Tuesday, December 22, 2009

262. The Baseline Scenarios -- 38: The Gap

As we've seen in the previous post, there is a striking resemblance between the distribution of P/B style (A1 - A4), as presented in the second of my mini-maps, and the distribution of tonal languages in the Old World. A very similar pattern presents itself when we consider the distribution of musical instruments, wood carving and mask making traditions, and negrito populations. In every case, we see essentially the same gap, in southwestern and southern Asia, where certain things we would expect to find, based on a smooth, uninterrupted southern migration from Africa to the Sahul, are missing.

As I've already mentioned, genetic anthropologist Stephen Oppenheimer has pointed to a very similar gap in the genetic evidence, which led him to propose a major bottleneck event stemming from the eruption of Mount Toba. While the timing of the Toba eruption makes Oppenheimer's interpretation problematic, the genetic evidence he cites does indeed suggest that something very unusual must have happened in South Asia very early on, at what might well have been a critical turning point in human history.

With all this in mind, let's return to the genetic map I presented two posts ago:


(Click on the above to enlarge the image.) The map, representing a phylogenetic tree based on mtDNA (mitochondrial DNA, representing female lineages only) is from a recently published article, Correcting for Purifying Selection: An Improved Human Mitochondrial Molecular Clock, by Pedro Soares et al., June 2009. (Among the co-authors, by the way, is Stephen Oppenheimer.) What makes this tree especially interesting is, first of all, the effort by the authors to correct for the effects of natural selection, and secondly, the introduction of timing estimates based on what the authors claim to be a new and improved molecular "clock." The number displayed next to each node represents their estimate for the origin of that node, in thousands of years. Thus, for example, the origin of M is given as 60.6, i.e., 60.6 thousand years ago.

Each node is color coded according to its most common geographical occurrence, as indicated in the key at the upper right. What's unusual here is the fact that the M haplogroups for South Asia (in violet) are presented farther to the right, and slightly lower, than the M haplogroups for East Asia (in blue), implying that the South Asian haplogroups appeared at a later date than those for East Asia. And sure enough, the date for South Asian M is 49.4, while that for East Asian M is 60.6. If the Out of Africa migration proceeded in an orderly, predictable fashion, from west to east, we would expect to find the oldest haplogroups in South, not East, Asia.

So once again we must ask: what gives?

The first point to be made regarding the new research is the fact that the dates they've come up with "render an out of-Africa dispersal prior to the Toba eruption in Sumatra at ~74 kya less likely." Given that Oppenheimer is one of the co-authors, this is a particularly significant conclusion. Nevertheless, the inconsistency between South and East Asia must be accounted for:

In the context of the southern-coastal-route model, it should be noted that although the distribution of haplogroup M has also been used to support the southern route model, the age of haplogroup M in India, at 49.4 (39.0; 60.2) kya, is significantly lower than in East Asia, at 60.6 (47.3; 74.3) kya . . . At face value, this could suggest an origin of haplogroup M in East Asia and a later migration back into South Asia, suggesting that it may have been a ‘‘pre-M’’ lineage that initially crossed South Asia. . . Southeast Asia may [therefore] be the point of origin of haplogroup M . . . Alternatively, if M dispersed with N and R through South Asia, M may have been caught up in a subsequent bottleneck and founder effect so that its age signals the time of re-expansion rather than first arrival (p. 752 -- my emphasis).

There will, of course, be a great many refinements and no doubt revisions of all the many phylogenetic trees that are now being published. Nevertheless it seems clear, not only from these results, but others already published, that the gap I've noted, based largely on cultural evidence, is reflected in the genetic evidence as well. While such a gap might at first seem to be a stumbling block in our understanding of the earliest migrations of modern humans out of Africa, it is also, as I see it, a tremendously important clue, which could explain a great many other mysteries, some of which are so obvious as to have been taken for granted by the great majority of investigators.

(to be continued . . . )


43 comments:

German Dziebel said...

"Nevertheless it seems clear, not only from these results, but others already published, that the gap I've noted, based largely on cultural evidence, is reflected in the genetic evidence as well."

This deepens my counter-out-of-Africa argument: not only do we not find African lineages outside of Africa and early M and N lineages in Africa (in this paper's terms "There are no ‘‘pre-M’’ or ‘‘pre-N’’ clades extant either within or outside Africa," p. 752), we don't find the earliest M lineages at the next stop along the hypothetical "southern" route (India).

Another inconsistency is that M1, which is considered the product of back migration into Africa, is located at the very root of the M macrohaplogroup. So if you move D lineages all the way to the right of the phylogeny and Indian M's all the way to the left next to M1, you'll see a perfect gradient from East Asia to South Asia to north and East Africa.

The same happens with N lineages: we have X, A and other outliers (according to the map) with eastern distributions, then a bunch of India-specific R lineages and then U6 in North and East Africa.

Y-DNA DE expansion back into Africa will correlate with mtDNA very well, I think, too. We discussed potential archaeological correlates earlier. That's how a "southern route" signature is supposed to look like. We're again left with the lack of a proof of similar solidity for an out-of-African migration.

Maju said...

I'm repenting (to some extent, so to say) of having told you about this study, Victor. Age estimates are a most inexact "science" with huge error margins and highly speculative, tentative, foundations. To me it borders pseudoscience, specially for how they "sell" it as "serious stuff".

As I said before the "sub-nodes" represent average estimates for the regional branches, which typically appear older for where each macro-haplogroup expansion lasted for longer, where each macro-haplogroup is more common. Reading age estimates as if they would be something like accurate and repeatedly proven and fine-tuned C-14 data is a brutal error, totally naive. These are nothing but educated guesses, cased into a more or less correct algorithm to make it appear more serious.

Mathematics is a useful tool but for the best and most logic algorithm to produce more or less correct results you need first to define a whole set of variables that in these cases are nothing but best guesses. Hence the results are also highly speculative, not hard data on which to build any theory. You can take them as notional reference but never as evidence of anything, as they can perfectly be totally wrong.

As we've seen in the previous post, there is a striking resemblance between the distribution of P/B style (A1 - A4), as presented in the second of my mini-maps, and the distribution of tonal languages in the Old World.

Sadly, I don't think you have proven that resemblance. Most P/B survivals are not correlated with tonal languages. The tonal language map (present day) may resemble your tentative reconstruction for the post-catstrophe musical groups spread (hypothetical remote past) but it does not resemble at all the present day presence of such survivals.

The more I think about it the more clear I see that. At first I was sympathetic to your claim but now I realize that I was being naive, that there is no obvious (nor seemingly subtle) correlation between P/B survivals and tonal languages distribution in present day terms. Only your hypothetical Paleolithic spread reconstruction fits but that is not hard data, just a hypothesis. And you can't build a theory on a mere hypothesis alone.

German Dziebel said...

"Age estimates are a most inexact "science" with huge error margins and highly speculative, tentative, foundations. To me it borders pseudoscience, specially for how they "sell" it as "serious stuff"."

Same for phylogeny. How can L0 be 149K years old, while L0k 11K years old? It's not just this paper, it's every paper. What, these Khoisan and Pygmy sequences were hybernating for more than 100K years?

German Dziebel said...

"Most P/B survivals are not correlated with tonal languages."

