None dare call it multiregionalism

Dienekes Pontikos resurfaces with a post, Out of Africa: a theory in crisis. The title is a bit hyperbolic. But in Dienekes’ defense, he’s been on this wagon for over ten years, and the evidence is moving in his direction, not against him. I think a little crowing is understandable on this part.

With that being said, I think the biggest rethinking that we’re doing is less about where modern humans arose (Sub-Saharan Africa, North Africa, the Middle East), but how they arose. Some geneticists are quite open to the idea of Eurasian (Neanderthal?) back-migration to Africa several times (and out of Africa several times). Others are positing that a “multiregional” model might actually be about the situation within Africa.

A simple stylized model of a rapid punctuated expansion of humanity which replaces other lineages in toto is no longer likely. There has been widespread admixture, even though the last major demographic wave seems to be overwhelmingly predominant, at least outside of Africa. But the whole process might result in a much more complex history than we had thought.

The multiregional model is probably wrong on the details. The history of our species is not really phyletic gradualism and anagenesis. But there are also many processes and dynamics which a multiregional model takes into account and anticipates that probably are important in a general sense toward understanding the origin of our species.


16 thoughts on “None dare call it multiregionalism

  1. Is there any book you’d receomend as an introduction to this topic? The one you linked ? (Last Human) or is it dated now?(newest edition seems to be 2007)

  2. I think the concept of a patchy metapopulation is becoming plausible. Butterflies often survive and evolve as metapopulations where adaptations become fixed in some habitats only until new migrants blow in. Each patch is a garden of experimentation. None of the individual habitats need ever fall completely out of the population, but adaptations may flow through them at differential rates since conditions inevitably vary from patch to patch, and the details of each habitat can constrain adaptation.

  3. There is no magic barrier at the Red Sea. When earlier more primitive hominids strode across the world like collosi, there is no way that more advanced “modern” humans would be bottled up inside of Africa for hundreds of thousands of years twiddling their thumbs.

    If it weren’t for cultural / political motives, Out of Africa would have been discarded long ago, if it were ever adopted in the first place. Out of Africa is the Creationism of the Left.

  4. “Out of Africa is the Creationism of the Left.”

    Whatever theory you like has to be able to explain why the genetic variation is over such a small range among humans. Compared to cats and dogs for instance, we are all pretty much the same.

  5. I just emailed you, Razib, but I thought that I would post my question here since this may be a better medium in which to respond.

    For speciation to occur within human subgroups, how much genetic distance would have to exist in terms of Fst value?

  6. For speciation to occur within human subgroups, how much genetic distance would have to exist in terms of Fst value?

    no value. fst is just % of total var in pool.

  7. @Afterthought
    There’s some truth in saying Out of Africa because of the… Out of Africa bottleneck. Of course, the name and the mystic behind it is misleading and tells little about the whole saga.
    And it wasn’t just easy for the Homo sapiens to venture Out of Africa either, precisely because there existed other human species competing for resources against them.

    @Walter Sobchak
    The low variation, in my opinion, has to do with our long lifespan and relative small quantity of children compared to other animals.
    The bottleneck is also part of the reason.
    But hey, so far looking at Ancient DNA, we have managed to discover that humans keep separating and reuniting as well due to migratory behaviours.
    If you want to see how humans can truly diverge from one another, look at the Karitiana and other isolated populations, who present Fst close to ~35% from everybody.

    There’s no rule for that, but to me, as long as we are able to interbreed, then we’ll continue to be classified as the same species – but this is a bad form of thinking, because it would classify Neanderthals, Denisovans, Rhodesiensis, Erectus, etc, as the same species as us, as populations interbreed with them in the past.
    I really don’t know, maybe a cap of ~40% Fst? We would be able to separate modern peoples right now with that, although only some small isolated groups. Maybe ~50% Fst.
    Now, other forms of classifications such as Race, Sub-Species or whatever are all fair game to me, with smaller and smaller Fst.
    I even defend the whole “racist” Macro-groups of the past: Caucasoid, Negroid, Australoid, Mongoloid and Amerinoid, as genetics pretty much proved them in my opinion (and we can see that there are populations who fall into “bridge” groups, such as Indians, which are pretty much Caucasoids and Australoids, the Eskimo, which are Amerinoids and Mongoloids, the Somali, which are Caucasoid and Negroid, and the list goes on).

