PDA

View Full Version : the western refugia Palaeolithic R1b concept that just will not go away



alan
09-06-2013, 08:48 AM
I see Maju still is a believer in R1b 'hid' in the west during the Neolithic

http://forwhattheywereweare.blogspot.co.uk/2013/09/hungarian-ancient-dna-and-origins-of.html?showComment=1378252973909

Davidski seems to see R1 as doing a pincer movement in the copper age. However, he rightly raises the issue of why are the two groups moving into a Neolithic collapse vacuum be the two closely related ones. That would seem a weird coincidence to me. I can see how the idea of a pincer movement comes about but the idea of them being two suddenly revived palae groups at the opposite ends of Europe just doesnt work unless the dating is horribly out. After all most of this so-called pincer movement is L11.

Even multiplying it by the Zhiv fudge rates would not make it older than 12000BC or so which is nothing like as old as we would expect from a palae lineage trapped in a western refuge. I have no doubt that a three fold fudge is impossible anyway based on the age that would make haplogroup I. Based on the latter the largest fudge I think is possible is doubling. That would make L11 a maximum of about 7000BC and the eastern L23 clades a maximum of about 10000BC. Even that is totally incompatible with the refugia models and far closer to the Neolithic spread. There is no pre-farming spread west noted in that timeframe. So, I do not really see why this palae R1b model still exists. It just doesnt make sense even when stretching to doubling the germline rates. Indeed, even if the age of P312 was trebled along Zhiv lines it still would only seem to come in with P312 about 11500BC and eastern L23XL51 clades coming in around 16000BC. Again this is totally incompatible with the refugia theory. The very latest possible east to west link in the palaeolithic that could have contributed to a western refuge is the Badegoulian c. 20000BC and even that is controversial. That would require a nearly 5 times fudge factor to make any sense of. That of course seems totally impossible as it would make the age of R as a whole impossibly early - over 100,000 years old.

So, in summary, there seems to be a basic logic problem with seeing palaelithic western continuity in R1b and it is surprising to see it still had adherents. I would not be totally shocked if it was found somewhere in the Neolithic though. The evidence to date is against this but it at least seems theoretically possible within the bounds of a viable fudge. That does not appear to be the case for a paleolithic western model.

One possible very outside bet within the double fudge factor range that could explain that sort of age, a distant link with the east but an absence in LBK is if R1b was somehow related to the strange group of late hunters, some adopting farming ideas, and pointed based pottery. It seems unlikely to me but I noticed there has been a very recent revival in the idea that the various groups with this type of pottery noted from France to Russia could be linked to an east-west movement just ahead of farming in some places and overlapping with it in others and linked to the emerging east-west north Eurasian diffusion of early pottery. There was a recent conference paper on this, I think by a Russian, suggesting some sort of pre-farming pointed based pottery wave from Russia spreading across Europe ahead of the SW Asian groups. However, I just cannot seem to find it again on the web. This is not the paper I had in mind but discusses the separate spread of pottery from farming in Eurasia

http://www.academia.edu/3754905/Gibbs_K_and_P._Jordan._2013._Bridging_the_Boreal_F orest_Siberian_Archaeology_and_the_Emergence_of_Po ttery_among_Prehistoric_Hunter-Gatherers_of_Northern_Eurasia._Sibirica_12_1_Sprin g_2013_1-38

I suppose in theory diffusion like that could have involved diffusion of some people associated with cermanic using hunters-gatherers. I am not convinced at all but it does provide a later hunter background, an east-west spread, a link back towards central asia and an arrival beyond eastern parts of Europe relatively late when some areas were already farming. This at least provides some echoes of some of the features of R1b phylogeny and delayed expansion in such a scenario. However, it is very low on clearcut evidence.

Mikewww
09-06-2013, 02:07 PM
I didn't really find those blogs helpful.

Am I missing something or are some of these bloggers trying to correlate Y DNA with autosomal and mt DNA mixed with either pro or con positions on migration versus continuity?

We've looked at that on other threads and I don't think we can correlate R1b with autosomal or mt DNA in any kind of consistent way.

