The Unz Review - Mobile
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 Gene Expression BlogTeasers
Native Americans Are Evolutionarily Elegant
Search Text Case Sensitive  Exact Words  Include Comments

James C. Chatters 2002 book

James C. Chatters 2002 book

By now you may have read the breaking news in The Seattle Times that Eske Willerslev’s group is going to publish genetic results on Kennewick Man. This “scoop” was obtained through the freedom of information act, which makes sense since Kennewick Man has been embroiled in political controversy since the beginning of its discovery by James Chatters in the 1990s. The issue is that morphologically the remains were not typical of contemporary Native Americans, which might cause some doubt as to the legitimacy of the social-political rights of the indigenous people of the region today. The social-political aspects have been beaten to death, and I am not particularly interested in that area. Rather, the science is more fascinating, if, somewhat less surprising in light of the results that are going to come out in the near future.

2019387254The most famous reconstruction of Kennewick Man is strange because it resembles British actor Patrick Stewart. Humans use phenotypes, morphology, to ascertain genetic relatedness when DNA is not available. In the 1990s DNA was not available. The inference by many researchers who had access to the remains was that Kennewick Man was different because his morphology may have resembled a person of European heritage. The controversy turned into such a circus that somehow Steve McNallen, arguably America’s foremost Northern European neo-pagan expositor, made claims on the remains on the same grounds as Native American people! Later scholars suggest that perhaps Kennewick Man was not so much European, as not typical of contemporary Native Americans (e.g., perhaps he was part of an early migration of basal East Eurasians related to the Jomon of Japan).

41VAznr2aiL._SX258_BO1,204,203,200_ If the Seattle Times report is correct, and I believe it is, Kennewick Man is part of the ancestral population to modern Native Americans. This should put to bed most of the political debate, since the results are likely to mollify many Native activists. But, there are still details to be fleshed out. A 2012 publication suggests that there was a secondary migration out of Eurasia, which resulted in the Na-Dene group which is common in the northern and western portions of North America. In contrast, Kennewick Man is likely to belong to the first ur-North Americans, who arrived as a relatively small population from Berengia ~15,000 years ago. This is the overwhelming majority of indigenous ancestry, and south of the Rio Grande basically the totality.*

Due for an update!

Due for an update!

The context here is important. One insight of modern ancient DNA is that there has been a great deal of population turnover over the past ~10,000 years, as well as admixture between disparate lineages. When Kennewick Man died ~9,000 years ago Europeans as we understand them did not exist genetically. All across Eurasia, Africa, and Oceania, the Holocene brought radical demographic turnover (with some exceptions such as the Andaman Islands and the deserts of southwest Africa). The New World was somewhat different, as I implied above. There were some demographic disruptions, but south of the Rio Grande, and across the eastern half of North America, the populations descend from a relatively homogeneous founder stock which arrived at the end of the Pleistocene. The fact that many remains seem “atypical” for the morphology of Native Americans is strong evidence of in situ evolution.**

Years ago a physical anthropologist told me that when you look at Amazonian natives they “looked” like Siberians. Yes, they had changed and adapted, but only somewhat. It illustrated to me the powerful constraint of limited genetic variation upon populations. Similarly, though there is variation in pigmentation among native populations in the New World, it is far less than you see in the Old World. Why? Perhaps it is a function of different (or lack thereof) of selective pressures. Or, perhaps the variation wasn’t there for selection in the first place? The history of the Old World has jumbled all our easy narratives. The New World may actually be a godsend because of the simple elegance of its demographic history.

* From my Twitter exchanges with Pontus Skoglund I believe there is some population structure in the founding “First American” group, though not a great deal.

** Admixture is an issue, but that can be obviated by genetic testing, as well as looking at early modern remains.

 
Email This Page to Someone

 Remember My Information



=>
Commenters to Ignore...to FollowEndorsed Only
    []
  1. It’s fascinating, but really how useful can one morphological sample be in working out these ancestral lineage?

    Read More
    ReplyAgree/Disagree/Etc.
    AgreeDisagreeLOLTroll
    These buttons register your public Agreement, Disagreement, Troll, or LOL with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used once per hour.
    Sharing Comment via Twitter
    //www.unz.com/gnxp/native-americans-are-evolutionary-experiments/#comment-852466
    More... This Commenter Display All Comments
  2. #1, N = 1 can be quite useful depending on context. thought experiment: imagine you found a skeleton in new south wales dating to 40,000 years ago which resembles modern day northeast asians (e.g., shovel shaped incisor, etc.). you’d have some explaining to do. in any case, with native americans a lot of the older skeletons look atypical. so it’s pretty clear that kennewick isn’t an aberration.

    Read More
  3. ” across the eastern half of North America, the populations descend from a relatively homogeneous founder stock which arrived at the end of the Pleistocene.”

    I read A book “The Comanches Death of A People” by T.R. Feherenbach. In the introduction he gives a general description of the Amerindians .One thing that struck me was that their languages were so splintered , he gives a number of 140 language stocks with no common root. If I understand correctly the European languages although unintelligible to each other have a common Indo-European root . One stock.
    These tribes were all living side by . He also mentions that they each had distinctive natural behaviors which weren’t shared with the others surrounding them. The book was written in 1974 so I suppose some of it could be outdated .
    I guess my question is if as you say there was homogenous founder stock why such big and varied differences in language and culture, especially language ?

    Read More
    • Replies: @Tobus
    @donut:why such big and varied differences in language and culture, especially language


    When a small population enters a massive expanse of unpopulated and highly habitable land, it spreads out and forms isolated groups of independent civilisations/cultures that don't need to compete for the same land. If the area is small (like Europe) or when it reaches population saturation (like Africa) individual populations must compete (or cooperate!) and thus mix with each other, sharing language, culture and genes. There's evidence of this happening in Europe for at least the last 5,000 years, while Native Americans have some of the highest between group Fst values in the world, suggesting relative isolation of the different subpopulations for about 15,000.
    , @jtgw
    Language families don't necessarily correspond to genetic families in the first place, though they often do simply because for the most part children learn their native language from parents and neighbors who are typically genetically related to them. But secondly, language relationships can only be demonstrated for languages that have diverged up to about 10,000 years ago, and the ancestral stock of most Native Americans arrived in America well before that. The reason is that you can only demonstrate relationships between sets of inherited words and other morphemes (e.g. prefixes and suffixes), but languages also replace their inherited morphemes with borrowed words, coinages and by other innovations. After about 10,000 years, most languages will retain too few ancestral morphemes to allow any plausible relationship to be demonstrated.
    , @CupOfCanada
    Just because a link can't be found doesn't mean it doesn't exist. Indo-European is 5,000 years old or so, and has been written down for roughly 3,000 years. For Amerindian languages, we're talking a common origin potentially 13,000 years ago, with the first written sources 1,700 years ago, and in that case only for one language. There's 11,000 years+ that we can't document and that is very hard to reconstruct.

    If I understand correctly the European languages although unintelligible to each other have a common Indo-European root . One stock.
     
    Not all languages in Europe are Indo-European. There are also Basque, Kartvelian, North Caucasian, Mongolic, Turkic, Uralic (these last 3 may be related) and Semitic languages. So not one identifiable stock by any stretch. And the Americas are much larger than Europe.

