Last Survivor

Over about 2500 years, the Middle Eastern farmers occupied almost all of Europe. That migration seem to have a single origin:  genetically they seem quite similar, even as far as north as southern Sweden.

The indigenous foragers didn’t disappear instantly, and some seem to have survived amidst the dominant farmers for a long time, without much gene flow. This pattern looks similar to what we’ve seen with Pygmies and Negritos.  In these contemporary examples, the encapsulated foragers have lost their original language. Probably those persisting European hunter-gathers also adopted the language of the surrounding farmers.

If this picture is correct, it seems likely that almost all of Europe, circa 4000 BC, spoke related languages, languages descended from a language that originated in the Fertile Crescent. The language map probably looked something the Bantu languages in Africa.

And then things went bad for the EEF peoples.   There is only one surviving language that seems likely to be member of that old radiation – Basque.  There were other languages that were probably in that family and lasted for quite a while: Iberian, Nuragic,  probably Tartessian,  maybe Etruscan, possibly Minoan and Eteocypriot.   The evidence we have on those other languages generally stems from the Classical era: by that time, these languages had had maybe five or six thousand years to diverge.

This means that Theo Vennemann is probably right about a Vasconic substratum, only it should be considerably more extensive than he suggests. The population in question originated near the heartland of Middle Eastern agriculture (First Farmers would be a good ethnonym), rather than spreading from a glacial refugium like Spain after the LGM.

How to prove this?  Sounds tough.  We may need another Rosetta stone.





This entry was posted in European Prehistory, Linguistics. Bookmark the permalink.

88 Responses to Last Survivor

  1. Pingback: Last Survivor | Research Workshop |

  2. j3morecharacters says:

    Are the Basques the last of the EEF peoples? Homologues of another loser race, the pygmies? I have some difficulty in accepting that because Basques are indistinguishible from Southern French people, well built and blond, and in nothing inferior to other Europeans.

    • minoritymagnet says:

      Basques are not a loser race, they were fierce enough to keep their language as opposed to the peoples around them. Pygmies did not manage that, the Bantu keep them as inferior slaves. They adopted their language and depend on their Bantu masters for survival.

      • j3morecharacters says:

        Read again: That is exactly what I am saying.

      • minoritymagnet says:

        You’re right. Skimming over is not enough as a non-native speaker.

      • Ian says:

        Perhaps they were just too slouch to learn a new language and adopt the newer trends brought by their neighbors. As Jose Luis Borges said, they historically excelled in nothing, except in milking their cows.

    • gcochran9 says:

      Most of southern Europe is EEF, and maybe half of northern Europe.

    • B&B says:

      CS Coon on the Basques:

      “On the whole, these craniological data indicate three facts: (1) the Basques are basically Mediterranean (in the wider sense) racially, with some brachycephalic accretion. (2) This accretion is for the most part Dinaric and only to a minor extent directly Alpine. Morphologically the Basque crania show many resemblances to those of Serbo-Croats and of some South Germans. Collignon’s comparison between French Basques and the southwestern French makes this distinction clear. (3) The Basques, through inbreeding, ethnic solidarity, and the possession of a recognized national ideal type, have developed a character istic physiognomy, the essential features of which are nasal prominence and a narrowness of the median sagittal facial segment, and of the mandible.

      Collignon believed, and Montandon follows him, that the French Basques are freer from modern mixture than are the Spanish Basques. This may perhaps be true, since neither the round-headed tendency of the French Basques nor their relatively high incidence of blondism can be wholly explained as local acquisitions. The Basques, as a whole, represent an ancient and subsequently specialized mixture of Mediterraneans and Atlanto-Mediterraneans with partially blond Dinarics, and it is just as possible that different Basque sub-groups differed originally in amount of Dinaric blood as that the modern Spanish Basques have been altered through Spanish mixture.

      Both the Atlanto-Mediterranean and Dinaric elements mentioned were present as early as the Copper Age in North Central Spain, where they were partially identified with the early Bell Beaker culture. The Keltic Iron Age racial type of Britain, which the living Spanish Basques so closely? resemble, was produced originally in southern Germany from a combination of Nordics with Bell Beaker or other Dinarics, and imported into England where Mediterranean and Atlanto-Mediterranean elements, as well as some Bronze Age Dinaric factors, were already present. The mixture of similar ingredients in different places produces similar results. Seen in the light of modern physical anthropology, the Basques are still interesting, and perhaps romantic, but no longer mysterious.”