At least in linguists' perceptions, tones are recurrent parallel innovations, and they may occur independently even within one single family. For instance, Edward Vajda, who established the connection between Ket and Na-Dene, believes, with Michael Krauss for Na-Dene, that tones in Ket and in a few Athabascan languages, including Navajo, are parallel developments (see his "A Siberian link with Na-Dene languages"). There very well may be some tendencies in the linguistic structure that favor the emergence of tones but tones as surface manifestations are not reliable for long-distance comparisons.

Maju said...

How can L0 be 149K years old, while L0k 11K years old? -

I haven't paid any attention to this particular clade yet but technically it is perfectly possible.

What such a long stem means is that the lineage did not successfully expand in all that time, remaining a "private" (tiny) lineage until it found recently such an expansive opportunity. In all that time it has naturally accumulated many mutations, which make up the long stem.

Instead other lineages, most probably, show shorter stems, often as short as only one mutation. This means that they expanded shortly after their parent lineage did, or even in that same process or a sub-scene of it.

DocG said...

Maju: "Reading age estimates as if they would be something like accurate and repeatedly proven and fine-tuned C-14 data is a brutal error, totally naive."

Stephen Oppenheimer, Vincent Macaulay, and Martin B. Richards are respected people in this field, with long track records. Their paper was published in a professional, peer-reviewed journal. You sound like German, who also finds it very easy to dismiss whatever doesn't suit him. If the geneticists are full of crap then why are we bothering with any of this stuff?

"Most P/B survivals are not correlated with tonal languages."

And you know this how? Actually there IS a very strong correlation in New Guinea between Papuan speakers and P/B vs. Austronesian speakers who don't sing in this manner. Not every Papuan language is tonal, so I can't say for sure, but many Papuan languages are in fact tonal, and Austronesian ones are not. In any case, direct comparisons on a language by language basis don't mean much because Austronesian languages are relatively recent and may have replaced tonal ones. Same with Indo-European languages in Europe.

"The tonal language map (present day) may resemble your tentative reconstruction for the post-catstrophe musical groups spread (hypothetical remote past) but it does not resemble at all the present day presence of such survivals."

The only hypothetical instance is the Indus valley, which I put on that map long before I was aware of the language distribution. Otherwise, the correlation holds as a reflection of present day survivals of P/B, at least insofar as the initial Out of Africa migration is concerned. The presence of P/B in the circum-Arctic represents a much later development and also a relatively minor one since it pertains only to throat singing, which is not typical for these groups.

The gap is real, Maju. YOu can argue over what it means, but you can't deny the correlations, which reflect, for the most part, current conditions.

Maju said...

Stephen Oppenheimer, Vincent Macaulay, and Martin B. Richards are respected people in this field, with long track records. Their paper was published in a professional, peer-reviewed journal. You sound like German, who also finds it very easy to dismiss whatever doesn't suit him. If the geneticists are full of crap then why are we bothering with any of this stuff? -

Do you agree that even the smartest and wisest person can commit errors? I do. And that's why one has always to be critical and self-critical.

Anyhow Oppenheimer is no geneticist, he's a physician. Not that I mind because he does seem knowledgeable anyhow, but just for the record.

The issue of age estimates has been around for long. There are serious issues in the fundamentals of such equations, which even if all the relevant variables would be known with precision, would still be just statistical approximates because there is no objective way to know the exact time it took any specific mutation to evolve and consolidate.

What we know from SNPs is a phylogenetic tree: we know the structure, the nodes, the sequence (like M1 only happening after M)... but we can't know only from that the age with any kind of precision.

Geneticists have tried to go further and build approximative equations to estimate ages. Fine but these equations rely on assumptions that can perfectly be wrong, like the age of divergence of Pan-Homo (itself quite uncertain but surely quite older than the usual assumptions), mutation rates (a highly controversial matter, though this has advanced quite a bit) and effective population sizes (of which we just have no idea at all).

These methods often produce contradictory results. I believe I have already mentioned that study in which H11 is older than 40 Kya, while the usual estimates for H as whole are much younger (12-18 Kya), though not necesarily more correct. Furthermore more often than not the resulting age estimates make no sense with the known archaeological record, which is a true problem when dealing with well researched regions like Europe.

Finally there are some vested interests in promoting the belief that age estimates are accurate: there's a whole host of commercial DNA testing companies that want to give their customers the feeling that they can get to know with great precision their ancestors and for that they need to sell these educated hunches as hard data, which they are not.

When C-14 datation was introduced, it first had to provide some reliable results with ancient objects of known age. C-14 passed that scientific trial with good qualifications but the molecular clock hypothesis and its derived methods have never been proven in such or any other way. They are not validated methods, just hypothetical constructs. And, as I said before, you can't build any theory on a mere hypothesis nor you can sell a mere hypothesis as hard data.

I don't even question that the MCH has some merits, what I question is that attitude of selling highly speculative age estimates as if they would be factual information. That is misleading and a betrayal to the ethical code of science.

First prove the method scientifically and then I will accept such results as reliable. I've been waiting for years for that to happen... in vane.

And you know this how?-

Following the information you provided.

The last map of your musical hypothesis shows in red (A, B1) those cultures where P/B persists in some way, right? These cultures are distributed in totally different ways than the tonal language map.

Actually there IS a very strong correlation in New Guinea between Papuan speakers and P/B vs. Austronesian speakers who don't sing in this manner.

Maybe. But not elsewhere.

(cont.)

Maju said...

The gap is real, Maju. YOu can argue over what it means, but you can't deny the correlations...

If the correlation between tonal languages and P/B persistences does exist, then you have failed miserably in explaining it. Tonal languages exist in SE and East Asia but I see no red-coded (P/B) musical traditions there, excepting arguably one single instance of B1 in Philippines (Negritos I presume) and precisely in an area with no language tonality today.

Austroasiatic, Daic or Sino-Tibetan peoples do not have P/B traditions in your map, yet they are the main instance of tonal languages out of Africa.

I'm sorry to put you against the ropes. But not doing it would be failing to the science and honesty. I really can't see it.

Anyhow, unlike German or yourself, I don't have a hypothesis I need to justify: I just look at the puzzle and look for the best possible solutions (or remain perplex when I can't find them). Of course, now and then, one may feel it's the time to put some hypothesis forward for discussion and reality check, but one must be ready for it to be debunked, either totally or partially.

I think your approach is very interesting, as you know, but I also think that your hypothesis fails to explain what really happened in Eurasia after the OoA, with or without catastrophe. There must be some other, better explanation.

DocG said...

Maju: "there is no objective way to know the exact time it took any specific mutation to evolve and consolidate."

As you know, all such results are presented as estimates and the times are reported as ranges, not exact time points. What we see in the phylogenetic map represents an average or mean, though I admit the use of a single number can be misleading.

While absolute time estimates, even with ranges, are admittedly problematic, relative times are far less so. And what's relevant to me in this publication is not the absolute timing, but the inference that M in India appeared at a later date than M in SE Asia.
That is consistent with other reports on the genetic picture in South Asia which also present this region as problematic in OoA terms(see my most recent post). It's the problematic role of South Asia that interests me most, not the exact timing. Their report on the timing anomaly is just another example of some sort of problem in assessing the history of South Asia that might be linked to the gap that's so evident between Africa and SE Asia (assuming one is willing to see the obvious).

You can't accuse the authors of being self-serving, since the timing they present is inconsistent with humans in South Asia during the Toba eruption, which shoots down a major hypothesis of Oppenheimer's.

And if you are so eager to dismiss such results, then you will also be forced to consider the relevance of Toba as at least a possibility. And if humans were in fact in this region when Toba erupted, then we would certainly expect to see the sort of bottlenecks you would prefer to deny, even as a possibility.