    There's evidence of other now extinct languages of unknown affiliation that were wiped out with the various migrations from the Neolithic onwards. This process was still incomplete in the Americas at the time of contact.
  4. anon says:     Show CommentNext New Comment

    I guess my question is if as you say there was homogenous founder stock why such big and varied differences in language and culture, especially language ?

    Violence?

    Read More
  5. Razib – is this population structure you refer to the Algonquin group Reich’s paper refers to? I think the evidence for that structure is pretty persuasive really. I’d be curious to see where Kennewick Man would place on that tree in the Reich paper – though I’d wager he’d be basal to Zapotec1/Karitania and not the others. He’d be interesting to compare directly to Anzick too – I’d bet they were very closely related.

    Read More
  6. @donut
    " across the eastern half of North America, the populations descend from a relatively homogeneous founder stock which arrived at the end of the Pleistocene."

    I read A book "The Comanches Death of A People" by T.R. Feherenbach. In the introduction he gives a general description of the Amerindians .One thing that struck me was that their languages were so splintered , he gives a number of 140 language stocks with no common root. If I understand correctly the European languages although unintelligible to each other have a common Indo-European root . One stock.
    These tribes were all living side by . He also mentions that they each had distinctive natural behaviors which weren't shared with the others surrounding them. The book was written in 1974 so I suppose some of it could be outdated .
    I guess my question is if as you say there was homogenous founder stock why such big and varied differences in language and culture, especially language ?

    :why such big and varied differences in language and culture, especially language

    When a small population enters a massive expanse of unpopulated and highly habitable land, it spreads out and forms isolated groups of independent civilisations/cultures that don’t need to compete for the same land. If the area is small (like Europe) or when it reaches population saturation (like Africa) individual populations must compete (or cooperate!) and thus mix with each other, sharing language, culture and genes. There’s evidence of this happening in Europe for at least the last 5,000 years, while Native Americans have some of the highest between group Fst values in the world, suggesting relative isolation of the different subpopulations for about 15,000.

    Read More
    • Replies: @jtgw
    Johanna Nichols showed that geography often played a role in determining whether a given area had much or little linguistic diversity. Mountainous and coastal regions and islands, for example, typically have high diversity, e.g. California, the Caucasus, New Guinea, each of which boasts several unrelated language families and language isolates; steppe and grass plains, on the other hand, typically have low diversity, e.g. the Great Plains or Central Asia, dominated by just a few language families (Algonquian and Macro-Siouan for the former, Turkic and Iranian for the latter).
    , @donut
    Thanks
  7. Didn’t Greenberg establish that the Amerindian languages are related, consisting of three groups?

    http://en.wikipedia.org/wiki/Amerind_languages

    Read More
    • Replies: @jtgw
    His conclusions are not accepted by most historical linguists, the main criticism being that he did not properly control for chance resemblances in his proposed cognates, i.e. words that sound similar and mean the same thing may just be flukes, such as English and Mbabaram (Australian aboriginal language) dog, both meaning "dog" but completely unrelated historically. You can look up critiques by Lyle Campbell, William Poser and Donald Ringe for more details on why such proposals for long-range relationships haven't been successful.
    , @pheltz
    As jtgw notes, Greenberg's methods aren't reliable and his work doesn't really shed much light on the subject--he doesn't make real allowances for chance resemblances, or for borrowing and other contact phenomena (which to be fair are quite hard to control for), and he doesn't really build on the existing demonstrated language groups or the reconstructions thereof.

    None of this is to say that his conclusions are wrong. Obviously they're quite plausible, and the alternatives are a bit hairy.
  8. @donut
    " across the eastern half of North America, the populations descend from a relatively homogeneous founder stock which arrived at the end of the Pleistocene."

    I read A book "The Comanches Death of A People" by T.R. Feherenbach. In the introduction he gives a general description of the Amerindians .One thing that struck me was that their languages were so splintered , he gives a number of 140 language stocks with no common root. If I understand correctly the European languages although unintelligible to each other have a common Indo-European root . One stock.
    These tribes were all living side by . He also mentions that they each had distinctive natural behaviors which weren't shared with the others surrounding them. The book was written in 1974 so I suppose some of it could be outdated .
    I guess my question is if as you say there was homogenous founder stock why such big and varied differences in language and culture, especially language ?

    Language families don’t necessarily correspond to genetic families in the first place, though they often do simply because for the most part children learn their native language from parents and neighbors who are typically genetically related to them. But secondly, language relationships can only be demonstrated for languages that have diverged up to about 10,000 years ago, and the ancestral stock of most Native Americans arrived in America well before that. The reason is that you can only demonstrate relationships between sets of inherited words and other morphemes (e.g. prefixes and suffixes), but languages also replace their inherited morphemes with borrowed words, coinages and by other innovations. After about 10,000 years, most languages will retain too few ancestral morphemes to allow any plausible relationship to be demonstrated.

    Read More
    • Replies: @donut
    OK Thanks.
    , @ohwilleke
    The open access paper Sijia Wang, et al., "Genetic Variation and Population Structure in Native Americans" (PLOS Genetics 2007) provides a rather more optimistic prospect for using genetics to reconstruct the higher level connections between the many small linguistic families of South America to the east of the Andes Mountains, and one Andean language, on a largely tree-like basis that reproduces the linguistic relationships that are established while going further to suggest macrolinguistic family relationships that aren't very solidly established linguistically.

    It has been difficult to establish macrofamilies of languages linguistically in South America due to issues like a shortage of linguists to study a highly linguistically diverse region and poor documentation of many languages, key missing links that are extinct or moribund and not well documented, and great expected time depth separating related language families. An institutional culture among South American linguists that has favored a conservative approach that leans towards splitting rather than lumping (famous in Russian school linguists) is also a factor.

  9. @bob sykes
    Didn't Greenberg establish that the Amerindian languages are related, consisting of three groups?

    http://en.wikipedia.org/wiki/Amerind_languages

    His conclusions are not accepted by most historical linguists, the main criticism being that he did not properly control for chance resemblances in his proposed cognates, i.e. words that sound similar and mean the same thing may just be flukes, such as English and Mbabaram (Australian aboriginal language) dog, both meaning “dog” but completely unrelated historically. You can look up critiques by Lyle Campbell, William Poser and Donald Ringe for more details on why such proposals for long-range relationships haven’t been successful.

    Read More
  10. @Tobus
    @donut:why such big and varied differences in language and culture, especially language


    When a small population enters a massive expanse of unpopulated and highly habitable land, it spreads out and forms isolated groups of independent civilisations/cultures that don't need to compete for the same land. If the area is small (like Europe) or when it reaches population saturation (like Africa) individual populations must compete (or cooperate!) and thus mix with each other, sharing language, culture and genes. There's evidence of this happening in Europe for at least the last 5,000 years, while Native Americans have some of the highest between group Fst values in the world, suggesting relative isolation of the different subpopulations for about 15,000.

    Johanna Nichols showed that geography often played a role in determining whether a given area had much or little linguistic diversity. Mountainous and coastal regions and islands, for example, typically have high diversity, e.g. California, the Caucasus, New Guinea, each of which boasts several unrelated language families and language isolates; steppe and grass plains, on the other hand, typically have low diversity, e.g. the Great Plains or Central Asia, dominated by just a few language families (Algonquian and Macro-Siouan for the former, Turkic and Iranian for the latter).

