Monday, February 16, 2015

R1 and the Dispersal of Ceramic from the Far East

This map shows the spread of ceramic pottery1 from its birthplace in East Asia where it slowly treks to the West.

The test result of haplogroup [R1a in a Northwest Russian Karelian released this week] confirmed for me the probability that ceramic technology entered the West as a result of a population movement in the early Holocene.

Trek of Ceramic Technology, From Asia to the West (Jordan & Zvelebil, 2014)
This correlates well with the movement of a foreign knapping technology [here], solar religion and other materials.  Most importantly, the results of last week provide genetic evidence connecting ancient peoples who lived in the vicinity of the Siberian Pit and Combed Ceramic Tradition from Karelia (R1a) to Lake Baikal (R*).  Although the individuals of Mal'ta-Buret' were very early pre-ceramic people, the Lake Baikal area would eventually provide the precursor for the kind of pottery used by Karelia man.

In time it may be shown that the transmission of ceramic pottery technology to the West came as the result of a population movement of R1 lineages creeping from the Altai into the Urals and the Southern Caspian where the technology appears simultaneously.

Similarly, the transmission of [ceramic technology into the New World] probably comes from the Yenisei region in a somewhat later movement of R1's brother haplogroup Q's subclades moving in the opposite direction, both descended from P-M45.

Quickly, here's a some comments regarding ceramic's Central Asian trek by Jordan & Zvelebil:

"After c. 7,500 BC (9,500 BP), in the context of early post-glacial environmental conditions, pottery is dispersed further to the north-west, via the northerly route through central Russia, the Upper Volga, into Karelia and beyond, forming various local traditions of pointed-based pitted and combed ware, such as the Sperrings pottery of Finland, and entering the East Baltic and northern Scandinavia by about 5,000 BC"
"More tentatively, it is possible that as a part of this general process of ceramic dispersal, the production of pottery also spreads via a more southerly route from Central Asia along the eastern shore of the Caspian south, into northwest Iran and northern Syria."
Also from the Haak paper was the presence of R1b among an early Cardial farmer in Northern Iberia. This certainly shows that early R1b was in contact with either Byblian or Thessalonian farmers in the final days of the Pre-Pottery Neolithic B.  We know that the end of the PPNB saw sweeping changes from pottery cultures of Eastern Anatolia and further East.

*Update*  Added this map to help illustrate dates on the upper map (book is preview)

A couple of footnotes:

* We should be careful not to get wrapped around the axle of periodization and technological classification.  The Karelian and Samaran are referred to as belonging to the "Mesolithic" or "Neolithic" or "Hunter-Gatherer", none of which is equivalent those of Western Europe in terms of technology or age. Within a certain context they have meaning, otherwise they are confusing or misleading.  I would focus on calendar dates and understand the changes taking place in these regions that are being settled by a continuous stream of peoples from the Southeast.

(1)  Ceramic pottery is distinct from prehistoric ceramic technology.

Ceramics Before Farming:  The Dispersal of Pottery Among Prehistoric Eurasia Hunter-Gatherers. Left Coast Press.Jordan, Zvelebil (2009) [Link]


  1. If pottery arrived to the West (not just to Europe but also to West Asia) via Siberia, something I tentatively support, it did so with the migrations of proto-Uralic N1-carrier peoples. This reached Eastern Europe (archaeogenetically certified by mtDNA C presence) but how it reached West Asia and Greece is not yet explained, an might have well been via cultural diffusion with nearly no personal contact. In any case pottery arrived to "peninsular" Europe, with the rest of the Neolithic package, from Greece.

    Anyhow I see no logic in the obsession of attributing the origins of R, R1, R1a and R1b to Siberia: the geostructure strongly and repeatedly points to West Asia or (in the case of R and R1 probably Pakistan/India). Just because there's been some findings of R* in Siberia and peculiar sublineages of R1b in the Volga, findings that are mostly consistent with the modern geostructure, we cannot rush to argue anything, because most other regions (West Asia, South Asia, the Balcans and even Western Europe) remain largely unresearched. Lack of evidence is not evidence of lack.

