I would defer to how actual linguists wield that razor, and I believe that the consensus, to the extent there is one, is that clicks share a common origin. I could be wrong about that, and would defer to one of our resident linguists if one of them pops into this thread.
Here’s a pdf paper that discusses the genes and languages of “Khoisan” people. I won’t try to summarize it, but highlights include
[ul][li] The genes of Hadzabe and San are very divergent.[/li][li] Clicks may be advantageous in stealth hunter settings.[/li][li] [/li][QUOTE]
Two lines of evidence, rarity of clicks in human languages and complexity of the shared repertoire of clicks and accompaniments, suggest that independent invention of clicks in San and Hadzabe populations is an unlikely explanation for the observed genetic pattern.
[/QUOTE]
[/ul]
I suppose Ruken includes among “amateur woo” the historical linguistics work of Murray Gell-Mann, Nobel Prize winner, who has collaborated with a lecturer at this infamous “Department of Anthropological Sciences at Stanford.”
From that article (emphasis mine):
“Two lines of evidence, rarity of clicks in human languages and complexity of the shared repertoire of clicks and accompaniments, suggest that independent invention of clicks in San and Hadzabe populations is an unlikely explanation for the observed genetic pattern.”
That’s the key methodology: the existience of clicks and their rarity are two pieces of evidence, but their embedding within a complex shared system is a much stronger piece of evidence. That’s how historical linguistics works: not just identifying shared features, but measuring the likelihood of a common origin by weighing the evidence and testing alternative hypotheses (which they also address in this article). I’m not a fan of the Ruhlen & Greenberg, but I can’t fault their methodology on the linguistic side here. This article is much more about biology than language, but then again, so is this thread.
You are correct. Dr Gell-Mann is an amateur linguist, who has no qualification in historical linguistics or any related field, and whose opinions on the subject are worth about as much as yours or mine.
I am quite frankly amazed at the use of Occam’s Razor in this situation.
We have two of the most divergent (if not most divergent) groups of people that share a trait that is basically common to just those two groups. And people want to invoke Occam’s razor to say that the trait must have arose independently?
That, to me, is the antithesis of Occam’s razor.
For myself, I am pointing out that Occam’s razor is not a useful tool in this case in either direction. No one is arguing that these traits must have arisen independently, only that barring aditional evidence, the mere fact that a sound, even a rare one, exists in two sets of languages suggests nothing. Occam’s razor says that the simplest explanation is likely to be correct, but with phonemes, you have two problems:
-
“There is a relationship” and “there is no relationship” are equally likely. Having established that these are rare phonemes, “there is a relationship” is slightly stronger, but
-
“There is a genetic relationship” (common inheritance) and “there is a cultural relationship” (borrowing) are equally likely. Occam can’t help with two equal likelihoods.
Or that it arose much more recently than when the two populations diverged, and was subsequently borrowed.
But many other unrelated languages share rare traits.
Their data are more about biology than language. You cannot fault their methodology on the linguistic side because there isn’t one. They did some solid genetic work; African genetics are fascinating and I hope they continue to do more work in the area. Their fantasies about 40,000-year-old language relics are not serious linguistic theories. That doesn’t mean they aren’t correct. It would be really cool and I kind of hope that they are. But they don’t adequately support that conclusion.
Perhaps at this stage you could point us to actual published scholars who make these criticisms. This linkage between the click language isn’t obscure or something with only one published proponent.
So perhaps you could provide a cite for your position, having been shown one supporting the alternative?
What linkage? The fact that they have some phonemes in common?
The idea that this supports the conclusion that these phonemes existed in the ur-language of all humanity is absurd on its face.
I have written in multiple posts that they may have a common origin.
And as I’ve said, it could be true, but Knight et al are leaping given the data at hand.
That wasn’t the question. the question is, do any actual scholars disagree with Knight et al at all, or is it just you.
Saying that relativity “might be right” isn’t responsive in a thread where i have just claimed that Einstein and numerous other physicists are wrong.
So I ask once again, do any actual scholars agree with you?
Which position in particular are you interested in? I can check for you.
We can start with your assertion that “Knight et al are leaping given the data at hand”. Has any scholar in the history of the world agreed with that statement?