Nothing is impossible, but at least as it pertains to R1b, I think it takes some contortions to try to reason that R1b in Europe has a Paleolithic inhabitant. I can only think of it as being mesmerized by current high frequency distributions with a belief in continuity.

GoldenHind
09-06-2013, 02:36 PM
Some poorly informed person, probably in marketing (sorry if that is redundant), at BritainsDNA lists S116* (P312*) as "hunter-gatherer."

alan
09-06-2013, 03:53 PM
I found the paper I couldnt get my hands on earlier - the one that speculated that there was a non-farming 'Neolithic' wave from perhaps the Urals area earlier in some areas than farming. In eastern terminology the Neolithic just means pottery using whereas in the west we use the word to mean early farmers. Its one of those mathematical models that I am generally wary of. Nevertheless its worth discussing even if only to put the idea down.

http://arheologija.ff.uni-lj.si/documenta/pdf34/DPdavison34.pdf

Although most of these ceramic using hunter groups are from eastern and northern Europe, they do also mention similar groups in the west like La Hoguette and Roucadour in France. Others with similar ideas also discuss Limberg and Dutch and Belgian groups like Swifterbant. My feeling is that its easy in soft focus to point to a bunch of pottery using hunters and link the dots but the specifics make this unconvincing. On the other hand the idea of pottery spreading by diffusion among hunters from east to west well ahead of farming in its more eastern zone is accepted but in western Europe its more complex because it overlaps with the early farming period and could be a borrowing from the farmers. The pointed base pottery seems distinctive on the surface of it all but it has been argued in terms of Swifterbant that that is a minority shape and has a functional role.

I only bring all of this up because these kind of arguements involving a non-farming east to west spread will probably be used as the last refuge of the pre-farming R1b spread school of thought. I am not supporting the idea.



I see Maju still is a believer in R1b 'hid' in the west during the Neolithic

http://forwhattheywereweare.blogspot.co.uk/2013/09/hungarian-ancient-dna-and-origins-of.html?showComment=1378252973909

Davidski seems to see R1 as doing a pincer movement in the copper age. However, he rightly raises the issue of why are the two groups moving into a Neolithic collapse vacuum be the two closely related ones. That would seem a weird coincidence to me. I can see how the idea of a pincer movement comes about but the idea of them being two suddenly revived palae groups at the opposite ends of Europe just doesnt work unless the dating is horribly out. After all most of this so-called pincer movement is L11.

Even multiplying it by the Zhiv fudge rates would not make it older than 12000BC or so which is nothing like as old as we would expect from a palae lineage trapped in a western refuge. I have no doubt that a three fold fudge is impossible anyway based on the age that would make haplogroup I. Based on the latter the largest fudge I think is possible is doubling. That would make L11 a maximum of about 7000BC and the eastern L23 clades a maximum of about 10000BC. Even that is totally incompatible with the refugia models and far closer to the Neolithic spread. There is no pre-farming spread west noted in that timeframe. So, I do not really see why this palae R1b model still exists. It just doesnt make sense even when stretching to doubling the germline rates. Indeed, even if the age of P312 was trebled along Zhiv lines it still would only seem to come in with P312 about 11500BC and eastern L23XL51 clades coming in around 16000BC. Again this is totally incompatible with the refugia theory. The very latest possible east to west link in the palaeolithic that could have contributed to a western refuge is the Badegoulian c. 20000BC and even that is controversial. That would require a nearly 5 times fudge factor to make any sense of. That of course seems totally impossible as it would make the age of R as a whole impossibly early - over 100,000 years old.

So, in summary, there seems to be a basic logic problem with seeing palaelithic western continuity in R1b and it is surprising to see it still had adherents. I would not be totally shocked if it was found somewhere in the Neolithic though. The evidence to date is against this but it at least seems theoretically possible within the bounds of a viable fudge. That does not appear to be the case for a paleolithic western model.

