New blogs

Leherensuge was replaced in October 2010 by two new blogs: For what they were... we are and For what we are... they will be. Check them out.

Sunday, March 7, 2010

Indonesian Y-DNA is mostly Paleolithic


Found
at Dienekes'.

Tatiana M. Karafet et al., Major East-West Division Underlies Y Chromosome Stratification Across Indonesia. MBE 2010. Pay per view.

While I don't have at the moment access to this surely interesting paper and hence to the likely substantial details, the abstract sounds highly interesting:

The early history of Island Southeast Asia is often characterized as the story of two major population dispersals: the initial Paleolithic colonization of Sahul ~45 thousand years ago and the much later Neolithic expansion of Austronesian-speaking farmers ~4,000 years ago. Here, in the largest survey of Indonesian Y chromosomes to date, we present evidence for multiple genetic strata that likely arose through a series of distinct migratory processes. We genotype an extensive battery of Y chromosome markers, including 85 SNPs/indels and 12 Y-STRs, in a sample of 1,917 men from 32 communities located across Indonesia. We find that the paternal gene pool is sharply subdivided between western and eastern locations, with a boundary running between the islands of Bali and Flores. Analysis of molecular variance reveals one of the highest levels of between-group variance yet reported for human Y chromosome data (e.g., ?ST = 0.47). Eastern Y chromosome haplogroups are closely related to Melanesian lineages (i.e., within the C, M and S subclades) and likely reflect the initial wave of colonization of the region, while the majority of western Y chromosomes (i.e., O-M119*, O-P203, and O-M95*) are related to haplogroups that may have entered Indonesia during the Paleolithic from mainland Asia. In addition, two novel markers (P201, P203) provide significantly enhanced phylogenetic resolution of two key haplogroups (O-M122, O-M119) that are often associated with the Austronesian expansion. This more refined picture leads us to put forward a four-phase colonization model in which Paleolithic migrations of hunter-gatherers shape the primary structure of current Indonesian Y chromosome diversity, and Neolithic incursions make only a minor impact on the paternal gene pool, despite the large cultural impact of the Austronesian expansion.


Bad news in any case for the Neolithicists, again confirmation for the Paleolithicists like myself. This in a context like SE Asia where it's been argued more than anywhere else in favor of a massive population replacement in Neolithic times (based on craniometric data mostly). It seems it was not the case after all.

I wonder how does Karafet and colleagues treat the issue of the proposed MNOPS macro-haplogroup downstream of K, or if just limit themselves to describe the sublineages (such as M, S and O) regardless of upstream affiliation.

Notice that the discovered divide goes exactly along Wallace line, a major ecological divide, which exists because the continent reached in the Ice Age, when sea levels were much lower, as far East as Bali and Borneo, while the islands east of that line remained separated.


Understanding the ecological divide at Indonesia (from Wikipedia)

This divide only makes sense if navigation was limited at the time and makes no sense whatsoever if one wants to explain it within the context of Austronesian expansion, because for these formidable oceangoing sailors, who colonized as far as Hawaii, Easter Island and Madagascar, such barrier would have never been any obstacle at all.


Update (March 10):

I got a copy of the paper (thanks to Argiedude again) and the four phases proposed are as follows:
  1. Early colonization reaching to Sahul. Lineages: C*, K*, C2, M and S. Dated to c. 45 Ka ago.
  2. Second flow affecting only Sundaland. Lineages: O1a1, O2a, O3(xO3a3) and O1a. Dated to 30-15 Ka.
  3. Austronesian colonization. Lineages: O1a2 and O3a3(xO3a3b). Dated to 4-3 Ka.
  4. Late arrivals from India, West Asia and China. Lineages: H, L, J and O3a3b.

Of course, the dates attached to the two Paleolithic phases are to be taken with some caution.

Some F* (East Indonesia) and R (West Indonesia) were also detected and should be associated to phases 1 and 4 respectively.

The remnants of the first phase are found particularly in East Indonesia, Melanesia, Polynesia (as partial genetic extension of Melanesia) and Philippines, where K* still constitutes almost 50% of Y-DNA. There's also some C* and K* in West Indonesia (and notably, different source, some 25% K* among the Orang Asli of peninsular Malaysia) but it's much more patchy and diluted.

Of the second phase, I'd read that O2a and O3 are arrivals from mainland SE Asia but that O1 may have coalesced at Sundaland and flowed northwards to as far as Taiwan, where O1a2 constitutes some 90% of the Aboriginal Y-DNA, being the main haplogroup that can be directly linked to the Austronesian expansion in phase 3.

___________

Addendum (March 12): map by Argiedude showing the frequencies of the lineages belonging to the first colonization wave of Karafet, here described as "Australasian lineages" and also showing current distribution of Austronesian and Papuan languages (the latter a paraphyletic group):



_________________

Important update (Mar 14): rare H and J lineages:

Thanks to the keen eye of Ebizur and Argiedude (see the discussion here and at Dienekes' blog), two rare Y-DNA lineages have been spotted in the Karafet samples: these are H(xH1,H2) and J(xJ1,J2).

H(xH1,H2) is found in Bali (17/641 = 2.65%) along with some H1 and H2 (one case of each). This paragroup seems rare even in India.

J(xJ1,J2) is found in Bali too (2/641 = 0.3%) along with some J1 (6/641) and J2b (4/641). It is also found in Vietnam (1/70 = 1.4%), along with some J2(xJ2b) (1/70), this last one also found in Java, Sulawesi, Vanuatu and Han Chinese. J* is, according to Argiedude not really found anywhere else.

Someone by the name Finn uploaded the paper and supplements at Zoho Viewer and you can (at least by the moment), find them at these links: main, PDF supp, XLS supp.

_____________

Update July 3: Natsuya has got some further details via private communication with Tatiana Karafet. Specifically the distribution of haplogroups among the four different Han samples and STR haplotype sequences for O3b3b-M7. LINK.


200 comments:

DocG said...

"This divide only makes sense if navigation was limited at the time and makes no sense whatsoever if one wants to explain it within the context of Austronesian expansion, because for these formidable oceangoing sailors, who colonized as far as Hawaii, Easter Island and Madagascar, such barrier would have never been any obstacle at all."

Excellent point, Maju. There are significant musical differences between East and West along the Wallace line as well. Do you think this supports the notion that the Sahul could have been populated, initially, as the result of an accidental event, rather than a purposeful sea voyage?

Maju said...

Hi, Victor. Glad to see you around.

"Excellent point, Maju".

It's Karafet's point, it seems, and it's like someone calling your attention to something that you should have realized long ago but never thought of it. It's a very good point, indeed, but not mine.

"Do you think this supports the notion that the Sahul could have been populated, initially, as the result of an accidental event, rather than a purposeful sea voyage?"

I'd say not. There's way too much top level diversity in Sahul and, in general, East of Wallace Line to be the product of a single random founder effect. I think these migrations were pre-planned voyages among people already used to the sea.

However the "voyages of discovery", going the first time of all beyond the horizon (and finding there was something there) was probably accidental, maybe because of storms or tsunamis. Some of these shipwreck survivors managed to guess how to get back home and did so. Eventually some people in that group (or a related one) decided to (organizedly) migrate to such a virgin land where life looked so promising, doing it when the weather and currents were most favorable.

I don't know how many migrations of this kind there were but probably not too many.

However the population of Sunda were not necessarily as seagoing as those of Wallacea. But they could well be more and better armed (just a guess). They surely knew of each other, even if in a blurry way, and that means that they knew that the other side was not anymore virgin land. Demic pressure alone (drift) would easily keep controlled the genetic influence whatever few migrants may have crossed the Wallace Line in later times.

They could obviously make the journey but it was not an easy one. That coupled with normal demographic pressure would keep the divide as we find it now. Without this major obstacle the flows may have been more significative, particularly I guess those from the Asian mainland, where demographic pressure surely mounted up more easily.

It's not the same to "defend a border", even my mere reproductive means, if this is a mere line on flat ground or if this is a major geographic barrier that is very difficult to cross.

Ebizur said...

Here are all the MNOPS*-M526 results from Karafet et al. (2010):

Mainland/Southeast Asia
0/48 Taiwanese Aboriginals
22/48 = 45.8% Philippines
0/70 Vietnam
2/32 = 6.3% Malaysia
0/165 China (Han)
0/58 China (Miao)
0/51 China (She)
0/49 China (Tujia)
0/60 China (Yao)

Western Indonesia
0/641 Bali/Balinese
2/61 = 3.3% Java/Dieng(n=35)+composite group(n=26)
5/38 = 13.2% Sumatra/Toba
5/86 = 5.8% Borneo
0/60 Nias/Gomo(n=47)+Hilitobara(n=13)
0/74 Mentawai

Eastern Indonesia
24/394 = 6.1% Flores/Bere Manggarai(n=11)+Boawae(n=27)+Cibol(n=55)+Rampasasa(n=95)+Seso Borowa(n=30)+Wogo(n=35)+Woloara(n=29)+Wolotopa(n=46)+Bama(n=49)+Bena(n=17)
4/54 = 7.4% Sulawesi/Mandar
14/350 = 4.0% Sumba/Anakalang(n=50)+Kodi(n=44)+Lamboya(n=49)+Loli(n=38)+Mamboro(n=53)+Rindi(n=27)+Wanokaka(n=52)+Wunga(n=37)
3/92 = 3.3% Lembata/Hadakewa(n=46)+Waipukang(n=46)
9/28 = 32.1% Alor
3/9 = 33.3% Timor
5/30 = 16.7% Moluccas

Oceania
6/44 = 13.6% Maewo (Vanuatu)
0/10 Nasioi (Bougainville)
1/15 = 6.7% Papua New Guinea coast
5/33 = 15.2% Papua New Guinea highland
7/16 = 43.8% Micronesia
0/6 American Samoa
0/10 Rapa Nui (also known as "Easter Island")
1/24 = 4.2% Tahiti
0/12 Tonga
0/12 Western Samoa

This team has found the highest frequencies of MNOPS*-M526 in their samples from the Philippines (22/48 = 45.8%), Micronesia (7/16 = 43.8%), Timor (3/9 = 33.3%), Alor (9/28 = 32.1%), and the Moluccas (5/30 = 16.7%).

The Y-STR variance associated with MNOPS*-M526 is highest in Oceania (0.898), followed by Western Indonesia (0.720), Eastern Indonesia (0.698), and finally Southeast Asia (0.381). The Y-STR variance associated with MNOPS-M526 as a whole (including known subclades) is highest in Oceania (0.982), followed by Eastern Indonesia (0.738), Southeast Asia (0.604), and finally Western Indonesia (0.594).

Also of note:

Bali
6/641 = 0.94% K4-P261
(Unfortunately, the authors have not made it clear whether K4-P261 belongs to MNOPS-M526.)

China (Han)
2/165 = 1.2% N*-M231(xN1-LLY22g)

Ebizur said...

More curious results:

Vietnam:
1/70 J*-M304(xJ1-M267, J2-M172)
1/70 J2-M172(xJ2b-M12)
(I mentioned on Dienekes' blog yesterday that Karafet et al. had reported finding one instance of J-12f2(xJ2-M172) and one instance of J2-M172 in their Vietnamese sample as early as 2001. Now it appears that the J-12f2(xJ2-M172) individual from Vietnam also has tested negative for M267, so he is one of those very rare J* people.)

Bali
17/641 = 2.65% H*-M69(xH1-M52, H2-Apt)
1/641 H2-Apt
1/641 H1-M52
2/641 J*-M304(xJ1-M267, J2-M172)
6/641 J1-M267
4/641 J2b-M12
13/641 L-M20
2/641 Q1a3-M346
11/641 R2-M124
2/641 R1b1-P25(xR1b1b2-M269)
2/641 R1b1b2-M269
9/641 R1a1a-M17

These results are incredibly weird. For one, H*-M69(xH1-M52, H2-Apt) is rare in modern Indian populations, probably no more frequent than it is in modern Balinese. Most present-day Indian members of haplogroup H belong to its subclade H1a-M82, with small numbers in H1*-M52(xH1a-M82) and H2-Apt, and practically no one in H*-M69. Despite this fact, H*-M69 seems to be by far the most common form of haplogroup H Y-DNA in the Balinese. Furthermore, most haplogroup J Y-DNA in modern Indians belongs to J2a-M410 and J2b2-M241, but J1-M267 is the most common form of haplogroup J in the Balinese. Karafet et al. also have found two Balinese who belong to J*(xJ1, J2) like that Vietnamese individual, but their Balinese sample contains no potential J2a-M410. WTH? I suppose we must be seeing the traces of some odd founder effects.

terryt said...

Thanks for all that effort Ebizur. Now Maju, do you believe that MNOPS may have arisen somewhere near Wallacea?

Maju said...

"Here are all the MNOPS*-M526 results from Karafet et al. (2010)"

You must mean K*. Karafet doesn't mention MNOPS at any moment.

You have also said in another discussion that this high result of K* in Philippines is not normal and that other authors find only much smaller amounts.

"Now Maju, do you believe that MNOPS may have arisen somewhere near Wallacea?"

In truth, I don't know. The results of different researchers (cf. Ebizur at Dienekes) are very different.

Whatever the case we are talking of K*, we don't know if it's MNOPS or what. Philippines is not Wallacea in any case. The next higher apportion of K* is among the Orang Asli of Malaysia.

Maju said...

Ok, I just noticed Ebizur's comment at Dienekes':

"OMG! I just noticed that Karafet et al. (2010) have retested their K* Filipinos for M526 and found them to be positive"...

That's curious and would confirm my earlier hypothesis that SE Asian and Oceanian K* is possibly all MNOPS.

Wallacean? IDK. SE Asian? Yes.

Maju said...

And also interesting the H* thing. However it's hard to explain why only in Bali.

Ebizur said...
This comment has been removed by the author.
Ebizur said...

Maju said,

"That's curious and would confirm my earlier hypothesis that SE Asian and Oceanian K* is possibly all MNOPS."

It does look likely that nearly all haplogroup K Y-DNA in Southeast Asia and Oceania belongs to the MNOPS-M526 subclade. The rare exceptions are some L-M20 in Bali (13/641) and possibly also some K4-P261 in Bali (6/641), though the authors have not clarified whether K4-P261 is a subclade of MNOPS-M526 or not.

Maju said,

"Philippines is not Wallacea in any case. The next higher apportion of K* is among the Orang Asli of Malaysia."

After the Philippines, Karafet et al. have reported finding the highest frequency of MNOPS*-M526 in Micronesia, followed by various parts of Wallacea in eastern Indonesia (Timor, Alor, and the Maluku Islands). However, the sample sizes for Micronesia and these particular eastern Indonesian locales are very small.

22/48 = 45.8% Philippines
7/16 = 43.8% Micronesia
3/9 = 33.3% Timor
9/28 = 32.1% Alor
5/30 = 16.7% Moluccas

Even assuming that the 4/17 = 23.5% H5 (=K-M9(xO3-M122, O1a-M119, O2a-M95, P-M45, M1-M5)) in Su Bing's sample of Orang Asli were all MNOPS*-M526, the Orang Asli would only fall between Timor-Alor and the Maluku Islands in regard to the frequency of MNOPS*-M526.

Maju said...

So you also think that "Wallacea" is at the origin of MNOPS...

Well...

I'm skeptic. Karafet does not suggest any backflow within the first migration, that would include all MNOPS except NO and P, as well as all C in that area.

However she might have not yet reached that stage of "revelation" about MNOPS. Maybe you or Terry should write to her.

Or maybe she finds the data inconclusive.

There is a lot of Y-DNA diversity in Melanesia but there is very limited evidence suggesting a back-migration.

And this stands particularly true re. mtDNA, of which the only reasonable back-migration is that of N (pre-R, pre-N1'5, pre-N2 and pre-X). Additionally you can consider very limited B and F in South Asia.

Ebizur said...

Maju said,

"So you also think that "Wallacea" is at the origin of MNOPS..."

Where have I said so? The only evidence that could possibly be interpreted to suggest a Wallacean origin of MNOPS-M526 is the finding of greatest Y-STR variance on an M526 background in Oceania (0.898 for MNOPS*-M526, 0.982 for MNOPS-M526 as a whole (including known subclades)), but note that it would then be logical to claim an "Oceanian" (i.e. New Guinean/Melanesian/Micronesian/Polynesian in the context of the present study) origin of MNOPS-M526 rather than a Wallacean origin (which would be "Eastern Indonesian" in the context of the present study). In any case, there is no evidence to suggest that either of the major Eurafrasian-American subclades of MNOPS-M526 (i.e. NO-M214 and P-M45) has derived from an MNOPS-M526 ancestor who has passed through Wallacea or Oceania. The new finding of MNOPS*-M526 in Maritime Southeast Asia and Oceania indicates only that some MNOPS-M526 people encountered the ocean at some point and possessed (or developed) the seafaring technology necessary to reach the islands of Wallacea and Oceania (including New Guinea). Another branch of MNOPS-M526 males may have ventured into the ocean at a different place, and perhaps they may have evolved into another descendant haplogroup (or haplogroups) of MNOPS in a different set of islands.

Maju said,

"And this stands particularly true re. mtDNA, of which the only reasonable back-migration is that of N (pre-R, pre-N1'5, pre-N2 and pre-X). Additionally you can consider very limited B and F in South Asia."

Wow. I have absolutely no idea how you have reached this conclusion.

Maju said...

"Wow. I have absolutely no idea how you have reached this conclusion".

I have discussed Eurasian mtDNA several times all in this blog (last time here, see also in general label=mtDNA).

And Terry and I have discussed somewhat about it before. He claims that mtDNA R must be "Wallacean", while I say that it's clearly South Asian in origin. I say that N is probably SE Asian and he says it's Siberian or something like that.

"Where have I said so?"

I thought you implied that with the data about some local Wallacean MNOPS* frequency. However I understand that the data you posted at the discussion at Dienekes, you mentioned highest MNOPS* diversity at Oceania, followed by West Indonesia (Sundaland) and only then East Indonesia (Wallacea).

I'm not sure what is what you mean, really.

"The only evidence that could possibly be interpreted to suggest a Wallacean origin of MNOPS-M526 is the finding of greatest Y-STR variance on an M526 background in Oceania (0.898 for MNOPS*-M526, 0.982 for MNOPS-M526 as a whole (including known subclades)), but note that it would then be logical to claim an "Oceanian" (i.e. New Guinean/Melanesian/Micronesian/Polynesian in the context of the present study) origin of MNOPS-M526 rather than a Wallacean origin (which would be "Eastern Indonesian" in the context of the present study)".

I am in agreement with this. And I have come to "hate Terry's Wallacea" so much that I'd strongly prefer that diversity favored a Melanesian coalescence area for MNOPS, even if I originally proposed Indochina or Sundaland (and I'm still inclined to favor that region on parsimony grounds).

It was long ago my first idea, not about MNOPS (not yet described) but K as a whole but I found opposition at Ren's forum (Ren himself in particular) and I was persuaded that such apparent "cul-de-sac" as is Melanesia could not be at the origin of the largest lineage of Eurasia, that there must be some other origin. I still think that Melanesia and Sahul in general is a demographic cul-de-sac but maybe there was a moment early on that it was not.

However if there's some evidence supporting a Sundaland/Indochina/South China origin for MNOPS, I'd rather favor it. Very specially because of mtDNA, also because I believe Terry's hypothesis about boats being invented precisely at Wallacea is a total nonsense and finally because of archaeological issues as well (no particular evidence of East Asian lithics migrating to South or West Asia in the MP or UP).

Unlike Terry and you I have been placing most of my attention to the more informative mtDNA and also archaeology. As Y-DNA can make in principle large sweeps that are not available for mtDNA, I'd rather trust the latter in case of "conflict". In general mtDNA is a more solid reference for the overall ancestry.

"he new finding of MNOPS*-M526 in Maritime Southeast Asia and Oceania indicates only that some MNOPS-M526 people encountered the ocean at some point and possessed (or developed) the seafaring technology necessary to reach the islands of Wallacea and Oceania (including New Guinea)".

What about C? C*, C2 and C4 are aboudant over there, and I'd dare say that the pattern of distribution of C over all Asia looks the most "coastal" of all three Eurasian Y-DNA macro-lineages (C1 in Japan, C5 along the Indian coasts and C3 in NE Asia but not deep inland into Siberia, as N or Q).

Ebizur said...

Of course, it is also possible that the MNOPS-M526 ancestor(s) of NO-M214 and P-M45 has never left dry land. For example, NO-M214 might have originated in an MNOPS-M526 male in mainland China, while P-M45 may have originated in an MNOPS-M526 male in India, and the Maritime Southeast Asian and Oceanian MNOPS-M526 people may all descend from a guy who has set out into the sea from Vietnam or wherever (though it is also possible that Maritime Southeast Asian and Oceanian MNOPS*-M526 may be paraphyletic).

Maju said...

Of course, Ebizur. One of the problems is to define well the phylogeny because it depends too much on "luck", that is: on whether researchers set themselves to find out about such matters, spending resources and time to find out maybe just one SNP... if they do at all.

I'd find easier to explain, to some extent at least, if all Oceanian MNOPS would not be paraphyletic but a true haplogroup ("MS-plus"). But we can wait for maybe other 10 years before this issue is clarified - if it ever is.

Whatever the case, I do agree that MNOPS looks original of SE Asia or Sahul. But I prefer what was "mainland SE Asia" (Sundaland would work well) in the Ice Age for parsimony reasons.

You have mentioned a high level of MNOPS* diversity in West Indonesia (former Sundaland), right?

Ebizur said...

Maju said,

"What about C? C*, C2 and C4 are aboudant over there, and I'd dare say that the pattern of distribution of C over all Asia looks the most "coastal" of all three Eurasian Y-DNA macro-lineages (C1 in Japan, C5 along the Indian coasts and C3 in NE Asia but not deep inland into Siberia, as N or Q)."

Karafet et al. (2010) have presented some interesting data regarding the distribution of C-RPS4Y711(xC2-M38, C3-M217) in Southeast Asia and Oceania. It occurs most frequently in Borneo and the Mentawai Islands in western Indonesia, in Flores, Lembata, Sulawesi, and Timor of eastern Indonesia, and in the Yao and Tujia peoples of Guangxi and Hunan in southern China. In Oceania, they have found C(xC2, C3) only in one individual from coastal Papua New Guinea (1/15) and in one individual from Micronesia (1/16), but, as you may see, the sample sizes are very small.

The STR variance associated with C-RPS4Y711(xC2-M38, C3-M217) is highest in Eastern Indonesia (0.451; n=143), greatly outstripping the STR variance associated with this group in Southeast Asia (0.179; n=17) and Western Indonesia (0.131; n=39). Likewise, the STR variance associated with C2-M38 is also highest in Eastern Indonesia (0.572; n=313) as I have described a couple weeks ago, followed by Oceania (0.377; n=57). However, the STR variance associated with C-RPS4Y711 as a whole is greatest in Southeast Asia (0.676; n=43), followed by Eastern Indonesia (0.650; n=456), Oceania (0.460; n=59), and finally Western Indonesia (0.165; n=40). The high STR variance of C-RPS4Y711 as a whole in Southeast Asians is probably a consequence of a very ancient divergence between the C3-M217 clade and the C-RPS4Y711(xC2-M38, C3-M217) clade(s) that together comprise the Southeast Asian pool of C-RPS4Y711 Y-DNA.

As for C5-M356, it is not particularly coastal in its distribution, with some of its highest frequencies I have seen reported so far being in the Hindus of Chitwan District, Nepal (1/26 = 3.8% C3-M217, 2/26 = 7.7% C5-M356; Fornarino et al. 2009) and in the Tibeto-Burman-speaking Newar of the Kathmandu Valley in Nepal (2/66 = 3.0% C5-M356; Gayden et al. 2007).

Haplogroup C3 is spread very widely throughout central, eastern, and southern Siberia, Central Asia, and East Asia, and it also has been found in some populations from the Americas, Southeast Asia, the sub-Himalayan region, Southwest Asia, and Europe. The only evidence that might be taken to suggest a coastal origin of C3-M217 is its reputedly high STR variance in the North China-Korea-Manchuria-Amur region.

Maju said...

What you say rather ratifies me in favor of a mainland SE Asian (Indochina) origin for Y-DNA C as a whole, regardless that it may have got a secondary spread center at Wallacea.

As for C5, I was under the strong impression that it's most commonly found in East, South and West India, i.e. "peninsular" India. I'm not sure that the Nepali cases overrule this. In any case a riverine route along the Ganges is also possible and it doesn't look at all like the whole haplogroup has any connection to or origins in Central Asia, as Terry has happily claimed sometimes, using AfPak C3 as reference without any justification (he prefers a North Asian origin for C for some odd reason).

Ebizur said...

I have mentioned several times already that Y-DNA haplogroup C is more common in the north of South Asia than in the south, and it seems that this pattern should hold for the subclade C5-M356, too, though SNP-tested cases of C5-M356 are rare in the literature.

For comparison with the Nepali data that I have already presented, here are some data from Sengupta et al. (2006) regarding the distribution of C5-M356 in India:

West India total (25 Konkanasth Brahmin + 20 Maratha + 14 Nav Buddha):
2/59 = 3.4% C5-M356

North India total (18 Chamar + 19 Muslim + 29 Rajput + 14 Uttar Pradesh Brahmin)
2/80 = 2.5% C5-M356 (1/18 Chamar, 1/29 Rajput)

East India total (18 West Bengal Brahmin + 7 Tanti + 14 Santal + 13 Mahishya + 20 Lodha + 30 Ho + 5 Gaud + 11 Bagdi + 10 Agharia):
2/128 = 1.6% C5-M356

India total
2/728 = 0.27% C*-M216
11/728 = 1.51% C5-M356

South India total (30 Irula + 27 Koya Dora + 16 Kota + 30 Konda Reddy + 19 Kurumba + 8 Toda + 29 Pallan + 25 Vanniyar + 31 Vellalar + 29 Ambalakarar + 30 Iyengar + 29 Iyer)
2/303 = 0.66% C*-M216
4/303 = 1.32% C5-M356

Northeast India tribes total (4 Chakma + 30 Jamatia + 5 Mog + 27 Mizo + 21 Tripuri):
1/87 = 1.1% C5-M356

Central India tribes total (21 Halba/Indo-European + 30 Kamar/Dravidian + 20 Muria/Dravidian):
0/71 C-M216

It is quite clear from various sources of data that haplogroup C5-M356 is most common in the north and west of the Indian subcontinent, as well as in Nepal (which is quite northerly). It also has been found in Central Asia (cf. Karafet et al. 2008), in the Arabian Peninsula (United Arab Emirates and Saudi Arabia; cf. Cadenas et al. 2008 and Abu-Amero et al. 2009), and probably also in Turkey (western part of the Black Sea Region and in Istanbul Province; cf. Cinnioglu et al. 2004). I suspect that some part of the C-M130(xC1-M8, C3-M217) Y-DNA found in the PRC (Mandarin-speaking Muslims, Uyghurs from Urumqi and Yili, etc.) is also C5-M356 (cf. Xue et al. 2006).

Maju said...

I don't think that frequency of C3 is relevant, what I care is about diversity, particularly about basal diversity (number of top level clades). And most basal C subhaplogroups are in the south: C2, C4, C5 and most of C*.

Considering that the haplogroup is scattered between Indonesia/Sahul (2 clades plus C*), NE Asia (2 clades) and South Asia (essentially 1 clade), I find that an origin point at SE Asia makes total sense.

Similarly nobody argues anymore for the origin of R or R1b in Atlantic Europe: no matter how frequent it's there (almost 100% in some populations), the basal diversity is rather low. Only for a downstream clade such as R1b1b2a1 can it be argued such thing.

So, unless you're going to unveil high levels of basal diversity for C around Amour river or whatever, I'll stay with the SEA origin theory.

Btw, do you have any idea about where is C6 found?

The low frequencies C5 in South Asia are probably determined by the fact that, if this lineage back-migrated from SEA, it must have found a rather crowded South Asia, without much room for its own expansion. In contrast, C1 and C3 in NE Asia, as well as C2 and C4 at Sahul, surely participated at the early colonization, what favored their numbers greatly.

But, at the end of the day, what we have is five clades that can be simplified to dots or colored areas on a map, plus C*: you join the dots and get an area of most likely origin, which IMO falls by SE Asia or South China. If Y-STR diversity also supports that, then even better.

terryt said...

"And also interesting the H* thing. However it's hard to explain why only in Bali".

Bali is Hindu whereas the rest of Indonesia is Muslim. Might be significant.

"For example, NO-M214 might have originated in an MNOPS-M526 male in mainland China, while P-M45 may have originated in an MNOPS-M526 male in India, and the Maritime Southeast Asian and Oceanian MNOPS-M526 people may all descend from a guy who has set out into the sea from Vietnam or wherever"

That might be possible apart from a coupe of things. One is that the basal mutation must have happened in one individual, somewhere, and your idea requires separate MNOPS individuals (or clans) moving huge distances. That's more likely if groups are moving into uninhabited regions, but we're talking back migrations here. The other obstacle is that, at face value, the maps at Dienekes have immediately MNOPS-derived Y-hap P through Southern and Western Indonesia (possibly indicating a westward movement) and Y-hap NO heading north through Borneo, Vietnam, Thailand and Korea, as far as Mongolia.

http://www.cell.com/current-biology/image/S0960-9822(09)02067-3?imageId=gr2&imageType=large

"I believe Terry's hypothesis about boats being invented precisely at Wallacea is a total nonsense"

Would you accept 'greatly improved'?

"I still think that Melanesia and Sahul in general is a demographic cul-de-sac but maybe there was a moment early on that it was not".

I think the evidence shows that quite clearly.

"The high STR variance of C-RPS4Y711 as a whole in Southeast Asians is probably a consequence of a very ancient divergence between the C3-M217 clade and the C-RPS4Y711(xC2-M38, C3-M217) clade(s) that together comprise the Southeast Asian pool of C-RPS4Y711 Y-DNA".

A phenomenon often ignored when claiming 'diversity' equals 'origin'.

"It is quite clear from various sources of data that haplogroup C5-M356 is most common in the north and west of the Indian subcontinent"

Suggesting it came in from the west, not the east.

"The low frequencies C5 in South Asia are probably determined by the fact that, if this lineage back-migrated from SEA"

IF.

Maju said...

"Bali is Hindu whereas the rest of Indonesia is Muslim. Might be significant"

That's the first thing we have all thought about. However most of Indonesia was Hindu before islamization and, as Ebizur says, this H* clade is not the most common in India either.

So there's some mystery there.

"One is that the basal mutation must have happened in one individual, somewhere, and your idea requires separate MNOPS individuals (or clans) moving huge distances".

There must be a difference of many many generations between the mutations defining MNOPS and nodes such as P or NO. The stem leading to P in special is extremely long.

So the MNOPS founder and the P founder should have been separated by lots of time. This also applies to the other clades (to some lesser extent - or is it that way only because they are less well researched?).

"The other obstacle is that, at face value, the maps at Dienekes have immediately MNOPS-derived Y-hap P through Southern and Western Indonesia (possibly indicating a westward movement) and Y-hap NO heading north through Borneo, Vietnam, Thailand and Korea, as far as Mongolia".

It's "Chiaroni's maps" (link), rather than "Dienekes'", right?

What you describe is the real migration, more or less. However I can't see any P map. There is an MNOPS map, an O map and an R map but I can't find any P map (logically because P is almost by default tested sloppily as P "except something very specific downstream", so we hardly know what is R, what Q and what something else in most cases).

Maybe there was a map of P earlier but the supplementary information is now corrected and the older link doesn't work anymore. Whatever the case, it's just a map, and unless you know what exactly is reflected in it and how accurately...

Wait... ah, that other map, you mean! That map is obviously wrong: there's not 50% F* in China anywhere, unless you are including O3 in it. I never paid any attention to that map because of such obvious and brutal error.

"Would you accept 'greatly improved'?"

I don't really know. I'm inclined to think that the people who crossed to Sahul were relatively seagoing people at the beginning. I have even pondered if they might have invented the outrigger long before the Austronesians made it famous.

