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.

Wednesday, August 4, 2010

Genetic flow across the Strait of Gibraltar


Researchers from the University of Geneva try to address the somewhat complex problem of genetic flow across the Strait of Gibraltar, between SW Europe and NW Africa, in a new paper:


Mathias Currat et al., Human genetic differentiation across the Strait of Gibraltar. BMC Evolutionary Biology, 2010. Open access.

They analyze a number of genetic variables such as Y-DNA, mtDNA, blood groups ABO and Rh, and other antigen/antibody systems (HLA, MNS and GM) and produce a number of statistical analysis for them. However they acknowledge that their modeling on continuous, rather than punctual, gene flow, may have some difficulties providing clear-cut answers.

Still they do produce results that are in general favorable for the P scenario (flow since 20,000 years ago), excepting to some extent the Y-DNA, which is probably a secondary sex-biased element overall.

Above: The scenarios are described in page 10 and in the supplementary material. P is "Paleolithic" flow beginning 20,000 years ago (Ibero-Maurusian or Oranian culture) among small populations, N instead is "Neolithic" flow among larger populations. PN is intermediate and PNI is intermediate but also considering the Islamic expansion (though details are not provided).

They also produce results of quite greater affinity within each region (SW Europe and NW Africa) than between them, as expected.


Notice in this graph (and also fig. 3) how autosomal markers' curves and that of mtDNA (MT-HV1, in blue) are very similar, however while mtDNA tends to lesser homogeneity within regions, it strongly tends to higher homogeneity instead across the Strait. This I interpret as meaning greater affinity across the strait in the Paleolithic, before the primarily male expansions reflected in the Y-DNA, altered the scenario.

Sadly no graphs for each of the two regions is provided, what does not allow us to make a more detailed assessment, like whether the Y-DNA expansion happened at both coasts or not, and, if so, if they were fully comparable or somehow different. I say this because the apparent SW European origin of North African mtDNA H (c. 25%) detected by Cherni 2008, and probably other haplogroups like V and K, would be consistent with a more European-like population before the E1b1b1 expansion in the context of Capsian culture (in North Africa) which is probably associated also to the expansion of Afroasiatic languages. However I see no reason to support a similar male-biased expansion in SW Europe.

Let the authors conclude:

Conclusion

While contrasted conclusions were obtained by previous studies based mostly on single genetic loci, our study clarifies the role of the Strait of Gibraltar regarding its permeability to gene flow. Indeed, our multi-locus approach led us to take into account variations between loci when trying to infer past history of human populations around the Gibraltar area. We were thus able to show that the Y chromosome on one side, and HLA-DRB1 on the other side constitute two extreme cases of very strong and very weak (respectively) genetic differentiations between populations across the Strait. The lack of genetic differentiation for HLA-DRB1 is particularly interesting because it can be explained by balancing selection (with a coefficient of selection estimated here to be around 2%). Given the huge worldwide dataset available for this locus, a better understanding on the mechanisms of selection at HLA loci could be very helpful to the study of human evolution, and more generally MHC. Our results obtained for Gibraltar have to be confirmed by further studies in other areas, especially where gene flow between populations is reduced. This work thus constitutes a step forward towards a better characterization of the combined effects of selection and demography on the genetic structure of populations, and especially on their genetic differentiation.

130 comments:

aargiedude said...

Thanks, Maju, another great article I wouldn't have known about if it weren't for you, plus your opinions.

This past week I've been trying to understand why the frequency of R1b1b2 in Latin America, excluding E1b1a and Q so as to measure only their western y-dna, is much lower than would be expected given their supposedly Spanish-only origin. In trying to understand what's going on, I've produced a map showing the precise frequencies of R1b1b2 in Iberia. The thread is here:

http://www.worldfamilies.net/forum/index.php?topic=9556.0

Or a link to the map:

http://img508.imageshack.us/img508/5464/decypheringthewesternyd.gif

I'm sure you'll be interested in the topic itself, but what's pertinent to your post is, once again, we observe an east-west cline in the dna of Iberia, in this case of R1b1b2. It's very clear.

Here's a previous map I made about the distribution of y-dna E-M81 in Iberia:

http://i88.photobucket.com/albums/k178/argiedude/IberianM817201samples.gif

And the east-west cline seems also apparent in mtdna L, which I remember you believe probably has a major component of paleolithic origin. It would be interesting to look at the distribution of U6, also.

Maju said...

Let's see: if I understand you and your graphs correctly, you are saying that:

1. Latin Americans have in some cases lower R1b1b2 that the average Spaniard (but only in some cases less than the average Western Iberian).

This is logical because Latin Americans are largely mestizo (the only area strongly European is the South Cone and you skipped it) and also because it was not "Spain" but Castile which colonized the region. Spain was only formally created after the War of Spanish Succession, when the autonomy of the states of the crown of Aragon were suppressed and the Kingdom of Spain was for the first time formally described. Earlier, Aragonese subjects could not colonize America nor trade with it, at least in principle, just like Belgians or Italians could not either: it was a Castilian enterprise. Basques, even if highly autonomous were formally annexed to Castile, so they did took part.

Castile back then included Leon, Galicia, Asturias, Extremadura and Andalusia, one of the most densely populated areas. Its capital was in Valladolid (partly moved to Madrid since Philip II) but a secondary capital existed at Seville, which was also in charge of America and all other colonial affairs (Philippines in particular, which was a dependency of Mexico).

So for me your Latin American figures are no surprise, excepting the high ones of Ecuador maybe.

Unless you are excluding already all possible Native American and African Y-DNA before processing.

(follows)

Maju said...

2. Cline in Iberian R1b1b2.

I am aware of it. I already discussed it somewhat here and here, though my real focus was on the rarer clades of possible Neolithic or Celtic origin.

IMO, there was a particularly strong (well, R1b1b2 still makes up more than 50% almost everywhere, so not so strong) Neolithic and/or post-Neolithic colonization in West Iberia. This may have been reinforced by Celtic arrivals, specially in regard to haplogroup I (many doubts). Notice that where highest R1b1b2 is found is roughly where detected subclades, specially R1b1b2a2f, is also found.

Most interesting from your map for me is the case of Murcia (no data in Adams' paper), which shows quite high apportions of R1b1b2. I and Heraus had also reached certain consensus at his anthropometry blog that Murcians and in general East Iberians are more akin in looks to Basques, than West Iberians. It's probably related.

My reasoning is that Murcia lacks of Neolithic colonization (which is anyhow limited except in southern Valencian Country) and even of Sauveterrean/geometric Epipaleolithic influence, remaining in the Azilian/laminar stage. The same happens to a large extent in East Andalusia. This is also the area of El Argar civilization and most likely the origin of Iron Age Iberian culture (and language).

There is curious Basque sounding toponimy in all the area (Iliberri, now Granada; Andarax, the river of Los Millares, El Argar itself sounds Vascoid) but of course it must be Iberian, whatever the differences and similitudes. (I'm rather for Vasco-Iberian in linguistics but with a likely Paleolithic origin, at most Neolithic).

"And the east-west cline seems also apparent in mtdna L, which I remember you believe probably has a major component of paleolithic origin".

In North Africa. I am not sure when E1b1b1b (M81) arrived to Iberia, as its distribution correlates best with other likely Neolithic lineages, specially E1b-V13, what to me suggests that Cardium Pottery people arriving to Portugal (and by extension the Iberian NW later on) first went through admixture in North Africa (possibly North Morocco, where CP is also found).

Maju said...

Oops, first link did not make it ("here" without link above): it is this post.

Maju said...

Btw, I just noticed the shockingly high apportions of E1b1b1b-M81 in Cantabria. What's up with that? I know that there must be a relatively marked genetic barrier west of Bilbao but where did Cantabrians got so much M81 (nearby Asturians are also anomalously high in E in general in Adams' data and others but it's more marked for E-V13 and E*, I think).

Maju said...

Getting back to American R1b1b2. Now I understand that you are calculating it on "Western Y-DNA" only.

IDK, there must have been founder effects and notice that also that a lot of migrants were convert Jews and Muslims who tried that way to escape the pressure of the Inquisition. Later on, there was another wave in the 19th and 20th centuries that must have included a lot of people from other European origins: Germans, Italians, etc. Additionally early on North Africans were enslaved on religious grounds sometimes, though not sure how influential this could be overall.

aargiedude said...

it was not "Spain" but Castile which colonized the region.

Whoa! I wasn't aware of that. It's very significant.

Castile back then included Leon, Galicia, Asturias, Extremadura and Andalusia, one of the most densely populated areas.

Thank you very much. It's basic history, sure, but I was totally blind to this fact. I had assumed that the unification of Spain allowed it to then launch grand enterprises such as the conquest of America. Very interesting.

a secondary capital existed at Seville, which was also in charge of America and all other colonial affairs

That blows. So presumably the main port of departure, and thus region of origin of many of the men that participated in the crucial first half century of conquest of America would have been Andalucia.

Btw, I just noticed the shockingly high apportions of E1b1b1b-M81 in Cantabria. What's up with that?

Maca-Meyer (2003) found that the Pasiegos of Cantabria had 24% M81. But his other Cantabrian samples had 8% M81, which is also quite higher than the typical 3% or 4% found in most of Spain.

a lot of migrants were convert Jews and Muslims who tried that way to escape the pressure of the Inquisition

My initial angle on this subject was precisely that there must have been a lot of Sephardic Jews amongst the conquerors of America. Then I noticed something very interesting, in another haplogroup, that changed my point of view. I'm going to post that tomorrow.

Later on, there was another wave in the 19th and 20th centuries that must have included a lot of people from other European origins: Germans, Italians, etc.

This was extremely limited, in numbers and location. The overwhelming majority of Europeans went to Argentina and south Brazil. I don't think people of European ancestry ever constituted more than half of the population of Argentina, and in the case of Brazil, they probably never reached even 1/5 of the total population.

aargiedude said...

Back to the topic of this thread: I can't find the study, just an abstract at PubMed. Do you have a link?

Maju said...

"I had assumed that the unification of Spain allowed it to then launch grand enterprises such as the conquest of America".

Not really. There was of course a warrior caste in need of new occupations after the Reconquista but there were other choices such as the Mediterranean or Europe itself.

The main dynamo of exploration was without doubt Portugal, a relatively small country which, after the end of the Reconquista in the 13th century (Granada was just a leftover) had nowhere to go except to the sea. First they fought in Morocco, then they searched for the fabled gold of Sudan (now Mali) and began trading in slaves and organizing the first colonial plantation ever: Madeira. Eventually the went in search of the Spice Islands.

Colombus had an alternative plan but the Portuguese were already set, so he could only hope financing from Isabel of Castile (France and England were still too busy with their Hundred Years' War). Isabel postponed the decision until Granada was conquered and well, suddenly they were faced with an unexpected continent to plunder and conquest. Fiesta!

They arranged a treaty with Portugal and there was no other competence for almost a whole century. The Native American powers were no match: they were in the Chalcolithic Age.

Under Isabel's grandson and direct heir, Charles V, not only were Castile (with the American colonies) and Aragon (with Sardinia, Sicily, Naples and Milan) but also all the possessions of the crowns of Austria and Burgundy. Four large disparate realms and the diffuse joint title of Roman Emperor and King of Germany fell on him.

At that time Spain still meant the Iberian Peninsula, i.e. Hispania. When Charles' son, Philip II, also acquired Portugal (lost Austria and the HRE to his half-brother) the historical Habsburgian Spain was created and the name of Spain was more and more used for the complex monarchy, which had its see at Madrid. After the loss of Portugal (and its colonial empire) the name remained in use. However there was no polity of such name, just a dynasty.

When the dynasty came to its end in the early 18th century, it would have been inherited by the Dauphin Louis of France, eldest son of Louis XIV, the King Sun. If the "natural" dynastic course would have been allowed, France, Spain, most of Italy, as well as Belgium and Luxemburg, and most of America would have fallen under one single and absolutist monarch.

Naturally, the other European powers opposed, eventually leading to the War of Spanish Succession and the loss of the Low Countries and Italian possessions to Austria, as well as Gibraltar and Minorca to England. But for Spain as such it meant the birth of a unified state of that name under the Bourbon dynasty and the suppression of the self-rule of Aragon, Catalonia, Valencia and Mallorca, which had fought in the Austro-English camp (precisely to retain their self-rule). Basque provinces' immense autonomy was kept however until the 19th century.

There are not many people of Catalan or Aragonese ancestry in Latin America. There are some but surely they arrived in the last three centuries.

Maju said...

"So presumably the main port of departure, and thus region of origin of many of the men that participated in the crucial first half century of conquest of America would have been Andalucia".

Andalusia was without doubt important but the rest of the Kingdom also participated: both most famous conquerors, Cortés and Pizarro, were from Extremadura, an impoverished "marche" that produced many more colonists than its size would suggest. Basques, Galicians and Andalusians, with the occasional Portuguese or Italian, participated most often as sailors but also as land colonists, along with inland Castilians. But I'd say that Extremadura, inland Castile and Andalusia, were surely the main exporters of colonists. I haven't seen any study anyhow, just my best guesses.

"Maca-Meyer (2003) found that the Pasiegos of Cantabria had 24% M81. But his other Cantabrian samples had 8% M81, which is also quite higher than the typical 3% or 4% found in most of Spain".

Ah, Pasiegos. El Pas is an isolated valley near Asturias and León, cannot be extrapolated to the whole province. Now I understand.

"My initial angle on this subject was precisely that there must have been a lot of Sephardic Jews amongst the conquerors of America".

There were of course but how many Jews were in Spain in 1492? I know no figures but they cannot have been that many. Anyhow, as the colonists were converts we should see that influence, if they were numerous, in modern Spain and we don't (cf. Adams: almost no J1 in Iberia but more than 20% among Sephardic Jews).

"This was extremely limited, in numbers and location".

I have read that it was very important and is called by some "the whitening of Latin America", as earlier criollos (Americans of European ancestry) were a tiny minority.

Also consider Lebanese/Syrian in this wave. They are an important and rather affluent minority in many American countries.

"The overwhelming majority of Europeans went to Argentina and south Brazil".

And Chile and Uruguay...

But they had also their impact in places like Venezuela (quite strong), Mexico, Cuba, Paraguay... I don't know much about the details but that's what Latin Americans have told me. I'd read the relevant Wikipedia articles/sections if you want to make sure.

"... they probably never reached even 1/5 of the total population".

Can be enough to make an impact of the subtlety you indicate, I believe. Specially because you are measuring only people of male European ancestry, not all those of Native American or African paternal ancestry.

"... I can't find the study, just an abstract at PubMed. Do you have a link?".

Oopsa! I forgot to paste the link. Now it's corrected, check the post again.

Maju said...

Btw, I remember from my Spanish language classes in school, that it was argued that lowland areas in America, such as Venezuela, were colonized primarily by people from Andalusia, while highland areas like Mexico or Colombia were by people from Castile. This had, apparently something to do with American dialects and accents. I wouldn't forget Canary Islands in the equation as nearly every ship stopped there in their way to America.

Maju said...

Galicians were also very important in the 19th-20th century migration wave, up to the point that in many places Spaniards in general are called "Gallegos".

terryt said...

"P is 'Paleolithic' flow beginning 20,000 years ago (Ibero-Maurusian or Oranian culture) among small populations"

We'll assume they crossed by boat, but considering that humans had made it across Wallacea into Australia and New Guinea by 50,000 years ago (at least) 20,000 years is but yesterday. And remember that Y-hap R1b1b2's ancestor was Y-hap KMNOPS*. And Y-haps M and S, along with several unclassified Ks, were probably amoung these first people to cross Wallacea. That leaves roughly 30,000 years for improved boating to reach the western end of the Mediterranean.

But interestingly humans don't seem to have made it to many Mediterranean islands (apart from Crete!) until much more recently than 20,000 years ago.

Maju said...

Terry: Gibraltar Strait is 13 Km wide, quite less in the Pleistocene, Corsica or Ibiza, the closest islands to land, are out of sight.

I understand that being in sight or out of sight makes the difference, rather than having or not having "boating technology". However in your fabled Wallacea, they must have crossed several times those distances, what is pretty much incredible-but-true.

"And remember that Y-hap R1b1b2's ancestor was Y-hap KMNOPS*".

MNOPS, right. The asterisk is used not for the root but for paraphyletic unclassified ("other") siblings and therefore cannot be the ancestor, the same your brother is not your father.

"And Y-haps M and S, along with several unclassified Ks, were probably among these first people to cross Wallacea".

No, probably not. IMO it was people of Y-DNA C (mtDNA N and M). MNOPS and mtDNA R arrived later, I understand.

Similarly in Japan and Andaman it was people with Y-DNA D and mtDNA M (N in Japan too), again MNOPS (NO in this case) being a secondary arrival.

On thing we have to understand is that culture and haploid lineages are not exactly the same thing. Much less that particularly pragmatic aspect of culture which is technology.

There were no patents in the Pleistocene: people copied others at will. We still do in fact.

Maju said...

Btw, you may want to read Petraglia's latest paper while it's available at Scribd (I'd download it but you need a Facebook account and I don't want to have one as a matter of principle).

There are a lot of Middle Paleolithic sites in South Arabia dated to c. 80-70 Ka. And then again from 35 Ka onwards.

aargiedude said...

Maju:
When Charles' son, Philip II, also acquired Portugal


Wow, I had no idea. I just read about it on wikipedia:

"he [Philip II] made his nephew Albert of Austria his viceroy in Lisbon. In Madrid he established a Council of Portugal to advice him on Portuguese affairs, giving excellent positions to Portuguese nobles in the Spanish courts, and allowing Portugal to maintain autonomous law, currency, and government."

So from 1580 to 1640 Portugal was formally part of Spain, yet had such autonomy that it functioned almost like a separate country, even continuing to rule its overseas colonies? This is all rather, something, I have to say!

aargiedude said...

Maju:
the historical Habsburgian Spain was created and the name of Spain was more and more used for the complex monarchy, which had its see at Madrid


That is really something. Thanks.

Maju:
and the suppression of the self-rule of Aragon, Catalonia, Valencia and Mallorca, which had fought in the Austro-English camp (precisely to retain their self-rule). Basque provinces' immense autonomy was kept however until the 19th century.

