NEANDERTHAL LANGUAGE?

Dan Dediu and Stephen Levinson, from the Max Planck Institute for Psycholinguistics and the Radboud University Nijmegen, have published a paper (abstract) arguing that (in the words of this Jul. 9 Sci-News.com story) “essentially modern language and speech are an ancient feature of our lineage dating back at least to the most recent ancestor we shared with the Neanderthals and the Denisovans.” The abstract linked above says:

This reassessment of the antiquity of modern language, from the usually quoted 50,000–100,000 years to half a million years, has profound consequences for our understanding of our own evolution in general and especially for the sciences of speech and language. As such, it argues against a saltationist scenario for the evolution of language, and toward a gradual process of culture-gene co-evolution extending to the present day. Another consequence is that the present-day linguistic diversity might better reflect the properties of the design space for language and not just the vagaries of history, and could also contain traces of the languages spoken by other human forms such as the Neandertals.

Now, my instinctual reaction is “What a load of poppycock,” but both Dediu and Levinson have been mentioned with respect over at Language Log, and I’m certainly not competent to have an informed opinion. I’m posting this so that those better informed than I can weigh in, and in the hope that it might prompt one of the Loggers to address it.

Comments

  1. I haven’t read the article yet, but it’s not surprising that Neanderthals may have had language given that they interbred with anatomically modern humans in significant numbers within the last hundred thousand years. Hard to see how that would have happened if one (sub)species had language and the other didn’t.

  2. Er – since humans will pretty much shag anything with a hole, I find it very easy to see how that will have happened without language being necessary at all. And you certainly don’t need language to rape anybody …

  3. Trond Engen says

    I’ve always found the young date of origin implausible given the age of human settlement in Australia.

  4. I suppose it depends how much contact would have been necessary to produce the 1-4% Neanderthal DNA that modern non-African humans apparently have. If the answer is “more than a little”, well, of course it’s possible that forcible impregnation of language-having AMH women by languageless Neanderthals was a common event in the Paleolithic, but some kind of culturally mediated admixture sounds a lot more likely, I’d think.

  5. Maybe the “giants in the earth” of Genesis 6, produced when “the sons of God came in unto the daughters of men”, are a faint memory of robust Neanderthal hybrids?

  6. That part in the Bible about the giants has always intrigued me.

  7. George Grady says

    This is off-topic; I hope you’ll forgive me.
    When I was learning Russian back in college, one of my professors showed us an animated movie about a man with a helicopter pack who lived in a little boy’s chimney, or something like that. He caused all sorts of trouble, but things all worked out in the end. She told us it was really popular. Does anyone know what that was? My Google skills are failing me.

  8. Trond Engen says

    That has to be Astrid Lindgren’s Karlsson på taket. Imdb tells me that this was animated in Russian as Malysh i Karlson.

  9. “Maybe the “giants in the earth” of Genesis 6, produced when “the sons of God came in unto the daughters of men”, are a faint memory of robust Neanderthal hybrids?”
    Nah, it’s definitely a memory of space aliens having sex with humans and genetically engineering the species.

  10. Absurd idea that our own languages retain traces of Neandertal speech. Ugh!

  11. Trond Engen says

    Neandertaal.

  12. George Grady says

    Trond,
    That’s it! Thanks!

  13. “There were giants in the earth in those days” — there are giants in the earth these days, slowly being dug out; they’re called dinosaurs.
    Anyway I think it would probably be best to reserve judgment on this paper until we can see how it deals with Auel 1980, whose argument to the contrary is both persuasive and romantic.

  14. I’ve always wondered if legends about trolls and ogres in Scandinavia might not represent a vague folk memory of a time when we coexisted with Neanderthals. It doesn’t seem implausible that some lingering isolated groups of Neanderthals could have even survived in the woods of Northern Europe for some time until they gradually dwindled away.

  15. The Erste Allgemeine Verunsicherung addressed this in 1991, and if an Austrian punk band is to be trusted on questions of phylogenetics and the genesis of language, they seem to think the Neanderthals could speak 🙂 .

  16. Jeffry House says

    It is amazing to recognize that Neanderthals actually had Louis Vuitton handbags and down jackets.

  17. They were knockoffs; the Neanderthals couldn’t tell the difference.

  18. “I’ve always wondered if legends about trolls and ogres in Scandinavia might not represent a vague folk memory of a time when we coexisted with Neanderthals. ”
    Vanya, what is likelier is they represent a memory of the thousand year lag between the arrival of Neolithic farmers and the onset of intermarriage with the indigenous foragers. The genetic evidence indicates there was such a lag.

