Sunday, February 7, 2010

308. Aftermath 23: Australia and New Guinea

Even if we assume that the hypothetical scenario I've been presenting is more or less on-track, a particularly challenging problem remains: the origin of the remarkable musical style that, as far as I've been able to determine, is characteristic of just about every Australian Aboriginal group throughout all regions of the continent. While there do seem to be some intriguing similarities between the dance styles of the Chenchu and Australian Aboriginals, as noted in Post 299, I've never been able to find any musical practice anywhere in South Asia that resembles the "unison/ iterative/ one-beat" style of the Australians.

I've been accused of going too far, at times, to justify my "pet theory" of musical evolution, and my attempt to account for Australian Aboriginal music in terms of some sort of bottleneck event, possibly due to the Toba eruption, is seen as a perfect example of how I am forced to go to unlikely extremes in order to make my theory work. No matter how many times I repeat myself, it seems to do no good, but just for the record: I do not have a theory of musical evolution, or evolution generally, either "pet" or otherwise. I am, very simply, trying to make sense of what happened in early human history, in terms of the various types of evidence available to me. To that end, I find it convenient to explore certain possibilities, in the form of testable hypotheses, which I then try, as best I can, to put to the test.

Since by far the strongest and most promising evidence comes from the realm of population genetics, and the strongest and most promising hypothesis is the "Out of Africa" model, my principal effort for the last several years has been to explore as fully as I can what that model implies for both musical and cultural evolution -- to determine whether what I know or think I know about the musical picture especially is or is not consistent with what I am learning about the picture currently being painted by the intrepid geneticists. If I can't get the pictures to coordinate into some sort of focus, then I will be forced to consider some other alternative. And that will be fine with me. If I were in it for the robe and bowl,* the course of my entire life would have been very different, I can assure you. I am here strictly and exclusively for the Dharma. (Though a bowl full of rice -- in the form of a grant -- would, from time to time, be welcome.)

If there is a "pet theory" to be justified by the bottleneck I have in mind, it is not a theory of music, but the "Out of Africa" theory itself. Because the existence of Australian Aboriginal "unison/ iterative/ one-beat" style is, on its face, very difficult to explain as an "evolutionary development" from anything remotely African. And if we might want to explain it away as an "independent invention," we would still be left with the problem of why such an invention was needed in the first place, if the ancestors of the Australians already had a perfectly viable musical tradition, with roots, like the Australians themselves, in Africa. On its face, the profound difference between this style and the P/B style I've associated with our HBP ancestors, and the HMP migrants, is so great as to be more consistent with the very different multi-regional model.

According to the multi-regionalists, the great diversity we see in the world of today, genetic, morphological and cultural is due to the fact that modern humans developed more or less independently in different regions of the world, so that things we have in common, like language, music, ritual, myth, marriage customs, kinship systems, religion, etc., must be the product of convergent evolution, i.e., some sort of "destiny," based on certain universal properties of the human mind that cause it to develop along certain lines and not others. A somewhat softened version of this model claims that many of these similarities must be due to various processes of genetic and cultural interchange due to the continual migration of various peoples throughout the world over millions of years. And any differences, such as the many differences between P/B and "Unison/ Iterative/ One-beat," wouldn't need explaining, since they would be due to the fact that these two styles arose under completely different circumstances, in completely different regions of the world. Any version I've ever heard of the multi-regional model makes little sense to me, while the Out of Africa model makes a great deal of sense. So, if the music of the Australian Aborigines can't somehow be accounted for by some version of Out of Africa, then I have to admit I'm stumped and will need to start over.

(to be continued . . . )

*From the Mumonkan: The Sixth Patriarch was pursued by the monk Myõ as far as Taiyu Mountain. The patriarch, seeing Myõ coming, laid the robe and bowl on a rock and said, "This robe represents the faith; it should not be fought over. If you want to take it away, take it now." Myõ tried to move it, but it was as heavy as a mountain and would not budge. Faltering and trembling, he cried out, "I came for the Dharma, not for the robe."

9 comments:

German Dziebel said...

"If there is a "pet theory" to be justified by the bottleneck I have in mind, it is not a theory of music, but the "Out of Africa" theory itself. Because the existence of Australian Aboriginal "unison/ iterative/ one-beat" style is, on its face, very difficult to explain as an "evolutionary development" from anything remotely African."

Very well put. I might add that all the marked differences between Australian aborigines and Africans, be it music or, say, craniology and dentition, occur against the background of unmistakable similarities between Australian aborigines and other non-Africans (e.g., their mtDNA and Y-DNA lineages are clearly related to other Eurasian haplogroups). Hence, Australian aborigines are not an aberrancy in a sea of normal progressions from Africa to the rest of the world but rather a very clear example of how warped is the current story of an out of African migration based on the decreasing molecular and phenotypic diversities from Sub-Saharan Africa into Asia, Australasia and America. Populations outside of Africa are not a subset of African populations, hence if they appear so in genetic studies, genetic studies are likely wrong.