    1. "Anyhow I see no logic in the obsession of attributing the origins of R, R1, R1a and R1b to Siberia:"

      I really don't view 'most' haplogroups as having a terminal end in a particular location, especially not this one. As far as its exact trek, I really can't say but I do think that of the lineages in West Eurasia this one most out of place. Sure, there is some low frequency Q or N, but R is by far the odd man out in a room of IJK, G, E, F &C.

      BTW, I think this jives just fine for SW Asia and the Northern Indus and can explain the antiquity of R here as well.

    2. R is not the odd man, P1 is. P1 includes Q, R and P1*, most common towards Bengal.

      As Karafet recently demonstrated, K2 (formerly K(xLT)) must have coalesced in Sundaland (modern day Western Indonesia and Malaysia) and expanded from there in three directions: NO to the North, P1 to the West (P(xP1) remained in SE Asia) and what is now called K2a (incl. M, S and other Oceanian K) to New Guinea.

      If we track P1, we find the greatest basal diversity in South Asia (P1*, Q and R) and West Asia, particularly around Iran (greatest basal diversity for Q and R1a). R1b has a more westerly origin, it seems, but still in West Asia (not sure right now if Levant or Anatolia is the most likely origin). So it seems very apparent that P1 migrated Westwards through Northern India and had a founder effect or several within the colonization of Western Eurasia in the early Upper Paleolithic. The fact that we P1* (of the pre-R branch but not directly ancestral to R but very derived already) 24 Ka ago in Central Siberia, coupled with the fact of Q1 (and mtDNA X2) having to have gone through that area some millennia earlier in order to make its founder effect in Native Americans, underlines that pre-R and Q1 were already living lineages before the LGM.

      When we observe the drift trees of the Mal'ta study and the Underhill study on R1, we observe that the drift of Ma1's pre-R seems to be more recent in time than the R1b-M269 node, although older than the R1b-M412 (Euro-clade) one. So R1b as such was already live some 24 Ka ago, probably only beginning to branch out but real. And that was in West Asia necessarily (basal diversity obligues).


    3. Why did P1 back-migrated to the West from SE Asia? What other evidence we have for it?

      I explain this in this entry:

      It is from before the Karafet findings (so K2 is called MNOPS) and the focus is mtDNA but what can be seen is that the patterns that we can detect in mtDNA for the Eurasian colonization are mirrored very well by Y-DNA. In fact the Y-DNA K2 expansion corresponds very well to the mtDNA N/R one, even if the N part is in some areas rather associated with C patrilineages.

      So basically H. sapiens spread towards East Asia via India and then something happened that made room for a secondary expansion from SE Asia. We can't be sure of what was it but the Toba catastrophe looks like a good candidate and the timing seems to fit well with my archaeologically calibrated estimates.

      So I estimate that c. 70 Ka BP K2 expanded from SE Asia in three directions: NO to the North, K2a to the SE and P1 to the West. P1 derivates (R and Q, and maybe some P1* occasionally found in Africa) therefore were in West Asia in time to take part in the early Upper Paleolithic (and maybe be a central force in its genesis). Most important founder effects were: Q1 in the proto-Amerindians (who brought UP/mode 4 to East Asia via the Siberian-Mongolian corridor) and R1b somewhere in West Asia, which at some point spread to various areas: Central Asia, North-East Africa and Europe. The other main patrilineage associated with this West Asia centered UP expansion is IJ: I in Europe, J1 in the Levant but also across the Red Sea in NE Africa and J2 in Highland West Asia.

