Friday, June 11, 2010

Humans in the Philippines 67,000 years ago

So say Mijares and colleagues (2010), reporting the discovery of a small human third metatarsal ResearchBlogging.orgfrom Callao Cave in the northern Philippines. The paper present a brief overview of fieldwork conducted at Callao since 2003 that exposed Pleistocene deposits at the site. The age of the layer in which the metatarsal was recovered was obtained through Electron Spin Resonance (ESR) and Uranium Series (U-Series) on two cervid teeth, one of which yielded an age of 66 +11/-9 kya.

From Mijares et al. (2010: 7, Fig. 8, Copyright © 2010 Elsevier Ltd). The Callao specimen (A) is compared to H. sapiens (B) and H. habilis (C), and three non-human primates to the right.

The really interesting part of the paper comes when the author discuss the taxonomic attribution of the metatarsal. They compare it to various extant primates and show that it is a convincing Homo bone, aligning itself most closely with small-bodied populations, such as H. habilis or contemporary Philippine Negritos, the latter of which stand out as likely potential analogs of the hominin to whom the metatarsal belonged. That said, "the dimensions of the base of the bone and the section of the shaft are smaller, indicating peculiar proportions for the Callao metatarsal. At the mid-shaft, the shaft appears to be considerably smaller in the dorso-plantar direction than in the Negrito comparative sample. As shown by the reduced dimensions obtained for the dorso-plantar height and medio-lateral breadth of the proximal facet for the lateral cuneiform, the base is very small. It is the smallest of our sample, confirming the particular shape and proportions of the bone as seen from lateral and superior views" (Mijares et al. 2010: 8).

The authors emphasize that the peculiarities of the Callao metatarsal are unique in the panorama of known foot bones attributed to various Pleistocene Homo. Provocatively, they point out that the dimensions of the H. floresiensis third metatarsal from Liang Bua (LB 1) are very close to those of the Callao specimen (Mijares et al. 2010: 9). While they present this comparison as speculative, the implications of the exercise are clear: they're asking whether something like H. floresiensis could have been present at Callao ca. 67 kya, although they do cover their bases by emphasizing that the closest analog small-bodied humans known in the region today are Negritos.

What's a bit puzzling is their repeated discussion that the Philippines are east of Wallace's line. While I know there's a bit of debate over this, I've always understood the Philippines as being located west of Wallace's line, on the Asian side of things. Mijares et al.'s argument that the Philippines are "beyond Wallace's Line in Island Southeast Asia" appear to be a further manner of potentially linking the Callao specimen to those from Flores.

In any case, as the authors conclude, the Callao third metatarsal "documents the presence of a hominin species on the island of Luzon as early as 67 ka, and is testimony to a capability to colonize new territories across open sea gaps. The Philippine specimen also indicates that Flores was not the only island in Wallacea to be occupied by hominins more than 50,000 years ago" (Mijares et al. 2010: 9). Regardless of the precise taxonomic affiliation of that bone, it indicates a great time depth for human presence in that part of the Old World, and provides some thought-provoking evidence that seafaring must have been part of the hominin behavioral range by that time, something that seems to potentially have been the case in other parts of East Asia at that time.

Reference:

Mijares, A., Détroit, F., Piper, P., Grün, R., Bellwood, P., Aubert, M., Champion, G., Cuevas, N., De Leon, A., & Dizon, E. (2010). New evidence for a 67,000-year-old human presence at Callao Cave, Luzon, Philippines Journal of Human Evolution DOI: 10.1016/j.jhevol.2010.04.008



58 comments:

Maju said...

Interesting, thanks for posting this.

If it's a H. floresiensis, then no big surprise because they were already known to be in the islands of that area, right? (Initially for "human" I imagined H. sapiens, though it's not necesarily the case).

The main conclusion would be the one you suggest in the last paragraph: "seafaring must have been part of the hominin behavioral range by that time".

Remember also the case of the Cretan quartz handaxes (you blogged about that in march), which doubles the age for this phenomenon, regardless of the species of Homo involved. So I would say that yes: that some sort of boating with naval capability was part of the Homo sp. behavioral range already and very clearly so.

Martin said...

Cool, Julien! This was the first I heard of it. Little hairy people popping up all over the place.

Julien Riel-Salvatore said...

Maju -
I completely agree with you, but I'm still waiting to read the official publication of the Cretan handaxes before I make up my mind (due out in the next issue of Hesperia)...

If this find could be shown to be H. floresiensis, it'd be a huge deal, since the geographic range of the species would be expanded, which would perhaps militate against the idea that insular dwarfism was a major driver of the peculiar morphology of the 'hobbits.' I'm saying this because, if they were able to move around quite a bit, insularity may not have acted quite as strongly as has generally been argued to be the case on Flores. And if that's the case, then why are they so odd and diminutive?

Personally, I'm more convinced by a link to Negritos. But if that's the case, it is also major, in that it provides a very early date for the earliest presence of H. sapiens in the region, in addition to showing that considerable morphological variability already existed in our species almost immediately after the putative dispersal date from Africa...

Julien Riel-Salvatore said...

Martin -
indeed... though I reserve judgment on the hair for now ;)

terryt said...

"What's a bit puzzling is their repeated discussion that the Philippines are east of Wallace's line. While I know there's a bit of debate over this, I've always understood the Philippines as being located west of Wallace's line, on the Asian side of things".

Technically most of the Philippines lie east of Wallace's line. The only exception is the island of Palawan which has been connected to Borneo at times.
I had a very heated exchange with Maju over that fact. As far as I'm aware several species of monkey are the only Asian mammals to have entered the Phillipines. And no Sahul mammals reached there. So that makes the Philippines very much part of Wallacea.

"If this find could be shown to be H. floresiensis, it'd be a huge deal, since the geographic range of the species would be expanded, which would perhaps militate against the idea that insular dwarfism was a major driver of the peculiar morphology of the 'hobbits.'"

It doesn't really argue against it. They may not have been 'able to move around quite a bit'. Ini-directional migration may be the explanation, probably accidental. In which case insularity would still be a very good explanation for their dwarfism.

"Personally, I'm more convinced by a link to Negritos. But if that's the case, it is also major, in that it provides a very early date for the earliest presence of H. sapiens in the region"

Again not necessarily so. However it has even more interesting implications. Are the SE Asian Negritos at least partly descended from pre-sapiens Homo species?

Maju said...

"Are the SE Asian Negritos at least partly descended from pre-sapiens Homo species?"

Not that we know. While Filipino Negritos have not been as extensively researched as Malaysian ones or the various peoples of Sahul, there's nothing so far suggesting that they would be related to other ancient hominins, with the likely exception of the newly discovered small admixture with Neanderthals of all Eurasians equally.

Their small size may have been somewhat common in the past and is maybe related to metabolic issues (suggested for Pygmies in tentative connection with their jungle habitat, low in some key mineral), as they are not the only population worldwide to display such size range (Bushmen, other Negritos, some Mayas and, of course the extreme Pygmies).

By phenotype they make me think of some undefined type of archaic Eurasian, still with some links to African typology but also to West Eurasia and, of course, the peoples of southern India and Sahul. They are a very cool people to help imagine how early Eurasians could have been.

"I had a very heated exchange with Maju over that fact".

I think we can agree if we consider Philippines a distinct insular province on its own right. I have never seen any mention of Sahul animals of any sort arriving to Philippines, while Wallacea is defined for being a transitional zone between the Eurasian and Australian continents (and ecorregions), including marsupials such as the cuscus, as well as Australasian-specific rodents.

Philippines would seem more directly "connected" to Wallacea via Borneo-Palawan than directly.

terryt said...

"I have never seen any mention of Sahul animals of any sort arriving to Philippines"

Nor Sunda ones. Except a couple of monkeys and several deer species.

