Chapter Fourteen: Mysteries of Sahul

At the time of the Out-of-Africa migrations, water levels in the oceans were much lower than they are today, and as a result many of the islands of Island Southeast Asia were linked with the Malaysian mainland to form a single peninsula, called Sunda; and Australia and New Guinea were also linked to form a single continent, Sahul:

Figure 14.1 Sunda and Sahul

As you can see from the map, the low water levels meant that island hopping from Sunda to Sahul would not have been too much of a challenge -- especially since, as is now suspected, the Out of Africa migrants had already been doing much of their traveling by boat. Since some of the earliest archaeological evidence of modern human habitation comes from Australia, and since some of the arguably “oldest” populations (based on both their genetic and cultural makeup) now live in New Guinea and Australia, it stands to reason that Sahul must have been part of the Out-of-Africa migration.

But there is a problem. If Sahul were populated by Out of Africa migrants when both New Guinea and Australia were joined into a single landmass, and both regions had remained relatively isolated from then to now, as appears to be the case, we would expect the populations now living in both places to be quite similar, both morphologically and culturally. And we would assume they'd be closely related genetically as well. This, however, is not the case.

Melanesia overall, including both New Guinea and the closely related group of islands to its east, known as “Island Melanesia,” is highly differentiated culturally, whereas Australia is much more homogeneous, with almost all aborigines having distinctively “Australoid” features and sharing many traditions in common. Possibly because of the prolonged isolation of each group from its neighbors, due partly to geography, partly to endemic warfare, there are far more different languages and language families in New Guinea than anywhere else on Earth, while Australia is dominated by a single language family, called Pama-Nyungan, with all the others crowded into a relatively small area in the north, the region closest to New Guinea. The unusual distribution pattern for language families in Australia is visible in the following map (from the Wikipedia article, Indigenous Australian Languages).

Figure 14.2 Native Australian Language Families

The huge yellow region is where Pama-Nyungan languages are spoken, while the much smaller, multi-colored region to the north contains just about every other language family on the continent.

The African Signature

The musical picture for New Guinea and Island Melanesia is complex, with several different vocal styles and many different types of instruments. In several cases, we find P/B-related vocal styles, and also instances of instrumental hocket, especially with wind ensembles of pipes, panpipes, trumpets and flutes (see Chapter Nine, Audio Examples 17-22). In other cases we hear unison singing, and in still others, relatively simple part singing similar to that of Western Polynesia. The picture for Australian aboriginal music, on the other hand, is completely different, exhibiting a remarkably high degree of homogeneity and lacking any trace of an African signature.

Among the most compelling instances of the “African Signature” in New Guinea is Audio Example 17, already presented in Chapter Nine -- P/B style yodeled/interlock, as recorded by Steven Feld in the Southern Highlands of New Guinea: Audio Example 17:Bosavi Yodeling (from Bosavi: Rainforest Music of Papua New Guinea).

A somewhat different type of yodeled interlock can be heard among the Abau people of the Upper Sepik River highlands: Audio Example 53: Healing Song (from Songs & Dances from Papua New Guinea, track 5, recorded by John Thornley).

Here is an example of relatively straightforward “shouted hocket,” with yodeling, from the Huli people of the Southern Highlands: Audio Example 54:  Huli Yodeling (from Emap FM – Music from Oceania ).

A very similar type of shouted hocket can be heard in this recording of another highland group, the Dani: Audio Example 55: Dani (from Emap FM – Music from Oceania).

Among the Aka Pygmies of Africa a very similar type of hocketed interchange, called “esime,” functions as an interlude between more complex songs (Kisliuk 1998:41): Audio Example 56: Aka esime (from Musical Anthology of the Aka Pygmies, recorded by Simha Arom). Relatively simple hocketed performances of this type are classified as “haplogroup” A1 in the Phylogenetic Tree provided in Appendix B.

