Saturday, January 2, 2010

271. The Baseline Scenarios -- 47: The Migration

The picture of a continuous migration path from Africa through South Asia is reinforced by additional cultural information gleaned from the Cambridge Encyclopedia of Hunters and Gatherers, on seven Indian tribal groups, the Andamanese Islanders (Jarawa and Onge), the Birhor, Chenchu, Nayaka, Paliyan, Hill Pandaram and the Veddahs of Sri Lanka. The picture for most of these groups is roughly similar, with gathering and hunting (often with bows and arrows) supplemented with some form of swidden agriculture in all but the Andaman groups, who lack farming traditions. The term "egalitarian" is used in almost all instances to characterize their political and economic situation and all groups are described as generally non-violent and informally communal, with no permanent leaders. The Andaman Islanders are, of course, the best known and most remarkably "negrito" group, and indeed stand out both for their strong resemblance to African pygmies and their unique isolation, possibly dating all the way back to the GM itself.

Though it might seem a minor issue, the importance of honey gathering for literally every group should not be taken for granted. It is a pervasive theme in the culture of a great many hunting and gathering peoples, and the considerable degree of skill required strongly suggests a tradition rooted in a common ancestry, unlikely to have developed through independent invention or convergent evolution.

On the other hand, the importance of swidden agriculture (aka horticulture) among so many South Asiatic "hunter-gatherers" is more difficult to assess, especially as this highly idiosyncratic method of farming, so different from that of more "advanced" societies with whom they might have had contact, is so widely found among so many indigenous peoples worldwide. Since any form of agriculture seems so unlikely for HBC, and since the estimated date of the African exodus is so much earlier than the "official" estimates for the origin of agriculture (ca 10,000 ya), it seems highly unlikely that such a practice could have been spread to Asia via HMC. On the other hand, if it wasn't part of HMC, then by what means could it have spread among so many isolated peoples in so many different parts of the world? Which suggests the possibilty that some type of farming might have arisen much earlier than is usually thought. All I'll say on this matter for now.

Given the presence of "negrito" morphology in South Asia, as discussed in the previous post, and the many other hunter-gatherer groups in the region who, as we've seen, share so many traditions and core values with African hunter-gatherers, one would assume that a strong African connection would be apparent in the genetic evidence as well. The genetic picture is not that simple, however, and many questions remain.

In a paper dating from 2004, Genetic structure and affinities among tribal populations of southern India, H. Vishwanathan et al. report on their effort "to test whether the phenotypic similarities of some south Indian tribal groups to Africans represent a signature of close relationship to Africans or are due to convergence." Their research, based on a study of 24 autosomal (i.e., nuclear, rather than mtDNA or Y chromosome) markers,
showed that [tribal] Indian populations are closely related to one another, regardless of phenotypic characteristics, and do not show particular [genetic] affinities to Africans. We conclude that the phenotypic similarities of some Indian groups to Africans do not reflect a close relationship between these groups, but are better explained by convergence.
Since both "negrito" features and the many cultural affinities so strongly suggest an African connection among so many native peoples of tribal India, this result is perplexing. Coupled with some of the other findings I've already presented, which call certain features of the South Asiatic genetic picture into question, the genetic evidence from this region, with respect to the mainstream Out of Africa model, must, at the moment, be regarded as inconclusive.

But perhaps not so inconsistent with the cultural evidence after all, because, as I've already pointed out, there are inconsistencies in the ethnographic picture as well, particularly in the realm most relevant to my own research. While so many characteristics of HBC and HMC can be found among the tribal peoples of India, there is, as I've been emphasizing for some time now, a surprising gap in the musical picture. What I've been calling the "African Signature," i.e., the presence of musical elements associated with Pygmy/Bushmen style (P/B), are almost completely absent from both South Asia and the Near and Middle East, yet reappear in Southeast Asia and many other points to the east and south.

The situation is clearly summarized in the following table, drawn from the Cantometric database, showing the distribution of sung contrapuntal polyphony in the samples from Asia, Island Southeast Asia, Melanesia and Australia(Click on the image to get a clearer view.):

Note the dramatic differences in the representation of this very distinctive musical feature, with so many instances among indigenous groups in Southeast Asia, South China, Indonesia, Island Melanesia and New Guinea, and yet only a single instance among the entire 165 song sample for Tribal India, one other for the entirety of Village India, and no instances whatsoever for Northeast Asia (including Han China), the Near East, Middle East and aboriginal Australia.

