Tuesday, November 21, 2017

Maternal Genetic Makeup of the Iberian Peninsula (Szécsényi-Nagy et al, 2017)

Middle Neolithic.  France.

This paper includes the HVS-1 profiles of nearly three hundred ancient Iberians.  Rather than summarize the paper, let me ignore everything else and zero in on a single point.  The enigma of H1 and H3.

Multiple streams of evidence are starting to illuminate something that happened in the MN, possibly moving out of France.  From the Paris Basin to Paris Street, to barbarian Blatterhohleans and belligerent Baalbergers we see something that is curiously steeped in WHG-rich ancestry, high haplogroup H frequency, and the curious presence of R1b.  I call them the Violet Volken after the Fregel et al, 2017 K=8.

Studied Area. Snip from Fig 1
Before delving into this study, let's question the basis of a particular narrative.  It is that the rebound of WHG-related ancestry in the Middle Neolithic is the result of a slow process by which hunter communities were civilized and absorbed into the general farming population.  Does that even work mathematically based on population estimates?

But at gory Gougenheim (Beau et al, 2017) the maternal profiles don't support a situation in which shy hunters slowly grafted into the settled mainstream.  Instead, like Blatterhohle Cave we see what amounts to a 'deeply segregated society' based on the data available so far.  It's pretty clear who ruled the school in Gougenheim.  What if the spike in HG-related ancestry isn't local, shy hunters?  Let me answer that for you, it's probably not.

This study possibly adds more full to that fire.  Maybe in the strictest terms Olalde et al, 2017 is right that Maritime Iberian Beakers contributed little ancestry to Continental Beakers, but it may be more complex than that.  And this offers an intriguing thread to pull:
"...the proportion of haplogroup H is higher in the Iberian Early Neolithic (EN) (22.2%; n = 27) than in Central Europe (15%; n = 160), while the frequency of N1a is very low (3.7% compared to 9.4% in Central Europe). Another difference with regards to Central Europe is the occurrence of haplogroup N* in Neolithic NEI group"
Hervella et al, 2015 made a nearly identical comment concerning the Neolithic Balkans.  I'll revisit the Hervella "H" hypothesis in a moment.
Snip of Fig 2
It's unlikely that the high rate of H1 and H3 can be explained by a founder effect of the Danube farmers.  Just not enough time for a patriocally, high female-mobility-oriented culture to pile up only H1's in some remote pocket of Southern France.  That reasonably leaves the Cardial/epi-Cardial/Impressa or whatever tradition which can explain the Mediterranean littorals; or alternatively, as Hervella et al suggested, a third Neolithic expansion (more on this).
"The largest proportion of the H individuals belongs to the subhaplogroup H1 (65.1%), and the second largest group is subhaplogroup H3 (14%)... H3 is detected in Chalcolithic individuals from central, southeast and southwest Iberia. H1 is observed in each period and region, but more frequently in the Chalcolithic and Early Bronze Age than in the Neolithic. .... However it becomes apparent from our results that the H diversity in prehistoric Iberia is different from the H diversity of Central Europe, and more similar to the Neolithic populations in France.  Notably, common Central European subhaplogroups H5 and H7 have not yet been observed in Southwest Europe."
The below hints at what I've already suspected, that a large portion of the Steppe-related ancestry in Beakers of the North European Plain is owed to the Corded Ware Culture that it subsumed, probably from a combination of wife-buying and destruction.
"Interestingly, we also do not find evidence for influx in the East to West direction, as none of the investigated Chalcolithic individuals show ‘steppe ancestry’, which seen in contemporaneous Central European Corded Ware and Bell Beaker groups, suggesting that eastern influxes did not reach the Iberian Peninsula until later periods."
What this means is that the rapid rise of haplogroup H in Iberia is not connected to Steppe-related ancestry, at least before the regular Bronze Age.  This is an important and subtle point.  It means that Iberia was experiencing a surge-of-something that was also affecting other parts of Europe hundreds of years before the Beaker phenomenon. 