    Read More
  11. @Tobus
    @donut:why such big and varied differences in language and culture, especially language


    When a small population enters a massive expanse of unpopulated and highly habitable land, it spreads out and forms isolated groups of independent civilisations/cultures that don't need to compete for the same land. If the area is small (like Europe) or when it reaches population saturation (like Africa) individual populations must compete (or cooperate!) and thus mix with each other, sharing language, culture and genes. There's evidence of this happening in Europe for at least the last 5,000 years, while Native Americans have some of the highest between group Fst values in the world, suggesting relative isolation of the different subpopulations for about 15,000.

    Thanks

    Read More
  12. @jtgw
    Language families don't necessarily correspond to genetic families in the first place, though they often do simply because for the most part children learn their native language from parents and neighbors who are typically genetically related to them. But secondly, language relationships can only be demonstrated for languages that have diverged up to about 10,000 years ago, and the ancestral stock of most Native Americans arrived in America well before that. The reason is that you can only demonstrate relationships between sets of inherited words and other morphemes (e.g. prefixes and suffixes), but languages also replace their inherited morphemes with borrowed words, coinages and by other innovations. After about 10,000 years, most languages will retain too few ancestral morphemes to allow any plausible relationship to be demonstrated.

    OK Thanks.

    Read More
  13. @bob sykes
    Didn't Greenberg establish that the Amerindian languages are related, consisting of three groups?

    http://en.wikipedia.org/wiki/Amerind_languages

    As jtgw notes, Greenberg’s methods aren’t reliable and his work doesn’t really shed much light on the subject–he doesn’t make real allowances for chance resemblances, or for borrowing and other contact phenomena (which to be fair are quite hard to control for), and he doesn’t really build on the existing demonstrated language groups or the reconstructions thereof.

    None of this is to say that his conclusions are wrong. Obviously they’re quite plausible, and the alternatives are a bit hairy.

    Read More
  14. Strange how Greenberg’s work on African language families is so widely accepted while his work on Native American languages is vilified. Are peoples’ pre-existing ideas to blame? And strange how his three language groups has been shown to coincide with genetic evidence.

    Read More
    • Replies: @CupOfCanada
    Last paper I read, Greenberg's language families in the Americas had no predictive power on genetics actually. I can't recall off hand, but there was a paper that actually directly tested this.
    , @jtgw
    That Greenberg's reconstructions for African languages are widely accepted is news to me. Where did you get that from? Greenberg is the Peter Duesberg of historical linguistics.
  15. @donut
    " across the eastern half of North America, the populations descend from a relatively homogeneous founder stock which arrived at the end of the Pleistocene."

    I read A book "The Comanches Death of A People" by T.R. Feherenbach. In the introduction he gives a general description of the Amerindians .One thing that struck me was that their languages were so splintered , he gives a number of 140 language stocks with no common root. If I understand correctly the European languages although unintelligible to each other have a common Indo-European root . One stock.
    These tribes were all living side by . He also mentions that they each had distinctive natural behaviors which weren't shared with the others surrounding them. The book was written in 1974 so I suppose some of it could be outdated .
    I guess my question is if as you say there was homogenous founder stock why such big and varied differences in language and culture, especially language ?

    Just because a link can’t be found doesn’t mean it doesn’t exist. Indo-European is 5,000 years old or so, and has been written down for roughly 3,000 years. For Amerindian languages, we’re talking a common origin potentially 13,000 years ago, with the first written sources 1,700 years ago, and in that case only for one language. There’s 11,000 years+ that we can’t document and that is very hard to reconstruct.

    If I understand correctly the European languages although unintelligible to each other have a common Indo-European root . One stock.

    Not all languages in Europe are Indo-European. There are also Basque, Kartvelian, North Caucasian, Mongolic, Turkic, Uralic (these last 3 may be related) and Semitic languages. So not one identifiable stock by any stretch. And the Americas are much larger than Europe.

    There’s evidence of other now extinct languages of unknown affiliation that were wiped out with the various migrations from the Neolithic onwards. This process was still incomplete in the Americas at the time of contact.

    Read More
    • Replies: @jtgw
    If the first wave of migration into the Americas happened in a short space of time, it's highly plausible these ur-Americans spoke closely related languages or even one language, and therefore it's more plausible that all current Amerindian languages have a common ancestor. But all we can do is guess this based on the genetics: there is no direct linguistic evidence for such relationships. That's the point.
    , @donut
    Yes I've read that about the Basque and Finnish languages and others that are isolated and unrelated to any others around them . But what I find so fascinating is the question of how they came to be there .And I guess no one can answer that question , or can they
  16. @terryt
    Strange how Greenberg's work on African language families is so widely accepted while his work on Native American languages is vilified. Are peoples' pre-existing ideas to blame? And strange how his three language groups has been shown to coincide with genetic evidence.

    Last paper I read, Greenberg’s language families in the Americas had no predictive power on genetics actually. I can’t recall off hand, but there was a paper that actually directly tested this.

    Read More
  17. @terryt
    Strange how Greenberg's work on African language families is so widely accepted while his work on Native American languages is vilified. Are peoples' pre-existing ideas to blame? And strange how his three language groups has been shown to coincide with genetic evidence.

    That Greenberg’s reconstructions for African languages are widely accepted is news to me. Where did you get that from? Greenberg is the Peter Duesberg of historical linguistics.

    Read More
  18. @CupOfCanada
    Just because a link can't be found doesn't mean it doesn't exist. Indo-European is 5,000 years old or so, and has been written down for roughly 3,000 years. For Amerindian languages, we're talking a common origin potentially 13,000 years ago, with the first written sources 1,700 years ago, and in that case only for one language. There's 11,000 years+ that we can't document and that is very hard to reconstruct.

    If I understand correctly the European languages although unintelligible to each other have a common Indo-European root . One stock.
     
    Not all languages in Europe are Indo-European. There are also Basque, Kartvelian, North Caucasian, Mongolic, Turkic, Uralic (these last 3 may be related) and Semitic languages. So not one identifiable stock by any stretch. And the Americas are much larger than Europe.

    There's evidence of other now extinct languages of unknown affiliation that were wiped out with the various migrations from the Neolithic onwards. This process was still incomplete in the Americas at the time of contact.

    If the first wave of migration into the Americas happened in a short space of time, it’s highly plausible these ur-Americans spoke closely related languages or even one language, and therefore it’s more plausible that all current Amerindian languages have a common ancestor. But all we can do is guess this based on the genetics: there is no direct linguistic evidence for such relationships. That’s the point.

    Read More
    • Replies: @CupOfCanada
    Thing is, the genetic evidence is for at least 3 different migrations, and possibly a fourth, and these don't match with Greenberg's classifications. Algonquians are classified under Amerind by Greenberg but appear to be from a separate (perhaps simultaneous) migration to these ur-Americans.
  19. @jtgw
    If the first wave of migration into the Americas happened in a short space of time, it's highly plausible these ur-Americans spoke closely related languages or even one language, and therefore it's more plausible that all current Amerindian languages have a common ancestor. But all we can do is guess this based on the genetics: there is no direct linguistic evidence for such relationships. That's the point.