DocG said...

German: "Populations outside of Africa are not a subset of African populations, hence if they appear so in genetic studies, genetic studies are likely wrong."

So you are dropping your Out of America theory and opting for the Multiregional model? I'm surprised!

German Dziebel said...

"So you are dropping your Out of America theory and opting for the Multiregional model? I'm surprised!"

The advantage of blogging is that it allows for very quick exchange of ideas. It's disadvantage is that misunderstandings may be just as quick. I can't drop the out of America idea because, IMO, it fits best the interdisciplinary data at hand and provides plenty of room for science to evolve through self-reflection. However, I'm mindful of your and others' sensibilities on this blog, and try to diffuse my critique of the current version(s) of the out of Africa model into inviting other solutions as well, rather than prescribing only one single alternative.

DocG said...

If you still adhere to your Out of America model, German, then you must realize that you too have a problem. If "populations outside of Africa are not a subset of African populations" as you claim, then where is the evidence of the opposite situation: that African populations are a subset of populations outside of Africa? If neither group can be regarded as a subset of the other, then we are left with the multiregional model as our only alternative.

German Dziebel said...

"If you still adhere to your Out of America model, German, then you must realize that you too have a problem. If "populations outside of Africa are not a subset of African populations" as you claim, then where is the evidence of the opposite situation: that African populations are a subset of populations outside of Africa? If neither group can be regarded as a subset of the other, then we are left with the multiregional model as our only alternative."

The details still need to be worked out but what's clear to me is that, according to virtually ALL genetic systems (I have my feet firmly on the ground here), Amerindians and Australasian populations harbor the greatest inter-group diversities on the worldwide scale, with those progressively declining as one approaches Africa. In Africa, predictably, hunter-gatherers have higher inter-group diversities than agriculturalists or pastoralists. This means that African inter-group diversities are effectively a subset of non-African inter-group diversities. Research also shows that most of human species-wide inter-group divergence values are concentrated among small foraging populations. (Phylogenetically, it means that it's precisely among populations with the highest inter-group diversities that we should look for the earliest lineage splits.)

The radical divergence of some musical styles from one another, like P/B from breathless solo or unison iterative one-beat is a mirror image of this very early pattern of diversification observed in human populations. Late Pleistocene small foraging bands would drift away from each other and diverge dramatically. Later, as human populations colonized the continents, they began exponentially growing in size. As a result of this process the early diversified musical forms got either generalized on a continent-wide scale or replaced by newer styles. Africa doesn't have either breathless solo, or unison one-beat, while outside of Africa one finds those two plus P/B. Even this simplistic picture shows that Africa is a subset of non-African variation, at least as far as the earliest split(s) are concerned.

It's precisely in regions where we find high levels of inter-group molecular divergence that we also find 1) the co-presence of highly distinct musical traditions, sometimes within one island; 2) the distribution of a single style over wide geographic distances (often gapped, as newer forms replaced the old ones).

DocG said...

German: "The details still need to be worked out but what's clear to me is that, according to virtually ALL genetic systems (I have my feet firmly on the ground here), Amerindians and Australasian populations harbor the greatest inter-group diversities on the worldwide scale, with those progressively declining as one approaches Africa."

"Australasia" is too broad a category. And the devil is in the details for sure. Inter-group diversity is great in New Guinea and Island Melanesia, but not Australia, as far as I can see. Not linguistically, not musically and to my knowledge, and correct me if I'm wrong, not genetically either.

Moreover, while inter-group diversity may be great among South American Indians, in North America (north of Mexico) there would seem to be far less diversity and again the degree of musical diversity is also minimal, when compared with just about any other continent on Earth. So it's hard to see how inter-group diversity can support your argument.

As for the rest, as I've said before, the complete difference in our views of so many aspects of the problem make it impossible for me to respond meaningfully to your argument, based as it is on so many untested assumptions. Once again, I urge you to put your thoughts on this matter together, either in an extended essay or blog or book. You have interesting ideas, but I can't respond adequately to them here and this is not the place to evaluate them in any case.

German Dziebel said...

"Once again, I urge you to put your thoughts on this matter together, either in an extended essay or blog or book. You have interesting ideas, but I can't respond adequately to them here and this is not the place to evaluate them in any case."

I simply tried to answer your question, Victor.

""Australasia" is too broad a category. And the devil is in the details for sure. Inter-group diversity is great in New Guinea and Island Melanesia, but not Australia, as far as I can see. Not linguistically, not musically and to my knowledge, and correct me if I'm wrong, not genetically either.

Moreover, while inter-group diversity may be great among South American Indians, in North America (north of Mexico) there would seem to be far less diversity and again the degree of musical diversity is also minimal, when compared with just about any other continent on Earth. So it's hard to see how inter-group diversity can support your argument."