Finally to the Hervella "H" Hypothesis which is an attempt to explain the extreme rise in the haplogroup H right before the Late Neolithic.  Brotherton et al, 2013 postulated that this rise reasonably came from Southwest Europe into Central Europe with the Bell Beaker Culture.  But the Hervella team found that explanation not suitable after examining a similar mito-phenomenon in the much earlier Neolithic and Metal Age of the Balkan Peninsula.  Clearly there were farmer-like maternal profiles and then the exploding H-people.  This is what Hervella wrote in response to Brotherton in 2015:

"...none of the models studied to date have taken into consideration another possible and obvious explanation, namely a new wave of Neolithic migration into Europe through the ‘traditional route’ of the Balkan Peninsula. This new wave of Neolithic migrations are represented by Vinča and Dudeşti cultures (5500–5000 BC), that trace their origin in North-West Anatolia on the basis of ceramics features [28]. The Boian, Zau and Gumelniţa cultures from Middle-Late Neolithic (M_NEO) from Romania are the direct continuation of this cultural complex; the M_NEO group from Romania displayed differences in haplotype (S5 Fig) and haplogroup distributions (S4 Fig) with the Middle Neolithic from Central Europe."
That would put a 3rd Neolithic behind the Iron Gates, in the Wallachian Plains cultures that also gave rise to the Boian encrusters, who are nearly contemporary with cultural changes in the North of France and Germany around 4300.

What if that is the source of excess WHG in the Middle Neolithic in the West of Europe?


  1. So you're saying a population rich in yHG R1b, mHG H, and WHG ancestry is the source of the rise in WHG in west Europe after 5000 BC? Northern Bell Beaker got its Steppe ancestry and high mHG H frequencies from Corded Ware women. R1b L151 hit western Europe before H1/H3 and before the rebound of WHG. So how can those three things be related?

    1. L51 might have entered the North European plain at the beginning of the Middle Neolithic. If you look at the Neolithic cultures in this area such as Rossen and the Western TRB groups, there's intrusive elements such as right-facing oval pit burials and white-encrusted pottery.

      Already from some of those Western groups such as Baalberg or Blatterhohle there is the presence of R1b. I'm not sure if or how that marries with H1 and H3 who both seem to already be at a respectable frequency in the Iberia EN. But according to this paper the wiggle was wide enough that they just grouped MN and LN together as MLN. So it could be that H1 and H3, or a portion of that is actually MN, but who knows.

      The rise in HG is an actual thing. I think the NE Balkans is an interesting a suspect place, which is BTW, also a plain. So it's a suspect place in light of the Iron Gates Mesolithic.

    2. And to clarify, at Blatterhohle and Gougenheim there are clearly foreigners. No one has figured that out yet.

      But if Hervella is correct about H-folks penetrating the Eastern Balkans from NW Anatolia into an area with a large relict of Mesolithic ante-IronGate R1b, then it is a very interesting place to see a source for cultural shifts in the MN elite of the Western North European plain.

    3. This is the first time I've seen this Iron Gate Neolithic hypothesis really articulated and honestly, it is very attractive. There is also room to tie up some other loose threads here.

      First, there is almost surely a significant discontinuity in population genetics and language between first wave Anatolian EEF and the population that was in Anatolia immediately prior to the Hittites. Iron gate people could explain that.

      Second, it provides a solution to the discontinuity in Beaker population genetics between Iberia and the rest of Europe. Iron Gate Neolithic people expand from Anatolia to the Balkans to Iberia. This culturally linked swath of people provides the matrix into which the Bell Beaker culture expands from Iberia to points North and Northeast of Iberia on the strength of the edge that the Iron Gate Neolithic people had over their predecessors and additional Iberian innovations.

      Third, in places where the Iron Gate Neolithic did not reach, the more primitive farmers are no match for the Corded Ware people expanding off the European steppe. But, while steppe people also expand into Iron Gate Neolithic territory, these more advanced people (further bolstered by Iberian Bell Beaker innovations) have more staying power vis-a-vis the steppe invaders. This slows the advance of the steppe people enough that they don't make it all of the way to Iberia in the early to middle Bronze Age (after which the steppe migration runs out of gas, perhaps due to the abatement of whatever was pushing them off the European steppe in all directions) and it means that the Iron Gate Neolithic => Bell Beaker civilization they encounter integrates them as a numerous and prosperous wave of immigrants into their power structure (a bit like the Romans in the Greek-speaking and culturally Greek Eastern Roman Empire a baker's dozen centuries or so later; or like Catholic and Lutheran migrants to the Mid-Atlantic and Midwest who produced a huge population genetic shift in these regions while being subsumed culturally into the culture of the first founding European farmer populations there in the 19th and early 20th centuries), rather than entirely crushing the existing power structure as conquerors and replacing them entirely in culture and population genetics - something that fits the historic archaeological characterization of Bell Beaker cultural innovations as more like traders and missionaries than war parties.

    4. There are ways you could test this hypothesis in the near future.

      For example, we would expect the male-female ratio of steppe ancestry sources based on comparing X chromosomes to other autosomes in Western Europeans (ideally ancient, but modern wouldn't be irrelevant) to be significantly lower in Southern and Western Europe than in Eastern, Central and Northern Europe where a 5-1 to 14-1 ratio is observed in a recent study.