  3. Nick says:

    The Wikipedia entry on the Vasconic substratum says: “Vennemann also adduces evidence from genetics and blood types that show that the Basques share characteristics found throughout Central and Western Europe, especially in typical areas of retreat-like mountains.” Later: “The hypothetical Vasconic substratum has been largely rejected by historical linguists.”

    So what’s the deal here–are the historical linguists who reject the theory not taking into account the genetic evidence?

    • gcochran9 says:

      The genetic evidence is recent and I would guess that few linguists are familiar with it.

      • B&B says:

        Linguistic evidence can be evaluated by linguistics only, and the only language connected to Basque is its predecessor Aquitanian. A connection to the Artegnacian archaeological culture is intuitively true.

        Wierdly there is extensive Vasconic vocabulary in Guanche, that is unlikely to be the result of trading contacts since others like the Normans left no such vocabulary there.

        My thoughts were that the Vasconic vocabulary must be Cardial-associated to have reached North Africa before the Guanches left the mainland. There is no evidence of Vasconic in Iberia outside of the Basque country and Guanche was then a relict of former language diversity in the Maghreb before Berber proper replaced everything, spreading along the Roman limes and the trans-Saharan trade routes of the Garamantes.

        Proto-Berber can’t be earlier than the introduction of date palms, and Berber has a very atypical vocabulary for an Afro-Asiatic language. Try looking to North Africa for odd substrates that might have influenced Europe.

      • Anonymous says:

        “…Try looking to North Africa for odd substrates that might have influenced Europe…”

        Irish and Scottish Gaelic have Afro-Asiatic word order (VSO) superimposed with Celtic vocabulary. It’s quite possible that the Q-Celts overran AfroAsiatics… but where? In Ireland? Or Iberia?

      • dearieme says:

        “Wierdly there is extensive Vasconic vocabulary in Guanche, that is unlikely to be the result of trading contacts since others like the Normans left no such vocabulary there.” I don’t find that persuasive. If a first bunch of outsiders provided words to fill empty linguistic niches, then there might be no good reason to borrow words from a later bunch of outsiders.

    • pheltz says:

      In historical linguistics, even hard-line splitters typically don’t “reject” proposed relationships in the sense of affirmatively thinking the languages must not be related. The position is more: sure, it’s plausible for them to be related, and in fact the world would make more sense if it turned out they were, but it doesn’t seem to have left any perceptible trace.

      Isn’t that basically in agreement with what our host is saying? Yes, a fairly close relationship between Vasconic and Etruscan etc. would be unsurprising, a relief even. I think most historical linguists would have agreed with that, not only before the current genetic evidence was available, but before we even knew that DNA was responsible for heredity. But the languages themselves aren’t cooperating, at all. Which is odd, and frustrating.

      • Seth Long says:

        Very few historical linguists, in my experience, follow population genetics research. The idea that genetic evidence should guide linguistic reconstruction is more or less a foreign one, and since the comparative method of reconstruction becomes more and more tenuous the further back you go, most effort these days is focused on reconstructing proto-languages outside of Europe.

      • pheltz says:

        @Seth Long:
        All perfectly true, but my point is that you don’t *need* to know a thing about genetics to be motivated to find a Basque-Etruscan connection. Two pre-IE languages in the same part of Europe (with Vasconic substrates in southern France and in Sardinia, very close to Etruria)? Certainly seems like they should be related. But there doesn’t seem to be even a semi-credible cognate list floating around out there. Compare that with the substantial evidence produced for connections as distant in time as Afroasiatic or Dene-Yeniseian.

        • gcochran9 says:

          It might be that Etruscan is the product of a later population movement, and doesn’t go back to Old Europe. But it seems likely that many of the languages I mentioned do. But we know so little of them.

      • B&B says:

        there is a Baskic substrate demonstrated only in Gascon where the Aquitanians lived. Baskic does correlate only with the old culture of Artegnac.