But I don't see any particular reason to associate Y-DNA P with improved boating capabilities of any sort, so whatever extra navigational skills, they seem to have remained restricted to the East and seem best related with another macro-haplogroup: C, which MNOPS maybe copied the seagoing skills from.

You also have Y-DNA D at Japan and Andaman... it seems more of a regional skill/inclination/cultural trait (whose evidence is favored by the aboundance of islands over there anyhow) than anything related to any specific Y-DNA lineage.

Maju said...

"I think the evidence shows that quite clearly".

Not so clearly. Don't run so fast. We don't see Australian, Melanesian or Wallacean lineages at any significative frequency West of Wallace Line. It doesn't look like Sahulians took over the World, as you seem to claim. However relatives of some of them (of Melanesians specially, which are the ones with lots of MNOPS) did.

And, curiously enough Melanesians are the only ones with lots of mtDNA R (P specifically) in the region. So to me the issue seems most related to mtDNA R and it's less clear cut on the Y-DNA side and would relate IJK rather than strictly MNOPS (as IJK is what goes with mtDNA R in West Eurasia essentially).

"A phenomenon often ignored when claiming 'diversity' equals 'origin'".

Diversity must be treated with due care and focus on basal diversity and not just raw diversity (because a lineage with lots of derivatives will be necessarily more diverse than one with few). But basal diversity still supports a SE origin for Y-DNA C as far as I can see.

terryt said...

"I never paid any attention to that map because of such obvious and brutal error".

And there are horrendous errors in the Australian one as well. However Ebizur seems to accept P in SE Asia.

"The stem leading to P in special is extremely long".

But that doesn't mean it moved a huge distance over that time span. The fact that P is reasonably common in SE Asia in fact suggests it didn't move far. Just that it remained a private lineage for some time.

"I have even pondered if they might have invented the outrigger long before the Austronesians made it famous".

I doubt that the outrigger is paticularly old. And even the dugout may be not much more than 10,000 years old. In spite of your claims to the contrary making a dugout with stone tools and fire is a difficult process. Dugouts are quite often made around here these days, usually for ceremonial purposes rather than practical, and it's a major job, even with iron tools.

"I don't see any particular reason to associate Y-DNA P with improved boating capabilities of any sort"

I'm sure their expansion was not simply random though, and so has to be associated with something. That's the only thing I can think of. If you can come up with a better idea please let us know. Improved boating would certainly allow rapid expansion, and that seems to be what we see.

"seem best related with another macro-haplogroup: C, which MNOPS maybe copied the seagoing skills from".

Agree totally. C seems to be the first into Australia at least. And probaby, therefore, Wallacea. So must have had boats of some sort.

"it seems more of a regional skill/inclination/cultural trait (whose evidence is favored by the aboundance of islands over there anyhow) than anything related to any specific Y-DNA lineage".

There's a lot of truth in that. However any new technology probably usually starts off being handed from father to son. A 'meme', I think it is sometimes called. Of course eventually it achieves wider circulation as it becomes more widely known. And spreads further than the genes, or even the original haplogroup.

"It doesn't look like Sahulians took over the World, as you seem to claim".

Not 'Sahulians', 'Wallaceans'. There is a big difference, as you often point out. Sahulians in general kept moving east. All islands they could return to in the west were already inhabited. However on the mainland riverine habitats may have been relatively unexploited, so enabling a thin thread of expansion.

"However relatives of some of them (of Melanesians specially, which are the ones with lots of MNOPS) did".

But I strongly suspect that KMNOPS originated slightly further west than Melanesia. Guess where.

"And, curiously enough Melanesians are the only ones with lots of mtDNA R (P specifically) in the region".

I'm pleased that you see that now. Significant?

"basal diversity still supports a SE origin for Y-DNA C as far as I can see".

I don't think we're quite ready to examine that one yet. We still have some KMNOPS things to clear up.

Maju said...

"However Ebizur seems to accept P in SE Asia".

It's not what I've gathered (I quote from above: "P-M45 may have originated in an MNOPS-M526 male in India") but let's see what he has to say.

Personally I have absolutely no reason to think P as SE Asian. From what I've been able to gather P has a likely origin at South Asia (even Q looks "Iranian" - but uncertain)... and very few P is found further east, unless it's at NE Asia/America.

"But that doesn't mean it moved a huge distance over that time span".

But it allows it, as you need about one generation for each actual (not the same as known) Y-chr SNP to happen, maybe more.

That is a minimal time of 19 generations, which is about 600 years. Of course it was more, surely a lot more, time but in any case not anything instantaneous, as you seem to imply in your pseudo-reasonings.

"The fact that P is reasonably common in SE Asia in fact suggests it didn't move far".

It's not P(xQ,R) but, it seems, mostly R1a1 arrived from India recently (I think there are a few R-other and Q erratics but not many).

Instead, if you would have bothered even reading about haplogroup P at Wikipedia, you would have found that P(xQ,R) is rather common in Northern South Asia. Found at Orissa, Jarkhand, West Bengal, Nepal, Mizoram, Himachal, Andrah Pradesh, Gujarat and Maharastra.

Q and R are also very diverse in that region but with more westerly centroids.

"I doubt that the outrigger is paticularly old. And even the dugout may be not much more than 10,000 years old".

Can't say: this is a very theoretical discussion that can't be won in either direction without a time machine.

But what about leather canoes, similar to kayaks and those often made in Eurasia and America? They are lighter and require the same kind of technology you need to keep warm in cold weather: needles! Needles were known already by Kostenki people and are surely older.

What kind of craft do you think they used: mere rafts? Do you think they could get whole clans to cross Wallace and Liddeker lines with just that? I think they needed something more seagoing and the outrigger is a quite simple concept that greatly improves stability with or without hollowed out logs.

Maju said...

"I'm sure their expansion was not simply random though, and so has to be associated with something".

I'm sure about that but I have no clear idea of what that may be... except maybe the some sort of hurling weapon like the atlatl (but proto-Khoi used bow and arrow so meh), or some critically useful tool like the needle. It might have to do with sociology or whatever other subtle feature we can't grasp easily.

The case is that they did expand. And largely they did so overland.

"Improved boating would certainly allow rapid expansion"...

Through Central Asia? Skipping the Mediterranean islands?

Sorry but it doesn't fit well.

"However any new technology probably usually starts off being handed from father to son. A 'meme', I think it is sometimes called".

But technology, useful technology is quickly copied. Purely cultural features are more resistant to change but technology is primarily a practical matter, so if the neighbor does something better than you, you try to find how... normally it is as easy as just asking or watching.

"Not 'Sahulians', 'Wallaceans'. There is a big difference"...

There's some difference but Y-DNA C2 is what best defines Wallaceans genetically and it only expanded to the east of it (via New Guinea).

So same thing in the end.

"But I strongly suspect that KMNOPS originated slightly further west than Melanesia. Guess where".

I did the maths and my result is around Kuchin, Sarawak, Borneo island. That's my best guess: Sundaland.

""And, curiously enough Melanesians are the only ones with lots of mtDNA R (P specifically) in the region".

I'm pleased that you see that now. Significant?"

What have I been saying as of late about mtDNA R expanding eastwards with Y-DNA K (MNOPS)? Don't try to twist my discourse with cryptic meaningless questions.

""basal diversity still supports a SE origin for Y-DNA C as far as I can see".

I don't think we're quite ready to examine that one yet. We still have some KMNOPS things to clear up".

I have already gone through that: MNOPS coalesced most likely at Sundaland. I haven't worked out C to that detail but I'd say that SE Asia, maybe Sundaland too, where there's a lot of C*.

Time maybe for you to look to West Indonesia instead of being fixated with East Indonesia.

aargiedude said...

I agree with Ebizur about y-dna H. Here's the stats. Sengupta's Indo-Pakistani H samples are only 15% H(xH1,H2). Another big study, by Eaaswarkhanth, found only 10% H(xH1a,H2), and notice that he didn't test fo H1, so even that 10% might be further reduced. In contrast, the Indonesian H is 85% H(xH1,H2), or 17/20.

The only problem is that all 17 H(xH1,H2) are concentrated in Bali, which always draws considerations of some local founder effect or some other odd thing.

terryt said...

"What kind of craft do you think they used: mere rafts? Do you think they could get whole clans to cross Wallace and Liddeker lines with just that?"

Yes.

"Through Central Asia? Skipping the Mediterranean islands?"

I didn't realise there were no rivers or lakes in Central Asia. And if boating is so ancient how come the Mediterranean islands were not extensively settled long ago?

"I have already gone through that: MNOPS coalesced most likely at Sundaland"

Why do you consistently eliminate K from consideration? Is it because it would skew your results eastward?

Maju said...

"And if boating is so ancient how come the Mediterranean islands were not extensively settled long ago?"

It seems to me that because your mythical Wallaceans had not arrived, though Y-DNA MNOPS and mtDNA R had.

We are still waiting for them in fact. We need the Wallacean redemption, so we can go on vacation to Rhodes or Mallorca... meh!

Seriously: you know perfectly that Crete was colonized 130 Ka ago. So WTF!

"Why do you consistently eliminate K from consideration? Is it because it would skew your results eastward?"

It's because the haplogroup is called MNOPS, it seems. Even if it includes a lot of K*/K-number, it is not the same as K and it's more likely that these end up being MNOPS*/MNOPS-number than K, including L, T and MNOPS be renamed. ISOGG methodology rules and not your whim, luckily.

Mason said...

"Finn" is my previous account name :)

Anyway, good observations in this blog.

Best,
Natsuya

Mason said...

O1a2 isn't the majority of Taiwan aboriginals.

From data I saw before, Taiwan aborignal Y-DNA go like this:

O1a-M119* = 65.4%
O1a2-M110 = 16.7%
O3-M122* = 8.6%
O2a* = 1.9%
O2a1 = 2.5%

But the data was published some years ago. O1a1-P203 and O3a3-P201 are new found haplogroup. I assume that most of Taiwan aboriginal O1a*/O3* are actually O1a1/O3a3.

We shall know that Taiwan aboriginals are not homogenous. There are a least over 10 different populations of them, and all of them speak different languages.

The Amis tribe has 20% O3-M122*, which I suppose they're O3a3-P201.
The Bunun tribe has 58.8% O1a2 and 17.6% O2a1.
And Atayal tribe has 95.5% O1a-M119*, which I think they're mostly O1a1-P203.

http://web2.nmns.edu.tw/PubLib/NewsLetter/98/259/a-4.pdf

See the page 5 of the article, it's Chinese, but you can see haplogroup frequenies in 9 tribes of Taiwan aboriginals.

Mason said...

Another English paper referring to the origins of Island Southeast Asian Y-DNA, especially O1a-M119:

Paternal genetic affinity between western Austronesians and Daic
populations

Note: we have to know, when this paper was published, they had not found O1a1-P203 and O3a3-P201 yet, so the conclusions they made is not right. It's like we ignore the new downstream SNPs of R1b1b2, and compare all the STRs of different subclades, which is not very appropriate.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2408594/pdf/1471-2148-8-146.pdf

Maju said...

You seem to be right about Taiwanese O1a. I thought I was following Ebizur's posted data (at the Dienekes' discussion) but at some point I seem to have messed things.

This was probably because Karafet somehow considers O1a* mostly pre-Austronesian:

"However, sharing of 12-locus Y-STR haplotypes associated with P203 chromosomes was found between Taiwanese aboriginals and Indonesians (i.e., from Nias, Mentawai, Java, and Bali), while no such sharing was found between the Taiwanese aboriginal and mainland Southeast Asian P203 chromosomes in our sample. This suggests that some portion of Indonesian O-P203 chromosomes may have migrated from Taiwan".

I think I'm just getting overloaded with data and can't think straight anymore. I need to chew on this matter more calmly.

terryt said...

"It seems to me that because your mythical Wallaceans had not arrived, though Y-DNA MNOPS and mtDNA R had".

Doesn't make sense. If those haplogroups didn't actually originate in Wallacea they certainly originated close to it, especially KMNOPS.

"It's because the haplogroup is called MNOPS"

Where? Are you claiming that K is a totally separate haplogroup from MNOPS?

"This suggests that some portion of Indonesian O-P203 chromosomes may have migrated from Taiwan".

Almost certainly so, and makes perfect sense.

Maju said...

"If those haplogroups didn't actually originate in Wallacea they certainly originated close to it, especially KMNOPS".

Sure but maybe they were not as attached to boating as you claim.

As I have said before, the feat that deserves scrutiny and a good explainination is the clonization of Wallacea/Sahul.

But no matter how good they were at boating over there, such thing surely was not enough to give them an edge that would allow them to conquer the world (what, we all know, is the only worry of any forager clan: hunt to conquer!)

Maybe Asian superpowers, including their cousins of P and NO clans, had other strategical weapons such as foot massage (psychological weapon: let them relax and then cut their guts), greater apportion of triplets (you need large litters in order to conquest the world before others do), and needles, not just for sewing, but for torture sessions where to extract the know-how of every available technology from their enemies.

And maybe they kept memory of something more important than a mere canoe (oops, rusty raft, canoes need highly elaborated technology, I forgot): how to make fire with flint and yeast.

Of course, I'm half joking, being sarcastic but boats are just the tree, not the forest of Prehistory. You are so obsessed with the tree that...

"Where? Are you claiming that K is a totally separate haplogroup from MNOPS?"

Can you read what I say? It's its ancestor.

Ebizur said...

Maju said,

"You seem to be right about Taiwanese O1a. I thought I was following Ebizur's posted data (at the Dienekes' discussion) but at some point I seem to have messed things."

The only data regarding the Y-DNA of aboriginal populations of Taiwan that I have posted in the comments section of Dienekes' blog post about the recent article by Karafet et al. (2010) are the following:

Taiwanese aboriginals (Karafet et al. 2005)
1/48 = 2.1% C-RPS4Y711
3/48 = 6.3% O3-M122(xM134, LINE)
43/48 = 89.6% O1a-M119
1/48 = 2.1% O2a-M95

Taiwan aborigines (Hammer et al. 2006)
1/48 = 2.1% C3-M217(xC3c-M86)
3/48 = 6.3% O3-M122(xM134, LINE)
34/48 = 70.8% O1a-M119(xO1a2-M110)
9/48 = 18.8% O1a2-M110
1/48 = 2.1% O2a1-M111

Of course, both these sets of Y-DNA data have been derived from the same sample set of 48 Taiwan aborigines. According to the data provided by Hammer et al. (2006), only slightly over one fifth of all haplogroup O1a-M119 Y-DNA in Taiwan aborigines belongs to the subclade O1a2-M110. I never have claimed that a majority of Taiwan aborigines belong to O1a2-M110, but you may have been confused by a statement I have made in the past regarding the strong association between O1a2-M110 and speakers of Austronesian languages, which is correct. (For example, Karafet et al. (2010) have found 25/383 O1a1-P203, 1/383 O1a*-M119(xO1a1-P203, O1a2-M110), and 0/383 O1a2-M110 in their samples of non-Austronesian-speaking Han, Miao, She, Tujia, and Yao males from China. O1a2-M110 seems to be fairly strictly associated with Austronesian-speaking populations of Madagascar, Melanesia, and Maritime Southeast Asia, including Taiwan, the Philippines, and Indonesia, but Kayser et al. (2008) also have found O1a2-M110 in 1/100 males from Tuvalu in westernmost Polynesia.)

Anyway, here are all the Taiwan aborigine Y-DNA data I have encountered in the literature:

Taiwanese Aboriginals (Karafet et al. 2010; sample set is the same as that tested by Karafet et al. 2005 and Hammer et al. 2006)
1/48 = 2.1% C3-M217
3/48 = 6.3% O3a3-P201(xO3a3b-M7, O3a3c-M134)
34/48 = 70.8% O1a1-P203
9/48 = 18.8% O1a2-M110
1/48 = 2.1% O2a1-M111

Aboriginal Taiwanese (Tajima et al. 2004)
2/223 = 0.9% C3-M217
7/223 = 3.1% (N?)O-AS1(xO1a-M119, O3-M122)
28/223 = 12.6% O3-M122
186/223 = 83.4% O1a-M119

Atayal (Su Bing et al. 1999 & 2000)
7/24 = 29.2% H6(=O3-M122(xO3a3b-M7, O3a3c-M134))
1/24 = 4.2% H7(=O3a3b-M7)
1/24 = 4.2% H8(=O3a3c-M134)
13/24 = 54.2% H9(=O1a-M119(xO1a2-M50/M110/M103))
2/24 = 8.3% H10(=O1a2-M50/M110/M103)

Yami (Su Bing et al. 1999 & 2000)
2/8 = 25% H9(=O1a-M119(xO1a2-M50/M110/M103))
6/8 = 75% H11(=O2a-M95(xO2a1-M88/M111))
[Note that the Yami/Tao people are not really aboriginal inhabitants of the island of Taiwan/Formosa, but rather aboriginal inhabitants of Orchid Island/Lán Yǔ, which is a separate, small island currently under effective control of the ROC government.)

Paiwan (Su Bing et al. 1999 & 2000)
2/11 = 18.2% H8(=O3a3c-M134)
6/11 = 54.5% H9(=O1a-M119(xO1a2-M50/M110/M103))
3/11 = 27.3% H10(=O1a2-M50/M110/M103)

Ami (Su Bing et al. 1999 & 2000)
6/6 = 100% H9(=O1a-M119(xO1a2-M50/M110/M103))

Bunun (Su Bing et al. 2000)
1/9 = 11% H9(=O1a-M119(xO1a2-M50/M110/M103))
6/9 = 67% H10(=O1a2-M50/M110/M103)
2/9 = 22% H12(=O2a1-M88/M111)

Taiwan Aborigines total (24 Atayal + 8 Yami + 11 Paiwan + 6 Ami + 9 Bunun; Su Bing et al. 1999 & 2000)
7/58 = 12.1% H6(=O3-M122(xO3a3b-M7, O3a3c-M134)) [only in Atayal sample]
1/58 = 1.7% H7(=O3a3b-M7) [only in Atayal sample]
3/58 = 5.2% H8(=O3a3c-M134) [only in Paiwan and Atayal samples]
28/58 = 48.3% H9(=O1a-M119(xO1a2-M50/M110/M103))
11/58 = 19.0% H10(=O1a2-M50/M110/M103) [only in Bunun, Paiwan, and Atayal samples]
6/58 = 10.3% H11(=O2a-M95(xO2a1-M88/M111)) [only in Yami sample]
2/58 = 3.4% H12(=O2a1-M88/M111) [only in Bunun sample]

Ebizur said...

Taiwan (Hurles et al. 2005)
1/39 = 2.6% O-M175(xO1a-M119, O2-P31, O3-M122)
15/39 = 38.5% O1a-M119(xO1a1a-M101, O1a2-M50)
13/39 = 33.3% O1a2-M50
1/39 = 2.6% O2*-P31(xO2a-M95, O2b-SRY465)
2/39 = 5.1% O2a1-M88
5/39 = 12.8% O3-M122(xO3a3c-M134)
2/39 = 5.1% O3a3c-M134

Taiwan Aborigines (Kayser et al. 2008)
2/41 = 4.9% O-M175(xO1a-M119, O2a-M95, O3-M122)
18/41 = 43.9% O1a-M119(xO1a2-M110)
14/41 = 34.1% O1a2-M110
4/41 = 9.8% O3a-M324(xO3a3b-M7, O3a3c-M134)
1/41 = 2.4% O3a3c-M134
2/41 = 4.9% O2a-M95

Amis (Hui Li et al. 2008)
2/28 = 7.1% O-M175(xO1a-M119, O2a-M95, O3-M122)
12/28 = 42.9% O1a-M119(xO1a2-M110)
5/28 = 17.9% O1a2-M110
2/28 = 7.1% O2a-M95(xO2a1-M88/M111)
6/28 = 21.4% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)
1/28 = 3.6% O3a3c-M134(xO3a3c1-M117)

Pazeh (Hui Li et al. 2008)
3/21 = 14.3% K-M9(xM1-M5, O-M175, P-M45)
8/21 = 38.1% O1a-M119(xO1a2-M110)
4/21 = 19.0% O1a2-M110
3/21 = 14.3% O2a-M95(xO2a1-M88/M111)
3/21 = 14.3% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)

Makatao (Hui Li et al. 2008)
1/37 = 2.7% C-M130
1/37 = 2.7% K-M9(xM1-M5, O-M175, P-M45)
2/37 = 5.4% O-M175(xO1a-M119, O2a-M95, O3-M122)
26/37 = 70.3% O1a-M119(xO1a2-M110)
2/37 = 5.4% O2a-M95(xO2a1-M88/M111)
5/37 = 13.5% O3a3c-M134(xO3a3c1-M117)

Thao (Hui Li et al. 2008)
1/22 = 4.5% K-M9(xM1-M5, O-M175, P-M45)
18/22 = 81.8% O1a-M119(xO1a2-M110)
1/22 = 4.5% O1a2-M110
2/22 = 9.1% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)

Paiwan (Hui Li et al. 2008)
14/22 = 63.6% O1a-M119(xO1a2-M110)
6/22 = 27.3% O1a2-M110
2/22 = 9.1% O3a3c-M134(xO3a3c1-M117)

Atayal (Hui Li et al. 2008)
21/22 = 95.5% O1a-M119(xO1a2-M110)
1/22 = 4.5% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)

Rukai (Hui Li et al. 2008)
9/11 = 81.8% O1a-M119(xO1a2-M110)
2/11 = 18.2% O1a2-M110

Pyuma (Hui Li et al. 2008)
8/11 = 72.7% O1a-M119(xO1a2-M110)
1/11 = 9.1% O1a2-M110
1/11 = 9.1% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)
1/11 = 9.1% P-M45

Tsou (Hui Li et al. 2008)
16/18 = 88.9% O1a-M119(xO1a2-M110)
1/18 = 5.6% O1a2-M110
1/18 = 5.6% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)

Bunun (Hui Li et al. 2008)
1/17 = 5.9% K-M9(xM1-M5, O-M175, P-M45)
3/17 = 17.6% O1a-M119(xO1a2-M110)
10/17 = 58.8% O1a2-M110
3/17 = 17.6% O2a1-M88/M111

Saisiyat (Hui Li et al. 2008)
5/11 = 45.5% O1a-M119(xO1a2-M110)
1/11 = 9.1% O1a2-M110
1/11 = 9.1% O2a-M95(xO2a1-M88/M111)
1/11 = 9.1% O2a1-M88/M111
3/11 = 27.3% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)

Taiwan aborigine total (28 Amis + 21 Pazeh + 37 Makatao + 22 Thao + 22 Paiwan + 22 Atayal + 11 Rukai + 11 Pyuma + 18 Tsou + 17 Bunun + 11 Saisiyat; Hui Li et al. 2008)
1/220 = 0.45% C-M130 [only in Makatao sample]
6/220 = 2.73% K-M9(xM1-M5, O-M175, P-M45) [only in Pazeh, Bunun, Thao, and Makatao samples]
4/220 = 1.82% O-M175(xO1a-M119, O2a-M95, O3-M122) [only in Amis and Makatao samples]
140/220 = 63.64% O1a-M119(xO1a2-M110)
31/220 = 14.09% O1a2-M110
8/220 = 3.64% O2a-M95(xO2a1-M88/M111) [only in Pazeh, Saisiyat, Amis, and Makatao samples]
4/220 = 1.82% O2a1-M88/M111 [only in Bunun and Saisiyat samples]
17/220 = 7.73% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)
8/220 = 3.64% O3a3c-M134(xO3a3c1-M117) [only in Makatao, Paiwan, and Amis samples]
1/220 = 0.45% P-M45 [only in Pyuma sample]

Ebizur said...

Paiwan (Capelli et al. 2001)
51/53 = 96.2% Haplogroup H(=O1a-M119)
2/53 = 3.8% Haplogroup I(=O-M175(xO1a-M119, O2a-M95, O3-M122))

Bunun (Capelli et al. 2001)
12/50 = 24.0% Haplogroup G(=O2a-M95)
37/50 = 74.0% Haplogroup H(=O1a-M119)
1/50 = 2.0% Haplogroup L(=O3-M122)

Atayal (Capelli et al. 2001)
49/50 = 98.0% Haplogroup H(=O1a-M119)
1/50 = 2.0% Haplogroup L(=O3-M122)

Ami (Capelli et al. 2001)
3/53 = 5.7% Haplogroup G(=O2a-M95)
23/53 = 43.4% Haplogroup H(=O1a-M119)
2/53 = 3.8% Haplogroup I(=O-M175(xO1a-M119, O2a-M95, O3-M122))
25/53 = 47.2% Haplogroup L(=O3-M122)

Yami (Capelli et al. 2001)
1/40 = 2.5% Haplogroup C(=C-RPS4Y711)
11/40 = 27.5% Haplogroup G(=O2a-M95)
28/40 = 70.0% Haplogroup H(=O1a-M119)

Taiwan Aborigines total (53 Paiwan + 50 Bunun + 50 Atayal + 53 Ami + 40 Yami; Capelli et al. 2001)
1/246 = 0.41% Haplogroup C(=C-RPS4Y711) [only in Yami sample]
26/246 = 10.57% Haplogroup G(=O2a-M95) [only in Yami, Bunun, and Ami samples]
188/246 = 76.42% Haplogroup H(=O1a-M119)
4/246 = 1.63% Haplogroup I(=O-M175(xO1a-M119, O2a-M95, O3-M122))
27/246 = 10.98% Haplogroup L(=O3-M122) [only in Ami, Atayal, and Bunun samples]

Maju said...

Thanks for all that data, Ebizur. As I said, it was me who got things messed up, surely because of the claim by Karafet of only O1a2 (and some O1a* by haplotype) being truly derived from Taiwan (Asutronesian).

What I find more curious is that Karafet also claims O3a3(xO3a3b) to be Austronesian but all the O3 in TAs is like 10%, while among Indonesians and Filipinos (the other samples) it includes a much larger apportion of the Y-DNA. This is something that I don't understand, really.

terryt said...

"Maybe Asian superpowers, including their cousins of P and NO clans, had other strategical weapons such as foot massage"

You continue being totally stupid. Your choice of course.

"Can you read what I say? It's its ancestor".

It's part of the same haplogroup. K* is the ancestor but K1, K2, K3, K4 (or whatever you wish to call them), M, NO, P and S are equally part of the haplogroup.

"Karafet also claims O3a3(xO3a3b) to be Austronesian but all the O3 in TAs is like 10%, while among Indonesians and Filipinos (the other samples) it includes a much larger apportion of the Y-DNA. This is something that I don't understand, really".

Particular haplogroups are not necessarily closely associated with particular languages. What seems to have happened during the Austronesian expansion is that people on several different islands and neighbouring regions adopted the language (and presumably the associated improvements in technology) and then expanded in various directions themselves. It's impossible to say that a particular haplogroup is the exclusive marker of the Austronesians. It varies from place to place and presumably included several hapolgroups in each of those places.

Maju said...

"It's part of the same haplogroup. K* is the ancestor but K1, K2, K3, K4 (or whatever you wish to call them), M, NO, P and S are equally part of the haplogroup".

1. K* is not the ancestor, K* is a paragroup that includes all K-other lineages (and which can vary in extension depending on what SNPs are tested for).

2. K also includes L and T, besides the lineages you mentioned and whatever K* not included in them.

So K is (or was) the ancestor of MNOPS plus other lineages, including L, T and all other K, whether it's part of MNOPS or not. K is a set of ancestrally related Y-DNA lineages, a haplogroup.

Sometimes you seem to fail at really basic understanding of phylogeny and nomenclature issues. I want to believe you don't do that on purpose but you really need to improve that.

...

Re. Austronesian lineages, your answer is not helpful to pinpoint the origins of that particular haplogroup nor how it got involved as main actor within Austronesian expansion, which is what I would like to know.

Mason said...

Ebizur, I too appreciate your collection of Taiwan aboriginal Y-DNA data, very interesting.

Maju, although O3 is relatively lower than O1a among Taiwan aboriginals regarded as a whole.

But the Amis tribes have the highest O3 (20%~50%, and they could be mostly O3a3-P201) among all the tribes.

The Amis tribes are located along the eastern coast of Taiwan. It has been proposed that the ancestors of Amis were the main "out of Taiwan" seafaring farmers sailing southward to ISEA thousands of years ago.

Interestingly, some of Amis and other eastern coast tribes have similar oral traditions that mention their ancestors came from a southern island called "Sanasai".

I was thinking that the higher O3 among Amis may have been the influencies from the Philippines, and also that Amis and Filipinos both have high mtDNA B4a1a. Now I'm confused that where the origin of O3a3-P201 in Taiwan and ISEA is. Is it Taiwan or ISEA? I believe it's key to find out the origin of Austronesian languages.

Marie Lin and Jean Trejaut's team (from Mackay Memorial Hospital of Taiwan) is working on Taiwan aboriginal Y-DNA. There'll be a paper to be published soon, I suppose.

Best,
Natsuya

Maju said...

"Now I'm confused that where the origin of O3a3-P201 in Taiwan and ISEA is. Is it Taiwan or ISEA?"

According to Karafet's proposal it would be Austronesian-specific, taking part in that colonization. But maybe she's wrong.

Whatever the case, I just see no correspondence between the ISEA apportions of O3a3/O1a2 and the same apportion among Taiwan Aborigines, even in the most favorable tribal case. So either there was a marked founder effect in favor of O3a3, this haplogroup was borrowed from somewhere else (China?) or Karafet is wrong in this.

Maybe the origin of Austronesians was at ISEA? Philippines? Can't say.

Mason said...

From the paper:

"To explore the potential root of O-P201 chromosomes in Indonesia we calculated genetic distances (RST) based on O-P201* STR-haplotypes (supplementary table S4). The divergence between Filipinos/Taiwanese aborigines and Indonesians was insignificant and 10-fold less than that between Southeast Asians and Indonesians (0.027, P = 0.11 versus 0.349, P = 0.00). Genetic distances between Oceania and Philippines/Taiwanese aborigines were even lower (0.007, P = 0.35)."

"Importantly, despite its relatively low frequency on Taiwan (~6%), genetic distances based on STR variation associated with P201 chromosomes reveal a much closer relationship among Taiwanese aboriginals/Filipinos, Indonesians, and Oceanians than between any of these groups and mainland Southeast Asians (supplementary table S4). Therefore, we hypothesize that this new marker traces the large population expansion associated with the spread of Austronesian languages and culture."

Now the question is that we don't know O3a3-P201 went from Taiwan to ISEA, or ISEA to Taiwan.

Maju said...

Thanks for the quotation: that's it. I wish they'd use at least the first number of the ISOGG nomenclature because I tend to get messed with mutation names (which are meaningless in my brain) and, in this case, I was all the time thinking they meant O1a1-P203 and not O3a3-P201.

"Now the question is that we don't know O3a3-P201 went from Taiwan to ISEA, or ISEA to Taiwan".

IMO, from the higher frequency area to the lesser frequency one, unless diversity suggests otherwise. We're talking of post-Neolithic migrations, not Paleolithic founder effects on virgin lands.

However they might have taken it from some other area, I'd scan the Fujian area in SE China for possible matches, also all around (Gwandong, Ryu-Kyu...).

Luzon (Philippines) looks like a nice candidate anyhow for the admixture. Maybe the earliest Austronesians from Taiwan mixed there with some other group, native to ISEA (?), which adopted their language in form of Malayo-Polynesian and expanded it around.

Ebizur said...