  8. There’s no rule for that, but to me, as long as we are able to interbreed, then we’ll continue to be classified as the same species

    there’s no rule for species. you are describing the biological species concept. evolutionary geneticists like it. at least those who work on mammals, and a lesser extent other animals. plant systematists, not so much. ecologists, a lot less.

    (i like the BSC personally, but you know my biases)

    I really don’t know, maybe a cap of ~40% Fst? We would be able to separate modern peoples right now with that, although only some small isolated groups. Maybe ~50% Fst.

    if you want a genetic statistic, you need to do pairwise nucleotide differences or something like that. Fst is not good. you can have low Fst if your diversity is really really high (east africa btwn bantu groups) or high Fst if the diversity you have is much more partitioned btwn groups (papua).

    Fst is a bad measure for macroevolutionary stuff cuz the denominator (total variance) matters and varies a lot.

  9. Ah, comically predictable drive-by post from Afterthought lol

    I’d say the Multiregional model was wrong on fundamentals, not details. Regional modern human populations didn’t evolve in situ from preceding regional archaic human populations. Where ancient DNA is available (admittedly very few places) we see a rapid change between highly divergent populations. In modern DNA the overwhelming majority of the common ancestry of non-Africans appears to have been shared near the beginning of the Upper Palaeolithic, and we apparently descend from a rapid and highly successful population expansion at that time.

    Sure, now we can talk about the multiregional origin of Middle Palaeolithic AMHs, but Multiregionalism was an actual proper theory that made fundamental predictions which were falsified.

    Though I could be totally missing something here?

  10. There were back to Africa influxes. Why weren’t they more successful? And how successful were they? Semitic pulses into East Africa were powerful and lasting. What imprint did they leave? And why not more? Why not further? What were the Punic contacts along sea routes with West Africa and across the desert to the interior? How much genetic exchange happened? How much persists, and why wasn’t it more? Is the disease regimen the simple answer?

    How much of the genetic homogeneity of Eurasia is simply to the scale of the expansion from paltry pre-agricultural densities, along with the overlay of a single conquest dynamic – that of the IE expansion? Why were the pre-existing populations of Eurasia so much more easily overcome than those of Africa? A tropical disease regimen didn’t stop the Indo-Aryan expansion.

    For that matter, the persistence of the genetics of Lebanon is fascinating to me. It makes me wonder whether geographical populations are somehow resilient to penetration.

  11. @ryan: Why were the pre-existing populations of Eurasia so much more easily overcome than those of Africa?

    By who, the Bantu expansions or the Levantine Neolithic related expansion, (who I think had become some kind of pastoralists by the time they reached the edge of their expansion in SE Africa)? I’d guess the issue, for why the outcome seems different to India, was pretty simply no climate suitable crop package for sheer population density that compared to rice+tropical monsoon system of South Asia. Teff, I assume, did not have such an explosive potential as the West African package for expanding across Africa either, which happened later in history and led to them replacing other Africans, and that is all she wrote.

    The Horn is not really such a small territory to expand across compared to South Asia (if small by Bantu expansion standards), and population recent population growth has been intense (above 100 million in Ethiopia) the main thing relevant for question seems perhaps why population growth there was persistently fairly low on the eve of European colonisation (relative to India), and that seems like it would be rooted in the agricultural system.

  12. Thanks, Matt. That’s useful. I’m writing from a degree of ignorance, and looking for answers, not positing that my questions are significant challenges to anyone’s theory.

    I was thinking more of the Semitic kingdoms of the horn in antiquity and the various other Semitic and then Arabic realms that might have aspired to wider dominion in the ensuing 1300 years or so. And your explanation is probably part, perhaps all of the answer.