        Afroasiatic is far more recent than people think (see Blench again) and Dené–Yeniseian is a controversial macrofamily deserving a few comments. Since Xiongnu (Hunnish) was Yenisean the American type of language must formerly have been widespread through Asia in the past.

        Dené-Yenisean does actually have some lexical backing and note that words in both Yenisean and Na-Dené for bows and arrows are cognates. Even if the two families have no genetic relationships and the 36 or so cognate words are coincidences and wanderworter, I speculate that Na-Dené as we know it appeared late, under Asian influence, in association with the introduction of the Asian War Complex.

        When the AWC was imported semi-Asianised Americans were able to expand through the northern continent. In the Northwest Coast itself, Haida remained as a surviving isolate influencing Tlingit.

        Really Yenisean demonstrates that American-like languages were formerly present in much of Asia and that Asia continued to influence the languages and material cultures of the proximal New World. But a genetic link between Na-Dené and Yenisean remains unproven.

      • pheltz says:


        How far the Vasconic substrate reaches eastward in Provencal is not really the point. Two pre-IE languages in the same part of Europe = clear motivation to look for a connection.

        I’m not sure which Blench publication you’re referring to, but “Archaeology, Language, and the African Past,” 2006, doesn’t argue for a particularly short Afroasiatic chronology, and suggests a depth of 18,000 years (!) for Nilosaharan.

        If you’re arguing for a shorter time-depth than the consensus for Na-Dene (and my apologies if I misunderstand your position), along the lines of 1300 years to put it in line with the AWC rather than ~5000 years, I’d be interested to hear that argument in more depth. Even so the proposed time depth for Dene-Yeniseian is roughly another 8000 years. And if D-Y isn’t totally agreed upon, there’s at least some body of evidence to discuss.

      • B&B says:

        All I say about Na-Dene is that its speakers benefited greatly from the AWC and that the survival of Yenisean in Siberia demonstrates languages with this typology formerly to have been spoken in much of Asia (the Yenisei is not the most proximal part of Siberia.) Either this corresponds to a genetic relationship or constant contacts round the north Pacific saw the distribution of loanwords connected to the war complex.

        As to the probable time depth of Afro-Asiatic, Blench now sets it as 7,500 years.

  4. Greying Wanderer says:

    “If this picture is correct, it seems likely that almost all of Europe, circa 4000 BC, spoke related languages … And then things went bad for the EEF peoples.”

    I think this is mostly true but

    “The indigenous foragers didn’t disappear instantly, and some seem to have survived amidst the dominant farmers for a long time”

    I think the distribution of y DNA I shows indigenous foragers survived (after a forager to farmer transition) in two main regions: parts of Scandinavia and parts of the Balkans, and at least in the Scandinavian case dramatically increased their range as a component in later expansions

    If correct that would mean Germanic and the languages spoken in the Balkan refuge would have a different substrate to the rest of Europe i.e. those languages would have a forager substrate and the rest would have a Vasconic substrate.

    If so that would that not imply a Rosetta stone could be found by looking for a common substrate between Germanic and the language(s) spoken in the Balkan refuge area shown on the y DNA I distribution map that was different to the Vasconic one?

    • Greying Wanderer says:

      The Balkan refuge (if it is one) is somewhere around Croatia – Slovenia – Bosnia aka Dinaric Alps more or less.

      Coincidentally Scandis and Dinaric Alps people: unusually tall.

      • CamCorder says:

        Funnily enough, Basques have always been noted for their height and athletic build, compared to the surrounding Southern Europeans. But no I.

  5. dearieme says:

    “without much gene flow”: but Neanderthal levels of gene flow are enough to make a difference, it would seem.

  6. Cplusk says:

    Is it possible that Proto-Indo-Hittite speakers moved to Pontic-Caspian steppe through Caucasus from Eastern Anatolia and conquered/assimilated into a larger WHG-ANE people there? They would add the small levels of West Asian component into them and this new population would later expand into Europe.

  7. Greying Wanderer says:


    “The jentilak (‘Giants’), on the other hand, are a legendary people which explains the disappearance of a people of Stone Age culture that used to live in the high lands and with no knowledge of the iron. Many legends about them tell that they were bigger and taller, with a great force, but were displaced by the ferrons, or workers of ironworks foundries, until their total fade-out.”