There have been few reports of finding haplogroup O1a2-M50/M103/M110 Y-DNA in samples of populations outside of Taiwan, Maritime Southeast Asia (i.e. the Philippines, Indonesia, Malaysia), Melanesia, and Madagascar. The following are all the cases of which I am aware:

Austronesian speakers, but outside of the main geographical range of haplogroup O1a2
Tuvalu (Kayser et al. 2008)
17/100 = 17% C2a-M208
2/100 = 2% F-M89(xK-M9)
8/100 = 8% K-M9(xK3-P79, M1-M4, M2-M353, M3-P117, NO-M214, P-M74, S-M230)
28/100 = 28% K3-P79
1/100 = 1% O1a2-M110
41/100 = 41% O3a-M324(xO3a3b-M7, O3a3c-M134)
3/100 = 3% O2a-M95

Majuro (Su Bing et al. 2000)
[Majuro is the main atoll/capital of the Republic of the Marshall Islands in eastern Micronesia, just NNW of the Gilberts, which are, in turn, just NNW of Tuvalu.]
1/9 = 11% H4 (=F-M89(xK-M9))
6/9 = 67% H5 (=K-M9(xM1-M5, O1a-M119, O2a-M95, O3-M122, P-M45))
2/9 = 22% H10 (=O1a2-M50/M103/M110)

Majuro (Hurles et al. 2005)
1/11 = 9.1% F-M89(xJ-12f2.1, K-M9)
1/11 = 9.1% J-12f2.1
7/11 = 63.6% K-M9(xK1-M147, L-M61, M1-M4/M5, M2a-SRY9138, NO-M214, P-P27/M45, T-M70)
2/11 = 18.2% O1a2-M50

Kra-Dai speakers from southern China and Southeast Asia
Kam(=Dong) (XIE Xuan-Hua, LI Hui, MAO Xian-Yun et al., "Genetic Structure of Tujia as Revealed by Y Chromosomes," Acta Genetica Sinica, October 2004, 31 (10) : 1023-1029)
3/20 = 15% H1 (=Y*(xDE-DYS287, F-M89))
3/20 = 15% H6 (=O3-M122(xO3a3b-M7, O3a3c-M134))
3/20 = 15% H8 (=O3a3c-M134)
5/20 = 25% H9 (=O1a-M119(xO1a2-M50/M110/M103))
2/20 = 10% H10 (=O1a2-M50/M110/M103)
4/20 = 20% H11 (=O2a-M95(xO2a1-M88/M111))

Dong (Su Bing et al. 1999 & 2000)
2/10 = 20% H1 (=Y*(xDE-DYS287, F-M89))
1/10 = 10% H6 (=O3-M122(xO3a3b-M7, O3a3c-M134))
2/10 = 20% H8 (=O3a3c-M134)
2/10 = 20% H9 (=O1a-M119(xO1a2-M50/M110/M103))
1/10 = 10% H10 (=O1a2-M50/M110/M103)
2/10 = 20% H11 (=O2a-M95(xO2a1-M88/M111))

Northeastern Thai (Su Bing et al. 1999 & 2000)
1/20 = 5% H4 (=F-M89(xK-M9))
1/20 = 5% H5 (=K-M9(xM1-M5, O1a-M119, O2a-M95, O3-M122, P-M45))
1/20 = 5% H6 (=O3-M122(xO3a3b-M7, O3a3c-M134))
1/20 = 5% H7 (=O3a3b-M7)
1/20 = 5% H9 (=O1a-M119(xO1a2-M50/M110/M103))
1/20 = 5% H10 (=O1a2-M50/M110/M103)
9/20 = 45% H11 (=O2a-M95(xO2a1-M88/M111))
4/20 = 20% H12 (=O2a1-M88/M111)
1/20 = 5% H14 (=P-M45(xQ1a1-M120, Q1a3a-M3, R1a1a-M17))

Mollao (Li Hui et al. 2008)
3/30 = 10.0% C-M130
1/30 = 3.3% K-M9(xM1-M5, O-M175, P-M45)
4/30 = 13.3% O-M175(xO1a-M119, O2a-M95, O3-M122)
1/30 = 3.3% O1a-M119(xO1a2-M110)
1/30 = 3.3% O1a2-M110
19/30 = 63.3% O2a-M95(xO2a1-M88/M111)
1/30 = 3.3% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)

Zhuang, North (Li Hui et al. 2008)
3/22 = 13.6% O-M175(xO1a-M119, O2a-M95, O3-M122)
1/22 = 4.5% O1a2-M110
16/22 = 72.7% O2a-M95(xO2a1-M88/M111)
1/22 = 4.5% O3a1-M121
1/22 = 4.5% O3a3c1-M117

Mulam (Li Hui et al. 2008)
1/40 = 2.5% C-M130
5/40 = 12.5% D1-M15
3/40 = 7.5% F-M89(xK-M9)
2/40 = 5.0% K-M9(xM1-M5, O-M175, P-M45)
2/40 = 5.0% O1a-M119(xO1a2-M110)
10/40 = 25.0% O1a2-M110
12/40 = 30.0% O2a-M95(xO2a1-M88/M111)
3/40 = 7.5% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)
2/40 = 5.0% O3a3c-M134(xO3a3c1-M117)

Ebizur said...

Mon-Khmer speakers from southern China
Bolyu (Li Hui et al. 2008)
1/30 = 3.3% DE-YAP(xD1-M15)
1/30 = 3.3% K-M9(xM1-M5, O-M175, P-M45)
3/30 = 10.0% O-M175(xO1a-M119, O2a-M95, O3-M122)
3/30 = 10.0% O1a-M119(xO1a2-M110)
1/30 = 3.3% O1a2-M110
7/30 = 23.3% O2a-M95(xO2a1-M88/M111)
9/30 = 30.0% O3-M122(xO3a1-M121, O3a3b-M7, O3a3c-M134)
2/30 = 6.7% O3a3c-M134(xO3a3c1-M117)
3/30 = 10.0% O3a3c1-M117

Probably Mon-Khmer speakers from Southeast Asia, but possibly having some Chamic Austronesian or Southern Daic ancestry
Cambodian (Su Bing et al. 1999 & 2000)
1/26 = 3.8% H1 (=Y*(xDYS287, M89))
1/26 = 3.8% H3 (=D1-M15)
3/26 = 11.5% H4 (=F-M89(xK-M9))
3/26 = 11.5% H5 (=K-M9(xO3-M122, O1a-M119, O2a-M95, P-M45, M1-M5))
1/26 = 3.8% H6 (=O3-M122(xO3a3b-M7, O3a3c-M134))
4/26 = 15.4% H8 (=O3a3c-M134)
1/26 = 3.8% H9 (=O1a-M119(xO1a2-M50/M110/M103))
1/26 = 3.8% H10 (=O1a2-M50/M110/M103)
6/26 = 23.1% H11 (=O2a-M95(xO2a1-M88/M111))
3/26 = 11.5% H12 (=O2a1-M88/M111)
1/26 = 3.8% H14 (=P-M45(xQ1a1-M120, Q1a3a-M3, R1a1a-M17))
1/26 = 3.8% H16 (=R1a1a-M17)

If the findings of O1a2-M110 in Kra-Dai-speaking populations of both southern China and Southeast Asia (Northeastern Thailand in this case) are accurate, they may be considered evidence to support the hypothesis of Kra-Dai being the nearest sister taxon (in the stammbaum paradigm) or the nearest neighbor (in the sprachbund paradigm) to Austronesian. The Bolyu and Cambodian individuals may reflect Kra-Dai or Austronesian influence, respectively.

Maju said...

"If the findings of O1a2-M110 in Kra-Dai-speaking populations of both southern China and Southeast Asia (Northeastern Thailand in this case) are accurate, they may be considered evidence to support the hypothesis of Kra-Dai being the nearest sister taxon (in the stammbaum paradigm) or the nearest neighbor (in the sprachbund paradigm) to Austronesian. The Bolyu and Cambodian individuals may reflect Kra-Dai or Austronesian influence, respectively".

Sounds plausible.

Regardless of language, what I would infer is a possible coalescence of O1a2 in that area of Southern China, around Ghizou-Hunan-Guangxi, which may be the heartland of the Kradai super-ethnicity. Hunan is also the abode of East Asian Neolithic, what may be a better explanation for such minor presence among Austroasiatics (probably along with other lineages I am not able to identify right now).

Maju said...

It'd be interesting to compare with SE Chinese, who must retain at least a big deal of pre-Sinitic genetics and may be (I guess) a missing link in all this process.

terryt said...

"K is a set of ancestrally related Y-DNA lineages, a haplogroup".

Yes, and K1, K2, K3 and K4 are members of the KMNOPS haplogroup.

"Sometimes you seem to fail at really basic understanding of phylogeny and nomenclature issues".

And you alter your understanding depending on what point of view you are supporting at the time. Some time back you insisted that K1, K2, etc. were simply regional variants that had formed along with M, S, NO and P. Now that that claim conflicts with the belief you're currently supporting you are pushing the idea that somehow K1, K2, K3 and K4 belong to a group separate from MNOPS.

"Maybe the origin of Austronesians was at ISEA? Philippines? Can't say".

The Austronesian langiages in Taiwan appear to belong to two groups. Ayatalic is close to the Austronesian root, but other languages are closely related to Indonesian and Philippine languages. This strongly suggests there was a back-migration to Taiwan, probably from the Philippines, early on during the establishment of the Austronesian language. So we'd get a very varied mix of haplogroups from a very early stage. Therefore we really have no way of knowing if 'O3a3-P201 went from Taiwan to ISEA, or ISEA to Taiwan'. I have a diagram of the phylogeny of Austronesian languages in a book. I'll try to find a similar one on the net.

"they may be considered evidence to support the hypothesis of Kra-Dai being the nearest sister taxon (in the stammbaum paradigm) or the nearest neighbor (in the sprachbund paradigm) to Austronesian".

That used to be the accepted position. Vietnamese and Khmer were next closest neighbour, followed by Miao-Yao, and all were considered even more distantly related to Tibetan and Northern Chinese languages. But these days people are less ready to accept ancient language relationships. But the originally accepted pattern of language development suggests a southward movement.

"Hunan is also the abode of East Asian Neolithic, what may be a better explanation for such minor presence among Austroasiatics (probably along with other lineages I am not able to identify right now)".

Probably most of the O haplogroup, if the movement south implied by the old considerations concerning the language relationships is correct.

Ebizur said...

Maju said,

"Regardless of language, what I would infer is a possible coalescence of O1a2 in that area of Southern China, around Ghizou-Hunan-Guangxi, which may be the heartland of the Kradai super-ethnicity."

I don't know why you would infer such a thing from the data I have posted. Those are the erratic cases of haplogroup O1a2; the majority of extant haplogroup O1a2 individuals reside in Maritime Southeast Asia and speak Malayo-Polynesian languages. Most of the rest also speak Malayo-Polynesian (Malagasy, Marshallese, etc.) or other Austronesian (Bunun, Paiwan, etc.) languages. The correlation between O1a2 and Kra-Dai languages is very weak in comparison to the correlation between O1a2 and Austronesian languages, but the fact that O1a2 has been found sporadically in Kra-Dai-speaking ethnic groups of southern China and also in the northeastern Thai is interesting because it suggests that there has been some sort of interaction between the ancestors of many Austronesian-speaking ethnic groups and the ancestors of many Kra-Dai-speaking ethnic groups in a relatively recent (e.g. post-Neolithic) timeframe.

Maju said,

"Hunan is also the abode of East Asian Neolithic, what may be a better explanation for such minor presence among Austroasiatics (probably along with other lineages I am not able to identify right now)."

The only case of O1a2 that I know to be certainly associated with an Austroasiatic-speaking ethnic group is the Bolyu O1a2 individual reported by Li et al. (2008). As far as I know, the Cambodian O1a2 individual reported by Su et al. (1999 & 2000) is not necessarily a speaker of Khmer or any other Austroasiatic language; Chamic Austronesian-speaking people reside in Cambodia even at present, and their influence in southern Indochina has been very great historically, though speakers of Chamic languages are now minorities even in their historical homelands. Even if the "Cambodian" sample of Su Bing's team were composed solely of individuals who presently speak an Austroasiatic language (most likely Khmer), there are plenty of historically plausible scenarios that may account for the presence of an originally Austronesian or Kra-Dai Y-DNA lineage in an Austroasiatic-speaking inhabitant of modern Cambodia. The lack of O1a2 (or O1a of any sort) in most Austroasiatic-speaking populations even in southern China makes it quite unlikely that haplogroup O1a2 was present in the proto-Austroasiatic-speaking ancestral population.

cf. YANG Zhili, DONG Yongli, GAO Lu et al., "The distribution of Y chromosome haplogroups in the nationalities from Yunnan Province of China," Annals of Human Biology, January–February 2005:

Bulang (Shuangjiang, Yunnan)
2/28 = 7.1% C-M130
1/28 = 3.6% F-M89(xK-M9)
5/28 = 17.9% K-M9(xO1a-M119, O2a-M95, O3-M122, P-M45)
2/28 = 7.1% O3-M122(xO3a3b-M7, O3a3c-M134)
4/28 = 14.3% O3a3c-M134
14/28 = 50.0% O2a-M95(xO2a1-M88)

De'ang (Luxi, Yunnan)
2/16 = 12.5% F-M89(xK-M9)
3/16 = 18.8% K-M9(xO1a-M119, O2a-M95, O3-M122, P-M45)
2/16 = 12.5% O3-M122(xO3a3b-M7, O3a3c-M134)
7/16 = 43.8% O3a3c-M134
2/16 = 12.5% O2a-M95(xO2a1-M88)

Wa (Shuangjiang, Yunnan)
4/31 = 12.9% C-M130
1/31 = 3.2% F-M89(xK-M9)
4/31 = 12.9% K-M9(xO1a-M119, O2a-M95, O3-M122, P-M45)
7/31 = 22.6% O3-M122(xO3a3b-M7, O3a3c-M134)
8/31 = 25.8% O3a3c-M134
7/31 = 22.6% O2a-M95(xO2a1-M88)

Yunnan Austroasiatic total (31 Shuangjiang Wa + 28 Shuangjiang Bulang + 16 Luxi De'ang)
6/75 = 8.0% C-M130
4/75 = 5.3% F-M89(xK-M9)
12/75 = 16.0% K-M9(xO1a-M119, O2a-M95, O3-M122, P-M45)
11/75 = 14.7% O3-M122(xO3a3b-M7, O3a3c-M134)
19/75 = 25.3% O3a3c-M134
23/75 = 30.7% O2a-M95(xO2a1-M88)

Ebizur said...

Maju said,

"It'd be interesting to compare with SE Chinese, who must retain at least a big deal of pre-Sinitic genetics and may be (I guess) a missing link in all this process."

I mentioned yesterday that Karafet et al. had not found even one instance of O1a2-M110 in all their 383 samples of Han, Miao, She, Tujia, and Yao males from China. This covers practically all extant ethnic groups in the southeast of mainland China. Most representatives of haplogroup O1a-M119 in continental Asian populations (and, in fact, in most populations that include members of haplogroup O1a) belong to O1a1*-P203. O1a1*-P203 is also the most common Y-DNA haplogroup in aboriginal peoples of Taiwan. O1a2 is notable despite its relative scarcity even in its zone of maximal frequency (i.e. aboriginal Taiwan) because its distribution is limited almost completely to populations that speak an Austronesian or a Kra-Dai language.

Maju said...

"Yes, and K1, K2, K3 and K4 are members of the KMNOPS haplogroup".

You don't know that about K1. In fact you don't know that for any of those particular haplogroups, though, on light of Karafet 2010, the case for the SEA lineages (K2, K3 and K4) seems pretty solid.

"And you alter your understanding depending on what point of view you are supporting at the time. Some time back you insisted that K1, K2, etc. were simply regional variants that had formed along with M, S, NO and P. Now that that claim conflicts with the belief you're currently supporting you are pushing the idea that somehow K1, K2, K3 and K4 belong to a group separate from MNOPS".

You misinterpret me wildly.

Per the old phylogeny, K1, K2, K3 and K4 were basal K sublineages, exactly like L, M, NO, P and S.

But the discovery of MNOPS changes that perception, because a number of these K sublineages would now belong to a single subhaplogroup of K, namely MNOPS.

At the moment it seems these are M, NO, P, S and all the K-other in SEA and Oceania (what surely implies K2, K3 and K4, as well as other regional K*, but probably not K1 nor the K* found outside SEA/Oceania).

But I have explained this before. It is you who insist in misreading what I (and the researchers) say.

"The Austronesian langiages in Taiwan appear to belong to two groups. Ayatalic is close to the Austronesian root, but other languages are closely related to Indonesian and Philippine languages. This strongly suggests there was a back-migration to Taiwan, probably from the Philippines, early on during the establishment of the Austronesian language".

It's possible, I could not say. But I was actually thinking of Malayo-Polynesians, which is the only non-Taiwanese branch of the Austronesian linguistic family. My bad with the terminology.

What I'm suggesting is that maybe Austronesian (Taiwanese) arrival at Luzon, together with other ingredients that I cannot fathom in detail, would have been the real origin of the Malayo-Polynesian expansion, which is the only one that happened within the Austronesian linguistic group (all the rest is Taiwan tribals).

I mean that the origin of Malayo-Polynesian language and culture (and hence genetics) might be at Luzon and nearby islands and not directly at Taiwan.

This would potentially explain the dominant O3a3 Y-DNA among Austronesians outside Taiwan (but not in Taiwan). But leaves unexplained how did O3a3 arrived to Luzon.

"Probably most of the O haplogroup, if the movement south implied by the old considerations concerning the language relationships is correct".

But the old considerations are "old" for a reason.

Nowadays the simplistic straightforward associations between genes and languages are not commonly accepted. And certainly not by me.

Also you have again Karafet 2010 saying that most O in ISEA is Paleolithic, so...

Maju said...

"I don't know why you would infer such a thing from the data I have posted".

Mulam (a Kradai ethnicity from Guangxi, at the border with Hunan and Gizhou): 25% (10/40) O1a2. Kradai speakers from Southern China: 10% (2/20), Dong (from the three aforementioned provinces): 10% (1/10).

I may have exaggerated. Not sure.

"I mentioned yesterday that Karafet et al. had not found even one instance of O1a2-M110 in all their 383 samples of Han"...

When I say Chinese in anthropological or ethological discussions I normally mean Han. The rest (excepting Hui, a sui-generis Pakistani-like pseudo-ethnicity) are in principle not Chinese native speakers and hence not ethnic Chinese. It's language what makes the ethnicity primarily, not the passport or ID card, which is a political intrusive element.

Anyhow, how well did she sample SE Han in particular? There are a lot of them.

Also, do you recall if this lineage is found among Hainan ethnicities?

I'm just trying to figure out how this lineage sprouted from O1a. Maybe it did already in Taiwan and migrated from there to the interior but maybe it's the other way around.

Which is your hypothesis?

Mason said...
This comment has been removed by the author.
Mason said...

According to Karafet's main pdf, "mainland southeast Asia" samples include the samples in southern China, such as Han, Miao, She, Tujia, Yao. And Karafet said, from STR point of view, TAB/Filipino/Indonesian/Oceanian O3a3-P201* are closely related to each other, but far away from mainland southeast Asian O3a3-P201* including southern Chinese O3a3-P201*.

So there may be two theories: (1) Some O3a3* were separated from mainland O3a3* and isolated in Taiwan since Paleolithic, and later expanded to ISEA and Oceania after Neolithic. (2) Some O3a3* were isolated in ISEA (Indonesia?) since Paleolithic and expanded to Taiwan after Neolithic.

BTW, there are three major branches of Austronesian languages in Taiwan: Atayalic, Tsouic, Paiwanic.
And Paiwanic reveals some likely affinity with Malayo-Polynesian languages outside Taiwan.

Ebizur said...

Maju said,

"Mulam (a Kradai ethnicity from Guangxi, at the border with Hunan and Gizhou): 25% (10/40) O1a2. Kradai speakers from Southern China: 10% (2/20), Dong (from the three aforementioned provinces): 10% (1/10)."

I have divided the information on samples in which O1a2 has been found outside its normal geographic range into various categories, and I have placed a description of each of the categories ("Kra-Dai speakers from southern China and Southeast Asia," "Mon-Khmer speakers from southern China," etc.) in italics immediately before the data from the first sample that I have placed in that category. Thus, Kra-Dai speakers from southern China and Southeast Asia describes the category to which the Kam/Dong (x2), northeastern Thai, Mollao, Mulam, and northern Zhuang samples belong. Note that I have cited data on the Kam/Dong people ("Kam" being an endonym, and "Dong" being a Chinese exonym) from the articles by Su Bing et al. (1999 & 2000) and Xie Xuan-Hua et al. (2004). On the other hand, Li Hui et al. (2008) have not found any O1a Y-DNA of any sort in a sample of 38 Kam/Dong individuals.

In short: the 25% (10/40) O1a2 "Mulam" figure comes from Li Hui et al. (2008), the 10% (2/20) O1a2 "Kam" figure comes from Xie Xuan-Hua et al. (2004), and the 10% (1/10) O1a2 "Dong" figure comes from Su Bing et al. (1999 & 2000).

Maju said,

"When I say Chinese in anthropological or ethological discussions I normally mean Han. The rest (excepting Hui, a sui-generis Pakistani-like pseudo-ethnicity) are in principle not Chinese native speakers and hence not ethnic Chinese. It's language what makes the ethnicity primarily, not the passport or ID card, which is a political intrusive element."

I think you have underestimated the relevance of the Karafet team's data regarding the Y-DNA of the Miao, Yao, She, and Tujia peoples. The various Hmong-Mien peoples (Miao, Yao, She) and the Tujia are considered to have preceded the Han in south-central/southeastern parts of mainland China. These pre-Han ethnic groups should have absorbed any remnants of even earlier ethnic groups in the region that had failed to flee elsewhere (perhaps Kra-Dai peoples, which persist as ethnic entities in China only in the southwest, around Yunnan, Guizhou, and Guangxi, and on Hainan Island). Therefore, any substratal influence on the Y-DNA of modern Han males in south-central/southeastern parts of mainland China should be expected to be proximally related to Hmong-Mien peoples, Tujia, or something similar to one or both of those groups. A direct Austronesian influence on southern Hans is quite unlikely anywhere outside of Taiwan.

Maju said...

But the apportion of O3a3* (or O3 for the case) among Taiwanese Aboriginals is very low. It makes no sense to trace its origin in ISEA/Pacific Ocean to Taiwan.

Instead it's high in Philippines and Sumatra. Philippines stands now to my eyes as a more likely source of the Y-DNA involved in the Austronesian migration.

There was probably some bidirectional flow between Philippines and Taiwan at the beginning of the Austronesian expansion... with O3a3* flowing northwards and O1a* and O1a2 southwards. My best guess so far.

Maju said...

The above comment was meant for Natsuya.

For Ebizur now:

"Thus, Kra-Dai speakers from southern China and Southeast Asia describes the category to which the Kam/Dong (x2), northeastern Thai, Mollao, Mulam, and northern Zhuang samples belong".

I thought it was a different sample, specially because it only included two O1a2 individuals and the Mulam sample included 10. There's an error or misunderstanding somewhere.

"I think you have underestimated the relevance of the Karafet team's data regarding the Y-DNA of the Miao, Yao, She, and Tujia peoples. The various Hmong-Mien peoples (Miao, Yao, She) and the Tujia are considered to have preceded the Han in south-central/southeastern parts of mainland China. These pre-Han ethnic groups should have absorbed any remnants of even earlier ethnic groups in the region that had failed to flee elsewhere (perhaps Kra-Dai peoples, which persist as ethnic entities in China only in the southwest, around Yunnan, Guizhou, and Guangxi, and on Hainan Island). Therefore, any substratal influence on the Y-DNA of modern Han males in south-central/southeastern parts of mainland China should be expected to be proximally related to Hmong-Mien peoples, Tujia, or something similar to one or both of those groups. A direct Austronesian influence on southern Hans is quite unlikely anywhere outside of Taiwan.".

I do not mean an Austronesian influence (though who knows) but rather the pool from which the Taiwan O1a2 arose.

Anyhow, Fujian, for instance is larger than Portugal and has a population like Spain. Each of the Chinese provinces is like a medium sized country worth detailed scrutiny on its own right.

Also I don't think that Southern Han are mere immigrants from the North but rather, essentially, aculturized pre-Chinese southern nations, with just a small apportion of immigrant blood maybe. I just don't believe that pre-Chinese farmers would flee but that they were assimilated by the imperial ethnic identity, which was also in the interest of the Empire, logically, as it needed productive subjects not just empty lands.

I'm not making up anything: it's a well known fact that Southern Chinese (Han) are genetically different from Northern Chinese (Han).

Ebizur said...

Maju said,

"Anyhow, how well did she sample SE Han in particular? There are a lot of them."

Karafet et al. (2010)
China (Han)
10/165 = 6.1% C3-M217
1/165 = 0.6% D1-M15
1/165 = 0.6% G2a-P15
1/165 = 0.6% J2-M172(xJ2b-M12)
2/165 = 1.2% N*-M231(xN1-LLY22g)
12/165 = 7.3% N1-LLY22g(xN1a-M128, N1c1-M178) I'm fairly sure these are actually N1*-LLY22g(xN1a-M128, N1b-P43, N1c-M46)
2/165 = 1.2% N1c1-M178
1/165 = 0.6% N1a-M128
2/165 = 1.2% O3*-M122(xO3a-P197)
7/165 = 4.2% O3a-P197(xO3a3-P201, O3a4-JST002611)
14/165 = 8.5% O3a3-P201(xO3a3b-M7, O3a3c-M134)
3/165 = 1.8% O3a3b-M7
47/165 = 28.5% O3a3c-M134
18/165 = 10.9% O3a4-JST002611
1/165 = 0.6% O1a*-M119(xO1a1-P203, O1a2-M110)
15/165 = 9.1% O1a1-P203
8/165 = 4.8% O2*-P31(xO2a-M95, O2b-SRY465)
10/165 = 6.1% O2a-M95(xO2a1-M111)
9/165 = 5.5% O2a1-M111
1/165 = 0.6% Q1-P36(xQ1a3-M346)

Obviously, this is a composite of the Karafet team's samples of Hans from Shaanxi (northern China; n=44), Guangdong (southern China; n=40), and Taiwan (n=84 according to Hammer et al. (2006); n=82 according to Karafet et al. (2001)). Karafet et al. (2005) have used the same pooled sample of Hans with n=166; one of the O3-M122 Hans has gone missing in the interim between Karafet et al. (2005) and Karafet et al. (2010). Thus, nearly three quarters of the Han sample of Karafet et al. (2010) have been derived from southeastern and south-central areas, yet not even one instance of O1a2-M110 has been found.

Maju said,

"Also, do you recall if this lineage is found among Hainan ethnicities?"

Haplogroup O1a2 appears to be absent from Hainan. Li et al. (2008) have reported 0/34 O1a2-M110 in a sample of Qi (also known in Cantonese transcription as "Gei"), 0/27 O1a2-M110 in a sample of Jiamao, 0/31 O1a2-M110 in a sample of Cun, and 0/30 O1a2-M110 in a sample of Lingao, all of which are speakers of phylogenetically primitive Kra-Dai languages restricted to the island of Hainan. They also have reported 0/31 O1a2-M110 in a sample of Tsat, a demographically insignificant Chamic ethnic group native to the southernmost tip of Hainan Island. Su Bing et al. (1999 & 2000) have found 0/11 O1a2-M50/M103/M110 in a sample of "Li," which is a Chinese exonym under which most of the aforementioned Kra-Dai-speaking aboriginal populations of Hainan Island may be subsumed.

In "Paternal Genetic Structure of Hainan Aborigines Isolated at the Entrance to East Asia," Li Dongna, Li Hui, Ou Caiying et al. (2008) have included a column in their data table only for O1a*, though they have claimed in the paper's Materials and Methods section that they have tested M110. Anyway, haplogroup O1a2 must be very rare in populations of Hainan if it is present at all.

Ebizur said...

Maju said,

"I'm just trying to figure out how this lineage sprouted from O1a. Maybe it did already in Taiwan and migrated from there to the interior but maybe it's the other way around.

Which is your hypothesis?"

I do not recall having ever seen a median-joining network or even a spreadsheet of Y-STR haplotypes associated with O1a2-M50/M103/M110. The only relevant data I have are the STR variance data from supplementary table S3 of Karafet et al. (2010). These data suggest that O1a2-M110 is a very young haplogroup, with only its sister taxon, O1a1-P203, and eastern Indonesian F*-P14 exhibiting comparably low variance among the haplogroups included in the analysis. However, one must keep in mind that the populations in which haplogroup O1a2 has been found most frequently include the Taiwan aborigines and the Malagasy of Madagascar, so the probability that it has been spread at least in part by speakers of Austronesian languages should be about 100%. My current hypothesis for the origin of the O1a2 that seems to be sparsely distributed among Kra-Dai-speaking ethnic groups from the southwestern parts of mainland China to northeastern Thailand (barring any lab errors or funny business on the part of the Chinese researchers responsible for these reports) is that it has diffused from an early Austronesian-speaking source population into a certain subset of the early Kra-Dai-speaking population or population cluster, perhaps at a time, prior to its westward migration, when the Kra-Dai-speaking recipient population inhabited an area near the shores of the East China Sea or the South China Sea.

Ebizur said...

Maju said,

"I thought it was a different sample, specially because it only included two O1a2 individuals and the Mulam sample included 10. There's an error or misunderstanding somewhere."
I have included six samples in the category "Kra-Dai speakers from southern China and Southeast Asia": two samples of Kam/Dong, one sample of northeastern Thai, one sample of Mollao, one sample of northern Zhuang, and one sample of Mulam. You must have overlooked my identification of the first sample under the category's heading as Kam(=Dong) (XIE Xuan-Hua, LI Hui, MAO Xian-Yun et al., "Genetic Structure of Tujia as Revealed by Y Chromosomes," Acta Genetica Sinica, October 2004, 31 (10) : 1023-1029). That's really all there is to it. I apologize if my listing of the data has been too messy.

As for the differences among Hans of various regions, these seem to be clinal, with the northernmost Hans approaching Mongols and Manchus and the southernmost Hans approaching Vietnamese and Zhuangs. Assimilation of pre-Han populations seems to have left a mark on the Y-DNA pools of Chinese speakers in Guangdong and Guangxi, with haplogroup O2a-M95 being much more common among Hans in these parts of south-central/southwestern China than in southeastern (e.g. Taiwan) or northern areas (e.g. Shaanxi).

Ebizur said...

I have found another apparent discrepancy in the Chinese data of the Karafet & Hammer team:

Karafet et al. (2001)
Northern Han (Shaanxi)
2/44 = 4.5% P-DYS257.108(xQ1a3a-DYS199/M3, R-UTY2)
1/44 = 2.3% R1a1-SRY10831.2

Karafet et al. (2005)
Han ("The Chinese Han sample was composed of individuals from the Guangdong and Shaanxi provinces and of individuals from different parts of Taiwan (Karafet et al. 2001).")
1/166 = 0.6% Q1-P36
0/166 R-M207

Hammer et al. (2006)
Northern Han
2/44 = 4.5% Q1-P36
1/44 = 2.3% R-M207

Karafet et al. (2010)
China (Han)
1/165 = 0.6% Q1-P36(xQ1a3-M346)
0/165 R-M207

Maju said...

"...the populations in which haplogroup O1a2 has been found most frequently include the Taiwan aborigines and the Malagasy of Madagascar"...

Add Nias island, offshore of Sumatra, which has about the same Y-DNA composition as Taiwan Aborigines, but sure: that seems to be it.

"...the probability that it has been spread at least in part by speakers of Austronesian languages should be about 100%".

I am totally in agreement with that. I was trying to figure how O1a2 came to existance, from O1a, which may have got an ancestral homeland somewhere in SEA, maybe Sundaland. Trying to understand the possible Paleolithic flows in the area, according to Y-DNA genetics...

But guess that in due time.

"I do not recall having ever seen a median-joining network or even a spreadsheet of Y-STR haplotypes associated with O1a2-M50/M103/M110".

I just realized that the fig. 4 of Karafet's paper is precisely that: and the central node is more than 95% Eastern Indonesian (wow!), cross-hatched pattern, with less than 5% being Taiwanese Aboriginal. Other Taiwanese lineages seem derived and scattered, with the central nodes being East and West Indonesian almost totally.

"14/165 = 8.5% O3a3-P201
3/165 = 1.8% O3a3b-M7
47/165 = 28.5% O3a3c-M134"

This is an interesting detail: SE Han have significant O3a3(xO3a3b). Not as high as among Filipinos/Sumatrans but there's more here than among Taiwan Aborigines - and that also applies to the main component of the paragroup among Malayo-Polynesians.

Nevertheless, Karafet considers all this paragroup a mark of Austronesian expansion and is in fact most of the Y-DNA she can associate with this process.

There are two potential sources for this paragroup outside ISEA (having both O3a3* and O3a3c, see map S4 for a quick visual reference): Southern Han and Orang Asli.