    My understanding was that the Bantu expansion reached eastern Africa relatively early. I’m not clear on why colonizing such agricultural territorial wouldn’t have been an interesting venture for a set of Arabic Vikings, if you will. Or maybe it was, in ways that I don’t know about, leaving traces that I haven’t read about.

    I see a tweet from Razib saying we know more about the ancient genomics of Estonia than all of India or China. How much more true of Africa?

    I would be interested in better, book length sources on Africa, preferably history/archaeology, though an accessible book on genetics would be great.

    On a related note, after reading a NY Times article today on the slave trade in Benin, I looked for histories of that era in Benin / Dahomey, and found little that seemed likely to slake my interest.

  13. Thinking towards this from another direction, within 150 years of the Europeans reaching the Americas, they had overrun vast portions of it, imposing themselves as a new elite in many areas and establishing large scale settlement elsewhere. The discovery of the Americas was in ways a spin-off of the Portuguese ventures along the African coast. Yet it took hundreds of years before Europeans established more than trading stations anywhere in Africa. What was the difference? Was disease the primary competitive advantage of the Europeans vis-à-vis the Americas?

    Why could they not crack the African nut for so long? Why could the Arabs not do it? Why had no Semitic people ever been able to?

    The popular version runs that the Renaissance engendered the formation of superior European civilizations which quickly overran the world. But I think there’s an argument that disease was the only reason the Europeans overran the Americas. And that it was only a long dynamic in which new surplus resources from the Americas played a primary role, and also a complex one by allowing the surplus populations that could devote themselves to something other than farming and in-fighting (ie, inventing), that allowed European colonialism to span the world.

    In short, that the superiority of European civilizations was at least initially the product of conquest, not the determinant of it.

  14. There was no Arab power ever present in the Horn aside from a 50 year tributary status of Mogadishu and the southern Somali Benadir strip by Zanzibar, which really was a Somali Geledi sultan and clan puppeting the Omani govenor in Mogadishu, who was more a Geledi agent in service of the Omani Zanzibar.

    The Semitic migrations into the Ethiopian Highlands was a phenomenon that began around 4,000 years ago and origninated from Eastern Saudi Arabia and the Tihama region by the earliest wave of Semitic speakers into Arabia, the ancestral proto-Ethiosemites, who were predominatly J1-p56 and not J1-p58 like all other Semitic groups at the time.

    This group was mostly proto-Semitic in ancestry, and contributed around 23% of ancestry to modern Ethiosemites. The proto-Ethiosemites were likely 40% proto-Semitic and 60% Agew in ancestry. This proto-Ethiosemitic group founder effected for certian A3b2 lineages and E-M34 lineages of the Omotic type. They then assimilated Agew Central Cushitic groups in the Highlands as well as North Cushitic groups in Northern Eritrea related to the Beja. The far southern groups assimilated East Csuhitic groups of the Highland variety like Sidamos and the Hadiyya.

    All other Cushitic groups were unaffected by this Semitic migration, aside from modern Oromos and Highland East Cushitic groups. Somalis (Lowland East Cushitic) do not show this influence, as they are only >1% Iran Chl in ancestry. Same for South Cushitic groups like the Iraqw of Tanzania and the Bantu Tutsis.

    Also the Levant Neolithic had was not important to the Cushitic groups as all their Eurasian ancestry was Upper Paleolithic North African, not Natufian. 40% of Somali mtdna looks to be some ancestral pre-Natufian Upper Paleolithic West Eurasian Anatolian-Zagros related. This populations was ydna H2 and T1-L208 heavy with minor C1a. It seems only T1a survived the trek to Africa during the Upper Paloelithic. This West Eurasian group mixed with an expanding Paleolithic Highland Ethiopian population that was predominately pre-E-M35 or E-M215* (E1b1b*) and minor A3b2.

    These two populations mixed in the Nubia and Red Sea Egypt and Sudan starting around 28,000 years ago and they had stabilized around 18,000 years ago. E-M78 seems to have originated near a Lake Nasser refugium in the early Mesolithic late Paleolithic.