There are not many people of Catalan or Aragonese ancestry in Latin America. There are some but surely they arrived in the last three centuries.


Very interesting. Makes perfect sense.

I'd say that Extremadura, inland Castile and Andalusia, were surely the main exporters of colonists. I haven't seen any study anyhow, just my best guesses.

Aha, that's a key point. So it's not a known fact. Educated guesses yes, but detailed statistics of how many from where, no.

aargiedude said...

Ah, Pasiegos. El Pas is an isolated valley near Asturias and León, cannot be extrapolated to the whole province. Now I understand.

Totally agree. But what caught my attention is that Maca-Meyer also tested non-Pasiegos from the rest of Cantabria, and they had, I think, 8% M81, notably higher than elsewhere in Iberia. And a different set of samples from Cantabria, in the yhrd database, have 10% M81. So, definitely not 24%, but it still seems there's an unusually high rate of M81 in Cantabria overall, around 10%.

I wouldn't forget Canary Islands in the equation as nearly every ship stopped there in their way to America.

Some time ago I was looking at the western mtdna of Latin Americans and found out, thanks to the Canary-specific lineages of U6b1, that possibly half, maybe most, of Cuba's and Puerto Rico's western mtdna is probably of Canarian origin. On the other hand, interestingly, I didn't find even one sample of this specific lineage amongst continental Latin Americans.

Galicians were also very important in the 19th-20th century migration wave, up to the point that in many places Spaniards in general are called "Gallegos".

This is a local thing in Argentina, indeed, but nowhere else. Actually, it's more like a Buenos Aires thing.

----------------------------

And thanks for the link to the study. :)

Maju said...

I don't know how to explain the rather high rates of haplogroup E1b1b1 variants in both Asturias and Cantabria. One would expect these regions to keep much of the original lineages of the Paleolithic, which with all likelihood E is not. They should be more like Basques and other Pyrenean peoples, right?

If so, there must have been some sort of heavy recolonization at some point but when? And why would this carry specifically the E1b1b1 haplogroup and not an assorted mixture of various lineages from wherever the colonization originated?

An alternative possibility is that in the Solutrean genesis, there seems to have been a migration from somewhere in Iberia proper (Iberian facies of early Solutrean) to Asturias, however this did not affect Cantabria nor the Basque Country. IF the early Iberian Solutrean implied interaction with North Africa (Oranian genesis) and IF this interaction already included Y-DNA E, as seems to imply absorption of mtDNA U6 this might be an explantion. But SE Iberia does not display these genetic traits, so it must have happened in Portugal and nearby areas if anything. Or maybe in the plateau, where was once reported an abundance of Solutrean sites near Madrid, whose remains are now lost.

This would make some sense for Asturias but has some issues, mainly that it implies that E1b1b1 was already spread into NW Africa, not only E1b1b1b-M81 but also other variants. This would contradict my usual assumption of E1b1b1b spreading in the Epipaleolithic with Capsian culture.

Its expansion to Cantabria then might have been associated to Magdalenian of facies Cantabrian-A, which is virtually non-existent in the Basque Country (facies B, proto-Azilian). However facies B is also found in Cantabria.

I say this because only founder effects in a Paleolithic context might seem to explain the unusual frequencies, right? Because we know that the Romans forced Cantabri to move down to the valleys but we don't know of any massive settlement at any time, much less from an origin like North Africa that could explain these apportions.

"... maybe most, of Cuba's and Puerto Rico's western mtdna is probably of Canarian origin. On the other hand, interestingly, I didn't find even one sample of this specific lineage amongst continental Latin Americans".

That's interesting, thanks.

"This is a local thing in Argentina"...

Maybe but Galicians colonized elsewhere in that period. Fidel Castro is himself 100% Galician by immediate ancestry (just an example).

"So from 1580 to 1640 Portugal was formally part of Spain, yet had such autonomy that it functioned almost like a separate country, even continuing to rule its overseas colonies? This is all rather, something, I have to say!"

That was the general rule for all Habsburgian realms: they shared a common monarch and therefore a common foreign policy but in internal affairs they were essentially independent. The exception was Castile, whose fueros (constitution or statutes) were revoked by Charles V when Castilian gentry and cities revolted unsuccessfully against his "Flemish" rule (look up "comuneros"). But the same that Portugal could rule its own affairs, so did Aragon, Catalonia, Sicily or Flanders, depending on their own laws and traditions.

terryt said...

"No, probably not. IMO it was people of Y-DNA C (mtDNA N and M). MNOPS and mtDNA R arrived later, I understand".

I agree with that. However people began moving beyond New Guinea to the Northern Solomon Islands about 35,000 years ago and the main Y-haps found through that region are K-derived. So that implies some sort of improved boating technology by that time.

"There were no patents in the Pleistocene: people copied others at will. We still do in fact".

Almost certainly more so today than in the Paleolithic. And even today jobs (such as fishing or farming) and recreation pursuits (such as boating) tend to pass from father to son. With smaller populations such connection between haplogroup and technology is likely to have been even closer, although I agree completely that technology has traveled through the human population faster than have the haplogroups.

terryt said...

Thanks for that Petraglia paper although my computer doesn't like it for some reason. Interesting comments regarding the timing of modern humans, well in excess of 100,000 years. Of course we've both suspected that for some time. And the comment that the prehistory of the region is more complicated that usually assumed.

"There are a lot of Middle Paleolithic sites in South Arabia dated to c. 80-70 Ka. And then again from 35 Ka onwards".

You will understand that I find the change around 35,000 years particularly interesting.

"I don't know how to explain the rather high rates of haplogroup E1b1b1 variants in both Asturias and Cantabria".

I noticed that the original article you posted does tend to imply that the late Paleolithic movement across the Gibraltar Strait was from Africa to Europe, rather than in the other direction.

Maju said...

Are you telling me that crossing into the Northern Solomon islands is harder than crossing all that island labyrinth of Wallacea to New Guinea and Australia?

I don't see why? In this map, it does not seem to be harder than island-hopping through Malukku. Also AFAIK, those peoples are still mostly Y-DNA C2 (and mtDNA M), right?

Maju said...

"I noticed that the original article you posted does tend to imply that the late Paleolithic movement across the Gibraltar Strait was from Africa to Europe, rather than in the other direction".

Does it? It also deals with the fact of North Africa having large amounts of European mtDNA (which we know, by Safi, that are original from Europe, which is also consistent with all the archaeology we know). It does mention Oranian (Iberomaurusian), which may be derived from Iberian Gravetto-Solutrean. The opposite is impossible, though specific influences in design at later stages, such as the evolution of back-tipped points may be of African (Aterian) derivation.

I don't think it implies any direction for any possible flow, nor it can be derived from their data either.

terryt said...

"Does it?"

From the link (page 7):

"indicate that the strait was permeable to human migrations, with an estimated contribution of North-West Africa to Iberia of 18% for mtDNA compared to 7% for Y-chromosome lineages".

"Are you telling me that crossing into the Northern Solomon islands is harder than crossing all that island labyrinth of Wallacea to New Guinea and Australia?"

Evidently that is the case. The map you linked to technically doesn't include any of the Solomon Islands. Just New Britain and New Ireland, although they too may have been settled long after New Guinea (or especially Australia). Your map doesn't go as far as Bougainville even.

"Also AFAIK, those peoples are still mostly Y-DNA C2 (and mtDNA M), right?"

Partly correct. Y-DNA C2 moved in later, as Ebizur showed us some time ago. Y-haps M and K3 are the main early ones beyond New Guinea. Early mtDNA is presumably M-derived although B has come to dominate most of the region, possibly introduced with Y-hap C2 and the Austronesians.

Maju said...

"with an estimated contribution of North-West Africa to Iberia of 18% for mtDNA"

Cannot be. U6 has like 1/3 of those figures in Iberia, possibly even less. Inversely only mtDNA H makes up 25% or more of North African mtDNA and it's indeed of SW European derivation. As I have mentioned before, I understand that possibly as much as 50% or North African mtDNA has European origin.

"The map you linked to technically doesn't include any of the Solomon Islands".

You can see Boungainville by the East.

"Your map doesn't go as far as Bougainville even".

It does. It's cut in half by the east of New Guinea.

"Y-DNA C2 moved in later, as Ebizur showed us some time ago".

I'm not sure what Ebizur said or when but I'm pretty sure that when we discussed the Karafet 2009 paper, we were pretty much in agreement that C2 was there, in Wallacea and Melanesia, since the early colonization.

"... possibly introduced with Y-hap C2 and the Austronesians".

Your Austronesians are magical: they mutate haplogroups at convenience, depending where they live.

Seriously, I'm pretty sure that we discussed and consensuated that Austronesians must have got at some point a matrilocal organization or something of the like because, while the mtDNA does look Austronesian (B), most of the Y-DNA looks of Papuan/Melanesian origin (C2). Austronesians in other areas do not have C2, only in Wallacea, Melanesia and Polynesia, but not in Philippines, Taiwan, Micronesia, Sundaland or Madagascar.

C2 is an ancient Melanesian lineage, not something introduced by Austronesians.

terryt said...

"Cannot be".

Another example of your believing you know far more about a subject than do apparent experts in the field, simply because they disagree with your pre-existing belief? Isn't that the source of your argument with Dienekes?

"I'm not sure what Ebizur said or when but I'm pretty sure that when we discussed the Karafet 2009 paper, we were pretty much in agreement that C2 was there, in Wallacea and Melanesia, since the early colonization".

Ebizur was very definite that the distribution of C2 and C2a was very different. Quote, from what I printed off at the time:

"The frequency of C2*-M38 also shows a pattern opposite to that of its subclade, C2a-M208. C2*-M38 occurs most frequently in populations of Wallacea and Western New Guinea, whereas C2a-M208 occurs most frequently in populations of Remote Polynesia, but parts of the area around the midpoint between these two poles are practically devoid of both C2*-M38 and C2a-M208 ..."

He later wrote:

"Most extant C2*-M38(xC2a-M208) seems to be confined to eastern Indonesia, including the islands of Wallacea and the western half of New Guinea".

So it looks very much as though C2 was in Wallacea since the early colonization (in the form of C2), but not in Melanesia till later (in the form of C2a), and then very sparsely.

"C2 is an ancient Melanesian lineage, not something introduced by Austronesians".

Not Melanesian. In fact the dates Ebizur gives for the two clades fits the idea that C2 is ancient in Eastern Indonesia and Wallacea (35-45,000 years, or more) but the date for C2a fits exactly the Austronesian expansion (4-5000 years). So: C2a introduced by Austronesians.

"Your Austronesians are magical: they mutate haplogroups at convenience, depending where they live".

As does every other ethnic group. For example Jews, and Basques.

terryt said...

By the way, I've been meaning to bring your attention to some current news items.

I remember you claiming that fire could not have been a factor in the extermination of the Central Asian megafauna because it was impossible to set fire to the vegetation of the region. They're doing a pretty good job of it in Russia at the moment.

And as for your idea that Paleolithic humans lived on river flood-plains. A flood such as the current one in Kashmir and Pakistan would certainl,y have exterminated any Paleolithic populations foolish enough to live on such flood-plains.

I'm sure you'll claim 'global warming' as the explanation for these events, but surely climate has fluctuated extremely over our prehistory.

Maju said...

Authority is not evidence, Terry. Data, not opinions is what I feed on.

"that the distribution of C2 and C2a was very different".

C2a is just a subclade of C2 and there's nowhere where it could come but from New Guinea!

And the Polynesian-specific clade is anyhow not C2a but C2a1, a further level derivative and a founder effect.

C2a is found (review your copy of Karafet 2010 supplementary material) ONLY in some Polynesian islands (Vanuatu, Tahiti, Samoa) and in coastal Papua (the highest apportion: 2/15: 13%). It does not exist in Wallacea.

In Wallacea you do find aboundant C2(xC2a) but not a single case of C2a. C2(xC2a) is also found in highland Papua, what makes it almost necessary that the transition C2-C2a-C2a1 happened in New Guinea, where the three steps are found (or at least C2-C2a).

You have your brain so busy trying to find hints of "evidence" for your preconceptions that you totally miss the big picture. You seem not to learn for that reason.

"So: C2a introduced by Austronesians".

C2a1 was the clade introduced by Austronesians in Polynesia and outer Melanesia, after probably borrowing it from Papuans or otherwise people from New Guinea. Even if they took it from Wallaceans, that would still make the lineage pre-Austronesian because it's not found in the probable Austronesian homeland in Taiwan, Philippines and/or Borneo.

Apples and oranges, damnit!

"As does every other ethnic group. For example Jews, and Basques".

Not that I know. They only do that when they absorb massively other peoples, such as Yemenis (i.e. Yemeni Jews do not look at all like Euro-Mediterranean Jews in this aspect nor others, they look almost typical Yemenis). Otherwise, it's like arguing that L2 is English because it's found in Jamaicans.

Maju said...

"A flood such as the current one in Kashmir and Pakistan would certainl,y have exterminated any Paleolithic populations foolish enough to live on such flood-plains".

Oh, sure... like a tsunami would surely have exterminated the Onge... just that they did perfectly well: they are not domesticated farmers but genuine hunter-gatherer humans. You know: that's the difference between one of your cows and a genuine semi-wild cow like a betizu: the first die off in catastrophes, the second would survive without us perfectly well.

If the river goes high and you have a boat... you get on it and move out to higher ground. But that can only happen in hunter-gatherer conditions, farmers and urbanites, well actually no one, can't migrate freely anymore. That's a problem. Of course it can take a toll, but it's not such a big deal.

"I'm sure you'll claim 'global warming' as the explanation for these events, but surely climate has fluctuated extremely over our prehistory".

It's never been so hot, except maybe at some narrow period in the Neolithic. I mean, never since our species left Africa. It's been a lot colder though.

Ebizur said...

Maju, "C2a-M208" does not mean the same thing as "C2a*-M208." C2a1-P33 is a subclade of C2a-M208, and it may be subsumed in the category of C2a-M208 when contrasting C2a-M208 with C2*-M38. In fact, this is the only method of analysis that is possible when one is dealing with a data set in which C2a1-P33 has not been distinguished from its superset, C2a-M208, such as the data set of Kayser et al. 2008.

In any case, Terry is correct that currently available data do not indicate an ancient presence of C2a-M208 in Melanesia. In fact, C-M130 Y-DNA of any sort is practically absent from populations of the Solomon Islands archipelago according to all relevant data that I have seen; the only case reported so far is a single C2*-M38(xC2a-M208) individual sampled from the Teop, an Austronesian-speaking population of northern Bougainville Island.

North Bougainville & Buka total (Austronesian > Oceanic)
1/54 = 1.9% C2*-M38(xC2a-M208) [only in Teop]
4/54 = 7.4% F-M89(xG-M201, H-M69, I-M170, K-M9) [only in Saposa]
20/54 = 37.0% K-M9(xK3-P79, M1-M4, M3-P117, O-M175, P-M74, S-M230) [only in Saposa and Teop]
3/54 = 5.6% K3-P79 [only in Saposa and Teop]
1/54 = 1.9% M1b*-P87 [only in Saposa]
10/54 = 18.5% M1b1-P22
3/54 = 5.6% M3-P117 [only in Teop and Buka]
2/54 = 3.7% O-M175(xO1a-M119, O3-M122) [only in Teop and Buka]
8/54 = 14.8% O1a-M119
1/54 = 1.9% O3-M122 [only in Buka]
1/54 = 1.9% S-M230 [only in Buka]

Central Bougainville
Aita (Non-Austronesian)
15/18 = 83.3% M1b1-P22
3/18 = 16.7% K-M9(xK3-P79, M1-M4, M3-P117, O-M175, P-M74, S-M230)

South Bougainville
2/3 M1b1-P22
1/3 K-M9(xK3-P79, M1-M4, M3-P117, O-M175, P-M74, S-M230)

Western Province, Solomon Islands (Cox et al. 2006; linguistically approx. 83% Austronesian and approx. 17% non-Austronesian/Central Solomons Papuan as of 1999 census)
11/20 = 55.0% K-M9(xM1-M4/M106, O-M175, P-M74/92R7, S-M230)
9/20 = 45.0% O-M175

Malaita Province, Solomon Islands (Cox et al. 2006; linguistically approx. 100% Austronesian)
8/12 = 66.7% K-M9(xM1-M4/M106, O-M175, P-M74/92R7, S-M230)
1/12 = 8.3% S-M230
3/12 = 25.0% M1-M4/M106
(The authors have noted: "O-M175 was not identified in the sample from Malaita Province, where only Austronesian languages are currently spoken. Conversely, nearly half the men sampled from the Western Province are O-M175 carriers (45%), despite speaking a Papuan language. It is unclear whether this effect is an artefact of small sample size or an accurate representation of Y-chromosome diversity in Solomon Islands, but the finding again emphasizes that community studies can discern levels of variability that may be missed at higher scales of analysis.")

C2a-M208 may have originated in some indigenous population of New Guinea (perhaps the ancestors of the Dani/Lani group), but the major dispersal of this haplogroup through Remote Oceania certainly has been effected by Austronesian speakers within the last several millennia.

Maju said...

You may notice, Ebizur, that I said:

"the transition C2-C2a-C2a1 happened in New Guinea, where the three steps are found (or at least C2-C2a)".

Why? Because I am conscious that Karafet made no difference between C2a1 and C2a in general, so I cannot judge properly that.

Whatever the case C2 in general must have coalesced in Wallacea or East of it, so it's not something that Austronesians brought from their homeland around the South China Sea. We can't say that C2 in any variant is specifically Austronesian. In fact we can say exactly the opposite: that C2 was a clade that Austronesians picked up in their expansions. Whether they did in Wallacea, New Guinea or the Near Melanesian islands is less important.

What I do see is that the only region that has both phylogenetic levels (C2 and C2a) is New Guinea. Could be elsewhere around the big island? Maybe but does not make much of a difference.

"Terry is correct that currently available data do not indicate an ancient presence of C2a-M208 in Melanesia".

I have already said that for anthropological purposes, Wallacea is not substantially different from Melanesia. In any case the issue would be between Wallacea, New Guinea and near Island Melanesia (Solomon islands and such). It can never be attributed to the putative origin of Austronesians this side of Wallace Line.