Highland vs. Lowland

There are two large language families in New Guinea and Island Melanesia: Austronesian and "Papuan." The former is generally regarded as much more recent than the latter, as it is associated with groups thought to have originated somewhere in southern China or Southeast Asia that expanded during the last 4,000 years or so, first to Melanesia and then to Polynesia. Most of these newer populations settled along the northern and eastern coast of New Guinea, and on many other Melanesian islands. "Papuan" is the name given to the languages of those who were presumably already living in Melanesia when the Austronesians arrived. The so-called "Papuan" languages are actually a large group of unrelated language families -- along with several languages regarded as unaffiliated "isolates" -- spoken by people living, for the most part, in the interior highlands.

Geneticists Alan J. Redd and Mark Stoneking found two mitochondrial DNA clusters among Papua New Guinea highlanders with “coalescent time estimates of ~80,000 and 122,000 years ago, suggesting ancient isolation and genetic drift.” There are indications that “84% of the sample of PNG highlander mtDNA belong to these two clusters” ("Peopling of the Sahul: mtDNA Variation in Aboriginal Australian and Papua New Guinean Populations," American Journal of Human Genetics, 65, 1999, p. 808). This is only one of several such assessments, ethnological, linguistic and genetic, that associate the ancestry of the New Guinea highlanders with the original "Out of Africa" lineage, while most Austronesian speakers of the coastal and lowland areas are considered relatively recent arrivals from the north.

To determine whether the highland-lowland dichotomy could predict the African Signature, a Cantometric search for one of the most distinctive features of P/B, vocal interlock,  was conducted for the entirety of the New Guinea sample, with the following results:

Culture Name	Language Family	Location	#
BALIEM	Papuan	Highlands of Irian Jaya -- Baliem Valley	1
BIAMI	Papuan	Southern Highlands	2
BISORIO	Papuan	Highlands -- East Sepik Province	1
DANI	Papuan	Highlands of Irian Jaya -- Baliem Valley	5
EIPO	Papuan	Highlands of Irian Jaya -- Baliem Valley	10
HAMTAI	Papuan	Highlands -- Gulf Province	1
HULI	Papuan	Southern Highlands	2
KOVAI	Papuan	Umboi Isl., Morobe Province, Northeast Coast	1
MONI	Papuan	Highlands	1
OK	Papuan	Highlands of West Papua -- near Irian Jaya	3
YALI	Papuan	Highlands of Irian Jaya -- Baliem Valley	4

Table 14.1 Vocal Interlock -- New Guinea

Of the 11 groups above that “tested positive” for interlock, all are Papuan speakers and all but one are highlanders. Since 23 groups in the entire sample are identified as highland and 22 as coastal or lowland (with an additional 13 I have not yet been able to locate accurately), there does appear to be a strong correlation between the musical evidence and the genetic/linguistic evidence, distinguishing an indigenous highland population, with roots in the early Out-of-Africa migration, from a much more recently arrived coastal/lowland population, associated with Austronesian languages and culture.

The picture for Island Melanesia, is not so simple, however, as the “lowland/highland” dichotomy is not always clear, and many native “Papuan” groups now speak Austronesian languages. Nevertheless, Island Melanesia also contains many instances of the African Signature, as evidenced by Audio Examples 18 and 22 (see Chapter Nine), and the following, truly remarkable, recording of vocal interlock, accompanied by pipes, from the island of Buka, north of Bougainville, in the Solomon Islands:  Audio Example 57: Buka Singers with Pipes (from Emap FM – Music from Oceania).

Australian Homogeneity

Melanesian music is by no means limited to P/B-related styles, and is in fact one of the more diversified musical areas on Earth. In contrast, Australia is among the most musically homogeneous regions in the world. Australian aboriginal singing is characterized by tense, nasal vocal style, either solo or unison, the frequent iteration of single notes, with sticks or boomerangs beaten together to produce relatively simple one-beat rhythms or simple variants of the one-beat pattern:

Audio Example 58: Western Australia (from Emap FM – Music from Oceania).

From the Yuendumu Community, Central Australia: Audio Example 59: Traditional Song (from Traditional Aboriginal Music:Sounds from the Bush, Arc Music, track 20).

From Northeast Australia: Audio Example 60: CapeYork (from Emap FM – Music from Oceania).