    This mixed populations was either Beja-like to Somali-like, with northern groups being Beja-like and southern groups being Somali-like in their West Eurasian/Ancestral East African ratios. This group was the proto-Afroasiatic population of old.

    The proto-Cushites were from the southern groups that were Somali-like, and they were mostly E-V32, E-V1515, E-V22, T1a-L208, and A3b2-M13/M118. The Natufians and Levantines had 1 E-M78 and were all pre-E-M34 and the like. They had one E-M35 lineage (E-Z830) out of the dozens that existed at the time. E-M34 lineages also existed the time also, and were heavy in the proto-Omotic group. The proto-Omotics splintered from core Afroasiatic group by 13,000 ybp and had migrated into the Ethiopian Highlands by at least 5,000 ybp.

    The proto-Cushitic are also believed to have domesticated the Nubian and Somali donkeys around the Red Sea Hills. They only recieved sheeps, goats, and cattle from northern Afroasiatic imtermediaries, likely ancestral proto-Semites in the Delta with little gene-flow, just like the proto-Nilotic and Sudanic groups in Chad and Darfur did.

    So the Levant Neolthic had no direct impact on proto-Cushitic groups aside from indirect borrowing of domesticates like cattle, however there was absolutely no genetic influence whatsoever. But then we have indirect flow the other way as well, like Donkeys, and pottery – which originated from the Nilo-Saharans, and actual genetic influence like mtdna L2a1 and E-Z830/and all E-M35.

  15. @Ryan, I’m no expert on the area at all! Re: African ancient dna, read Skoglund 2017 (, Schlebusch 2017 (, Fregel 2017 ( and there’s a bit more from 2015 (

    Based on modern dna, this paper suggests some things about dates To an extent there’s also Schuenemann 2017, which is really for Egypt –

    We probably (and only very recently) know a bit more about Africa than China for the Holocene (last 10,000 years) but have nothing as old as the Tianyuan Man (40 kya) ancient dna from China, and Africa’s a big place, so it’s not quite comparable.

    (Presumably this is because there is a good history of Western researchers in Africa who are interested in modern human origins, while less so in Asia and possibly the Indians and the Chinese want to do the work themselves?)

    (I)t was only a long dynamic in which new surplus resources from the Americas played a primary role, and also a complex one by allowing the surplus populations that could devote themselves to something other than farming and in-fighting (ie, inventing), that allowed European colonialism to span the world.

    This is the kind of idea a la Kenneth Pomeranz of “ghost acres” and surplus wealth from the New World providing a boost to Western European economies and so that leads to science and technology and economic growth in a positive feedback and so ultimately allows further imperialism (which possibly also feeds back again).

    It’s a plausible idea, but I don’t think it’s currently so popular among economic historians and historians of science and innovation. See this piece by Mark Koyama for a quick insight –

    We’re really dealing with two parts of European colonialism.

    There is the part, the first phase, where they gained ascendancy over various peoples who were outside the cutting edge of Eurasian technology (which was on the cutting edge of world civilization, though, yes, there are arguments against a too linear view). Really this is the part that Iberian, English, French imperialism in the New World fell into. That more just required them to be roughly at the Eurasian cutting edge (roughly the same as India, China, Middle East, which they were), then to have incentives to expand across the oceans from national competition between European early states and finally the disease “advantage”.

    Then there is a second phase where European trading companies, posts and interests and then governments sort of grow from being small outposts to taking over territories which were formerly at the technological cutting edge, like India, or informal spheres of influence in China. This did require a decisive European technological edge and so more sophisticated system for innovation in science and technology. But like I say above, economic historians (the consensus) seem to be looking at this more as driven by competition between European states and as the result of specific technological innovations, internal economic drivers and cultural change, rather than due to inputs of wealth / land / resources from the New World or colonies generally. (But this is all pretty off topic!)

  16. ” as long as we are able to interbreed, then we’ll continue to be classified as the same species”

    That would mean that horses and donkeys are the same species. Maybe the rule should be to be able to interbreed without any loss of fertility such as is described by Haldane’s Law.


Comments are closed.