    Thing is, the genetic evidence is for at least 3 different migrations, and possibly a fourth, and these don’t match with Greenberg’s classifications. Algonquians are classified under Amerind by Greenberg but appear to be from a separate (perhaps simultaneous) migration to these ur-Americans.

    Read More
    • Replies: @jtgw
    Ah I see - fascinating! Well, my impression was that the genetic evidence was the strongest thing Greenberg had going for his theory, so it's interesting to see even that doesn't work. But I'd be surprised if Algonquian speakers immigrated 15,000 years ago: one would expect their daughter languages to have diverged too much for any recognizable relationship between them to be apparent. Also, by Algonquian do they mean Algic? The Algonquian languages in fact form just one branch of a family that also includes, amazingly, a couple of languages native to northern California (Yurok and Wiyot).
  20. @CupOfCanada
    Just because a link can't be found doesn't mean it doesn't exist. Indo-European is 5,000 years old or so, and has been written down for roughly 3,000 years. For Amerindian languages, we're talking a common origin potentially 13,000 years ago, with the first written sources 1,700 years ago, and in that case only for one language. There's 11,000 years+ that we can't document and that is very hard to reconstruct.

    If I understand correctly the European languages although unintelligible to each other have a common Indo-European root . One stock.
     
    Not all languages in Europe are Indo-European. There are also Basque, Kartvelian, North Caucasian, Mongolic, Turkic, Uralic (these last 3 may be related) and Semitic languages. So not one identifiable stock by any stretch. And the Americas are much larger than Europe.

    There's evidence of other now extinct languages of unknown affiliation that were wiped out with the various migrations from the Neolithic onwards. This process was still incomplete in the Americas at the time of contact.

    Yes I’ve read that about the Basque and Finnish languages and others that are isolated and unrelated to any others around them . But what I find so fascinating is the question of how they came to be there .And I guess no one can answer that question , or can they

    Read More
    • Replies: @jtgw
    Basque seems to be indigenous to the area, going by place names, and were already there in Roman times. I think most agree that they represent a pre-Indo-European stock in that part of Europe; there is no reason to think they came from anywhere else after Indo-European speakers arrived, which in Spain I think took place with the influx of Celtic speakers in the Iron Age. Before then is speculation, though I expect they can be placed plausibly among the Neolithic farmer migrants.

    Finnish is a Uralic language, and proto-Uralic was most likely spoken in central Russia, pretty much due north of where proto-Indo-European speakers lived, which was most likely eastern Ukraine.

    For Indo-European, the book to read now is David Anthony The Horse, the Wheel and Language.
    , @ohwilleke
    There are very solid theories in each case.

    * Basque is probably a relic community of what was once a vast Vasconic linguistic community in Western Europe and the Northern coastal areas of continental Europe associated with the archaeological Copper Age Bell Beaker culture that expanded from an origin in SW Iberia starting ca. 2900 BCE, with origins elsewhere, that filled the gap created by the collapse of first wave Neolithic agricultural societies and lead to massive expansion of Y-DNA R1b and mtDNA H (with a minor mtDNA X2 contribution) in the regions where it was powerful. This society kept Indo-European cultures associated with the contemporaneous Corded Ware people at bay for about 1000 years, but eventually, Indo-European Celts and Germanic populations prevailed around the time of Bronze Age collapse (ca. 1200 BCE) almost everywhere but Basque country, although the transition took time. There were additional Vasconic languages spoken near current Basque country at the time of first Roman contract. The evidence suggesting this hypothesis can't easily be summed up in a single comment. European languages in the former Bell Beaker region show some common Vasconic language substrate influences (e.g. residual base twenty number names like "seventeen") and traces of Vasconic language influences in place names.

    The source of the proto-Bell Beaker people who formed the Bell Beaker culture and expanded out of Southern Iberia is cryptic.

    Based upon genetic and linguistic hints (I don't claim that there is definitive evidence although it is plausible), and the archaeological cultures in existence at the time, my best guess would be that their origins were in the Southern Caucasus mountains (likely in the Kura-Araxes culture which was strongly culturally influenced by the better known Maikop culture (aka Maykop culture) (ca. 3700 BCE-3000 BCE) of the Northern Caucasus), possibly via the Minoan society or having a common origin with it, with Basque probably belonging to the same macrolinguistic family as Kartevelian at a time depth of about 5000 years (e.g. a similar time depth to linguistic common origin of Hindi and English).

    Basically, some of the main hints are: (1) Basque is a ergative language like the modern North and South Caucasian languages and a number of now extinct languages nearby like Sumerian and the Elamite language, (2) the proto-Basque population was rich in Y-DNA R1b and mtDNA H with a minor X2 contributions, something which is also present in the Southern Caucasus (which is much closer to the place of origin of these haplogroups geographically) and in Minoan ancient DNA, (3) the ABO blood type frequencies in the Southern Caucasus are very similar to those of areas like Scotland that had heavy Bell Beaker demic impact, and (4) the Southern Caucasus were home to one of the earliest metal age cultures of Europe at about the right time and the Bell Beaker people were known for their metal working. Most alternatives to this model have serious problems on the linguistic, archaeological culture timing, or genetic front. It is hard to tell if sea routes or overland routes were most likely. A sea route and an intermediate link in the Minoan culture is attractive because the cultural role of the Bull is prominent in both Minoan and the oldest known Basque cultures.

    * The origins of the Finnish language are much better understood and the subject of a recent post and discussion at this blog. It is not a language isolate and instead is part of a larger Uralic language family (named after a presumed homelands in the Ural mountains that divide North Asia from European Russia). It probably arrived in Finland around 1500 BCE coinciding with the advent of the Finnish Bronze Age, in a process of language shift that replaced an Indo-European language was part of the Corded Ware culture language family which replaced an even older and completely lost language of Finland's Comb Ceramic culture of maritime hunter-gatherers around ca. 2500-2300 BCE coinciding with the introduction of agriculture.

    Finnish shares a macro-language family with Estonian, Hungarian and a number of less well known languages of ethnic enclaves in Russia (e.g. Mari and Komi). We can make some quite informed guesses about which components of the current population genetics of Finland correspond to its Mesolithic, Indo-European Neolithic and Uralic Bronze migration strata, although Swedish and Russian Slavic migration to Finland in the historic era complicate this effort somewhat.

  21. @CupOfCanada
    Thing is, the genetic evidence is for at least 3 different migrations, and possibly a fourth, and these don't match with Greenberg's classifications. Algonquians are classified under Amerind by Greenberg but appear to be from a separate (perhaps simultaneous) migration to these ur-Americans.

    Ah I see – fascinating! Well, my impression was that the genetic evidence was the strongest thing Greenberg had going for his theory, so it’s interesting to see even that doesn’t work. But I’d be surprised if Algonquian speakers immigrated 15,000 years ago: one would expect their daughter languages to have diverged too much for any recognizable relationship between them to be apparent. Also, by Algonquian do they mean Algic? The Algonquian languages in fact form just one branch of a family that also includes, amazingly, a couple of languages native to northern California (Yurok and Wiyot).

    Read More
    • Replies: @CupOfCanada
    I meant Algonquian because I don't think any Yurok and Wiyok were included in the study as far as I am aware, so I didn't want to assume anything for Algic as a whole.