"while Wallacea is defined for being a transitional zone between the Eurasian and Australian continents (and ecorregions)"

So, because the Philippines have neither Sahul nor Sunda species they are exactly 'a transitional zone between the Eurasian and Australian continents'.

"including marsupials such as the cuscus, as well as Australasian-specific rodents".

Just a few Wallacean islands very close to New Guinea/Australia have marsupials such as the cuscus. Not the Philippines, as far as I'm aware. And the 'Australasian-specific rodents' arrived in Australia/New Guinea from Wallacea, and diversified there, rather than being originally a product of Sahul.

So, rather than considering the 'Philippines a distinct insular province on its own right' the islands fit precisely any definition you care to use for Wallacea.

"Philippines would seem more directly 'connected' to Wallacea via Borneo-Palawan than directly".

I very much suspect it's the other way round. The Philippines have provided the main connecting route between Sunda and Sahul. To my way of thinking the original Homo sapiens route to Australia/New Guinea was probably via Palawan to the Phillipines, then south to Sulawesi, Halmahera and New Guinea. The sea crossings required are less than those required from Timor.

I agree that the above is not really important. What is important is whether these ancient island-dwellers crossed Wallace's line deliberately or accidently. I still tend to plump for accidently at this stage.

German Dziebel said...

""Personally, I'm more convinced by a link to Negritos. But if that's the case, it is also major, in that it provides a very early date for the earliest presence of H. sapiens in the region"

Again not necessarily so. However it has even more interesting implications. Are the SE Asian Negritos at least partly descended from pre-sapiens Homo species?"

There probably were several iterations of short-stature populations in South and Southeast Asia. What the authors found could be a population of modern HSS which, however, is unrelated to a later wave of Austronesian-speaking Negritos.

terryt said...

"What the authors found could be a population of modern HSS which, however, is unrelated to a later wave of Austronesian-speaking Negritos".

There is no doubt at all that Negrito groups that speak Austronesian languages have adopted those languages. It's unknown what languages they spoke before Austronesian's expansion. Perhaps it was related to some New Guinea or Andaman Islands language.

German Dziebel said...

"There is no doubt at all that Negrito groups that speak Austronesian languages have adopted those languages. It's unknown what languages they spoke before Austronesian's expansion. Perhaps it was related to some New Guinea or Andaman Islands language."

What's your proof of that? I may have missed some recent genetic studies on Negritos, but from the point of view of linguistics, at least one Andaman language, namely Ongan, was showed recently to be related to Austronesian. I forwarded you the link, didn't I? I have no doubt that there were several waves of Austronesian-speaking populations occupying different niches and either maintaining a foraging lifestyle or switching to agriculture, but I do need to see some good evidence that African Pygmies or Southeast Asian Negritos lost their original languages.

terryt said...

Here is an interesting map. It shows the tarsier's distribution in the Philippines. This is the species Maju confused with the cuscus. It's placental, a primate, and presumably entered the Philippines via Borneo.

http://en.wikipedia.org/wiki/File:Philippine_Tarsier_geodistrib.png

Perhaps the tarsier has been exterminated in Luzon, but perhaps it never reached there. This would suggest the tarsier enetered via the Sulu Archipeligo. This is therefore the most likely route for another, this time widespread, mammal.

But the publication linked to here examines human remains recovered in Luzon, not Mindanao. But Mindanao and Sulawesi lie on any route between the Philippines and Flores. Unfortunately Mindanao is not the safest place for anyone at present, let alone scientists. And people in parts of Sulawesi are not remarkably contented either. So, evidence of absence etc.?

"but I do need to see some good evidence that African Pygmies or Southeast Asian Negritos lost their original languages".

I think the evidence for Pygmies having adopted the languages of their neighbours is pretty overwhelming. It's possible their neighbours have adopted Pygmy languages of course, but these languages most often belong to widespread families. Besides each separate group of Pygmies speak basically unrelated languages. The same situation applies for SE Asian Negritos. Only some speak Austronesian languages. And the various Negrito groups are almost ceertainly remant survivors of an earlier population, if they are in fact connected to each other at all.

"I have no doubt that there were several waves of Austronesian-speaking populations"

Almost certainly so. But the Austronesian languages appear to have been spread by sea, and Negritos are mostly inland inhabitants, even occupying dense forest. The main exception is the Austronesian languages spoken in inland regions of Borneo, but no Negrito groups survive on that island. I'll grant Ongan may be related to Austronesian but is that situation representative of the Andamans as a whole? I doubt it.

Maju said...

"This is the species Maju confused with the cuscus".

I did not confuse anything of the like, Terry: cuscus in Wallacea and New Guinea, tarsier in Philippines. The tarsier is a placentarian mammal and a remote cousin of us: a primate.

German Dziebel said...

"I think the evidence for Pygmies having adopted the languages of their neighbours is pretty overwhelming."

Hmm. You're confusing agricultural products with languages.

The truth is that there's very little evidence that Pygmies have adopted the languages of their neighbors. It's the wishful thinking of some pre-historians who want to believe that tribes of short-statured humans are some kind of living fossils. It's remarkably suspicious that all short-statured populations should have lost their languages without a trace. In order to ascertain language replacement, you have to unearth substratum effects in both the languages of the colonizers and the languages of the colonized, especially, when it comes to foragers vs. agriculturalists in such specialized areas as hunting and gathering terminology. Bahuchet did find some signs of a substratum in some Pygmy languages but this is still very tentative. Read "Are the African Pygmies an Ethnographic Fiction" by linguist Roger Blench. I can e-mail you this article.

As the Ongan-Austronesian connection suggests, it's not that Pygmies and Negritos lost their original language but rather some speakers of Niger-Congo, Nilo-Saharan, Austronesian and Austroasiatic languages lost their original phenotype as they moved into tropical forests. What the presence of foragers among the speakers of these languages tells us is that the formation of these language families must pre-date agriculture.

Andrew Oh-Willeke said...

At the most recent minimum sea level in the Homo Sapiens Out of Africa era, Sundaland and the Philippines (which would have been mostly a single large island then) would have been much closer than they are now, and there are probably been times since Homo Erectus Out of Africa that the two land bodies have been even closer.

The underwater gaps would have been far shorter than those that would have had to be crossed to enter New Guinea; probably similar to or narrower than the English Channel. These channels would have been shallow and protected from winds that produce high seas and transoceanic waves by mountain chains on all sides. The land on the other side of the water would have been visible at all times. The channel connecting the Sula Sea and the Celebes Sea and between the Sula Sea and the Pacific Ocean would have been comparable in many ways to the Dardanelles and Bosporus straits with the Sula Sea analogous to the Marmara Sea, although the straits would have been wider.

In short, it doesn't take very advanced boating technology to get across. A really good swimmer could do it, and a floating log or makeshift raft could get a family across on a good day at low tide.

The line that Alfred Russel Wallace actually drew in the 19th century put Sundaland and the Philippines on the same side of the line (i.e. the West side), so terryt is strictly wrong on that point as is the article. This doesn't, however, necessarily mean that Sundaland and the Philippines are in the same biogeographic region. Antonio Pigafetta's biogeographical line of 1521 which coincides with the of Alfred Wallace also recorded the biological contrasts between the Philippines and the Maluku Islands.

A 57,000-77,000 years ago date (simply using the mean date without the range isn't appropriate when the range is so wide) is consistent with the generally accepted range for Out of Africa dates, and evidence from the population of the Americas show that paleolithic coastal populations could move the kind of distances involved in a matter of one or two thousand years tops. Even Mathilda, who is a strong proponent of early Out of Africa, doesn't argue for earlier than 55,000-60,000 years ago for Australia

It is also reasonable to assume that Philippine Negritos are descendants of the oldest modern human population in the Philippines. H. floresiensis in the Philippines would be a stretch that I would expect pretty definitive evidence to prefer over Philippine Negritos. The twist at the bottom of the medial in the image also suggests that the bones are on the modern end of the continuum.