      You mention also "G, E, F & C". The reality is that these lineages played in general just minor roles:
      → F* and C1 are oddballs and don't seem to have made any major impact anywhere. It's possible that they were slightly more common in the Upper Paleolithic but probably not too frequent anyhow, judging on the available data.
      → E, or more exactly E1b is an African lineage which has basically two highly derived variants outside Africa: M78, clearly associated to the Afroasiatic Mesolithic expansion to the Levant (Harifian), whose derived sublineage V13 had some significant role associated to G2a in the European Neolithic. The other is Moroccan-centered M81, which had a minor impact in Western Iberia and an oddball Welsh village; it could be Neolithic or it could even be Paleolithic (backflow from the Oranian genesis).
      → G is the only one which can be related to P1 and IJ and may have arrived to West Asia with them in the early UP. However it's clear that it had limited impact (Palestine, Caucasus) before it became more important (G2a specifically) in the European Neolithic. It's an interesting lineage but it does not overwhelm anything else.

    4. I'm only referring to R1, not its parent or P. I largely agree with the model for the dispersion of K2, but regardless of how far splinters of R reached in West Asia in the Paleolithic, it seems that the majority of R lineages entered West-of-the-Caspian fairly late. It seems that the Pottery Neolithic in Northern Mespotamia or Karelia is best explained by population movements from further East, whever they emanate. I think these can be plausibly yoked ultimately to Central Asia, north or south.

    5. IMO R1b was "West of the Caspian" as you put it early on, at the very least in the Levant-Anatolia arch. What we don't know yet is when and how its various sublineages spread to Africa, Europe and Central Asia. I see no reason to imagine R1b originating further East than that: the top ranking basal diversity is clearly concentrated in the Eastern Mediterranean and not further East:

      R1b-V88: forms an arch from Chad to the Levant and to Italy
      R1b-M355: found in Turkey
      R1b-P297: scattered in West Asia, Europe and Central Asia, two subclades:
      → R1b-M73: Central Asia, Bashkirs
      → R1b-M269: with origin in either the Balcans or West Asia, two sub-branches:
      →→ R1b-Z2103: mostly West Asia and now it seems also some in Eastern Europe
      →→ R1b-L51: essentially in Europe (some backflow to Anatolia/Armenia but clearly derived), two major subclades:
      →→→ R1b-S116: SW Europe, with projection to NW Europe, West-Central Europe and NW Italy (arguably associated with the Franco-Cantabrian region)
      →→→ R1b-U106: NW Europe, most common in the Netherlands (arguably associated with Doggerland and other North Sea areas)

      In synthesis:
      → R1b-root is clearly centered in the Western areas of West Asia (Anatolia/Levant)
      → → R1b-V88 is surely also centered in that area, although the clade demands some greater research, especially in Africa
      →→ R1b-P297 is arguably original from West Asia (some improved research would be required or, if it does exist, I'm not aware)
      →→→ R1b-M73 instead already seems centered in Central Asia
      →→→ R1b-M269 seems again centered in West Asia, and it's plausible that so are its immediate subclades

      Is West Asia a sink, as David claims rather happily? Well, I can't believe it, more so when there's no other plausible origin when the haplogroup is properly considered on its hierarchical diversity within the geography.

    6. Thanks for showing that. Ancient Y DNA is pushing towards an east European and north Eurasian origin, but such a scenario doesn't seem to be able to explain R diversity in west Asia and south Asia.

    7. Ancient DNA will always point toward cold places. Finding Mal'ta Boy's R* in Siberia tells us little about where R originated.

  2. This makes sense.

    West Asia is overrated, and was basically a sink for R, R1, R1a and R1b.

    1. Your ideological one-liners disappoint me a lot, David.

    2. You're missing the point. The middle east better at preserving diversity, not necessarily because groups came from there but because whatever does get there stays there. The steppe however, is subject to more population turnover so you are more likely to lose the diversity that was once there.

      Out of the very few samples from pre-neolithic North Eurasia we find R* and early R1(a/b). These individuals are also linked by a common autosomal heritage and all lack discernible near east components. The adna surely points to the north despite the early R1b in Spain and the lack of testing in the Near East.