One possible very outside bet within the double fudge factor range that could explain that sort of age, a distant link with the east but an absence in LBK is if R1b was somehow related to the strange group of late hunters, some adopting farming ideas, and pointed based pottery. It seems unlikely to me but I noticed there has been a very recent revival in the idea that the various groups with this type of pottery noted from France to Russia could be linked to an east-west movement just ahead of farming in some places and overlapping with it in others and linked to the emerging east-west north Eurasian diffusion of early pottery. There was a recent conference paper on this, I think by a Russian, suggesting some sort of pre-farming pointed based pottery wave from Russia spreading across Europe ahead of the SW Asian groups. However, I just cannot seem to find it again on the web. This is not the paper I had in mind but discusses the separate spread of pottery from farming in Eurasia

http://www.academia.edu/3754905/Gibbs_K_and_P._Jordan._2013._Bridging_the_Boreal_F orest_Siberian_Archaeology_and_the_Emergence_of_Po ttery_among_Prehistoric_Hunter-Gatherers_of_Northern_Eurasia._Sibirica_12_1_Sprin g_2013_1-38

I suppose in theory diffusion like that could have involved diffusion of some people associated with cermanic using hunters-gatherers. I am not convinced at all but it does provide a later hunter background, an east-west spread, a link back towards central asia and an arrival beyond eastern parts of Europe relatively late when some areas were already farming. This at least provides some echoes of some of the features of R1b phylogeny and delayed expansion in such a scenario. However, it is very low on clearcut evidence.

alan
09-06-2013, 04:28 PM
I agree Mike. However I think the spread of eastern European connections among the late hunters through the the medium of pottery among hunters will be their final fallback position even though the extending of this to late hunters in western Europe beyond the north European plain is much more questionable due to the fact farmers had arrived nearby those groups by the time the pottery becomes apparent there. Certainly, no matter how one dresses it up such a fall back position, it would still essentially require an east-west spread in the early Neolithic period with the only difference being that it was from the east European plain and involved pottery using late hunters instead of SW Asia involving farmers.

I suspect a line of arguement might go that the phylogeny of R1b in a latitudinal sense would broadly fit the spread of this late hunter pottery from the east of Russia from c. 15000BC to western Europe around 5000BC.

www.ffzg.unizg.hr/arheo/ska/tekstovi/neolithic_dispersal.pdf‎

I dont think it works when you look at the big differences among the pottery using hunters but I suspect that line of reasoning might be a fall back position for those who are allergic to the idea of farmers or copper workers. The fact that one of the groups sometimes mentioned is in the Pyrenees will be another attraction.


I didn't really find those blogs helpful.

Am I missing something or are some of these bloggers trying to correlate Y DNA with autosomal and mt DNA mixed with either pro or con positions on migration versus continuity?

We've looked at that on other threads and I don't think we can correlate R1b with autosomal or mt DNA in any kind of consistent way.

Nothing is impossible, but at least as it pertains to R1b, I think it takes some contortions to try to reason that R1b in Europe has a Paleolithic inhabitant. I can only think of it as being mesmerized by current high frequency distributions with a belief in continuity.

R.Rocca
09-06-2013, 05:46 PM
The biggest hurdle against a Palaeolithic presence for R1b in Western Europe is the same today as it has been for the last five years - ancient DNA simply does not support it. Even if not all 'refigia' are tested for, you would at least expect some of it to have trickled into other cultures and have shown up by now.

Curious
09-06-2013, 05:50 PM
As a new poster, I hope I won't be accused of hijacking the thread by returning to the concept of when R1b entered those areas where it predominates today (Britain, Ireland and the Basque country of Spain and France). I've read the genetic arguments as to why R1b is more recent than had been assumed, and why it must have originated further east, but I've also noticed that the field of DNA research is evolving rapidly, and what was gospel a few years ago is overturned now. So I'd be more convinced by the comments about lack of ancient R1b in those areas where it's strongest now, except for the fact that there don't seem to be any Y DNA samples prior to 4500 BP for Britain, Ireland or the Basque country. None, nada. I find a claim that R1b wasn't present to be unconvincing when there's no evidence one way or the other. Or are folks treating all of Western Europe as a unity and assuming that Y DNA profiles for Germany, eastern France or eastern Spain can be taken as evidence for Ireland, Britain and the Basque country prior to 4500 BP, despite the current differences in DNA signatures? If so, is the R1b find for Germany proof that R1b was present in Britain, Ireland or the Basque country? Or can we perhaps stop referring to lack of any evidence about Y DNA in a particular as proof of the absence of R1b in that area and admit that the evidence is solely genetic? If so, I'll have to admit to being at a disadvantage in the argument, since I'm not a geneticist. Perhaps the current models are correct, but I won't be convinced until I see evidence as to what lineages actually were anciently present in those areas where R1b now predominates.