Terry is not saying that it comes from Wallacea, he is saying that it was introduced by Austronesians. Read what he says again, please, before giving him the reason somewhat unwarrantedly.

"C2a-M208 may have originated in some indigenous population of New Guinea (perhaps the ancestors of the Dani/Lani group), but the major dispersal of this haplogroup through Remote Oceania certainly has been effected by Austronesian speakers within the last several millennia".

In this we agree. The issue is where did the lineage originate, not which language/culture was its vector into remote previously uninhabited islands.

My whole point is that C2 in general and C2a in particular is not an Austronesian-original nor even an Austronesian-specific lineage at all. C2a may be a Melanesian-Polynesian lineage to a large extent but that's it.

Maju said...

"Terry is correct that currently available data do not indicate an ancient presence of C2a-M208 in Melanesia".

Based on what? Sorry to go over this again but, if C2 is 45 Ka old (per Karafet's wave A timing) and there are two know downstream nodes to present, C2a can perfectly be 30 Ka old and C2a1 some 15 Ka old. Just divide 45/3. It is a very rough calculation but a valid one anyhow.

You could fine-tune a bit, I guess, using STR methods but I'm not aware anybody has used them in ways we can discern the whole process, right?

Maju said...

And if C2 is like 60 Ka old (per the Callao metatarsal - still very tentative), then it's all even older: C2a could be 40 Ka old and C2a1 even as much as 20 Ka old.

terryt said...

"C2a is just a subclade of C2 and there's nowhere where it could come but from New Guinea!"

Not so. More likely to have come from Eastern Indonesia. Ebizur mentioned that C2a in Melanesia diversified no more than 4-5000 years ago, so presumably it arrived a little before then, possibly in the form of C2 which it then largely replaced.

"In Wallacea you do find aboundant C2(xC2a) but not a single case of C2a".

Yes. That's what Ebizur said. Please read what I actually wrote.

"Terry is not saying that it comes from Wallacea, he is saying that it was introduced by Austronesians. Read what he says again, please, before giving him the reason somewhat unwarrantedly".

I assumed you would understand that I claimed it traveled east with Austronesians. Whether it formed part of any 'original' Austronesian population or not I have no way of knowing. However it seems obvious it joined the Austroneians before their major expansion.

"I have already said that for anthropological purposes, Wallacea is not substantially different from Melanesia".

I think Ebizur has shown they are substantially different. The people in the two regions even look quite different.

"C2a could be 40 Ka old and C2a1 even as much as 20 Ka old".

'Could be'. But that is hardly evidence that they are. And that from someone who earlier said, 'Data, not opinions is what I feed on'.

"Oh, sure... like a tsunami would surely have exterminated the Onge..."

Oh. My apologies. I didn't realise that the Onge lived on a flood plain.

"If the river goes high and you have a boat... you get on it and move out to higher ground".

Is that so? Then why haven't all those Pakistanis done that? After all even farmers can have a boat.

Maju said...

"Not so. More likely to have come from Eastern Indonesia".

Nobody has detected C2a so far in Wallacea. Eastern Indonesia is a confusing term, as it would include all West Papua.

"I assumed you would understand that I claimed it traveled east with Austronesians".

"East"? East where?! C2a westernmost known location is Papua-New Guinea! In spite of C2 being abundant in Wallacea, and of them being tested for C2a by Karafet's team, not a single case was found.

"However it seems obvious it joined the Austroneians before their major expansion".

Only before arriving to Polynesia and far Melanesia. Earlier we do not know (and IMO unlikely).

"'Could be'. But that is hardly evidence that they are".

Indeed. Just pointing to that possibility. You were giving for granted something that is also undemonstrated.

"I didn't realise that the Onge lived on a flood plain".

You don't seem to realize that floods don't commit genocide: they just cause trouble (and are yet another reason for people to need boats).

"Then why haven't all those Pakistanis done that? After all even farmers can have a boat".

Do they actually have one? Do you see them fishing in the waters? They are of course moving to high ground but they are not the masters of their lives anymore, nor there is probably much game left in that country.

Whatever the case, they will survive in most cases, as happened in the past. People are much more sturdy than you seem to believe.

I don't know if there are floods in New Zealand but here (and more towards the Mediterranean or the Danubian plain) they are common occurrence: they damage river basin infrastructures and maybe kill a few in the worse cases but that doesn't seem like anything it would wipe out a hunter-gatherer tribe. It certainly does not take a meaningful toll in terms of percentage of human lives.

The greatest risk may be lack of drinking quality water for a while in fact. We are used to see the images of the few who resist on top of their roofs but much more dangerous is water quality after the flood. Still people, overall, survives that too.

terryt said...

"'East'? East where?! C2a westernmost known location is Papua-New Guinea! In spite of C2 being abundant in Wallacea"

You don't seem to have yet grasped the fact that C2 and C2a are different haplogroups. Confusing them is like claiming Y-hap R1a is the same as MNOPS. I've said before, but I'll say it again, Ebizur gave the dates for Melanesian Y-hap C2a as around 4500 years ago. So it's most likely that C2a derives relatively recently from C2 in Eastern Indonesia, perhaps, or perhaps not, including Irian Jaya.

"Do they actually have one? Do you see them fishing in the waters?"

On what grounds do you claim that all Paleolithic people had boats then? A little consistency wouldn't go amiss.

"They are of course moving to high ground but they are not the masters of their lives anymore"

And such, surely, would have been the same in the Paleolithic. But they didn't have air drops in those days, so extinction would be very likely.

"they damage river basin infrastructures and maybe kill a few in the worse cases but that doesn't seem like anything it would wipe out a hunter-gatherer tribe".

Flood plains are exactly that. They have been formed by floods, often very large ones. Not just destructive of infrastructures.

"The greatest risk may be lack of drinking quality water for a while in fact".

Again, same in the Paleolithic.

"We are used to see the images of the few who resist on top of their roofs"

And how long would they survive there without outside help?

Maju said...

"You don't seem to have yet grasped the fact that C2 and C2a are different haplogroups".

They are not really different haplogroups, because C2a is a subhaplogroup of C2.

"Confusing them is like claiming Y-hap R1a is the same as MNOPS".

I am not confusing them. The only thing I am saying is that C2a is a derivate of C2, exactly as P (rather than R1a) is a derivate of MNOPS or MNOPS is a derivate of K.

Anyhow, it's irrelevant for the discussion, because my whole point is that there is no known C2a west of New Guinea, so I repeat: "east of where?"

"Ebizur gave the dates for Melanesian Y-hap C2a as around 4500 years ago".

I don't know if or why that would be true. Can you elaborate on WHY that claim would stand? It seems central to your argumentation, so please explain why do I have to accept that claim?

"On what grounds do you claim that all Paleolithic people had boats then?"

I don't claim anything, but I find most likely that people who could be living by the river, would have boats to exploit it properly.

Anyhow, one of the main issues with the current Pakistan floods and floods in general is that most settlements are built just by the river, in areas it naturally floods. IF hunter-gatherers would do that, I would imagine they can properly exploit the riverine environment and therefore had boats. Anyhow, most Paleolithic sites I know of are actually on higher ground, also in riverine environments. Of course this may be because floods erase the remnants in other places but it also implies that people used at least sometimes vantage high places to make camp and overlook their territory.

"And such, surely, would have been the same in the Paleolithic. But they didn't have air drops in those days, so extinction would be very likely".

Well, it is clear that you and me see this matter in very different ways. I think that they'd move to higher ground, if they were not there already and take advantage of the scared animals. Food should not be a problem, water might be though.

I don't think extinction can be caused by a mere flood. In my way of thinking it'd need at least a number of accumulating pressures, including belligerant pressure by other humans. Also we'd have to ponder the size of the group we are talking about because a 20 people band can be easily wiped out accidentally but it becomes much harder when we start thinking of a 500-1000 people (or larger) ethnic group.

If one group dies out, there's always another nearby that survives and can then take advantage of the abandoned niche to expand.

Obviously we are thinking in very different terms overall, what makes our conclusions radically different.

terryt said...

We have rather drifted away from the topic so I'll keep it short.

"I don't know if or why that would be true. Can you elaborate on WHY that claim would stand?"

I'll leave Ebizur to comment on that. It's from the information he posted way back, which I printed off. I'm fairly sure we three agreed at the time, although perhaps it was just me and him who agreed.

"Of course this may be because floods erase the remnants in other places but it also implies that people used at least sometimes vantage high places to make camp and overlook their territory".

Paleolithic hunter-gatherers 'mainly' lived on higher places overlooking their territory is the most likely explanation. It is most likely that exploitation of riverine environments with boats is a late Paleolithic development, although I concede it probably dates back at least 50,000 years in the East.

"I don't think extinction can be caused by a mere flood".

Quite possible in the Paleolithic, especially because, as you suggest, we're probably talking about groups of 20 people in a band who 'can be easily wiped out accidentally'.

"If one group dies out, there's always another nearby that survives and can then take advantage of the abandoned niche to expand".

Exactly. Population replacement.

Maju said...

"I'll leave Ebizur to comment on that".

I read that as saying: "I have no damn idea, Ebizur save me!"

"... although perhaps it was just me and him who agreed".

Perhaps, indeed. I'm pretty sure I did mention back then as well, my "brute logic" (but effective enough) of the clades potentially being much older than your claims.

"... because, as you suggest, we're probably talking about groups of 20 people in a band who 'can be easily wiped out accidentally'".

I'm not suggesting that. I am in fact suggesting the opposite: larger ethnic groups ("tribes") divided in many of such 20-30 people economic units, with people changing band often, as it is known to happen among the Hadza, a close relative of the pre-OOA population that are still hunter-gathering. They change bands for all kind of reasons but specially because of disagreements and arguments. The ethnic group is large enough to allow for such a continuous and eventually circular flow.

These larger ethnic groups of many hundred people, distributed in many small operative units is in fact our "genetic unit" rather than the fluctuating band. Of course the band must not be ignored either because I can imagine that often migrations were initiated by just a band or two, eventually becoming a larger ethnos if successful.

"Exactly. Population replacement".

No. Because the neighbor group are their "cousins", they are the same population (in most cases at least) or (in other cases maybe) a distinct but still related and convergent population by means of growing interpersonal ties because of the simple cause of neighborhood.

Ebizur said...

Haplogroup C2a1-P33 probably has sprung from its C2a-M208 root not long before the settlement of westernmost Polynesia according to Murray P. Cox, Alan J. Redd, Tatiana M. Karafet et al., "A Polynesian Motif on the Y Chromosome: Population Structure in Remote Oceania," Human Biology, October 2007, v. 79, no. 5, pp. 525–535:

"Given its Polynesian specificity, the P33 mutation probably arose just before, or during, the region’s initial settlement. Archeological remains of a founding Polynesian settlement at Nukuleke, Tonga, were radiocarbon-dated to about 2,900 years b.p. cal. (Burley and Dickinson 2001). Although the Tongan and Samoan archipelagos of Western Polynesia were colonized rapidly, there was a hiatus before settlement farther eastward (Kirch 2000). The age of the P33 mutation should match this timeframe within the current limits of genetic dating. We inferred the TMRCA of extant P33 lineages using variation in 23 Y-chromosome STRs under a coalescent-based geometric-geometric mutation model (Watkins 2007). The P33 transition likely arose within a broad temporal window from 7,500 to 1,500 years b.p. (95% confidence interval; mean 4,500 years b.p.). Although encompassing a large chronological spectrum, this interval is consistent with radiocarbon estimates for the earliest Polynesian settlements."

This team's most likely estimate of the TMRCA of C2a1-P33 (4,500 years) must be the source of the number that Terry has been quoting.

However, the same team also has published an estimate of the TMRCA of C2a-M208 as a whole:

"The spatial and temporal distributions of ancestral haplogroups also enlighten the early history of ancestral Polynesians. A relative paucity of C-M208*(xP33) suggests that the marker may have developed not long before P33. Expanding Austronesian-speaking populations probably assimilated M208 individuals (Green 1991, 2003) as they moved eastward along New Guinea’s northern coast about 3,500 years b.p. (Spriggs 2003). Descendants of these admixed communities later swept Melanesian markers, such as the lineages considered here, to high frequency in the remote islands of Polynesia (Cox 2006a; Cox and Lahr 2006). Consequently, M208 must have arisen before (Kayser et al. 2003) or during this population expansion. Using variation in 12 Y-chromosome STRs in M208-derived individuals, we infer an origin for the M208 mutation between 12,300 and 3,900 years b.p. (95% confidence interval; mean 8,100 years b.p.), consistent with earlier Bayesian estimates (95% confidence interval 19,700 to 2,800 years b.p.; mean 6,900 years b.p.; Kayser et al. 2003)."

As for C2*-M38(xC2a-M208),

"The distribution of C-M38* (xM208, P33) is centered on eastern Indonesia and Melanesia (Underhill et al. 2001) and has been equated with the C-RPS4Y/DYS390.3del variant in earlier studies (Kayser et al. 2003). Bayesian dates for this mutation yielded upper confidence limits with considerable antiquity: 30,300 to 4,500 years b.p. (95% confidence interval; mean 10,600 years b.p.) (Kayser et al. 2003). Despite a reduced chronological range, we also infer a late Pleistocene ancestor for M38-derived lineages, between 37,200 and 28,800 years b.p. (95% confidence interval; mean 33,000 years b.p.). Coupled with this marker’s spatial distribution, there seems little doubt that the ultimate paternal ancestors of Polynesian P33 carriers once lived in Melanesia or its immediate environs (Kayser et al. 2003)."

Ebizur said...

The following is their description of their Y-STR testing and TMRCA estimation method:

"C-M38*(xM208, P33) and C-M208*(xP33) individuals were screened with a suite of 12 Y-chromosome short tandem repeats (STRs): DYS385a, DYS385b, DYS388, DYS389I, DYS389II, DYS390, DYS391, DYS392, DYS393, DYS426, DYS438, and DYS457. C-P33 individuals were screened with 11 additional Y-chromosome STRs: DYS442, DYS446, DYS447, DYS453, DYS454, DYS455, DYS456, DYS458, DYS570, DYS576, and DYS607. The entire set of 23 Y-chromosome STRs was used for analyses of P33-derived individuals. ... TMRCA (time to the most recent common ancestor) dates were inferred using Y-chromosome STR data under a geometric-geometric mutation model (Watkins 2007). Analysis was conditioned on three parameters: α (= 0.1), which determines the geometric distribution of Y-chromosome STR step changes; β (= 1.05), which sets the likelihood of mutation to a geometric dependence on microsatellite length; and k (= 13), which represents the mean Y-chromosome STR repeat length in the data set. Confidence intervals (95%) were applied as 1/n^(1/2), where n is the number of Y-chromosome STR loci. Unscaled coalescent time units were converted to chronological dates using a mean male intergeneration interval of 31 years (Fenner 2005) and a mean Y-chromosome STR mutation rate (averaged across multiple studies) of 2.1 × 10^–3 per generation (Gusmão et al. 2005)."

Terry does seem to have mistaken their estimate of the TMRCA of C2a1-P33 for an estimate of the TMRCA of C2a-M208, but otherwise both Maju and Terry appear to be saying much the same thing. Also quoting Cox et al. 2007,

"We traced individual histories of C-P33 lineages using Y-chromosome STR variation to tease apart recent Polynesian history. Clear spatial structuring is observable in the network (Figure 2). From a broad geographic perspective the most closely related C-M38*(xM208, P33) individuals were found on the islands of Flores and Sulawesi in eastern Indonesia. The next most closely related men, C-M208*(xP33) carriers, inhabit coastal Papua New Guinea. The most basal P33 lineage was found in Samoa, but it may represent a generic Western Polynesian founding lineage given the limited number of Tongans in our sample (n = 12). For similar reasons, related P33 lineages may some day be identified in Fiji (cf. Figure 2; Kayser et al. 2006), a likely staging ground for the settlement of Polynesia. Regardless, this overall pattern fits a movement of ancestral Polynesians trending generally eastward from (or through) Island Southeast Asia, along coastal New Guinea, and out into the wider Pacific (Hurles et al. 2002; Kayser et al. 2003). We predict that similar evolutionary patterns may be observed in other Y-chromosome lineages carried by Polynesians, such as O-M122."

In other words, the Y-STR profiles of members of C2*-M38(xC2a-M208) are quite diverse, but they are geographically concentrated in Wallacea, New Guinea, and a few islands of Melanesia proper (i.e. mainly or exclusively Austronesian-speaking islands of the Southwest Pacific that are neither Micronesian nor Polynesian), with a few stray individuals in non-Wallacean parts of Indonesia (1/40 Southern Borneo in Kayser et al. 2003, 1/74 Mentawai in Cox et al. 2007). In Melanesia, Maewo Island of Vanuatu has been found to have a fairly high frequency of C2*-M38 (9/44 = 20.5% according to Cox et al. 2007), and this paragroup also has been found with notable frequency in Fiji (9/105 = 8.6% according to Kayser et al. 2008), but it seems to occur in only perhaps 1% (1/107) of the male population of the Solomon Island archipelago according to data from Cox et al. 2006 and Scheinfeldt et al. 2006. C2*-M38 also has been found in approximately 1% of males from Polynesia proper (3/332 total, or 1/29 Tonga, 1/61 Samoa, and 1/77 Cook Islands according to Kayser et al. 2008).

Ebizur said...

C2a-M208 seems to occur sporadically among both Austronesian speakers and non-Austronesian speakers in New Guinea, and it has been found in about 10% to 15% of sampled Austronesian-speaking males from a few Melanesian "satellites" around the eastern end of New Guinea (e.g. the Admiralty Islands, which are located off the northeastern coast of New Guinea, and the Trobriand Islands, which are located off the southeastern coast of New Guinea), but it has been found with high frequency in Near Oceania only among members of the Dani and Lani of the highlands of Western New Guinea (Kayser et al. 2003) and in two of a sample of five Austronesian-speaking Irarutu males from the Bomberai Peninsula of Western New Guinea (Mona et al. 2007). C2a-M208 has not been observed any further west than the West Papuan-speaking Karon of the Bird's Head Peninsula as far as I know (Mona et al. 2007 have reported finding C2a-M208 in 1/22 = 4.5% of a sample of Karon males). C2a-M208 has been found only rarely in the major islands of the Bismarck Archipelago (e.g. New Britain, New Ireland) and in Maewo Island of Vanuatu (1/44 C2a-M208(xC2a1-P33) according to Cox et al. 2007), and, as I have mentioned before, it has not been found at all in any of the Solomons. C2a-M208 reappears with moderate frequency (approx. 10% to 20%) in the westernmost parts of the Fijian-Polynesian-speaking realm, such as Fiji and Tuvalu. As one looks further out toward the more typically Polynesian areas of the Pacific, C2a-M208 becomes the predominant Y-DNA haplogroup in the form of its subclade, C2a1-P33. In order to discover the proximal origin of the majority of Polynesian Y-DNA, I think it is important for someone to determine how many of the cases of C2a-M208 that Kayser et al. 2008 have reported finding in 14/105 = 13.3% of their Fijian sample also exhibit the P33 mutation.