    3. R1a or R1b still could have originated in the Near East but I don't find it very likely that we would be finding the stuff were finding in the north, so efficiently.

    4. Why would West Asia be better at preserving diversity. Are you suggesting that West Asian populations are somehow "frozen in time"? The issue of West Asian demographic changes, based on aDNA, remains to be fully clarified but the little data we have from the Syrian Euphrates suggests that there were major shifts there as well (although admittedly that crossroads may have been particularly sensible to demic changes). In any case, from the archaeological viewpoint, we do see important migrations, from outside as well as internal. From outside particularly: (1) African influences in the Mesolithic, (2) IE influences in Kura-Araxes → Anatolian IEs, (3) IE influences Indo-Aryan (Mittani first, Iranian later), (4) IE influences from Europe (Greek, "Sea Peoples" and Phrygio-Armenian flows, as well as Celtic at a later stage).

      I see no reason at all in any case to imagine that R1 was flowing to West Asia at all stages of its development. If anything, we should see specific founder effects, if what you imagine is true. For example terminal R1b-L51 branches are apparent in Turks and Armenians, indicating late flows from Europe, but that can't explain upstream R1b diversity, nor it can explain that of R1a, nor the greater basal diversity of R1 and R around Pakistan. What you suggest is just not-parsimonious: Occam's Razor is strongly against it.

  3. "This map shows the spread of ceramic technology from its birthplace in East Asia"

    Ceramic pottery of course.

    I would love to know where the technology itself first originated.
    Of course we have the ceramic Venus of Dolní Věstonice of 29,000–25,000 BCE.

    1. Thanks for the clarification. I meant 'pottery'. I'll update the post.

  4. Something important to remember is Haak 2015 didn't even find a reason to present the possibility of east Asian ancestry in the C1g bearing Mesolithic Karelian. It doesn't look like he had east Asian ancestry. Just like with Y DNA Q and C, mtDNA C1 may not be "east Asian" even though based on modern pops it appears that way.

    1. Whatever the case mtDNA C clearly coalesced in NE Asia, so its derived branches found in the Epipaleolithic and Neolithic of Eastern Europe (C1, C4 and C*) must have arrived ultimately from there. The simplest explanation is that they arrived with Y-DNA N1 (i.e. with proto-Uralic peoples migrating westward via the Taiga). What Bellbeakerblogger mentions about the westward migration of the concept of pottery (ultimately from China) underlines this pattern of late UP/Epipaleolithic migration through Siberia to the West. Everything converges and says: proto-Uralic cold climate specialists.

      Why did the cold climate specialists originate in East Asia? Well, actually the first cold climate specialists probably originated in West Eurasia (Central Asia), and Ma1 or the proto-Amerinds that migrated eastwards from Altai belonged to that group. However the LGM (permafrost line as far south as Budapest and Beijing) must have been a game changer, and after the LGM a new specialist and quite dynamic population arose instead from East Asia. In the West the ice sheet prevented people from migrating northwards too fast, but in Siberia there was no such ice sheet (rather dry cold climate, with many rivers and glacial lakes) or it was limited to certain areas, see: , so the N1 peoples of, say, Manchuria, soon found a living in Siberia and then moved west once the Mansi Lake allowed, surely c. 10 Ka ago.

      See also:
      → (reconstruction of Y-DNA N expansion)
      → (Neolithic China Y-DNA, showing commonality of N in the North early on)

    2. PS- Notice that mtDNA C belongs to a larger haplogroup known as M8, whose first branching was into M8a (exclusive of NE Asia) and CZ (common in NE Asia but with a wider range across Siberia, Central Asia and America).

    3. I saw some mtDNA data for Somalians. In addition to mtDNA haplogroup N1, there was also an instance of haplogroup C. I have no way of verifying the reliability of the data however.