alan
09-06-2013, 06:24 PM
There is more to it than that. The phylogeny or chain of SNPs makes it essentially impossible for western R1b clades like DF27, L21 etc to have come from anywhere other than the east end of Europe or nearby areas of Asia and those clades are WAY downstream from those in somewhere like Iran etc. Counting SNPs or calculating STR variance indicates that the western R1b clades are apparently only about a quarter of the total age of R1. The age of R1 is currently placed about 18,000 years ago - a quarter of that is 4500 years ago or 2500BC. Now even if the anchor date for R1 is too young it cannot be older than the likely date of the permanent settlement of modern humans into central Asia/Iran and the western European clades cannot be more than a quarter of that. So its incredibly unlikely that the young clades of R1b found in western Europe could be much older than 8000 years old.

So, at an outside stretch it is theoretically possible that R1b could have spread west around then with early Neolithic peoples but the western ice age hunter-gatherer refugia model appears to be mathematically impossible. Although the hunter-gatherer Franco-Cantabrian ice age refugia option appears dead and an association with the first farmers to reach the west looks unlikely, I do not think everyone has totally written off the possibility of something like a middle Neolithic movement from the east c. 5000BC. The copper age model is the favourite at present but we are all aware there is enough doubt on absolute dating to make ruling out something like a spread west in the middle Neolithic impossible.


As a new poster, I hope I won't be accused of hijacking the thread by returning to the concept of when R1b entered those areas where it predominates today (Britain, Ireland and the Basque country of Spain and France). I've read the genetic arguments as to why R1b is more recent than had been assumed, and why it must have originated further east, but I've also noticed that the field of DNA research is evolving rapidly, and what was gospel a few years ago is overturned now. So I'd be more convinced by the comments about lack of ancient R1b in those areas where it's strongest now, except for the fact that there don't seem to be any Y DNA samples prior to 4500 BP for Britain, Ireland or the Basque country. None, nada. I find a claim that R1b wasn't present to be unconvincing when there's no evidence one way or the other. Or are folks treating all of Western Europe as a unity and assuming that Y DNA profiles for Germany, eastern France or eastern Spain can be taken as evidence for Ireland, Britain and the Basque country prior to 4500 BP, despite the current differences in DNA signatures? If so, is the R1b find for Germany proof that R1b was present in Britain, Ireland or the Basque country? Or can we perhaps stop referring to lack of any evidence about Y DNA in a particular as proof of the absence of R1b in that area and admit that the evidence is solely genetic? If so, I'll have to admit to being at a disadvantage in the argument, since I'm not a geneticist. Perhaps the current models are correct, but I won't be convinced until I see evidence as to what lineages actually were anciently present in those areas where R1b now predominates.

Curious
09-06-2013, 07:49 PM
I've already admitted that I don't know enough about genetics to argue about when R1b evolved or where it appears to have evolved, alan, although I did read something recently about some geneticists questioning whether STR variances are as good a predictor as they had assumed. So, as I inferred, I'll trust the genetic argument about R1b if it holds up over time. I was just trying to cure a few folks from adding, at the end of a learned dissertation about genetics and R1b, a comment along the lines of "and if R1b was ancient in Atlantic Europe, why isn't there any evidence that it was there along with other Y haplotypes". At this point, there doesn't seem to be any evidence one way or the other about any Y haplotype in Atlantic Europe prior to the IE expansion. Since the genetic arguments do sound convincing, as far as I can tell, it may well be that when enough testing of old bones is done we'll find that R1b was indeed lacking in Paleolithic and Neolithic Atlantic Europe. But it appears that the testing hasn't been done yet.