I am open to ALL possibilities, Maju, but they have to make sense and they have to have at least potential explanatory power.

There are many things that for me require explanation which for you are taken for granted as simply the way evolution works. You see no problem and no need to look for explanations, whereas I do NOT take these things for granted. Cultural drift based on straightforward migrations can not produce the patterns we see. If you aren't bothered by that, then of course my struggles to explain the patterns must seem to you as completely pointless.

DocG said...

Maju: "The last map of your musical hypothesis shows in red (A, B1) those cultures where P/B persists in some way, right? These cultures are distributed in totally different ways than the tonal language map."

OK, I think I see where the problem lies. There is only so much detail I could cram into those maps, so certain things can be misleadig. Actually I'm pleased to learn that you've been studying that map so carefully.

To clarify: Map 2 actually presents the current distribution of P/B in Africa, SE Asia, Island SE Asia and Melanesia. The only hypothetical is the positioning of A3 and A4 in northern Pakistan. Most of this is omitted from the last map simply to avoid clutter. The last map is intended to represent later migrations, not the complete current world distributions of these styles.

B1 is an offshoot from P/B, a variant that's commonly found today also in SE Asia and Melanesia and also Central and S. America. But it is measurably different from A1-A4 and is NOT found in Africa.

The only hypotheticals in any of these maps are the placement of A3 and A4 in Pakistan and the placement of B2 in South Asia, which represents the hypothetical origin point of a style that could have migrated to other parts of Asia and is no longer typical of South Asia. Everything else you see represents the current distribution of these styles as gleaned from recordings made in the 20th and 21st centuries.

DocG said...

Maju: "Tonal languages exist in SE and East Asia but I see no red-coded (P/B) musical traditions there, excepting arguably one single instance of B1 in Philippines (Negritos I presume) and precisely in an area with no language tonality today."

I apologize for not explaining these maps more fully. It's not surprising that you are misreading them. The African signature of P/B currently abounds in precisely the same regions where we see such a saturation of tone languages, i.e., Africa, SE Asia and Melanesia. This is reflected in the second map, but not the first, where it is omitted to avoid clutter.

As for the Philippines, as you know this is largely now an Austronesian speaking region, but was clearly not always so. If all the surrounding languages are tone languages, it can probably be safely assumed that there were tone languages in the Philippines as well prior to the spread of Austronesian.

But that's not the most important point, because that is open to interpretation. What's most important is the striking similarity in the overall picture when we compare, say, my second map with the map of tone language distribution. This is NOT open to interpretation, it represents the state of affairs in both world music and linguistics until very recently.

"Austroasiatic, Daic or Sino-Tibetan peoples do not have P/B traditions in your map, yet they are the main instance of tonal languages out of Africa."

There are many instances of A1-A4 and B1 in this region. Whether there is a direct correlation between tone languages and these styles is yet to be determined, as I only became aware of this possibility very recently. But even if there isn't a one to one correlation in every case, the overall pattern is not easily ignored. Unless you prefer to lead your life with eyes wide shut.

Maju said...

... relative times are far less so...

Only in the case we are discussing nodes in the same branch. We know that M1 and M2 are younger than M necesarily because they are their descendants but it's not really possible to compare with great accuracy the relative age of M1 and M2. It may be more clear in some cases, when "sister" lineages have very different length at their stems but otherwise...

Anyhow, what Soares' paper does is akin to this example: a family has 5 daughters but they live in two different villages nowadays. The oldest and the two youngest daughters live in village A, while the second and third daughters live in village B. So you get maybe that the sisters living in village A average 25 years, while those living in village B average 27. But still the elder daughter and the majority of daughters (highest diversity) live in village A.

Village A is of course South Asia, while village B is Eastern Eurasia (including Sahul). Which is the original village of the family? Hard bet, but I'd say that village A is more likely to be the ancestral homeland.

You can't accuse the authors of being self-serving...

I am not accusing anyone of anything. I'm just providing a valid criticism of their alleged findings.

And if you are so eager to dismiss such results, then you will also be forced to consider the relevance of Toba as at least a possibility. And if humans were in fact in this region when Toba erupted, then we would certainly expect to see the sort of bottlenecks you would prefer to deny, even as a possibility.

Well, the archaeological evidence in South Asia is strongly suggestive of pre- and post-Toba continuity (same tools and culture under and above the Toba ash layer: they survived, they kept the culture, a culture that seems strongly related to Southern African MSA, per Petraglia, which we have discussed before). Hence even Toba might not have been enough to cause a true bottleneck (i.e. the killing of a vast majority of people, above 90%, preferably above 99%).

But anyhow I am not even questioning the bottleneck hypothesis, what I'm saying is that all your model for Eurasia, as you have exposed in the last posts, is inconsistent, even with the bottleneck hypothesis. The three pieces of the puzzle (genetic, linguistic and musicological) don't seem to fit with each other: each is telling a totally different story... apparently.

I feel that you're forcing their convergence without really making it work anyhow. They just refuse to converge.

(cont.)

Maju said...

To clarify: Map 2 actually presents the current distribution of P/B in Africa, SE Asia, Island SE Asia and Melanesia. The only hypothetical is the positioning of A3 and A4 in northern Pakistan. Most of this is omitted from the last map simply to avoid clutter. The last map is intended to represent later migrations, not the complete current world distributions of these styles.

Yes, this seems to be part of the misunderstanding. Do you have a better map of real present day distribution of P/B derived styles?

A crucial issue is, anyhow, that at least Sino-Tibetan and Daic families are tonal AND they come from modern day China (it's not totally impossible that Austroasiatic also does), where there doesn't seem to be any P/B survivals and yet fully belongs to the Eastern Eurasian line of genetic spread, which surely has its origins in SE Asia. Excepting the issue of tonality in language, the same can be said of most of NE Asia and (by extension) America.

I may be missing some fine detail but I still think that your model needs a lot of fine-tunning before it works well.

Maju said...

I get the problem in my interpretation. I was assuming that the last map was more or less the current spread.

However I still have issues, as mentioned in another thread. All those tonal-language families are known to have spread recently from China, or, in the case of Austroasiatic maybe Indochina. They do not belong to the Negrito area (Sundaland-Wallacea) more than Austronesian (at least it is not something we can assume as a fact).

China has no or almost no P/B survivals anywhere, even if most people there speaks tonal languages. On the other hand, the only genuinely Negrito preserved languages (Andamanese) are not tonal.

So I still protest that the P/B survivals' area and the tonal languages' area don't seem correlated, at least not clearly.

What's most important is the striking similarity in the overall picture when we compare, say, my second map with the map of tone language distribution. This is NOT open to interpretation, it represents the state of affairs in both world music and linguistics until very recently.

I don't see them being parallel. Sorry. They match only with a lot of wishful thinking, if at all.

There are many instances of A1-A4 and B1 in this region. Whether there is a direct correlation between tone languages and these styles is yet to be determined, as I only became aware of this possibility very recently.

That would be crucial to prove your point, I believe.

But even if there isn't a one to one correlation in every case, the overall pattern is not easily ignored.

I'm saying and I insist that the overlap between P/B and tonal languages is forced, and not at all obvious. Maybe you are right, but so far, IMO, you have failed to explain it in a convincing way.