Andrew Oh-Willeke said...

@German Driebel

The population genetic evidence that African Pygmies are genetically distinct from other Africans and are more closely related genetically to the Khoisan than to their linguistic neighbors is pretty definitive and further dates divisions of pygmy populations into subgroups. Conventional mutation rate dating puts the Pygmy-Khoisan v. West African split at 70,000 years ago, more or less. There is also evidence of substrate language in African pygmy speakers of Niger-Congo languages which they speak now (mostly phonetic). And, the African pygmies don't speak just any Niger-Congo languages, they speak Bantu family Niger-Congo languages. This is a language family whose time and place of pre-expansion origins (which is not directly adjacent to Pygmy territory) is well documented linguistically to a tiny area on the Niger-Cameroon border just a few thousand years ago.

There are a couple of language families in the Andamans, one connected to Austronesian, and the other not. The population genetic evidence that they are not Austronesian (based on studies of hair samples) is pretty convincing.

There is also good population genetic evidence that South Asian tribal populations are distinct from other South Asian populations and prior in time, although the divisions within populations of "everybody else" in South Asia are more controversial.

The strongest evidence that the Philippines Negritos did not speak an Austronesian language (as they do now) is that the time and place where the Austronesian languages originated is probably the most definitive one in all of linguistics and is supported by population genetic evidence. If the Philippines Negritos didn't lose their language and instead were Austronesians who moved into the tropical forests, then you have a new mystery to explain who lived in the Philippines for the previous 60,000 years and you have to assume that they were completely wiped out.

"It's remarkably suspicious that all short-statured populations should have lost their languages without a trace."

Two populations (one of the Andaman Island groups and a group from Nepal) have not. Those that have lost their languages are vastly outnumbered by their neighbors. Papuan and many small aboriginal Australian languages still survive, but many of those populations are only a couple of hundred years or less away from first contact with outsiders (as opposed to thousands of years for Niger-Congo and Austronesian exposed populations).

Example of language loss by small populations that are vastly outnumbered by their neighbors are well documented in North America, South America, North Asia, Africa, and Europe in the historical era and are frequent as we speak. There are probably fewer than 1,000,000 pygmies, Khoisan, uncontacted Amazonians, and Asian Negritos alive today out of 6,000,000,000 people on the planet. More than half of the languages in the world are on the verge of extinction. When you are outnumbered 600:1 or more, don't have formal language recognition from the government, have small base populations, and are economically marginalized, losing our language is hardly surprising.

German Dziebel said...

"The population genetic evidence that African Pygmies are genetically distinct from other Africans and are more closely related genetically to the Khoisan than to their linguistic neighbors is pretty definitive and further dates divisions of pygmy populations into subgroups. Conventional mutation rate dating puts the Pygmy-Khoisan v. West African split at 70,000 years ago, And, the African pygmies don't speak just any Niger-Congo languages, they speak Bantu family Niger-Congo languages..."

How do you like the title of this paper by Quintana-Murci et al. Maternal traces of deep common ancestry and asymmetric gene flow between Pygmy hunter–gatherers and Bantu-speaking farmers //PNAS 2008?
Pygmies and Bantu are genetically related and speak related languages.

"There is also evidence of substrate language in African pygmy speakers of Niger-Congo languages which they speak now (mostly phonetic)."

This evidence doesn't exist. All Bantu- and Ubangian-related (notice: not just Bantu!) Pygmy languages derive phonetically from proto-Bantu and proto-Ubangian.

"The population genetic evidence that they are not Austronesian (based on studies of hair samples) is pretty convincing."

Very good. Whose language did they adopt? In fact, I was using Ongan as an example of a short-statured population that speak a language related to Austronesian, hence of roughly the same age as other Austronesian languages.

"The strongest evidence that the Philippines Negritos did not speak an Austronesian language (as they do now) is that the time and place where the Austronesian languages originated is probably the most definitive one in all of linguistics and is supported by population genetic evidence. If the Philippines Negritos didn't lose their language and instead were Austronesians who moved into the tropical forests, then you have a new mystery to explain who lived in the Philippines for the previous 60,000 years and you have to assume that they were completely wiped out."

May be they were. Why not? Maybe Homo floresiensis lived in the Philippines. In any case, your theory allows for Austronesians to move into the tropical forest - otherwise how did the Negritos borrow their language. So, both theories agree that Austronesian-speakers moved into the tropical forests. But then you're making an assumption, which is not parsimonious and entirely arbitrary, namely that some of these Austronesian-speakers used to speak a different language. Only because they are foragers and have a short stature. This information has nothing to do with linguistics.
Don't repeat the mistakes of 19th century scholars who classified "races" on the basis of such disparate attributes as grammatical structure and hair texture.

Do we really have Pygmy populations in Papua New Guinea? There're accounts of Pygmy populations in Australia, but nobody makes a claim that they lost their languages, although they could just as easily as you do for Africa or Southeast Asia. Your method of arbitrarily assigning language replacement to groups that look "ostensibly ancient" to you can take you anywhere: why not suggest that all Finno-Uralic-speakers are former Indo-European speakers? The only way to prove a language replacement is to use the data from languages themselves. If there're no substratum effect detectable in them, just leave them alone. At best, it's a hung jury situation.

"Example of language loss by small populations that are vastly outnumbered by their neighbors are well documented in North America, South America, North Asia, Africa, and Europe in the historical era and are frequent as we speak."

Of course language loss is real. However, our attention is constantly distracted by the myths of language loss that we tend to attribute to phenotypically "inferior" populations. In Africa, we better focus on other populations such as Laal, which may be better representative of Pleistocene Africa than Pygmies.

German Dziebel said...

In addition to Roger Blench's article I quoted above, you could read Alan Fix's work on the Semang and their agricultural neighbors. One of his most recent papers is published in "Forager-traders in south and southeast Asia: long-term histories," edited by Kathleen D. Morrison and Laura Lee Junker. His conclusion is that phenotypical variation between Negrito foragers and agriculturalists is continuous, rather than discrete and probably represents descent from a single ancestral population speaking the same language.

terryt said...

"it's not that Pygmies and Negritos lost their original language but rather some speakers of Niger-Congo, Nilo-Saharan, Austronesian and Austroasiatic languages lost their original phenotype as they moved into tropical forests".

That could be true in some cases, such as in Borneo. In fact it is extremely likely if humans have only relatively recently been able to exploit jungle habitats. But in that case the Negritos would not represent ancient relict populations. They would be close genetically to their neighbours. Unfortunately for your argument:

"The population genetic evidence that African Pygmies are genetically distinct from other Africans and are more closely related genetically to the Khoisan than to their linguistic neighbors is pretty definitive and further dates divisions of pygmy populations into subgroups".

Pretty convincing evidence that the African Pygmies at least have been long established in their jungle homes.

"I did not confuse anything of the like, Terry: cuscus in Wallacea and New Guinea, tarsier in Philippines".

Sorry for the misunderstanding. But the cuscus is by no means distributed widely through Wallacea. Just a few islands very close to Sahul. Just as a few Sunda mammals have managed to reach neighbouring islands in Wallacea.

Maju said...

Andrew: you are right. But it's futile to argue with German Dziebel and his beliefs. He will have no respect for truth and manipulate the data in order for it to feel with his expectations. He knows perfectly that Pygmies have nearly no genetic relation with other Africans but he will argue otherwise, just because. I suggest that in doubt read his own sources, which never say what he claims he does, at least not the genetic ones.