    From Reich et al 2012:

    The 47 most southern Native American populations are consistent with descending from a single peopling event (all statistics |Z|<3; Table S5). However, 5 Northern Native American (NNA) populations—Ojibwa, Cree, Algonquin, Cheyenne and Chipewyan—have Z-scores 3-6 standard errors from expectation, and are also outliers in population structure analyses (Figure 1B and Figure 2). Further examination of the values of the 4 Population Test statistics demonstrates two distinct patterns of relationships to Arctic and East Asian populations among these 5 NNA groups. The statistics for four of the NNA (Cheyenne, Cree, Ojibwa and Algonquin) are highly correlated (average r2 =0.72; Figure S6) and indicate a closer relationship of these populations to the Inuit than to any Asian group (Figure 2B and Table S6). By contrast, statistics involving the Chipewyan are not correlated to the other four NNA (r2 =0.05; Figure 2C; Table S6), suggesting distinct gene flows with Asians. Globally, these findings show that Native Americans break into three broad groups: the 47 Native American populations from Meso-America southward, the Inuit along with 4 NNA populations with whom they appear to have exchanged genes, and the Chipewyan who speak a Na-Dene language. This is consistent with the controversial22 three migration wave model of Greenberg which views Inuit and Na-Dene languages as markers for distinct migrations from Asia9, although not with the purest form of that model which would specify that the Inuit and Chipewyan represent sister groups to some Siberian populations, whereas in fact they cluster with Native Americans (Figure 1C), consistent with subequent admixture within the Americas. Intriguingly, Greenberg’s hypothesis that Na-Dene marks a distinct migration with Asia has been supported by recent linguistic work that shows that Na-Dene language have a link with Siberian Yeniseian languages23. The group of Siberian populations with which the Chipewyan show the strongest genetic affinity includes the Ket, the sole living speakers of Yeniseian (Table S6).
     
    Link to the paper: http://genetics.med.harvard.edu/reich/Reich_Lab/Welcome_files/2012_Nature_NativeAmericans.pdf

    If I recall correctly, the fundamental idea that Greenberg presents is more or less supported by genetics, but that his classification of specific groups and the relationships between those groups is not entirely correct. Also, a lot of it rests on the assumption of a 1:1 correspondence between genetic and linguistic origins, which isn't always true. For example, when my family switched from speaking Scots Gaelic to speaking English ~180 years ago, we didn't also change our DNA.

    Achilli et al (2013) covers this reasonably well IMHO, which is freely available here. http://www.pnas.org/content/110/35/14308.full.pdf

    Basically the distinctiveness of the Dene is also shared with other northern groups like the Yakima and Algonquins, and the relationship between the Dene and the Yenisiens and Siberians could be due to a back migration from Alaska rather than the reverse. It's a good paper and well worth the read IMHO.
    , @Greg Pandatshang
    My guess is that linguists talking about really old Algonquian topics might say Algonquian when, in principle, they mean Algic, out of habit. Or maybe they reject the Algic theory, although I think most Algonquianists do accept it.

    Talking about Algonquian genetics is a whole other kettle of fish. There might not be such a thing as Algic genetics.

    Edward Vajda, in a talk that's available on Youtube, mentioned that he believes that the linguistic ancestors of Algonquian speakers came as a distinct migration at around the same time as the first settlement of the New World. He sees them as the last of the first wave. I went into a little more detail about this in a comment on this blog about a year ago. Razib was dubious. http://blogs.discovermagazine.com/gnxp/2013/11/long-first-age-mankind/#.VMBVqPV33To

    In the same talk, Vajda comments that he was a little surprised that the comparative method works on Dene-Yeniseian comparisons, given the enormous time depth involved. Prima facie, he would have expected the ornate verb morphology of these languages to be relatively more labile, but, empirically, it looks like it has in fact been remarkably durable over the millennia.

    So, apparently, Vajda believes that standard comparative reconstruction can sometimes work quite a bit further back the conventional wisdom of 6 to 8 thousand years. Vajda is taken seriously by linguists in general, but that doesn't necessarily mean that they mostly agree with him on these particular points, of course.
  22. @donut
    Yes I've read that about the Basque and Finnish languages and others that are isolated and unrelated to any others around them . But what I find so fascinating is the question of how they came to be there .And I guess no one can answer that question , or can they

    Basque seems to be indigenous to the area, going by place names, and were already there in Roman times. I think most agree that they represent a pre-Indo-European stock in that part of Europe; there is no reason to think they came from anywhere else after Indo-European speakers arrived, which in Spain I think took place with the influx of Celtic speakers in the Iron Age. Before then is speculation, though I expect they can be placed plausibly among the Neolithic farmer migrants.

    Finnish is a Uralic language, and proto-Uralic was most likely spoken in central Russia, pretty much due north of where proto-Indo-European speakers lived, which was most likely eastern Ukraine.

    For Indo-European, the book to read now is David Anthony The Horse, the Wheel and Language.

    Read More
  23. @jtgw
    Basque seems to be indigenous to the area, going by place names, and were already there in Roman times. I think most agree that they represent a pre-Indo-European stock in that part of Europe; there is no reason to think they came from anywhere else after Indo-European speakers arrived, which in Spain I think took place with the influx of Celtic speakers in the Iron Age. Before then is speculation, though I expect they can be placed plausibly among the Neolithic farmer migrants.

    Finnish is a Uralic language, and proto-Uralic was most likely spoken in central Russia, pretty much due north of where proto-Indo-European speakers lived, which was most likely eastern Ukraine.

    For Indo-European, the book to read now is David Anthony The Horse, the Wheel and Language.

    Thank you . Ordered it

    Read More
  24. @jtgw
    Ah I see - fascinating! Well, my impression was that the genetic evidence was the strongest thing Greenberg had going for his theory, so it's interesting to see even that doesn't work. But I'd be surprised if Algonquian speakers immigrated 15,000 years ago: one would expect their daughter languages to have diverged too much for any recognizable relationship between them to be apparent. Also, by Algonquian do they mean Algic? The Algonquian languages in fact form just one branch of a family that also includes, amazingly, a couple of languages native to northern California (Yurok and Wiyot).

    I meant Algonquian because I don’t think any Yurok and Wiyok were included in the study as far as I am aware, so I didn’t want to assume anything for Algic as a whole.

    From Reich et al 2012:

    The 47 most southern Native American populations are consistent with descending from a single peopling event (all statistics |Z|<3; Table S5). However, 5 Northern Native American (NNA) populations—Ojibwa, Cree, Algonquin, Cheyenne and Chipewyan—have Z-scores 3-6 standard errors from expectation, and are also outliers in population structure analyses (Figure 1B and Figure 2). Further examination of the values of the 4 Population Test statistics demonstrates two distinct patterns of relationships to Arctic and East Asian populations among these 5 NNA groups. The statistics for four of the NNA (Cheyenne, Cree, Ojibwa and Algonquin) are highly correlated (average r2 =0.72; Figure S6) and indicate a closer relationship of these populations to the Inuit than to any Asian group (Figure 2B and Table S6). By contrast, statistics involving the Chipewyan are not correlated to the other four NNA (r2 =0.05; Figure 2C; Table S6), suggesting distinct gene flows with Asians. Globally, these findings show that Native Americans break into three broad groups: the 47 Native American populations from Meso-America southward, the Inuit along with 4 NNA populations with whom they appear to have exchanged genes, and the Chipewyan who speak a Na-Dene language. This is consistent with the controversial22 three migration wave model of Greenberg which views Inuit and Na-Dene languages as markers for distinct migrations from Asia9, although not with the purest form of that model which would specify that the Inuit and Chipewyan represent sister groups to some Siberian populations, whereas in fact they cluster with Native Americans (Figure 1C), consistent with subequent admixture within the Americas. Intriguingly, Greenberg’s hypothesis that Na-Dene marks a distinct migration with Asia has been supported by recent linguistic work that shows that Na-Dene language have a link with Siberian Yeniseian languages23. The group of Siberian populations with which the Chipewyan show the strongest genetic affinity includes the Ket, the sole living speakers of Yeniseian (Table S6).