China and Indochina seem key to your language-music correlation hypothesis now. We will have to ignore Negritos for this because they now speak foreign languages (except in Andaman). So you basically have:

1. The China/Indochina tonal languages area.

2. Papua (this part seems consistent but it's not enough alone)

3. The scattered remnants in NE Asia/America and Europe, which don't correlate.

So I'd suggest you to pay due attention to the China/Indochina area if you want to lean the scales in your favor.

German Dziebel said...

Luis: "Anyhow, unlike German or yourself, I don't have a hypothesis I need to justify: I just look at the puzzle and look for the best possible solutions (or remain perplex when I can't find them)."

Not in my eyes. From the very beginning you behaved like an out-of-Africa watchdog, which instantly threw me off. Glad to see you are developing an "opposable mind." But you do need to generate more ideas of your own and keep beefing them up with interdisciplinary evidence.

Victor: "You sound like German, who also finds it very easy to dismiss whatever doesn't suit him."

I actually spent years educating myself in several fields of knowledge in very good schools, juggling various academic and applied projects and working hard on amassing and classifying my own data. The only field I'm struggling with is comparative musicology. I ask people to prove their ideas to me, not simply build "most parsimonious" trees. Unfortunately geneticists, who have access to large grants (Spencer Wells got 4MM for his Genographic Project) have developed a knowledge infrastructure that is hard to understand and criticize. That's why many scholars simply stay away from them or accept what they say at face value. Like you do. Again, this leaves genetic theories untested against other fields of knowledge but an impression is crafted that all sciences are in agreement. Geneticists also need to report on those grants in a linear and consistent manner. Out of Africa has been their stable deliverable for many years. It's indeed the most economical and marketable idea that has emerged in human origins research in the 20th century. But I have no confidence that it's true in essence.

DocG said...

Maju: "The three pieces of the puzzle (genetic, linguistic and musicological) don't seem to fit with each other: each is telling a totally different story... apparently.

I feel that you're forcing their convergence without really making it work anyhow. They just refuse to converge."

I am, as always, looking for testable hypotheses and this one is certainly testable. It could be wrong, but if it's wrong then there is still a lot to be explained, because Out of Africa implies a continuity so if we see discontinuity then either there is something wrong with Out of Africa or some other factor that hasn't been considered yet that caused these discontinuities. It's not up to me to prove they are there because they are obviously there, regardless of whether all of them are related or not. The burden of proof is also on you and others who support OoA to explain them. Or else confess that you can't say for sure that this migration actually occurred.

Whether or not all four aspects of my hypothesis are connected with one another, and can be adequately correlated, there is most certainly a gap in the continuity between Africa and SE Asia and the gap is most certainly in South Asia. If you want to deny it then it is you who is forcing things into a neat package, not me.

DocG said...

Maju: "A crucial issue is, anyhow, that at least Sino-Tibetan and Daic families are tonal AND they come from modern day China (it's not totally impossible that Austroasiatic also does), where there doesn't seem to be any P/B survivals and yet fully belongs to the Eastern Eurasian line of genetic spread, which surely has its origins in SE Asia."

You are saying that the tonal languages of SE Asia originate in China? And that prior to that they were not tonal? How do you know that? Where in China? And when did this happen? If you can demonstrate that for me, then I'll forget about trying to connect P/B and tonal languages. But the gap will STILL remain, nonetheless and will still need to be accounted for.

And by the way, the African signature does indeed extend to China, though it is no longer reflected in the vocal music (except for South China, where most of the indigenos groups are located). We see it in the importance of panpipes, especially historically, and also in the many gong chime ensembles, also found largely in archaeological sites rather than living traditions.

Japanese Gagaku, which is organized according to so-called Colotomic structure, reflecting hocket as a basic principle, has its origin in China, though it is no longer practiced there I don't think.

You are assuming you know more than you do, Maju, and making all sorts of unwarranted assumptions. But if you can explain to me how all tone languages of SE Asia derive from Mandarin Chinese, all will be forgiven. :-)

DocG said...

Maju: "So I still protest that the P/B survivals' area and the tonal languages' area don't seem correlated, at least not clearly."

Clearly they are correlated, as should be evident from my map and the tone language map. We see P/B survivals in essentially the same regions as tone language: SS Africa, SE Asia and Melanesia. And if I were to provide you with a more exact map, including all the many examples of instrumental hocket, the overlap would be even more striking.

But you are assuming that by "correlated" I mean something different from what I've actually claimed, in fact something far more ambitious than what I've claimed.

I am NOT claiming that P/B is directly correlated with tone language on a group by group basis. In fact I'm not even claiming a correlation between P/B and tone language. (Though I find such a possibility intriguing.) What I'm saying is that there is a correlation between the distributions of certain features, including P/B and tone language that reveals a very interesting and hard to explain gap in the southern migration path, a gap that needs to be explained.
And I am trying to figure it out. It's the gap I want to focus on for now, not other possibilities that are interesting but at the moment secondary.

Even if I had the information on tone language from every single group that has P/B-related traditions, it wouldn't mean much because, as you know, there has been a great deal of language change in SE Asia, due mainly to the expansion of Austronesian but no doubt also to other linguistic expansions that have taken place since these areas were first populated with modern humans.

It's even possible, as you imply, that much of the tone language we see in SE Asia is a relatively late development due to the influence of Chinese. If that's the case that would interest me very much and would make a difference certainly.

But even if you could demonstrate that, the correlation I've pointed to would remain -- and the gap I've pointed to would remain. It would have a different meaning, yes. But it would still need to be explained.

You assume a lot about me, Maju, which is too bad, since usually you've been a very sympathetic and understanding commentator. I have no ax to grind over this issue. The only aspect of any of this that I would insist on would be the Pygmy/Bushmen connection -- for me that is completely solid. As for the rest, it's part of an exploration that I'm conducting to see how far it's possible to take all the various evidence and the ideas growing out of it, and see if it might be posssible to build a coherent history of early human culture.

I have some ideas I find interesting and I want to explore them. And since you have not yet convinced me that these ideas are all wet, I will continue to explore them, thank you. However, if you can suggest a viable alternative that makes more sense, believe me I will have no trouble exploring YOUR approach.

Maju said...

From the very beginning you behaved like an out-of-Africa watchdog, which instantly threw me off.

Not sure what you mean by this. OoA is more than sufficiently demonstrated, whatever German thinks. I am persuaded that this is that way, and so is everybody but you.

I do hold that, barring whatever new research can find, the currently known haploid phylogenies are perfectly valid and very much unquestionable. However if genetic research would some day show otherwise, I'd have to change my opinion. But I'm not going to change it only on alleged cultural correlations that are too easy to misunderstand and misinterpret: they are not on their own valid scientific evidence.

OoA is as solid as Heliocentrism or the Law of Gravity. Questioning it may be an entertaining intellectual exercise for a high IQ "stoned" group of friends with nothing better to do... but is essentially pointless, specially when no new hard evidence is questioning it.

You may be thrown off by my critcisms, German, but in truth I am probably one of the few people who is paying you any attention (even if often reluctantly).

Maju said...

Victor: I think you might want to create a new map, showing clearly the current status of P/B remnants in Greater Eurasia. Of course this is a request for my convenience, you don't have to if it doesn't suit you (I can always browse the blog carefully and build one myself).

We see P/B survivals in essentially the same regions as tone language: SS Africa, SE Asia and Melanesia.

I'm excluding ultra-Saharan Africa for this discussion because it's Eurasian processes what matter here, right?