  19. “I’ve always wondered if legends about trolls and ogres in Scandinavia might not represent a vague folk memory of a time when we coexisted with Neanderthals. ”
    Vanya, what is likelier is they represent a memory of the thousand year lag between the arrival of Neolithic farmers and the onset of intermarriage with the indigenous foragers. The genetic evidence indicates there was such a lag.

  20. Well, neither can we, much of the time. In Harry Turtledove’s “Down in the Bottomlands”, the Straits of Gibraltar have been sealed for the last five million years, and the Mediterranean is Death Valley (or the Dead Sea) writ large. As a consequence, Strongbrow polities still apparently occupy most of Europe, with Highheads found in Africa and perhaps elsewhere. Turtledove wrote it before the DNA evidence on Highhat-Strongbrow interbreeding, so the story assumes they aren’t interfertile (see also rishathra).
    There is also the fact that human pubic lice are not closely related to human head and body lice, but are rather an offshoot of gorilla lice. Hmmmm.

  21. The authors use high estimates of the ages of certain language families to argue for low estimates of the rate of language change. But these aren’t uncontroversial. The high estimate of the age of the Indo-European languages leaves problems with the common words for chariot parts. One of the sources they cite has, I think, been skewered on Geocurrents.
    The authors also identify facial expressions as an important mode of communication. This is a problem in day-to-day life, because, for many of us, eye contact is painful, we can’t read most people’s facial expressions, and most people can’t read ours. However, in combination with signing and writing, I think this shows that language can cross between channels.

  22. I have just finished reading the article.
    My advice to others: Don’t waste your time.
    Since this may seem like a harsh statement, let me exemplify:
    On page 11 they conclude that “it is reasonable to assume that Neanderthal languages were most probably tonal”. Whence comes this conclusion, you might ask? Why, from one of the authors (Dediu) having observed in an earlier article a correlation between the frequency of certain genes and the use of phonemic tone (p.10).
    In turn, because genes associated with non-tonal languages are absent from known samples of Neanderthal DNA, they therefore conclude that Neanderthal languages were probably tone languages. QED.
    What part of “Correlation is not causation” don’t they understand, I wonder?
    They also assume (p.12) that if Neanderthal languages survived or influenced non-Neanderthal languages this would have left tangible traces in known languages of Europe and the Middle East. This is a leap of faith which remains unexplained and, I would maintain, is unjustifiable.
    After all, the later spread of Afroasiatic throughout North Africa and the Middle East (from what seems likely to have been an Urheimat in the Sudan or Ethiopia) must have eliminated whatever earlier languages once existed there.
    Might this pre-Afroasiatic substrate have left its mark, then? Well, it is interesting to note that the non-subsaharan members of the family (Berber, Semitic, Egyptian) are non-tonal, unlike a majority of Cushitic and Chadic languages.
    Since this distribution of phonemic tone is the *polar opposite* of what their “model” predicts, I can only repeat my earlier advice: don’t waste your time.

  23. If you take the position, which I do, that something that could reasonably be called language is quite ancient in the human lineage, it raises the question of what are the distinguishing characteristics of “modern” versus “ancient” languages.
    I suppose this might have been dealt with somewhere in the literature, but not being an academic, I’ve never seen it. Without that, and without a firm grasp of the genetic architecture of our language ability, I see most of this as simple arm-waving and theorizing in the absence of evidence.
    With respect to TR’s opening comment and the responses: I took it to mean that the genetic architecture of the brain would be to different for a cross to survive, not that there wouldn’t be attempts.

  24. Don’t waste your time.
    Thanks for confirming my expectations!

  25. As a side note, William Golding’s 1955 novel The Inheritors is a very original fiction about what Neanderthal life might have been like. Golding has them use a form of English, but presumably just so he could tell the story the way he wanted, not because he thought that’s what they spoke. John Self’s excellent review describes the book as “built from the language up”.

  26. Trond Engen says

    There’s much to be said for coming up with imaginative new ideas and laying them out for public evaluation. There’s very little to be said for adding new layers of fantasy to what’s already been refuted. So we have Dediu et al. supporting Atkinson et al. quoting Dediu quoting Atkinson quoting Renfrew. Or something like that. One might think that the reception given by historical linguists to any one of those papers would have made other biologists or psychologists or whatever stop and think, but instead it seems that there’s a whole community of para-linguistics being built, of crossover scientists supporting, or surpassing, eachother with new leaps of faith, blissfully unaware or willfully ignorant of linguistic counter-evidence.

  27. it seems that there’s a whole community of para-linguistics being built, of crossover scientists supporting, or surpassing, each other with new leaps of faith, blissfully unaware or willfully ignorant of linguistic counter-evidence.
    Yes, exactly. It’s infuriating.