    Giant cro-magnon women maybe 🙂

    • Niels says:

      @Greying Wanderer

      Whatever happended to the EEF peoples happended before anybody had any knowledge of iron. So the legend seems to be unlikely. Back then people didn’t even have bronze. Hightech of that Age was copper. Don’t think that would be such a big advantage.

      • Greying Wanderer says:

        true, i was mostly joking about the giant cro-magnon women although on the other hand i think you could argue the key point of the legend is knowledge of metal so the detail of whether the metal was copper, bronze or iron might have changed over time.

  8. CamCorder says:

    There is a soccer team in Spain called Athletic Bilbao which has a Basques-only policy, players without Basque heritage are not allowed to play there. Yet and despite losing their best players to the big teams they have constantly played in the first league and often with good results, even on international level.

    Not bad for having to pick your players out of population with a size of 3,000,000. Also, Argentinia, Uruguay, Chile and Columbia have a large Basque heritage (>10%) and those are also good soccer teams. Especially Uruguay which is a very small country but also very successful.

    • Greying Wanderer says:

      If you look at the teams who have consistently appeared in the top 4 or top 8 of the soccer world cup the distribution is very odd. It’s almost all western Europe and south America.

  9. Kate says:

    I don’t know how coincidental this is but, the ten lowest national Gini coefficients of the late 2000s are, in ascending order:
    Slovenia, Denmark, Norway, Czech R., Slovak R., Belgium, Finland, Sweden, Austria, Hungary.

  10. Kate says:

    I expect I’m in the remedial class but here goes,
    When you say ‘from the middle-east’, you don’t mean arabs, do you, you mean otzis, is that correct?
    Is it incontrovertible that otzis descend from a line that moved north from egypt and that, the possibility that otzis descend from a westward migration from SW Asia, is no longer a possibility, in light of genetic evidence?

    • gcochran9 says:

      The farmers settling Europe look to have originated in northern Syria, or thereabouts. Farmers actually started colonizing Europe before Egypt.

    • Matt says:

      Speaking of Egypt and Vennemann, Vennemann thought that the Early Neolithic people spoke Afroasiatic languages, yet we’re thinking here they spoke languages more like Basque.

      So where are we considering Afroasiatic languages to come from? And how were they present in much of the Middle East (not Sumer) by the time we have a historical record. Something analogous to the Indo-European invasions, only perhaps by pastoralist adapted EEF?

      • B&B says:

        Afro-Asiatic comes from the horn of Africa because that’s where the oldest branches remained, and the oldest branch of AA lacks its own pastoralist terminology. From there, Chadic went westward with the Tenerians and North Afroasiatic represents a migration northwards. Berber diverged from Semitic in the Nile Delta probably and proto-Semitic appears in the extreme northern limit of AA expansion, right where the Saharo-Arabian world meets the Palaearctic realm.

        The reconstructed vocabulary for proto-Semitic places it in northern Syria or thereabouts, and this fits Amorite being the first branch of Semitic to diverge. It also fits shared vocabulary between Semitic languages and PIE because speakers of both languages would’ve been in indirect contact early on.

        Sumerian is a language isolate but it was spoken either by a people ‘on top of’ the ‘Ubaids or the ‘Ubaids who moved into southern Iraq, because there is identifiably a non-Sumerian substrate in Sumerian place names. Sumerian itself shows evidence of areal contact with the NEC languages of the Caucasus and also with Hurro-Urartian.

      • minoritymagnet says:

        If you would mind answer some question of an interested linguistics layman?

        How do the semitic languages in south arabia (mehri, soqotri etc.) and ethiopia (amharic etc.) fit into this picture? Back migration over the red sea? What do you suppose proto-semites looked like? Nilotic as people from the horn of africa or rather caucasian?

      • B&B says:

        Proto-Semitic people were Syrians and the bearers of South Semitic into Ethiopia from Arabia were also white people.