    Link to the paper: http://genetics.med.harvard.edu/reich/Reich_Lab/Welcome_files/2012_Nature_NativeAmericans.pdf

    If I recall correctly, the fundamental idea that Greenberg presents is more or less supported by genetics, but that his classification of specific groups and the relationships between those groups is not entirely correct. Also, a lot of it rests on the assumption of a 1:1 correspondence between genetic and linguistic origins, which isn’t always true. For example, when my family switched from speaking Scots Gaelic to speaking English ~180 years ago, we didn’t also change our DNA.

    Achilli et al (2013) covers this reasonably well IMHO, which is freely available here. http://www.pnas.org/content/110/35/14308.full.pdf

    Basically the distinctiveness of the Dene is also shared with other northern groups like the Yakima and Algonquins, and the relationship between the Dene and the Yenisiens and Siberians could be due to a back migration from Alaska rather than the reverse. It’s a good paper and well worth the read IMHO.

    Read More
  25. @jtgw
    Ah I see - fascinating! Well, my impression was that the genetic evidence was the strongest thing Greenberg had going for his theory, so it's interesting to see even that doesn't work. But I'd be surprised if Algonquian speakers immigrated 15,000 years ago: one would expect their daughter languages to have diverged too much for any recognizable relationship between them to be apparent. Also, by Algonquian do they mean Algic? The Algonquian languages in fact form just one branch of a family that also includes, amazingly, a couple of languages native to northern California (Yurok and Wiyot).

    My guess is that linguists talking about really old Algonquian topics might say Algonquian when, in principle, they mean Algic, out of habit. Or maybe they reject the Algic theory, although I think most Algonquianists do accept it.

    Talking about Algonquian genetics is a whole other kettle of fish. There might not be such a thing as Algic genetics.

    Edward Vajda, in a talk that’s available on Youtube, mentioned that he believes that the linguistic ancestors of Algonquian speakers came as a distinct migration at around the same time as the first settlement of the New World. He sees them as the last of the first wave. I went into a little more detail about this in a comment on this blog about a year ago. Razib was dubious. http://blogs.discovermagazine.com/gnxp/2013/11/long-first-age-mankind/#.VMBVqPV33To

    In the same talk, Vajda comments that he was a little surprised that the comparative method works on Dene-Yeniseian comparisons, given the enormous time depth involved. Prima facie, he would have expected the ornate verb morphology of these languages to be relatively more labile, but, empirically, it looks like it has in fact been remarkably durable over the millennia.

    So, apparently, Vajda believes that standard comparative reconstruction can sometimes work quite a bit further back the conventional wisdom of 6 to 8 thousand years. Vajda is taken seriously by linguists in general, but that doesn’t necessarily mean that they mostly agree with him on these particular points, of course.

    Read More
    • Replies: @ohwilleke
    Taking the best available archaeological, linguistic and genetic evidence together (with archaeological dating of the oldest clearly Na-Dene archaeological remains in Alaska playing the greatest role in setting the date and the linguistic and genetic data merely playing a confirming role), a best estimate of the time of arrival of the Na-Dene people in North America is ca. 1500 BCE, a time depth of 3500 years (similar to the time depth of Welsh and Italian from each other), which makes a Dene-Yeniseian connection by the comparative method much more plausible than it is if one assumes a deeper time depth of ca. 6000-8000 years.
    , @jtgw
    I believe Algic was established in a similar way to Dene-Yeniseian: by establishing correspondences between the inflectional affixes, since most of the morphological roots had been replaced. So that would give a big time depth to Proto-Algic just as with Proto-Dene-Yeniseian. However, the time depth is not as great as with the rest of Amerindian, is it? The Proto-Algic/Algonquian speakers arrived at the tail end of the first wave, while the Proto-Dene arrived considerably later than that.

    Vajda is taken seriously and was working off of Johanna Nichols' idea that languages with complex morphological systems may provide better evidence for very old relationships with other languages with similarly complex morphologies, since the morphemes in question, e.g. nominal and verbal affixes, are replaced at a slower rate on average than lexical items like noun and verb roots. It's similar to how correspondences between Latin, Greek and Sanskrit are most apparent when you set their declensions and conjugations side by side and see the remarkable correspondences in the endings.
    , @CupOfCanada
    Per the Achilli et al (2013) paper I linked above, Dene mitochondrial DNA suggests a time depth closer to 4,000 years. So Vajda's assumption that Dene-Yenisien's links go back more than 6-8,000 years may just be incorrect.

    For convenience. Here's the comment Razib replies to in 2013:

    >Vajda argues against the idea (which I believe is the received opinion) that the first wave of settlers in North America consisted of a single, small group of hardy survivors who barely made it out of the Arctic before stumbling into Eden.

    Razib's reply:

    >the genetics leans strongly against this. e.g. in this paper all 'first americans' from north to south america seem to be same distance from MA-1.

    >but fascinating comment. this is one area i think that non-genetic fields can clarify and illuminate greatly.

    I don't know want to speak to Razib's present thoughts on this, but I would like to point out that this comment predates the Anzick-1 paper. All populations in the Americas are more or less equally distant to MA-1. Not all populations are equally distant to Anzick-1 though - Algonquian, Dene and Inuit groups are more distant to Anzick-1 than more southern populations. As shown below:

    http://www.nature.com/nature/journal/v506/n7487/fig_tab/nature13025_F2.html

    So I think genetics has caught up with Vajda when it comes to Algonquians. Very few populations north of the Mexican border were sampled though, so it may be a larger group than just Algonquians that share ancestry and possibly linguistic connections with this "second" (perhaps parallel) migration. I believe Wakashan groups appear to have ancestry from this other wave too, so Algonquian–Wakashan may be correct.
  26. @donut
    Yes I've read that about the Basque and Finnish languages and others that are isolated and unrelated to any others around them . But what I find so fascinating is the question of how they came to be there .And I guess no one can answer that question , or can they

    There are very solid theories in each case.