      Y-DNA R1b should pre-date steppe ancestry consistently from pre-Hittite Anatolia to the Balkans to at least Northern Italy to Southern France to Iberia.

      You should see Iron Gate Neolithic DNA profiles in prestige graves before before the arrival of people with steppe ancestry and in at least the early period after steppe ancestry.

      You should see an archaeological culture boundary between remains with Iron Gate Neolithic genetics and those with EEF genetics, and should see evidence a trade routes and bride exchange and common burial practices and archaeobotanic innovations that connect the Iron Gate Neolithic region in a quite narrow time frame of a few centuries in any given place.

      Precision carbon dating should show a slower pace of advance for steppe peoples (categorized by material culture litmus tests) in the Iron Gate Neolithic area than elsewhere in Europe.

      The Iron Gate Neolithic ought to be occurring after the local collapse of first wave EEF agriculture, but before the early Bronze Age.

      The Iron Gate Neolithic should lack some element of the food production or technology package necessary for success in colder climes that Corded Ware people did have.

      The HG component of Iron Gate Neolithic autosomal genetics should look more eastern on an east-west PCA cline compared to other HG components than true Mesolithic HG from far Western Europe since some of the HG was at least picked up in the Balkans if not in Anatolia. You also ought to see a Balkan to Iberian trend in autosomal haplogroup tracing.

      Iron Gate Neolithic people should lack the lactose persistence gene which increasingly looks like it expanded from NW Europe - LP is still pretty rare in the Iron Gate Neolithic region compared to other parts of Europe.

    5. "For example, we would expect the male-female ratio of steppe ancestry sources based on comparing X chromosomes to other autosomes in Western Europeans"

      A poster on Eurogenes made a similar comment, although I can't remember who. It was concerning the sex ratio estimates for steppe ancestry. I'm not sure how to figure that, but I did notice early on that Mittle-Saale and Elbe Beakers in big numbers looked like a combination of 1/2 CWC and 1/2 something else. But, we shall see.

    6. "Iron Gate Neolithic territory, these more advanced people (further bolstered by Iberian Bell Beaker innovations)"

      In particular, Iron Gate Neolithic people probably receive their next major tech boost of metal technology (proximately) at least up to Copper and maybe early Bronze from Iberia rather than from the East. This could be tested by looking at the alloys of early metal artifacts in this region and sourcing the place of origin based on its metal composition.

    7. @bellbeakerblogger - "A poster on Eurogenes made a similar comment, although I can't remember who."

      That was me.

      Mathieson 2017 does this by doing separate qpAdm runs for the X chromosome vs the whole autosome and comparing the two.

      How they did the actual comparison I'm not entirely sure, but I think they set up a system of equations and generated a parameter space from that?

      David said he tried doing something similar though and ran into problems due insufficient data from the X chromosome alone. There seem to be some problems that Mathieson ran into as well, as for example the Balkans Bronze Age comes back as 291% Anatolia Neolithic and -198% Yamnaya... so... yeah...

      Thanks for articulating this so well. I've been thinking along similar lines for a while and you've added a lot of pieces to the puzzle.

  2. Good idea. If reliable one could add a force behind this late neolithic migration: the flooding of Black Sea. (If there was a flooding of ancient Black Sea there MUST have been migrations.)

    1. Makes sense. The time frame is at least suspect for the formation of the Danube Delta, regardless of how gradual that may have been. The area behind the Iron Gates was certainly inhabited by R1b, a majority it seems. The destruction of the Danubian riparian zone or the expansion of wetlands could have had a positive or negative effect on population size. It may have affected all of the cultures of the Circum-Pontic region.

    2. I don't think the Black Sea's sea level fits the bill. We need timing ca. 5000 BCE to 4300 BCE (6950 BP to 6250 BP more or less) which is wrong for Black Sea flooding if I recall correctly. This is basically a wet Holocene optimum period that immediately followed a bad arid period.

      Also, an Iron Gate Neolithic surge has to have a root with the major impact in the Balkans (which has a Mediterranean basin coast not a Black Sea coast) and Anatolia (which gains elevation from sea level along its Black Sea coast very quickly). A rising Black Sea level mostly affects the north shore of the black sea, not the Anatolian coast (and the western coast of Anatolia is in the Aegean not the Black Sea).