If so there are, as I outlined before, the following cases of P/B survivals:

1. Papua (tonal languages)
2. Parts of SE Asia but not China (some imperfect overlap with tonal languages, which might well have arrived from China in the Holocene)
3. The Arctic-American group/s (no tonal languages mostly)
4. The West Eurasian group (no tonal languages mostly)

The overlap is extremely imperfect in my opinion. You may be right or not in these two elements being related but the case has not been proposed in clear incontrovertible terms for Greater Eurasia.

And if I were to provide you with a more exact map, including all the many examples of instrumental hocket, the overlap would be even more striking.

That would be really interesting to see.

But you are assuming that by "correlated" I mean something different from what I've actually claimed, in fact something far more ambitious than what I've claimed.

You have been very absolute and insistent in your claim on this parallel, as something almost unquestionable, obvious... virtually the same thing. What I say is that I don't see it clear, not for Greater Eurasia.

(cont.)

Maju said...

I am NOT claiming that P/B is directly correlated with tone language on a group by group basis. In fact I'm not even claiming a correlation between P/B and tone language.

I understood from your last posts that you were claiming that. Maybe I misunderstood you but maybe also you emphasized this matter a lot, as if it would be some sort of extra evidence and not just a possibility.

What I'm saying is that there is a correlation between the distributions of certain features, including P/B and tone language that reveals a very interesting and hard to explain gap in the southern migration path, a gap that needs to be explained.

If this is what you actually say, then I'm ok with it. Not sure if tonality may have evolved and devolved several times, as German claims, but I agree that it is an interesting possibility.

Even if I had the information on tone language from every single group that has P/B-related traditions, it wouldn't mean much because, as you know, there has been a great deal of language change in SE Asia, due mainly to the expansion of Austronesian but no doubt also to other linguistic expansions...

Agreed. Although Austronesian expansion only applies to Sundaland/Wallacea. There have been almost for sure other linguistic expansions: at least Daic and Tibeto-Burman, from what is now China. We also don't have any evidence that Negritos spoke tonal languages before Austroasiatic and Austronesian arrival (and Andamanese languages are rather evidence in the opposite direction).

It's even possible, as you imply, that much of the tone language we see in SE Asia is a relatively late development due to the influence of Chinese.

I don't say it's Chinese influence as such. What I say is that all SE Asian tonal languages have roots in modern mainland China and expanded from there. The only possible exception would Austroasiatic, which is at least as old as Neolithic in the area (and might be argued to be rooted in Hoabinhian - though very controversially) but is for sure not native to Sundaland/Wallacea.

What I say is that the tonal languages area of Eastern Asia is not specifically SE Asian in any way, but rather of the China area. There are doubts re. Austroasiatic but only this one, and in any case don't apply to the southern (now mostly insular) half of SE Asia, where it's clearly a Holocene arrival as well.

...

I don't think there is any axe-grinding about this. I just happen to be quite unconvinced of the parallel you draw (or my understanding of it) and I express this disconformity the best I can. You (or anyone) should not only expect "sympathy and understanding", as I have a very critical nature. While this attribute may not be the best for social relations, it's really nice for scientific discussion and I always hope to be valued for it.

However you now clarify that it's not meant to be a perfect parallel but just another sign of "the gap". Then I can agree, yes.

German Dziebel said...

"OoA is more than sufficiently demonstrated, whatever German thinks. I am persuaded that this is that way, and so is everybody but you. "

Consensus is a poor predictor of truth. Out of Africa hasn't been scientifically demonstrated, otherwise I would've seen it. People who created out of Africa know only one thing, they aren't well rounded in all the disciplines involved, they dismiss what they don't understand. Your current debate with Victor attests to the fact that out of Africa has serious problems. I suggest a solution.

"You may be thrown off by my critcisms, German, but in truth I am probably one of the few people who is paying you any attention (even if often reluctantly)."

Your style of writing has nothing in common with criticism, as you lack an understanding of how trees are constructed, how genetic diversity evolves and how languages and kinship systems change. You have been a liability, rather than an asset to an academic discussion so far. But you're making some progress. Keep it up, we'll get there everntually.

DocG said...

Maju: "Not sure if tonality may have evolved and devolved several times, as German claims"

Tonality, or any other cultural attribute for that matter, can only evolve or devolve within a given society. It can't do so en masse, with the same changes taking place in several unconnected societies at once -- unless all these societies were subject to the same external forces (such as the Toba explosion, for example, or some other large-scale event).

So it's very difficult to see how large-scale differences, such as the differenced between Africa, South Asia and SE Asia, could result from local evolutions-devolutions on a group by group basis. That sort of thing would produce checkered results, more or less equally distributed in all regions.

Maybe there should be a field within anthropology called "distribution studies," because so few anthropologists seem to understand the meaning of different distribution patterns. Large scale differences in distribution clearly have a very different history than small-scale differences.

DocG said...

Maju: "What I say is that all SE Asian tonal languages have roots in modern mainland China and expanded from there."

Wikipedia: "The Kradai or Kra-Dai languages, also known as Daic, Kadai,[1] or Tai-Kadai, are a language family of highly tonal languages found in southern China and Southeast Asia. The diversity of the Kradai languages in southeastern China, especially on Hainan, suggests that this is close to their homeland."

Are these languages derived from modern Chinese? Looks to me as though modern Chinese could be derived from one of them.

Also from Wikipedia: "Kradai-speaking populations originated in the southern part of East Asia and then migrated northwards and eastwards with Kam-Sui probably being the oldest."

The evidence from New Guinea can't be ignored either, though it's not strictly speaking SE Asia. Many of the highland Papuan languages are tonal. The languages of most of the coastal groups are Austronesian, i.e., much more recent, and are of course not tonal.

So there is certainly enough evidence of archaic tonal languages at the far reaches of the southern route to contrast rather strongly with the absence of any sort of tonal languages in S. Asia.

DocG said...

Maju: "However you now clarify that it's not meant to be a perfect parallel but just another sign of "the gap". Then I can agree, yes."

OK, good. What's important to me is that this very strange looking large scale difference of distribution reveals a gap that can apparently be accounted for only through historical analysis. It could not have developed simply through normal channels of evolution or devolution, which would have produced a very different distributional pattern.

Maju said...

Are these languages derived from modern Chinese?.

When did I say that? Please!

Modern China is a geographic region and the southern half of it (at least) was not originally Chinese-speaker. We know that Chinese language expanded from North to South as the Chinese state expanded itself. It did so over other ethno-linguistic peoples, who were often absorbed. Some important remnants of them still persist and all are directly related to the languages modernly spoken in SE Asia (Kradai, Tibeto-Burman, Austroasiatic and Asutronesian families).

I understand that there is some evidence that strongly suggests that these linguistic groupings migrated from North to South. For example, of five primary subdivisions of Kradai, four are found in South China with only Daic being more common in Indochina.

Tibeto-Burman is even argued to have migrated from Northern China (within the Sino-Tibetan phylum, somewhat controversial)

The only real doubt is about Austroasiatic which appears old enough in the region to be at least Neolithic (some would argue surely that it coalesced in the Hoabinhian culture but I'm not really sure).

So I am not claiming that tonality in these language families is derived from Chinese language but very possibly from the geographical area that corresponds with modern China. I only have serious doubts about Austroasiatic in this regard, which was almost for sure partly replaced by Daic and Burmese languages in relatively recent times.