I must add that the Cantometric evidence is just the tip of the iceberg, since there are a great many instances of similarly organized instrumental ensembles with a very similar distribution throughout Southeast Asia, Indonesia, etc., yet all but totally absent from South Asia, the Near and Middle East, Central and Northeast Asia (with the exception of Japan and ancient China).

(to be continued . . . )


German Dziebel said...

Hill Saora (Sora), a Munda population that has counterpoint in your table, also has a little bit of level tone in their language (see Donegan, Rhythm and the Synthetic Drift of Munda, 2004). Another Munda language that has tones (level tones again) is Korku (see Zide, Korku Low Tone and the proto-Korku-Kherwarian Vowel System, 1966). In this regard, Sora and Korku are similar/different from Mon-Khmer, who have highly developed contour tones. As a side note, Munda and Mon-Khmer branches of Austroasiatic are highly divergent in all domains of language.

Also, on a different note, I just came back from the American Museum of Natural History in NYC and they mention in one of the captions a traditional African "talking drum" that "imitates the tonal pattern of a tonal language." I don't know which one they meant but I thought this could lead you into an interesting exploration into the functional connections between linguistic tones and polyphonic music.

DocG said...

Thanks very much, German, for the heads-up on the Hill Saora. To be honest, I didn't realize they were an exception to the musical rule in India until I did that query, earlier today. It looks as though they are also an exception linguistically, one of the few groups in India with tone language (?) And the other info on related Mon-Khmer languages is also interesting.

And yes, I know about talking drums. They can "talk" specifically because the languages they "talk" in are tonal. So it's not like Morse code, they actually "speak" by replicating the tonal contour and rhythm of the statement the drummer wants to make. Not always understandable to all native speakers, but other drummers can follow it and there are many in-jokes as a result.

African vocal melodies are also very much affected by linguistic tone and this has been pretty thoroughly studied. P/B vocalizing is mostly with nonsense vocables, by the way, so this music has much more freedom than other African vocal music where text is more important.

And yes, I've been thinking a lot about the relation between tone language and music, though not polyphonic music specifically. But I do think that if music had an influence on the early development of language, which I believe it did, then it makes sense for the earliest languages to have been tone languages.

German Dziebel said...

"But I do think that if music had an influence on the early development of language, which I believe it did..."

You may be interested in looking at Piraha in South America, which Everett describes as a language that "can be whistled, hummed, or encoded in music. In fact, Keren Everett believes that current research on the language misses much of its meaning by paying little attention to the musical aspect of it (prosody). Consonants and vowels may be omitted altogether and the meaning conveyed solely through variations in pitch, stress, and rhythm. She says that mothers teach their children the language through constantly singing the same musical patterns." (wiki)

See also

See, I can easily flip your "tree" upside down and still retain all the most interesting aspects of your thinking. (smiley face.)

German Dziebel said...

After further research, it appears that in South Asia, apart from a few IE languages likely influenced by Burushaski and, of course, the Brahmaputran branch of Sino-Tibetan, Munda is the only language family that has a couple of languages with tones. Dravidian languages lack them altogether. Unlike Munda and Brahmaputran, Dravidian languages have no continuation in Southeast Asia, the area in which tones are common. If Nostraticists are right, the wider connections of Dravidian are in northern Eurasia, the area with no tones. Munda, on the other hand, is a South Asian branch of Austroasiatic, hence its tones may have the same origin as Mon-Khmer.
It's also noteworthy that all proposed cultural connections between South Asia and Australia involved Dravidians, rather than Munda.

manju said...

It's also noteworthy that all proposed cultural connections between South Asia and Australia involved Dravidians, rather than Munda.

Mundas' main Y-haplogroup is O2a which they share with their other Austro-Asiatic cousins. However, their matrilineage is overwhelmingly South Asian. Dravidians have mostly South Asia specific Y-Haplogroups like H, L1, R2. Though Middle Eastern J2b is also observed substantially.

The mtDNA(M42) that connects South Asians and Australians is observed among Mundas and Dravdians tribes in Eastern regions where they are in close contact with Mundas.

German Dziebel said...

Thanks Manju. I was trying to find a pattern that would align linguistic and music "gaps" and lack thereof in South Asia and Australia, but apparently it's not supported by genetics. E42 must be the result of gene flow from Munda into Dravidians.