I think that Karafet decided in favor of this haplogroup because O3a3c is found in Tahiti (3/24) along O3a3*, a much more common para-lineage among Austronesians, what also happens in Philippines but not elsewhere in the Austronesian world.

I'm even tempted to discard O3a3c from the Austronesian pattern (though Tahiti is hard to explain) but, even doing that, the remaining O3a3* should have an either Southern Han, Filipino or "Sundaland" origin. It can't be Taiwanese (and less so when even O1a2 seems ISEA originated, per above).

"As for the differences among Hans of various regions, these seem to be clinal"...

Almost everything in Eastern Asia seems clinal.

"Assimilation of pre-Han populations seems to have left a mark on the Y-DNA pools of Chinese speakers in Guangdong and Guangxi, with haplogroup O2a-M95 being much more common among Hans in these parts of south-central/southwestern China than in southeastern (e.g. Taiwan) or northern areas (e.g. Shaanxi)".

I would not only look at O2a but also and maybe very specially to the many sublineages in O3, as well as to the meaningful near-lack of C3 in southern Han. C3 levels are probably a good quick reference marker for Northern Han demic penetration in fact, as they cannot be attributed just to recent Altaian penetration, necessarily minimal, but most likely to ancient sources of at least Neolithic origins.

"I have found another apparent discrepancy in the Chinese data of the Karafet & Hammer team"...

Aren't these different samples?

Ebizur said...

Maju said,

"I just realized that the fig. 4 of Karafet's paper is precisely that: and the central node is more than 95% Eastern Indonesian (wow!), cross-hatched pattern, with less than 5% being Taiwanese Aboriginal. Other Taiwanese lineages seem derived and scattered, with the central nodes being East and West Indonesian almost totally."

Oh, yes, you are right! "FIG. 4.—Median-joining network for haplogroup O-M110 based on variation at 12 Y-STRs." This new paper by Karafet et al. (2010) really does contain an abundance of interesting new information. However, you must note that Karafet's sample of Taiwan aborigines includes only nine representatives of O1a2-M110, and all nine of these have presented distinct 12 Y-STR haplotypes. Also, the nine Taiwan aborigine O1a2 haplotypes are spread out on several different branches of the median-joining network, so Taiwan still should be considered a very likely candidate for the place of origin of haplogroup O1a2.

Maju said,

"This is an interesting detail: SE Han have significant O3a3(xO3a3b). Not as high as among Filipinos/Sumatrans but there's more here than among Taiwan Aborigines - and that also applies to the main component of the paragroup among Malayo-Polynesians."

Remember, the Karafet team's pooled "Han" sample contains (approximately) 44 individuals from Shaanxi Province, which is nowhere near southeastern China; it is actually a north-central/northwestern province. I can only add that, according to Hammer et al. (2006), their Taiwan Han sample contains the greatest apportion (approx. 29/84 = 34.5%) of O3-M122(xO3a3c-M134), followed by their Shaanxi Han sample (11/44 = 25.0%), and finally their Guangdong Han sample (4/40 = 10.0%). So, yes, it is most likely that Taiwan Han have a relatively high frequency of O3a3-P201(xO3a3b-M7, O3a3c-M134), but I cannot be sure about this without knowing the distribution of O3a4-JST002611 and O3a-P197(xO3a3-P201, O3a4-JST002611) among the three regional Han samples.

Maju said,

"I think that Karafet decided in favor of this haplogroup because O3a3c is found in Tahiti (3/24) along O3a3*, a much more common para-lineage among Austronesians, what also happens in Philippines but not elsewhere in the Austronesian world."

Tahiti, like many other Pacific islands, historically has hosted a significant population (ca. 10% or more of the total population of the island at some times) of ethnic Chinese, most of whom were males from southern parts of China who married native Polynesian females. Today, most descendants of these Chinese immigrants to Tahiti have been assimilated into the local population, so it should not be surprising to find a Y-DNA haplotype recently derived from China in a Polynesian-speaking male from Polynesia. The Philippines, of course, likewise has a great deal of recent influence from China, particularly from Fujian and Guangdong provinces. Most (or even all) of the O3a3c-M134 haplotypes in the Philippines and Oceania may reflect recent Chinese and (particularly in Micronesia) Japanese influences.

Ebizur said...

Maju said,

"I would not only look at O2a but also and maybe very specially to the many sublineages in O3, as well as to the meaningful near-lack of C3 in southern Han. C3 levels are probably a good quick reference marker for Northern Han demic penetration in fact, as they cannot be attributed just to recent Altaian penetration, necessarily minimal, but most likely to ancient sources of at least Neolithic origins."

I agree that haplogroup C3 should not be overlooked. However, you are incorrect that this haplogroup is rare in southern Han populations. C3-M217 is one of the haplogroups that has been found with very similar frequency among Hans from all regions of China, and, in fact, among their three samples of Han Chinese, Karafet et al. have found C3-M217 with highest frequency in their sample of Hans from Taiwan and with lowest frequency in their sample of Hans from Shaanxi, though I do not think the differences are significant in this case.

Maju said,

"Aren't these different samples?"

Ostensibly, no. In the present study, Karafet et al. have reused what appears to be the same (minus one individual) pool of Han samples that they have used for their 2005 article about Bali, and they have described it in that article as being "composed of individuals from the Guangdong and Shaanxi provinces and of individuals from different parts of Taiwan (Karafet et al. 2001)," citing their article from 2001, in which the three samples of Hans have been tabulated separately.

Maju said...

"... so Taiwan still should be considered a very likely candidate for the place of origin of haplogroup O1a2".

Fair enough.

Still the (yes: highly diverse) TA haplotypes are distributed only in 4/10 branches, plus the central node and at least in one of these cases it appears derived from via West Indonesia.

Counting only central node and "inner ring" (most basal nodes):

Taiwan Aborigines: 2/11
East Indonesia: 3/11
West Indonesia: 5/11
Other (mainland SEA, South China incl.): 1/11
Unclear (basal node not sampled): 3/11

They don't add up to 11 because some are shared between two regions.

The latter "unclear" group are:
- MSEA/Taiwan
- MSEA/Taiwan/West Indonesia
- MSEA/West Indonesia

So, reconstructing:
- TAs: 4/11 (including core)
- East Ind.: 3/11 (including core)
- West Ind.: 7/11
- MSEA: 4/11

Inconclusive but West Indonesia is the most basally diverse, however it does not include the central node. I'm again speculating about Philippines.

"Remember, the Karafet team's pooled "Han" sample contains (approximately) 44 individuals from Shaanxi Province, which is nowhere near southeastern China; it is actually a north-central/northwestern province".

Ok, I take notice. Still...

"Tahiti, like many other Pacific islands, historically has hosted a significant population (ca. 10% or more of the total population of the island at some times) of ethnic Chinese, most of whom were males from southern parts of China who married native Polynesian females".

I had absolutely no idea. If so we can safely exclude this lineage from the Austronesian expansion.

"C3-M217 is one of the haplogroups that has been found with very similar frequency among Hans from all regions of China, and, in fact, among their three samples of Han Chinese, Karafet et al. have found C3-M217 with highest frequency in their sample of Hans from Taiwan and with lowest frequency in their sample of Hans from Shaanxi, though I do not think the differences are significant in this case".

Aha. That's interesting. I thought it was a northern trait.

I was just briefly looking at Hong Shi 2005 and I do get the impression that all or most O3 subclades are southern, and can't really find a northern specific marker among them (except O3a* probably).

However you might have a sharper eye.

In any case, it does seem like the Southern Han (and to some extent also Northern Han) share their Y-DNA with the pre-Han ethnicities of the South. So it looks difficult to discern based on haplogroups... but not just between North Han and South Han but in most cases also between Han and non-Han.

Mason said...

O3-M122 and most (maybe all) of its subclades have southern origins. O3a3c-M134 also has a southern origin, this clade was previously considered as Sino-Tibetan origin, but its oldest haplotypes are found in more southern populations, such as Daic. My personal Y-DNA is O3a3c* (M134+ M117- P101-), but my paternal line could very likely be southern non-Han assimilated into Han hundreds or thousands of years ago. (I'm a Taiwan Han.)

Those O3a* in Hong Shi's paper could mostly be today's O3a3-P201* and O3a4-002611, which weren't found yet back then.

Mason said...
This comment has been removed by the author.
terryt said...

"But the discovery of MNOPS changes that perception, because a number of these K sublineages would now belong to a single subhaplogroup of K, namely MNOPS".

Surely there's no evidence at all that all the Ks belong to a single separate lineage within MNOPS.

"At the moment it seems these are M, NO, P, S and all the K-other in SEA and Oceania (what surely implies K2, K3 and K4, as well as other regional K*"

So you agree that some K lineages, at least, belong in MNOPS (or whatever you wish to call it). So why do you consistently ignore these lineages when considering the place of origin for said MNOPS haplogroup?

"What I'm suggesting is that maybe Austronesian (Taiwanese) arrival at Luzon, together with other ingredients that I cannot fathom in detail, would have been the real origin of the Malayo-Polynesian expansion, which is the only one that happened within the Austronesian linguistic group (all the rest is Taiwan tribals)".

That's very likely. In fact it seems that some of those 'Malayo-Polynesian' languages moved back to Taiwan as well. As Natsuya said, 'And Paiwanic reveals some likely affinity with Malayo-Polynesian languages outside Taiwan'. I'd guess that the Austronesian expansion is the result of a complex of minor movements around the region, especially between the Philippines and Taiwan, before the major expansion.

"Each of the Chinese provinces is like a medium sized country worth detailed scrutiny on its own right".

That really does sum up the problem.

"I don't think that Southern Han are mere immigrants from the North but rather, essentially, aculturized pre-Chinese southern nations, with just a small apportion of immigrant blood maybe".

It's almost certainly more than 'just a small apportion', but not enough to obscure the 'well known fact that Southern Chinese (Han) are genetically different from Northern Chinese (Han)'. But the movemnt south is before the establishment of the Empire, so 'productive subjects' were not actually needed, just empty lands.

"These data suggest that O1a2-M110 is a very young haplogroup"

And the same presumably holds for all the downstream O mutations. Their southern diversity belongs to a stage after their relatively recent arrival from the north. All basal O haplogroups are also found in the north. For example O2b is northern and O2a southern, but O2a probably formed from O2* that had moved south. So O1a2 could well be Taiwanese in origin.

"it is most likely that Taiwan Han have a relatively high frequency of O3a3-P201(xO3a3b-M7, O3a3c-M134)"

So O3a is probably Han, not indigenous South Chinese.

"C3-M217 is one of the haplogroups that has been found with very similar frequency among Hans from all regions of China"

And is almost certainly northern in origin. This provides yet more evidence for substantial southerly movement through China.

Mason said...

to terryt:

As I mentioned in the previous post, O3a* in Hong Shi's paper could actually be P201 and 002611, which haven't been found back in 2005.

The most common O3 subclades found in Han populations are O3a3c-M134*, O3a3c1-M117, O3a3-P201*, and also 002611. The real O3a* (with M324, P197, but no more downstream markers) is rare.

http://www3.interscience.wiley.com/cgi-bin/fulltext/117956695/PDFSTART

Table 4 of the paper contains Y-DNA haplogroup frequencies of Japanese (in different parts of Japan), Koreans, and also Taiwan Han people.

They use different haplogroup names in the paper:

O3e* = O3a3c-M134*
O3e1 = O3a3c1-M117
O3-021354* = O3a3-P201*

Mason said...

Note: Hong Shi's 2005 paper also used different haplogroup names from today's.

O3b = today's O3*
O3a* = O3a-M324*(P197*)
O3a4 = O3a3b-M7
O3a5 = O3a3c-M134
O3a5b = O3a3c-M134*
O3a5a2 = O3a3c1-M117*

Ebizur said...

Judging from a comparison between the data tables of Hammer et al. 2006 and Karafet et al. 2010, the erstwhile NO*-M214 may be a phantom paragroup after all. The two N*-M231(xN1-LLY22g) Han individuals reported by Karafet et al. 2010 correspond to the individuals represented by the 1/40 = 2.5% NO-M214(xN1-LLY22g, O-M175) Guangdong Han and the 1/44 = 2.3% NO-M214(xN1-LLY22g, O-M175) Shaanxi Han data reported by Hammer et al. 2006. Xue et al. 2006 have not found any instance of NO-M214(xN1-LLY22g, O-M175) among their Hans sampled from Harbin, Heilongjiang (n=35), Yili, Xinjiang (n=32), Chengdu, Sichuan (n=34), Lanzhou, Gansu (n=30), and Meixian, Guangdong (n=35). This means that N*-M231(xN1-LLY22g) has been found in one Han individual from Guangdong in south-central China and another Han individual from Shaanxi in northwestern/north-central China, but there is now no evidence for any presence of haplogroup NO*-M214(xN-M231, O-M175) in any Han Chinese population. All other purported cases of NO* actually have been tested to be NO-M214(xN1-LLY22g, O-M175), so any number of these may turn out to be N*-M231(xN1-LLY22g) as well.

Inferred from comparison between Han Chinese data sets of Hammer et al. 2006 and Karafet et al. 2010:

Shaanxi Han
1/44 = 2.3% N*-M231
3/44 = 6.8% N1*-LLY22g
1/44 = 2.3% N1a-M128

Guangdong Han
1/40 = 2.5% N*-M231
6/40 = 15.0% N1*-LLY22g

Taiwan Han
3/82 = 3.7% or 3/84 = 3.6% N1*-LLY22g
2/82 = 2.4% or 2/84 = 2.4% N1c1-M178

Maju said...

"Those O3a* in Hong Shi's paper could mostly be today's O3a3-P201* and O3a4-002611, which weren't found yet back then".

The Northern Han marker(s) should be apparent, in principle, in Kararafet's study as more frequent among Han (mix N-S sample) than among other ethnicities. And I can only identify the following that follow such criteria: N (all clades), O3a3c-M134 and some single individual erratics in the J and Q haplogroups. C3-M217 is rather high but it's higher among the Tujia (a Sino-Tibetan ethnicity original from East Sichuan, rather "northernly") O3a4-JST002611 is also meaningful but shared with the Miao (Hmong-Mien family), O2a1-M111 is also somewhat high among Han but again overshadowed by another Hmong-Mien people: the She.

It would seem like the Han have no single Y-DNA marker but rather are a patchwork of their neighbors. Nevertheless, there might be downstream haplogroups or STR haplotypes that might be specific to them. Whatever the case and, unless further evidence shows the opposite, and barring maybe some localized settlements, it seems to me that there was no meaningful demic colonization but rather a gradual absorption of more and more neighbors into the hegemonic imperial ethnicity.

My two cents anyhow... and open to alternative ideas (backed with some data).

Maju said...

"Surely there's no evidence at all that all the Ks belong to a single separate lineage within MNOPS".

What kind of objection is that to what I said?:

"But the discovery of MNOPS changes that perception, because a number of these K sublineages would now belong to a single subhaplogroup of K, namely MNOPS".

"A single subhaplogroup [of K], namely MNOPS" is not the same as "a single separate lineage within MNOPS".

You read what you want to read not what is written, which is quite worrisome.

"So you agree that some K lineages, at least, belong in MNOPS (or whatever you wish to call it). So why do you consistently ignore these lineages when considering the place of origin for said MNOPS haplogroup?"

I do not. I considered all of Karafet's K*/MNOPS*, both as one single paragroup and as locally different paragroups. The result was the same: Borneo.

That's because there's a lot of unclassified MNOPS* at Sundaland and East of it. Overall it does push the composite centroid from Peninsular Malaysia to Borneo... but not to your darling Wallacea. I also considered with and without Micronesia/Polynesia and again the result was not different.

"It's almost certainly more than 'just a small apportion'"...

Would you mind to back this claim with some real data. As I posted elsewhere, I can't find any single meaningful lineage that is Han-specific.

"But the movemnt south is before the establishment of the Empire, so 'productive subjects' were not actually needed, just empty lands".

This is absolutely absurd and shows a total ignorance of the history of China and therefore the Han ethnicity. You are every day less and less rigorous, what makes discussing with you a quite unpleasant activity - certainly an unproductive one.

"So O3a is probably Han, not indigenous South Chinese".

Hong Shi et al., Y-Chromosome Evidence of Southern Origin of the East Asian–Specific Haplogroup O3-M122. AJHG 2005.

Maju said...

"As I mentioned in the previous post, O3a* in Hong Shi's paper could actually be P201 and 002611, which haven't been found back in 2005".

This haplogroup is shared with the Miao, among other less relevant ethnicities, and these have a clearly higher apportion. Also Vietnamese. Per Karafet, O3a4-JST002611 is found:

- Han (TB): 18/165 (11%)
- Miao (HM): 12/58 (21%)
- Vietnamese (AA): 10/70 (14%)
- Tujia (TB): 4/49 (8%)
- Philippines (AN): 1/48 (2%)

So it looks more like yet another Southern O haplogroup.

Ebizur said...

Maju said,

"The Northern Han marker(s) should be apparent, in principle, in Kararafet's study as more frequent among Han (mix N-S sample) than among other ethnicities. And I can only identify the following that follow such criteria: N (all clades), O3a3c-M134 and some single individual erratics in the J and Q haplogroups."

Among the Karafet team's samples of Han Chinese, haplogroup N Y-DNA occurs most frequently in the sample from Guangdong (17.5%), second most frequently in the sample from Shaanxi (approx. 11%), and least frequently in the sample from Taiwan (approx. 6%). Having such data, it is difficult to treat Chinese haplogroup N Y-DNA as a "Northern Han" marker.

Mason said...
This comment has been removed by the author.
Mason said...

http://www3.interscience.wiley.com/cgi-bin/fulltext/117956695/PDFSTART?CRETRY=1&SRETRY=0

According to Table 4 of the paper, this is the frequency of O3a4-002611 in Japanese, Koreans, Taiwan Han people:

Japanese(Asahikawa): 3.1%
Japanese(Kanto): 0.7%
Japanese(Nagoya): 3.4%
Japanese(Western): 2.1%
Japanese(Okinawa): 6.3%
Korean: 0.6%
Taiwan Han: 8.5%

Mason said...

O3a3-P201* (also known as 021354*) in three populations:

Japanese(Asahikawa): 2.1%
Japanese(Kanto): 3.6%
Japanese(Nagoya): 6.0%
Japanese(Western): 6.2%
Japanese(Okinawa): 5.7%
Korean: 9.8%
Taiwan Han: 5.6%

And O3a3c-M134 (O3e*+O3e1* in Table 4) in three populations:

Japanese(Asahikawa): 4.4%
Japanese(Kanto): 7.3%
Japanese(Nagoya): 6.8%
Japanese(Western): 10.4%
Japanese(Okinawa): 4.6%
Korean: 27.3%
Taiwan Han: 29.0%

Mason said...
This comment has been removed by the author.
Mason said...

to Maju:

As the data I posted above, O3a4-002611 is seen in all over East Asia. Its origin remains unknown. I believe this lineage as well as O3a3-P201* and O3a3c-M134 is bonanza which we can use to clarify the prehistory of East Asia.

Maju said...

The data is interesting but I see that:

O3a3-P201* is clearly higher towards the South, what, in addition to what was said before, doesn't make it look like a Han or otherwise NE Asian marker at all.

In West Indonesia and Mainland/SEA (with or without Han) it is around 7%, what is also in line with what is found among Taiwan Han and Okinawans and well above what Japanese proper or Koreans have in any region. So I'd rather think Japanese O3a3* is of southern "coastal" origin.

Philippines has a much higher percentage of this lineage (25%) by the Karafet data (guess that even more in other datasets with less K*/MNOPS*). Sumatra has 55% (the highest figure I could spot). Other high results are Tonga (44%), Micronesia (19%) and Tahiti (12%)... but they are surely "recent" founder effects from, I guess, Philippines.

This clade looks more "Austronesian" than O3a3c, at least in Polynesia/Micronesia. But the origin (or at least the highest frequencies) look rather from ISEA.

As for O3a3c-M134, this I already said it's the one looking more Han-like (if any), even if shared with the Tujia (still a relatively "northernly" ethnicity). But notice that, by Hong Shi 2005, it's also very high among various southern ethnicities, specially around Yunnan (and also has peaks at Hunan and Fujian).

It's not particularly high among Northern Han (7-20%), a little higher among Southern Han (14-22%), but it's very dominant among many diverse southern ethnicities: Nu (66%), Tujia (45%), Yunnan Tibetan (77%), Dulong (32%), Shui (53%), Yunnan Yao (29%), Deang (44%), Yunnan Manchurian (56%)...

So kinda we need some other candidate, I fear. Even if O3a3c looks somewhat more "northernly" it still looks too southernly.

Maju said...

Ebizur: fair enough. But still it's a marker that shows ups more commonly among Han, something no other marked does.

Ebizur said...

Maju said,

"O3a3-P201* is clearly higher towards the South, what, in addition to what was said before, doesn't make it look like a Han or otherwise NE Asian marker at all.

In West Indonesia and Mainland/SEA (with or without Han) it is around 7%, what is also in line with what is found among Taiwan Han and Okinawans and well above what Japanese proper or Koreans have in any region. So I'd rather think Japanese O3a3* is of southern 'coastal' origin."

You seem to have confused O3a3*-P201 or O3a3-P201(xO3a3b-M7, O3a3c-M134) with O3a4-JST002611. According to the data that Natsuya has posted regarding the prevalence of these haplogroups in Japanese, Korean, and Taiwan Han populations, haplogroup O3a3*-P201 is most common in Koreans (approx. 10%), followed by western Japanese, central Japanese from Nagoya, Okinawans, and Taiwan Han with essentially equal frequency (approx. 6%), and it is somewhat less common in eastern Japanese from Kanto and Hokkaido. However, most of the data provided by Nonaka et al. 2007 are predictions of haplogroup membership based on Y-STR haplotype, so they should be referenced with due caution.

Haplogroup O3a4-JST002611, on the other hand, seems to occur most frequently in Hmong, Vietnamese, and Han populations. It also may be found in 5% to 10% of Tujia and Okinawans and in fewer than 5% of Japanese from other parts of Japan, Filipinos, and Koreans. Karafet et al. 2010 have not found any instance of O3a4-JST002611 in their samples of She and Yao people from southern China, Taiwan aborigines, Malaysians, Indonesians, and Oceanians.

Mason said...

to Ebizur:

I suppose O3a3-P201* found in different regions, such as Korea and Indonesia, may belong to different subclades which haven't been identified yet.

Mason said...

According to Hong Shi et al. 2005, O3-M122 is nearly 30,000 years old, and even O3a3c-M134 is about 25,000 years old. I think O3a3-P201* would be very old as well.

Mason said...

to Maju:

"Han" people were formed not ealier than 2,000 years ago. When "Han" first appeared in the historical records, they already had diverse haplogroups.

We have to know that most O3 subclades we see today are 20,000~30,000 years old, far predating Han, and even Sino-Tibetan as well as all the other language families in East Asia.

But I suppose, O3a3c-M134 and some O3 subclades were the leading haplogroups among early Han people.

Maju said...

Even if the lineages are old, I would have presumed that the people who gathered at the Yellow River to form the Northern Chinese Neolithic cultural continuum, eventually leading to the formation of China as state and the Han or Chinese as ethnicity, would have some specifics, whichever they are.

Specifics that I would imagine different, at least to some degree, from those of the southern Neolithic area around Hunan and other southern East Asian peoples.

However we can't find that specificity. Maybe it exists at further downstream levels but it's not apparent on what we can see. Maybe a haplotype NJ graph of O3 might give some clues. Only if we can detect such Han or (Chinese or Yello River people) specifics can we judge clearly whether there was any migration southwards as the empire expanded or not.

So far I have seen nothing, so I'm tempted to discard any meaningful migration... but without any marker that defines "Northern Chinese" we can't be totally sure.

Maju said...

Also, on a side note, I just got notice of a new paper (open access) focused on an Austroasiatic hunter-gatherer enthicity of Thailand, the Mlabri, but that includes some interesting comparisons of autosomal DNA between diverse East Asian ethnicities, from Japan and China to various SEA ethnicities.

I find particularly interesting the NJ tree of figure 4, that places most Hmong-Mien closer to Chinese and Japanese than to other SEA ethnicities. The other more or less clear groupings are:
- Tai Kadai and about half of Austroasiatics
- Karen (TB) and the other half of Austroasiatics
- Htin and Mlabri (both Austroasiatics)

A different tree (ML) is at figure 7, that clusters in four somewhat different branches:
- Japanese, Chinese and Hmong Mien
- Tai-Kadai (closest to the one above)
- Austroasiatics and Karen
- Mon (AA) - totally apart from the rest

It's like you can discern several SEA groups but can't discern NEA peoples from them (or at least from some of them, specially Hmong-Mien).

Maju said...

"You seem to have confused O3a3*-P201 or O3a3-P201(xO3a3b-M7, O3a3c-M134) with O3a4-JST002611. According to the data that Natsuya has posted regarding the prevalence of these haplogroups in Japanese, Korean, and Taiwan Han populations, haplogroup O3a3*-P201 is most common in Koreans (approx. 10%)"...

It seems I did. But only re. the data posted by Natsuya, and not the Karafet data (checked both now).

Still Philippines and Sumatra are higher and (in wait from a NJ haplotype tree or some other clarification) it still gives the impression that this lineage has a southernly origin in ISEA.

Only O3a3c-M134 looks rather northernly (and the 27% of Koreans ratifies this).

terryt said...

"Would you mind to back this claim with some real data".

Here, for the fourth or fifth time, are some references to the physical appearance of the Hoabinhian people of South China and SE Asia:

http://books.google.co.nz/books?id=5eEASHGLg3MC&pg=PA286&lpg=PA286&dq=hoabinhian+melanesia&source=bl&ots=E_qu0HH7ek&sig=oylw1B3EBWdfWTJpeg9JtYiMh7Y&hl=en&ei=tMOmS9aRDJOMtAOtzcEi&sa=X&oi=book_result&ct=result&resnum=2&ved=0CAgQ6AEwAQ#v=onepage&q=&f=false

From that, 'Hoabinhian skeletons are reported from a number of Vietnamese and Malaysian caves, all conforming to an Australian/Melanesian range in pheneotype rather than to a Mongoloid one'.

http://www.jstage.jst.go.jp/article/ase/116/3/116_201/_article

Abstract: 'An excavation at the cave site of Hang Cho in northern Vietnam resulted in the discovery of a terminal Pleistocene human skeleton in a relatively good state of preservation. The material culture from this site belongs to the pre-ceramic Hoabinhian period. An AMS radiocarbon date on a tooth sample extracted from this individual gives a calibrated age of 10450 ± 300 years BP. In discussions of the population history of Southeast Asia, it has been repeatedly advocated that Southeast Asia was occupied by indigenous people akin to present-day Australo-Melanesians prior to the Neolithic expansion of migrants from Northeast Asia into the area. Cranial and dental metric analyses were undertaken in order to assess the biological affinity of early settlers in this region. The results suggest that the Hang Cho skeleton, as well as other early or pre-Holocene remains in Southeast Asia, represent descendants of colonizing populations of late Pleistocene Sundaland, who may share a common ancestry with present-day Australian Aboriginal and Melanesian people'.

Note: 'the Neolithic expansion of migrants from Northeast Asia into the area'.

So the Hoabinhain people originally looked Melanesian, or Australian. Certainly not Mongoloid. And you have claimed, yourself, that the 'Negritos' represent a remnant of the pre-Mongoloid population in SE Asia. The Mongoloid look in the region is post-Hoabinhian.

"As I posted elsewhere, I can't find any single meaningful lineage that is Han-specific".

That's a problem. The consensus here is that there is no movement at all of male haplogroups from the north into SE Asia. The conclusion must be that just women moved in large numbers, and changed the phenotype. Or the genotype moved unaccompanied by either male or female haplogroups. Either scenario is unlikely.

We're left to assume that the SE Asians changed substantially through some unknown mysterious process, sometime after 10,000 years ago, to become substantially more like their Northeast Asian contemporaries (Mongoloid-looking).

terryt said...

"So far I have seen nothing, so I'm tempted to discard any meaningful migration..."

I can understand people in the region changing without any migration. But changing to look more like the people further north?

Maju said...

You are basing everything in anthropometry but you lack of a contemporary reference from "further north". There's no single pre-Neolithic skull that conforms to the Mongoloid archetype.

Also I'm not so sure that Hang Cho skull looks particularly Melanesian or "Australoid": he looks just archaic, like all skulls everywhere before c. 15 or 13 Ka ago. He's quite ortognathous, for example, while Papuans and Australian Aboriginals are extremely prognathous. Also he shows nothing of the ruggedness of typical Australian Aboriginal skulls, nor the depressions at the sides of the forehead either. You'd have major difficulties classifying this guy as Australoid or Papuoid.

The main problem for anthropometric conjectures of the kind you make is that you don't provide either for a possible source and that all known skulls further north before Neolithic are similarly non-Mongoloid.

Anyhow, even the typical Australian Aboriginal skull type is recent and does not show up before c. 7 Ka ago.

Maju said...

Take a look at this paper by Brown. It's very interesting: there's no "Mongoloid" but also no "Australoid" skull in East Asia before Neolithic.

However they could be close to Jomon, Ainu and Eskimo types by PC analysis. And yes, Eskimos cluster apart from either "Americanoids" and "Sinoids" in this graph.

In the caricature iconography, it might be said that Upper Cave 101 might resemble Buriats, while Minatogawa and Liujiang are closest to each other and might remind something of Jomon types (but not modern Japanese, not at all).

In the hierarchical graph, Minatogawa is the most Mongoloid, or should I say modern?, of all the archaics, clustering again close to Jomon. But the rest of the archaic skulls are all apart, while Australian Aboriginal moderns lay well inside the modern "Mongoloid" group, further inside it than the Buriat sample.

It's a total mess: there are no "Mongoloids" before Neolithic... except in America, arguably.

terryt said...

Thanks for the Peter Brown link.

"There's no single pre-Neolithic skull that conforms to the Mongoloid archetype". And
"all known skulls further north before Neolithic are similarly non-Mongoloid".

As Peter points out in the paper. But his paper doesn't consider Hoabinhian skulls at all. And we're specifically talking about SE Asia (Indonesia) here.

"It's a total mess: there are no 'Mongoloids' before Neolithic... except in America, arguably".

Interesting statement on p. 120: 'Is it a possibility that migration across the Bearing [sic.] Straits went in two directions and the first morphological Mongoloids evolved in the Americas?' Personally I doubt the Americas, but it could easily be somewhere north of China. So the Mongoloid phenotype may actually have evolved even further north than China. In which case it would be an immigrant into China. This would explain why the Aborigines, Jomon, Ainu and Eskimo are all outliers.

"there's no 'Mongoloid' but also no 'Australoid' skull in East Asia before Neolithic".

Makes sense if the above is correct.

"he looks just archaic, like all skulls everywhere before c. 15 or 13 Ka ago".

Certainly not 'Mongoloid' though. So the Mongoloid phenotype is immigrant into SE Asia. No-one claims it evolved there and moved north, becoming more 'distinctive' as it did so. So what haplogroup(s) was involved?

"even the typical Australian Aboriginal skull type is recent and does not show up before c. 7 Ka ago".

From the paper, 'Late Pleistocene Aborigines are stll clearly recognisable as Aborigines'. So the recent 'typical Australian Aboriginal skull type' is recognisably connected to ancient Australian ones. Unlike ancient Chinese skulls. And certainly not Hoabinhain skulls and modern SE Asian populations.

"Also he shows nothing of the ruggedness of typical Australian Aboriginal skulls"

Almost certainly more 'Melanesian' than 'Australian'.

Another interesting claim in Peter's paper is this: 'Sometime after Xujiayao (Chen et al. 1982; Wu and Wu 1985) modern people appear in China, with the earlier hominid fossils ... anatomically intermediate between H. erectus and H. sapiens'. How could that be?

Maju said...

"But his paper doesn't consider Hoabinhian skulls at all".

Nope. But it considers an Australian skull from 10 Ka. Anyhow, it'd be the same: there are no "Mongoloids" (if such category makes any sense at all, what I doubt) before Neolithic.