      • minoritymagnet says:

        So you’re saying that (nilotic) proto-afro-asiatic speakers (or speakers of a meta-subbranch) migrated to the levant where they imposed their language to “caucasian” peoples there becoming proto-semitic over time. These semites migrated and took over mesopotamia, arabian peninsula and eventually crossed the red sea to ethiopia. So, semitic languages should have a non-afro-asiatic substratum, do they?

      • B&B says:

        Nilotic is not the same as Afro-Asiatic, and the first speakers of the latter were likely of the type thats represented by the (Chadic?) Tenerians and the (Egyptian) Badarians.

        Everywhere there will be substrate languages but the question is asked more often of other parts of the world than others so that its often asked of Germanic, Celtic and Sumerian but not often of Japonic. Within AA the question is mostly asked about the lexical far-outlier Berber, but less often of Semitic to my knowledge except for South Semitic.

  11. Anonymous says:

    Supposedly the Emperor Claudius compiled an Etruscan dictionary. It seems possible that a copy might turn up some day.

  12. Karl Zimmerman says:

    Wouldn’t the supposed non-Uralic substrate in the Saami languages be evidence of the languages spoken by the Western Hunter-Gatherers?

    Regardless, there’s obviously a lot of extant languages left in the Caucasus which could be EEF. The same is probably true for Dravidian – and given the West Eurasian segment of Indian DNA seems to link them closest to the Caucasus, one would expect to see some linguistic relations here. Perhaps even isolates like Burushaski and Nihali, although I think these are more likely to be hunter-gatherer remnants, with the former intrusive from somewhere to the north.

    One thing I find myself wondering is why the story of agriculture in East Eurasia seems so different from West Eurasia, as the agricultural era happened later, yet the languages with clear farmer antecedents are much more diverse. Even if you hold the Austric hypothesis to be true at the broadest level (e.g, encompassing Austronesian, Austro-Asiatic, Tai–Kadai, and Miao-Yao) Sino-Tibetan languages don’t seem related to them at all. And it doesn’t explain why the agriculturalist Koreans and Japanese ended up speaking entirely different (Altaic?) languages.

    • Jim says:

      Greenberg includes Korean, Japanese and Ainu in his Euro-Asiatic group but I recall that he says that aren’t necessarily closer to Altaic than other Euro-Asiatic language.

      So the Japanese postposition “ni” really is “in” spelled backwards.

      • B&B says:

        Altaic may be about areal contact not genetic descent. If you’re going to use Altaic for Mongolic, Turkic and Tungusic you might as well include Korean and Japonic as well.

      • B&B says:

        Ainu is a language isolate but the inferred pattern of its spread and relatively recent origins suggests an Okhotsk rather than a Jomonese origin. Ergo the Emishi were not Ainu people.

      • dearieme says:

        When I was a schoolboy, that people was referred to as The Hairy Ainu. Language was a bit more colourful then.

    • B&B says:

      Interestingly a suite of folklore and words connects upland peoples from the Pyrenees across to the Hindu Kush. There might be reason to place Vasconic, both NWC and NEC, and Burushaski in a macro-family though with such long isolation its now difficult to demonstrate. The words include vocabulary related to food production.

    • B&B says:

      Is this thesis of interest to you?

  13. chrisdavies09 says:

    There appears to have also been quite substantial migration into Southern Europe directly from North Africa in the Neolithic, with a portion of this genetic contribution reaching Northern Europe. If one looks at HLA data it is obvious, and there are strong hints in the mtDNA and Y DNA data.
    Fror example, HLA haplotype: A*29:02-B*44:03-C*16:01-DRB1*07:01-DQB1*02:01
    [Strong linkage disequilibrium].
    Found at varying frequencies in populations right across Europe from Iberia and Italy to Switzerland, France, British Isles, Low Countries, Germany, Scandinavia, etc.
    Peak European frequencies are in Gipuzkoa, Basque country (10%); Girona, Catalonia (7.5%), and Balearic Isles.
    All of its components and significant portions of the haplotype itself are found at high frequencies between Senegal and Cameroon, while the full haplotype is found at high frequency in North West Africa [between Morocco and Tunisia] but at zero frequency in Near East/Middle East.
    The wide distribution of the haplotype in Europe but complete lack of equilibration suggests entry <10kya, probably from the Maghreb [Berbers]. I believe it is a marker for pastoralists arriving in Europe in the Neolithic.
    I don't have HLA data for Syria yet, but I do have Turkey, Israeli Arab Druze, Palestinians, Jordanians, Iraqi Kurds, Lebanese, Saudi Arabians etc. Various other HLA haplotypes found in these populations can also be traced into Europe following different routes, again probably in the Neolithic.