      We also don't need nearly as intense a push as we do for the steppe migration off the European steppe which was a multi-generational male dominated near replacement surge. The mtDNA H1 and H3 frequencies absolute do increase, but it isn't replacement either. A back of napkin estimate would be that the introgression percentage from Iron Gate Neolithic people over two or three centuries (6 to 10 generations) is probably on the order of 20%-45%, weighted towards the lower end of that range if the Iron Gate Neolithic technology package gave people with that culture a selective edge over Old Europeans. My guess would be population pressure from migrations of Levantine E1 people north driven by the tail end of the arid period of the 6th millennium BCE, as one of the main pushes that chain reaction style pushes people out of Anatolia/Aegean/Balkan turf that is facilitated by a break through in farming technique/diversification that solve the problems in farming technique that led to the farming collapse a few centuries after the arrival of the first wave Neolithic (perhaps crop rotation with legumes and keeping fields fallow part of the time and organic fertilizers which combined would combat soil exhaustion, as soil exhaustion is a plausible strong contributing cause of the first round of farming collapse).

    3. I'm not sure Neolithic societies really need a push beyond Malthusian constraints. They would be driven to expand their range out of necessity. I think all you need is a slightest edge to form an opportunity.

  3. The Black Sea flood is ca. 5600 BCE (too early) and as I note, mostly floods land on the north shore of the Black Sea, not much in Anatolia or even the Bosphorus Strait, and none at all in the Balkans. https://en.wikipedia.org/wiki/Black_Sea_deluge_hypothesis

    This is a good fit as a push factor for R1b-V88 people to head to the Nile and beyond (and in other directions as well where they are more diluted). But, it is a poor fit for an Iron Gate Neolithic push factor.

    The Iron Gate Neolithic expansion ought to post-date the 8.2 ky event, although if this happens closer to 4300 BCE than 5000 BCE, and is also not a great fit for the 5.9 ky event, which is a bit after the proposed Iron Gate Neolithic expansion, although it is a better fit. https://en.wikipedia.org/wiki/5.9_kiloyear_event

    "It may have contributed to the decline of Old Europe and the first Indo-European migrations into the Balkans from the Pontic–Caspian steppe. Around 4200–4100 BCE a climate change occurred, manifesting in colder winters in Europe. Between 4200–3900 BCE many tell settlements in the lower Danube Valley were burned and abandoned, while the Cucuteni-Tripolye culture showed an increase in fortifications, meanwhile moving eastwards towards the Dniepr. Steppe herders, archaic Proto-Indo-European speakers, spread into the lower Danube valley about 4200–4000 BCE, either causing or taking advantage of the collapse of Old Europe." If the Iron Gate Neolithic starts ca. 4200 then we have Indo-Europeans on their tail pushing them out and a fit with pre-IE cultural disruption as the OP notes.

    One can see the Iron Gate Neolithic as a pilot wave, a bit like the non-IE Etruscan migration into Northern Italy just a century or so ahead to the Italic language IE migrants on their tail.

  4. Great post. This speaks to many of my big picture questions. Needless to say, it doesn't answer them, but it does help me organize my recent and still superficial knowledge of the subject. Thanks.

  5. Awesome post! If the Iron Gate Neolithic turns out to have had a major impact on Western Europe, do you think it would make sense to see in the associated cultures the linguistic progenitors of the Iberian, Tartesian and Basque peoples as well as perhaps the Picts?

  6. This culture and migration could be the possible origin for vasconic languages? what do you think?
    Thank you!

  7. No! The proposed third Neolithic expansion.

  8. From Olympus Mons:

    "It just has been published the R1b-M343 (xP312xU106) Y-DNA tree by Sergey Malyshev.

    I don’t really know them, don’t know how credible this things are. However two things are remarkable.
    a. They, like Genetiker, put ATP3 as M269 and,
    b. most amazingly, add MC337A, Monte Canelas (Portugal), Late Neolithic/Chalcolithic also as R1b-M269.
    c. Atp3 in north Spain 3400-3100 bc and MC337 is 3200-2900bc in the most remote southwestern point of Europe.

    Remember that Martiniano et al published several Portugal Late Neolithic/Chalcolithic individuals as I2a1b giving the impression they dominate this Period in Portugal and MC337A as nothing reported for him. Then, the following period, middle bronze age, as full of R1b1a2a1a2 (so, P312). Hence the arrival of Steppe (or Balkan east bell beakers) R1b timeline mantra. Though, these 3 bronze age P312 actually had no Steppe component to them. But it did not deter the narrative.

    If, I repeat If, this turns out to be correct then we have In Portugal, further way as one gets from steppe. 3200BC (M269) ------- son 1800BC (P312) ----- grandson R1b-DF27 (without Z195, so no East to west cline). In the same 100 miles area.

    Not really the story we are been told, is it?"