Blake, you seem to be confusing the validity of the argument with the validity of its conclusions. You don’t need to be an expert to assess the former—anyone who has a smattering of logical thought can do that—but you do to assess the latter.
No one disputes that the African click languages have a common origin. What is disputed is the time and the nature of that origin. Greenberg and Ruhlen are operating well beyond the margins of safety in historical linguistics, and I don’t think either of them would dispute that were they around to be asked. See Wikipedia (Merritt Ruhlen) or Mass Comparison for an example of some of their reception. They may be right, but they’re hardly mainstream and they may well be wrong.
Let’s take a look. Knight et al have been cited 66 times since the paper was published over 10 years ago.
If we exclude self-citations by the authors and their frequent collaborator Sarah Tishkoff (now at UPenn, congrats on the new job), we’re left with 36 citations. It’s a good thing these scholars are such hot-rods, otherwise we wouldn’t have much to work with.
Of those 36, 31 are on genetics (including papers on the demography of Jamaica, the native identity of the Seaconke Wampanoag Tribe of Massachusetts, and really cool one questioning the supposed population bottleneck in the Penultimate Glacial Period*)
1 is on nutrition and disease,
1 is on navigation,
1 is on Khoesan skull morphology
The remaining two address language. We’re in luck! This paper has been highly cited by experts on language.
The first, DOI: 10.1371/journal.pone.0035289
Charles Perreault (Santa Fe Institute) and Sarah Mathew (Centre for the Study of Cultural Evolution, Stockholm U.) accept Knight et al’s estimates of genetic age and linguistic ages and add in click languages’ phonemic diversity to estimate that human language arose 242-163 kya. They do not address if the clicks have been conserved since the Hadzabe and Ju|’hoansi genetically diverged. It actually seems like a good paper if anyone is interested in checking it out, and the authors fill it with caveats and explanations of why their results could be way off.
The second, DOI: 10.1111/j.1749-6632.2009.04423.x
Michael Corballis (Dept of Psychol, U. Auckland) wrote a review in a book series that I don’t have access to, so I can’t tell you anything, sorry.
Despite being published by expert linguists in a best-seller linguistics journal, this paper has completely passed under the radar of the authors’ fellow linguists!
Or it could be that some geneticists published a genetics paper in a bio journal. It contains wild speculation not supported by the data they collected. Since their linguistic conclusions mean fuck-all, the paper has been ignored by linguists, since you don’t get tenure by paying attention to this stuff, and no journal editor I know would touch a paper that solely focused on bashing a paper that nobody read in the first place.
So has any scholar in the history of the world agreed that Knight et al are leaping given the data at hand? No, no scholar in the history of the world has agreed. Other than this scholar, nobody cares. And no, I’m not going to write a paper about it either.
*Completely unrelated to today’s episode, but this one slipped by me when it came out and I want to read it. I thought I’d share:
Resequencing Data Provide No Evidence for a Human Bottleneck in Africa during the Penultimate Glacial Period
By:Sjodin, P; Sjostrand, AE; Jakobsson, M; Blum, MGB [a bunch of Swedes]
in MOLECULAR BIOLOGY AND EVOLUTION
Volume: 29
Issue: 7
Pages: 1851-1860
DOI: 10.1093/molbev/mss061
Ruken: Are you a linguist, or do you have a degree in that subject? No snark intended, it just seems that you are implying you are in that last post, but I’d like to clarify that.
Um, no. The Finns and the Russians have been neighbors only for the past 1 000 years, at most, as that is when the Slavs entered Northwestern Russia that was populated by myriad Finno-Ugric-speaking indigenous peoples - culturally, linguistically and genetically close to the Finns. True, Finns are distinct from most European peoples, but not from their true Northeastern European neighbors, most of which were displaced / assimilated / exterminated by the Slavic expansion.
If the only two groups with click languages in the world were, say, Euskara (Basque) and Haida, then one might think “How, odd. That certainly seems random.” But there definitely doesn’t seem to be anything remotely random about these two groups. Far from it.
Clicks and very high genetic separation? Too much going on to assume randomness.
And to suppose language diffusion long after genetic separation, that is hardly a simplifying assumption.
That really only leaves one simple explanation.