    • gcochran9 says:

      I doubt this. And looking at single loci when whole genomes are available is a mistake.

      • chrisdavies09 says:

        Well there is the ‘Mozabite’ portion in the Spanish genome.
        For me it is rather interesting how the distribution of the HLA A29-Cw16-B44-DR7-DQ2 haplotype appears to closely mirror the map of megalithic culture in Neolithic Europe, along with early pastoralism, presence of U6 and L1/L2/L3 mtDNA markers, presence of E-M81 subclade of E1b1b Y-DNA, etc. Neolithic migrations from North Africa directly into Europe definitely appear to have occurred; the genetic impact is smaller than that of the migrations from the Middle East though.

      • Sandgroper says:

        I picked up a bit of North African in my genome from somewhere. No idea where.

    • B&B says:

      10kya there were no Berbers, but there existed the Mesolithic Capsians. Consult Chamla about their racial type and connections to the Natufians of Palestine. Irish finds their dental traits close to the Kabyles.

  14. spandrell says:

    If EEF languages had been diverging for 6,000 years before the IE came in, they had probably diverged beyond all recognition.
    The Bantu languages are a huge mess of utterly unintelligible languages, and it’s only been 2.5k years.
    But isn’t ergative common to Caucasian languages and Basque? Funny how that disappeared all over Europe.

  15. spandrell says:

    I wonder if there’s any data coming out on the population structure in China?

    Agriculture seems to start in China around 8k BC, then all sorts of Neolithic cultures appear around, with millet farmers in the Yellow River and rice farmers in the Yangtze.

    Pottery seems different all over the place until the Erligang bronze-age culture around 1600BC conquered the whole plain area. First Bronze artifacts are found further West in the oasis corridor, so they probably came from further West.
    Then around 1200BC we have proof of chariot use and literacy from the Shang ruins, in a culture obviously derived from the Erligang conquerors.

    The Shang empire and the successor Zhou had a similar system of fortified garrison towns, controlling an apparently different ethnic stock of farmers in the surrounding countryside.

    I wonder how Cochran’s theory of genetically distinct warriors taking over the European farmers maps to the evidence in China. There are signs of an ethnic warrior elite lording over the indigenous farmers, but no real data on how genetically distinct they were, nor signs of massive disruption of agricultural patterns.

    • SpaghetiMeatball says:

      We have evidence that the people living in some settlements in the oasis corridor – tarim basin in the taklamakan and some nearby places – were part of one of the earliest indo-european waves from the volga-ural. I think I read somewhere about early chinese writing about coming across “strange men with blue eyes, and red hair. They looked like red monkeys.” Or something along those lines. These people might have introduced chariots to the shang. Maybe the shang were even partially descended from them.

      I think they originally referred to all IE speakers from the steppe as the yuezhi. Here’s a sculpted yuezhi warrior’s head:

      And we all know about the tarim mummies.

      • SpaghetiMeatball says:

        Or maybe not. Just did some quick googling to find out the shang started out from northeastern china, so they probably didn’t have any indo-european precedents.

      • spandrell says:

        Chariots aren’t attested in China before the 1200 BC Anyang ruins, and one thinks if the Tarim mummies had chariots they would’ve used them.

      • SpaghetiMeatball says:

        Well, the tarim people lived from about 2000-1000BC.
        And chariots were attested around the time in the steppe. Those were the first chariots.
        So either the early chinese states independently invented chariots or some of the relatives of the tarim guys introduced to the chinese later.

      • anonymous says:

        “…I read somewhere about early chinese writing about coming across “strange men with blue eyes, and red hair…”

        The Kirgiz were a red-haired, green eyed population til they were darkened in the middle ages by Islamic and Chinese introgression. Some Kirgiz still show these traits.