  28. Linguists (as a group, not individuals!) have no one to blame but themselves for the rise of pseudo-linguistics. When one group of scientists abandons a legitimate part of their subject, a power vacuum results, and another group of scientists is free to occupy it.

  29. Trond, Hat: it seems to me that the truly infuriating thing isn’t the cult-like navel-gazing, something which some mainstream “schools” (sects?) of linguistics are definitely guilty of (ge*COUGH*tivism, anyone?): what IS infuriating is the total disregard for basic methodological soundness (cf. the example I gave above).
    Articles whose foundations you disagree with, but which are internally logical and clear, can be of some use. Suppose, for example, that a scholar offered a novel model of language spread and used Atkinson & Gray as the source for Indo-European (as Dediu and Levinson do). IF said work is otherwise sound, it might be worth testing the novel model in question using more reliable sources for Indo-European than Atkinson & Gray.
    But Dediu and Levinson’s “work” does not just yield garbage because they rely solely on unreliable “authorities” on historical linguistics. It also yields garbage because the conclusions they arrive at do *not* follow from the data. So the source and accuracy of the data really do not matter.
    Hence my conclusion above: Don’t waste your time. This article is not merely flawed: there is nothing in it that could be salvaged.

  30. Trond Engen says

    Jim: the thousand year lag between the arrival of Neolithic farmers and the onset of intermarriage with the indigenous foragers
    I meant to add something here. There’s a growing body of archaeological evidence for two distinct cultures even in central Scandinavia all through the Viking Age and into the Medieval, each with its own international network. In the vast regions of forests and mountains the Finns, i.e. the Sami, left graves showing connections with Finnic tribes as far east as the Ural. Along the coasts and in the fertile flatlands the Norse were part of a Baltic and European system.
    Now, by all accounts both Germanic and Finnic are newcomers to Scandinavia relative to the Neolithic revolution, but it’s intriguing how the two and their relation seem to continue something much older, and how both had external ties long and strong enough to have led to a language shift that can’t (yet) be traced archaeologically.

  31. I’ve now read the article, and I disagree with Etienne’s “don’t waste your time”. As far as I can tell, they do a good job of supporting their main claim — “Neandertals probably had language” — with evidence from several different fields. (Of course, I can’t evaluate much of this evidence, and I tend to believe their claim anyway, so maybe I wasn’t reading as critically as I could have been.) The stuff about tonality, possible Neandertal language influence, and the antiquity of IE is all pretty tangential; the first two are presented merely as additional conjectures (and the second as something that might be investigated, not as an assumption), and although they do buy into Atkinson et al.’s ideas on IE, I don’t see that anything in their argument depends on that. Now, it may be that their archaeological/genetic/anthropological arguments are open to criticism, but that’s a different story.

  32. With respect to TR’s opening comment and the responses: I took it to mean that the genetic architecture of the brain would be to different for a cross to survive, not that there wouldn’t be attempts.
    Exactly – would a human-Neandertal hybrid have the cognitive capacities to survive and reproduce in a human society if half its genes came from a creature that was incapable of speech? Again, anything’s possible, but it doesn’t seem that likely.

  33. Okay, I’ve read the paper now too, and I take an intermediate position between Tom and Etienne. As I see it, there are four claims:
    1) Neanderthals could have had language. Fairly well established.
    2) Neanderthals did have language. Not clearly established, but not unreasonable either.
    3) We can determine concrete things about Neanderthal languages (e.g. they were probably tonal). Bogus, but essentially trivial.
    4) The Neanderthal-language assumption lets us conclude things about existing languages we didn’t know before (e.g. the diversity of existing languages can only be explained by linguistic crossovers from Neanderthal languages). Even more bogus, and much more serious. I can’t agree that this part is peripheral: the whole point of establishing 1 and 2 seems to be to license these speculations.
    So while the paper is not worthless, it does not and cannot do the work that its authors want it to do.

  34. “The men of Whee [i.e. Bree] were frequently mistaken for Neanderthals, a confusion which the latter deeply resented.” —Bored of the Rings

  35. The Neanderthal languages were probably tonal

    The authors must not have heard of tonogenesis, the development of tones in non-tonal languages. Tones as distinctive accompaniments of syllables can arise from specific combinations of vowels and consonants (since those sounds tend to influence each other), followed by weakening or loss of consonants so that the tone of each vowel (influenced by the earlier neighbouring consonant) remains. See wikipedia on “tone (linguistics)”, especially the section on “Origin”, toward the end.

Speak Your Mind

*