"Interesting statement on p. 120: 'Is it a possibility that migration across the Bearing [sic.] Straits went in two directions and the first morphological Mongoloids evolved in the Americas?' Personally I doubt the Americas, but it could easily be somewhere north of China".

Actually he does not dwell in the "Mongoloid" nature of NA skulls. Luzia is the oldest one (only 11 Ka ago) and is again said to be "Australoid", what in practice means: "we have no damn idea, she looks archaic". Kennewick man is from 8-5 Ka ago and still not typically Mongoloid. Not even most modern NAs can be said to be typically Mongoloid, even if some (scattered) groups are more clearly so.

I don't think that there is any reason to consider that there is a single Mongoloid type, much less that it has "migrated" from some unknown place, precisely where no skulls are found and, much less, that such place was in "the North" (why? Neolithic is a "southern" phenomenon: the far north does not allow for farming nor has any Neolithic record)).

I read someone speculating that the Yellow River Neolithic might have originated from Sichuan Mesolithic, what he related with the expansion of Sino-Tibetan. This may correlate with the detected Y-DNA affinities between the Tujia and the Han... but can't be extrapolated to other populations.

I am thinking that there was maybe an "Austro-Tai" (Austronesian+Kradai, which share an autosomal component, as well as with Southern Han) in Hunan area but that would exclude Asutroasiatics and would mean these are rooted in the Hoabinhian and would have been involved in their own Neolithic facies. Tentative, eh!

Whatever the case, I don't think there's enough evidence (or rather any clear evidence of any kind) to support massive migrations. Chienese cluster here and the peoples of the Pacific arch cluster there, along with Amerindians. Siberian peoples don't cluster well with either group. We may well be misled by some traits that can easily be transformed with minimal blood dilution (or even without any dilution at all).

"So the Mongoloid phenotype is immigrant into SE Asia".

That makes no sense whatsoever. It's much more likely that it coalesced there, even if it only did recently, maybe for mere epigenetic reasons, triggered by nutrition or whatever.

"From the paper, 'Late Pleistocene Aborigines are stll clearly recognisable as Aborigines'. So the recent 'typical Australian Aboriginal skull type' is recognisably connected to ancient Australian ones".

Look at the graphs. He probably means the Roonka skull which is not dealt with there and belongs to 7 Ka ago only.

East Asian Neolithic is older than that.

"Almost certainly more 'Melanesian' than 'Australian'".

For me, he looks "Caucasoid", if anything. He has no prognathism, so he can't cluster with Melanesians, Australians nor Africans, except maybe North Africans. He could cluster well with some dolicocephalic Iberians I know, hehe.

He even has a markedly divided chin, a trait almost only found in Europe.

"Another interesting claim in Peter's paper is this: 'Sometime after Xujiayao (Chen et al. 1982; Wu and Wu 1985) modern people appear in China, with the earlier hominid fossils ... anatomically intermediate between H. erectus and H. sapiens'. How could that be?"

It's probably not for real. Remember that the Chinese Academy was strongly multirregionalist. But last time I checked multirregionalism is dead.

terryt said...

"Remember that the Chinese Academy was strongly multirregionalist".

Peter Brown is Australian, and very much opposed to the multiregionalists.

"He probably means the Roonka skull which is not dealt with there and belongs to 7 Ka ago only".

Why are you putting words into his mouth? Read his article again.

"I don't think that there is any reason to consider that there is a single Mongoloid type"

So why is it so easy to recognise that a particular person comes from East Asia?

"That makes no sense whatsoever. It's much more likely that it coalesced there"

And that's why the Mongoloid phenotype becomes more 'pronounced' the further north you go?

"there are no 'Mongoloids' (if such category makes any sense at all, what I doubt) before Neolithic".

And there are no Mongoloids in SE Asia until very late in the Neolithic.

"the far north does not allow for farming nor has any Neolithic record))".

Are you absolutely sure that the Mogolian and Turkic-speaking people who moved west some 1000 years ago, and left their descendants so widespread in Central Asia, were farmers?

Maju said...

"So why is it so easy to recognise that a particular person comes from East Asia?"

Because they usually have "chinky eyes". There are exceptions though, the former owner of the Chinese shop across the street had no epicanthic fold and looked very ambiguous. Certainly not your typical East Asian, except for the yellowish tinge of the skin maybe.

So what does they make them "Mongoloid"? Peculiarities of the skin first of all, which can't be identified in any skull.

The "Mongoloid" skulls once and again show a wide range. See for example the Structure run Dienekes did: while Norwegians and Egyptians clustered together at all levels, "Mongoloids" diverged in many clusters from very early (ref1,ref2).

Not to mention that Andeans, for example, don't look Mongoloid at all, with those big noses, often more "Caucasoid" than any West Eurasian and rare epicanthic folds.

In the exercise at Dienekes (see ref 1 above), I could count 6 different "Mongoloid" types, not counting Polynesians (who diverge very clearly at the beginning), while the Caucasoid cluster remains compact even at K=14.

I can identify the following real groups:
- Andeans (Peru)
- Californians (Santa Cruz)
- Pacific Coastal East Asians
- Buryat
- Eskimo
- Anyang (Neolithic Chinese)

In Brown's paper again Chinese and Coastal peoples cluster separately, while Eskimos and Buryats always cluster on their own, with Buryats even more distant than Australian Aboriginals.

So it seems to me that the "Mongoloid type" is just skin-deep, at least to a large extent.

Btw, I was watching yesterday a documentary on South Africa and the people was sooo extremely diverse, that some almost looked more akin to East Asians or West Eurasians than to the person sitting just beside them. They were all in the Negroid-Khoisanid range by usual standards but some looked more like some Eurasians than their peers, specially if we ignored some skin-deep traits such as skin color.

Maju said...

"Peter Brown is Australian, and very much opposed to the multiregionalists".

But that's not his claim but one by Chinese scholars in the 1980s that he merely mentions passing by.

"Why are you putting words into his mouth? Read his article again".

Seems he says that but the graphs don't say that, nor says the skulls I have seen at Peter Brown's site either. So...

"And that's why the Mongoloid phenotype becomes more 'pronounced' the further north you go?"

It does not seem to be the case, with Buryats and Eskimos consistently clustering apart.

<<"there are no 'Mongoloids' (if such category makes any sense at all, what I doubt) before Neolithic".

And there are no Mongoloids in SE Asia until very late in the Neolithic. >>

Not sure what you mean. What is "very late" and why you claim that. Are there Neolithic "non-Mongoloids" in SEA in the Neolithic? References, please.

"Are you absolutely sure that the Mogolian and Turkic-speaking people who moved west some 1000 years ago, and left their descendants so widespread in Central Asia, were farmers?"

LOL, the horse was not domesticated (barring the possible Magdalenian episode) until c. 5-4000 BCE. What are you talking about?

I know that the only hunter-gatherers remaining in non-tropical Eurasia are precisely in Siberia, including some cousins (Nganasan) of other herder peoples, as are the Finnic ones.

You always brush to your side, never presenting evidence (even if your claims are far-fetched) and ignoring facts as Buryats being more distant than Australian Aboriginals from the main "Mongoloid" cluster.

That's not a honest intellectual discussion. I'm tired of arguing with single-minded freaks, really.

Maju said...

Let's check more in detail your (Brown's?) claim of Australian continuity. From the paper:

"a combined
terminal Pleistocene group from Coobool Creek, Nacurrie and Kow
Swamp"

That's the Pleistocene (late Pleistocene) sample. You can check all three (though Nacurrie has no specific page) at Brown's site.

Kow Swamp is dated to 13-10,000 BP. Coobool Creek to 14-7000 BP. No idea about Nacurrie but seems it should be from the same time frame at the very end of the Pleistocene.

For some reason, even if Brown criticizes others for not using Keilor, he also decided to ignore it in this paper. Keilor is the less Australian looking of all and rather resembles Liujiang and other Eurasian archaic remains (Cro-Magnon?)

Whatever the case, we have a sample that is biased (if anything) in favor of continuity with modern AAs.

But the results are that

"All of the recent Australian Aborigines were correctly allocated
to their group, and only 1 of the terminal Pleistocene Aborigines was
allocated to the recent Aboriginal group".

Ouch!

"Recent and terminal
Pleistocene Australian Aborigines can be clearly distinguished from each
other, but not to the same extent that Upper Cave 101 is separated from
modern and Neolithic Mongoloids".

So they are different, even if not as different as Upper Cave 101.

"Neither Liujiang or Upper Cave 101 are Australoid in appearance".

Aha!

"The first split is between Upper Cave 101 and all of the other samples. Liujiang, followed by the terminal
Pleistocene Australian Aboriginal group are the next to split and they
are separated from the next branch containing Minatogawa 1 and the
Jomon".

This refers to the ML tree, where all archaic samples cluster apart, while modern Australian Aborigines cluster inside the Mongoloid cluster, at least if Buryats are Mongoloids.

So the main difference is not between Australoids and Mongoloids but between Pleistocene and Holocene. How curious!

Actually the main difference is between UP-101 and everybody else.

Then between Liujiang and everybody else.

Then between the combined Late Pleistocene sample from Australia and everybody else.

So Pleistocene Australians are very different from everybody, though less different than UC-101 and Liujiang. By comparison modern AAs are "typical Mongoloids", so to say. Certainly more typical than Buryats.

"There is also nothing particularly “Australoid” about the morphology of Upper Cave 101, Liujiang and Minatogawa 1. If by “Australoid” Coon (1962) was suggesting like recent Australian
Aborigines, then tooth dimensions and facial prognathism alone make
them very unlikely candidates (Brown 1989)".

How come? Coon wrong? Impossible: he's God reincarnated! [sarcasm intended: I think Coon was just a stupid Nordicist whose typology is useless]

"No one should be surprised that these three fossils fall outside
the modern East Asian range of variation, after all terminal Pleistocene Australian Aborigines also fall outside the recent Aboriginal range in cranio-facial size and, to a lesser degree, shape (Brown 1989, 1992)".

Uh, oh!

I couldn't find your quote. No doubt it's there somewhere but I did find lots of quotes that say otherwise and in general illustrate my viewpoint and do not contradict it at all.

terryt said...

"Because they usually have 'chinky eyes'".

Yes. They're distinctive. And they all look similar to each other in many respects, and easily identified as East Asian. What's your problem?

"I could count 6 different 'Mongoloid' types"

They vary, but are still recognisable as East Asians.

"the former owner of the Chinese shop across the street had no epicanthic fold and looked very ambiguous".

Almost certainly from southern China.

"So the main difference is not between Australoids and Mongoloids but between Pleistocene and Holocene. How curious!"

You find it curious only because it doesn't fit your belief. I find it prefectly easy to understand. And what have Australain Aborigines to do with the change in SE Asia? The people in SE Asia before the Mongoloid entry are far more likely to have resembled contemporary populations in New Guinea. The fact that all populations have changed since that time is hardly surprising. The relevant detail is that through South China and SE Asia (and much of North China) they have changed towards a single type: Mongoloid.

"Are there Neolithic 'non-Mongoloids' in SEA in the Neolithic? References, please".

I've already provided you with adequate references, but you prefer to re-interpret the evidence to suit your own belief, so I'm wasting my time.

terryt said...

"Not sure what you mean. What is 'very late' and why you claim that".

So back to the original subject.

From a link I provided earlier: 'it has been repeatedly advocated that Southeast Asia was occupied by indigenous people akin to present-day Australo-Melanesians prior to the Neolithic expansion of migrants from Northeast Asia into the area'. So it seems he at least has no problem with a southerly movement within China as far as SE Asia.

Interesting comment in Wiki (re. Y-hap O3) regarding southward migration: 'Among all the populations of East and Southeast Asia, Haplogroup O3 is most closely associated with those that speak a Sinitic, Tibeto-Burman, or Hmong-Mien language. ... The Sinitic and Tibeto-Burman language families are generally believed to be derived from a common Sino-Tibetan protolanguage, and most linguists place the homeland of the Sino-Tibetan language family somewhere in northern China. The Hmong-Mien languages and cultures, for various archaeological and ethnohistorical reasons, are also generally believed to have derived from a source somewhere north of their current distribution, perhaps in northern or central China'.

So Y-hap O3 is probably northern in Origin.

A further quote from Wiki regarding Y-hap O as a whole: 'Haplogroup O* lineages, which belong to Haplogroup O but do not display any of the later mutations that define the major subclades O1, O2, and O3, can still be detected at a low frequency among most modern populations of Central Asia and East Asia. For example, a broad survey of Y-chromosome variation among populations of central Eurasia found haplogroup O-M175*(xO1a-M119,O2a-M95,O3-M122) in 2.5% (one out of 40 individuals) of a sample of Tajiks in Samarkand, 4.5% (1/22) of Crimean Tatars in Uzbekistan, 1.5% (1/68) of Uzbeks in Surkhandarya, 1.4% (1/70) of Uzbeks in Khorezm, 12.5% (2/16) of Tajiks in Dushanbe, 1.9% (1/54) of Kazakhs in Kazakhstan, 4.9% (2/41) of Uyghurs in Kazakhstan'

So could you please remind me: which of these groups is southern? Admittedly some could actually belong to labeled haplogroups.

And O2 divides into northern and southern groups. Again Wiki: 'Haplogroup O2 is also notable for the fact that it can be divided into two major subclades that show almost completely disjunct distribution. One of these subclades, Haplogroup O2a (M95), is found among some (mostly tribal) populations of South and Southeast Asia, as well as among the Khmers of Cambodia and the Balinese of Indonesia. ... The other major subclade, Haplogroup O2b (SRY465, M176), is found almost exclusively among the Korean,Japanese,Thai,Vietnamese and Indonesian'.

So did O2 move north or south? The O2b, at least, in the south is probably associated with the (possibly coastal) movement south of microlithic technology that eventually gave rise to the Austronesian expansion. And the O2a may have moved south through China over land. The diversity of O in Indonesia is as likely to be because of separate routes of entry rather than being proof of origin.

Maju said...

<<"So the main difference is not between Australoids and Mongoloids but between Pleistocene and Holocene. How curious!"

You find it curious only because it doesn't fit your belief.>>

It fits my understanding reasonably well. There was irony intended in that exclamation.

If all moderns, including non-Mongoloids cluster together versus the archaics, then the change is one happening in time not space. There are no migrations (at least not needed), just change that needs some other explanation, such as diet or climate.

"I find it prefectly easy to understand".

Really? Let me guess: some magical people from some remote place where nobody has found anything yet, like the tundra, expanded and colonized not just East Asia to the last corner but also Australia.

"And what have Australain Aborigines to do with the change in SE Asia?"

That's the crux of the matter. They are more "Mongoloid" than Buryats but not any closer to their likely ancestors of the Murray River late Pleistocene! Just like East Asians are not like Upper Cave 101, Minatogawa nor Liujiang.

"The people in SE Asia before the Mongoloid entry are far more likely to have resembled contemporary populations in New Guinea".

Nope: low prognathism.

"The relevant detail is that through South China and SE Asia (and much of North China) they have changed towards a single type: Mongoloid".

Or you could say towards a single type: Australoid.

The key is that we do not find such single type anywhere unless you define it as "present day human". "Mongoloids" are highly diverse and "Australoids" cluster with them.

"From a link I provided earlier: 'it has been repeatedly advocated that Southeast Asia was occupied by indigenous people akin to present-day Australo-Melanesians prior to the Neolithic expansion of migrants from Northeast Asia into the area'".

He does seem to have a problem with that when he declares that Liujiang is not Australoid.

He does not seem persuaded either that Wadjak 1 and 2 (Neolithic/Epipaleolithic people from Java) are Australoids. Even if he may consider possible the affinity with Keilor, Keilor is itself proclaimed non-Australoid and again strikingly ortognathous.

What is somewhat clear to me is that Keilor, Liujiang and Wadjak are similar, distinct from any modern morphotype, with more robust skulls than modern types and reminding of Cro-Magnons if anything.

The robustness of these and all other archaic skulls, may provide for a clue in the change of morphotype: because one could easily expect lighter skulls to grow to different shapes than robust ones. This robustness or lack of it may easily be coded by a single gene, which in turn may be affected by environmental epigenetics, such as diet rich or lacking in animal protein.

So I can't accept that skull morphotypes alone say anything, at least nothing clear.

Maju said...

"Haplogroup O3 is most closely associated with those that speak a Sinitic, Tibeto-Burman, or Hmong-Mien language".

Or Austronesian, lol.

That's an unsourced opinion that only makes sense when we consider other O lineages and based only in frequence (i.e. Sino-Tibetans have less O2 and O1, that is: less diversity).

"So Y-hap O3 is probably northern in Origin".

Pseudo-science at its worst!

Look at the subclades and where there is most O3 diversity. I'm sure that your darling Wikipedia links to some relevant papers, some of which you should surely read before opening your mouth with such self-complacient statements.

For instance the Hong Shi 2005 paper, titled precisely: 'Y-Chromosome Evidence of Southern Origin of the East Asian–Specific Haplogroup O3-M122'.

Please!

"A further quote from Wiki"...

Enough Wiki. Read a paper or two!

"So could you please remind me: which of these groups is southern?"

O1, O2 and O3 are.

"So did O2 move north or south?"

No. It represents a wave of S->N coastal migration. The fact that it's found among so diverse SE Asian peoples is at least some evidence, right? Moreso when its brother lineage O2a is only found in SE Asia.

Even Y-DNA D has a southern origin.

Everything in East Asia (at least) has a likely southern origin. Nothing that I can think of has a northern origin (and has expanded southwards in any significative manner). It may sound exaggerated but that's what the data says.

Ebizur said...
This comment has been removed by the author.
terryt said...

"I couldn't find your quote. No doubt it's there somewhere but I did find lots of quotes that say otherwise and in general illustrate my viewpoint and do not contradict it at all".

You didn't look very hard. It's in the discussion, at the end of the article. His conclusions are obviously of no interest to you at all. So when you write:

"That's the crux of the matter. They are more 'Mongoloid' than Buryats but not any closer to their likely ancestors of the Murray River late Pleistocene! Just like East Asians are not like Upper Cave 101, Minatogawa nor Liujiang",

you demonstrate you have no idea at all of SE Asian prehistory.

"There are no migrations (at least not needed), just change that needs some other explanation, such as diet or climate".

Across a whole region of widely diverse habitats?

"The fact that it's found among so diverse SE Asian peoples is at least some evidence, right?"

Yes. A huge spread of population. A migration.

"Or you could say towards a single type: Australoid".

You could say that, but it would be rubbish.

"'Mongoloids' are highly diverse and 'Australoids' cluster with them".

You are the only person in the world who would confuse a 'Mongoloid' with an 'Australoid'.

Ebizur said...

Truth be told, I do not recall having ever seen even one confirmed case of O*-M175(xO1-MSY2.2, O2-P31, O3-M122), just as I have never seen a confirmed case of NO*-M214(xN-M231, O-M175). However, Hurles et al. 2005, Hammer et al. 2006, Regueiro et al. 2006, and Xue et al. 2006 have reported finding the following cases of O-M175(xO1a-M119, O2-P31/M268, O3-M122):

2/34 = 5.9% Li (Xue et al. 2006)
3/52 = 5.8% Manchu/Liaoning (Hammer et al. 2006)
1/20 = 5.0% Zhuang/Guangxi (Hammer et al. 2006)
1/30 = 3.3% Han/Lanzhou (Xue et al. 2006)
1/31 = 3.2% Oroqen (Xue et al. 2006)
1/33 = 3.0% North Iran (Regueiro et al. 2006)
1/39 = 2.6% Daur (Xue et al. 2006)
1/39 = 2.6% Taiwan aborigines (Hurles et al. 2005; "The Taiwanese population sample represents individuals pooled from four different aboriginal groups.")
1/105 = 1.0% Tibet (Hammer et al. 2006)
1/149 = 0.7% Mongolia (Hammer et al. 2006)

I have heard that at least one case of O1*-MSY2.2(xO1a-M119) has been found in Mongolia, so I guess that the 1/149 = 0.7% O-M175(xO1a-M119, O2-P31, O3-M122) Mongolia figure of Hammer et al. 2006 probably represents O1*-MSY2.2(xO1a-M119).

Maju said...

I'm not the only person in the world: I'm taking that from the Brown's paper we are discussing here, in particular the ML tree (fig. 5).

"You didn't look very hard".

I typed "Australoid" in the search feature read all paragraphs that included that word and quoted here most of them for you. Now try reading and understanding them.

I should have typed "Aborigines" but does it make any difference? It's one sentence versus all the damn article. I can't see a single datum that justifies that claim.

"Across a whole region of widely diverse habitats?"

Across the whole world maybe, because neither East Asians (no East Asians at all) nor Australian Aborigines look at all like their ancestors.

I have no idea why but the fact is that skulls seem to be much more plastic than what you think. There's no way to get anything straight just from skulls. We need genes and cultural artifacts.

Maju said...

And by the way: it's not "across a whole region". I have argued all the time here, that Mongoloids are hyper-diverse. You can see that in Brown's paper, you can see that at Dienekes' own work and you will see that when you look at them, unless you are of the type that thinks "all Chinese look the same".

Ebizur said...

By the way, it seems that Hurles et al. 2005, Hammer et al. 2006, and Xue et al. 2006 have tested only P31 as a marker of haplogroup O2. Regueiro et al. 2006 have tested only M268.

Maju said,

"No. It represents a wave of S->N coastal migration. The fact that it's found among so diverse SE Asian peoples is at least some evidence, right? Moreso when its brother lineage O2a is only found in SE Asia."

O2a-M95 has been found in Pakistan (Pashtun), Uzbekistan (Uzbeks in the Fergana Valley), northern China (Han, Evenk, Bonan, Dongxiang, and especially Qiang, Daur, and Salar), and Japanese from all parts of Japan. Cases of O2a-M95 also have been found in India, Nepal, and Oceania, but I will grant you the possibility that these may be ascribed to known or suspected recent influences from Southeast Asia.

Maju said...

"... but I will grant you the possibility that these may be ascribed to known or suspected recent influences from Southeast Asia".

Thanks. But the last thing I would dare claim is that it's "recent". Though in some cases it may be.

My only point is that the human dispersal in East Asia looks from all angles (mtDNA, autosomal DNA and even Y-DNA) a south to north process, possibly in two or more wavefronts that can be roughly identified with the the various major Y-DNA lineages.

Was there some back-migration North to South? Possibly but we still have to clarify its extent and timing and is not the massive obvious replacement that Terry defends, based on obsolete theories.

There's no clear north-south march, much less a massive one. We can't identify the lineages involved, nor associate them linearly (nor even loosely in most cases) with linguistic groups and other "marches of the titans".

There is also no clear identification of any possible source population for the Mongoloid morphotype or rather morphotypes. So, while maybe risky, but considering all the rest, specially genetics, I prefer to think that there was no such replacement, much less from some unidentified northern population and much less with all northern representatives clustering apart, in some cases further apart than Australian Aboriginals.

Instead there may have been processes of growing interaction, maybe with more limited migrational flows, that helped that apparent homogenization of morphotypes by mere normal admixture (i.e. sexual reproduction) plus whatever unidentified factors, which may well be environmental. It's a complex and poorly understood matter and I don't wish to discuss it further. Besides we really don't know anything about the most definitory "Mongoloid" traits, such as epicanthic fold or skin pigmentation, among fossil remains, which anyhow are scarce and maybe even not representative.

terryt said...

"just as I have never seen a confirmed case of NO*-M214(xN-M231, O-M175)".

Perhaps the haplogroup broke abruptly (after the M214, P188, P192, P193, P194 and P195 mutations) into four: O1, O2, O3 and N.

"Moreso when its brother lineage O2a is only found in SE Asia."

I'll remind you that O's brother lineage, N, is spread right across Northern Eurasia.

"There's no way to get anything straight just from skulls. We need genes and cultural artifacts".

Exactly. So why are you placing so much emphasis on skulls?

"Across the whole world maybe, because neither East Asians (no East Asians at all) nor Australian Aborigines look at all like their ancestors".

Maju. If that was the case why would Peter Brown claim in his summary, 'Late Pleistocene Aborigines are stll clearly recognisable as Aborigines'? The change was not huge in Australia, unlike SE Asia.

"I have argued all the time here, that Mongoloids are hyper-diverse".

Of course they are. Just like any other widespread group. However, as a group, they tend to have straight black hair, yellowish skin, relatively hairless bodies and a pronounced eyefold. You are claiming that during the Neolithic this suite of characteristics spread suddenly and miraculously across the whole of Eastern Eurasia, from Indonesia in the south to the Russian Far East in the north, without any migration or genetic expansion of any sort being involved. This is difficult to accept. Surely a change through migration is a much more believable scenario. Especially as there is evidence for such an expansion, in spite of what you believe.

"But the last thing I would dare claim is that it's 'recent'".

Relatively recent. Almost certainly with the Austro-Asiatic or Tibeto-Burman language expansions.

"is not the massive obvious replacement that Terry defends, based on obsolete theories".

Just because those theories have become generally accepted over the last forty years doesn't mean they are obsolete. They have not been overturned at this stage. Nor are they likely to be, I would guess. And the genetic evidence can certainly be interpreted to uphold the idea, although you prefer, for some obscure reason, to deny that possibility.

"There's no clear north-south march, much less a massive one".

And there's no clear south to north one either. Humans have probably been moving north and south through the region since H. erectus times.

"There is also no clear identification of any possible source population for the Mongoloid morphotype or rather morphotypes".

I would suggest scientists check out any ancient human remains they might find along the Tibet/Inner Mongolia/China border region.

"It's a complex and poorly understood matter and I don't wish to discuss it further".

I can understand your motive there. It's because your belief system can't cope with the sort of hypothesis require to explain such a widespread change of pheneotype. Unlike the 'obsolete theories' you so despise.

Maju said...

"So why are you placing so much emphasis on skulls?"

Am I? All this section was triggered by you because you argued:

1. That skulls "demonstrate" that there was replacement

2. That this replacement had a north-south direction

Brown clearly demonstrates that there is no known source for this hypothetical replacement, certainly not in the north. And now you seem to agree skulls don't matter so much.

Then finito (I hope).

"Perhaps the haplogroup broke abruptly (after the M214, P188, P192, P193, P194 and P195 mutations) into four: O1, O2, O3 and N".

That is not what the phylogeny says. N seems to be a minor lineage from the old times or pre-O (NO) that made its way to the North and thrived there.

"Maju. If that was the case why would Peter Brown claim in his summary, 'Late Pleistocene Aborigines are stll clearly recognisable as Aborigines'? The change was not huge in Australia, unlike SE Asia".

You can insist in that all you want but that's not what the data says. He just looks at the skulls and gets that subjective impression somehow but nothing in his own work supports it.

Look at the data because I'm not accepting selected sentences that amount to nothing and are cotradicted by others along the whole text, as quoted above.

"However, as a group, they tend to have straight black hair, yellowish skin, relatively hairless bodies and a pronounced eyefold".

A lot of "Mongoloids" have curly hair (Tibet, Amazonia, etc.) It's somewhat atypical but not a definitory trait.

"yellowish skin"

This trait must have coalesced (founder effect?) early in the human dispersal process because the known pigmentation traits of East Asians seem to arise then (ref- Coop'09).

"a pronounced eyefold"

Epicanthic fold. A trait that is not totally absent in other populations (Khoisan, North Europeans, etc.) nor universally present among East Asians, much less Native Americans.

Maju said...

"You are claiming that during the Neolithic this suite of characteristics spread suddenly and miraculously across the whole of Eastern Eurasia"...

No, please! I say that such traits are flesh and were probably widespread before Neolithic. Whatever the case it's something that skulls will never tell but the patterns of the derived East Asian MC1R haplotype do suggest that these traits are old, very old.

"Just because those theories have become generally accepted over the last forty years doesn't mean they are obsolete".

They are obsolete because thy clash frontally with the genetic evidence and are based on nothing but conjectures.

"And the genetic evidence can certainly be interpreted to uphold the idea"...

I don't think it can. Nor you have made any meaningful effort for it. Just brief sentences without much (or any) backing.

So please: start your own site and dedicate some time to elaborate on this stuff, instead of just arguing and arguing in cricles without evidence.

"And there's no clear south to north one either".

Yes. It's very clear in the deep layers: all the diversity accumulates in the south and all haplogroups seem to originate there at one or another step (or most commonly all).

You need really to have strong phobias and filias: a general "Hyborianism", so to say, an obsession with waves of humans arising once and again at the Arctic. I think this is caused by implicit racism, acknowledged or not in very Coonish style. And in any case it's not supported by anything nor makes any sense (because the arctic areas can never support enough people to displace the "masses" living at the tropics).

"It's because your belief system can't cope with the sort of hypothesis require to explain such a widespread change of pheneotype".

The cycle of circular arguing is completed. Earlier you said: "So why are you placing so much emphasis on skulls?" It's only because you insist, of course. And now you close your post insisting again.

For me the data evidences that there is now enough variability to imagine every possible scenario. There is lack of sampling in modern SEA peoples anyhow.

Andrew Oh-Willeke said...

I have to agree that the Bali results don't seem exceptional in light of the history (per Wikipedia on Bali):

"The Hindu Majapahit Empire (1293–1520 AD) on eastern Java founded a Balinese colony in 1343. When the empire declined, there was an exodus of intellectuals, artists, priests and musicians from Java to Bali in the 15th century."

While on a big picture India seems quite admixed, it is full of pockets of ideosyncratic heritages preserved for long periods by caste endogamy, and the pool from which the Hindus colonization took place probably came from a few castes in one or two areas. The isolated places where these Y-DNA haplotypes are seen in modern India were probably the places from which the Bali colonists were drawn and I wouldn't be surprised to see historical records or cultural links between them if you knew what you were looking for. The J1 is more likely an artifact of Arab Muslim trade or religious links.

Andrew Oh-Willeke said...

Sanasai Island is aka Green Island:

"Located in Pacific Ocean southeast of Taiwan, Green Island is under the governance of Taitung County authorities. It is the northernmost volcano island of Luzon arc, and has an area of 17 square kilometers during low tide. Situated in the sea at a distance of 33 kilometers off the southeastern coast of Taitung City (18 nautical miles), it is 73 kilometers north of Lanyu (41 nautical miles). Green Island was known as Fire-Burned Island, Chicken Heart Island, Qing-zi Island or Sanasai during the old days.

 Green Island is 4 kilometers long from the north to the south and 3 kilometers wide from the east to the west. Outline of the island is an inverted trapezoid shape. There are lots of hills on the island; 60% of the total area is land with slopes above 30%. Huoshao Mountain, the highest peak on the island (281 meters), is located in the center of the island. It descends gradually towards the coast with many precipitous sea cliffs. Most of the plain areas are zonally distributed along the coast. Broad flat areas are concentrated around Gong Guan and Zhong Liao where have been developed as the major villages.

 The archaeological excavations have proved that human activities at Green Island can be traced back to 4,000 years ago and indigenous cultures had been found around the island since then till the very late period. In 1799 (the 4th year of Jia-Qing Reign, Qing Dynasty), Chen Bi-xian, a fisherman from Siaoliouciou, led a group of people to Green Island to exploit the land. Later, after Han people moved in, fights began to occur between them and Dawu people. As the exploitation of the Han people, Dawu people gradually moved backward to the area around Zhong Liao and finally withdrew from Green Island. During Qing Dynasty, Han people mainly relied on agriculture, supplemented by fishery and lumbering and this had made this island an autonomic and self-supporting closed society which was not under the control of Qing Government. During the Japanese colonization period, although the introduction of skipjack tuna of pole and line fishery and Katsuo industry had promoted the development of commerce and fishery, forests suffered unprecedented destruction. After 1961, Cheng Gong Fishing Port and Fu Gang Fishing Port (originally called Jialulan Port or Qielan Port) were successively accomplished and deer-raising industry also started to become prosperous. However, since 1991, the rapid development of sea and air transport, the fast growth of tourist population and the changes of social structure have all brought great impacts on the environment and ecology."

The island is about 200 km North of Batan Island in the Phillipines, and given ancient sea levels, it would have been hard to colonize in the paleolithic from the Phillipines. There is no time in the last couple million years that this would have been connected by land to there, or even within visual distance, but it would have been a visual distance trip at time in the human era of low sea levels and is mostly shallow sea from Formosa.