    4. Maju,

      Somehow Uralic speakers in Europe became roughly 30% EEF. This makes it hard for me to believe their language existed and was widespread before they received EEF ancestry. But I know nothing about archaeology, I just wanted to remind you of that.

    5. Many Uralic speakers in Europe, Finns for example, have become through the millennia some 90% European genetically (although their Y-DNA remains largely N1). If that includes 30% EEF is something I don't necessarily subscribe, as Haak et al. admit it's difficult to quantify the apportions of admixture (the tendencies are clear but the fractions are approximative at best). But it's indeed possible that admixture with "Corded Ware types" south and west of them have given something like that fraction of EEF affinity you mention.

      Other European Uralic speakers like the remote Nenets remain much more strictly Siberian/East Asian in their genetic make up.

    6. Something I came to agree with Kristiina (herself Finnish and very knowledgeable of Uralic and related genetics, as well as archaeology) was that the process of dilution of the East Asian genetic baggage of European Uralics could be likely explained, in a simplified way, assuming two neighboring populations, one (A) south of the other (B), and with an effective population size ten times larger (because of climate). Let's say A has Ne=1000 and B has Ne=100. Each generation they casually exchange brides within a patrilocal context, let's say 10 brides: the impact of these 10 brides in B is massive, while in A is less relevant. Soon B becomes almost like A in autosomal and mtDNA markers but not in Y-DNA (because of patrilocality), A is only weakly affected by this exchange however.

      You can make more complex models with four or more populations, which surely approach better the reality of south-north interactions in NE Europe, but the essence is the same. So that's probably how Uralic populations that once were like Nenets (more or less) are now Estonians and Finns. And that's how some minor Uralic genetics (particularly autosomal and mtDNA) became integrated in more southernly populations like early Indoeuropeans, who scattered it around.

      It may seem puzzling on preliminary look but it's no mystery when you think about it. Probably proto-Amerindians went through a similar process in NE Asia before crossing into America.

  5. "I would love to know where the technology itself first originated".

    Firing clay objects is probably quite ancient but the oldest actual pottery used to be considered Japan, but more recently the focus seems to have shifted to China itself. I have long believed that ceramics spread from somewhere near Japan all the way to at least western Asia by diffusion, not separate development. This seems to provide further evidence for that belief. Perhaps it also explains the infusion of the Na-Dene languages into America and the apparent relationship with Ket. It may even lend some support to the concept of 'Dene-Caucasian'.

    "how it reached West Asia and Greece is not yet explained, an might have well been via cultural diffusion with nearly no personal contact"

    I certainly accept the technology moved beyond the genetic expansion.

    " K2 (formerly K(xLT)) must have coalesced in Sundaland (modern day Western Indonesia and Malaysia) and expanded from there in three directions: NO to the North, P1 to the West (P(xP1) remained in SE Asia) and what is now called K2a (incl. M, S and other Oceanian K) to New Guinea".

    Maju, didn't we some time ago get into a very heated argument on this very subject?

  6. M335 is more common in some Central Asian groups than in Anatolia.

    1. That sounds intriguing. My first notice of this lineage was that it had been sampled in Jordan, Wikipedia says Turkey instead. Do you have any reference for the Central Asian claim?

  7. The numbers on the map merely refer to chapters in the source book and are not themselves very meaningful. It would be helpful to include a timeline from the source of the relevant dates, which do indeed tend to show an East to West migration of ceramic pottery technology prior to the adoption of farming.

    1. I added a second map. The book is a preview, which blows.
      #7 would be nice to read, unfortunately we're living off the land.

    2. The second map is really powerful and helpful. Thanks. I missed it in the preview.

    3. The second map is really powerful and helpful. Thanks. I missed it in the preview.

  8. This comment has been removed by the author.

  9. This comment has been removed by the author.

  10. If you're interested in seeing a particular nonpreviewable section of the book, let me know: v {period} funnene 76 [no space in between] {AT} g mail