Also from Wikipedia: "Kradai-speaking populations originated in the southern part of East Asia and then migrated northwards and eastwards with Kam-Sui probably being the oldest".

Are you trying to insult my intelligence? I can perfectly read the many "citation needed" and "vague" clauses added by disgruntled users to that most confusing section. I have also been a Wikipedian myself in the past and I know that bad articles/sections can and do exist.

Just for safety, I checked the Spanish Wikipedia and, know what?, it says exactly what I thought:

"Las lenguas tai-kadai se originaron en el sur de China, que es el hogar de la mayoría de las subfamilias del grupo. Hablantes de lenguas tai se desplazaron hacia el sudeste asiático en tiempos históricos, fundando naciones que más tarde se convertirían en Tailandia y Laos".

Translation: "The Thai-Kadai [Kradai] languages originated in Southern China, which is the homeland of most subfamilies of the phylum. Speakers of Thai [Daic] languages migrated to SE Asia in historical times founding nations that eventually would become Thailand and Laos".

I am not going to work researching this matter further right now (if you happen to insist I'd consider doing it) but I have never before read any such hypothesis. I used to be member of an anthropology forum with very knowledgeable East Asians and all thought that Kradai has a South Chinese origin and that Austroasiatic was the only language phylum that could be considered native to mainland SE Asia, if any among the modern ones.

Maju said...

The evidence from New Guinea can't be ignored either...

True. But it is not too conclusive, specially with other pre-Neolithic survivor languages of the area south of Indochina (Andamanese, Australian Aborigines) not being tonal. Of these Papuans are also the only agriculturalists.

Whatever the case, what we have in SE Asia in this aspect is a big question mark, rather than the "clear evidence" you have suggested.

So there is certainly enough evidence of archaic tonal languages at the far reaches of the southern route to contrast rather strongly with the absence of any sort of tonal languages in S. Asia.

Well, actually you have two: Burusho and Munda (Austroasiatic). And all the rest is Dravidian and Indoeuropean (except for certain very minor language in the verge of extinction - I have to check that one). Indoeuropean is clearly a recent arrival and Dravidian is disputed: some have argued that spread only in Neolithic times.

However I'm willing to concede in this because you may well be somewhat correct, even if I am not sufficiently certain that you are. But I cannot but warn you that the conclusions you have reached re. tonal languages are not as solid and clear as you'd wish.

I'm really sorry to "sabotage" your last addition to the puzzle's possible solution but I feel I must. Science is built upon critical thought.

A possibility, even a probability, is not an incontrovertible fact.

Maju said...

The other language of India I mentioned before is Nihali. I don't know if it's tonal or not but has been strongly influenced by an Austroasiatic language (and secondarily Indoeuropean). It is a language isolate and only spoken (1991) by some 2000 tribal people.

You may want to research it a bit.

DocG said...

DocG: Are these languages derived from modern Chinese?.

Maju: "When did I say that? Please!"

And I quote:
Maju: "What I say is that all SE Asian tonal languages have roots in modern mainland China and expanded from there."

You're confusing me, Maju. Either you believe they originated in modern Chinese or not. It sounds like you just claimed that the former is so. If that's not true, then please explain exactly what it is you DO mean.

Maju: "For example, of five primary subdivisions of Kradai, four are found in South China with only Daic being more common in Indochina."

But southern China is a well known center of indigenous culture and many of these people have a very different type of culture entirely than that of China "proper." They certainly have very different musical styles.

Maju: "So I am not claiming that tonality in these language families is derived from Chinese language but very possibly from the geographical area that corresponds with modern China."

Yes. And so . . . ? What is your point? Are you saying that the map of tone languages appearing on the WALS website is misleading, that SE Asia is NOT a center of tone language after all, that this is merely some illusion produced by the fact that "all SE Asian tonal languages have roots in modern mainland China and expanded from there"? If so, then I suggest you contact the editors immediately to inform them of their error. And if this is not what you mean, then why are you wasting our time with these observations?

Here is another Wikipedia quote (http://infao5501.ag5.mpi-sb.mpg.de:8080/topx/archive?link=Wikipedia-Lip6-2/39573.xml&style):
"Tone is frequently an areal rather than a genetic feature: that is, a language may acquire tones through bilingualism if influential neighboring languages are tonal, or if speakers of a tonal language switch to the language in question. For example it is generally accepted that tone spread to the Chinese languages through the influence of another language family, most likely Miao-Yao . . . Something very similar happened in Vietnamese, probably under the influence of Tai-Kadai languages."

I know you're suspicious of Wikipedia, but this person appears to be authoritative, and seems to be saying the opposite of what you are saying. I'm not saying he is right and you are wrong. But to be brurally honest, I see no reason to accept your opinion over this one. As I see it, this is a question that is far from settled. But since the experts who produced the WALS maps also seem to believe that the distribution of tone languages presented there represents something meaningful, I will accept their view of this rather than yours, at least for now, Maju. Same with the genetic evidence. Sorry for being so "brutal," but there it is.

Maju said...

DocG: Are these languages derived from modern Chinese?.

Maju: "When did I say that? Please!"

And I quote:
Maju: "What I say is that all SE Asian tonal languages have roots in modern mainland China and expanded from there."


Exactly: there is not a single word about Chinese language.

You're confusing me, Maju. Either you believe they originated in modern Chinese or not.

China is not the same a Chinese language.

But southern China is a well known center of indigenous culture and many of these people have a very different type of culture entirely than that of China "proper." They certainly have very different musical styles.

All southern China was once, not so long ago not Chinese ("indigenous" or whatever you want to call it).

Yes. And so . . . ? What is your point? Are you saying that the map of tone languages appearing on the WALS website is misleading, that SE Asia is NOT a center of tone language after all, that this is merely some illusion produced by the fact that "all SE Asian tonal languages have roots in modern mainland China and expanded from there"?...

Yes.

Notice that misleading is not the same as wrong. The map is perfectly correct (AFAIK), just that you are letting yourself being misled by the illusion so many dots on SE Asia create.

If so, then I suggest you contact the editors immediately to inform them of their error.

Are you kidding? The map is correct, it is YOU who is letting himself mislead out of wishful thinking. I even allowed myself to fall to the optical illusion at the beginning. But soon I realized I had to recant and call your attention on the subject.

For example it is generally accepted that tone spread to the Chinese languages through the influence of another language family, most likely Miao-Yao...

This is new to me but in any case Mia-Yao is a language group of Southern Chinaz, probably much more widespread in the past.

Something very similar happened in Vietnamese, probably under the influence of Tai-Kadai languages.

I don't know what to think, but this guy is in fact claiming that Austroasiatic (i.e. Vietnamese) are not true tonal languages (but "tonalized" by the indfluence of Kradai). It only makes clear that you need to research the matter further before making the claims you are making so lightly.

I know you're suspicious of Wikipedia...

I'm not suspicious of Wikipedia as such, I just know that the quality of articles varies a lot and some are blatantly wrong or one-sided. It is a good first hand reference but you have to be a bit careful, specially if there are no citations and/or the text seems to contradict other information. If there are citations, I normally prefer to look at them if they are open access.

But to be brurally honest, I see no reason to accept your opinion over this one.

I am not giving any opinion: I am warning you of feebleness in the identification of tonal languages as dominant in mainland SE Asia (Indochina for short). They are now but they may perfectly not have been in Paleolithic.

I say and mantain that the center of gravity of all those East Asian tonal languages appears to be in Southern China (which has nothing to do by itself with Chinese language, which is original from Northern China), not Indochina.