Likely the original Formosans sent a contingent there either from Formosa or China, which then recolonized Formosa.

Andrew Oh-Willeke said...

The J* in Bali and Vietnam probably both derive from Socotra where that is the predominant type, during Muslim empire expansion to Indonesia. Socotra would have been on the Indian Ocean trade route.

Maju said...

"The J* in Bali and Vietnam probably both derive from Socotra"...

Or not. J* is not a type it's the odd box.

Unless you have evidence that it's the case by haplotype, J* means nothing other than it's part of the very extended haplogroup J.

terryt said...

"The Hindu Majapahit Empire (1293–1520 AD) on eastern Java founded a Balinese colony in 1343. When the empire declined, there was an exodus of intellectuals, artists, priests and musicians from Java to Bali in the 15th century."

I'd be very surprised if the Y-hap H turns out to be anything other than an introduction from around that time. And the J (whatever haplogroup it is) is also likely to have been introduced by Arab traders.

Mason said...

Maju, there are some evidences which may support that O2a-M95* in Indonesia is pre-Austronesian but not necessarily Paleolithic.

It was suggested there was a wave of Neolithic dispersal from Indochina into ISEA before the Austronesian expansion from Taiwan. This wave of people might have brought O2a-M95* to western Indonesia and Borneo.

Crossing the Luzon Strait: Archaeological Chronology in the Batanes Islands, Philippines and the Regional Sequence of Neolithic Dispersal

Atholl Anderson

http://beta.nmp.gov.tw/main/07/7-3/3-3-2/2.025-045.pdf

"In reviewing the chronological data in relation to linguistic and archaeological evidence, the existence of at least two neolithic dispersals can be proposed (Figure 5). Neolithic I, if it can be called that, may be represented by expansion from South China through Thailand and Vietnam then through Malaya to Borneo, if not more widely, of basket or cord-marked ceramics amongst other types (but amongst which red-slipped pottery is scarce or absent). This seems to have occurred relatively early and it has been associated, in part at least, with the expansion of Austroasiatic languages (Higham 2004). It is not necessarily a neolithic defined exclusively by agricultural expansion. At Gua Sireh, in Borneo there is rice at about 2500 BC but at Nong Nor, in coastal Thailand, and quite widely in coastal Vietnam, there are middle-Holocene sites containing polished stone tools and pottery, but no sign of food production (Higham 2004, Su 1997)."

http://konglong.5d6d.com/userdirs/8/6/konglong/attachments/month_1006/1006180517cd2a7b2bb46c8dc1.png

Maju said...

Thanks for the link, Natsuya. It is a very interesting paper indeed.

The Neolithic expansion model reflected in the graph you selected would indeed fit reasonably well with the autosomal clustering found in the HUGO Consortium paper (discussed here and here). Though I'm unsure right now on what could this mean in terms of Y-DNA lineages (would have to review the whole matter - I lost track and forgot many critical details).

However it's worth noticing that Andersson himself does not seem very persuaded by the specifically Neolithic, much less demic, nature of these flows:

The conventional proposition, one of the demic-diffusion class of arguments (Cavalli-Sforza 2002:81), is that agricultural productivity fuelled levels of population increase which, in turn, propelled continuing dispersal (Diamond and Bellwood 2003). There are several problems here. One is the scarcity of evidence for early neolithic cereal agriculture in island Southeast Asia and the plausibility of arguments which suggest that it would have been difficult to transplant (Paz 2002). Another is that current evidence of the neolithic II expansion, suggests that at a regional level it occurred almost instantaneously in archaeological time, from about 1700 BC in the Philippines to 1500 BC in
Maluku and New Guinea, and what is more, through a series of very large islands with diverse agricultural possibilities. Unless the initial neolithic population was huge then the rate of growth cannot have been sufficient to drive dispersal so rapidly by the demographic mechanism inferred
.

Mason said...

Here's some more:

THE NEOLITHIC AND AUSTRONESIAN EXPANSION
WITHIN ISLAND SOUTHEAST ASIA AND INTO THE PACIFIC

Matthew Spriggs

http://arts.anu.edu.au/arcworld/resources/papers/Spriggs/2007/Spriggs_b2007.pdf

Given that most of the diversity of Austronesian languages is in Taiwan, any model linking the spread of that language family and the spread of Neolithic cultures would require that the dates for the beginning of the Neolithic in Taiwan should be substantially earlier than those in related areas to the south. They are indeed earlier by somewhere around 1,000 to perhaps 2,000 years. In a sentence: the diversity and pattern of subgrouping of the Austronesian languages are exactly mirrored in the chronology and pattern of nearly all of the Island Southeast Asian and Western Pacifi c Neolithic early pottery-using cultures. The one tracks the pattern of the other. The only area investigated so far where this may not be the case is in the major part of Borneo, and there are both linguistic and archaeological clues (see below) which have been suggested in the past as showing some earlier involvement of this area in the spread of a second, independent Neolithic movement linked to the separate spread of the Austro-Asiatic languages down through the Malay Peninsula. Hence we may need to talk of Island Southeast Asian “Neolithics.”

The early Neolithic cultural assemblages from Gua Sireh, Niah Cave and the Upper Birang River have in the past been suggested as having more in common with those of the Malay Peninsula and southern Thai Neolithic sites than the other sites in Island Southeast Asia discussed so far (Bellwood 1997: 236-241). In relation to this other Neolithic tradition Bellwood noted that the Malay Peninsula Neolithic was “a completely separate cultural entity from the islands” (Bellwood 1997: 219). The early Sarawak and Kalimantan sites contain cord or basketry-wrapped paddle impressed pottery, and the typically Island Southeast Asian red slip and/or incised wares are virtually absent. Later assemblages seem to fit much better the usual Island pattern, possibly suggesting a change in cultural and perhaps linguistic affiliation over time (cf. Chazine 2003: 49-50). Bellwood noted that there is a claimed Austroasiatic linguistic substratum in some Borneo languages and Austroasiatic influence is also suggested for Sumatra, but may be of much later date (Bellwood 1996: 483). These putative influences would of course provide a further link to Mainland Southeast Asia where languages of this family are found. More recently Bellwood has reconsidered his interpretation and now believes that the links of the earliest cord-marked pottery from Borneo are likely to be back to the Fine Corded Ware of Taiwan, a tradition that occurred in sites in Taiwan along with red-slipped pottery in its latest phases (Bellwood, pers. Comm. 2006). His research on this issue is ongoing.

Maju said...

Interesting. Thanks again.

For Spriggs it seems to be all a matter of rapid Neolithic diffusion from Taiwan, which produces a problem, as far as I can see: that the arrival of Neolithic to Taiwan is virtually simultaneous with the arrival to Philippines and Indonesia.

This process does not seem able to explain the linguistic one of Austronesian... unless the migration/cultural spread began not in Taiwan but in mainland SE China.

If so we might be in front of two related but different processes: the Malayo-Polynesian and the Taiwanese one. The high diversity of native Taiwanese languages might be then explained by a much more plural process of cultural/demic transference than the one leading to Malayo-Polynesian formation. This may also explain the differences in Y-DNA, I guess.

Still I miss Striggs dealing with the problem of Melanesian Neolithic and possible flows from Indochina. What he does is simply restricting all to the Austronesian paradigm and ignoring all the rest. It is important, I believe to understand how Neolithic arrived/evolved locally (without pottery probably) to New Guinea and also whether there were other flows via Malaysia from Indochina.

terryt said...

"Still I miss Striggs dealing with the problem of Melanesian Neolithic and possible flows from Indochina. What he does is simply restricting all to the Austronesian paradigm and ignoring all the rest".

What he seems to be missing is any consideration of the Hoabinhian and the (somehow) connected Toalean. For Toalean see:

http://www.highbeam.com/doc/1G1-84343057.html

These two cultures are earlier than the Austronesian (they date from 10k or more, and last until the accepted Austronesian arrival), and the Toalean even reached the Andamans.

No-one has any idea what language is associated with the cultures but it may explain, 'Bellwood noted that there is a claimed Austroasiatic linguistic substratum in some Borneo languages and Austroasiatic influence is also suggested for Sumatra, but may be of much later date'. It certainly explains the two-stage model of expansion in the region, and possibly 'that the arrival of Neolithic to Taiwan is virtually simultaneous with the arrival to Philippines and Indonesia'. Sea lanes were already open and so the Austronesians were able to spread very rapidly.

Maju said...

There is total oversight of the previous archaeology. Unlike in Europe, where research on Neolithic usually came after Paleolithic was rather well known, here scholars seem to have a very poor idea of Paleolithic, Epipaleolithic and Early Neolithic and all is focused on Late Neolithic of Austronesian affinity. So it's like we can't really judge.

In Europe we can see that Danubian Neolithic is essentially (not totally) a cultural replacement and we can see that, instead, Mediterranean Neolithic is largely arrival of pottery, agriculture and some animals onto pre-existent Epipaleolithic cultures. And we know to a large extent the exceptions too. That we know because we know what was before. In SE Asia and particularly insular SE Asia we cannot do that because we are dealing with a Late Neolithic on nothing known... though we do know that there was something before, right.

The article on Toalean is nice but is focused on a particular island: Sulawesi. We need to know what happened, from the archaeological viewpoint, in Java, Sumatra, Borneo, Luzon, etc. and very importantly New Guinea. How are their Epipaleolithic and Neolithic processes related or not.

Because it'd be not the first time that a culture spreads without much genetic impact. And if it does in a matter of just few centuries, maybe a few decades, and almost simultaneously into all the Malay Archipelago, then we really have a problem justifying a radical demic change.

It'd be more easy to compare with the European colonization instead, which arrived and left, causing an economic, social and cultural impact but not any sort of demic replacement at all. Not in Indonesia but in other areas, it even changed the dominant language.

Maybe you could say it's a much radical comparison and I'd have to concede, but it's a counterpoint to the model of sudden demographic expansion, which is also inspired in other realities of modern European colonialism (Australia, USA, Argentina, Brazil).

I don't think that the Neolithic peoples of the South China Sea area had the capability to do as Industrial Era states... in either sense. But if anything I'd be inclined to think in terms of elite domination than massive demic replacement, specially considering the short time involved. More so when a lot of Indonesian peoples show strong genetic affinities to Austroasiatics or Papuans (HUGO Consortium paper) and that Karafet seems to think that the bulk of the Y-DNA in Sundaland is pre-Neolithic.

terryt said...

"I'd be inclined to think in terms of elite domination than massive demic replacement"

Quite probably true.

"In SE Asia and particularly insular SE Asia we cannot do that because we are dealing with a Late Neolithic on nothing known... though we do know that there was something before, right".

Maybe. I've several times pointed you towards articles that suggest the densely forested regions of SE Asia were sparsely, if at all, inhabited. It may not have been until people arrived who were prepared to set fire to the jungle, and indulge in basic agriculture, that we can talk of a substantial population through much of the region.

"The article on Toalean is nice but is focused on a particular island: Sulawesi".

There is information elsewhere, including the claim that the Toalean was the first into the Andamans. I'll see what I can find when I have more time.

"Karafet seems to think that the bulk of the Y-DNA in Sundaland is pre-Neolithic".

But she could be wrong.

Maju said...

"including the claim that the Toalean was the first into the Andamans"...

The first Neolithic. But most Andamanese used to be/still are Paleolithic. So not really relevant.

What we need is archaeology of the Epipaleolithic and Early Neolithic in the big islands. I don't think these were unpopulated ever.

"But she could be wrong".

Of course, but my impression is that MC chronologies tend to err by being way too short, not too long. So I find extremely difficult that her chronology of 30-15 Ka can be cut to a mere 4 Ka, specially when she has detected another (third) wave of that age.

Mason said...

About the bulk of Indonesian O lineages which came from Indochina, such as O2a-M95*, this is what Karafet said in their paper:

This paragroup is widespread in south and southeast Asia (Karafet et al. 2005) and it is unclear when it initially entered western Indonesia.

The ancient time to the most recent common ancestor
17 (TMRCA) of this lineage and Asian distribution supports the hypothesis of a pre-Austronesian incursion of O-M95* into western Indonesia from southeast Asia (Kumar et al. 2007).


It has also been suggested that O-M95* chromosomes appeared in Indonesia after the initial colonization of the Pacific by Austronesian farmers and that the high frequency of O-M95* in Bali and Java may reflect an even more recent influx of males from the Indian subcontinent (e.g., possibly concomitant with the spread of Hinduism and the establishment of Indian kingdoms in the first millennium) (Karafet et al. 2005).

Even if the TMRCA of O2a-M95* is very ancient, it doesn't necessarily mean the arrival of O2a-M95* in Indonesia is Paleolithic. Becasue the common ancestor probably lived in Paleolithic Indochina or southern China.

For example, if TMRCA of a group of American R1b-M269men dates back to Paleolithic, it doesn't mean their common ancestor arrived in Americas during the Paleolithic.

Mason said...

Here's a paper written by Higham:

A Review of Archaeology in Mainland Southeast Asia

http://viewer.zoho.com/docs/vadbbck

Maju said...

"Even if the TMRCA of O2a-M95* is very ancient, it doesn't necessarily mean the arrival of O2a-M95* in Indonesia is Paleolithic. Becasue the common ancestor probably lived in Paleolithic Indochina or southern China".

Makes sense because the variance is quite lower in West Indonesia. However it's almost the same in East Indonesia as in the mainland (and in both cases it's a rare lineage). It would allow for this lineage to have arrived in a later time frame (applying a proportion, I get c. 10 Ka).

"the high frequency of O-M95* in Bali and Java may reflect an even more recent influx of males from the Indian subcontinent"...

This does not make any sense, right? Because this lineage is rare in South Asia, excepting the probably not involved Austroasiatic tribals.

It rather would seem to support a localized founder effect in Java (and by extension Bali), maybe with Hoabinhian "Neolithic".

"Here's a paper written by Higham: A Review of Archaeology in Mainland Southeast Asia"

Thanks, I downloaded it and will read later.

terryt said...

"The first Neolithic. But most Andamanese used to be/still are Paleolithic".

Not so. The article claimed 'the first people'. That's it. May just refer to some islands (perhaps the Onge) if German's claim their language is related to Austro-Asiatic is correct.

"So I find extremely difficult that her chronology of 30-15 Ka can be cut to a mere 4 Ka, specially when she has detected another (third) wave of that age".

The later end of that (15K) easily fits the Hoabinhian/Toalean. And even 30K is long after Australia was settled, so even at that date they're not the first people to SE Asia. The third wave at 4K is probably Austronesian, as I think you'll agree.

"It has also been suggested that O-M95* chromosomes appeared in Indonesia after the initial colonization of the Pacific by Austronesian farmers"

That makes sense. But I agree that:

"the high frequency of O-M95* in Bali and Java may reflect an even more recent influx of males from the Indian subcontinent"

Doesn't make sense. O-M95 in India is much more likely to be a result of movement in the other direction, from SE Asia.

"However it's almost the same in East Indonesia as in the mainland (and in both cases it's a rare lineage)".

That could be explained through a large number of immigrants into East Indonesia, just as the rapid expansion data imply.

terryt said...

"Here's a paper written by Higham: A Review of Archaeology in Mainland Southeast Asia"

Thanks for that Natsuya. Nothing too surprising for me, however some will no doubt disagree with the inclusion of Austronesian in the Austric language family. The author certainly sees a substantial expansion of these languages from the north into SE Asia. And a relatively unpopulated SE Asia before that time. Also some good maps, notably the sea level and language distribution ones. Interesting ideas about the Austric languages spread into India.

I think the data suggest that the Y-haps C and K(xO) are early in SE Asia. But the Os are more recent, perhaps some arrived before the Neolithic as we understand that term. But from further north.

Maju said...

I'm not buying into fringe theories on "recent" origin of Andamanese, unlike the Nicobarese who speak Austroasiatic and are exclusively Y-DNA O, the Andamanese speak isolate languages and are exclusively Y-DNA D (ref), what denotes their extreme antiquity.

What someone who claims, against all evidence, that humankind originated in America says is absolutely irrelevant. I have discussed with him much more than you and he really twists and distorts everything to his convenience: read the references (if he provides any) not his claims, because both in most cases are not related at all (or only very vaguely in the best cases).

Maju said...

"I think the data suggest that the Y-haps C and K(xO) are early in SE Asia. But the Os are more recent"...

Since the discovery of MNOPS (including, it seems, all SE Asian and Oceanian native K-other), it becomes evident that NO (and O) is what MNOPS evolved into in SE Asia north and west of Wallace Line. The data does not suggest in any way that O arrived from anywhere else at all.

O is by far most diverse in SE Asia overall and this paper establishes that some of its sublineages are most diverse in Sundaland in fact. It is particularly the case of O1a-M119 and its subclade O1a1-P203, suggesting, IMHO, that this lineage coalesced in Sundaland an emigrated to the mainland, in a reflux within the diffuse 30-15 Ka period that Karafet assigns to most O flows in the area.

Taiwanese O1a2 probably reflects a migration FROM ultimately Sundaland at some point in the Paleolithic. That's what I think after pondering the overall diversity of O1a.

Not yet researched but it's very possible that what can be said of O1a applies to all O1 because it includes nearly all of this haplogroup. And that means one of the three branches of O.

The other two O2 and O3 also coalesced in SE Asia but in the mainland, somewhere between Sanghai and Bangkok surely. That makes O a very ancient SE Asian lineage, albeit starring an expansion of its own over the other ancient lineages of the area: C and D specially. Alternatively these two would have been minoritary since the beginning succeeding by localized founder effects in the periphery only, much like other MNOPS subclades (N, P, M and S specially).

In synthesis: O is not "more recent" in SE Asia, certainly not more recent than the rest of K, which is all MNOPS in that area, just more successful.

terryt said...

"I'm not buying into fringe theories on 'recent' origin of Andamanese"

As far as I know there is absolutely no evidence that the Andamanese are particularly ancient there. In fact I've noticed a marked decrease in the number of such claims these days.

"the Andamanese speak isolate languages and are exclusively Y-DNA D (ref), what denotes their extreme antiquity".

Neither of these facts actually denote 'extreme antiquity', certainly not a remnant of the great southern migration. The Y-hap may have arrived from the north at any time, and New Guinea languages are extremely varied yet some are said to have arrived in New Guinea just something like 10K years ago.

I completely accept that the Nicobarese are more recent to their islands than are the Andamanese. But it seems strange that the Andamanese were unable to reach the Nicobar islands, which are closer to the Andamans than the Andamans are to the mainland.

terryt said...

"In synthesis: O is not 'more recent' in SE Asia, certainly not more recent than the rest of K, which is all MNOPS in that area, just more successful".


So the apparent large influx from the north suggested in the Higham paper brought in no outside haplogroups?

Maju said...

"Neither of these facts actually denote 'extreme antiquity', certainly not a remnant of the great southern migration".

I don't know where your mind is but mine is clearly focused on the "second migration" starred by Y-DNA O (and N surely too) from somewhere in SE Asia into SE Asia and beyond (specially NE Asia). The process depicted by Karafet as the 30-15 Ka ago flows (fig. 5-B).

There's no remnant of this process among Andamanese nor of the later recent flows (fig. 5-C and 5-D). So Andamanese, like the Ainu and other peripheral East Eurasians (Papuans, Australian Aborigines) and a good deal of the ancestors of other peripheral peoples of that area (East Siberians, Japanese, Tibetans) must belong to the pre-O migrations, those with Y-DNA D, C and MNOPS(xO) [or (xNO) or (xNO,P)].

Andamanese are a very clear case of this pre-O East Eurasia and you are just playing revisionist without any foundations.

"The Y-hap may have arrived from the north at any time"...

Y-DNA D is clearly of SE Asian origin (ref) and that's particularly clear for the D* branch of the Andamanese. It has even been claimed that D is the oldest Y-DNA lineage east of Bengal.

"... and New Guinea languages are extremely varied yet some are said to have arrived in New Guinea just something like 10K years ago".

I don't buy that either. There's always some smartass linguist who loves to push things. Each time I discuss with a linguist he/she has totally different opinions than the previous. Linguistics, when pushed too far, just gets totally wild and subject to the ideological preferences of the linguist.

So I only take seriously the consensus and, if anything, those non-consensual theories which have some acceptance, I have analyzed a bit myself and make sense in the broader context. No Nostratic for me and certainly no Sino-Caucasian, thanks. No ubiquitous "Paleolithic Indoeuropean" everywhere from Portugal to Bengal either.

"But it seems strange that the Andamanese were unable to reach the Nicobar islands, which are closer to the Andamans than the Andamans are to the mainland".

Maybe they were expelled. Or maybe the Andaman Islands were closer to the mainland in the Ice Age. Actually that seems to be the case in Karafet's map A and B. Notice that the Irrawaddy is a powerful river with a broad delta, that extends on a large now submerged platform.

Maju said...

"So the apparent large influx from the north suggested in the Higham paper brought in no outside haplogroups?"

I have just read the paper up to the Bronze Age section in order to reply to this.

I must say that your supposition remains highly speculative and that Thailand Neolithic dates from at least 5500 BCE, which is pretty much contemporary with Chinese Neolithic (500 to 1500 years more recent if the date is to be taken literally but not particularly related in cultural aspects to Yangtze or Pearl River Neolithic cultures).

Other areas are less researched and may still yield surprises (they should in fact). No data is provided for the key issue here: Island SE Asia and Melanesia.

Furthermore, a lot of the speculation on flows from the North relies on a Renfrewist hubris where the Neolithic is attempted to be interpreted on linguistic reconstructions that are clearly excessive. The Austric hypothesis is still nothing but that and there is absolutely no reason to imagine that Austroasiatic languages were ever spoken in modern China. In fact, modern and known historical distribution strongly suggest an origin in Indochina, being the most likely candidate for the main language group of the distinct Indochinese Neolithic.

I'd adventure the possibility that Hmong-Mien might have been the main language group of Yangtze Neolithic and Kradai that of the Pearl River group. However it's also possible that Kradai was the language of Yangtze Neolithic. Chinese is likely to have coalesced in the Northern Neolithic (Yellow River) group instead. Tibeto-Burman is clearly peripheral (Sichuan-Yunnan highlands).

But with all this we are still far from discerning what happened in SE Asia and particularly in Island SEA and Papua before and during the Neolithic.

terryt said...

"What someone who claims, against all evidence, that humankind originated in America says is absolutely irrelevant".

He's probably refering to this paper (by a linguist, I'm afraid):

http://email.eva.mpg.de/~blevins/pdf/webpub2007a.pdf

"Since the discovery of MNOPS (including, it seems, all SE Asian and Oceanian native K-other), it becomes evident that NO (and O) is what MNOPS evolved into in SE Asia north and west of Wallace Line"

And N arrived in Northern Asia by spaceship?

"Thailand Neolithic dates from at least 5500 BCE"

That's along way from 30,000 years, and yet no haplogroups were introduced with it?

Maju said...

"And N arrived in Northern Asia by spaceship?"

By walking. N is highly diverse in Southern China and probably originated there, however the haplogroup had its greatest success in the Far North, probably because there it was not out-competed by O (whatever the reasons for the success of O).

"That's along way from 30,000 years, and yet no haplogroups were introduced with it?"

I can't at this stage discuss which haplogroups were introduced, at what levels and when. It's very possible anyhow that cultural difusion of Neloithic really happened with little genetic input, as probably happened in Europe, Africa and so many other places.

Furthermore, I'd say that no/small input of Neolithic genetics is the null hypothesis, specially when archaeology does not suggest a clear cultural migration or finds that this cultural flow happened on local cultural continuity. I expect that any claim of demic replacement beyond minor inputs in such cases is sufficiently proven.

Recentists have that belief that every single cultural flow, specially Neolithic, implies mass migration and demic replacement. This may be the case in some particular areas and time frames but just because they believe it, it does not have to be true.

I demand proof to support such claims and such proof in most cases fails to be produced. Hence they hide in some peculiar interpretations of the molecular clock that magically makes nearly all lineages to "become" hyper-recent (in their minds at least) but that's simply not acceptable as proof: it's twisting science to fit one's preconceptions. Similarly other such manipulations of the data, like ignoring phylogenetic distinctions, are used in order to pretend that the replacement hypothesis is correct (case of the Balaresque paper).

It is of course possible that the circumstances of East Asia are different from that of Europe, and also that the circumstances of different subregions are different within each of these areas. All these details must be carefully assessed in each case to produce, when possible, the most likely results.

But, in principle, I remain highly reluctant to accept mass replacement hypothesis at face value: they are mostly ideological and reliant on too many one-sided assumptions rather than on solid data.

terryt said...

"But with all this we are still far from discerning what happened in SE Asia and particularly in Island SEA and Papua before and during the Neolithic".

We actually have quite a bit of research on the subject. Here are some links to Bellwood's work which, hopefully, you can access:

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

And

http://www.questia.com/googleScholar.qst;jsessionid=MkmJtLJMy5Jb8Jt4d2mc213WQvbTyzV1TQNGVn2SxdPz37xJp189!-289140146!1566849592?docId=5001400547

You might be able to do something with this:

http://books.google.co.nz/books?id=6kDm5d3cMIYC&printsec=frontcover&dq=Southeast+Asian+prehistory+bellwood&source=bl&ots=QjC5eiPuHO&sig=tFUFRad9oBbBrqjYvKxtXH9onSg&hl=en&ei=GSYkTLG-N4PlnQeK45GTDQ&sa=X&oi=book_result&ct=result&resnum=1&ved=0CBoQ6AEwAA#v=onepage&q=Southeast%20Asian%20prehistory%20bellwood&f=false

"the haplogroup had its greatest success in the Far North, probably because there it was not out-competed by O"

So it walked there, but O was unable to do so.

Maju said...

Thanks. Not sure if it's very useful because when I try to look at Bellwood's review of anything pre-Austronesian it's a non-available page. :(

terryt said...

Sorry about that. I tried a few and they seemed to work, but that may have been just the Austronesian ones. I'll see what I can do. But Bellwood is very respected in SE Asian prehistory so you may be able to Google something useful. I've read several of his books over many years. That's basically where I've formed my ideas from (with the help of several NZ geneticists and anthropologists).

Maju said...

Relax, we are not going to untangle every other detail of Asian Prehistory in this blog. Of course if you happen to stumble on key info, please post it... but we will survive not knowing for sure some obscure details.

terryt said...

"Relax, we are not going to untangle every other detail of Asian Prehistory in this blog".

But the exchange of information sure as hell helps us immensely. For example your comment regarding the cuscus, and my looking at the tasier' distribution, have indicated to me that humans are most likely to have first crossed Wallace's line between Mindanao and Sulawesi.

Maju said...

I don't understand this logic of you about the tarsier, really. The tarsier is a placentarian mammal and a primate, albeit in the most remote branch from ours, whose origins lie in Afroeurasia; while the cuscus is a marsupial whose origins lie in Australasia.

terryt said...

"The tarsier is a placentarian mammal and a primate, albeit in the most remote branch from ours, whose origins lie in Afroeurasia; while the cuscus is a marsupial whose origins lie in Australasia".

Precisely the point. Each animal has invaded Wallacea from the respective neighbouring region. This indicates that Mindanao is not impossibly remote from Sunda and Halmahera is not impossibly remote from Sahul. The easiest route between Sunda and Sahul for any mammal would probably therefore involve those two regions.

German Dziebel said...

Luis: "What someone who claims, against all evidence, that humankind originated in America says is absolutely irrelevant..."

Luis again: "Take a look at this paper by Brown. It's very interesting: there's no "Mongoloid" but also no "Australoid" skull in East Asia before Neolithic."

Peter Brown, p. 120: "At present the earliest people with a generalised East Asian cranial morphology are probably found in the
Americas. Is it a possibility that migration across the Bearing Straits
went in two directions and the first morphological Mongoloids evolved
in the Americas?"

By now, it should be clear to everybody in the blogosphere that everything that Luis Aldamiz says about the out-of-America theory and about me is a libel. But that's not why I'm here. My former professor of craniology, Alexander Kozintsev, published an interesting paper in which he argued that some Bronze-Age South Siberian skulls still preserve aspects of typical Amerindian (not Paleoindian) cranial morphology. See http://www.ncbi.nlm.nih.gov/pubmed/9988381.

Although in my book "The Genius of Kinship" I didn't address the possible origins of Mongoloids in the New World, the presence of all four Amerindian mtDNA markers in South Siberian populations at very low frequencies is a very good sign of admixture. It's also noteworthy, as Berezkin detected in the distribution of folkloric motifs, that South Siberia shares with North America some very unique narrative features (Berezkin. Mythological Links Between South Siberia and North America // Archaeology, Ethnography and Anthropology of Eurasia 2003). In general, per Vladimir Napolskikh, North America tends to contain more archaic versions of the same folkloric motifs than Siberia. Then, of course, we have the Na-Dene-Ket linguistic connection. Not to mention the fact that Jomon skeletons show mtDNA D1, which is typical for America, but is missing from later Asian populations.

So, the evidence for at least some level of late Pleistocene-early Holocene admixture from the new World to the Old World is pretty consistent. The question is of course what was the magnitude of this admixture. Peter Brown's work suggests that the whole Siberian-Amerindian nexus detected by geneticists in the form of mtDNA A, C, D and Y-DNA C3 and Q may be the result of a relatively recent admixture from the New World to the Old World. This would make mtDNA haplogroup A and Y-DNA C3 as localized in America (prior to the admixture event) as, say, mtDNA haplogroup S and Y-DNA haplogroup C4 are restricted to Australia. Notably, A, C and D aren't found in Europe or Australasia and in (South)east Asia they peter away as one goes south.

Luis: "I'm not buying into fringe theories on "recent" origin of Andamanese..."

Terry" "As far as I know there is absolutely no evidence that the Andamanese are particularly ancient there...

Luis: "the Andamanese speak isolate languages and are exclusively Y-DNA D (ref), what denotes their extreme antiquity".

Terry: "Neither of these facts actually denote 'extreme antiquity', certainly not a remnant of the great southern migration..."

Terry is right. For a collection of cross-disciplinary data contradicting the myth of the relic status of short-statured populations around the world, see my discussion with Terry and Andrew at http://averyremoteperiodindeed.blogspot.com/2010/06/humans-in-philippines-67000-years-ago.html_

Terry: "Not so. The article claimed 'the first people'. That's it. May just refer to some islands (perhaps the Onge) if German's claim their language is related to Austro-Asiatic is correct."

Austronesian, not Austroasiatic, per Blevins. But you do hint at a possibility that the presence of Austroasiatic languages in India (the Munda group) and the presence of para-Austronesian language(s) in the Andaman islands go back to the same wave of migration.

Maju said...

But the tarsier never made it to Wallacea. There's nothing in all that suggesting a link between the Asian mainland (Sundaland) and Wallacea via Philippines. I don't say this is impossible but the direct route across Wallace Line is at least as possible and IMO much more likely.

I know it's not always the case but in school we all learned that the shortest way between two points is the straight line, right? So you'll need some solid evidence to persuade me of the opposite and nothing in all what has been mentioned here suggests so.

Maju said...

German: by now you should know I'm not interested at all in wasting my time discussing (in pointless circles) with you. I have done that in the past and it's a total waste because you are so extremely partisan of your wacko hypothesis that you will twist everything and arbitrarily select decontextualized snippets of info without any respect for the facts.

If you think this claim is a libel, feel free to sue me.

Now get lost.

Thanks.

German Dziebel said...

"you will twist everything and arbitrarily select decontextualized snippets of info without any respect for the facts."

Yes, it's a libel. Not being trained in the scientific method or well-read in the relevant topics, you don't even know what a context or a fact is. Do I want to sue you? No.

"because you are so extremely partisan of your wacko hypothesis."

Only to counteract the out of Africa hysteria and the peopling of the America obsession that permeate the academic and blogging circles.

Maju said...

German: your attitude of disqualifying others on stupid things as having a university degree, says a lot about the kind of mentality you have.

It's just sick.

I don't judge people for their titles, medals or whatever but for what they say and do. Only that way you can know if you are before a dumb and ignorant academic or an intelligent and knowledgeable common person. Your presumptuous academic elitism is totally wrong and your stupid accusation that I know not the scientific method is even wronger.