He's a warrior of absurd pseudo-anthropolgy, not any scientifically-minded person, He does so because he believes, against all fossil and genetic evidence, that humankind arose in America and spread from there, so he's forced to challenge everything that is common sense to the rest. It's a total waste to argue with him.

I believe I must issue the warning because, while Terry knows about him (and doesn't seem to mind), I believe Andrew and Julien do not.

terryt said...

"so terryt is strictly wrong on that point as is the article".

I think you will find that more and more scientists are suggesting the line runs to the west of the Philippines. Those islands, except for Palawan, have never been connected to Sula. So, even though 'The channel connecting the Sula Sea and the Celebes Sea and between the Sula Sea and the Pacific Ocean would have been comparable in many ways to the Dardanelles and Bosporus straits with the Sula Sea analogous to the Marmara Sea' they were basically impassable for placental mammals.

"His conclusion is that phenotypical variation between Negrito foragers and agriculturalists is continuous, rather than discrete and probably represents descent from a single ancestral population speaking the same language".

Surely gene flow between the two populations would give the same result.

terryt said...

"But then you're making an assumption, which is not parsimonious and entirely arbitrary, namely that some of these Austronesian-speakers used to speak a different language".

It's impossible that both the Onge and the Philippine Negritos spoke Austronesian languages when they first arrived (c. 60,000 years ago)in their respective island habitats. Any language they might have originally had in common would have diverged so much that the two languages would today be unidentifiable as belonging to a single family. Austronesian began diverging no more than 5-6 thousand years ago. So it must be true 'that some of these Austronesian-speakers used to speak a different language'. Nothing to do with 19th century racism.

German Dziebel said...

"It's impossible that both the Onge and the Philippine Negritos spoke Austronesian languages when they first arrived (c. 60,000 years ago)in their respective island habitats."

Terry, we don't know when they arrived in their habitats. That's what we're trying to understand. But you already assume that Negritos must be 60K years old. Then of course you end up hypothesizing language replacement. But these are not facts but a pile of assumptions. If Ongan is related to Austronesian and Ongans are short, then Ongans must be some 5-8,000 years old. Unless, of course, Ongans again lost their very distinct language and adopted an Austronesian language. But this will be too much to assume, right?

"Nothing to do with 19th century racism."

Not in your case, of course. But the origin of this ideas is with 19th century Victorian science. Modern scholars question it all over the place. See my references.

"which never say what he claims he does, at least not the genetic ones."

Luis, I know for a fact that you don't read anything. But at least make an effort to absorb the following title: "Maternal traces of deep common ancestry and asymmetric gene flow between Pygmy hunter–gatherers and Bantu-speaking farmers." In all conventional phylogenies, Pygmies belong with the rest of Sub-Saharan Africans, while Khoisans are outliers. The Khoisans speak languages utterly unrelated to Nilo-Saharan or Niger-Congo, while all Pygmy groups speak Nilo-Saharan and Niger-Congo languages. There's a perfect concordance between linguistics and genetics. In a recent post, Julien made an argument for an interdisciplinary approach to human origins. This is exactly what I practice.

Your attempts to police my forum contributions are amusing. Just a reminder: I'm a man with two doctorates in relevant scientific disciplines, while you're a half-literate amateur believing in the myth of an African Eve.

German Dziebel said...

"Surely gene flow between the two populations would give the same result."

Apparently, this "gene flow" failed to conceal the Pleistocene phenotype of the Negritos from your eagle eye.

German Dziebel said...

Terry: "Any language they might have originally had in common would have diverged so much that the two languages would today be unidentifiable as belonging to a single family. Austronesian began diverging no more than 5-6 thousand years ago."

Andrew: "Conventional mutation rate dating puts the Pygmy-Khoisan v. West African split at 70,000 years ago, more or less."

Both genetics's molecular clock and linguistics' glottochronology are very provisional dating systems. Although the dates for Austronesian are well-established (there're issues with linguistic subgrouping, though, so we still have to be careful), in Africa the situation is a bit less clear. It's possible that African language families are older than 5-6,000 years, while African molecular clades are younger than 140,100, 70,000 years. The Khoisan language family is often dated at 18,000 years. Khoisan physical morphology is older than 5K but younger than 36K (see Stynder et al.). Early to mid-Holocene South
African Later Stone Age human
crania exhibit a distinctly
Khoesan morphological
pattern // South African Journal of Science 103, July/August 2007. When geneticists date the Khoisan vs. the rest split at 140K, it's at odds with other sciences. How in the world can Khoisans split from the rest of proto-humans 100 years before the emergence of modern human behavior in the archaeological record and still have all the accoutrements of a behaviorally modern humans. Geneticists' dates are, therefore, clearly suspect.

German Dziebel said...

Apologies for the hasty writing. The last sentence should read: "When geneticists date the Khoisan vs. the rest split at 140K YBP, it's at odds with other sciences. How in the world can Khoisans split from the rest of proto-humans 100K years before the emergence of modern human behavior in the archaeological record and still come out with all the accouterments of a behaviorally modern human population? Geneticists' dates are, therefore, clearly suspect.

German Dziebel said...

Terry and Andrew: another relevant source for our discussion of the origin of Negrito populations is Mapping Human Genetic Diversity in Asia by The HUGO Pan-Asian SNP Consortium, et al. Science 326, 1541 (2009).

On p. 1545 it literally says: "Although this study does not disprove a two-wave model of migration, the evidence from our autosomal data and the accompanying simulation studies (figs. S29 and S30) point toward a history that unites the Negrito and non-Negrito populations of Southeast and East Asia via a single primary wave of entry of humans into the continent."

Just like in Africa, short-statured foraging populations in SE Asia share with their agricultural neighbors both genetic and linguistic common ancestry.

terryt said...

"while Terry knows about him (and doesn't seem to mind)"

I firmly believe argument, discussion and different evidence help firm up our ideas about any given subject. And German does provide some interesting evidence.

Back to the really interesting bit:

"Regardless of the precise taxonomic affiliation of that bone, it indicates a great time depth for human presence in that part of the Old World"

It places the conventional OoA at 50k in extreme doubt. I know Maju and I both agree that the OoA must be older than that, and German doesn't accept it at all, but many must now have to adjust their thinking on the subject.

Maju said...

"It places the conventional OoA at 50k in extreme doubt".

Very few people if any has been working with such recent dates in quite a long time. Genetic evidence is overwhelmingly supportive of older colonization of Tropical Asia before West Eurasia, what necessarily implies older dates. 70 ka is much more the kind of "conventional wisdom" within the "recent coastal route migration" model (for example Karafet 2008).

"... but many must now have to adjust their thinking on the subject".

We still don't know if is a Homo sapiens or a Homo floresiensis. By the moment it's just unclear evidence and I'd say that Ljujiang or the claimed oldest dates for the colonization of Australia are so far more conclusive, even if not too much more.

There's also a jaw with chin in South China dated to c. 110,000 Ka ago, which should be more suggestive, yet nobody seems moved by it.

The key issue with this finding is that it adds evidence for some navigation capability at those or earlier dates (Crete quartz axes).

terryt said...

"Very few people if any has been working with such recent dates in quite a long time".

I disagree. We still see it mentioned without qualification often enough.

"There's also a jaw with chin in South China dated to c. 110,000 Ka ago, which should be more suggestive, yet nobody seems moved by it".

Doesn't fit the paradigm?

"The key issue with this finding is that it adds evidence for some navigation capability at those or earlier dates (Crete quartz axes)".

I'm still far from convinced for claims of such early 'navigation capability'. You've claimed elephants could swim the distances required, so how come they failed to cross the lesser distance to the Philippines? I still would suggest that very ancient human sea crossings are accidental rather than deliberate.

Andrew Oh-Willeke said...