Maju said...

Let's go back on the discussion because I'm beginning to go nuts on whether we are debating the presence of C2 in Melanesia, that of C2a or the age of C2a1 (Polynesian clade).

Terry said: "Y-DNA C2 moved in later, as Ebizur showed us some time ago".

Terry said: "Early mtDNA is presumably M-derived although B has come to dominate most of the region, possibly introduced with Y-hap C2 and the Austronesians".

This was the origin of the discussion.

What I say is that C2 is not "Austronesian" (South China Sea) but coalesced originally East/South of Wallace Line (i.e. not in Eurasia proper but in Wallacea or beyond). And that, considering its relatively strong presence in highland New Guinea, we are before a case of likely ancient introduction in Papua and Near Island Melanesia (Solomon and such). C2 should have coalesced among the first human arrivals to the scattered region beyond Wallace Line.

I said: ""I'm not sure what Ebizur said or when but I'm pretty sure that when we discussed the Karafet 2009 paper, we were pretty much in agreement that C2 was there, in Wallacea and Melanesia, since the early colonization".

Terry replied: "Ebizur was very definite that the distribution of C2 and C2a was very different".

I said: "C2a is just a subclade of C2 and there's nowhere where it could come but from New Guinea! ... And the Polynesian-specific clade is anyhow not C2a but C2a1, a further level derivative and a founder effect".

Terry insisted: "So: C2a introduced by Austronesians".

In New Guinea and Melanesia, from context. Though it's difficult to follow Terry's contextualization sometimes.

(continues)

Maju said...

Following with the discussion review:

Ebizur then jumped to the discussion wit the following: "In any case, Terry is correct that currently available data do not indicate an ancient presence of C2a-M208 in Melanesia".

Uh? Then it must have coalesced among mermen because no C2a is known west of Papua.

Then he contradicted the previous sentence: "C2a-M208 may have originated in some indigenous population of New Guinea (perhaps the ancestors of the Dani/Lani group), but the major dispersal of this haplogroup through Remote Oceania certainly has been effected by Austronesian speakers within the last several millennia".

I don't care about remote Oceania (because after all the discussion on Paleolithic boating is about near Oceania, not Tahiti), so guess that in the end Ebizur and I might be in agreement.

Or not? Clarification needed.

The sloppiness of geographic terms such as "Melanesia" that for some mean just Island Melanesia east of New Guinea, for others every single island from Wallace Line to Kanaky (New Caledonia) makes the situation, and most normally includes both New Guinea and Island Melanesia (but excluding Wallacea) makes the discussion confusing.

Also jumping from C2 to C2a and then to C2a1 is like an elephant trap: first is wide, then narrower and then not only most narrow but also has little to do with the beginning of the discussion at all. And I am the elephant, of course.

So, guys, go to the grain:

1. where do you think that C2 coalesced? (my opinion is that in Wallacea anywhere from 60 to 40 Ka ago)

2. where do you think that C2a (and not C2a1!) coalesced? (my opinion is that in New Guinea or not too far from there, some 40 to 20 Ka ago)

So, when Ebizur claimed that "Terry is correct that currently available data do not indicate an ancient presence of C2a-M208 in Melanesia", I replied "based on what?" Question that got no answer.

Can you please be more direct, Ebizur? Conveniently ignoring the crux of the matter and surrounding the discussion with a net of pretty much irrelevant data is not what I call a direct, honest answer.

So based on what? And by "Melanesia" you mean exactly what?

Maju said...

Digging in your quotes, maybe you have tried to answer in this last batch. For instance:

"A relative paucity of C-M208*(xP33) suggests that the marker may have developed not long before P33".

Well, 15% in some communities, which don't even seem to be Austronesian-speakers, is not "paucity", as per wikitionary's explanation (smallness, meagerness).

Can these people accept that the migrating Austronesian wave through northern New Guinea (probably a very small group) was affected by a meaningful founder effect by taking in Papuan (or otherwise pre-Austronesian Melanesian) males (or at least their offspring)? That's the only logical explanation!

"Using variation in 12 Y-chromosome STRs in M208-derived individuals, we infer an origin for the M208 mutation between 12,300 and 3,900 years b.p. (95% confidence interval; mean 8,100 years b.p.), consistent with earlier Bayesian estimates (95% confidence interval 19,700 to 2,800 years b.p.; mean 6,900 years b.p.; Kayser et al. 2003)."

Let's assume for a minute that these MC estimates are correct. It would mean that C2a might have coalesced (mean) 8-7000 years ago, what is 4-5000 years earlier than Lapita culture and in general the Austronesian expansion out beyond Philippines.

So it has nothing to do with the Austronesian expansion. So it must have been in Melanesia (the only place where it does exist besides far Oceania) before Austronesians even began expanding at all.

Of course, you can push the matter to the lowest estimates (4-3,000 years ago) but that's pushing things a lot. And to my understanding that usual MC estimation methods brutally push timelines towards the present it looks even more far fetched. I'd rather tend to favor the largest bounds instead of 20-12 Ka ago (if anything).

So C2a is pre-Austronesian with all likelihood, even by the usual MC estimates.

"Terry does seem to have mistaken their estimate of the TMRCA of C2a1-P33 for an estimate of the TMRCA of C2a-M208, but otherwise both Maju and Terry appear to be saying much the same thing".

I would not be surprised, really. Sometimes is like bumping heads for the sake of it. Add a couple of arbitrary usage of words like "Wallacea" and "Austronesian" and the scenario is set for a long, maybe pointless, argument.

But I think that the key matter here is whether we can say that C2 and C2a are of Austronesian origin and dispersal, not in Far Oceania but in Near Oceania and Wallacea (i.e. areas settled already in the Paleolithic), which is what matters for our original discussion.

And my conclusion is again that nope. No way.

"C2a-M208 has not been observed any further west than the West Papuan-speaking Karon of the Bird's Head Peninsula as far as I know (Mona et al. 2007 have reported finding C2a-M208 in 1/22 = 4.5% of a sample of Karon males). C2a-M208 has been found only rarely in the major islands of the Bismarck Archipelago (e.g. New Britain, New Ireland) (...), and, as I have mentioned before, it has not been found at all in any of the Solomons".

Exactly my point: C2a coalesced most probably in New Guinea before Austronesian expansion, from either local Papuan C2 or maybe pre-Austronesian Wallacean C2 (this last would require archaeological support of some sort).

Cheers.

terryt said...

"I read that as saying: 'I have no damn idea, Ebizur save me!'"

No. It means that I realise Ebizur knows far more about the subject than either you or I do.

"The sloppiness of geographic terms such as 'Melanesia' that for some mean just Island Melanesia east of New Guinea, for others every single island from Wallace Line to Kanaky (New Caledonia) makes the situation, and most normally includes both New Guinea and Island Melanesia (but excluding Wallacea) makes the discussion confusing.

'Melanesia' is actually the islands east of New Guinea out to Fiji, or the beginning of Polynesia. Fiji is borderline. And sometimes New Guinea is included, but even within island Melanesia there are two separate regions; An earlier Papuan-speaking region (Northern Solomon Islands) and a region inhabited later by Austronesian-speaking people (the rest).

"Haplogroup C2a1-P33 probably has sprung from its C2a-M208 root not long before the settlement of westernmost Polynesia according to Murray P. Cox, Alan J. Redd, Tatiana M. Karafet et al."

Ebizur. Merely quoting experts is not going to sway Maju in the slightest. He already has his own beliefs and they're not going to change. But your comment means that C2a-M208 had made it to Eastern Melanesia a little before 2,900 years b.p. (given the apparently rapid Austronesian progress from the Bismark Archipeligo).

"What I say is that C2 is not 'Austronesian' (South China Sea) but coalesced originally East/South of Wallace Line (i.e. not in Eurasia proper but in Wallacea or beyond)".

I don't think anyone claims a 'South China Sea' origin for the Austronesians. There is debate as to whether it was Taiwan or the Philippines, or a combination, including points south of those islands. That leaves open the possibility of an actual Wallacean origin for the Austronesians.

"And that, considering its relatively strong presence in highland New Guinea, we are before a case of likely ancient introduction in Papua and Near Island Melanesia (Solomon and such). C2 should have coalesced among the first human arrivals to the scattered region beyond Wallace Line".

Where do you get the 'relatively strong presence in highland New Guinea' from? As far as I know it's not very common at all there. It's mostly south Wallacean. So rather than 'beyond Wallace Line' it coalesced in wallacea. As Ebizur wrote, 'In other words, the Y-STR profiles of members of C2*-M38(xC2a-M208) are quite diverse, but they are geographically concentrated in Wallacea, New Guinea, and a few islands of Melanesia proper (i.e. mainly or exclusively Austronesian-speaking islands of the Southwest Pacific that are neither Micronesian nor Polynesian)'. In other words it's prtesence in Melanesia is strong mainly in those regions settled by Austronesian-speaking people, not Papuan-speaking. So it expanded into wider Melanesia with the Austronesians.

"Exactly my point: C2a coalesced most probably in New Guinea before Austronesian expansion, from either local Papuan C2 or maybe pre-Austronesian Wallacean C2"

Possibly. But as you quote Ebizur as saying: 'C2a-M208 has not been observed any further west than the West Papuan-speaking Karon of the Bird's Head Peninsula ... C2a-M208 has been found only rarely in the major islands of the Bismarck Archipelago (e.g. New Britain, New Ireland) (...), and, as I have mentioned before, it has not been found at all in any of the Solomons'. So even on the Bird's head it seems very likely that itwas introduced by Austronesians who later adopted their neighbours' language.

Maju said...

"'Melanesia' is actually the islands east of New Guinea out to Fiji"

Well, not. Sometimes the term is used that way but, as you admit, New Guinea is also included very very often. And yet in genetics and other disciplines such as linguistics it may include Wallacea or parts of it.

In fact when the word Melanesia(n) appear in pop. genetic contexts almost invariably means Papuans or Papuans and some other nearby group(s).

So I'd suggest you choose a better term for that such as "island Melanesia" or whatever.

"I don't think anyone claims a 'South China Sea' origin for the Austronesians. There is debate as to whether it was Taiwan or the Philippines, or a combination, including points south of those islands".

That's what I mean essentially.

However some non-Westerners seem to favor instead a West Indonesian origin of some sort, at least that's what I find on occasion when I read random articles on the matter. And yet some other times links with mainland South China and Vietnam have been suggested.

So South China Sea area is the quasi-perfect description for all that array of suspects.

"Where do you get the 'relatively strong presence in highland New Guinea' from?"

From Karafet 2010. Table S2:

Papua-NG highlands: C2*: 2/33 (6%). No other C clade reported.
Papua-NG coastal: C* 1/15 (7%), C2a(xC2a1) 2/15 (13%). No other C clade reported.

"As far as I know it's not very common at all there. It's mostly south Wallacean".

Also, more frequent in fact. In Wallacea (Eastern Indonesia) in the paper, it's 313/957 (33%), only C* reported besides that in the C haplogroup.

The key to my reasoning is that both C2 and C2a are only found side by side East of Wallacea, in Papua or the islands. But sure: if you wish to push their ages to the last few years and imagine that Wallacea was influenced by Austronesian expansions before Melanesia (New Guinea incl.), then you can make it somehow be "Austronesian".

But it's totally force-fed.

"So rather than 'beyond Wallace Line' it coalesced in wallacea".

Which is beyond Wallace Line from the viewpoint of most. What is what 'coalesced in Wallacea'? I forgot. If you mean C2, I can agree in principle but if you mean C2a it seems a baseless claim.

None of them seem particularly "Austronesian" to me in any case.

"But as you quote Ebizur as saying: 'C2a-M208 has not been observed any further west than the West Papuan-speaking Karon of the Bird's Head Peninsula ... C2a-M208 has been found only rarely in the major islands of the Bismarck Archipelago (e.g. New Britain, New Ireland) (...), and, as I have mentioned before, it has not been found at all in any of the Solomons'. So even on the Bird's head it seems very likely that itwas introduced by Austronesians who later adopted their neighbours' language".

Don't you realize how contradictory is this claim? Isn't it much more logical that it was borrowed from Papuans or other Melanesians by passing founder groups of Austronesians of the Lapita culture?

It is not found West of Papua and its immediate ancestor is not found this size of Wallace Line, where Austronesians originated without almost doubt. So it really beats me how can you make an Austronesian origin claim, really.

En fin.

terryt said...

"Papua-NG highlands: C2*: 2/33 (6%). No other C clade reported.
Papua-NG coastal: C* 1/15 (7%), C2a(xC2a1) 2/15 (13%). No other C clade reported".

That's hardly a 'relatively strong presence in highland New Guinea'. In fact it's presence is generally stronger in coastal regions. And we find no separate New Guinea C haplogroup such as C6. So it's extremely unlikely that these highland Y-haps C2 and C2a were amoung the first people to enter New Guinea. They are almost certainly more recent immigrants (less than 10,000 years at least).

In the pages I copied way back Ebizur wrote:

"Taken together, these three TMRCA estimates for C2*-M38 suggest that it is most diverse in the Lesser Sunda Islands and the Maluku Islands of Wallacea in eastern Indonesia".

He quotes at least 30%, or more, through much of that region. So C2* almost certainly coalesced there, not in New Guinea. We're left with Y-hap C2 in Wallacea and Y-hap C4 in Australia, but no separate early Y-hap C in New Guinea. So other haplogroups were first into New Guinea. Ebizur also wrote:

"It appears that both C2*-M38 and C2a-M208 are quite rare in Island Melanesia (especially the Solomon Islands, including Bougainville), but on the rare occasions when haplogroup C does occur in Island Melanesia, it is about equally likely (and in Vanuatu, perhaps even more likely) to belong to C2*-M38 as it is to belong to C2a-M208".

He quotes 17.5% for C2* in Vanuatu, and 2 or 3% for C2a. Now. We know that Vanuatu was first settled (about 4000 years ago) by Lapita pottery-using people. So the C2a mutation may not be much older than that. That fits the dates Ebizur gave way back for C2a. It's universally acknowledged that the Lapita people are associated with the eastward expansion of the Austronesians. So. The Austronesian language was carried east beyond the Northern Solomons by Austronesian-speaking people, just as I said.

"Can these people accept that the migrating Austronesian wave through northern New Guinea (probably a very small group) was affected by a meaningful founder effect by taking in Papuan (or otherwise pre-Austronesian Melanesian) males (or at least their offspring)? That's the only logical explanation!"

To some extent that is the accepted explanation. However New Guinea itself is not particularly likely to have been the source of Y-hap C. Any Y-hap C there is more likely to have entered New Guinea with the Austronesian movement along the north coast. The Lapita people spread from the Bismark Archipeligo. Now, that obviously doesn't mean that the Austronesian expansion as a whole started there, but it is quite likely that Y-hap C2a developed there, from C2* as C2 moved rapidly from Wallacea, skirting the northern coast on New Guinea, and then out into the Pacific. Accompanied by some Y-hap O certainly. The reason why (as I quote above) that 'both C2*-M38 and C2a-M208 are quite rare in Island Melanesia (especially the Solomon Islands, including Bougainville)' is that it's again universally acknowledged that after the Lapita people had moved into the wider Pacific groups of more 'Melanesian-looking' people followed along behind, bringing Y-haps K and M.

terryt said...

"Papua-NG highlands: C2*: 2/33 (6%). No other C clade reported.
Papua-NG coastal: C* 1/15 (7%), C2a(xC2a1) 2/15 (13%). No other C clade reported".

That's hardly a 'relatively strong presence in highland New Guinea'. In fact it's presence is generally stronger in coastal regions. And we find no separate New Guinea C haplogroup such as C6. So it's extremely unlikely that these highland Y-haps C2 and C2a were amoung the first people to enter New Guinea. They are almost certainly more recent immigrants (less than 10,000 years at least).

In the pages I copied way back Ebizur wrote:

"Taken together, these three TMRCA estimates for C2*-M38 suggest that it is most diverse in the Lesser Sunda Islands and the Maluku Islands of Wallacea in eastern Indonesia".

He quotes at least 30%, or more, through much of that region. So C2* almost certainly coalesced there, not in New Guinea. We're left with Y-hap C2 in Wallacea and Y-hap C4 in Australia, but no separate early Y-hap C in New Guinea. So other haplogroups were first into New Guinea. Ebizur also wrote:

"It appears that both C2*-M38 and C2a-M208 are quite rare in Island Melanesia (especially the Solomon Islands, including Bougainville), but on the rare occasions when haplogroup C does occur in Island Melanesia, it is about equally likely (and in Vanuatu, perhaps even more likely) to belong to C2*-M38 as it is to belong to C2a-M208".

He quotes 17.5% for C2* in Vanuatu, and 2 or 3% for C2a. Now. We know that Vanuatu was first settled (about 4000 years ago) by Lapita pottery-using people. So the C2a mutation may not be much older than that. That fits the dates Ebizur gave way back for C2a. It's universally acknowledged that the Lapita people are associated with the eastward expansion of the Austronesians. So. The Austronesian language was carried east beyond the Northern Solomons by Austronesian-speaking people, just as I said.

"Can these people accept that the migrating Austronesian wave through northern New Guinea (probably a very small group) was affected by a meaningful founder effect by taking in Papuan (or otherwise pre-Austronesian Melanesian) males (or at least their offspring)? That's the only logical explanation!"

