PDA

View Full Version : western steppes blocked from both east steppes and SW Asia in upper Palaeolithic



alan
07-14-2013, 08:33 PM
I was reading recently about the khvalynian Sea - the hugely extended Caspian of c. 15000 to 9000BC. It clearly would have been a major barrier between the western steppes and central Asia and must have had a major impact on cultural, genetic and linguistic development. I was going to post on this before but I forgot. I was reminded as I am re-reading Anthony (I wanted to read as many papers by other authors before re-reading it) and he mentions it too and makes the same point. This is a recent paper on the subject

http://ubuntuone.com/p/19zG/

In addition a spillway linked the Caspian and Black Sea and blocked off the Caucasus route south at the same time. Indeed it formed part of a chain of linked seas from the Black to the Aral Sea in this period.

That basically boxed the western steppe zone from movement from both the east and south througout a significant chunk of the upper-Palaeolithic until the period that farming started to rise in SW Asia. That makes the choice of R1b location of being either north or south of this east-west chain of seas (Black-Azov-Caspian-Aral) even more stark. Wherever it was in 15000BC it would still have been in 10000BC. If this is the case then this barrier could be as old as R1b itself (Karafet puts R1 at 16500BC) and stood almost from the time of origin of R1b to the time P297 came about. Interestingly the article states that upper Palaeolitic settlement of modern humans was even delayed by this barrier from penetrating some areas like the Caucasus and central Asia until after 10000BC. The article later tries to narrow the period somewhat to a point commencing somewhat later.

Regardless this appears to mean that any movement south as per regugia models would have had to taken place before this event or would not have been possible until nearly the start of the Neolithic in SW Asia. It makes the choice of locations for where R1b overwintered very very stark. It either had to be among the upper Palaeolithic groups on the Ukraine steppe area (or westwards) OR in SW Asia.

alan
07-14-2013, 09:33 PM
The highest confidence period for this phenomenon c. 1200BC-9000BC still would predate P297.

Rathna
07-15-2013, 01:02 AM
It makes the choice of locations for where R1b overwintered very very stark. It either had to be among the upper Palaeolithic groups on the Ukraine steppe area (or westwards) OR in SW Asia.

Westward, as I think having demonstrated ad abundantiam. And not only R1b but also R1a-M420 and subclades, not known in Asia.
I thank you for this posting, Alan.

AJL
07-15-2013, 01:56 AM
R1b but also R1a-M420 and subclades, not known in Asia.

On the contrary. (Underhill 2009)

DMXX
07-15-2013, 02:05 AM
Westward, as I think having demonstrated ad abundantiam. And not only R1b but also R1a-M420 and subclades, not known in Asia.
I thank you for this posting, Alan.

I see samples with...

R*-M207
R1*-M173
R1a1*-SRY1532 (as AJL pointed out, we know there's R1a*-M420 too in Iran in Underhill et al.)
R1b*-M343
R1b1a1-M73
R1b1a2*-M269
R1b1a2a*-L23
R1b1a2a1a-M412

How can we have proper debates when people aren't even aware of the data which opposes their arguments? I haven't seen such a chronological breakdown of R subclades anywhere outside of West Asia.

http://2.bp.blogspot.com/-_mypYjUAwqw/UAjZdNrHAqI/AAAAAAAAAMY/nxbhHviubow/s1600/study4.png

Rathna
07-15-2013, 02:31 AM
I see samples with...

R*-M207
R1*-M173
R1a1*-SRY1532 (as AJL pointed out, we know there's R1a*-M420 too in Iran in Underhill et al.)
R1b*-M343
R1b1a1-M73
R1b1a2*-M269
R1b1a2a*-L23
R1b1a2a1a-M412

How can we have proper debates when people aren't even aware of the data which opposes their arguments? I haven't seen such a chronological breakdown of R subclades anywhere outside of West Asia.


DMXX, you can see that Iran gets pretty all the haplogroups. The solutions are two: either it is the origin of everything (and it is a possibility) or it was a crossroad where haplogroups from East and West encountered.

About R1a-M420 (I don't see clearly in your map, where it seems R1a1) is documented by some paper and I spoke about it in the past, but unfortunately we haven't their STRs to compare with the Western European ones documented largely by SNPs and STRs, and I hypothesized that they had come from the Russian plane and we could answer this only by their STRs or deepen SNPs tests. But what I think is important it is that it has never been found in India, and if it were present in Iran from its beginning perhaps we should have found there too.
We are searching for truth at a thin scale. See what I said about the R1b1 found in Central Asia and about Indian Raza.

DMXX
07-15-2013, 02:58 AM
That's a fair point, but we don't even see this sort of R subclade diversity in Central Asia. Things over there seem to be a couple steps upstream with regard to para-haplogroups. You do see R* there as well, but also things like R2*-M479, which I personally haven't seen in Iran.

That trend has been there since the mid-2000's and hasn't changed with more data. Iran most certainly did serve as a crossroads, you are absolutely correct on that point, but we have to remember it's a mountainous country with all sorts of isolated pockets where older layers of demography can remain. This was the means by which the Zoroastrians maintained their faith following the spread of Islam.