"I think you will find that more and more scientists are suggesting the line runs to the west of the Philippines."

A biogeograpical line surely does separate the Philippines from Sundaland. But, that biogeographical line is not fairly described as "the Wallace line," because the term "Wallace line" refers to a specific line drawn on a historical map by a specific historical figure, Alfred Wallace.

One might call the line you describe "the Sunda line," for example.

Andrew Oh-Willeke said...

"[O]ur attention is constantly distracted by the myths of language loss that we tend to attribute to phenotypically "inferior" populations."

Language loss is something that has been experienced by almost everybody, without regard to phenotype. Etruscan and Pictish and Akkadian and Minoan and most of the Paleolithic languages of the neighbors of the Han Chinese and the Iberian Celtic languages and Sumerian are languages every bit as dead as the ancestral languages of the Pygmies and Philippines Negritos.

The issue is not phenotype, it is being ruled or economically/culturally dominated for some extended period of time by someone who speaks a different language.

Kingdoms and empires are bad for the languages of the non-ruling classes in those kingdoms and empires. The vast majority of people in North America and South America trace their descent mostly to people who did not speak English, Spanish, Portugese or French. Yet, these are the overwhelmingly dominant languages of North America and South America and few others are stable, let alone thriving. There are millions of Americans of Asian, Italian, Greek, German, Scandinavian, Native American, or African, descent who know not a single word of the languages that their ancestors spoke in 1450 CE or even 1800 CE.

Latin is not the birth language of anyone today despite having been the dominant language of the largest empire in the world just 1700 years ago. The language of the people who proximately brought down the Roman empire (Gothic) is dead as well. There are no detectable genetic traces of the people who brought the Hungarian language to Hungary, but its predecessor language is gone. Hittite and its descendants are no longer spoken despite their one time regional dominance in Asia Minor. Tocharian and Avestan and Elamite are just as dead.

Linguistic dating isn't absolute. But, it does accurately show relative age and can be calibrated with dates of other archeological events.

terryt said...

"One might call the line you describe 'the Sunda line,' for example'.

Fair enough. Julien wrote, 'What's a bit puzzling is their repeated discussion that the Philippines are east of Wallace's line'. Perhaps what they meant was to include the Philippines in Wallacea. Wallacea is not defined as being either east or west od Wallace's line. Rather it is the region between Sunda and Sahul, and Wallace's line runs through it.

As for your comments on language replacement. Most of us would agree. Well put.

German Dziebel said...

"Language loss is something that has been experienced by almost everybody, without regard to phenotype....The issue is not phenotype, it is being ruled or economically/culturally dominated for some extended period of time by someone who speaks a different language."

Well, in the case of Negritos and Pygmies there's no evidence for language loss. Hence, a common - albeit not universal - assumption that language loss happened rests on a myth that small-statured foragers must have been dominated by their taller agricultural neighbors.

German Dziebel said...

"Linguistic dating isn't absolute. But, it does accurately show relative age and can be calibrated with dates of other archeological events."

I'm not sure exactly what you mean here, Andrew. There's a common belief that in most cases a first-order language family is no older than 5-6,000 years. There's no radiocarbon type of certainty behind this inference, and in a number of cases scholars working on individual families go over this time range in estimating the age of "their" language families. There're schools of thought (e.g., the proponents of the Indo-European continuity theory or the theory by Rogers et al. advocating for a remarkable similarity in the patterns of distribution of North American Indian families and Pleistocene (!) geographical zones.

But what seems to be certain is that past 5-6,000 years (i.e., beyond the time horizon of Austronesian, Indo-European and, with younger dates, Bantu) all attempts to correlate linguistic families with archaeological sites become utterly unreliable.

In addition to absolute age measures for language families based on the assumption of constant rate of lexical loss there are relative estimates based on the levels of phylogenetic and structural-type diversity on regional and continental levels. What is promising is not so much the mapping of languages onto archaeological cultures but the mapping of languages onto molecular clades and lineages (with or without kinship structures sandwiched between the two data patterns). There are reasons to believe that some form of language classification (first-order language families, macro-families, structural types, etc.) more or less accurately represent historical events and are well-mappable onto molecular clades. Gene flow and language replacement may mess up the overall picture in places but these secondary processes should become more easily detectable as research progresses.

The main point to be aware of is that languages are not epiphenomenal to "hard data" such as genes, tools and bones and cannot be easily forced into convenient scenarios of replacement. In addition to the myths of the original Pygmy and Negrito languages, I keep hearing people offhandedly attributing the lack of indigenous North American levels of linguistic diversity in North(east) Asia and Europe to some kind of massive Holocene language replacements. In both cases, no evidence is being presented but rather the lack of a need for it is assumed.

German Dziebel said...

More interdisciplinary data on the African Pygmies and the Negritos.

In full accordance with linguistics, craniology also supports common ancestry between Bantu-speakers and (Bantu-speaking) Pygmies, while Khoisans stand apart.

"In addition, although the “inter-ethnic” variation of the Pygmies was probably under-estimated, most of the latter despite their small sample size were often localized within or close to the variation of (Bantu-speaking) Central Africa. This fact also supported previous morphological observations as well as the hypothesis of a common origin for both Pygmies and Bantu-speakers, as it was proposed by Hiernaux (1976) and Froment (1993)." (Ribot, "Differentiation of modern sub-Saharan African populations: craniometric interpretations in relation to geography and history," 2004, http://bmsap.revues.org/3873).

As for Negritos, their dental pattern, which is an important prehistoric marker, falls within the general Southeast Asian Sundadonty type and shares no special proximity to either Africa or Australia. Again, in accordance with linguistic data. See "Negritos, Australian Aborigines, and the proto-sundadont dental pattern: The basic populations in East Asia, V," by
Tsunehiko Hanihara, 1992.

One could, like Terry does, invoke gene flow, but in the light of cross-disciplinary evidence, common ancestry would be the most parsimonious solution.

Anonymous said...

Wallacea is between Wallace's Line and Lydekker's Line. Wallace's Line doesn't run through Wallacea! Get over it. Please identify the authors who currently think otherwise.

Anonymous said...

Huxley's Line already exists for the "Sunda Line" being proposed.

Andrew Oh-Willeke said...

"I keep hearing people offhandedly attributing the lack of indigenous North American levels of linguistic diversity in North(east) Asia and Europe to some kind of massive Holocene language replacements."

We have overwhelming evidence of massive Holocene language replacement in the historic era. In many cases people were there and wrote down the story of what happened. When they didn't, the evidence is still often overwhelming.

From historic records: We know why people all over the world speak Arabic (the Islamic empire's expansion starting in the 7th century CE). We know why people all over Europe speak languages descended from Latin (the Roman Empire). We know why the Turkish people speak Turkish (language replacement from Anatolian Indo-European languages under the Seljek Turks). The first Turkic, and then Mongol expansions and language replacements from East to West in Central Asian and Siberia are documented. We have documentation of how the Tocharian language was replaced, although its origins are more obscure. We know why many people in Finland speak Swedish (the Swedes invaded and ruled Sweden for centuries). We know why people speak Spanish, Portugese, English and French in the Americas, and why English is spoken in Australia and New Zealand. We know the history of the expansion of the Slavic languages. We know that the Chinese forcefully expanded and forced those speaking other languages to learn Chinese or flee. We know how English came to be the language that it is today and where it came from. We know how Minoan was replaced by Greek. We known from detailed written records how the Semitic language Akkadian replaced Sumerian and how the Indo-European language Hittite replaced the non-Indo-European language Hattic. The death of Etruscan in favor of Latin was chronicled by Romans who knew the last people to speak that language. We know how and when and why Coptic ceased to be a living language in Egypt.