As far as Australian diversity is concerned, the lack of diversity in Australia isn't a matter of history but of geology. Had Sahul been one landmass, we would have had 2 large language families (Trans-New-Guinean and Pama Nyungan) and a myriad of small families and language isolates. A very normal situation for other areas with high diversity. In South America, for instance, we have Cariban and Arawakan, in Central/North America Oto-Manguean and Uto-Aztecan. It's because the Sahul broke apart 25K years ago that Australia looks "homogeneous but ancient."

Virtually all genetic systems attest to America writ large (North and South as aggregate) as having the largest inter-group diversity values. (BTW, this is not an assumption but a recurrent observation, the question is of course how to interpret it.) I don't know if North America has been assessed independently. And, no, I don't have separate data for Australia.

And, yes, North America is less diverse linguistically and musically than South America. But even there we have polyphonic California and the Northwest Coast vs. unison iterative one-beat elsewhere, plus drone polyphony is the Woodlands, as well as a large number of instruments in the Southwest. As North America was heavily impacted by the Ice Age and was half covered by permafrost until about 10K years ago, lower levels of diversity there are to be expected.

Amazonia is the most diverse linguistically in the world because it has the absolutely largest number of language isolates, which is the ground-zero of diversity estimates.

DocG said...

German: "I simply tried to answer your question, Victor."

You're right. I apologize.

As for the rest, I don't share your view that relative linguistic diversity is a meaningful measure of the relative antiquity of different culture areas. Not that a high level of diversity doesn't imply great age, since it certainly would take time for a great many different languages and language families to develop in any one region. But the opposite isn't true -- lack of linguistic diversity can be due to many different factors -- especially since archaic diversity can be wiped out whenever some powerful, aggressive or influential culture expands.

Moreover, unlike genetic mutations, linguistic "mutations" don't necessarily accumulate. While S. America is certainly linguistically diverse, which strongly suggests that this region has been populated for tens of thousands of years, New Guinea is at least as diverse, if not moreso -- and Australia, which is probably at least as "old" as New Guinea, is dominated by only one single language family. So I don't see how some formula can be applied based on language, to determine where modern humans originated. The most promising clue, as I see it, is not linguistic diversity, but the distribution of tone languages worldwide. But even on that score, there is much to be learned before firm conclusions can be drawn.

German Dziebel said...

""While S. America is certainly linguistically diverse, which strongly suggests that this region has been populated for tens of thousands of years, New Guinea is at least as diverse, if not moreso -- and Australia, which is probably at least as "old" as New Guinea, is dominated by only one single language family."

I explained the Australian situation earlier as caused by geological/oceanographic factors (the "break-down" of the Sahul), not by linguistic history. It's further noteworthy that Australian languages of the Pama-Nyungan family are extremely diverse lexically.

"So I don't see how some formula can be applied based on language, to determine where modern humans originated."

This formula is the most basic principle of historical linguistics: languages diverge through time and isolation and form tree-like structures. There're competing models (wave theory, etc.), which have some merit, no doubt, but the basic principle remains valid. Without this principle, there's no linguistic classification. The trick is that linguists agree that, while all languages are somehow related, exact groupings are unclear. Hence, we have to stay with relative diversity measures for now, but they are very telling, IMO. It's really not just about America vs. the rest, but about America, Asia and Australasia vs. Europe and Africa. The contrast is unmistakable and can't be dismissed.

Areas with high levels of linguistic diversity also correlate with specific grammatical and morphological features (e.g., high morphological complexity found in small families and in isolates, head-marking syntactic structures, etc.) that are underrepresented in areas with low linguistic diversity (Africa and Europe). So, it's a systemic phenomenon, not just arithmetics. The distribution of tone languages is just the tip of typological iceberg that WALS-type of studies have unearthed over the past 20 years.

Finally, there's non-coincidental relationship between areas with high language diversity and high levels of inter-group molecular diversity attested in mtDNA, Y-DNA, classical markers, autosomes. It means in ALL genetic systems studied to date. There's a corresponding correlation between areas with low genetic diversity and high intra-group molecular diversity (compare the number of mtDNA haplotypes is higher in Africa and Europe vs. Asia and America).

There's a very logical connection between the levels of linguistic and molecular diversity: small populations drift apart and develop high levels of morphological complexity and linguistic diversity. If original human demes were small (and this is a cornerstone of the single origin replacement model of human origins), then it very well may be that the genetic structure associated with small populations persevered in America, parts of Asia and Australasia, while Europe and Africa are the colonized areas.

I'm more and more convinced that the reason people don't take levels of linguistic diversity seriously is because of a long-term cultural bias that only certain "hard" disciplines are capable of answering questions about human origins. The hard disciplines, meanwhile, have grown softer over the past 20 years: it's not lithics and skulls that define our thinking about human dispersals but very fine biological matter called genes. Languages could be our next stop."