    * Basque is probably a relic community of what was once a vast Vasconic linguistic community in Western Europe and the Northern coastal areas of continental Europe associated with the archaeological Copper Age Bell Beaker culture that expanded from an origin in SW Iberia starting ca. 2900 BCE, with origins elsewhere, that filled the gap created by the collapse of first wave Neolithic agricultural societies and lead to massive expansion of Y-DNA R1b and mtDNA H (with a minor mtDNA X2 contribution) in the regions where it was powerful. This society kept Indo-European cultures associated with the contemporaneous Corded Ware people at bay for about 1000 years, but eventually, Indo-European Celts and Germanic populations prevailed around the time of Bronze Age collapse (ca. 1200 BCE) almost everywhere but Basque country, although the transition took time. There were additional Vasconic languages spoken near current Basque country at the time of first Roman contract. The evidence suggesting this hypothesis can’t easily be summed up in a single comment. European languages in the former Bell Beaker region show some common Vasconic language substrate influences (e.g. residual base twenty number names like “seventeen”) and traces of Vasconic language influences in place names.

    The source of the proto-Bell Beaker people who formed the Bell Beaker culture and expanded out of Southern Iberia is cryptic.

    Based upon genetic and linguistic hints (I don’t claim that there is definitive evidence although it is plausible), and the archaeological cultures in existence at the time, my best guess would be that their origins were in the Southern Caucasus mountains (likely in the Kura-Araxes culture which was strongly culturally influenced by the better known Maikop culture (aka Maykop culture) (ca. 3700 BCE-3000 BCE) of the Northern Caucasus), possibly via the Minoan society or having a common origin with it, with Basque probably belonging to the same macrolinguistic family as Kartevelian at a time depth of about 5000 years (e.g. a similar time depth to linguistic common origin of Hindi and English).

    Basically, some of the main hints are: (1) Basque is a ergative language like the modern North and South Caucasian languages and a number of now extinct languages nearby like Sumerian and the Elamite language, (2) the proto-Basque population was rich in Y-DNA R1b and mtDNA H with a minor X2 contributions, something which is also present in the Southern Caucasus (which is much closer to the place of origin of these haplogroups geographically) and in Minoan ancient DNA, (3) the ABO blood type frequencies in the Southern Caucasus are very similar to those of areas like Scotland that had heavy Bell Beaker demic impact, and (4) the Southern Caucasus were home to one of the earliest metal age cultures of Europe at about the right time and the Bell Beaker people were known for their metal working. Most alternatives to this model have serious problems on the linguistic, archaeological culture timing, or genetic front. It is hard to tell if sea routes or overland routes were most likely. A sea route and an intermediate link in the Minoan culture is attractive because the cultural role of the Bull is prominent in both Minoan and the oldest known Basque cultures.

    * The origins of the Finnish language are much better understood and the subject of a recent post and discussion at this blog. It is not a language isolate and instead is part of a larger Uralic language family (named after a presumed homelands in the Ural mountains that divide North Asia from European Russia). It probably arrived in Finland around 1500 BCE coinciding with the advent of the Finnish Bronze Age, in a process of language shift that replaced an Indo-European language was part of the Corded Ware culture language family which replaced an even older and completely lost language of Finland’s Comb Ceramic culture of maritime hunter-gatherers around ca. 2500-2300 BCE coinciding with the introduction of agriculture.

    Finnish shares a macro-language family with Estonian, Hungarian and a number of less well known languages of ethnic enclaves in Russia (e.g. Mari and Komi). We can make some quite informed guesses about which components of the current population genetics of Finland correspond to its Mesolithic, Indo-European Neolithic and Uralic Bronze migration strata, although Swedish and Russian Slavic migration to Finland in the historic era complicate this effort somewhat.

    Read More
  27. @Greg Pandatshang
    My guess is that linguists talking about really old Algonquian topics might say Algonquian when, in principle, they mean Algic, out of habit. Or maybe they reject the Algic theory, although I think most Algonquianists do accept it.

    Talking about Algonquian genetics is a whole other kettle of fish. There might not be such a thing as Algic genetics.

    Edward Vajda, in a talk that's available on Youtube, mentioned that he believes that the linguistic ancestors of Algonquian speakers came as a distinct migration at around the same time as the first settlement of the New World. He sees them as the last of the first wave. I went into a little more detail about this in a comment on this blog about a year ago. Razib was dubious. http://blogs.discovermagazine.com/gnxp/2013/11/long-first-age-mankind/#.VMBVqPV33To

    In the same talk, Vajda comments that he was a little surprised that the comparative method works on Dene-Yeniseian comparisons, given the enormous time depth involved. Prima facie, he would have expected the ornate verb morphology of these languages to be relatively more labile, but, empirically, it looks like it has in fact been remarkably durable over the millennia.

    So, apparently, Vajda believes that standard comparative reconstruction can sometimes work quite a bit further back the conventional wisdom of 6 to 8 thousand years. Vajda is taken seriously by linguists in general, but that doesn't necessarily mean that they mostly agree with him on these particular points, of course.

    Taking the best available archaeological, linguistic and genetic evidence together (with archaeological dating of the oldest clearly Na-Dene archaeological remains in Alaska playing the greatest role in setting the date and the linguistic and genetic data merely playing a confirming role), a best estimate of the time of arrival of the Na-Dene people in North America is ca. 1500 BCE, a time depth of 3500 years (similar to the time depth of Welsh and Italian from each other), which makes a Dene-Yeniseian connection by the comparative method much more plausible than it is if one assumes a deeper time depth of ca. 6000-8000 years.

    Read More
  28. @jtgw
    Language families don't necessarily correspond to genetic families in the first place, though they often do simply because for the most part children learn their native language from parents and neighbors who are typically genetically related to them. But secondly, language relationships can only be demonstrated for languages that have diverged up to about 10,000 years ago, and the ancestral stock of most Native Americans arrived in America well before that. The reason is that you can only demonstrate relationships between sets of inherited words and other morphemes (e.g. prefixes and suffixes), but languages also replace their inherited morphemes with borrowed words, coinages and by other innovations. After about 10,000 years, most languages will retain too few ancestral morphemes to allow any plausible relationship to be demonstrated.

    The open access paper Sijia Wang, et al., “Genetic Variation and Population Structure in Native Americans” (PLOS Genetics 2007) provides a rather more optimistic prospect for using genetics to reconstruct the higher level connections between the many small linguistic families of South America to the east of the Andes Mountains, and one Andean language, on a largely tree-like basis that reproduces the linguistic relationships that are established while going further to suggest macrolinguistic family relationships that aren’t very solidly established linguistically.

    It has been difficult to establish macrofamilies of languages linguistically in South America due to issues like a shortage of linguists to study a highly linguistically diverse region and poor documentation of many languages, key missing links that are extinct or moribund and not well documented, and great expected time depth separating related language families. An institutional culture among South American linguists that has favored a conservative approach that leans towards splitting rather than lumping (famous in Russian school linguists) is also a factor.

    Read More
    • Replies: @jtgw
    I do remember reading in Nichols' 1994 survey of linguistic diversity that South American native languages were especially poorly documented and that her measures of diversity in that region may not be accurate. I suppose a gene-language correlation model that has been adequately trained on comparisons of undisputed genetic and linguistic lineages could be useful in stimulating research into long-distance relationships among South American languages and language families. The problem with research like Cavalli-Sforza's is that the long-distance linguistic relationships he used were the product of Greenberg and Ruhlen's research and not accepted by the vast majority of linguists.
  29. @Greg Pandatshang
    My guess is that linguists talking about really old Algonquian topics might say Algonquian when, in principle, they mean Algic, out of habit. Or maybe they reject the Algic theory, although I think most Algonquianists do accept it.