Evidence of Bantu expansion and language replacement from Cameroon-Nigeria's border with evidence of a Khoisan substrate in parts of Southern and Eastern Africa is extremely solid, although some of it is slightly pre-historic. Evidence of the Indo-European languages of India being derived from Sanskrit from around 1500 BCE to 2000 BCE and replacing prior languages in much of India is very strong although the historical sources are only legendary. The anthropological and linguistic evidence for the origins and expansion of Celtic languages replacing the languages spoken before the Celtic languages is compelling. The details of the Austronesian language expansion are iron clad (and in a large number of places arrived in parts of the world with no prior human inhabitants in the Pacific and Madagascar, for example).

There are blind spots. There is not a consensus on the origins of the Dravidian languages. The early expansion of the Uralic languages is less well documented than we might hope. But, there is no reasonable doubt that there has been massive Holocene era language replacement on every continent.

German Dziebel said...

"But, there is no reasonable doubt that there has been massive Holocene era language replacement on every continent."

You are right: on every continent. The argumentative stance I was countering, however, tends to posit massive Holocene language replacements for Siberia (and Africa, when it comes to that) but not for America.The expansion of Na-Dene, Algic, Iroquoian, not to mention Uto-Aztecan, Quechua and others brought about language extinctions as well. Low population density in the Americas and the isolation of tribal demes posed great threats to their survival even without larger neighbors coming to dominate them. More importantly, European colonization from 1492 resulted in the loss of hundreds of Amerindian languages (California and Texas in North America are especially noteworthy), and no record of them exists. So, we don't need to go as far as the Holocene in search for language extinctions in the Americas.

The argument I mentioned is totally biased, and if read by American Indians also probably culturally insensitive. To attribute the differences in the levels of linguistic diversity between America and Siberia to Holocene extinctions in Siberia in order to bolster the archaeological argument for Amerindian recency (which has always had legal implications for American Indian tribal sovereignty) without a mention of the most recent language extinctions in America may come across as a colonial bias.

It's important to understand that there 140 language stocks in the Americas (compare 20 in Africa) and the world largest number of linguistic isolates and small language families. The associated grammatical structures and kinship structures are archaic. This can't be (and shouldn't be) dismissed as an artifact of one-sided Holocene extinctions in the Old World.

What it tells me is that America preserves Upper Pleistocene demographic, population and social structure that in the Old World was disrupted by massive population expansions and global colonizations from some 40-50K on (and possibly earlier). But these "massive" population processes and the concomitant accrual of molecular diversity didn't result in the replacement of the original languages in the Old World but rather in the expansion of existing families across vast geographic expanses. (That's what we would expect to find in the Americas if America was indeed a recently colonized continent.) This expansion of course led to the loss of some pre-existing linguistic diversity and in the shrinking of the geographic ranges of less successful language families but the magnitude of replacement was insignificant in comparison to the magnitude of expansion.

Note that the largest number of languages in the world belongs to Niger-Congo (followed by Austronesian). This is an exact correspondence to the levels of molecular diversity in Sub-Saharan Africa. But the basal linguistic diversity in Africa is very low (only 20 families and isolates). So, the Niger-Congo linguistic and genetic phenomenon is an artifact of population growth and is not indicative of age. In America, we see the opposite situation: basal linguistic diversity is huge, while demographic constraints have kept molecular diversity at Upper Pleistocene levels.

German Dziebel said...

(contd.)

For a good discussion of the linguistic puzzle of the Americas see, e.g., Roger Blench's website and his paper "Accounting for the pattern of Amerindian Languages." http://www.rogerblench.info/RBOP.htm

When we realize that Pygmies speak Niger-Congo and Nilo-Saharan languages with little to no "substrate effects," that craniologically and odontologically they cluster with their respective linguistic relatives, that their kinship structures are also highly transformed, we can't help but suspect that the postulation of 70,000 years for the divergence of Pygmies followed by 67,000 years of their isolation in tropical forests and finally the 3,000 of co-existence with Bantu that supposedly resulted in the complete loss of their very old languages is a myth perpetuated by geneticists. Most likely, the divergent mtDNA L lineages and Y-DNA B lineages found in the Pygmies are of relatively recent origin. Their divergence was caused by some adaptive pressures stemming from the demands of their new habitat that resulted in the acceleration of mutations in Pygmy mtDNA and Y-DNA lineages.

American Indian mtDNA and Y-DNA haplogroups, on the other hand, are associated with very divergent language families and hence (in certain critical nucleotide positions) are likely very old. The actual genetic diversity in American Indian tribes is hard to assess due to a large number of private polymorphisms that have never expanded past their clan or family of origin. See, e.g., Extensive mitochondrial diversity within a single Amerindian tribe, by Ward et al. //PNAS 1991. It comes as no surprise that geneticists keep discovering additional Amerindian haplogroups (such as "basal" M) in ancient remains and in undersampled populations.

terryt said...

"Please identify the authors who currently think otherwise".

Mijares, A., Détroit, F., Piper, P., Grün, R., Bellwood, P., Aubert, M., Champion, G., Cuevas, N., De Leon, A., & Dizon, E. for a start. As Julien said, 'What's a bit puzzling is their repeated discussion that the Philippines are east of Wallace's line'. Peter Bellwood, for one, has long considered the Philippines to be basically part of Wallacea.

"Most likely, the divergent mtDNA L lineages and Y-DNA B lineages found in the Pygmies are of relatively recent origin".

You could be onto something there. I've long argued with Maju as to whether humans have a long history in the jungle or not. To me there certainly seems to be evidence for lack of human habitation in the SE Asian jungle. So Pygmies may have relatively recently moved into such a habitat.

"To attribute the differences in the levels of linguistic diversity between America and Siberia to Holocene extinctions in Siberia in order to bolster the archaeological argument for Amerindian recency (which has always had legal implications for American Indian tribal sovereignty)"

I don't see how it can have legal implications. The Amerindians were obviously there long before Europeans arrived.

"The actual genetic diversity in American Indian tribes is hard to assess due to a large number of private polymorphisms that have never expanded past their clan or family of origin".

That could easily be the result of closer sampling. The American Indians have probably been sampled more closely than most other indigenous people. And one of Dienekes' recent posts claims identifiable differences between neighbouring villages in parts of Europe. Surely that's the same phnemenon.

German Dziebel said...

"So Pygmies may have relatively recently moved into such a habitat."

Bingo! Myth debunked.

"I don't see how it can have legal implications. The Amerindians were obviously there long before Europeans arrived."

That's a long story, and probably shouldn't have opened a can of worms here. The special status of American Indian tribes (tax exemptions incl.) in an otherwise immigrant nation derives from their claim for collective land ownership, cultural distinctiveness and aboriginal status. Back in the late 19th century under the Dawes Act, their lands were partitioned into individual chunks. Needless to say, it became possible because of the dramatic reduction of North American Indian population size to 200,000 people. Concomitantly, the dominant scholarly opinion was that these tribes are no more than 5,000 years old (Hrdlicka). They were deemed immigrants into North America just like any other ethnicity. The fact that they technically preceded Europeans was of little value because all immigrants, regardless of the timing of their entry to the U.S. (5K earlier before the Europeans or 5K after the Europeans) were accorded the same status. In the 1930s, under Collier's New Deal, this changed and American Indian patrimony was reaffirmed. Concomitantly, the antiquity of man in the Americas was extended to 12K on the strength of Clovis projectile point culture.

These days geneticists experience huge difficulties in obtaining gene samples from North American Indian tribes as the latter perceive that geneticists are appropriating their blood, defaulting on their promises to provide medical solutions to the tribes and are further re-purposing and misusing it to "prove" the immigrant status of American Indian tribes. They think of the Bering Strait theory as a myth that contradicts their "origin stories." See a glimpse of these debates in http://www.nytimes.com/2010/04/22/us/22dna.html?adxnnl=1&adxnnlx=1277478142-RDaiEuL034LJBDL2IhfoCA

It's a fascinating topic. Nowhere else I believe science, religion and politics are as intertwined as in the issues around American Indian origins.