To some extent that is the accepted explanation. However New Guinea itself is not particularly likely to have been the source of Y-hap C. Any Y-hap C there is more likely to have entered New Guinea with the Austronesian movement along the north coast. The Lapita people spread from the Bismark Archipeligo. Now, that obviously doesn't mean that the Austronesian expansion as a whole started there, but it is quite likely that Y-hap C2a developed there, from C2* as C2 moved rapidly from Wallacea, skirting the northern coast on New Guinea, and then out into the Pacific. Accompanied by some Y-hap O certainly. The reason why (as I quote above) that 'both C2*-M38 and C2a-M208 are quite rare in Island Melanesia (especially the Solomon Islands, including Bougainville)' is that it's again universally acknowledged that after the Lapita people had moved into the wider Pacific groups of more 'Melanesian-looking' people followed along behind, bringing Y-haps K and M.

Maju said...

Ok, we are discussing in circles. I don't agree with every detail but I doubt it's really important.

terryt said...

"I don't agree with every detail but I doubt it's really important".

OK. But unless you're prepared to claim humans have been in Vanuatu for something like 50,000 years, (rather than just 5000 years at most), you're going to have a great deal of trouble arguing convincingly that Y-hap C2 coalesced in New Guinea. C2 and C2a are much more likely to have entered that island together at the time of the Austronesian expansion. Although Ebizur claimed a date of about 10,000 years for Melanesian C2-M38(xC2a-M208) he mentions, as a possibile explanation for 'rather long TMRCA' for Wallacean C2*, that it could be 'a paraphyletic assemblage of subclades that have expanded recently from immigrating founders with widely divergent STR haplotypes'. To me this is much more likely to explain the 10,000 year dating for northern (and Highland) New Guinea C2* and C2a.

Maju said...

"... unless you're prepared to claim humans have been in Vanuatu for something like 50,000 years, (rather than just 5000 years at most), you're going to have a great deal of trouble arguing convincingly that Y-hap C2 coalesced in New Guinea".

I don't see how this makes any sense.

Anyhow I don't mean to persuade you of anything. I am ok with having a reasoning that passes my own filters and not your pointless exaggerated demands. New Guinea and Solomon Is. were colonized in the Paleolithic, long before Austronesians existed as such. Austronesians did not originate at any area with Y-DNA C2. They have nothing to do with each other except in scattered Oceania but that's mere "recent history".

terryt said...

"I don't see how this makes any sense".

Humans obviously first arrived in Vanuatu speaking an Austronesian language. They also obviously carried Y-hap C2. They also obviously arrived there no more than 4-5000 years ago. The same haplogroup is spread right from southern Wallacea, across Northern New Guinea and down into Vanuatu. So C2 had not been in New Guinea long enough to form a separate version there before it reached Vanuatu.

"New Guinea and Solomon Is. were colonized in the Paleolithic, long before Austronesians existed as such".

But obviously by no members of Y-hap C. Colonized by Y-hap K.

"Austronesians did not originate at any area with Y-DNA C2".

Not exactly true. There is continuing argument as to where the Austronesians actuall originated. However it's almost a certainty that they didn't originate as a single small group in a single small region. Austronesians developed somewhere around Taiwan, the Philippines and points further south. Almost certainly from a complex series of influences. So the place of origin includes the region of southern Wallacea, where Y-hap C was certainly present from as early as 50,000 years ago. And a region where various Austronesian languages are spoken today.

Maju said...

C2 is also found in Highland New Guinea. I'd certainly ignore deep sea islands from all this discussion and focus only on what could be (and actually was) populated in Paleolithic times.

"There is continuing argument as to where the Austronesians actuall originated".

Which is your theory? Because different models should be interpreted in different ways. If you want to make Austronesians to originate in Wallacea, for instance, then this would have deep implications for haplogroups traveling or not to more northernly and westernly areas of Austronesian language/culture and hence for the overall model of Austronesian expansion.

You cannot have all the advantages of all theories and none of their setbacks. You need to consider each one separately.

"Almost certainly from a complex series of influences".

A most vague claim, very appropriate to appropriate all the "advantages" and get rid of the "setbacks". But has a setback however: it denies genetic personality to the Austronesian core, essentially saying that Austronesian-specific genetics do not exist.

It also needs much development, including archaeological and linguistic elaboration.

terryt said...

"C2 is also found in Highland New Guinea".

Yes. And is almost certainly a relatively recent arrival there. If you're looking for C2 in New Guinea try the Bird's Head. Reasonably common there, but that region is geographically closely connected to Wallacea, so it presumably arrived there from wallacea at the same time as it began expanding more widely. C2 is Wallacean, not New Guinean.

"focus only on what could be (and actually was) populated in Paleolithic times".

Of course New Guinea was 'populated in Paleolithic times', but obviously not by any Y-hap C individuals. When a C6 was announced I was more than happy to accept Paleolithic C in New Guinea. But it now seems obvious that the C in New Guinea is C2, the Wallacean variety. If C was Paleolithic in New Guinea it would have become as distinctive as C4 in Australia, C1 in Japan, C3 in Northeast Asia and C5 in the hill country of Pakistan, Kashmir, Nepal and Northern India. It's not, so it's not of ancient ancestry in New Guinea.

"If you want to make Austronesians to originate in Wallacea, for instance, then this would have deep implications for haplogroups traveling or not to more northernly and westernly areas of Austronesian language/culture and hence for the overall model of Austronesian expansion".

It's extremely unlikely that the Austronesians originated on a single island. Far more likely is the concept they developed from the complex interplay of a series of populations. This is supported by the fact that no single haplogroup defines them. They were a complex of different populations from early in their history.

"But has a setback however: it denies genetic personality to the Austronesian core, essentially saying that Austronesian-specific genetics do not exist".

Thank you. That is true, so you can finally see it. No Austronesian-specific genetics exist, especially in the matter of haplogroups.

"It also needs much development, including archaeological and linguistic elaboration".

Archeologists and linguists probably know more about the origin and spread of Austronesians than they do about any other population. And their basic ideas have not been substantially altered for about 30 years. The eatward-moving group ultimately left from the Admiralty Islands (or somewhere very nearby) and the westward-moving group left from somewhere in Borneo. Different haplogroups are involved in the two movements. So the Austronesians had become very mixed by the time they expanded.

Maju said...

"No Austronesian-specific genetics exist".

So no genetic component can be said to be "Austronesian", right?

"Archeologists and linguists probably know more about the origin and spread of Austronesians than they do about any other population. And their basic ideas have not been substantially altered for about 30 years. The eatward-moving group ultimately left from the Admiralty Islands (or somewhere very nearby) and the westward-moving group left from somewhere in Borneo".

So the Eastern group was essentially Melanesian and not properly from the South China Sea area. Fair enough.

How do you or "archaeologists" connect the Admiralty Islands to Borneo? It may be obvious from the linguistic viewpoint but it's not so obvious from other viewpoints, including genetic and archaeological ones.

Otherwise you may want to begin speaking of Polynesians and Malays as different populations and not anymore with that ultra-vague, complacent and misleading "Austronesian" tag.

Maju said...

Even if we are way off topic, let's recapitulate on Austronesians.

Following this composite tree of Austronesian languages (the various ones used to construct it can be found here), it becomes very apparent that

1. Austronesian languages are probably original from Taiwan (all but one top-level branches are exclusive from the island)

2. The main (largest) branch is a single top-tier branch called Malayo-Polynesian, which includes all non-Taiwanese languages.

3. Malayo-Polynesian includes many basal dialects, distributed geographically as follows:
·Philippines-Marianas-Sea Gypsies: 7
·Borneo-Malay-Madagascar: 4
·Java: 3
·Sumatra: 3
·Bali-Lesser Sunda: 1
·Wallacea-Melanesia-Oceania: 2

This distribution is strongly suggestive of a primary role in Malayo-Polynesian genesis and spread of Philippines and nearby islands, specially the big ones of former Sundaland (Borneo, Java and Sumatra).

Wallacea in contrast seems secondary and rather marginal in the process. However it would seem that Oceanic languages are derived from the northern group (Eastern MP) in this area, suggesting a semi-independent core and flow for proto-Polynesians in the area of Halmahera.

So we have the following levels:

3. Central-Eastern MP, with a probable core in Wallacea

4. Eastern MP with a probable core near Bird's Head peninsula (Halmahera?)

5. Oceanic

Oceanic is divided in 6 basal subfamilies, all of which except Yapese, represented in Melanesia and all but Yapese and Temotu, represented in Near Melanesia.

Hence we can easily conclude that Oceanic languages originated probably in Near Melanesia, where they probably picked that characteristic C2a genetic marker (which has nothing to do with non-Oceanic or at least non-Eastern MP Austronesians).

terryt said...

"So no genetic component can be said to be 'Austronesian'", right?"

Correct. Although collectively they look some some sort of hybrid cline between East Asian and Melanesian-looking people, with some resembling either extreme to some extent.

"So the Eastern group was essentially Melanesian and not properly from the South China Sea area".

Not even 'essentially'. Even at the eastern extremity New Zealand Maori do not look 'Melanesian'. They still possess more East Asian elements than do Melasians, even from as far east as Fiji. It's generally quite easy to identify separately Eastern Polynesians, Western Polynesians, Fijians, Melanesian Islanders and New Guinea Natives. The more 'Melanesian' look of islander from west of Fiji is considered to be the result of subsequent movement following behind the Lapita people.

"How do you or 'archaeologists' connect the Admiralty Islands to Borneo? It may be obvious from the linguistic viewpoint but it's not so obvious from other viewpoints, including genetic and archaeological ones".

But the linguistic connection is very convincing. It's extremely doubtful that Austronesian is the original language of the Admiralty Islands so those islands must have some connection to further east. The same basically holds for Borneo. The Austronesian languages are relatively recent in both places.

"Otherwise you may want to begin speaking of Polynesians and Malays as different populations and not anymore with that ultra-vague, complacent and misleading 'Austronesian' tag".

The problem is, as I mentioned before, Polynesians and Malays do look more like each other than, for example Polynesians look 'Melanesian' of Malays look 'East Asian'. So there is some genetic connection even thouth the haplogroups are, to quite an extent, separate. As Dienekes is constantly at pains to point out, appearance doesn't necessarily coincide with haplogroup.

"1. Austronesian languages are probably original from Taiwan (all but one top-level branches are exclusive from the island)"

The basal language is, but many languages in Taiwan seem to originate in the Philippines. There was an early back migration.

"2. The main (largest) branch is a single top-tier branch called Malayo-Polynesian, which includes all non-Taiwanese languages".

And many Taiwanese languages. I haven't looked at your link yet but I'd be very suspicious if they have neglected that connection.

"3. Malayo-Polynesian includes many basal dialects, distributed geographically as follows:
·Philippines-Marianas-Sea Gypsies: 7
·Borneo-Malay-Madagascar: 4
·Java: 3
·Sumatra: 3
·Bali-Lesser Sunda: 1
·Wallacea-Melanesia-Oceania: 2"

I agree that nthe language spread south from the Pilippines into Wallacea, but you can't leave Wallacea out of the Austronesian geographic region. And note the connection: Wallacea-Melanesia-Oceania. Where might it have started?

"However it would seem that Oceanic languages are derived from the northern group (Eastern MP) in this area, suggesting a semi-independent core and flow for proto-Polynesians in the area of Halmahera".

Now we're getting somewhere. Not New Guinea.

"Hence we can easily conclude that Oceanic languages originated probably in Near Melanesia, where they probably picked that characteristic C2a genetic marker"

How do you leap the chasm and reach that conclusion? Isn't Halmahera much more likely as the source? And where before Halmahera? I agree the C2a probably originated in near Melanesia (probably the Admiralty Islands) but it did so as part of the C2* expansion, for which Wallacea is the likely source.

Maju said...

"How do you leap the chasm and reach that conclusion? Isn't Halmahera much more likely as the source?"

No, the Halmahera area is not part of Oceanic languages but of a related branch only. Also there's no C2a anywhere in Wallacea. So IMO C2a was picked in Papua or Near Island Melanesia and so a private founder lineage in this haplogroup became the typically Polynesian C2a1 haplogroup.

"Even at the eastern extremity New Zealand Maori do not look 'Melanesian'".

They do not have much Melanesian mtDNA, right? Guess it's the mothers who define populations - much more than fathers in any case. Look: Central Africans have Y-DNA R1b in large apportions yet they do not look like Europeans, many Indians and Eastern Europeans share Y-DNA R1a yet they are quite different, etc.

"... but many languages in Taiwan seem to originate in the Philippines. There was an early back migration".

This is something I'd like to read more closely about. It's not impossible but the diversity levels are so different (all in Taiwan, only one branch in the rest) that it's difficult to believe that Taiwan was not at the origin of the Austronesian family of languages.

terryt said...

"there's no C2a anywhere in Wallacea. So IMO C2a was picked in Papua or Near Island Melanesia"

Doesn't follow at all. C2a obviously developed from C2* at some stage, presumably at some stage in C2*'s migration. The fact that no C2a is found in Wallacea surely shows the direction of that migration, eastward and not westward. And you earlier made the connection:

"·Wallacea-Melanesia-Oceania: 2"

Many people have refered to the Austronesian eastward migration as 'the fast train to Polynesia', so we can assume a rapid expansion from some sort of Austronesian homeland out into Polynesia as far as Tonga and Samoa. So, C2 coalesing in Wallacea, not New Guinea.

"and so a private founder lineage in this haplogroup became the typically Polynesian C2a1 haplogroup".

And C2a1 formed during that 'fast train to Polynesia', almost certainly somewhere in Vanuatu. Although it could be present there as a back-migration from Western Polynesia. But again it appears to be a product of the directionality of the fast train.

"They do not have much Melanesian mtDNA, right?"

I don't think any has been recorded in Maori. Just mtDNA B. But I agree with you to quite an extent. Along the route of the fast train the migrating people did mix with and pick up genes from other people. There is a Melanesian element in the Eastern Polynesians but for all we know the C2 population in Wallacea may have looked much like present day Melanesians, rather than like the present population there.

"Central Africans have Y-DNA R1b in large apportions yet they do not look like Europeans, many Indians and Eastern Europeans share Y-DNA R1a yet they are quite different, etc".

Correct. And Dienekes has just posted information on Y-hap R1b where it appears that the movement of the haplogroup is far more complicated than anyone (including me) had ever imagined.

"This is something I'd like to read more closely about. It's not impossible but the diversity levels are so different (all in Taiwan, only one branch in the rest) that it's difficult to believe that Taiwan was not at the origin of the Austronesian family of languages".

I couldn't get your link to work yesterday but I'll try and find some information on that connection between Philippine languages and Taiwan. But one of the reasons for the diversity of Taiwanese Austronesian languages is the back-migration of languages originating outside that island. On the other hand I have no doubt at all that the Austronesian languages developed first in, and spread from, Taiwan. If it is so that all languages outside Taiwan belong to a single branch then we have just two basic branches in the Austronesian language family, Ayatalic (in Taiwan) and the rest (including some in Taiwan).

terryt said...

I've tried to find information about back-migration from the Philippines to Taiwan, unsuccessfully. My information was based on books by Bellwood and others on the subject, some years old by now. I may be able to scan something and send it. What I did find was this comment at Wikipedia, 'In a recent treatment, all Austronesian languages were classified into 10 subfamilies, with all the extra-Formosan languages grouped in one subfamily and with representatives of the remaining 9 known only in Taiwan [4]'. So there may have been a shuffle, although I'd like to see Peter Bellwoods perspective on that. The reference is Blust R (1999) Subgrouping, circularity and extinction: some issues in Austronesian comparative linguistics. In: Zeitoun E, Jen-kuei Li, P (eds) Selected papers from the Eighth International Conference on Austronesian Linguistics. Academia Sinica, Taipei, pp 31–94. However he seems to make no real attempt to discover wider connections. After all the conference was in Taiwan.

But at another Wiki link I found these relevant comments:

"The seminal article in the classification of Formosan—and, by extension, the top-level structure of Austronesian—is Blust (1999)".

So he is mainly an expert on just the Taiwanese languages.

"Speaking very broadly, the Austronesian languages can be divided into three groups of languages: Philippine-type languages, Indonesian-type languages and post-Indonesian type (Ross 2002)".

No mention of a distinct Taiwanese group.

"The Malayo-Polynesian languages are frequently included within Blust's Eastern Formosan branch due to their shared leveling of proto-Austronesian *t, *C to /t/ and *n, *N to /n/, their shift of *S to /h/, and vocabulary such as *lima 'five' which are not attested in other Formosan languages".

A wider connection that Blust seems to ignore. Later the article restricts the possible extra-Taiwan connection to just the Paiwanic group from the south of the island.

The link:

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

Maju said...

"C2a obviously developed from C2* at some stage, presumably at some stage in C2*'s migration".

Or after that migration. Haplogroups, even if small, need some time (generations and generations) to coalesce. As we are talking here of two "onion's layers" C2a* (Melanesian only) and C2a1 (Polynesian only) --and as there's no C2a in Wallacea-- C2a must have coalesced in Melanesia.

You can't have C2a1 if you don't have C2a before it. The same that you can't build the 3rd floor unless the 2nd already exists.

This has actually very little to discuss. It's a must or as much so as we can reconstruct.

It is anyhow a clear case of founder effect (whose detail we can't really reconstruct... without a very snoopy time-machine).

"... we can assume a rapid expansion from some sort of Austronesian homeland out into Polynesia"...

What happened to the Lapita culture intermediate stage? It gives the perfect prehistorical context for the kind of dynamics we see.

"And C2a1 formed during that 'fast train to Polynesia', almost certainly somewhere in Vanuatu".

Well, I don't know if Vanuatu is the place or or if it was Tonga or somewhere else... but what is clear is that Polynesian-specific C2a1 arose from Melanesian-specific C2a, which is non-existent in Wallacea or anywhere else in the Austronesian zone.

"I couldn't get your link to work yesterday"...

Works fine for me (re-link). It needs JavaScript activated. Seems a nice site for linguistic reference in general.

terryt said...

"but what is clear is that Polynesian-specific C2a1 arose from Melanesian-specific C2a"

Which arose from C2* which certainly exists 'in Wallacea' and many places in the eastern 'Austronesian zone'.