        And of course Genghis Khan and some of his children have long been claimed as redheads. The Khan’s Y-dna is clearly Asian yet there could well have been introgression of redheaded females into the Mongols from the nearby Kirgiz.

      • Sandgroper says:

        If you are interested in the Shang, check out this historically attested sweetheart. I have become somewhat obsessed – there are not actually that many truly historically attested female military leaders. Hua Mulan was probably mythical – if not, she was not Han. Fu Hao was definitely real, and definitely Shang, whoever they were. According to the oracle bones, she was killed by an ‘enemy officer’, so might not have been leading from the rear, as it were.

      • The fourth doorman of the apocalypse says:

        Hua Mulan was probably mythical – if not, she was not Han.

        I think the story of Hua Mulan can be rejected out of hand from what I understand from the summaries I have seen in Chinese. Any female who dressed as a male in those armies would be found out pretty quickly and would become the sexual plaything of the officers … (indeed I was told a joke in Chinese about Hua Mulan that suggests many Chinese males think it is joke worthy.)

        Of course, the fact that Fu Hao is attested to in oracle bones etc simply tells us that such a person existed, they do not really tell us anything about the veracity of her alleged exploits. It is not like the emperor could have had no incentive to embellish someone’s record.

      • Sandgroper says:

        They also found her unlooted tomb, which contained, inter alia, a large number of weapons (very unusual for a female tomb) and the remains of a number of people who were presumably sacrificed during her burial, and the oracle bones do indeed tell of more than just her existence.

        You might want to try being a bit less of a dick.

  16. DK says:

    What about Kartvelian languages? Smack in the middle of Caucasus, a closely related group of non-Indo-European languages with no obvious relationship to anything else (other than a tenuous connection to Basque).

  17. The fourth doorman of the apocalypse says:

    On a different aspect of the last survivor, this response to Steve Sailer’s posting about a review of Wade’s book says:

    The reason is that if we take a population that varies on some characteristic, such as height, and eliminate the bottom 10 percent of the height distribution in one generation then we will for all time change the average height of that population. This is because we have changed the average underlying average genotype of the population

    However, it seems to me that a single generation of truncated selection will not eliminate alleles for low height from the gene pool, only reduce the number of individuals carrying them, and unless many more generations of such truncated selection are applied, the population could well regress to the mean of the original population over time.

    Can anyone point me to analyses of this?

    • gcochran9 says:

      It reduces the gene frequency ( in the population) of alleles for low height. And as long as there is no further selection, they stay reduced. People will be shorter.

      In many cases, there is stabilizing selection in nature. If you bred a rat population that was shorter than average, then released them into the wild, the odds are that they would gradually go back to typical rat size, because that’s the size favored by selection. Feral animals often converge to certain phenotypes: pigs, horses, etc. In the lab, they might well stay short.

      You know, hardly anyone in public life understands this. Except Freeman Dyson, of course.

      • The fourth doorman of the apocalypse says:

        It reduces the gene frequency ( in the population) of alleles for low height. And as long as there is no further selection, they stay reduced. People will be shortertaller.

        I agree, but am wondering if there is some sort of impulse response that can be stated in generations?

      • The fourth doorman of the apocalypse says:

        And as long as there is no further selection, they stay reduced. People will be shortertaller.

        OK, that is the important criteria … now I understand. It’s kinda like the US Gov selecting for deadbeat dads.

      • Jim says:

        People sometimes erroneously think of “regression to the mean” as a casual effect.

      • Matt says:

        Although with in humans and IQ, in reality, there might be more an effect of people making their own environment.

        Remove the bottom 10%, people are smarter, they make a more demanding (or less demanding) system to suit their ability, this fixes the selective optimum higher (or lower).

        Perhaps particularly the case with potentially mercantile groups. Remove the bottom x, the inbreeding survivors move disproportionately into jobs y, this keeps them at ability level z.

        Or perhaps I am overstating the power of this trait to allow humans to reshape their own environment (or at any rate to change the environmental optimum) compared to other traits.