"The American Indians have probably been sampled more closely than most other indigenous people."

Not necessarily. It could be the opposite, per above.

Anonymous said...

@terryt
Nice try. Your source is the paper that makes the very mistake we're discussing! Ha. I've corresponded with the authors of the Philippines metatarsal paper, and they agree that they should have used "Huxley's Line" instead. BTW, it was Huxley who formalized "Wallace's Line" in the first place! There are various "lines" that have tried to make biogeographic sense of this archipelago; no need to make up new ones.

German Dziebel said...

"And one of Dienekes' recent posts claims identifiable differences between neighbouring villages in parts of Europe. Surely that's the same phnemenon."

This is exactly what geneticists working with classical markers noticed about South American tribes: the high level of inter-village differentiation due to fission-fusion and kin selection effects. There's a slew of publications (e.g., The genetic structure of a tribal population, the Yanomama Indians, by Neel et al.) documenting it. They are now largely forgotten but Salzano recently tried to restore the historical truth in an article in Current Anthropology "The Fission‐Fusion Concept" (2009). Early mtDNA studies largely confirmed this pattern (see Ward et al. 1991, with the dates of 78,000 for the internal differentiation in a single American Indian tribe, a figure much larger than that for the !Kung Bushmen). The Out of Africa model of human dispersals is based on an assumption of unstructured panmictic ancestral population (the Green et al. (2010) Neandertal study squarely puts this assumption on the line), on the sampling of isolated individuals, rather than clans, families or villages and on the interspecific (human-chimp divergence), rather than intrafamilial calibrations of the molecular clock.

terryt said...

"There are various 'lines' that have tried to make biogeographic sense of this archipelago"

Exactly the point I was making: the Philippines are definitely not part of Sunda. Nor are Halmahera and Sulawesi part of Sahul. The two end (the Philippines and Halmahera) are opposite ends of a biological cline through Wallacea (however widely you might care to define that region).

"no need to make up new ones".

You obviously didn't notice that it wasn't me who proposed a new name.

"Early mtDNA studies largely confirmed this pattern (see Ward et al. 1991, with the dates of 78,000 for the internal differentiation in a single American Indian tribe"

But that doesn't mean at all that the particular tribe has been genetically isolated all that time. Just that it has become made up of people whose common ancestry (where-ever that was, probably several completely different places) goes back that far. Just as Maju makes the mistake of assuming the region of apparent greatest diversity of haplogroups strongly indicates region of that haplogroup's origin. Say for example that members of a haplogroup spread out and diversify into several separate haplogroups. Later, if members of many of these separate haplogroups are involved in a wider expansion into a new region we will have a great diversity of the haplogroup in that new region. But it by no means proves the haplogroup as a whole originated there.

"Not necessarily. It could be the opposite, per above".

They were certainly heavily sampled in the early days. They were a population easily accessible to students at universities in the USA, where much of the early genetic work was done.

German Dziebel said...

"But that doesn't mean at all that the particular tribe has been genetically isolated all that time. Just that it has become made up of people whose common ancestry (where-ever that was, probably several completely different places) goes back that far. Just as Maju makes the mistake of assuming the region of apparent greatest diversity of haplogroups strongly indicates region of that haplogroup's origin. Say for example that members of a haplogroup spread out and diversify into several separate haplogroups. Later, if members of many of these separate haplogroups are involved in a wider expansion into a new region we will have a great diversity of the haplogroup in that new region. But it by no means proves the haplogroup as a whole originated there."

Yes, you're right here, Terry. It's indeed possible. One fact of diversity or lack thereof by itself doesn't mean much. That's why I tend to look at a problem from the perspectives of different disciplines and data streams.

"They were certainly heavily sampled in the early days. They were a population easily accessible to students at universities in the USA, where much of the early genetic work was done."

I'm not sure about that. You may be right.

terryt said...

"I'm not sure about that. You may be right".

It certainly holds for mitochondrial DNA for starters. I mean the letters A, B, C and D were applied first to the American Indian haplogroups. It was not until some time later that closely related haplogroups were identified through much of East Asia. At that point E was nominated to represent a SE Asian haplogroup, F for a widespread East Asian haplogroup and G, a Northeast Asian haplogroup. It's not until the letters had run all the way to H that it became necessary to label European haplogroups. And then not until L for African ones. And the order in which they were studied and labeled led to the extraordinary necessity of having to label one of them L0.

terryt said...

Regarding Wallacea:

http://en.wikipedia.org/wiki/Wallacea

From the article:

"The islands of Wallacea lie between Sundaland (the Malay Peninsula, Sumatra, Borneo, Java, and Bali) to the west, and Near Oceania including Australia and New Guinea to the south and east".

The philippines are part of neither region. And:

"The Philippines (excluding Palawan which was part of Sundaland) are usually but not always considered a separate region from Wallacea".

So it's not just me and the authors of the current paper. It seems that people in this part of the world include the Philippines but those in Europe and North America prefer to exclude them.

http://actazool.nhmus.hu/48Suppl2/newwallace.pdf

Have a look at the map near the beginning of the article.

German Dziebel said...

Terry: "It certainly holds for mitochondrial DNA for starters. I mean the letters A, B, C and D were applied first to the American Indian haplogroups. It was not until some time later that closely related haplogroups were identified through much of East Asia."

I'm still not convinced. The chronologically first mtDNA study (Johnson et al. 1983) was based on a worldiwide sample, and the American Indian sample was the one that was damaged. Then they were re-tested and it turned out (already after the paper was published) that Amerindians had the ancestral type at 100% frequency. Then, I have a AJHG paper from, again, 1983, entitled "Amino Acid Change Associated with the Major Polymorphic Hinc II
Site of Oriental and Caucasian Mitochondrial DNAs" by Blanc et al. which has no mention of Amerindian samples, but has African, European and Asian ones. They didn't have the names for those "haplotypes" yet (only the restriction sites) but the underlying phenomena were described on the basis of worldwide samples and Amerindian samples weren't the most prominent ones.

There was a period in the 1990s, I remember, when Siberian natives were undersampled over Amerindians. Hence, the earliest published dates for the divergence of American Indian haplogroups are much higher than the corresponding dates for the divergence of the same haplogroups in Siberian populations. But I think it concerned only Siberia vs. America and not America vs. Asia as a whole.

What is also true is 1) that geneticists used sequences that are found at high frequencies in American Indians as examples of automatically derived ones because archaeologists "proved" that American Indians are recent colonists. This simplified the construction of phylogenies; 2) that lots of early "African" samples came from African-Americans who were technically American populations for several hundred years. So, all the earliest announcements about the detected "diversity" of African populations were already compromised by the de facto situation whereby America was the most diverse continent.

occamseraser said...

Ernst Mayr penned a thoughtful review about "Wallace's Line" back in 1944 in the Quarterly Review of Biology 19: 1-14. He distinguishes between the "original" line of Wallace that excludes the Philippines and the one modified by Huxley that is more inclusive. I'm trying to chase down a reference therein by Scrivenar et al (1943)in the Proc. Linn. Soc. London that apparently argues that the faunal contents of Wallacea (however one defines its boundaries) is not so much a transition zone between Sunda and Sahul as it is a blend of several different biomes in between.

terryt said...

"the faunal contents of Wallacea (however one defines its boundaries) is not so much a transition zone between Sunda and Sahul as it is a blend of several different biomes in between".