    Talking about Algonquian genetics is a whole other kettle of fish. There might not be such a thing as Algic genetics.

    Edward Vajda, in a talk that's available on Youtube, mentioned that he believes that the linguistic ancestors of Algonquian speakers came as a distinct migration at around the same time as the first settlement of the New World. He sees them as the last of the first wave. I went into a little more detail about this in a comment on this blog about a year ago. Razib was dubious. http://blogs.discovermagazine.com/gnxp/2013/11/long-first-age-mankind/#.VMBVqPV33To

    In the same talk, Vajda comments that he was a little surprised that the comparative method works on Dene-Yeniseian comparisons, given the enormous time depth involved. Prima facie, he would have expected the ornate verb morphology of these languages to be relatively more labile, but, empirically, it looks like it has in fact been remarkably durable over the millennia.

    So, apparently, Vajda believes that standard comparative reconstruction can sometimes work quite a bit further back the conventional wisdom of 6 to 8 thousand years. Vajda is taken seriously by linguists in general, but that doesn't necessarily mean that they mostly agree with him on these particular points, of course.

    I believe Algic was established in a similar way to Dene-Yeniseian: by establishing correspondences between the inflectional affixes, since most of the morphological roots had been replaced. So that would give a big time depth to Proto-Algic just as with Proto-Dene-Yeniseian. However, the time depth is not as great as with the rest of Amerindian, is it? The Proto-Algic/Algonquian speakers arrived at the tail end of the first wave, while the Proto-Dene arrived considerably later than that.

    Vajda is taken seriously and was working off of Johanna Nichols’ idea that languages with complex morphological systems may provide better evidence for very old relationships with other languages with similarly complex morphologies, since the morphemes in question, e.g. nominal and verbal affixes, are replaced at a slower rate on average than lexical items like noun and verb roots. It’s similar to how correspondences between Latin, Greek and Sanskrit are most apparent when you set their declensions and conjugations side by side and see the remarkable correspondences in the endings.

    Read More
  30. There’s a relatively recent paper by Dennis Stanford and Stephen Oppenheimer called “Solutrean hypothesis: genetics, the mammoth in the room”. It’s incredibly low quality. They managed to misunderstand pretty much everything they possibly could, but it does contain some interesting information about haplogroup X2 in the Americas.
    It is largely restricted to Algonquian speakers and there is not just one but two distinct varieties of it, the rather common X2a and the very rare X2g which have a common ancestor ~20000 years ago in the Old World. My guess would be that they are markers of a pre-Clovis people , but from Siberia.

    Read More
  31. @ohwilleke
    The open access paper Sijia Wang, et al., "Genetic Variation and Population Structure in Native Americans" (PLOS Genetics 2007) provides a rather more optimistic prospect for using genetics to reconstruct the higher level connections between the many small linguistic families of South America to the east of the Andes Mountains, and one Andean language, on a largely tree-like basis that reproduces the linguistic relationships that are established while going further to suggest macrolinguistic family relationships that aren't very solidly established linguistically.

    It has been difficult to establish macrofamilies of languages linguistically in South America due to issues like a shortage of linguists to study a highly linguistically diverse region and poor documentation of many languages, key missing links that are extinct or moribund and not well documented, and great expected time depth separating related language families. An institutional culture among South American linguists that has favored a conservative approach that leans towards splitting rather than lumping (famous in Russian school linguists) is also a factor.

    I do remember reading in Nichols’ 1994 survey of linguistic diversity that South American native languages were especially poorly documented and that her measures of diversity in that region may not be accurate. I suppose a gene-language correlation model that has been adequately trained on comparisons of undisputed genetic and linguistic lineages could be useful in stimulating research into long-distance relationships among South American languages and language families. The problem with research like Cavalli-Sforza’s is that the long-distance linguistic relationships he used were the product of Greenberg and Ruhlen’s research and not accepted by the vast majority of linguists.

    Read More
  32. @Greg Pandatshang
    My guess is that linguists talking about really old Algonquian topics might say Algonquian when, in principle, they mean Algic, out of habit. Or maybe they reject the Algic theory, although I think most Algonquianists do accept it.

    Talking about Algonquian genetics is a whole other kettle of fish. There might not be such a thing as Algic genetics.

    Edward Vajda, in a talk that's available on Youtube, mentioned that he believes that the linguistic ancestors of Algonquian speakers came as a distinct migration at around the same time as the first settlement of the New World. He sees them as the last of the first wave. I went into a little more detail about this in a comment on this blog about a year ago. Razib was dubious. http://blogs.discovermagazine.com/gnxp/2013/11/long-first-age-mankind/#.VMBVqPV33To

    In the same talk, Vajda comments that he was a little surprised that the comparative method works on Dene-Yeniseian comparisons, given the enormous time depth involved. Prima facie, he would have expected the ornate verb morphology of these languages to be relatively more labile, but, empirically, it looks like it has in fact been remarkably durable over the millennia.

    So, apparently, Vajda believes that standard comparative reconstruction can sometimes work quite a bit further back the conventional wisdom of 6 to 8 thousand years. Vajda is taken seriously by linguists in general, but that doesn't necessarily mean that they mostly agree with him on these particular points, of course.

    Per the Achilli et al (2013) paper I linked above, Dene mitochondrial DNA suggests a time depth closer to 4,000 years. So Vajda’s assumption that Dene-Yenisien’s links go back more than 6-8,000 years may just be incorrect.

    For convenience. Here’s the comment Razib replies to in 2013:

    >Vajda argues against the idea (which I believe is the received opinion) that the first wave of settlers in North America consisted of a single, small group of hardy survivors who barely made it out of the Arctic before stumbling into Eden.

    Razib’s reply:

    >the genetics leans strongly against this. e.g. in this paper all ‘first americans’ from north to south america seem to be same distance from MA-1.

    >but fascinating comment. this is one area i think that non-genetic fields can clarify and illuminate greatly.

    I don’t know want to speak to Razib’s present thoughts on this, but I would like to point out that this comment predates the Anzick-1 paper. All populations in the Americas are more or less equally distant to MA-1. Not all populations are equally distant to Anzick-1 though – Algonquian, Dene and Inuit groups are more distant to Anzick-1 than more southern populations. As shown below:

    http://www.nature.com/nature/journal/v506/n7487/fig_tab/nature13025_F2.html

    So I think genetics has caught up with Vajda when it comes to Algonquians. Very few populations north of the Mexican border were sampled though, so it may be a larger group than just Algonquians that share ancestry and possibly linguistic connections with this “second” (perhaps parallel) migration. I believe Wakashan groups appear to have ancestry from this other wave too, so Algonquian–Wakashan may be correct.

    Read More
  33. […] results point to a Native-American heritage.'” – from dienekes. see also from razib: Native Americans Are Evolutionarily Elegant – “One insight of modern ancient DNA is that there has been a great deal of population […]

    Read More

Comments are closed.

PastClassics
What Was John McCain's True Wartime Record in Vietnam?
The “war hero” candidate buried information about POWs left behind in Vietnam.
The evidence is clear — but often ignored
What the facts tell us about a taboo subject
While other top brass played press agents for the administration’s war, William Odom told the truth about Iraq—though few listened.