You disqualify yourself every time you type something almost without exception.

And as I don't want you coming over here to say nonsense or insults, I'm going to censor all your future posts. You are warned so don't bother posting anything more.

Bye.

terryt said...

"But the tarsier never made it to Wallacea".

Depends entirely on what you call Wallacea. Seems Ernst Mayr was quite happy to include the Philippines in Wallacea (see Occamseraser's latest comment here:

http://averyremoteperiodindeed.blogspot.com/2010/06/humans-in-philippines-67000-years-ago.html

"but the direct route across Wallace Line is at least as possible and IMO much more likely".

The gap between Mindanoa and Sulawesi is a 'direst route across Wallace Line'. Wallace's line runs between them, whether you care to include the Philippines in Wallacea or not. In fact it's far more direct, and shorter, than any other route you might care to propose.

"in school we all learned that the shortest way between two points is the straight line, right?"

And the shortest distance between islands across Wallace's line is between Mindanao and Sulawesi, and the islands between them.

Maju said...

"Depends entirely on what you call Wallacea".

In doubt use Wikipedia. ;)

I will use the mainstream meaning. And it'd be nice if you assume that same meaning too for the sake of understanding and not wasting your own and others' time in meaningless terminological arguments.

"The gap between Mindanoa and Sulawesi is a 'direst route across Wallace Line'".

Fat fingers, eh? No prob, we all have bad days. :p

The gap between Mindanao and Sulawesi is not a route between Sundaland and Wallacea in any case but between the Filipino and the Wallacean provinces, both similarly separated from mainland Asia/Sundaland.

Discussing about terminology won't change the facts of geography.

"And the shortest distance between islands across Wallace's line is between Mindanao and Sulawesi, and the islands between them".

It's not true because Bali and Lombok are much closer for sure and the Makassar strait was surely much narrower in the Ice Age. It's not true because the Talaud islands between Mindanao and Sulawesi are not particularly close to either shore nor make the Mindanao-Sulawesi distance as such smaller at all.

But anyhow, even if it would be true, it would not matter much because any animal trying to cross between Sundaland and Wallacea would still have to cross at least three straits and not just one.

If you could at least provide a single example of land animal (other than Homo sp.) that is found naturally in both Philippines and Wallacea, you could show you have a point beyond nomenclature but so far you are just arguing on names and nothing else.

You actually seem to be arguing out of mischievous intent (arguing for the sake of arguing) rather than in a honest search for the factual reality. It's not a "game" I enjoy, really.

terryt said...

"You actually seem to be arguing out of mischievous intent (arguing for the sake of arguing) rather than in a honest search for the factual reality. It's not a 'game' I enjoy, really".

Not so. But you're being completely stupid. You're so convinced of your own beliefs you can't look at reality. Look at this comment:

"Discussing about terminology won't change the facts of geography".

I totally agree. And yet it's you who claims:

"The gap between Mindanao and Sulawesi is not a route between Sundaland and Wallacea in any case but between the Filipino and the Wallacean provinces, both similarly separated from mainland Asia/Sundaland".

That's exactly what I have been trying to draw your attention to. So who here is trying to 'change the facts of geography'?

"for the sake of understanding and not wasting your own and others' time in meaningless terminological arguments".

So why don't you follow your own advice?

"If you could at least provide a single example of land animal (other than Homo sp.) that is found naturally in both Philippines and Wallacea"

Surely you realise that Wallace's line is just a name (see above). Named because no land animals, apart from humans, have crossed it. On the other hand very few have made it from Sunda to the Philippines either, so the Philippines, too, are a separate biogeographic region from mainland Asia. It's irrelevant whether we call them Wallacean or not.

"It's not true because the Talaud islands between Mindanao and Sulawesi are not particularly close to either shore nor make the Mindanao-Sulawesi distance as such smaller at all".

The Talaud islands have nothing at all to do with it. They're way off to the east.

"It's not true because Bali and Lombok are much closer for sure and the Makassar strait was surely much narrower in the Ice Age".

And an ice age would surely expose much of the strip of submerged land between the northern tip of Sulawesi and the Sarangani Islands, off the southern tip of Mindanao. But obviously not sufficiently for animals without boats to be able to cross.

"But anyhow, even if it would be true, it would not matter much because any animal trying to cross between Sundaland and Wallacea would still have to cross at least three straits and not just one".

If it goes via the Philippines it only has to cross one: the gap between Mindanao and Sulawesi.

(continued)

terryt said...

"It's not true because Bali and Lombok are much closer for sure and the Makassar strait was surely much narrower in the Ice Age".

So let's look at this theory of yours. For several reasons a route to Australia through Java and Flores is very unlikely.

Granted, ancient humans managed to cross Wallace's line and colonise Flores. However these people remained isolated long enough to become what many see as a separate species. What's more this separate species appears to have survived until as recently as 12,000 years ago. This suggests modern humans didn't reach Flores until at least this time, even if the Hobbits' disappearance is the result of a volcano nearby.

So what happens when we turn back west across Wallace's Line to Java? We know that ancient humans, in the form of Homo erectus, had reached the western side of Wallace's line. Possibly as long ago as a million years. What's more some scientists claim that this H. erectus survived in Java until as recently as 30,000 years ago. Certainly long after humans had reached Australia.

Yet this post:

http://averyremoteperiodindeed.blogspot.com/2010/06/humans-in-philippines-67000-years-ago.html

claims modern humans in Luzon as long ago as 67,000 years ago (if not longer). Surely, then, this region fits better as a source for a settlement of Australia 50-60,000 years ago. Especially as it's just a hop, skip and a jump across Wallace's line from Mindanao to Sulawesi. With the technology required for this feat it would have been child's play to reach New Guinea and Australia.

Maju said...

Terry said: "But you're being completely stupid".

And yet:

Me: "The gap between Mindanao and Sulawesi is not a route between Sundaland and Wallacea in any case but between the Filipino and the Wallacean provinces, both similarly separated from mainland Asia/Sundaland".

Terry: "That's exactly what I have been trying to draw your attention to".

WTF! You said exactly:

"Precisely the point. Each animal has invaded Wallacea from the respective neighbouring region. This indicates that Mindanao is not impossibly remote from Sunda and Halmahera is not impossibly remote from Sahul. The easiest route between Sunda and Sahul for any mammal would probably therefore involve those two regions".

So I gather that you are claiming all the time that, for some mysterious reason, the shortest route between Bali and Lombok or Borneo and Sulawesi and is via Mindanao.

So can you make up your mind? >:(

Terry says: "Surely you realise that Wallace's line is just a name (see above). Named because no land animals, apart from humans, have crossed it".

That's not exact: elephants crossed it and other animals probably too. It's just a very noticeable ecological border: the westernmost border of animals of Australian origin:

"West of the line are found organisms related to Asiatic species; to the east, a mixture of species of Asian and Australian origin are present".

Terry says: "On the other hand very few have made it from Sunda to the Philippines either, so the Philippines, too, are a separate biogeographic region from mainland Asia. It's irrelevant whether we call them Wallacean or not".

It is relevant because no Australian animals are found in Philippines, all are of Asian origin. Hence Philippines is not part of the transitional province of Wallacea but a distinct province on its own right, a subprovince of Asia with no transitional character whatsoever.

Terry says: "The Talaud islands have nothing at all to do with it. They're way off to the east".

Maybe. I was "mislead" by this map. In any case, the only thing in the wide marine divide between Mindanao and Sulawesi are these Talaud islands and the Sangighe islands, both of which are south of Wallace Line and belong administratively to North Sulawesi.

So if you are arguing for some sort of direct route between Mindanao and Wallacea, be it Sulawesi or Malukku islands, these islands (Talaud and Sangighe) are the only thing in between other than water.

Terry says: "And an ice age would surely expose much of the strip of submerged land between the northern tip of Sulawesi and the Sarangani Islands, off the southern tip of Mindanao".

LOL, the Sarangani Islands are just slightly off the coast of Mindanao. Even if they were then a peninsula, it would not make any difference.

Terry insists: "If it goes via the Philippines it only has to cross one: the gap between Mindanao and Sulawesi".

First it has to go to Philippines (crossing straits and another ecological line), which is one problem or two, and then cross what was maybe the largest Wallace Line gap of the three relevant straits (which is yet another problem of big size in itself). That's not the point of lesser resistence but the most difficult path of all.

Circular logic that of yours based on nothing but stubbornness.

Maju said...

Terry says: "What's more this separate species appears to have survived until as recently as 12,000 years ago. This suggests modern humans didn't reach Flores until at least this time, even if the Hobbits' disappearance is the result of a volcano nearby".

That's a happy claim. I just don't believe it. H. floresiensis and H. sapiens may well have shared the island by using different environments, such as jungle and coast, for instance.

Of course only archaeology will solve the matter but SE Asian archaeology has so far too many gaps.

In general the Java-Lesser Sunda route is considered the default one in all reconstructions I have seen of the colonization of Sahul, with the one via Sulawesi-Malukku considered only as second possiblity. I have never ever read anyone arguing for a Filipino route before you did.

"Yet this post (...) claims modern humans in Luzon as long ago as 67,000 years ago (if not longer)".

I discussed it here - for the record.

The species adscription of the metatarsal bone is not clear yet. It fits with small H. sapiens, with H. floresiensis and even with H. habilis.

I also want to believe it is a modern human fossil but I do not know yet for sure. It's good to keep a prudent attitude and not let oneself being carried by mere wishful thinking. In science at least, reason must always rule over emotions.

"Surely, then, this region fits better as a source for a settlement of Australia 50-60,000 years ago. Especially as it's just a hop, skip and a jump across Wallace's line from Mindanao to Sulawesi".

Just because you wish so?

If it is a H. sapiens bone, what we don't know yet, the people must have arrived via Borneo (Sundaland), right?

So, if they were in Sundaland and Luzon (not even in Mindanao!) they still could have crossed (in theory) by any of the three Wallace Line straits and nothing says they did by the easternmost one, which is also the widest one by far.

You fall too much and too often to wishful thinking. It's a human flaw but you are particularly stubborn about it, what is not of any help, really.

terryt said...

Unfortunately for your credibility your hysterical outburst here is full of inconsistencies and contradictions, and most of the remainder is wrong. For example:

"You fall too much and too often to wishful thinking".

Yet you wrote:

"That's a happy claim. I just don't believe it. H. floresiensis and H. sapiens may well have shared the island by using different environments, such as jungle and coast, for instance".

'Just because you wish so'? And you believe that statement is not 'a happy claim'?

"Hence Philippines is not part of the transitional province of Wallacea but a distinct province on its own right, a subprovince of Asia with no transitional character whatsoever".

But you then claim, regarding the possibility of humans having reached the Philippines:

"First it has to go to Philippines (crossing straits and another ecological line), which is one problem or two"

Hang on. If the Philippines are 'a subprovince of Asia' surely that is no problem. Certainly the primitive tarsier made it. As did several species of deer. So, make up your mind. Either the Philippines are separate from Asia, in which case we may as well call them part of Wallacea, or they're part of Asia, in which case the fact that humans reached them quite early is no surprise at all.

"That's not exact: elephants crossed it and other animals probably too".

And that is 'proof' that humans reached Sahul via Flores? Tell me, exactly how far did these elephants get once they'd actually managed to cross Wallace's line? The same distance as the Hobbits, surely. This indicates that Flores is not the beginning of an easy route through Wallacea.

"So I gather that you are claiming all the time that, for some mysterious reason, the shortest route between Bali and Lombok or Borneo and Sulawesi and is via Mindanao".

Not necessarily the shortest, but it certainly looks as though it's the easiest. And it is certainly the region of Wallacea where Sunda species (the tarsier) come closest to Sahul species (the cuscus). Perhaps technically they come as close across the gap between Borneo and Sulawesi, but there are no possible island connections directly between these two islands even at times of lowest sea level. On the other hand the Lesser Sunda Islands are virtually devoid of either Sunda or Sahul mammalian species.

"Maybe. I was 'mislead' by this map. In any case, the only thing in the wide marine divide between Mindanao and Sulawesi are these Talaud islands and the Sangighe islands"

Not mislead. You just didn't bother to look at it properly. Try again and take note of the fact that the Sangighe islands lie on the same submarine ridge as Sarangani. Much of this ridge, except for the deeper trench near Sarangani, would have been exposed as islands during severe ice ages.

"First it has to go to Philippines (crossing straits and another ecological line), which is one problem or two, and then cross what was maybe the largest Wallace Line gap of the three relevant straits (which is yet another problem of big size in itself). That's not the point of lesser resistence but the most difficult path of all".

So you can see that the above statement is absolutely incorrect.

terryt said...
This comment has been removed by the author.
terryt said...
This comment has been removed by the author.
Maju said...

Terry said: 'Just because you wish so'? And you believe that statement is not 'a happy claim'?

No happy claim here: just considering the various possibilities objectively on their own merit. I'm not making any claim: just stating an obvious possibility that you are dead set to reject... based on nothing.

But you then claim, regarding the possibility of humans having reached the Philippines:

"First it has to go to Philippines (crossing straits and another ecological line), which is one problem or two"

Hang on. If the Philippines are 'a subprovince of Asia' surely that is no problem"
.

Ok. You want to mix apples and oranges again, right?

I insist: Philippines is not a TRANSITIONAL region between two continents, as Wallacea IS. Philippines is only related to Asia, not to Australia!

But arriving to Philippines as you know very well (because you have raised the point more than once) requires crossing a strait, exactly as arriving to Wallacea, be it Lombok or Sulawesi. So it is an additional obstacle to Wallace Line, which still has to be sorted even if anyone managed to reach to Mindanao.

You must think I'm some sort of idiot or something because Williams Line is great if you can get it into a sentence that ends up miraculously including Philippines in an expanded Wallacea but it's irrelevant if it's an extra obstacle for migration from Asia in addition to Wallace Line. WTF!

So for you it's great that there's an obstacle to reach Philippines if that increases the chances of getting the archipelago in your mystified Wallacea but it's non-existent if, as it happens, is an extra obstacle for any sort of migration via Philippines.

You first reach to conclusions and try to find what arguments you can use to defend that ideological goal you have set yourself to. I'm not accepting that lack of honesty anymore: I expect a fair and honest intellectual exchange not some fanatic who puts the cart (pre-determined conclusion) before the horse (facts) once and again.

Go preach somewhere else and come back if you ever manage to put the facts before the conclusions.

"That's not exact: elephants crossed it and other animals probably too".

And that is 'proof' that humans reached Sahul via Flores?


No. Who said it's proof? Who is saying that this or that route MUST be the one? I'm just saying that your proposal looks very much unlikely because it includes more than double of the obstacles than the other possible routes and flows tend to breach by the points of less resistance. IMO the point of less resistance is the Lesser Sunda chain and alternatively the Makassar strait but certainly not Philippines.

Could they still have chosen for some weird reason the hardest of routes, the one you defend like the idealized medieval knight would fight for his lady or his god, beyond any reason and any hope? Yes they could. But it's most unlikely and you have provided no evidence.

Extraordinary claims demand extraordinary evidence and you can't even provide ordinary one.

...

Maju said...

...

Not necessarily the shortest, but it certainly looks as though it's the easiest.

No matter how I look at it it is the hardest of all routes: two ecological divides (major straits) instead of just one have to be crossed!

"And it is certainly the region of Wallacea where Sunda species (the tarsier) come closest to Sahul species (the cuscus)".

NO! That region is Wallacea proper. Tarsiers are not closer to cuscus than orangutans are in any case. What a ridiculous logic you are using... I'm getting fucking tired of this circus, really.

Not mislead. You just didn't bother to look at it properly.

Gratuitous accusation. I'm not even bothering with this part.

Try again and take note of the fact that the Sangighe islands lie on the same submarine ridge as Sarangani.

This is the first time that you make some sense in many many posts. It does not constitute evidence (we'd have to look at a good bathymetry with a scale) and still ignores Williams Line, which is as great an obstacle as Wallace Line or almost.

You still have two major straits to cross instead of just one for the same goal, you still have to take a detour from the straight line, and you have not provided any sort of systematic evidence in any case, like proper measures of the straits' width at different sea levels.

Instead of being so vehement and fanatic of your intuition you should work a bit in order to gather data that could support it (or maybe not).

And if you'd come with that data instead of wordplays, I'd take you more seriously and would not get annoyed so much.

"So you can see that the above statement is absolutely incorrect".

No. I do not. Look at a decent bathymetry, measure with a ruler the exact distance at the various straits at different possible sea levels. Then maybe you find data that supports your hypothesis or not. That map is not any evidence because it's not clear enough, just an approximation.

Do your damn fucking homework instead of wasting my time.

Maju said...

Not sure if you're working on it or just ignoring the matter altogether.

But I've bothered making a couple of searches and found that, effectively, the Philippines route to Wallacea is the less likely of all.

What I found is:

1. This map (source) with sea depths of 200 m. marked in red for all the area.

2. This other map (source) of the same zone (but less extension) that confirms that the bathymetric depths are correct.

200 m. is quite more than the normally claimed drop in sea levels (c. 120 m.) but is a safe maximum. Definitively anything under 200 m. has been under the sea all the time (notwithstanding tectonics).

This leaves:

- Lombok strait: narrow, potentially passable in canoe or raft.
- Makassar strait: narrow, specially towards the south, potentially passable in canoe or raft.
- The strait between Mindanao and Sulawesi: very wide. It's populated by some small islands, mentioned before but never connected to any major land mass. The largest water span is much larger than the others considered. Probably not used in migrations before the Austronesian period because of it's greater difficulty.

Q.E.D (unless you provide some other data I don't know of).

Mason said...

According to Karafet's email, here's Y-SNP data of four specific Han populations collected as one "Han" in the paper.

http://viewer.zoho.com/docs/fezif

Taiwan Han (n=81):
6/81 = C3-M217
1/81 = D1-M15
3/81 = N1-LLY22g
2/81 = N1c1-M178
2/81 = O3-M122*
5/81 = O3a-P197* (O3a-M324*)
12/81 = O3a3-P201*
1/81 = O3a3b-M7
8/81 = O3a4-002611
20/81 = O3a3c-M134
1/81 = O1a-M119*
8/81 = O1a1-P203
6/81 = O2-P31*
5/81 = O2a-M95*
1/81 = O2a1-M111

Guangdong Han (n=40):
2/40 = C3-M217
1/40 = N-M231
6/40 = N1-LLY22g
1/40 = O3a-P197* (O3a-M324*)
1/40 = O3a3b-M7
2/40 = O3a4-002611
9/40 = O3a3c-M134
6/40 = O1a1-P203
4/40 = O2a-M95*
8/40 = O2a1-M111

Sanxi Han (n=42):
2/42 = C3-M217
1/42 = G2a-P15
1/42 = J2-M172
1/42 = N-M231
3/42 = N1-LLY22g
1/42 = N1a-M128
1/42 = O3a-P197* (O3a-M324*)
2/42 = O3a3-P201*
1/42 = O3a3b-M7
7/42 = O3a4-002611
18/42 = O3a3c-M134
2/42 = O2-P31*
1/42 = O2a-M95*
1/42 = Q1-P36*

Guizhou Han (n=2):
1/2 = O3a4-002611
1/2 = O1a1-P203

Maju said...

Nice, Natsuya, thanks. Not sure what conclusions are possible but I'm adding that link to the main post as update (so it gets some slightly higher visibility than buried under so many comments as this thread has caused).

terryt said...

"No matter how I look at it it is the hardest of all routes: two ecological divides (major straits) instead of just one have to be crossed!"

Are you still claiming the Sulu Archepeligo as a 'major strait', a major 'ecological divide'? As I've consistently tried to point out to you: several mammals have made the crossing without the benefit of boats. And we now have evidence of the early presence in the Philippines of some sort of human, modern or otherwise.

"So it is an additional obstacle to Wallace Line, which still has to be sorted even if anyone managed to reach to Mindanao".

And everyone surely agrres that humans needed boats to cross Wallace's Line. Humans had been able to walk that far.

"I expect a fair and honest intellectual exchange not some fanatic who puts the cart (pre-determined conclusion) before the horse (facts) once and again".

I am by no means the first to suggest such a route. However if you wish to credit me with being first when it's proved I'm quite happy to accept.

"Lombok strait: narrow, potentially passable in canoe or raft".

But that's just the first of the many difficult crossings. It's not simply a matter of reaching Flores and then finding connected islands all the way to Australia. Besides which you're also claiming an early modern human presence in the region without a single item of evidence.

"Tarsiers are not closer to cuscus than orangutans are in any case. What a ridiculous logic you are using..."

Your lack of both English and biological understanding is hampering you immensely. I have never claimed the tarsier and cuscus are close biologically. They come closest physically between the Philippines and Sulawesi. So, that is the region where Sunda and Sahul species are closest GEOGRAPHICALLY. Understand that?

"Q.E.D (unless you provide some other data I don't know of)".

Neither of the maps you link to are at all useful. In the first a name covers the region in question, and the second doesn't even include the relevant region. This is a better map:

http://docs.google.com/viewer?a=v&q=cache:lqWiL1dh3EMJ:www.internalwaveatlas.com/Atlas2_PDF/IWAtlas2_Pg263_Celebes.pdf+celebes+sea+bathymetry&hl=en&gl=nz&pid=bl&srcid=ADGEEShmfnCEbLHzKUbUdzSTG7LWB2Li43lmvcsYjGjhTs3eZbNWWysRaQm1OQgEV25JOV_yBtTX3zX76relFQSpggsD9A22TZaRZBu2_i48oT2Retbyn9zoh57Mwkoyz905DHd6M15i&sig=AHIEtbTRGlPDboDs6twfbBJKCIXXYS79pQ

You can see a string of islands along the Sangihe Arc. Presumably they are at present above 120 metres below sea level. I think it's safe to assume they were larger during ice ages, and other islands would have emerged along the arc.

Maju said...

"Are you still claiming the Sulu Archepeligo as a 'major strait', a major 'ecological divide'?"

Well, there's where Huxley Line is, right? At least that's what someone claimed at AVRPI the other day. And that's the border you claimed for Wallacea. I'd agree that Philippines are at least somewhat apart from the Asian Mainland as in the Ice Age (Sundaland), that there was no land connection ever and hence it was always a meaningful barrier.

Whatever the case, it is less relevant, because even if crossing to Philippines implied no Sea crossing at all, getting into Wallacea from Philippines does and it's the widest of all available crossings by far.

"But that's just the first of the many difficult crossings. It's not simply a matter of reaching Flores and then finding connected islands all the way to Australia".

Doesn't matter. Whatever the route into Sahul, it must have gone through Wallacea. However there are two possible routes:

1. The Southern route: Sundaland (Bali) > Lombok > Lesser Sunda Chain > Sahul.

2. The Northern route: Sundaland (Borneo) > Sulawesi > Malukku > Sahul

Either one is better than via Philippines since the beginning.

"Besides which you're also claiming an early modern human presence in the region without a single item of evidence".

I'm not claiming anything except what is obvious: that to get to Sahul people had to go via Wallacea. The presence in Wallacea is inferred from the presence in Sahul, which we know is quite old, though there are some divergences on the exact date (min. 48 Ka ago).

"Your lack of both English and biological understanding is hampering you immensely. I have never claimed the tarsier and cuscus are close biologically".

Erm... I meant closer in a geographical sense: in kilometers away!

"In the first a name covers the region in question"...

Doesn't hide much: it is evident that the Sangighe islands were totally separated from Sulawesi and form each other also in the Ice Age, only growing slightly in size. The sea gap between Sulawesi and Mindanao was not significantly narrowed, while the one between Borneo and Sulawesi was.

If you can find something better... it's you who is making the extraordinary claim and therefore who has to provide the extraordinary evidence.

I'm just doing your work.

"You can see a string of islands along the Sangihe Arc. Presumably they are at present above 120 metres below sea level. I think it's safe to assume they were larger during ice ages, and other islands would have emerged along the arc"

Your link does not seem to be helpful. The bathymetry's highest mark is of 500 meters, more than four times deeper than the sea levels in the coldest periods of the Ice Age.

Maju said...

Found something quite good: this site has many bathymetries, including Malukku Sea, which is the name of the strait we are discussing. They use a color graded scale rather than contour lines which is slightly less precise but good enough.

Direct link to the PDF version of the Malukku Sea Bathymetry. As the 120 m. depth line is at the light to middle pink transition (intense pink is under 250 m already, caution!) it becomes obvious that the islands only increased in size slightly with the Ice Age sea levels and were never linked between them nor the larger islands.

Maju said...

Oops, that was the link to Sulawesi (Celebes) Sea. Maluku Sea is this one. Both show the are we are discussing anyhow.

Maju said...

And oops again: Malukku Sea is not the name of the strait, it seems but of the basin south of it. Does anyone knows what's the name of the sea area separating Philippines and Sulawesi?

terryt said...

"Does anyone knows what's the name of the sea area separating Philippines and Sulawesi?"

I've told you: 'the Sangihe Arc'. It's a region where the sea becomes shallower.

"Your link does not seem to be helpful. The bathymetry's highest mark is of 500 meters, more than four times deeper than the sea levels in the coldest periods of the Ice Age".

So you're claiming that the islands at present above sea level on the Sangihe Arc were submerged at more than 500 metres during the ice age? Surely you can at leasdt try to talk sense.

"Either one is better than via Philippines since the beginning".

And I keep telling you it does not. According to my atlas the greatest distance between the islands on the Sangihe Arc is about 80 kilometres. That's much less than any one of several crossings required on any route from Bali to New Guinea.


"And that's the border you claimed for Wallacea. I'd agree that Philippines are at least somewhat apart from the Asian Mainland as in the Ice Age (Sundaland), that there was no land connection ever and hence it was always a meaningful barrier".

I've been telling you that forever. And also telling you that some people include the Philippines as part of Wallacea. Even Wikipedia admits that. I agree that there is no point in arguing the point any further. Your mind is a closed bone box.

terryt said...

"Both show the are we are discussing anyhow".

Thanks for finding those maps but only by ignoring the parts of that ridge at present above sea level can you support your case.

"it becomes obvious that the islands only increased in size slightly with the Ice Age sea levels and were never linked between them nor the larger islands".

I have never claimed they were 'linked'. As for your comment, 'the islands only increased in size slightly with the Ice Age sea levels', if you'd bothered to check you would have seen that Pulau Sangihe certainly increased in size.

Most of the islands shown on my map are not named, but some are. For example Salebabu:

http://www.fallingrain.com/world/ID/31/Salebabu.html

And so on:

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

Ebizur said...

Maju asked,

"Does anyone knows what's the name of the sea area separating Philippines and Sulawesi?"

The sea that is bounded by Mindanao and the Sulu Archipelago of the Philippines on the north, Borneo/Kalimantan on the west, Celebes/Sulawesi on the south, and the Sangir Archipelago/Sangihe Islands on the east is called the Celebes/Sulawesi Sea.

Maju said...

"So you're claiming that the islands at present above sea level on the Sangihe Arc were submerged at more than 500 metres during the ice age? Surely you can at leasdt try to talk sense".

Above sea level measures are not a bathymetry, Terry. I worked on that. Bathymetry only refers to underwater depth measures, the same that bathyscaphe or bathysphere do not refer to airplanes.

At least you could try to make sense yourself.

"And I keep telling you it does not".

Repeating the same song once and again is not any reasoning and it's boring.

"According to my atlas the greatest distance between the islands on the Sangihe Arc is about 80 kilometres. That's much less than any one of several crossings required on any route from Bali to New Guinea".

I don't care, as it's still wider than Lombok or Makasar straits at Ice Age levels, and it is. That's what matters first of all.

Once you are in Sulawesi or Lombok things then we are not comparing access to Wallacea but which route across Wallacea, a wholly different discussion.

"Thanks for finding those maps but only by ignoring the parts of that ridge at present above sea level can you support your case".

I don't have a case: you do! I'm just dismissing your case as unfounded and meaningless.

I am not ignoring anything, the Sangihe islands are still far away from Philippines. 80km? Maybe, I measured the whole strait and it's 2 latitude degrees wide, what is 120 nautical miles or 222 km.

"I have never claimed they were 'linked'"

You said:

"And an ice age would surely expose much of the strip of submerged land between the northern tip of Sulawesi and the Sarangani Islands , off the southern tip of Mindanao".

You were wrong. Admittedly I was also wrong because I thought first you could be right on this.

But nope.

I (not you) have looked for materials that provide evidence in either direction and it's clear now that the sea between Mindanao and Sulawesi was practically the same in the Ice Age, unlike Makassar strait and other areas.

If you would have done your homework on this earlier, you would have saved me work and we would wasted less time bumping heads for sure.

But for you the facts don't seem to matter but to "prove" something you believe in. So instead of just admitting the reality and thanking me for doing the work you did not do yourself, you are instead looking for some other explanation, some "but" that you can use in your pseudo-scientific endeavor.

First it was the pretext of a tag that hid... well, nothing of importance really. And now is that you find that the "strait" is 80 km wide (at least, the whole passage is more than 200 km wide).

By comparison Bab-el-Mandeb, that you consider impassable has only 30 km, some 10-15 in the Pleistocene. Gibraltar has 14 km., only slightly less in the Pleistocene. And lombok strait is only 18 km wide, roughly what Makassar strait had in the Pleistocene at its narrowest point too. And the crossing into Crete implied also distances of less than 20 km.

80 km is just too wide. Not only because of the distance but because of the crucial factor that you can't see the other side at departure: you therefore cannot know that there is a land you can cross to at the other side.

Actually you'd be able to see the horizon south of the northernmost Sangihe is. from a good height (like 700 m., perfect weather) in theory but the islands themselves wold be so small that it's highly unlikely that nobody would ever spot them or, if they did, think they being land and not a whale or whatever.

Not absolutely impossible maybe but well above the range of other strait crossings we know or suspect of at Paleolithic technological level and time frame.

Maju said...

About the name of the "strait" between Mindanao and Sulawesi, I can't find any name anywhere. Sulawesi Sea refers to the area west of this passage and Sarangahi Arc refers to the islands, not the water body.

It doesn't seem to have a name nor be considered any strait nor sea on its own right.

terryt said...

"I measured the whole strait and it's 2 latitude degrees wide, what is 120 nautical miles or 222 km".

But no individual island is more than 80km from it's nearest neighbour.

"80 km is just too wide ... Actually you'd be able to see the horizon south of the northernmost Sangihe is. from a good height (like 700 m., perfect weather)"

Not just the horizon. It is really easy to see very small islands 80km away from less than 700 metres altitude. I do it every day. I agree that is only the first step, but:

"Once you are in Sulawesi or Lombok things then we are not comparing access to Wallacea but which route across Wallacea"

Getting to Lombok is again just the first step. Several crossings beyond Lombok involve gaps of more than 150 kms.

"Repeating the same song once and again is not any reasoning and it's boring".

And I give up. You're obviously so committed to your pet theory you refuse to consider any other possibilities. 'Pig-headed' we call it.

Maju said...

See: Wikipedia: Horizon.

At 828 m over sea level (one of the examples) you can see the horizon at 103.75 km

So I estimated that for 80 km it'd be something like 700m. But it's in fact more like 600 m.

Whatever the case you won't spot the "tiny" island (by effect of the distance) in the huge ocean unless you know what you are looking for.

"Getting to Lombok is again just the first step. Several crossings beyond Lombok involve gaps of more than 150 kms".

It doesn't matter because you can always cross to Sulawesi directly from Borneo, which is necessarily a lot easier than doing from Philippines.

Anyhow I'd check that claim of you but right now the link to the maps site is not responding. 150 km does not seem realistic. Nobody would have ever crossed into New Guinea or Australia if that would be the case. I suspect that you are measuring distances without consideration of Ice Age actual land and sea sizes.

Maju said...

"And I give up. You're obviously so committed to your pet theory you refuse to consider any other possibilities".

"My pet-theory"? Which one?

Here we have a saying about thieves believing all to be like them, if you know what I mean.

terryt said...

"'My pet-theory'? Which one?"

Your belief that humans crossed the Bab al Mandab by boat, travelled all the way to SE Asia with boats, and then moved along the Sunda Islands to New Guinea and Australia by boat.