Exactly. And the origins of those biomes is very revealing. The flora is mostly of Asia origin evidently. Some members of the Sunda fauna made it to the Philippines, especially to Mindanao, and some members of the Sahul fauna made it to Halmahera and Sulawesi. But neither region is fully a member of either Sunda or Sahul.

And I'd take what Ernst Mayr says on the subject very seriously.

German Dziebel said...

On the issue of common ancestry between the short-statured foragers Pygmies and the agricultural Bantu, in addition to craniological and linguistic evidence quoted above, we have compelling mtDNA ("The L1c haplogroup of mitochondrial DNA retains a signature of a phase common to the ancestors of the two groups, while encompassing some specific sub-clades which can mark
their divergence" [Batini et al. Phylogeography of the human mitochondrial L1c haplogroup: Genetic signatures of the prehistory of Central Africa, p. 642] and Y-DNA evidence (Bantu and Pygmies belong to two sublades [B2a and B2b] of the same B2 lineage [Barnielle-Lee et al. "Genetic and Demographic Implications of the Bantu Expansion: Insights from Human Paternal Lineages," 2009]). There're of course traces of secondary Bantu and Khoisan lineages in Pygmies and Khoisan and Pygmy lineages in Bantu, but the core B2 nexus is shared exclusively between the two populations that also share linguistic kinship. It's noteworthy that Y-DNA shows the divergence of the B2b lineage at 11K and B2a lineage at 5K, which suggests that the emergence of Pygmies as a distinct population is relatively recent. Coalescence dates for mtDNA L1c are astronomical (90K years) and should be dismissed. Batini's inference that "our findings allow us to draw a scenario where, at least 90kya, ancestors of Western Pygmies and Bantu were still forming a genetically coherent hunting-gathering population" is completely ridiculous and flies in the face of all other lines of evidence.

terryt said...

"Batini's inference ... is completely ridiculous and flies in the face of all other lines of evidence".

Not necessarily so. If the pymies hadn't invaded the jungle by 90kya it's quite possible.

"while encompassing some specific sub-clades which can mark
their divergence"

Do they postulate any dates for the divergence of those Pygmy-specific sub-clades?

"which suggests that the emergence of Pygmies as a distinct population is relatively recent".

I think we're both in agreement on that. However even in the 'traditional' OoA it seems from Maju's work on mtDNA L that the Pymies were yet to become adapted to jungle when the Lines M and N emerged.

German Dziebel said...

"Do they postulate any dates for the divergence of those Pygmy-specific sub-clades?"

Yes, they do. The Pygmy clades L1c1a, L1c4, L1c5 are dated at 29, 37 and 37K (rounding up), respectively, the Bantu clades L1c1b, L1c1c, L1c2 are dated at 59, 46 and 55K, respectively. So, as you can see, the Bantu clades are older than the Pygmy clades. What do you make of that?

"Not necessarily so. If the pymies hadn't invaded the jungle by 90kya it's quite possible."

So, the Pygmies and the Bantu existed as a single population outside of the jungle 90K speaking dialects of the same language and sharing the same skull morphology. Then the Bantu split off and diversified between 60 and 55K. (Did they begin toiling the soil at that time?) Pygmies stayed dormant for another 30K years and then moved to the jungle to become short. But they still retained the same skull morphology. (Knowing that, say, Mongoloids are less than 10K old, a rather unique situation, isn't it?) Until 3K, however, the former brethren sharing the same L1 lineage didn't know of each other's existence. Then the Bantu developed agriculture and expanded widely, including some areas of the tropical forest, where they found the Pygmies who were now short. Pygmies lost their newly evolved language and adopted the Bantu language to restore the original situation in which the two populations spoke the dialects of the same language.

This is pure fantasy, Terry. This is Lord of the Rings. This is Harry Potter.

But if we drop the ridiculous dates that geneticists derive from the abstract chimp-human divergence point, not from the actual divergence rates within human families, clans and populations, we'll end up with a more realistic scenario. A population of proto-Niger-Congo (proto-Bantu, proto-Ubangian) speakers began expanding within the last 10K years. Smaller subpopulations migrated into the tropical forest, retained their foraging lifestyle and evolved short-stature, larger subpopulations first switched to agriculture and then re-entered the tropical forest at 3500K with a novel economy and a few novel mtDNA and Y-DNA lineages at hand. Sex-based asymmetric gene flow mixed up the older and the newer lineages across the subpopulations. Linguistic classification within Bantu (including Aka, and other Pygmy groups) mirrors genetic divergence within the original L1c clade.

"However even in the 'traditional' OoA it seems from Maju's work on mtDNA L that the Pymies were yet to become adapted to jungle when the Lines M and N emerged."

Total nonsense. Luis Aldamiz is a confused amateur. L lineages are interesting from the point of view of the divergences of incoming Eurasian populations within Africa but M and N lineages are much older than L lineages. M and N lineages are associated with 90% of world language diversity. L lineages are associated with just 10% of it. Linguistic methodology always begins at present and works itself back into the past ("reconstruction"). This has resulted in robust language classifications that cover the whole globe. Genetics is based on a putative species divergence moment (humans and African chimps) and then tries to interpret intraspecific population history in the last 100-50K years in this light. The results are utterly unreliable because convergence (homoplasy) are not well sorted out and the rates of change are unclear.

Maju said...

"Luis Aldamiz is a confused amateur".

In genetics, Dziebel, we are equally amateur. Your speciality is nothing but kinship anthropology and on the rest you are as qualified as I am (or probably much less).

Julien Riel-Salvatore said...

OK - there's been some interesting discussion on this topic, but unless comments are going to target the subject of the post, they will be not be posted. Specifically, personal attacks will not be tolerated.

JRS

terryt said...

"unless comments are going to target the subject of the post, they will be not be posted".

So back to the subject, 'Humans in the Philippines 67,000 years ago'.

Regardless of whether we consider the Philippines to be part of Wallacea or not we know that 'A biogeograpical line surely does separate the Philippines from Sundaland' (as Andrew Oh-Willeke said). Call it Huxley's Line if you must, or a branch of Wallace's Line if you wish, it seems likely that humans crossed to the Philippines via the Sulu Archipeligo. Perhaps by boat.

From the link in Julien's original blog:

"evidence of watercraft and adaptation to higher altitudes should receive greater consideration for reconstructions of modern human behavior, particularly in East Asia".

Note, 'particularly in East Asia'. For several reasons I believe it unlikely that boating goes way back in human prehistory. I can easily see it as having developed in the east, perhaps specifically in the Philippines. Seven thousand islands is a lot to experiment with.

The speciation of ancient humans on Flores argues against wider contact for that island, or ancient boating in the region. Neither Gibraltar nor the Bab al Mandab have neighbouring islands suitable for the opportunity for perfection of salt-water boating technology. Even in Crete the neighbouring islands remained uninhabited at the time. This is unlikely to have been the case had the humans that reached Crete done so through any sort of efficient boating ability.

terryt said...

Oh, the Komodo dragon provides another argument against the route to Australia as having been along the Sunda Islands. Wiki yet again:

http://en.wikipedia.org/wiki/Komodo_dragon

From the article:

"Their unusual size has been attributed to island gigantism, since there are no other carnivorous animals to fill the niche on the islands where they live.[4][5] However, recent research suggests that the large size of komodo dragons may be better understood as representative of a relic population of very large varanid lizards that once lived across Indonesia and Australia, most of which, along with other megafauna,[6] died out after contact with modern humans".

This strongly suggests that modern humans arrived in 'the Indonesian islands of Komodo, Rinca, Flores, and Gili Motang' relatively recently. If humans had been present on those islands since the Paleolithic the komodo dragon would surely not have survived there. I expect we'll now see some wierd and contorted explanation for how the dragon may have survived prolonged human presence only on those islands though.