Anglecynn
09-06-2013, 08:00 PM
I didn't really find those blogs helpful.

Am I missing something or are some of these bloggers trying to correlate Y DNA with autosomal and mt DNA mixed with either pro or con positions on migration versus continuity?

We've looked at that on other threads and I don't think we can correlate R1b with autosomal or mt DNA in any kind of consistent way.

Nothing is impossible, but at least as it pertains to R1b, I think it takes some contortions to try to reason that R1b in Europe has a Paleolithic inhabitant. I can only think of it as being mesmerized by current high frequency distributions with a belief in continuity.

This is true. Although R1b does kinda correlate with Atlantic/Western European components that show up as the major component up most of the Atlantic fringe, and also with the so called 'Gedrosian' components - That is - relatively speaking - quite high in places very high in R1b like Ireland. And also that the western European autosomal cluster is more closely related to the eastern European one than the Mediterranean one. So it looks like at this component has spread from east to west it has gradually changed through contact with one that has spread from the near east to the Mediterranean, which happens to coincide with the westernmost areas of Europe. Although it is far from a solid connection i think it is a hint anyway.

alan
09-06-2013, 08:38 PM
These autosomal clusters are composites though and could be hugely misleading. I dont take them very serious at the moment. Certainly not until some sort of consensus on how it divides up.

alan
09-06-2013, 08:43 PM
Funny thing is I actually think that is the weakest of the many arguements against early R1b in the west, be it palaeolithic or early Neolithic. I dont think the ancient DNA sample is spread around enough or big enough to be sure about much yet. However, on many other grounds I dont think a pre-Neolithic R1b presence in the west is possible.


The biggest hurdle against a Palaeolithic presence for R1b in Western Europe is the same today as it has been for the last five years - ancient DNA simply does not support it. Even if not all 'refigia' are tested for, you would at least expect some of it to have trickled into other cultures and have shown up by now.

alan
09-06-2013, 08:49 PM
I agree that no case can be closed officially until a decent ancient DNA database exists but I think its practically impossible that R1b was pre-farming in the west. Other than that no other options come so close to being ruled out and several options remain until a fair bit more ancient DNA appears. There is a lot of paradox in modern distributions.



I've already admitted that I don't know enough about genetics to argue about when R1b evolved or where it appears to have evolved, alan, although I did read something recently about some geneticists questioning whether STR variances are as good a predictor as they had assumed. So, as I inferred, I'll trust the genetic argument about R1b if it holds up over time. I was just trying to cure a few folks from adding, at the end of a learned dissertation about genetics and R1b, a comment along the lines of "and if R1b was ancient in Atlantic Europe, why isn't there any evidence that it was there along with other Y haplotypes". At this point, there doesn't seem to be any evidence one way or the other about any Y haplotype in Atlantic Europe prior to the IE expansion. Since the genetic arguments do sound convincing, as far as I can tell, it may well be that when enough testing of old bones is done we'll find that R1b was indeed lacking in Paleolithic and Neolithic Atlantic Europe. But it appears that the testing hasn't been done yet.

Anglecynn
09-06-2013, 10:21 PM
These autosomal clusters are composites though and could be hugely misleading. I dont take them very serious at the moment. Certainly not until some sort of consensus on how it divides up.

Most of them are composites. For example the main autosomal cluster that dominates in places like Ireland is a slightly more Mediterranean version of the main cluster that dominates in places like northern Germany. And it's more the combination of these different components and their amounts that tells you how certain countries relate to each other. Using Ireland as an example again, the main thing that links it to other places very high in R1b in Europe is this western component (and sometimes the Gedrosian component), but in all other respects they are overall more genetically similar to a country that has maybe just 40% R1b, for example. So it looks like R1b left a clear autosomal trace, but that it is definitely not the dominant one.

I don't know, all just speculation. But yeah i would maintain what Mike already said - That you can't really assign an autosomal component to a haplogroup. I think it is just a matter of finding the best fit, although that is more a matter of finding evidence to fit, rather than actual evidence.