The case for Sundaland/Wallacea is even more murky.

Didin't you get my last post on Nihali (re. India)?

DocG said...

Maju: "I don't know what to think, but this guy is in fact claiming that Austroasiatic (i.e. Vietnamese) are not true tonal languages (but "tonalized" by the indfluence of Kradai)."

As I understand it, most Austroasiatic languages are not tonal, with Vietnamese as an important exception. If this is in fact the case, the Wikipedia explanation makes sense.

But from what you've been saying I had the impression you believed that ALL tone languages in this whole region extending from southern China to SE Asia were due to the influence of modern Chinese. So I'm relieved to learn that this was a misunderdstanding on my part.

Maju: "I am warning you of feebleness in the identification of tonal languages as dominant in mainland SE Asia (Indochina for short). They are now but they may perfectly not have been in Paleolithic."

A great many things have obviously changed since the Paleolithic. We have no way of knowing for sure what any of these groups were speaking so long ago.

What I've been exploring (NOT asserting but exploring) is the possibility that the tone languages we now find in this general region (not only SE Asia but southern China, the Malay Peninsula, Melanesia, etc.) could be survivals from the tone language that was most likely spoken by HMP.

Whether these languages survive in SE Asia or southern China is of no interest to me whatsoever. What's important is their presence among indigenous peoples in the southeastern region of Asia. AND the strong contrast with what we see in South Asia, where there is an enormous gap with very little trace of tone language.

If all the tone languages of southern China and SE Asia were due to the relatively recent influence of the modern Chinese language(s), that WOULD matter, and that is what I thought you were claiming. AND it made little sense to me, so I'm glad this was a misunderstanding.

Maju said...

As I understand it, most Austroasiatic languages are not tonal, with Vietnamese as an important exception.

You may be right in this (at least in part, some other AA languages also seem tonal) but this would only make even less solid your claim of tonal languages being ancestral in SE Asia (excluding South China), because of all them only Austroasiatic languages are known to have some pedigree in the region (Neolithic at least).

But from what you've been saying I had the impression you believed that ALL tone languages in this whole region extending from southern China to SE Asia were due to the influence of modern Chinese.

Not at all, you were misreading me (much to the annoyance and confusion of both). All I said (and meant) was that Southern China (or China in general, if we include the Chinese language family) looks a more likely center of gravity for the tonal languages of the whole region.

So I'm relieved to learn that this was a misunderdstanding on my part.

And I'm relieved that you finally got it. :)

A great many things have obviously changed since the Paleolithic. We have no way of knowing for sure what any of these groups were speaking so long ago.

Precisely. The only of these phylums we can know for sure that was there in the Paleolithic is the Andamanese language family/-ies (not tonal, but very much isolated anyhow). Austroasiatic may also be very old in SE Asia but, as you made me notice now, is not all tonal. The rest are all recent arrivals of the last two millennia or so.

Whether these languages survive in SE Asia or southern China is of no interest to me whatsoever.

It is very important because the known flow of cultures and languages in the last millennia in all the region has been from North to South, not the opposite.

This seems central to the ability to prove or not the claim that SE Asia was originally tonal-speaker.

... the Malay Peninsula...

According to the map you posted first of tonal languages, there is no such thing in the Malay peninsula. Only one Austroasiatic language is spoken there today and doesn't seem to be tonal either (adding against AA phylum being originally tonal as this language surely arrived to Malaysia c. 13,000 years ago, with the formation of the Senoi in a Hoabinhian context). Your newest map on the matter does not say differently.

The more I look into the matter the less SE Asian tonal languages appear to me.

DocG said...

Maju: "The only of these phylums we can know for sure that was there in the Paleolithic is the Andamanese language family/-ies (not tonal, but very much isolated anyhow)."

The Andamenese are the only negrito group that would have been affected by Toba and in fact they would have been in the brunt of the worst of it. So if Toba is our engine of culture loss, then the Andaman Islands would have lost a great deal. And that seems to be the case. Not only do they lack a tonal language, their musical culture, so far as I've been able to determine, is extremely simple, to the extent that many songs appear to be monotonal, just the same note sung to different words. And they are apparently completely without musical instruments of any kind, except for a percussion platform. No one has done a systematic study of their music and there are only a few recordings, so my assessment may change in future.

But they are certainly not the only groups in that general region who were most likely "in place" during the Paleolithic. We have any number of highland New Guinea peoples as well, and many of their languages have been identified as tonal.

New Guinea is an extremely important indicator, along with parts of Island Melanesia, because it is one of the few places where Austronesian languages have spread without completely obliterating all the languages that preceded it. So if we find tonality among even some of the Papuan languages that suggests that many other languages now replaced by Austronesian languages might well have also been tonal.

DocG said...

DocG: "Whether these languages survive in SE Asia or southern China is of no interest to me whatsoever."

Maju: "It is very important because the known flow of cultures and languages in the last millennia in all the region has been from North to South, not the opposite."

I'm not interested in what happened in the last millennia, unless it can be proven that tone language originated and spread during that period, a theory which I don't think you could defend.

If Out of Africa is correct and the southern route is the correct one (which is the theory I have been exploring), then ALL people in East Asia originated in either SE Asia or even farther south. And this theory is reinforced by the recent genetic study I referenced earlier, on the origins of East Asians. So, if they were speaking tone languages, those languages would have spread from south to north. And the ones in the south, most of them, would have been obliterated by the recent spread of Austronesian languages. Which is probably why it looks as though those languages are confined mostly to southern China. If you are looking for an artifact, that's most likely it. The presence of tone language in highland New Guinea tends to confirm the above hypothesis, because this is one of the few places where we can peek behind the viel of Austronesian culture to see what was there earlier.

German Dziebel said...

I read around a bit, and it appears that tonal languages in Asia and in America (more hypothetically, as these languages are less known) are believed to have evolved multiple times. There are known triggers and diffusion paths that generated tones independently in Mayan, Athabascan, Ket and Sino-Tibetan languages. What seems to be peculiar about African tones is that there're no known triggers and diffusion paths for tones there.

This means that in Africa tones may have evolved earlier than in many parts of Asia and America, and the phonetic triggers have become murky with time. There's a chance that some languages in Asia retained their tones from the time when Asia was originally peopled (out of Africa?) but the fact that Australia lacks them tells me exactly what I knew originally, namely that tonogenesis is a pretty sporadic phenomenon occurring at different times and different places independently.

See Childs, Introduction to African Languages, 2003, pp. 85-86, with other literature.

Maju said...

Sincerely, I give up trying to persuade you of the feebleness of this part.

I don't see how Papua is more important than Australia or Andaman or the (seems originally non-tonal) Austroasiatic language family. It's an arbitrary choice of you.

I'm not interested in what happened in the last millennia, unless it can be proven that tone language originated and spread during that period, a theory which I don't think you could defend.

Spread in Indochina? Yes, I think it is very defensible that it happened that way.

But whatever...

DocG said...

German: "I read around a bit, and it appears that tonal languages in Asia and in America (more hypothetically, as these languages are less known) are believed to have evolved multiple times."

If that were the case in the Old World, then the distribution pattern would be very different and in fact much more random looking. This does seem to be the picture in the New World and what you've read may apply for the most part there. From what I've been reading, it would seem that tone language in Asia doesn't so much "evolve" as diffuse due to the influence of neighboring or conquering peoples.

DocG said...