Anyway, I found this relatively concise Peter Bellwood link which should tell you everything you're ever likely to want to know about Austronesians:

http://epress.anu.edu.au/austronesians/austronesians/pdf/ch05.pdf

I'll extrtact a few parts from the article. For a start it seems I'm not the only one who thinks their prehistory has wider relevance:

"I have never been convinced the European so-called 'colonial' experiences were totally unique to the past two centuries."

"Austronesian origins are here presented as an example of a frequent phenomenon in world prehistory, whereby populations who develop agriculture in regions of primary agricultural origins are provided with essential economic advantages over surrounding hunter-gatherers."

"a number of claims have been made for ancient genetic relationships between An [Austronesian] and other Asian mainland language families"

"differing opinions were presented linking Austronesian genetically with Thai-Kadai, Austroasiatic and Sino-Tibetan ... all these four language families might have been geographically contiguous in the early centuries of agricultural development in China"

"The Neolithic 'revolutions' of China evidently occurred in two culturally-connected regions. The first, in the basin of the Yellow River ... The second, in the middle and lowere Yangzi basin"

(All this surely indicates a substantial southward movement from those river basins. The only possible genetic footprint of such a movement is the various Y-hap Os. Therefore Y-hap O did not coalesce in SE Asia. It is invasive to the region, and is the main source for the East Asian element of the modern SE Asian phenotype.) Refering to the period before the Neolithic expansion from China:

"in the rainforests of Sundaland there probably would have been only very sparse settlement"

"The Austronesian expansion into Peninsular Malaysia surely occurred long after the expansion of Austroasiatic cultivators ... southward from Thailand during the third millenium BC."

"it is worth remembering that much of the early expansion of the Austronesian-speaking peoples was through Wallacea, especially the Philippines and Sulawesi with their manifold satellite islands. It is amoungst the more watery topography of Late Pleistocene Wallacea, rather than land-bridged Sundaland, that one might expect pre-Austronesian maritime traditions to have flourished and to have been transmitted to later arrivals"

(Note that he includes the Philippines in 'Wallacea', as he does in the following extract)

"Indeed, in order to settle the Wallacean islands (Philippines, Sulawesi, Lesser Sundas, Moluccas), as also Australia and New Guinea, the original Pleistocene colonists must have had some degree of seafaring capacity, even if rudimentary, by at least 40,000 years ago. Did the Austronesians learn a number of seafaring skills from them ..."

"how difficult it is to reconstruct terms for boats and navigation for Proto-Austronesian (as opposed to Proto-Malayo-Polynesian)"

(continued)

terryt said...

It is also obvious that Bellwood envisages progressive improvements in boating technology over the whole period from the first crossing of Wallace's Line:

"Many of these people's [Pre-Austronesians on islands north and northwest of New Guinea], like their Wallacean cousins, also knew how to cross sea, as witnessed by the Late Pleistocene discovery and distribution of New Britain obsidian ... However their seafaring skills were perhaps more limited than those of the later Austronesian colonists"

"3000 BC; Proto-Austronesian expansion to the northern Philippines; improvement of seafaring technolgy"

"Second/first millenium BC? Beginnings of mobile maritime (proto-sea nomad?) adaptations around the Sulu and Sulawesi Seas"

"These, in turn, might have laid the groundwork for:
Middle and late second millenium BC; Lapita colonization of remote Oceania to as far as Tonga and Samoa. Seafaring skills were here developed further amidst an ever-expanding vista of uninhabited islands"

(The only real disagreement I would have with him is when he lists a number of reasons for the Austronesians' rapid expansion. Although he does write refer to 'sheltered inshore fishing' on new islands he ignores the fact that those fish would have been more numerous and larger than were those remaining on islands they'd already colonised. Shellfish would also have been larger, more numerous and accessible. And unoccupied islands would have had huge colonies of nesting seabirds as well as flightless birds on many islands. Finding previously unoccupied islands has been called the prehistoric equivalent of winning the lottery.)

Maju said...

"Which arose from C2*"

But earlier. Otherwise we'd be unable to differentiate C2a from C2a1. Mutations need time to coalesce (i.e. not just to happen but to become stablished in a number of individuals of reproductive age). You want all the structure to unfold in few centuries and to leave the only intermediate traces among Melanesians but be original from somewhere else.

I say that is absurd.

And I say that twisted logic is potentially triggered by (unconscious?) racism, as happens with the weirdo theories on E1b1b1 origins, etc.) Trying to deny their role to "black" peoples in order not to be (in this case your admired Polynesians) somewhat "black" by ancestry (against the factual evidence). Patriarchal mythical-thought is also intertwined (as paternal ancestry seems to be much more a matter of such discussions than maternal one).

Overall it's ghosts and not reason what are driving your "logic".

"... Peter Bellwood link which should tell you everything you're ever likely to want to know about Austronesians"...

Thanks for the link. Though I doubt at this time there's anything of relevance left to know about Austronesians. You have been extremely insistent on these peoples.

"I have never been convinced the European so-called 'colonial' experiences were totally unique to the past two centuries."

That's another cultural problem, along with racism and patriarchy. Plainly: some peoples' history is so altered by colonization and genocide of the natives that they seem to think it's normal and that it also happened all the time since humans exist.

They are surely wrong, specially because:

1. no such huge scale hyper-fast demic movements nor genocides could be organized before modernity.

2. primitive peoples, including Neolithic peoples before the Chalcolithic stage, had no polities that could organize much.

3. the time frames considered fully allow for total cultural change at the populations "under attack". Native Americans for instance displayed a clear adaptative drive before colonization and modernized themselves (naturally). Racist genocides in a modern massive context could occasionally stop that process but in the time of millennia and with the technologies and numbers available in the Neolithic, this would plainly be impossible at large scale.

"All this surely indicates a substantial southward movement from those river basins. The only possible genetic footprint of such a movement is the various Y-hap Os".

Can be cultural, bringing little genetics with it. People copy each other, including technology and language.

"Therefore Y-hap O did not coalesce in SE Asia. It is invasive to the region"...

I don't see how this can be proven. Actually I find difficult to see such a massive continental-scale migration (and genocide) happening in Neolithic conditions. Plus we see almost no specific Austronesian genetic signature.

I have not fully solved the matter but it's obvious that some O subclades are native from SE Asia, as is the haplogroup as a whole (specially if you include South China in SE Asia, as usual). It also seems that the age of haplogroup O (and all its relatives like M, N, P, Q, R, S) is clearly Paleolithic.

Some local replacement may have happened (Mentawai are very much like Taiwan aborigines in their genes, for instance) but this cannot be generalized t the whole region, not with the available evidence certainly. Surely Austronesian flows moved haplogroups around somewhat and I am unsure of the extent of these flows in Sundaland but the situation becomes more clear as you move eastward across the Wallace Line and eventually into the open Ocean.

Well...

Maju said...

Read Bellwood's paper. Essentially he is denying any sort of pre-Austronesian Neolithic in the Sundaland region (for instance) and also making too many assumptions about how primitive Neolithic peoples might outcompete in a genocidal manner the supposed pre-existent foragers.

It doesn't seem too different from the Neolithic colonization hypothesis in Europe, which mostly seem to make little sense, specially on light of the genetic evidence but also of some of the archaeological evidence, such as the toolkit continuity in most Cardium Pottery sites, which signals indigenous peoples going agricultural rather than colonists.

In the case of SE Asia we have more limited archaeological evidence but I can only imagine that in due time it will inform us of much more complex and less hyper-simplistic colonialist scenarios. One thing the local archaeology already says anyhow is that there was Neolithic before Austronesians, what simply lays Bellwod's conjectures to rest, as local Neolithic peoples, as happened with Papuans, would be able to resist new arrivals with that agricultural background (providing numbers and specially the mindset of farmers, already interested in "productive" economy).

terryt said...

"You want all the structure to unfold in few centuries"

And that's what Ebizur's information indicated. C2a is not ancient, a little more than 4000 years. Exactly what we would expact as part of the Austronesian expansion.

"Essentially he is denying any sort of pre-Austronesian Neolithic in the Sundaland region"

Far from it. He quite specifically proposes that Austroasiatic-speaking people preceded them:

"The Austronesian expansion into Peninsular Malaysia surely occurred long after the expansion of Austroasiatic cultivators ... southward from Thailand during the third millenium BC."

and:

"perhaps it can be hypothesized that the majority of the agriculture populations of Thailand, Indochina and Peninsular Malaysia at this time spoke languages related most closely to the modern Austroasiatic family."

He has already mentioned that this family too was a product of southward movement from the Chinese Neolithic.

"It doesn't seem too different from the Neolithic colonization hypothesis in Europe, which mostly seem to make little sense"

You're obviously free to believe what you like.

"One thing the local archaeology already says anyhow is that there was Neolithic before Austronesians, what simply lays Bellwod's conjectures to rest"

So you're wrong again. In fact that is his explanation for why the Austronesians made so little impact on the Sunda mainland. You are certainly a real expert at reading an article and not comprehending anything that conflicts with your beliefs.

Maju said...

"C2a is not ancient, a little more than 4000 years".

I disagree.

And if it'd be recent it'd be still older than C2a1, and hence C2a and C2a1 would still have coalesced in Melanesia.

"Far from it. He quite specifically proposes that Austroasiatic-speaking people preceded them"...

He's disregarding any Neolithic south of Malaysia peninsula, that's what I meant.

"He has already mentioned that this family too was a product of southward movement from the Chinese Neolithic".

What there's no evidence of. No Austroasiatic languages are spoken in South China. Also where did other "Chinese Neolithic languages" like Daic, Hmong, Tibeto-Burman or Sinitic come from?

I am of the opinion that Autroasiatic is not the product only of the Neolithic diffusion from South China but rather a native SE Asian language family. Daic would be the native language of South China, while Sino-Tibetan would be of the inner lands around Sichuan and possibly related Sinitic from the North Chinese Neolithic (if it was not Korean or whatever).

However I suspect that Austroasiatic peoples made inroads into Java, Borneo, Sumatra and beyond with their Neolithic and their language. Unless it is a Paleolithic component what we are seeing in these islands.

I don't see clear that Neolithic spread = migration and genocide. In fact I usually see many signs of the opposite interepretation: it is largely a know-how spread by short or middle distance migrants who are normally minority at their area of arrival, (unless it is a desert island) and who assimilate (more ore less intensely) the locals easily because of their superior economy (however not as radically superior as guns vs. spears, steam machines vs. roaming bisons - not at all).

We can see this process of assimilation of local forager tribals into agriculture in a more or less forced or willing manner everywhere. It's true that foragers do not typically like working but it's all a matter of contact and oportunity, and specially watching your kind do it too. There were many millennia of transition everywhere for the process to be smooth. Also early agriculturalists were not as radically different from hunter-gatherers... they just lived more sedentarily and complemented the hunt with crops and cattle.

I just don't buy Neolithic replacement theories unless it's clearly demonstrated in the archaelogical plane, as may be the case with the Danubian Neolithic (but not elsewhere).

"So you're wrong again".

No I'm not and if I got an euro for each time you said I'm wrong while not being, I'd have some nice hundred bills by now.

And if I got an euro for each time you mistook A for B (Peninisular Malaysia for the whole Sundaland area) in order to make it fit into the argumentation and forge an appearance of an error on my side and confirmation on your side, I'd have thousands already.

terryt said...

"I disagree".

Only because it doesn't fit your belief.

"He's disregarding any Neolithic south of Malaysia peninsula, that's what I meant".

Not so. He says specifically, 'Austroasiatic-speaking cultivators might originally have colonised onwards quite rapidly into parts of Sumatra and western Borneo if one takes a broad view of the pollen evidence and some linguistic substratum hints'.

"No Austroasiatic languages are spoken in South China".

Possibly not actually in China.

http://en.wikipedia.org/wiki/Austro-Asiatic_languages

From the link:

"Austro-Asiatic languages have a disjunct distribution across India, Bangladesh and Southeast Asia, separated by regions where other languages are spoken. It is widely believed that the Austro-Asiatic languages are the autochthonous languages of Southeast Asia and the eastern Indian subcontinent, and that the other languages of the region, including the Indo-European, Kradai, Dravidian and Sino-Tibetan languages, are the result of later migrations of people".

But Bellwood seems not to accept that Austroasiatic languages are 'autochthonous languages of Southeast Asia and the eastern Indian subcontinent', but that they are simply the earliest of the Chinese Neolithic languages to move south, although he agrees that its distribtution has been broken up by later movements.

And we have another language family with a discontinuous distribution: Miao-Yao, or Hmong as you have called it:

http://en.wikipedia.org/wiki/Hmong-Mien_languages

I have seen claims it is related to Austroasiatic, in which case it would be the South Chinese branch of Austroasiatic.

"Also where did other 'Chinese Neolithic languages' like Daic, Hmong, Tibeto-Burman or Sinitic come from?"

His argument is that they are all related and moved south with the Central Chinese Neolithic.

"I just don't buy Neolithic replacement theories"

But you've already written:

"It's true that foragers do not typically like working"

It's not so much they don't like working. Foraging is hard work too. It's quite easy to envisage how agriculturists could replace huntergatherers without involving any level of genocide for two main reasons. First, a great deal of evidence shows that hunter-gatherers are very reluctant to settle and become agriculturists. Second, agriculture allows a greater density of popualtion. So even if the number of huntergathers in a region remains constant they will soon be greatly outnumbered by the agriculturists.

So, Maju, you're wrong again.

Maju said...

"But Bellwood seems not to accept that Austroasiatic languages are 'autochthonous languages of Southeast Asia and the eastern Indian subcontinent', but that they are simply the earliest of the Chinese Neolithic languages to move south, although he agrees that its distribtution has been broken up by later movements".

It's just his opinion. He is not my messiah.

I am of the opinion that Austroasiatic did indeed expand with Neolithic but from SE Asia and not China. The fact, for instance, that the red AA autosomal structure component is not really found in South China clearly indicates that this component originated south of the PRC's border (discussed here and here, and largely confirmed here).

I think you are trapped in obsolete models of East Asian prehistory based on very much speculative constructs based only on skull shapes, which we know are largely plastic and not merely a genetic product. I really cannot accept that without further supporting evidence of other type.

"And we have another language family with a discontinuous distribution: Miao-Yao, or Hmong as you have called it".

Yes. I cannot easily take a clear side for Daic or Hmong-Mien on which is the "real" Yangtze River Neolithic population derived. It is possible that one is more strictly the Yangtze branch and the other maybe corresponds to the Pearl River's "corded ware" Neolithic (a derivate of the previous but with a distinct personality). Or whatever (I really do not know for sure).

"I have seen claims it is related to Austroasiatic"...

Well, there are so many fringe linguistic speculations that is not really worth considering but the well established ones.

A possibility could be for each of the three continental SE Asian language families to have evolved in the Paleolithic in nearby but distinct regions, maybe Yangtze, Pearl River and Mekong. Another possibility would be that the AA family formed in Indochinese Neolithic (Thailand and such) and expanded northwards, while Yangtze Neolithic with Daic language expanded southwards and Hmong-Mien got "sandwiched" between both, surviving only "miraculously" (miracles do not exist, so there must be some reasons - just that I do not know them).

"His argument is that they are all related and moved south with the Central Chinese Neolithic".

There's no such thing as "Central Chinese Neolithic", just South and North AFAIK (Yangtze and Yellow River, roughly, with some original variants towards the coast but related anyhow). It seems now pretty clear that the language of the Northern Neolithic must have been Sinitic (not Sino-Tibetan as a whole but specifically Sinitic, with Tibeto-Burman being a more inland and less clearly agricultural branch around Sichuan). The problem would be clarifying, if possible the cultural and linguistic characteristics of the Yangtze and Indochinese Neolithics. But I'd say that Kradai and Austroasiatic are nice bets respectively.

Maju said...

"Foraging is hard work too".

Very different still. The kind of scheduled work required to farm is really hard work in comparison with the forager's lifestyle. Foragers don't talk in terms of work but of things they do in an often semi-magical context. Farmers talk of work as the center of their lives. Foragers' life may be "harder" in the sense of unpredictability but farmers' life is predictably much harder overall because it's almost only work and more work, day after day, season after season. Maybe not now with industrial agriculture but certainly in the past.

"First, a great deal of evidence shows that hunter-gatherers are very reluctant to settle and become agriculturists".

Yes but that is in the short/mid term. What about when the process takes centuries and millennia and when there is not a strictly dividing line between agriculture/pastoralism and foraging? In the West Mediterranean, for instance, Neolithic arrived c. 5000 BCE but it did not arrive clearly to most of the Atlantic, just a few hundreds kilometers away until almost two millennia later. Also in the Western Med we do have clear evidence of foragers becoming agriculturalists, even if the time frame was shorter, and that encompasses most of Cardium Pottery sites.

So I understand that in early Neolithic conditions the transition was easier in many cases, partly because the dividing line was not so strict and partly because the evidence clearly supports such a transition form foraging to farming.

Modern foragers tend to resist agriculturalization in part because they know that, in a hierarchical society their status would be lowest in most cases. Still the transition happens at individual and even group level often enough for most of the local genetic pool to make the transition in few centuries, even if there's a hardcore group that totally refuses change. But in non-hierarchical early Neolithic, when farming was largely a mere complement to hunting, the transition should be much easier.

"Second, agriculture allows a greater density of population. So even if the number of huntergathers in a region remains constant they will soon be greatly outnumbered by the agriculturists".

Well, it certainly allows for greater densities but also restricts mobility (agriculture is essentially sedentary, so the most mobile groups would be those that still keep intermediate abilities and are not strictly dependent of agriculture only). Also there are ecological conditionants, which I believe were critical in the slow down of Neolithic advance to Atlantic Europe (too humid) and even in ill-understood cultural changes like the one at the root of the Danubian Neolithic.

The Danubian case is a clear case where population replacement is a clear possibility. But it is also a clear case where there is no front wave advancing from the Balcanic Neolithic but actually a strong cultural founder effect first in NE Hungary (Eastern Linear Pottery) and then in West Hungary, East Austria and Moravia (Western Linear Pottery, Danubian Neolithic senso stricto), which is at the origin of all the expansion into Germany, Bohemia and Poland, plus the hybrid expansion into Belgium, North France and parts of SE Europe (over Balcanic Neolithic and other pre-existent Neolithic cultures).