More or less the same general performance style pervades the entire continent, though occasionally one hears something more complex, with traces of polyphony. The only important musical instrument is the Didgeridoo, which was traditionally found only in the west and may be a relatively recent innovation. Drums, plentiful in Melanesia, are all but absent in Australia. The above descriptions are deceptive, however, as Australian singing and Didgeridoo playing are among the most sophisticated musical art forms in the entire world. Many of the texts that go with these songs are also remarkable examples of highly sophisticated, allusive and complex poetry.

Considering the importance of Australia as the bearer of the earliest archaeological evidence of modern humans outside of Africa, evidence which so strongly supports the Out-of-Africa model, the absence of any trace of the “African signature” in any of its indigenous music is difficult to explain. If the Out of Africa migrants were singing and playing in some version of P/B style, then what could have happened when they got to Australia that made them lose their musical traditions and develop such different ones? And since, as we’ve seen, we do in fact find many instances of the African signature in New Guinea and Island Melanesia, its absence in Australia is especially difficult to understand. Coupled with all the other evidence for major discrepancies, morphological, genetic, linguistic, etc., we are faced with an extremely perplexing mystery.

I would like to propose an explanation that might resolve all or most of the contradictions, which again, like so much else in this book, should be seen as exploratory, speculative and provisional. Let’s begin with some provocative clues:

A Divided History

It has long been thought that the Tasmanians, tragically exterminated during the initial stages of the colonial era, might have been direct descendants of an initial wave of immigration that preceded the entry of australoid peoples. This notion was revived by anthropologist Joseph Birdsell and his associate Norman Tindale, who promoted what they called a "tri-hybrid" theory of Australian history involving three successive waves of migration. According to Birdsell, the first immigrants, the "Barrineans," were Negritos, and it is their remains we see in the "gracile" Mungo Lake skeletons, the earliest (ca 45,000 ya) fossil remains in Australia. The next wave were what he called the "Murrayians," with "caucasoid" features resembling the Ainu. And the last wave were the "Carpentarians," the now dominant "australoids," with affinities to the australoids of India. Birdsell's research confirmed the almost mythic existence of Pygmies in Australia, which made it logical for him to conclude that they were most likely descended from the original "Barrineans."

Figure 14.3 Joseph Birdsell with adult Australian Pygmy

For Birdsell, the early Tasmanians, who may have had a similar morphology, judging from various remains, had also been Negritos, and therefore must also have been descended from the earliest immigrants. Birdsell's "tri-hybrid" theory has been disputed and is no longer a part of mainstream anthropology, possibly due to "political correctness" concerns, as it flew in the face of a popular movement promoting the idea that all aboriginals were descended from the original inhabitants of the continent.

Male vs. Female

An ongoing theme in the genetic story from this part of the world is a surprising male-female distinction, and Australia is no exception. The findings reported in a paper of 2003, by Max Ingman and Ulf Gyllensten, Mitochondrial Genome Variation and Evolutionary History of Australian and New Guinean Aborigines, reveal

a striking difference between the genetic history of females and the reported history of males in the Australian Aboriginal population. . . Kayser et al. (2001) proposed that the high frequency of a unique [Y chromosome] haplotype in Australia is the result of a population expansion that started from a few hundred individuals. In this case, the predominance of a unique Y-chromosome haplotype in Australia would be the result of a founder effect. However, there does not appear to be a corresponding loss of genetic diversity resulting from a bottleneck seen among mitochondrial lineages (p. 1604 -- my emphasis).

In other words, the major discrepancy between Australian Y and mtDNA diversity suggests a bottleneck in the former, yet none in the latter, which seems puzzling -- unless males and females have a very different history on this continent. (Remember, the Y chromosome is found only in males and can represent only male lineages.)

Our mitochondrial [i.e., female line] data imply that some lineages from the populations of Australia and New Guinea have shared a common history since the initial colonization of Sahul. . . . [However,] [t]he lack of a common Y-chromosome haplotype found both in Australia and in the New Guinea highlands (or in any other Melanesian population) argues against the concept that the New Guinean and Australian populations are derived from the same migration event (Kayser et al. 2001). However, the Australia-specific Y chromosome haplotype could have arisen after the colonization of Sahul and therefore is absent in other populations. (my emphasis)

Passage from India?