"I suspect that you are measuring distances without consideration of Ice Age actual land and sea sizes".

Yet you can claim:

"you can always cross to Sulawesi directly from Borneo, which is necessarily a lot easier than doing from Philippines".

Definitely not true. Try looking at bathymetry between those islands. But Borneo to the Philippines is reasonably easy, via the Sulu Archipeligo.

"150 km does not seem realistic. Nobody would have ever crossed into New Guinea or Australia if that would be the case".

That's pretty much the distance often mentioned. Yet humans did manage to reach Australia. So perhaps they didn't go via that route.

"So I estimated that for 80 km it'd be something like 700m. But it's in fact more like 600 m".

The Mokohinau Islands are clearly visible (on a clear day) from a friend's house at an altitude of much less than 200 metres. They are 70km away from the house. So either your calculation is out or Wiki is wrong.

"Whatever the case you won't spot the 'tiny' island (by effect of the distance) in the huge ocean unless you know what you are looking for".

You're obviously not a practical person are you. Do you have a job? You certainly won't spot a little island from sea level until you're fairly close. But if you've already seen it in the distance you will know in which direction it lies. Besides which it's probably less important to know where a particular new island is than it being useful to be able to see your home island once you've reached it accidently. You can then go home and tell your relations that you've found an uninhabited tropical island. A situation that has been described as equivalent to winning the lottery these days.

And we can be fairly sure that any crossing from Mindanao to Sulawesi would not have been made by people new to salt-water boating. As well as the main islands of Mindanao and Luzon the Philippines consist of more than 7000 islands. Plenty of places to practice salt-water boating. And we can probably assume that humans had moved through the Sulu Archipeligo by boat. So even if humans did leave Africa with boats (which I think is doubtful) we can be fairly sure they greatly improved them in the Philippines, a region ideal for perfecting open water boating.

The reason I'm doubtful about OoA boating is that neither the Bab al Mandab or Gibraltar are close to any islands suitable for perfecting the technique. And the islands near Crete seem not to have been inhabited at the same time as Crete. Unlikely if the humans that reached Crete had efficient boats.

Maju said...

"Your belief that humans crossed the Bab al Mandab by boat, travelled all the way to SE Asia with boats, and then moved along the Sunda Islands to New Guinea and Australia by boat".

What's wrong with that? It's obvious that they could do it, as demonstrated by the presence of Homo sp. in Crete and Flores before the arrival of our species. That's not a "pet theory" but a quite reasonable conjecture, which is quite mainstream.

What I have quite clear is that in all this time you have provided absolutely no evidence against it, just hammering once and again around fuzzy "Wallacean" conjectures that don't even make much sense most of the time.

You are the one making the rather unbelievable claim that our kin could never have crossed such a narrow passage as the Red Sea because they had no boats or, in other words: they were dumb because they had been living in the Eritrean coasts for many millennia already and had crossed the Nile and the Sudd swamps and surely many other broad rivers, lakes and waterways populated by dangerous animals like crocodiles impeding swmming.

It is you who is claiming that people "could fly" over the waterways and you are doing so against all available evidence.

When I draw a map of Africa I normally skip the rivers and even lakes, but they are there. How could people cross the Nile without boats? What about the Zambezee? And in Asia already, how could people cross the Ganges, the Mekong, the Obi, the Bosphorus channel (which has always been a waterway) the Euphrates without some sort of efficient boat or raft?

And how could sufficient numbers of Homo sp. arrive to Flores or Crete in order to establish a viable population that lasted tens of thousands of years? Swimming across the open sea with pregnant women and small children?

No kidding please!

"Try looking at bathymetry between those islands".

I have POSTED several links to bathymetries between those islands already and the Makassar strait was narrow in the Ice Age by the southern extreme!

And you still have the Lombok way anyhow.

"But Borneo to the Philippines is reasonably easy, via the Sulu Archipeligo".

It's not clearly true but, anyhow, it doesn't matter because that only brings you further away from Sulawesi, not closer.

"That's pretty much the distance often mentioned".

Why don't you look at actual bathymetries? Check the useful link I posted before and that is again available.

Maju said...

"The Mokohinau Islands are clearly visible (on a clear day) from a friend's house at an altitude of much less than 200 metres. They are 70km away from the house. So either your calculation is out or Wiki is wrong".

You may be being inaccurate.

The approximate formula is d=√12.7xh (where h is in meters and d in kilometers), so the height at which you can see d=80 km is:

h=(80)²/12.7=504 m.

And for d=70, h=386 m, which fits quite well with my experience of being able to see the coast from Zugarramurdi mountains, which are over 300 m high and some 50 or 60 km from the coast.

So either your friend's home is higher or the islands are closer or you are looking at the wrong islands.

"And we can be fairly sure that any crossing from Mindanao to Sulawesi would not have been made by people new to salt-water boating".

That is reasonable but remember that people were living at the coast since some 100,000 years ago or more (can't recall exact date), just that not in Borneo but in Eritrea. Maybe they improved their skills in SE Asia but that's not too important.

The whole point of the "coastal migration" theory is that lineages that clearly coalesced in East Eurasia but seem to have arrived there "miraculously" (Y-DNA C and D, mtDNA N), did that because they were specialist coast dwellers and, sure, boaters, who were "all the time" canoeing from mangrove to beach and then again to another mangrove, as need or whim dictated. They did not exploit initially other environments and other groups did instead (Y-DNA F, mtDNA M).

It's arguable in the details, which we cannot really get to know easily, but it makes some good sense.

"The reason I'm doubtful about OoA boating is that neither the Bab al Mandab or Gibraltar are close to any islands suitable for perfecting the technique".

You can consider Arabia peninsula "an island" (and there are also smaller islands along Bab-el-Mandeb and along the Eastern coast of Africa.

I don't know why would you need islands anyhow. Often boating is faster than walking and if you live by the coast or wide rivers you sure need boats for everyday life. Waterways are the "roads" of the time when there were no roads (nor bridges!). It was that way all the time till recently. Waterways are almost never borders (nor in the archaeological record either) but often arteries, cultural and life centers.

You can see that today too: people lives almost exclusively by coasts and rivers and not on top of the mountains or the middle of deserts. That's for a reason: food and communication.

But for you it's a fetish to think that they did not have boats. Well, what can I say: you can believe what you wish but if you are going to insist again on your theory, the least you could do is to work humbly for some time in the details, the why of that belief if any. You can persuade anyone with arguments such as "I doubt that..." or "I don't think so". You need evidence of some sort and I see it nowhere.

Maju said...

And btw, look at the bathymetry of the Red Sea (blue line is 100 m). Bab, el Mandeb was nothing in the Ice Age and there was another strait north (Hanish islands, which was a peninsula in that time) not wider at all. It was like the Bosporus and the Dardanelles even with a "Marmara Sea" of sorts in between them too!

terryt said...

"You may be being inaccurate".

Do you mean to say it's not the Mokohinau Islands I can see? Don't forget that very few islands, apart from coral atolls, are actually flat.

"What's wrong with that? It's obvious that they could do it, as demonstrated by the presence of Homo sp. in Crete and Flores before the arrival of our species".

There is no indication whatsoever that those humans arrived in either region by boat. Other animals made the same crossing, at the same time, and it's doubtful those other animals arrived by boat.

"or, in other words: they were dumb because they had been living in the Eritrean coasts for many millennia already and had crossed the Nile and the Sudd swamps and surely many other broad rivers, lakes and waterways populated by dangerous animals like crocodiles impeding swmming".

I've certainly never claimed they were 'dumb'. Besides which it seems that, until relatively recently, the Nile and the Sud were actually obstacles to human expansion, not main highways. And any sort of boating on rivers and swamps is a totally different activity that salt-water boating.

"How could people cross the Nile without boats? What about the Zambezee? And in Asia already, how could people cross the Ganges, the Mekong, the Obi, the Bosphorus channel (which has always been a waterway) the Euphrates without some sort of efficient boat or raft?"

And what evidence do you have that they actually did cross those waterways by boat before about 10,000 years ago?

"That is reasonable but remember that people were living at the coast since some 100,000 years ago or more (can't recall exact date), just that not in Borneo but in Eritrea".

It's quite possible to live on a coastline and not use boats. Of course it's unlikely to happen these days because the knowledge of boating is so widespread. But that knowledge had to start somewhere. And that somewhere is almost certain to have been where the knowledge provided a huge advantage for those with it, and able to perfect the tecnique. Which rules out the Eritrean coast. Use boats once, and that's it. Whereas strings of islands is a perfect environment to develop the technique.

terryt said...

"The whole point of the 'coastal migration' theory is that lineages that clearly coalesced in East Eurasia but seem to have arrived there 'miraculously' (Y-DNA C and D, mtDNA N), did that because they were specialist coast dwellers and, sure, boaters, who were 'all the time' canoeing from mangrove to beach and then again to another mangrove, as need or whim dictated".

The second part of that is completely guesswork on your part. And generally speaking mangroves are not a very productive human habitat. My own guess is that there was nothing 'miraculous' about Y-DNA C and D, and mtDNA N's arrival in the east. They arrived by land. By what route we are yet to discover. Anyway it's impossible to believe that 'They did not exploit initially other environments'. So I agree that 'It's arguable in the details, which we cannot really get to know easily', but your theory certainly makes no sense at all.

"Often boating is faster than walking and if you live by the coast or wide rivers you sure need boats for everyday life".

Agreed. Once you have them.

"people lives almost exclusively by coasts and rivers and not on top of the mountains or the middle of deserts".

Yes. And it's obvious why. Well, it is to me if not to you. Nothing to do with boats. We need to drink water. That's why 'Waterways are almost never borders (nor in the archaeological record either) but often arteries, cultural and life centers'.

"And btw, look at the bathymetry of the Red Sea (blue line is 100 m). Bab, el Mandeb was nothing in the Ice Age and there was another strait north (Hanish islands, which was a peninsula in that time) not wider at all. It was like the Bosporus and the Dardanelles even with a 'Marmara Sea' of sorts in between them too!"

And desert on both sides. Prime human habitat, I'm sure.

Maju said...

"And generally speaking mangroves are not a very productive human habitat".

Mangroves are excellent for hunter-gatherers (mostly fishers in fact), specially if they have boats to move around quickly.

"... but your theory certainly makes no sense at all".

It's not "my" theory but the mainstream "coastal migration model".

"Agreed. Once you have them".

Once you have rivers, boats come naturally. You don't need to be Einstein to invent the boat, just some real persons with the need.

"Nothing to do with boats. We need to drink water".

It's not only us who need the water but animals and plants too, which we eat and use for everyday needs.

Also using rivers and other waterways to move around can increase the range of potential hunting zones a lot with the same effort.

"And desert on both sides. Prime human habitat, I'm sure".

Not desert: Yemen is not desert nor is nearby Asir. In fact the South Arabian coastal zone is not desert, the desert is further inland and north.

Coast is a prime, highly productive, human habitat and it is attested by archaeology as being used since long ago in the particular case of Eritrea.

terryt said...

"It's not 'my' theory but the mainstream 'coastal migration model'".

And it's obviously wrong. The Komodo dragon's survival tell's us so:

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".

Interesting, isn't it. So their survival 'in the Indonesian islands of Komodo, Rinca, Flores, and Gili Motang' surely indicates that modern humans didn't arrive in those islands until relatively recently. Therefore it's extremely unlikely the route humans took to Australia was through Nusa Tenggara. Of course I'm sure you'll be able to come up with some convoluted (and completely unbelievable) tale to account for the Komodo dragon's selective survival on just those islands.

"Once you have rivers, boats come naturally. You don't need to be Einstein to invent the boat, just some real persons with the need".

I've spent some time searching and can find no reference to African Pygmies using boats. Surely if boats were an advantage to anyone they'd be useful to Pygmies as they moved through the dense jungle. And most SE Asian Negritos are not noted for their extensive use of boats either. Besides, you don't need to be an Einstein to invent the horse and cart, but surely you're not going to tell me that it was used in the Paleolithic?

"Mangroves are excellent for hunter-gatherers (mostly fishers in fact), specially if they have boats to move around quickly".

The NZ Maoris, surely an accomplished boating people, rarely used mangroves. I assume you've never seen a mangrove forest up close. It's virtually impossible to travel far into one by boat. Or even walk through one.

"Yemen is not desert nor is nearby Asir. In fact the South Arabian coastal zone is not desert"

Near enough to desert as far as I'm concerned. And if it's as productive as you claim how come it's not densely populated today?

terryt said...

Interesting comment in the original abstract you quoted:

"We find that the paternal gene pool is sharply subdivided between western and eastern locations, with a boundary running between the islands of Bali and Flores".

Exactly where we find the Komodo dragon survived. Presumably the boundray is not so well-defined further north. And you may find this link concerning boats in pre-European Australia interesting. Especially considering it's universally accepted the Australian Aborigines must have reached Australia by boat:

http://www.archaeology.arts.uwa.edu.au/about/research/bowdler/offshore_islands_&_maritime_explorations

Maju said...

Why would us have to destroy the dragon? AFAIK the natives live in harmony with it, at least in some islands.

"A belief held by many natives of Komodo Island is that Komodo dragons are actually the reincarnation of fellow kinspeople and should thus be treated with reverence".

They don't kill the dragons but offer them sacrifices.

"And most SE Asian Negritos are not noted for their extensive use of boats either".

That's a reasonable argumentation but it must be said that neither Australian Aborigines nor Tasmanian ones nor even the Guanches of the Canary islands are famous for their boating skills. Hoewever they must have arrived to all their historical destinations using boats or rafts (unless you believe in systematic divine intervention).

"The NZ Maoris, surely an accomplished boating people, rarely used mangroves".

From Wikipedia: Ecological value of mangroves:

"Comparisons of the productivity of mangroves from different latitudes worldwide suggest that productivity and plant biomass decreases with increasing latitude. From this global pattern it is expected that mangroves in New Zealand, near their southern geographical limit would have relatively low productivity compared to their tropical equivalents".

"I assume you've never seen a mangrove forest up close. It's virtually impossible to travel far into one by boat. Or even walk through one".

You assume right: I have never seen a mangrove in real life.

However I have seen many documentaries of people living in coastal areas by mangroves and exploiting them with the help of boats, either in high tide (fishing) or, maybe more commonly, in low tide (seafood foraging).

"Near enough to desert as far as I'm concerned. And if it's as productive as you claim how come it's not densely populated today?"

Yemen and Asir were known as Arabia Felix (Happy Arabia) in Antiquity for a reason. The highest densities of population by far in all the region were there.

Even today Yemen is a populated country, with some 24 million people, almost as may as the huge and oil-rich Saudi Arabia (that would not be able to support as many people without the oil-derived wealth). In Antiquity it was the only really important region of all Arabia Peninsula.

No kidding: what do you know of Yemen?

"Its relatively fertile land and adequate rainfall in a moister climate helped sustain a stable population, a feature recognized by the ancient Greek geographer Ptolemy, who described Yemen as Eudaimon Arabia (better known in its Latin translation, Arabia Felix) meaning "fortunate Arabia" or Happy Arabia".

Yemen is not and has never been a desert. It's likely that in some periods at least it was even more hospitable than today but seldom worse.

Furthermore, all the coastal strip between Yemen proper and Oman, which is semi-arid, is anyhow habitable and has always been so.

And it had an advantage that may had promoted expansion: that it did not have any Neanderthals who would have dissuaded our ancestors from going farther.

Even if migrant Homo sapiens would have gone through the Crescent Fertile, somehow dribbling the Neanderthal hold of the territory, they would have been forced to cross desertic zones upon arrival to Iran, again having the best options of survival if they took the coastal route, where they could always find seafood and temperatures are generally warm/hot, that is: adequate to our species' natural nakedness without need of dramatic innovations such as rather sophisticated clothing and needles, which are a much greater technological barrier than mere boats, in my opinion.

Maju said...

"We find that the paternal gene pool is sharply subdivided between western and eastern locations, with a boundary running between the islands of Bali and Flores".

Terry added: "Exactly where we find the Komodo dragon survived".

You realize that the previous quote is a periphrasis for Wallace Line in all its extension, right?

Or maybe you don't?

As for the link you provide, it's interesting, of course, but what does it mean?

To me it means, first of all, that Paleolithic huntergatherers were never oceangoing sailors as Austronesians. There's a whole Era of difference between these two realities and I hope some day you realize that: you can't compare high seas Neolithic/Metal Ages' sailors such as Austronesians with huntergatherers.

You should better look at huntergatherer boaters such as Inuits.

It seems obvious to me that, considering these processes happened across many generations (millennia), peoples acquired and lost technologies according to their needs. Considering the many "stranded" peoples with alleged low boating abilities, as mentioned before, we must accept that either God exists and moves people around at whim (most unlikely) or they once had greater boating abilities than the ones we have been able to observe and record.

A quote from that link: "Some 40,000 years ago or more, ancestral Aborigines negotiated water barriers up to 65km to populate Greater Australia".

You were claiming a few days ago 150km!!!

65 km is already a major feat: much greater than any other known or suspected Paleolithic crossing (unless you believe in the Solutrean-Clovis hypothesis).

But it is smaller than your claims of 80km between Mindanao and the small islands north of Sulawesi. That's why most people discard that route.

terryt said...

"Why would us have to destroy the dragon?"

I don't know why. But it seems that as the first people spread through Indonesia to Australia they did so.

"Comparisons of the productivity of mangroves from different latitudes worldwide suggest that productivity and plant biomass decreases with increasing latitude".

So? I could have told you myself that even in New Zealand mangroves are highly productive ecosystems. They are highly productive precisely because they are so well protected. Nowhere in that article does it mention human exploitation of mangroves. Just destruction, and the need for their protection.

"Furthermore, all the coastal strip between Yemen proper and Oman, which is semi-arid, is anyhow habitable and has always been so".

And that's why the most common Y-hap there is the derived J? Not a basic haplogroup?

"You realize that the previous quote is a periphrasis for Wallace Line in all its extension, right?"

Quite. And they draw attention to the sharp 'boundary running between the islands of Bali and Flores' in contrast to the more diffuse boundary further north. Although the Austronesian expansion has clouded the issue I'm sure you will find this link interesting:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1876738/

From the link:

"All the ISEA populations studied have high levels of diversity, suggesting that they have maintained a comparatively large size over time and have not undergone substantial amounts of drift. The most diverse group is from Banjarmasin in Borneo (haplotype diversity 0.979), and the least diverse is the Tenggerese from Java (0.904)".

Strongly suggests that Java was settled later than Borneo. And:

"Haplogroup Q is most common in the easternmost locations studied (reaching 29% in Alor, where Papuan as well as Austronesian languages are spoken76), but, at 3% of the sample as a whole, it is found as far west as Borneo, indicating long-range gene flow from Near Oceania into ISEA. This may suggest traces of the voyaging corridor proposed by Terrell and Welsch,77 although it is unclear how far back in time this influence extends.9 It is worth considering the possibility that this may be a genetic trace of a conduit into ISEA for the root and tuber crops, perhaps of New Guinean origin,9 that arguably contributed far more to a change in subsistence in the Neolithic period of the region than did the introduction of rice farming from the north or west".

Settlement of Nusa Tenggara may even date to that time.

"You should better look at huntergatherer boaters such as Inuits".

But those 'huntergatherer boaters' spread no more than 3-4000 years ago, presumably because around that time they developed their boats.

"It seems obvious to me that, considering these processes happened across many generations (millennia), peoples acquired and lost technologies according to their needs".

I agree with that 100%, certainly must be so in the case of the Aborigines. Yet you also claim:

"we must accept that either God exists and moves people around at whim (most unlikely) or they once had greater boating abilities than the ones we have been able to observe and record".

We have very little way of knowing how advanced the boating technology was that allowed people to reach Australia. However we do know that in island SE Asia fairly effective boating technology goes back many thousands of years.

"You were claiming a few days ago 150km!!!"

We don't know what route the author is postulating. It's my understanding that the direct route from Java to New Guinea requires crossings of 150 kms, even at times of lowered sea level.

terryt said...

Some older interesting observations on the subject at John Hawks' site:

http://johnhawks.net/weblog/reviews/archaeology/upper/jerimalai_press_report_timor_2006.html

"The article discusses the significance in terms of a possible demonstration that the Timor route was taken by early Australian colonists, rather than the northern route via Sulawesi -- although it by no means rules out the northern route".

But:

"The find, however, raised big questions, such as why modern humans appeared to have bypassed Flores on their way to Timor".

So the survival of the dragons is significant, and you can see I'm not the only one who has a wider perspective on the subject.

More on the subject:

http://johnhawks.net/weblog/reviews/archaeology/upper/australia/australia_seafaring_conference_balter_2007.html

From that one:

"The article begins with a suggestion that the first inhabitants of Flores floated there on vegetation rafts by accident -- channel crossings being otherwise impossible for Lower Paleolithic hominids"

And another link, this one by Peter Bellwood, even older I'm afraid:

http://www.archaeology.org/9703/etc/specialreport.html

From it:

"If Australia was first reached from Timor, as seems likely, then a final sea crossing of about 55 miles, involving movement out of sight of land, would also have been required".

Note: miles, not kilometres. That's why I strongly suspect that the crossing to Sahul wasn't made from Timor, but from Halmahera to what is now New Guinea. And also from that link:

"Indeed, on Java new dates from the Ngandong and Sambungmacan sites suggest that Homo erectus may have survived far longer than previously believed, perhaps to as recently as 25,000 years ago"

Perhaps 25,000 is a bit recent, but 35,000 seems accepted these days. But it again suggests that modern humans didn't arrive there till around that time. Unless they bypassed that island too. And another interesting comment:

"These adzes suggest that manufacture of dugout canoes was technically possible by 13,000 years ago"

So Peter Bellwood obviously doesn't accept an ancient date for dugout canoes.

Maju said...

Mangrove usability.

"I first wrote this page a year ago, in early 2005, after my first real, searching visit into the mangroves, when I was, frankly, intimidated by the very idea of visiting a noisome swamp.

Since then, I've found that the local people actually choose to build their homes right in the middle of them, and that mangroves are far more productive than I imagined".

Yemen: we know that there have been Y-DNA swaps in other regions. We have mtDNA evidence in North Africa indicating colonization from Europe but if you look at Y-DNA you see almost nothing of that. Same with Yemen.

Was Java settled after Borneo? I don't know. There was no Java nor Borneo anyhow when Sundaland was settled.

Inuits expanded indeed in the last two millennia. It's irrelevant because they are just an example: we know of many other coastal dwelling, boater, cultures historical and prehistorical, some of them, like Magdalenian, with striking technological similitudes with Inuits (those proto-harpoons are not for picking the nose).

You claim:

"However we do know that in island SE Asia fairly effective boating technology goes back many thousands of years".

How many? Please don't tell me about late Neolithic Austronesians again, ok, as they are not comparable at all.

Maju said...

About the distance needed to cross the various straits into Australia, I'd suggest you do the following:

1. Open the previous link you posted

2. Open the map at fig. 1

3. Take a piece of paper and copy the scale carefully, possibly marking also 50 km marks.

4. Measure the straits.

I did and I don't see a single strait measuring more than 65 km in the usual routes. The strait south of Mindanao is still at least 80 km wide instead.

You are proposing the hardest of all possible routes!

Maju said...

About John Hawks' opinions, they are just his opinions. You two agree? Fine, now try to write a comment in his enlightened blog. ;)

While it's perfectly possible that the migration route went through Sulawesi-Malukku-New Guinea, this does not mean anything in favor of Midnanao being in it. The shortest route between Borneo and Sulawesi by far is via the southern part of the Makassar strait, which used to measure some 15 km (10-20, the scale does not have enough detail) in the Ice Age. The Bor-Sul-Mal-NG route simply must go through Makassar strait.

"Note: miles, not kilometres".

55 miles (some 90-100 km) is much more than I can measure in the map at Hill's paper.

Whatever the case, I have nothing against the Sulawesi route, my problem is about reaching Sulawesi from Mindanao, as you propose, when directly from Borneo is a lot easier.

"Perhaps 25,000 is a bit recent, but 35,000 seems accepted these days. But it again suggests that modern humans didn't arrive there till around that time. Unless they bypassed that island too".

Java was not an island back then. There's no special reason why the two species could not have shared the Sundaland peninsula, is it?

Could it be after all that our "Neanderthal" genes are in fact Erectus? Just a wild hunch.

"These adzes suggest that manufacture of dugout canoes was technically possible by 13,000 years ago"

I don't know for sure but you seem to lack imagination. While adzes may have made dugout canoes easier, they are not a strict requirement, you can perfectly use a regular axe for instance, always aided by fire, of course, which makes most of the job.

Also there are other alternatives to the popular dugout canoes: rafts and leather boats are the most commonly known ones.

Maju said...

PS- I have been refining the measures of the straits to be crossed by either route at Pleistocene sea levels, per Hills' map, and I get:

1. Northern route includes crossings of as much as 40-45 km between Sula and Buru and Ambon and Misool, then part of the Sahul continent by connection with Bird's Head peninsula in modern New Guinea island.

2. Southern route's widest strictly necessary crossing is some 50 km wide between Leti island and what was then the coast of Australia, NW of modern Darwin.

Crossing directly from Timor would require a navigation of some 65 km. but if they previously hopped to Leti the crossing is some 15 km shorter.

The southern route's first crossing is at Lombok and the northern one is at southern Makassar strait, both with some 15 km width across the Wallace Line. Island hopping between North and South Malukku would imply longer crossings, as would island-hopping through Aru islands into Sahul platform from Timor.

So, I gather that either the scenario implies two routes, branching at Sundaland or a single migration through, somewhat more likely, the northern route.

My hunch is two routes, mostly because the Papuan and Australian gene pools are quite different but who knows?

Maju said...

Erratum: "as would island-hopping through Aru islands into Sahul platform from Timor".

I mean "through Tanimbar island into Sahul platform near modern Aru".

terryt said...

"Since then, I've found that the local people actually choose to build their homes right in the middle of them, and that mangroves are far more productive than I imagined".

And did you even bother to look at those houses? Surely Paleolithic people didn't have the technology to build them.

"we know of many other coastal dwelling, boater, cultures historical and prehistorical, some of them, like Magdalenian"

Maju, don't be stupid. The Magdalenian arose long after humans had reached Australia, let alone SE Asia.

"How many? Please don't tell me about late Neolithic Austronesians again, ok, as they are not comparable at all".

Again. Stop being stupid. Boating in SE Asia goes back long before the Austronesians first formed. Austronesians were simply the product of a long period of gradually improved boating in the region.

"The shortest route between Borneo and Sulawesi by far is via the southern part of the Makassar strait, which used to measure some 15 km (10-20, the scale does not have enough detail) in the Ice Age".

That's obviously a very inaccurate map to give that distance. my map shows more like 60 kms, certainly still not an impoosible distance I agree.

"my problem is about reaching Sulawesi from Mindanao, as you propose, when directly from Borneo is a lot easier".

And even that is certainly a lot easier than going through Lombok and Flores.

"Java was not an island back then".

We don't know that for sure. The land between Borneo and Java may have been low-lying mangrove forest rather than open plain. And I remember seeing something recently that strongly suggested modern humans moved originally from the Malay Peninsular to Borneo, rather than to Sumatra. My conclusion at the time was that the low-lying region northeast of Sumatra was impassable at the time. Again probably being mangrove swamp.

"There's no special reason why the two species could not have shared the Sundaland peninsula, is it?"

It's unlikely they shared Java though, especially as there is as yet no evidence modern humans were there at the time.

"Could it be after all that our 'Neanderthal' genes are in fact Erectus? Just a wild hunch".

A 'wild hunch' that I believe to be quite likely. Especially seeing we both now agree that Y-hap KMNOP began somewhere in SE Asia (although we still disagree as to exactly where in that region).

"While adzes may have made dugout canoes easier, they are not a strict requirement, you can perfectly use a regular axe for instance, always aided by fire, of course, which makes most of the job".

I invite you to go out into the forest and try to make a dugout using the method you propose. An axe and fire is certainly sufficient to cut the tree down in the first place. But try the next step with just those tools.

"Southern route's widest strictly necessary crossing is some 50 km wide between Leti island and what was then the coast of Australia"

On my fairly accurate map that crossing is more like 125 kms. And the crossing from Timor is about the same.

"My hunch is two routes, mostly because the Papuan and Australian gene pools are quite different but who knows?"

I agree that scenario is quite likely. In fact one of the links I posted mentions a movement around 30,000 years ago, presumably by people with improved boats, who moved out beyong New Guinea and took the cuscus to the Admiralty Islands (amoung other places). But note: that is many years after Australia was first settled. And I also suspect that people from New Guinea are largely made up of people from this second movement. They were jungle-adapted people.

Maju said...

"And did you even bother to look at those houses? Surely Paleolithic people didn't have the technology to build them".

Erm? I know they are a more modern type of home but what's your point anyhow? There's a wide variety of options, including living on "docked" rafts/barges, whose lines you can take off at any moment to go to another such spot if there's trouble at home or whatever other reason (seasonal changes, whatever).

Of course this kind of lifestyle can't be proven... nor disproved either.

They could also build homes similar to those in the picture, except that using logs, branches and palm leaves instead of "modern" boards. It's not that complicated, is it?

"Maju, don't be stupid. The Magdalenian arose long after humans had reached Australia, let alone SE Asia".

It is still an Upper Paleolithic culture and one that has its roots in the Early UP of West Eurasia (Aurignacian and related cultures).

You are in denial: first Inuits are not a valid example because they are recent, then Magdalenians are not a valid example because they are not sufficiently old, no matter these two cultures are from cold waters, which need high specialization and that our knowledge of what preceded them is limited (we know that Solutreans had fishing hooks and that early European Homo sapiens had a diet rich in fish).

What are you asking for a time machine and a photo. You would question that too because you are a fanatic.

What have you to say in regards to the abundant evidence of inhabitation at coastal sites already in Africa, long before the OoA? Nothing will serve as evidence for your stubborn rejection of evidence and your lack of anything of the like.

You are wrong and you are anti-scientific. I'm fed up!

"Again. Stop being stupid. Boating in SE Asia goes back long before the Austronesians first formed".

Really? Evidence?

"That's obviously a very inaccurate map"...

It is not and I have produced consistent bathymetries that you can check, your map seems the one to be wrong if anything.

You are producing no evidence and demanding more and more. Stop that!

"And even that is certainly a lot easier than going through Lombok and Flores".

"Stop being stupid", as you say.

Just because you can repeat your idiotic claims once and again, that does not make them more real. Not at all. It just makes you less and less credible.

You are bordering troll-ism. Produce some data or give up. I'm fed up!

Maju said...

"We don't know that for sure".

What?! Are you now denying the existence of Sundaland?

"My conclusion at the time was that the low-lying region northeast of Sumatra was impassable at the time. Again probably being mangrove swamp".

Mangrove only lives in the tidal zone, not hundreds of kilometers inland.

In this map the blue line means 100 m under sea level (today), what in most of the Pleistocene was some 20 m above sea level. No mangrove can live there.

"It's unlikely they shared Java though, especially as there is as yet no evidence modern humans were there at the time".

There's no direct evidence of AMHs anywhere in the whole SE Asian region before 20 Ka or so. So?

"A 'wild hunch' that I believe to be quite likely".

I'm already repenting. You 'believe'? I panic at such kind of statements. You believe too much too early without due consideration.

Actually, unless Neanderthals were themselves admixed with Erectus intensely, this would make no sense. We should not detect Erectus genes by mere comparison between two H. ergaster derivatives.

"I invite you to go out into the forest and try to make a dugout using the method you propose. An axe and fire is certainly sufficient to cut the tree down in the first place. But try the next step with just those tools".

Actually, I'd say that it should be easier to hollow the trunk with a hand axe (no fire) than to pull down a tree with it.

I don't know how to use fire in this so I'm just thinking in getting an axe and start carving the log. It's a matter of patience and doing it while the wood is still fresh.

I (very roughly) estimate a month of 4 hrs. journeys. You can positivize your stubbornness into such things. Wood is workable.

"On my fairly accurate map that crossing is more like 125 kms"

Does your "fairly accurate" map even include a bathymetry? Or are you looking at modern day coasts? I can't tell.

Why don't you upload a scanned copy online so we can judge?