So we have an inflexion point at least at the Middle Danube. Even if there was a demic replacement in some of these areas, it was not any sort of simple wave front but there were founder effects and meaningful cultural changes which almost without doubt implied at least partly pre-existent local ex-foragers.

"So, Maju, you're wrong again".

I still think I am probably right.

terryt said...

"I still think I am probably right".

The remark was actually about your ideas concerning the Austronesians, not the topic you have adopted here. Most of which I agree with.

"So I understand that in early Neolithic conditions the transition was easier in many cases, partly because the dividing line was not so strict and partly because the evidence clearly supports such a transition form foraging to farming".

You would still have a time gap before the residents made the change. During which time incoming haplogroups would become more numerous than the residents'.

"agriculture is essentially sedentary, so the most mobile groups would be those that still keep intermediate abilities and are not strictly dependent of agriculture only"

Not necessarily. The agricultural group would still expand through the most desirable region first, leaving hunter-gatherers along the way, to join in or not as they saw fit.

"a strong cultural founder effect first in NE Hungary (Eastern Linear Pottery) and then in West Hungary, East Austria and Moravia (Western Linear Pottery, Danubian Neolithic senso stricto), which is at the origin of all the expansion into Germany, Bohemia and Poland, plus the hybrid expansion into Belgium, North France and parts of SE Europe (over Balcanic Neolithic and other pre-existent Neolithic cultures)".

That's much the same scenario I argued for concerning the Austronesian expansion. Which Y-haplogroup is 'Austronesian'? In both cases new people joined in and old haplogroups mutated into new versions.

terryt said...

"It's just his opinion. He is not my messiah".

No. But the genetic evidence does support his scenario if we're prepared to concede that Y-hap O is not ancient SE Asian. I'll come back to that.

"just South and North AFAIK (Yangtze and Yellow River)"

But they're very close to each other and we might as well consider them as one.

"A possibility could be for each of the three continental SE Asian language families to have evolved in the Paleolithic in nearby but distinct regions, maybe Yangtze, Pearl River and Mekong".

Why not all three in separate regions within the Yangtze and Yellow River Neolithic?

"I am of the opinion that Austroasiatic did indeed expand with Neolithic but from SE Asia and not China".

So where did the SE Asian Neolithic come from?

Let's see if we can find any connection between genes and the Austroasiatic, Hmong-Mien and Munda languages. If we start with India we see that Y-hap O2a is the best candidate. It is found through India from Andhra Pradesh to the Pashtun region. Perhaps Y-hap O2a carried the SE Asian Neolithic and the SE Asian-connected languages into India. But the languages spread nowhere near as far as the Y-chromosome.

Outside India O2a is spread through island and mainland SE Asia. There it is particularly associated with speakers of Austroasiatic and Kradai. Surprise, surprise. But again the Y-hap seems to have traveled further than the language because by the time O2a reached Madagascar it was very much associated with Austronesian.

So as you say O2a's distribution is probably the product of a SE Asian Neolithic expansion. But where had that O2a originated?

continued

Maju said...

We are just discussing in circles. We just do not agree. Assume it.

terryt said...

This is only indirectly connected to Gibraltar, but I think it is applicable to prehistoric migration in general.

Both Y-hap O2a and its relation O2b are scattered through much of China. But O2b is especially common in Korea and Japan. Is the connection between O2a and O2b north and south via the coast? Or do they both originate at the eastern end of the Yangtze and Yellow River Neolithic, one going north the other south? The two haplogroups are perfectly placed to encourage such a viewpoint.

Can we now say that O2 is Yangtze and Yellow River Neolithic in origin and spread with the early Yangtze and Yellow River Neolithic? But in the south O2a next participated in a second expansion, that of the Austronesians. So can we call it Austronesian?

When we turn to to the eastern end of the Austronesian expansion we find a close association with Y-hap O3, not O2. Perhaps we can make a case for O3 being one of the original SE Asian haplogroups. Now, Y-hap O3 must have been widespread through much of China long before it reached Tonga. Could it be possible that Y-hap O3 coalesced on an island somewhere between Tonga and Thailand? And then spread north? Or is O3's distribution a product of post-Paleolithic expansion in the opposite direction? On the Asian mainland Y-hap O3 reaches as far north as Tibet, and in SE Asia is especially associated with the Hmong-Mien. Perhaps it is in the same situation as O2, except that O3 expanded from the western end of the Yangtze and Yellow River Neolithic. So far from being an ancient haplogroup in SE Asia it seems it took some time for Y-hap O3 to reach SE Asia, but when it did it took off: east.

We're left with the ancient origin of the Austronesians, before Y-haps O2 and O3 (not to mention C2) had hijacked it. In Taiwan we find a third Austronesian haplogroup: O1. Y-hap O1 is certainly common through Wallacea, from the Philippines in the north to the Lesser Sunda Islands in the south. So the evidence for a connection to the Austronesians is pretty good.

Did Y-hap O1 originate in Taiwan? Surely it entered Taiwan from mainland China. O1 is found through much of South China, and in SE Asia it, like O2, is associated with the Kradai languages. It looks very much as if Y-hap O1 has also moved south with a Yangtze and Yellow River Neolithic expansion, possibly to the east of Y-hap O3's expansion and to the west of Y-hap O2's.

Another reason we might be inclined to associate Y-hap O with China rather than with SE Asia is because of its interesting relationship with Y-hap N. Was there, or was there not, an ancient haplogroup NO? Either way what we see is a clear geographic demarkation line separating N and O. Y-hap N occupies much of the region north and west of the Inner Mongolian high country, while O occupies much of the region to the south of that high country.

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

"Can we now say that O2 is Yangtze and Yellow River Neolithic in origin and spread with the early Yangtze and Yellow River Neolithic? But in the south O2a next participated in a second expansion, that of the Austronesians. So can we call it Austronesian?"

No. Re-read Karafet's paper.

Besides, O diversity is highest towards the south for all meaningful subclades.

terryt said...

"No. Re-read Karafet's paper".

I find it very interesting that you are very quick to dismiss the conclusions of papers that show your belief to be wrong, yet cling like a frightened baby to papers that happen to support your belief. Has it occurred to you that Karafet may be wrong? Certainly other researchers see the situation differently. Consider the interesting fact noted in the Karafet paper that Y-hap O generally didn't cross Wallace's line. Surely if O had been in SE Asia from the time humans first crossed the line it would have done so too. That must suggest to you that it is a late arrival in the region.

Just for half a minute open your mind to the possibility that Bellwood is correct, and then read what I posted yesterday.

terryt said...

"O diversity is highest towards the south for all meaningful subclades".

That is hardly surprising. Have you any idea at all what sort of environment they were moving through? I haven't seen South China from the air but I can imagine. I have seen the Laos/Vietnam border region. I've flown over most of New Zealand and seen some big mountains from the air, I've flown over the Pyrenees and the Atlas. The mountains in the L/V border make all these look like mere pimples, even though in some cases the others may be higher. I'd guess that each valley and each ridge in South China would have developed its own version of any haplogroups that moved in.

Maju said...

"Has it occurred to you that Karafet may be wrong?"

Of course. I just happen to agree with her. She is also one of the best population geneticists.

"Certainly other researchers see the situation differently".

There's always some different opinion. I have to make up my own.

Most TRMCA estimates for O make it 30-40 Ka old anyhow, what is totally incompatible with your Neolithic hypothesis.

Besides, the diversity of the haplogroup (as happens with others) is highest towards the South, indicating it expanded within the "coastal migration" process. However it may represent a secondary flow (related to mtDNA R probably) and this is not yet sufficiently well explained in archaeological terms. Much of the relevant archaeology may well be sunk under the sea today though.

"Consider the interesting fact noted in the Karafet paper that Y-hap O generally didn't cross Wallace's line. Surely if O had been in SE Asia from the time humans first crossed the line it would have done so too".

I already explained this to you. Repeating:

1. There is a physical major barrier: Wallace Line.

2. There were already people with similar technological level at the other side of the barrier, so any rare arrival in either direction would be absorbed and its genetic markers "drifted out".

Only a steady, massive, demic flow could overcome this situation but, seemingly, this did not happen (until much later and only to some degree).

"Just for half a minute open your mind to the possibility that Bellwood is correct"...

He's obviously wrong. Read what I posted a few days ago on that myself. He is oversimplifying reality in order to make his Austronesians an epical people of sorts. It's ridiculous Anglosaxon-style (mediatic rather than truly academic) pseudo-research. He's ignoring everything else and, as you denounced earlier, he seems to be hiding data, such as Filipino Negrito DNA because it does not fit with his model. That's not scientific but manipulative.

Maju said...

"I'd guess that each valley and each ridge in South China would have developed its own version of any haplogroups that moved in".

Well, that's a convenient ex-post-facto explanation you just made up.

The reality is that India and SE Asia (including South China) show that high diversity and in many areas these hills and mountains you mention simply do not exist.

It seems to me that you idealize the "Great Hunt" lifestyle of the steppes. But actually this was necessarily a derived and minority lifestyle.

Not just humans are best adapted to tropical climate naturally but also you see that the peoples of the Eurasian tropical belt have not lost the original blackness typical of this environment. Instead people migrating to America through the North did lose it and never recovered that original tropical trait. This is clear evidence (in addition to other) for a tropical belt migration.

This issue of pigmentation is very important because we do see two different patterns of depigmentation in the East and the West, which are totally consistent with a South to North migration by two different routes, east and west of South Asia. It's not just a matter of diversity but also a matter of global patterns.

You are basically saying: people went through the steppes (just because) and then they went extinct or nearly extinct, allowing for a radical reduction of diversity. I say: nonsense! This is putting the cart before the horses.

terryt said...

"I already explained this to you. Repeating"

And I've already explained why none of those theories of yours are valid.

"He's obviously wrong".

I just happen to agree with him. He is also one of the best SE Asian anthropologists. And most in this region would agree with him. Genetics is by no means the only evidence. It must be matched with archeology and anthropolgy, as I'v read you claim several times.

"He's ignoring everything else and, as you denounced earlier, he seems to be hiding data, such as Filipino Negrito DNA because it does not fit with his model".

As far as I know no-one has published the definitive Filipino Negrito data. We await Spencer Wells' publication.

"The reality is that India and SE Asia (including South China) show that high diversity and in many areas these hills and mountains you mention simply do not exist".

What? That's so ridiculous it doesn't deserve a response.

Maju said...

"Genetics is by no means the only evidence. It must be matched with archeology and anthropology, as I've read you claim several times".

And I insist here. He is conveniently ignoring all the pre-Austronesian Neolithic of the islands. That's absolutely abusive.

"As far as I know no-one has published the definitive Filipino Negrito data. We await Spencer Wells' publication".

Oops, payed it on the wrong guy. Sorry.

"That's so ridiculous it doesn't deserve a response".

Are you telling me that most of Thailand is not flat? And to a lesser extent Cambodia and the Mekong Delta? And all the large areas of continental platform SE of Thailand up to Borneo now submerged in spite of the sea only rising 120 m., clearly insuffiecient to sink any meaningful relief?

In South China itself, Yunnan is certainly a mountainous country but further East the relief is much gentler, specially along the Pearl and Yangtze lower courses. And again the large coastal platform to the East.

So what is "so ridiculous" about clarifying that it is a variegated region, much as any other?

You are avoiding reality.

terryt said...

"He is conveniently ignoring all the pre-Austronesian Neolithic of the islands".

And I've consistently pointed out to you that he's not. If you keep insisting he is, then we're wasting our time.

"Are you telling me that most of Thailand is not flat? And to a lesser extent Cambodia and the Mekong Delta?"

But that is not the region for which you claimed, 'O diversity is highest towards the south for all meaningful subclades'. The region with the highest O diversity is precisely the region that is most mountainous, such as South China and the region immediately the south (including northern Thailand).

"further East the relief is much gentler, specially along the Pearl and Yangtze lower courses. And again the large coastal platform to the East".

And don't forget the Hwang Ho just to the north. It and the Yangtze are precisely the regions where Bellwood proposes that all the existing SE Asian have their origin.

"So what is 'so ridiculous' about clarifying that it is a variegated region, much as any other?"

What was 'so ridiculous' was your claim that most of South China and SE Asia were flat ('The reality is that India and SE Asia (including South China) show that high diversity and in many areas these hills and mountains you mention simply do not exist'). A quote here, specifically concerning Burma but the comments really apply more widely through the region, from E. R. Leach, who fought in Burma with the locals during WWII and later became an anthropologist:

"The immediate neighbourhood of these great rivers and of their principal tributaries is low-lying, flat and fertile; away from the rivers the country is generally mountainous, often precipitous".

Everybody, except you, accepts that variation is greatest in mountainous regions. For example languages of New Guinea and the Caucasus are particularly varied, presumably because the mountains prevent prolonged interaction between populations. Speciation is greatest in mountainous regions for most genera. Even in Italy genetic differentiation between populations is greatest in the hilly regions. Surely it makes sense that we should find greater diversity of haplogroups in the mountains of China and SE Asia.

Maju said...

"And I've consistently pointed out to you that he's not".

You said Malaysia (the peninsula, I understand), not the islands. Whatever the case, his whole thesis is dependent on Austronesians spreading Neolithic, which is not the case.

"But that is not the region for which you claimed, 'O diversity is highest towards the south for all meaningful subclades'. The region with the highest O diversity is precisely the region that is most mountainous, such as South China and the region immediately the south (including northern Thailand)".

Mountains or islands may act as you say but I am looking at the whole regional pattern. Overall and almost without exception the pattern of scatter of all haplogroups (male or female) is from South to North. Y-DNA O is most diverse in SE Asia without doubt and also is O3.

"And don't forget the Hwang Ho just to the north. It and the Yangtze are precisely the regions where Bellwood proposes that all the existing SE Asian have their origin".

I just happen not to believe in such ideas of population replacement unless there's clear evidence. I do not because in all cases I know it happens that, in the end, the replacement was not really strong, if apparent at all. And in the case of SE Asia the evidence is scant if any, with a lot of evidence suggesting otherwise.

"What was 'so ridiculous' was your claim that most of South China and SE Asia were flat"...

That is not what I said! I said:

"The reality is that India and SE Asia (including South China) show that high diversity and in many areas these hills and mountains you mention simply do not exist".

And I stand for that.

Your quote for Burma is precisely confirming what I said: 'The immediate neighbourhood of these great rivers and of their principal tributaries is low-lying, flat and fertile'. And in the past as now it fed most of the people surely.

But Burma is anyhow just a fraction of all SE Asia. I wish we had a better archaeological understanding of the whole region and not just isolated fragments but the case is that SE Asia, specially in the Ice Age, had vast swaths of flat land, a good deal of it submerged now.

"Surely it makes sense that we should find greater diversity of haplogroups in the mountains of China and SE Asia".

It makes some sense, but not for high level haplogroups, which represent, signal, the deep past. Otherwise it'd be as if the small isolated communities of the mountains would have been pouring all the time onto the plains, when actually mountains act as refuge but never that I know as source of demographic expansions (there are some Caucasus-obsessed freaks but they are surely wrong).

Much more likely is that people tended to expand from the valleys and plains into the mountains when and where possible, because these are the ones with demographic power, "rich" and therefore producing more children some of which would eventually need to adventure in search of new lands (or will need to find refuge in the mountains).

I understand that the general flow of people in pre-industrial conditions is from high densities to low densities and not otherwise. Hence mountains are not the reason why there is more diversity at high level (and not low level) clades in SE Asia. It is valleys and other rich areas, mostly lowland, where the origin of these expansions was.

terryt said...

Again you appear to be deliberately misreading the link.

"Whatever the case, his whole thesis is dependent on Austronesians spreading Neolithic, which is not the case".

No. Bellwood is quite specific that the Neolithic was first carried into SE Asia by AUSTROASIATIC speakers. The Austronesians arrived later.

"It is valleys and other rich areas, mostly lowland, where the origin of these expansions was".

The book on Burma I mentioned earlier contradicts that statement. The valleys were settled later, by wet rice cultivators. The mountains had already been settled by dry rice cultivators.

"I just happen not to believe in such ideas of population replacement unless there's clear evidence".

Ah, your 'beliefs' again. But there's actually no need to postulate 'population replacement' anyway. If an incoming Neolithic group has more children that the local hunter gatherers the ratio of the two will change. It's quite possible that Y-haps C, F and K represent the earlier, thinly spread hunter-gatherers, and they are the earliest haplogroups anywhere in SE Asia: mainland, Malayan Peninsular or islands. It's just that the Y-hap Os eventually came to dominate them by numbers.

"Y-DNA O is most diverse in SE Asia without doubt and also is O3".

I realise you are only prepared to accept what Wikipedia says when it suits your belief, but that's not what it says on the subject:

"Some researchers believe that it [Y-hap O3] first appeared in China approximately 10,000 years ago".

Presumably from O* (more of that soon). And further:

"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. Haplogroup O3 comprises about 50% or more of the total Y-chromosome variation among the populations of each of these language families. 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. The Tibetans, however, despite the fact that they speak a language of the Tibeto-Burman language family, have high percentages of the otherwise rare haplogroups D1 and D3, which are also found at much lower frequencies among the members of some other ethnic groups in East Asia and Central Asia".

Sounds pretty much like Chinese Neolithic, and no mention of SE Asia as its origin.

terryt said...

To continue. Regarding O2 wiki says:

"Besides its widespread and patchy distribution, Haplogroup O2 is also notable for the fact that it can be divided into two major subclades that show almost completely disjunct distribution".

Difficult to account for under the scenario you propose, but very simple if its orgin is also the Chinese Neolithic.

And O1:

"The peak frequency of Haplogroup O1 is found among the aborigines of Taiwan, precisely the region from which linguists have hypothesized that the Austronesian language family originated".

But it obviously didn't arise on Taiwan in the Paleolithic. It must have arrived there from somewhere else.

"A slightly weaker correlation is observed between Haplogroup O1 and the Han Chinese populations of southern China, as well as between this haplogroup and the Kradai-speaking populations of southern China and Southeast Asia".

Ah, perhaps now we see a SE Asia Paleolithic origin for just this O haplogroup. But:

"The distribution of Daic languages in Thailand and other parts of Southeast Asia outside of China has long been believed, for reasons of traditional linguistic geography, to reflect a recent invasion of Southeast Asia by Daic-speaking populations originating from southeastern China"

So, again, its not SE Asian in origin but it may have arisen from the southern end of the Chinese Neolithic.