For many years, anthropologists have speculated regarding what appear to be striking physical similarities between Australian aboriginals and the Vedda of southern India and Sri Lanka, many of whom have a distinctly “australoid” physiognomy. While such comparisons have often been dismissed, recent findings suggest that they could have a genetic basis after all – but, interestingly enough, only on the male line.

In an article titled Gene Flow from the Indian Subcontinent to Australia: Evidence from the Y Chromosome, by Alan Redd et al., 2002, the authors present “strong evidence for an influx of Y chromosomes from the Indian subcontinent to Australia . . .”:

In sum, we found that 50% of the Y chromosomes sampled from aboriginal Australians [haplogroup C*] share common ancestry with a set of Y chromosomes that represent less than 2% of the sampled Indian subcontinent paternal gene pool. . . . (p. 676)

While only 2% of the male gene pool for India might seem insignificant, it's important to remember that the C* haplogroup is found only among certain tribal peoples in south India and Sri Lanka (where we find many australoid types today). It would be very strange indeed if the figure were much higher than 2%, since Australian aborigines bear little physical or cultural resemblance to East Indians generally. However, the figure shoots up to 50% in Australia, a remarkably strong representation. While these results are indeed suggestive, the connection may be relatively recent. According to their estimates, C* dates only to the mid-Holocene, roughly 8,000 years ago, which places this particular migration well past the Out of Africa exodus.

It would be much easier to argue for an Indian-Australian cultural connection if there were any distinctive musical similarities between Tribal India and Aboriginal Australia, but that does not seem to be the case. However, I recently came across a remarkable Youtube clip of dancing among the Chenchu hunter-gatherers of South India that strongly resembles certain types of Australian Aboriginal dance: Video Example Seven -- Children of the Forest:

(http://www.youtube.com/watch?v=sQSDpRX8bZk&feature=related )

While the very opening contains some interesting moves, the most remarkable similarities with Australia can be found at the 1:30 and 2:30 marks.

Compare the above with the Australian Aboriginal dancing seen in portions of Video Example Eight: Dance During Initiation Ceremony (skip to roughly 30 seconds in):

(http://www.youtube.com/watch?v=V6hKzvdtIcg&feature=related )

A Hypothetical Reconstruction

On the basis of the analysis presented above, along with a considerable amount of additional evidence not presented here, but available via my blog, Music 000001 (see especially Posts 297-310), I’ve been able to put together an admittedly very speculative hypothesis, roughly consistent with Birdsell’s “trihybrid” theory, which could account for all or most of the odd discrepancies, morphological, genetic, linguistic and musical, between New Guinea and Island Melanesia on the one hand, and the Australian Aboriginals on the other. Here’s what I think could have happened:

1. Early entry into Sahul by island hopping from Sunda, in the immediate wake of the Out of Africa migration. The earliest immigrants would have been a small band of HMP (Hypothetical Migrant Population) descendants, who would have retained an African morphology and an African culture and value system, based largely on HMC. These early immigrants would not have been seriously affected by the population bottleneck(s) I've associated with the Toba eruption (or some equally devastating event), as they would presumably have been living far enough to the east of India at the time to be only minimally affected, and therefore would have retained their original African characteristics to at least some significant degree.

2. It seems reasonable to think in terms of a fairly rapid expansion along the coast of the entirety of Sahul, followed by a very long period of stability, in which these relatively peaceful and cooperative hunter-gatherer descendants of HBP and HMP could have lived together in harmony for literally tens of thousands of years.