Maju: "Sincerely, I give up trying to persuade you of the feebleness of this part."

Good, because I've been getting very impatient with your continual nitpicking over this issue. Clearly there are people in what could be called "Greater Southeast Asia" who have spoken "in tones" for a very long time -- and they include some of the oldest and most isolated societies, such as the southern Chinese indigenes and highlanders of New Guinea.

Whether there are other groups in the same general area who do not have tone languages is totally beside the point as far as my explorations are concerned. The overall picture presented in the WALS map may be misleading to some extent, but nevertheless there are more than enough native tone languages in this general area to present a very real contrast to what we see in South Asia. And that is all that concerns me.

You consistently assume that I am making some sort of case that I am NOT trying to make and then find all sorts of piddling exceptions that have nothing to do with me.

You seem to have picked up some really bad habits from reading too much in the archaeological literature where petty nitpicking of this sort is a kind of occupational disease. I have no interest in playing that sort of game. I'm not going to argue with you over whether it's a proven fact that some tone language or other is truly indigenous or comes from somewhere else, that would be pointless. I'm looking for large-scale patterns and in this case there is a huge pattern that's obvious. You can't see the forest for the trees, Maju, which is really a shame.

DocG said...

DocG: "I'm not interested in what happened in the last millennia, unless it can be proven that tone language originated and spread during that period, a theory which I don't think you could defend."

"Spread in Indochina? Yes, I think it is very defensible that it happened that way."

Why Indochina? Why are you so focused on just this one area? I could't at the moment care less about whether tone languages are native to Indochina. As well you know.

Tone language either developed independently in this entire region (which includes southern China, Indochina, Malaya, Indonesia, Melanesia, etc.) or else represents a survival from the language of HMC. I see it as a very exciting and promising question to explore, but you seem to see it as some sort of threat that must be debunked before it goes too far. Typical for an archaeologist, I must say. Which is why everyone in this field always seems to be at everyone else's throat. What a shame!

German Dziebel said...

"If that were the case in the Old World, then the distribution pattern would be very different and in fact much more random looking. This does seem to be the picture in the New World and what you've read may apply for the most part there. From what I've been reading, it would seem that tone language in Asia doesn't so much "evolve" as diffuse due to the influence of neighboring or conquering peoples."

Duh! Tones evolved in Asia independently of Africa, then spread around. That's why they look "non-random" to you. Tones evolved in America independently of Asia and didn't spread around because American Indian populations are more isolated from each other than Asian populations are from each other. Some tones in America probably died out without spreading. The Ket language for which we have data is in Asia. It evolved tones independently of either African, or Southeast Asian or American sources of tones. This language is near dead now, so are its tones.

Victor, you constantly confuse places of origin and routes of migration with places of survival and pathways of innovation. In fact, in "The Genius of Kinship," the chapter about Darwin, Lyell and other 19th century theorists of "origins" is precisely about these important distinctions.

Maju said...

Ehm, just for the record, Indochina = mainland SE Asia, not just former French Indochina as some people seem to believe. Indochina includes that plus Thailand and Burma. Hence it's a major and central subdivision of SE Asia.

The other regions that can be included in SE Asia are Sundaland (the continental platform south of the Kraa isthmus), Wallacea (the islands between Wallace and Lydekker lines, for some also including Philippines) and Southern China (excluding the Tibetan Plateau, which is rather part of Central Asia or a region on its own right).

The "entire region" appears fragmented in the genetic landscape, probably because it was a crossroads at the Eurasian expansion, with different groups of people moving (at intergenerational level, of course) through it in several directions and it has clearly different prehistory and history after that. It is not like we should not make any distinctions. At least I do make them.

I'm looking for large-scale patterns and in this case there is a huge pattern that's obvious.

It is not THAT obvious: there are tonal languages in Papua and China (with spread into Indochina in recent times) but there are also non-tonal language phylums in that same area dating (probably) from the same process of Eurasian expansion. The pattern is mixed, even if tonal languages are more common in the East than in the West of the continent, where the pattern is not that mixed and tonal languages are very rare.

You can't see the forest for the trees, Maju, which is really a shame.

I can see the forest perfectly: it is a mixed forest not a simplified pine plantation.

It is you who is looking at one or two trees and saying: "look, all this forest is pines". It is not, there are many other trees too that are not pines.

DocG said...

Maju: "The pattern is mixed, even if tonal languages are more common in the East than in the West of the continent, where the pattern is not that mixed and tonal languages are very rare."

Exactly. You seem to think that I'm arguing for a clear picture of tone language dominance in the East and I'm not. The picture is mixed, granted. But the presence of tone language in this region can't be denied, it is there. And if it's not autochthonous in the East, but due to "independent invention," then why don't we find similar "independent inventions" in other parts of Asia? Or Europe? There's a huge chunk of the world totally devoid of tone languages. Which is strange if it's so easy for a non-tonal language to morph into a tonal one.

As I see it, the simplest and most logical explanation for East Asiatic tone languages is that they were always there from the beginning, as long as modern humans were there. And if some of them began as non-tonal and became tonal, that's most likely due to the influence of tone languages already in place, in the same region.

Maju said...

It's very possible in my opinion that tonality was in East Asia from the beginning. However, as you suggested indirectly, it might well have been circumscribed to one or two populations, like the Kradai, which in turn influenced others (like Austroasiatics or Han Chinese). Or, as I suspect, it may have been a common feature of the geographical region that we know modernly as China (as well as, separately, of New Guinea - which also has other peculiarities of its own).

If tonality was limited before Neolithic to parts of China and Papua, it would mean just two or so founder effects in an ocean of non-tonality or low tonality. In any case, sure, these founder effects were somewhat stronger initially and/or more influential in the long run than in the West (South and West Eurasia) but this could even be mere random distribution.

Such kind of founder effects are also apparent (though not necesarily parallel or correlated) in the genetic landscape, apparently dating from the early Eurasian Expansion so it's not anything that surprises me.

However, going a step further, I suspect (not sure) there might be a correlation between tonality and spread of some lineages such as mtDNA N and/or Y-DNA MNOPS, all of which are weak in South Asia.

Is tonality correlated with this linguistic feature and/or with some genetic feature. Naturally it is very hard to say but I would really have liked that you presented a case-to-case comparison to see if there is effectively any correlation at all or is just another of those semi-random founder effects.

You will surely ask why do these founder effects accumulate in Eastern Greater Eurasia? Of course the mini-bottleneck or pseudo-bottleneck hypothesis is a possibility but another reason is that the early spread of Eurasians was not the "U" (plus Sahul) we see now but rather a horizontal "T" (-|) pattern with its three poles in South Asia, Middle East Asia and Sahul, while West and Central Eurasia were closed to our species because they were under Neanderthal control and their accessible areas were rather arid possibly. This spread had the center of gravity in Indochina but the true origin in South Asia. So while East Asia and Sahul/Indonesia have some affinity this affinity is anyhow weak, very weak. Even the affinity between the various isolated remnant groups we know of south of Indochina is very limited and extremely old, so it's not like they are too correlated just for being east of South Asia.

With this diversity of Eastern populations rooted already in the early Eurasian Expansion, it is only logical to find the high diversity of local founder effects of all sorts, in genetics as in tonality and probably also in music.

India had its own early colonization led by one specific group (mtDNA plus Y-DNA F), which also participated in the colonization of the East but in the East we see that other groups had better opportunities themselves and hence co-participated in the colonization of these diverse regions, quite irregularly.