Your point of a crossroads definitely applies to the larger cities (such as Esfahan and Shiraz), but with somewhat endogamous groups resident in Iran (Assyrians, Gilakis) the strength of that point greatly reduces, particularly in historical times. We can only speculate as to which lineages had "swept through" Iran prior to this.

Rathna
07-15-2013, 04:10 AM
That's a fair point, but we don't even see this sort of R subclade diversity in Central Asia. Things over there seem to be a couple steps upstream with regard to para-haplogroups. You do see R* there as well, but also things like R2*-M479, which I personally haven't seen in Iran.

That trend has been there since the mid-2000's and hasn't changed with more data. Iran most certainly did serve as a crossroads, you are absolutely correct on that point, but we have to remember it's a mountainous country with all sorts of isolated pockets where older layers of demography can remain. This was the means by which the Zoroastrians maintained their faith following the spread of Islam.

Your point of a crossroads definitely applies to the larger cities (such as Esfahan and Shiraz), but with somewhat endogamous groups resident in Iran (Assyrians, Gilakis) the strength of that point greatly reduces, particularly in historical times. We can only speculate as to which lineages had "swept through" Iran prior to this.

In fact we now know that pretty all the R1* found in Asia are R2-M479*. Perhaps you know that VV, who probably was the strongest opposer of my theories (and he is of Italian descent!), founded an Haplogroup R1 Project at FTDNA, thinking that this validated his theories and were against mine or others. That project failed with the discovery of M479.
Of course Iran is very important at a genetic level and many areas are very conservative, but if you give a glance to the last paper of Behar et alii about the rare haplogroup (mtDNA) N1a3 and N3, I have spoken about also here in this forum, you can see how can we understand the matter: hg. N3 is certainly born in Iran and expanded from there, but it is found rarely elsewhere, and this demonstrates that the expansion from Iran was little in these last 10,000 years, even though it seems that agriculture was born on the Zagros mountains.
I take this declaration of Behar (et alii) like a palinode of all his reasearch for demonstrating that hg. mtDNA H, K, R0a etc are of Near East origin: perhaps you know that I have written a lot about their ancient presence in Europe, and also in this case it isn't said that these haplogroups weren't present in both places from ancient times.
Certainly the R1b1 of Raza is in India from its beginning like haplogroup, only that it couldn't be the ancestor of R-V88+ and R-L389+, for lacking YSC224. We'll see from the next tests if things will change.

alan
07-15-2013, 11:38 AM
I had been aware of a longterm block that the massively expanded Caspian Sea had c. 15000BC to 9000BC (give or take a millenium-dates vary depending on who you read) but the block to north-south relations caused by the chain of inland seas from the Black to the Aral hadnt fully registered. I suppose I always thought that the Caucasus or routes down the side of the west shore of the Caspian/east shore of the Black Sea remained open. However forr part of this period (again this varies depending on who you read) the Black Sea and Caspian were also linked. We are not talking a river but something extremely wide and deep. Clearly it has implications for R1b although more certainty of the duration of the Black-Caspian link would be helpful.

Certainly I think its fair to say that in east-west terms the Caspians massive extension (especially to the north) meant a profound cultural and probably genetic barrier was established not long after Karafets date for R1 as a whole. So I think its very likely IF R1 was located north of the Caucusas in the steppe area then it must have been on the west side. I dont think too many people would see R1 as waiting until after 9000BC to cross west of the Caspain and archaeology certainly finds no support for a movement like that. I have an open mind on this though.

The more tricky question is the north-south aspect of this. The Black Sea-Caspian period of connection may not have been as lengthy as this but I am not confident from the report that this is very well established. If this barrier was only present for the more limited time around say 12-9000BC then parts of R1a and R1b (presumably P25*) could have had found themselves partly on both sides of this north-south divide. That would have created divergent lines seperated from each other for at least a few millenia. This of course could have effected R1a as much as R1b. So much depends on the exact dating of the north-south aspect of the water barrier.

It is curious that P297 dates to around the period this abated although we also know that this line did next to northing for the next 5 or 6 millenia given the lack of P297* until M73 and then M269 arose. That is for me the strongest arguement that R1b was stranded on the north side of this boundary rather than in SW Asia. The report seems to rule out an intermediate position in the Caucasus or central Asia pre-9000BC as it observes a lack of upper Palaeloithic settlement in the area which it attributes to these water barriers (presumably combined with the mountain barriers which much have been impassible in the Upper Palaeolithic.