3. On the basis of the genetic evidence for both a highly contrastive history for males and females in Australia and a close Y chromosome association between Indian and Australian australoid populations, we can posit a second migration, occurring many thousands of years later, of mostly male australoid hunter-gatherers, whose ancestry would have stemmed from the South Asian centered bottleneck posited in Chapter Ten. According to Redd et al (2002),

The divergence times reported here correspond with a series of changes in the Australian anthropological record between 5,000 years ago and 3,000 years ago, including the introduction of the dingo; the spread of the Australian Small Tool tradition; the appearance of plant-processing technologies, especially complex detoxification of cycads; and the expansion of the Pama-Nyungan language over seven-eighths of Australia. Although there is no consensus among anthropologists, the former three changes may have links to India, perhaps the most relevant of which is the introduction of the dingo, whose ocean transit was almost certainly on board a boat. In addition, Dixon noted some similarities between Dravidian languages of southern India and Pama-Nyungan languages of Australia.

4. We can now extrapolate backward to speculate on how the arrival of these strangers could have led to the conditions we now see. And the first thing to consider is the fact that, in order to produce the largely australoid population we see in Australia today, the more recent immigrants would have to have mated with the “native” women, probably forcibly at first, and at the same time largely either killed, displaced or enslaved the native men, wherever they encountered them. This would explain the “different histories” of males and females we see in the genetic evidence. The mtDNA (female line) picture would not reflect the presence of men from a completely different population, but the Y chromosome evidence would -- and that does seem to be the case. Over time, as the more aggressive and belligerent newcomers expanded throughout the continent, the original inhabitants would have done what so many relatively non-aggressive, non-competitive, non-violent peoples have done throughout history -- retired to easily defended or undesirable refuge areas. This would explain the special status of Tasmania, which could have served as a last stand for some of the natives as they retreated southeast to the point farthest away from the most likely point where the newcomers would have arrived, the northwest. And since Tasmania was originally a kind of peninsula with a fairly narrow land bridge, that might have worked for them as a last line of defense until the sea level rose and they became completely isolated.

5. Since Australia is relatively flat and easily traversed, the indigenous males would not have had much chance of survival, but could have been hunted down and slaughtered or enslaved, and their women appropriated as wives. Northeast Queensland contains a tropical forest, which was until recently, according to Birdsell's research, the home of a few small groups of Pygmies, who may have originally retreated to this area as a refuge, possibly many thousands of years ago. But the most obvious refuge area would have been to the north, in what is now New Guinea, and it is the highlands of New Guinea that we can posit as the most likely refuge area for the majority of the retreating natives. If the newcomers had arrived while New Guinea was still attached to Australia, the refugees could have made their way north by land, but if the sea had already separated the two regions, they could have retreated in boats or rafts, at least while the distance was not too great. The australoid invaders would have followed them, and at that time taken over the New Guinea coast, while the natives retreated into the highlands.

6. The next important event in the history of this region is the advent of the so-called “Austronesians,” who are thought to have migrated to various points in New Guinea and Island Melanesia anywhere from 6,000 to roughly 4,000 years ago. The newly arrived Austronesians appear to have displaced most of the australoids along the coastal regions to the north and east. Their only recourse would have been a retreat into the highlands, which would therefore have come to harbor a mixed population, partly of “negrito” and partly of australoid origin. Since these groups would have formerly been bitter enemies, it's not difficult to see how the endemic warfare we now see in the New Guinea highlands could have originated at this point, although many of these populations seem ultimately to have merged, both physically and culturally.

If my scenario is correct, then the current situation in the former Sahul could be described in the following terms:

1. in Australia, the descendants of males from the second wave of migration and females from the first, with australoid morphology, speaking, for the most part, a Pama-Nyungan language; 2. in New Guinea, descendants of the original “negrito” settlers, with a degree of australoid intermixture, now surviving mostly in the highlands, but also along portions of the coast, living as foragers and part-time horticulturalists, speaking a wide variety of very different “Papuan” languages, and retaining at least some of their original African traditions, including, in some cases, P/B-related musical styles; 3. in New Guinea, descendants of population 1, formerly based on the New Guinea coast, now living for the most part in the highlands as forager/horticulturalists, possibly intermixed with population 2, both biologically and culturally -- also speaking “Papuan” or in some cases Austronesian languages; 4. relatively recent Austronesian immigrants, speaking Austronesian languages, and inhabiting, for the most part, the northern and eastern coastal and lowland areas of New Guinea.