Overall though, autosomal genetics can be pretty reliable. If it can take a bunch of data, compare it to another bunch of data and tell you what country you are from accurately, i'm pretty convinced it's useful. (And it can do that)

alan
09-06-2013, 11:41 PM
Yes I would agree with a lot of what you posted there. I do suspect though that Ireland's (and probably also Wales and Scotland's) huge R1b and L21 total owes an awful lot to a few large clusters associated with Medieval chiefs and their lineages. I suspect Ireland was no so different from the rest 2000 years ago. The main difference is that in the Celtic fringe (and you could also say this for the basques) the chiefs lineages have run riot because they were not interupted by later invaders and have simply multiplied up over the last 2000 years whereas those elites were disinfranchised by Romans, Germanics etc elsewhere much earlier.

I recall seeing that the main strong autosomal division in Europe is north-south rather than east-west. Consequenly the Irish, British, Germans, Poles, some Scandis etc cluster close together and are much more remote from the Med. I suspect that is to do with the entire north having a lot more pre-farming genes than the Med. areas. As you noted, the main difference is that there is a little more Neolithic farmer blood in western Europe than among the Slavs but they are still much closer to each other than the Med. groups. The British autosomal maps posted recently are getting down to minutae of differences between peoples who are actually very close on a European scale.


Most of them are composites. For example the main autosomal cluster that dominates in places like Ireland is a slightly more Mediterranean version of the main cluster that dominates in places like northern Germany. And it's more the combination of these different components and their amounts that tells you how certain countries relate to each other. Using Ireland as an example again, the main thing that links it to other places very high in R1b in Europe is this western component (and sometimes the Gedrosian component), but in all other respects they are overall more genetically similar to a country that has maybe just 40% R1b, for example. So it looks like R1b left a clear autosomal trace, but that it is definitely not the dominant one.

I don't know, all just speculation. But yeah i would maintain what Mike already said - That you can't really assign an autosomal component to a haplogroup. I think it is just a matter of finding the best fit, although that is more a matter of finding evidence to fit, rather than actual evidence.

Overall though, autosomal genetics can be pretty reliable. If it can take a bunch of data, compare it to another bunch of data and tell you what country you are from accurately, i'm pretty convinced it's useful. (And it can do that)

alan
09-06-2013, 11:46 PM
I forgot to say that my own feeling is that R1b was a very modest contrubutor of autosomal DNA by the time it reached the Atlantic. I generally feel that the autosomal mix in places like Ireland was largely set in place by the Meso-Neolithic mix and generally speaking I think the proportions of those two things is what makes up most of the variation in Europe other than places that were really on the highway of movements and the empires from Asia in later times.


Most of them are composites. For example the main autosomal cluster that dominates in places like Ireland is a slightly more Mediterranean version of the main cluster that dominates in places like northern Germany. And it's more the combination of these different components and their amounts that tells you how certain countries relate to each other. Using Ireland as an example again, the main thing that links it to other places very high in R1b in Europe is this western component (and sometimes the Gedrosian component), but in all other respects they are overall more genetically similar to a country that has maybe just 40% R1b, for example. So it looks like R1b left a clear autosomal trace, but that it is definitely not the dominant one.

I don't know, all just speculation. But yeah i would maintain what Mike already said - That you can't really assign an autosomal component to a haplogroup. I think it is just a matter of finding the best fit, although that is more a matter of finding evidence to fit, rather than actual evidence.

Overall though, autosomal genetics can be pretty reliable. If it can take a bunch of data, compare it to another bunch of data and tell you what country you are from accurately, i'm pretty convinced it's useful. (And it can do that)

Curious
09-08-2013, 03:34 PM
There's something in BBC News about 10,000 year old human bones from Cumbria. Now, if only they'd try to extract some Y dna. To be honest, I'd expect something like I or N rather than R1b - all this talk about counting SNPs and calculating STR variances is starting to convince me, even though I don't really understand it. But it would be nice to have at least one actual example of early Y dna from the Atlantic Fringe.