Nothing is certain but I think its clearly something that needs to be factored in in terms of possible hiding places for R1b (and R1a) in its early phases.

alan
07-15-2013, 01:51 PM
Its taken a bit of digging to find something of a general guide to late upper Palaeolithic in eastern Europe. This is not exactly bang up to date but its not ancient and the best overview I could find free on the net so far

http://books.google.co.uk/books?id=nXuqgInMOXIC&pg=PA228&lpg=PA228&dq=eastern+epigravettian&source=bl&ots=UmwiLzRCAE&sig=vM6_RX1hS0z_aDTO5zy6phOzo50&hl=en&sa=X&ei=YfvjUcJPiPDSBYTZgfAJ&ved=0CFoQ6AEwBw#v=onepage&q=eastern%20epigravettian&f=false

If the logic of the previous paper I posted on the watery barriers of the extended Caspian to east-west movement and north-south too is followed and its observation of a lack of convincing pre-10000BC upper Palaelithic settlement of modern humans in the area between the Black Sea and Caspian around Caucasus and central Asia is taken into account along with DNA dating/branching then it would make the late eastern 'epigravettian' (considered an unsatisfactory term) groups of eastern Europe one of the most likely homes for R1 and earlier forms of R1a and R1b. It would be nice to get a handle on the Ice Age position of our ancestors. I am not saying its the only option but its a very strong one, all things considered. The only alternative would be a position in SW Asia but that does not fit the lack of branching or the lack of ancient DNA evidence for involvement in early farming.

alan
07-15-2013, 05:36 PM
This paper discusses the continuing lack of Upper Palaeolithic sites in NW and wider northern Iran in its end summary and notes it is not until the epi-Palaeolithic (close to the Neolithic) that sites suddenly become common

inhttp://academia.edu/1583906/Biglari_F._and_V._Jahani_2011_The_Pleistocene_Huma n_Settlement_in_Gilan_Southwest_Caspian_Sea_Recent _Research_Eurasian_Prehistory_8_1-2_3-28 the area.

This is in line with the observations of the paper that promted this thread. That is further food for thought about the possible location of R1b before c. 10000BC. Seems like refugia in north Iran is less likely than one would have thought from the climate/environment alone.

As noted above the eastern gravettian and epigravettian people were pretty hardy sorts who seem to have adapted to pretty grim conditions. As far as I can see the most central dates for the blocking of the southerly escape route between the Caspian and Black Sea place it more centred about 12000-10000BC give or take a few centuries so it appears not moving south would have been a choice rather than a neccessity. However, the devil is in the detail and I imagine the Caucasus would have been a horrible barrier in the Upper Palaelithic as would that northern ridge in Iran along the Caspain and the north end of the Zagros. South may have looked pretty horrible standing at the northern steppe-north Caucasus interface at the LGM and may have looked like towering walls of ice with a lot less hunting options. How would they know that there was a different world in the SW of Asia beyond this barrier? Then for a time just when the weather improved for a couple of millenia, melt waters meant the Caspian expanded hugely and the Caspian-Black Sea link removed the option of moving south. The timing of the appearance of the epi-palaeoithic sites in the central and east Caucasus, north Iran and central Asia seems to coincide with the final removal of that barrier. Maybe it also was linked to the Younger Dryas around the same time. However, its hard not to conclude that these areas are much less likely to be refugia for R1b than the Ukraine.

J Man
07-15-2013, 06:14 PM
Is it not possible that R1b had a refuge in other parts of the Near East during the Upper Paleolithic and Mesolithic? That is in Anatolia and the Levant possibly? Don't most people today seem to think R1b spread into Europe with Neolithic farmers originating in the Near East?

alan
07-15-2013, 06:44 PM
Is it not possible that R1b had a refuge in other parts of the Near East during the Upper Paleolithic and Mesolithic? That is in Anatolia and the Levant possibly? Don't most people today seem to think R1b spread into Europe with Neolithic farmers originating in the Near East?

I was really looking at the pre-Neolithic location of R1b. The dates of the main European clades are post-date 3500BC which is about 3000 years too late for Neolithic farmers. Ancient DNA has shown an absence of R1b in Europe so far. In general R1b of the type found in the farming world of Neolithic doesnt branch and multiply in any significant way until 3500BC and beyond eastern Europe not much until after 3000BC. A location in the heart of the Neolithic farming world in Mesopotamia and Anatolia, the Levant etc just does fit R1b at all. In many ways R1b and R1a resemble each other in the age of their branching and absense in Neolithic Europe. Some people try and make these clades contrast too much when in fact their early stages look very similar.

Eliminating the location of R1b from SW Asia, Anatolia, northern Iran, the south Caucasus and central Asia and the Asian steppe before 10000BC only seems to leave a location in the epigravettian groups around the north and west shores of the Black Sea looking feasible.

All of this of course is dependent on the variance based age estimates being in the right ballpark but if they are then R1b looks to have been off-scene and somewhere north and east of the farming revolution.

J Man
07-15-2013, 06:55 PM
^ I suppose we will need ancient DNA results to prove or disprove this of course just like anything else.

alan
07-15-2013, 07:01 PM
^ I suppose we will need ancient DNA results to prove or disprove this of course just like anything else.

Very true. I think most people would find the biggest yDNA lineage in western Europe arriving as late as 3000BC or even later very hard to understand. However, it has been shown to be possible and that is the way the ancient DNA data is going at the moment. I think there is enough Neolithic yDNA already from various cultures to conclude the absence must be indicating real absense or rarity in that period in Europe. So far the ancient DNA and the dating suggested for R1b clades in Europe are in agreement. But yes we need more to be totally confident and there are a lot of cultures not sampled yet.