  18. The fourth doorman of the apocalypse says:

    OT, but in Why you can thank your mother for your IQ the author says:

    As I also pointed out in the previous post, there are at least 150 genes linked to intelligence on the X chromosome, and verbal IQ is definitely known to be X-linked. A recent study found that compared to male, female identical twins vary more on measures of social behaviour and verbal ability thanks to differential X-inactivation. This has a number of important implications. One is that as medical geneticists have long suspected, you get your intelligence predominantly from your mother, not your father—especially if you are male.

    Now, I would have expected that with X-inactivation being random, and given the number of cells involved when the brain is developing, any dosage dependent effects would be the same for identical twins to within about 1 in 10^-10 or something like that.

    What am I missing, or is he blowing smoke?

    • gcochran9 says:

      As for getting your intelligence from your mother, not your father: it isn’t true. Now it might be that you get a little more than 50% from your mother – say 52.5 %. I could believe that.

      “Because lyonization occurs fairly early in development, many of the progeny of a single embryonic stem cell are grouped together in the adult, forming patches.”

      In calico cats, the patches are big enough to see. In the brain, it might depend on where a given patch is: how important the cells, what part of the brain, how big the patch. etc.

      • The fourth doorman of the apocalypse says:

        I guess we could assume that the neuroectoderm develops from a single cell, but that would seem to contradict both Wikipedia and the reduced variance in female IQ compared to males:

        The nervous system is derived from the ectoderm – the outermost tissue layer – of the embryo. In the third week of development the neuroectoderm appears and forms the neural plate along the dorsal side of the embryo. This neural plate is the source of the majority of neurons and glial cells in the mature human.

      • Zerg says:

        Extremely anecdotal evidence leads me to believe that raw intelligence might be mostly from Mom, but that thoughtfulness (depth) depends on Dad. I wonder whether there’s a genetic basis for thoughtfulness, as opposed to intelligence. These two characteristics (intelligence and thoughtfulness) are certainly at least somewhat independent of each other. Intellectual character (as opposed to intellectual ability) is probably somewhat environmentally influenced, though. Imagine a kid bouncing back and forth between divorced parents — a typically semi-illiterate modern career mom and a sensitive nerd loser dad who reads the kid all of Tolkien, the Iliad and Odyssey, the Aeneid, etc. — wouldn’t the kid’s mind have turned out very differently by the time he grows up if he hadn’t spend a lot of time with his loser dad? (Same as the “What if I’d grown up in 7th century Arabia” scenario, I guess.)

      • IC says:

        Pure reduction and speculation here. The question can be easily answered by collecting datat from IQ mismached couples.

        • gcochran9 says:

          If there were any significant difference from 50-50 we’d already know. There isn’t.

          Not that any possible result would keep people from blathering on and on about it.

    • Greying Wanderer says:

      If i was a billionaire I’d be doing that – searching for sunken temples in the areas that were most recently flooded after the LGM. If there were any to the SW of Britain e.g. near the Scilly Isles, they’d possibly have a ton of gold artifacts as well.

  19. The fourth doorman of the apocalypse says:

    OT, but Sam Wang responds to Dienekes’ objection to his characterization of Cluster in his (Wang’s) criticism of Wade’s book:

    @dienekesp Hmmm, reading your work. You are not the target audience for this! This was originally commissioned for a lay audience. now shoo!

    Hmmm, so it is OK to be wrong/confusing when you are writing for a general audience?

    • Kate says:

      A question please:

      Supposing someone wanted to see if the AAA’s social construction of geographic ancestry had any genetic basis, so they selected K=5 to reflect 5 populations isolated geographically until around 1500 by the Sahara, the Urals, the Indian and Pacific oceans and the Bering Strait. And they found that 5 clusters emerged.

      Is that how the programme works? In which case, why do liberals think the answer is pre-determined and, why do realists not explain that they are testing the liberals’ own environmentally-derived categories ?

      Or, is the the data manipulated more than through the K number in order to obtain 5 clusters?

  20. I have literally no idea how accepted that idea is among the scientific community, but this pretty much confirms what you were saying, doesn’t it?

  21. Matthew M. Robare says:

    Blasted Egyptians. If they had been the slightest bit curious about their world, like the Classical Athenians, they could have recorded all this stuff because they were trading and fighting with all the peoples we have questions about. But no, they had to be deliberately obtuse and incurious.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s