with the sea, possibly caused by a tsunami but not anything like what you suggest.

The sea level rise was gradual. Even a massive sea level rise nowadays, as forecasted by scientists, would not be sudden: even if it can destroy our coastal cities, most people would just be able to, calmly and gradualy move to a higher ground.

McG said...

I suggest you look at some of my references re doggerland and the east coast of scotland. there was a tsunami off the coast of norway that dumped millions of tons of debris in that area. In any case, I cannot prove a catastrophe, its just that there is a discontinuity in R1b from 12 to 13 at 393 at about the time of the flood, about 6K BC. the annals of the time all reflect a major disaster occurred in the Black Sea region and others. I respect your opinion so I´ll leave it at that.

pconroy said...

McGregor,

The problem with "The Flood" is that there were many floods, many times, and there are flood legends in many cultures that inhabit coastal or lake-shore sites.

terryt said...

"The ark on Mt Araraat wasn´t there on holiday?"

Oh, so which Mt Ararat are you refering to? Urartu or Eridu? Or another one. To me Eridu makes the most sense. After all it's on a natural rise above the Mesopotamian floodplain, and the earliest temple in the region (where fish were ritually eaten on religious occasions) is found there. So, although Pconroy mentions many floods, the one that has come down to us through the Old Testament myth is likley to be one people survived on the higher ground at Eridu. Fits Gilgamesh as well.

zadeh79 said...
This comment has been removed by the author.
Maju said...

Afghanistan? I know of no single genetic study in that country - but of course I am very interested, can you direct me to such study? That is, assuming it does exist and is not an extrapolation you make from Pakistanis.