And even regarding O* Wiki says:

"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".

Nothing about SE Asia.

Maju said...

"No. Bellwood is quite specific that the Neolithic was first carried into SE Asia by AUSTROASIATIC speakers".

Again mixing stuff. He only means mainland SE Asia. I quote, as you seem you still need to re-read the whole article:

"perhaps it can be hypothesized that the majority of the agricultural populations of Thailand, Indochina and Peninsular Malaysia at this time spoke languages related most closely to the modern Austroasiatic family".

(He seems to have some confused concept of what Indochina means, btw, because it includes all mainland SE Asia south of the Chinese border, and not just former "French Indochina").

He briefly admits that:

"Indeed, Austroasiatic-speaking cultivators might originally have colonized onwards quite rapidly into parts of Sumatra and western Borneo if one takes a broad view of the pollen evidence and some linguistic substratum hints".

But immediately goes on to make a clear cut difference between the Malay Archipelago and its surroundings:

"The pre-existence of such agricultural
populations in at least Malaya and Vietnam can explain why the mainland of Southeast Asia was only a region of small-scale prehistoric settlement by
Austronesians, just as, under rather different cultural circumstances, was New Guinea".

Not a word on the islands between Malaysia and Papua, which are the main issue here. In fact, he is implying that Austronesians could expand into island SEA because there were no (or very few) farmers.

However there are other possible explanations and there are indications that farming pre-dates Austronesian arrival in all or most of these islands. He's just taking the simplistic model of demic wave so loved of North European classical views of human historical reality (this people replaces that people radically, instead of this culture replaces that culture but essentially the people remains the same). This model is oversimplistic and highly unlikely in all cases.

Maju said...

"The book on Burma I mentioned earlier contradicts that statement. The valleys were settled later, by wet rice cultivators. The mountains had already been settled by dry rice cultivators".

That's only Neolithic (again cross-reasoning through the ages!). Anyhow the data for lowlands, specially flood basins, is likely to be incomplete even in the presence of a good archaeological survey (which I'm sure it's not the case for Burma), because floods hide the evidence under thick layers of sediment over time.

"But there's actually no need to postulate 'population replacement' anyway. If an incoming Neolithic group has more children that the local hunter gatherers the ratio of the two will change".

Nobody denies that Neolithic communities have some greater potential for expansion, if they do not end up dead by spearing. The question is how did it happen in reality, specially upon a time when the distinction between primitive Neolithic and advanced Paleolithic was not so big. Would not foragers, on light of dwindling resources and higher Neolithic productivity, gradually turn into farmers/herders before the few colonists have time to really expand? They do nowadays, even if reluctantly, so I don't see why they would not in the past. And in many cases archaeological and genetic evidence supports this transition, which, of course, took many generations.

In the SEA case, I think that Karafet's interpretation is much more sound than what you and Bellwood seem to claim.

"Some researchers believe that it [Y-hap O3] first appeared in China approximately 10,000 years ago".

Obsolete claim. When it was established that O3 is like 30 Ka old (several authors) and most diverse in the south than in the north, the whole simplistic scheme fell to pieces. However if by "China" they mean somewhere in Southern China they may be somewhat close to reality because at the moment I am not sure if it coalesced in Indochina or South China, though the high diversity of O1 and O3 itself in island SEA is rather suggestive of an Indochinese origin maybe.

Whatever the case, it happened long ago in the late phases of the main Eurasian expansion of humankind, not recently.

"Haplogroup O2 is also notable for the fact that it can be divided into two major subclades that show almost completely disjunct distribution".

What's the problem with this?

""The peak frequency of Haplogroup O1 is found among the aborigines of Taiwan"...

But it's not really diverse.

""The distribution of Daic languages in Thailand and other parts of Southeast Asia outside of China has long been believed, for reasons of traditional linguistic geography, to reflect a recent invasion of Southeast Asia by Daic-speaking populations originating from southeastern China"

So, again, its not SE Asian in origin but it may have arisen from the southern end of the Chinese Neolithic".

This cultural and political (not demic!) expansion is clearly post-Neolithic, same as Burmese one. It's proto-historical in fact.

Ebizur said...

The O*-M175 Y-DNA from Wells et al. 2001 is not O*-M175 in the strict sense, but rather O-M175 excluding all known subclades for which Wells et al. have screened their samples. In this case, the relevant individuals (most of whom are Iranic, Turkic, or Korean speakers as far as I recall) probably should be O2b, O2*(xO2a, O2b), or even O1*(xO1a).

terryt said...

"This model is oversimplistic and highly unlikely in all cases".

That is simply your belief, and it is your belief simply because you wish it so. Anyway, as I said, there's no need to postulate population replacement. Just more numerous or more fecund immigrants. You even admit as much:

"Nobody denies that Neolithic communities have some greater potential for expansion"

OK. We now agree.

"He seems to have some confused concept of what Indochina means, btw, because it includes all mainland SE Asia south of the Chinese border, and not just former 'French Indochina'"

No. It's you who are confused. The only parts of mainland SE Asia that are not 'former French Indochina' are Thailand, Burma and the Malay Peninsular. All the remainder was French untill WWII. Laos, Vietnam and Cambodia.

"Not a word on the islands between Malaysia and Papua, which are the main issue here. In fact, he is implying that Austronesians could expand into island SEA because there were no (or very few) farmers".

Probably because that was in fact the case. Most of the islands 'between Malaysia and Papua' are today populated by Austronesian-speaking people, amoung whomm Y-hap O1 is very common: the mark of the Austronesians.

"However there are other possible explanations and there are indications that farming pre-dates Austronesian arrival in all or most of these islands".

True. The Hoabinhian. But the population on the islands (as opposed to the mainland) was probably not huge, and so the islands were easily populated by the more agriculturally oriented Austronesians.

"That's only Neolithic (again cross-reasoning through the ages!)".

Yes. And, in spite of your belief, they were basically the first there in any great numbers.

"Anyhow the data for lowlands, specially flood basins, is likely to be incomplete"

Perhaps so. But in the region we're concerned with the flood basins were unlikley to have been heavily populated before the Neolithic. The rainforest there is much denser and more impenetrable than any forest you might have seen in Europe.

"Would not foragers, on light of dwindling resources and higher Neolithic productivity, gradually turn into farmers/herders before the few colonists have time to really expand?"

Quite likely be absorbed, but only after the original 'few colonists' had expanded considerably.

"They do nowadays, even if reluctantly"

And they usually form a minority in the resulting population. Exactly the situation we find today in SE Asia between haplogroup O on the one hand and haplogroups C, F and K on the other.

"Obsolete claim".

I knew you wouldn't like it. It conflicts with your belief. However I notice you don't hesitate to call on Wiki when it supports your belief.

"I think that Karafet's interpretation is much more sound than what you and Bellwood seem to claim".

Only because of your pre-existing belief. Bellwood has been working in various parts of SE Asia and Australia much longer than has Karafet.

"In this case, the relevant individuals (most of whom are Iranic, Turkic, or Korean speakers as far as I recall)"

Not SE Asian? Amazing. tends to suggest that it's not just the expansion of Y-hap O1 and the Daic languages that is 'clearly post-Neolithic, same as Burmese one'. In other words all the O expansions are probably 'proto-historical in fact', or just a little earlier.

Ebizur said...

Wells et al. 2001
O-M175(xO1a-M119, O2a-M95, O3-M122)

14/45 = 31.1% Korean

1/16 = 6.3% Tajik/Dushanbe, Tajikistan

2/41 = 4.9% Uighur/Kazakstan

1/22 = 4.5% Crimean Tatar/Uzbekistan

1/40 = 2.5% Tajik/Samarkand, Uzbekistan

1/54 = 1.9% Kazak/Kazakstan

1/68 = 1.5% Uzbek/Surkhandarya, Uzbekistan

1/70 = 1.4% Uzbek/Khorezm, Uzbekistan

The frequency of O2b-SRY465 in published studies' Korean samples averages 30%, so all fourteen O-M175(xO1a-M119, O2a-M95, O3-M122 individuals in Wells' sample of Koreans probably belong to haplogroup O2b.

I recall that there is a Mr. Zhappas from Kazakhstan who has been posted on ySearch as a member of haplogroup O2. The Zhappas clan is a subgroup of the Baiuly, which is in turn a subgroup of the Alshyn, the westernmost people group that has conglomerated with other groups living to their east to form the modern Kazakh ethnic group. Geographically, the Baiuly traditionally have inhabited the land along the lower reaches of the Ural River, with a few clans living further south along the eastern side of the Caspian Sea or further west in the so-called European part of Kazakhstan. Even among the Alshyn clans, Zhappas is supposed to be the most (or one of the most) westerly, so I am at a loss to explain the origin of this man's Y-DNA. Based on Mr. Zhappas' Y-STR haplotype, can anyone determine whether he most likely belongs to O2a, O2b, or O2*(xO2a, O2b)?

Maju said...

"Anyway, as I said, there's no need to postulate population replacement. Just more numerous or more fecund immigrants".

Erm. If there were, at some space-time, like 5000 years ago in Mediterranean Iberia/France, tens of thousands of hunter-gatherers and a few hundreds at best of colonists, then all the initial demographic advantage would take too long to have a strong effect. The actual reality (apparent in both the archaeological and genetic record in the example) is that natives became agriculturalist before the few colonists could really build their numbers.

This is somehow natural for most scenarios in Early Neolithic, when farming was still primitive and had rather low productivity. Only where there were very low densities of hunter-gatherers, as in the Balcans, could colonists build up enough numbers.

Primitive farming could not probably duplicate numbers each generation but surely had a much more modest vegetative growth rate. Let's say 1.2% per generation.

At this rate, in 17 generations (some 500 years), they would have grown (if the ecology allowed it, what I doubt) up to (roughly) 25 times the original number. 100x25=2500, still a minority.

This is of course a mere "hat trick", because no real conditionants, such as ecological limits, competition between various groups, including foragers and recycled farmer natives are considered. The natural process of admixture between colonists and native (farmers primarily) is not considered either.

The dates of culturally native Cardium Pottery sites are not, AFAIK, really much younger than those that can be interpreted as true colonies, which were anyhow concentrated in some specific areas (South Valencian Country particularly in SW Europe, plus a few scattered colonies elsewhere maybe). So we are in this case before a quite clear example of, mostly, native "aculturation" or better "cultural transformation" with the advent of Neolithic.

"No. It's you who are confused".

No. What I say is that Indochina (mainland SE, Indochina peninsula) is not the same as "French Indochina" but includes also Thailand, Burma and peninsular Malaysia. In this Bellwood falls in the usual error of believing that "Indochina" means "former French Indochina" only, what shows some geographic ignorance, acceptable for a teenager or an uncultivated person but not for a scholar.

For clarity: he never uses the term "French Indochina" but just "Indochina" and then adds "Thailand, Burma..." as if these were not part of Indochina (they are). He shows there his ignorance of geography and history. As simple as checking any encyclopedia but nope...

If you do not understand such basic concepts, how can you pretend to unravel the prehistory of a region you are unable to name or describe properly?!

Maju said...

"But the population on the islands (as opposed to the mainland) was probably not huge"...

This would be an interesting issue to clarify. Today Indonesia is very densely populated, specially Java, but what do we know of the Prehistoric densities? Probably not much.

If what you claim here would be true, then you would have a point, but I doubt you can demonstrate it because we know very little of the material prehistory of the region (very poor archaeological record so far).

So why do you make a claim you cannot demonstrate or even back with data at all?

"the flood basins were unlikley to have been heavily populated before the Neolithic".

Another totally unjustified claim. Why would river basins, some of the richest environments, be depopulated? It only seems to me an absurd claim with no support at all.

"And they usually form a minority in the resulting population".

You are comparing across ages again here (early Neolithic/latest Paleolithic vs. the Industrial Era). Primitive farming was not at all that productive.

Anyhow we see clearly that even Iron Age peoples such as Bantus advancing on forager areas, absorbed loads of the local genetics. Just recently it has been found that Mozambicans are more native than Bantu, possibly a Twa-specific component (Great Lakes' Pygmies). And we are talking of peoples equipped with all the Iron Age technologies, the same that have predominated everywhere until a few centuries ago, technologies able to speed up deforestation and to very effectively farm the lands (much better than with digging sticks and flint sickles) and of course able (not just by technology but also by political organization) to cause death and slavery at relatively large scale (though the real militarist degeneration of some Bantu groups only happened under the Slave Trade peculiar conditions, it seems).

Anyhow, if you look at Patin's autosomal structure, you clearly see that Mozambicans are at least 50% pre-Bantu, with many being up to 90% that.

I could continue but not really worth it.

Maju said...

@Ebizur: what percentage of that same O* category is found in Mongols and other related NE Asian "micro-Altaics" (specially Tungusic peoples)?

I understand that, if those O* of Central Asia have a Turco-Mongolic origin, they should be found at meaningful levels in Mongols and possibly Tungus, additionally to Koreans.

Anyhow, in the Joshung Jung paper, there was some hint that Koreans might be central to the formation of NE Asian peoples. Difficult to judge because the oversampling of (South) Koreans could well be distorting the results but it's a hypothesis I'd like to explore in the context of (probably) "coastal migration" in the peopling of East Asia.

terryt said...

"Today Indonesia is very densely populated, specially Java".

But the very dense population is entirely the product of agriculture, and that's a product of the Neolithic. Besides which who is now 'comparing across ages again here (early Neolithic/latest Paleolithic vs. the Industrial Era)'.

"but what do we know of the Prehistoric densities? Probably not much"

No, not much. But the reason for that is most likely because it was very sparsely settled.

"Why would river basins, some of the richest environments, be depopulated? It only seems to me an absurd claim with no support at all".

Try living in one for a while.

"Primitive farming was not at all that productive".

We're talking about intensive rice cultivation here, which is very productive.

"For clarity: he never uses the term 'French Indochina' but just 'Indochina' and then adds 'Thailand, Burma...' as if these were not part of Indochina (they are). He shows there his ignorance of geography and history. As simple as checking any encyclopedia but nope..."

So what's your problem? Just because he uses 'Indochina' as shorthand for 'French Indochina'?

"acceptable for a teenager or an uncultivated person but not for a scholar"

You're obviously totally unfamiliar with Peter Bellwood and his work.

"If you do not understand such basic concepts, how can you pretend to unravel the prehistory of a region you are unable to name or describe properly?!"

So you really do know more about the region than Bellwood. Why am I not surprised? I'm obviously wasting my time so we should terminate this discussion.

"all fourteen O-M175(xO1a-M119, O2a-M95, O3-M122 individuals in Wells' sample of Koreans probably belong to haplogroup O2b".

Very likely, I agree.

Maju said...

"But the very dense population is entirely the product of agriculture, and that's a product of the Neolithic".

How do you know? Maybe Java and nearby areas were also highly productive in the Paleolithic. Anyhow, Neolithic is pre-Austronesian in Java almost for sure and the "Asutroasiatic" genetic component found by the HUGO paper in that zone clearly supports this idea.

"No, not much. But the reason for that is most likely because it was very sparsely settled".

Probably not. The reason is mostly that developing countries do not have many resources to spend in archaeology and additionally there may be a Muslim prejudice against archaeology and prehistory, because only what happened since Mohammed really matters. This last at least is very apparent in North Africa, where this way of thinking is used to hide the pre-Islamic and pre-Arab history of the region, but it's very possible it also weights in other Muslim countries like Indonesia.

I find just absolutely striking that South China is producing some very informative data and that SE Asia is not. Some other countries like Thailand, Vietnam or Philippines have also produced some archaeological record but others not at all. This is not because the modern political borders had any relevance in the Paleolithic or Neolithic. It is because archaeological research, very underdeveloped in general, it is not being taken seriously by the respective governments and only now and then some rare findings happen, mostly at the hands of foreign researchers (except in China).

"Try living in one for a while".

LOL, I live at a river basin. Nearly everybody does: it's a natural pattern of human settlement: rivers and coasts are generally the best places to live, as they concentrate most natural resources.

"We're talking about intensive rice cultivation here, which is very productive".

You are talking of that maybe. If so, it'd be a second Neolithic wave and not the original one. Whatever the case it can only be that much productive. IDK about rice farming but in Medieval Europe (and that is very advanced "Neolithic", with heavy plow and crop rotation), farmers could only produce that much. The surplus can be easily quantified by how many made up the non-farmer castes/classes, which were never more than 10% of the population and often many less. This is the kind of surplus farming produces: small, specially as population grows and less desirable fields are put to produce (some places are obviously better than others, sometimes much much better).

"So what's your problem? Just because he uses 'Indochina' as shorthand for 'French Indochina'?"

That he does not know what Indochina means. That's clear ignorance.

"So you really do know more about the region than Bellwood".

At least I know what Indochina is.

terryt said...

Just to check the possibility that you may be correct after all I’ve tracked down the paper you mention that claims Y-hap O3 is at least 30,000 years old, and moved north from SE Asia:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1226206/

Some interesting statements stand out immediately:

“we estimated the ages of the major O3-M122 subhaplogroups by use of the coalescence method developed by Zhivotovsky”.

I seem to remember you maintaining that this method under-estimates age. So can I presume you would place the beginning of O3’s diversity at more like 50,000 years?

“The O3-M122 has the highest frequency (41.8% on average) (fig. 2) in East Asians, especially in Han Chinese (52.06% in northern Han and 53.72% in southern Han)”

Hmmm. Does that suggest that the various Y-hap O3s actually expanded from the Han homeland, or does it suggest that O3 is an immigrant, from the south?

“The Yangtze River was used as the geographic border to separate the SEAS and NEAS. In the SEAS, there are 14 Tibeto-Burman–speaking populations with a recorded history of migration from northern China ~3,000 years ago”

So the Tibeto-Burman-speaking people cannot be used to represent the original SE Asian population, although we must concede that the language could have spread without any haplogroup movement.

“the Hmong-Mien populations have a recorded history of admixture with Han populations, although they are often considered southern populations”

So it seems that the Hmong-Mien-speaking people cannot be used to represent the original SE Asian population either. A large proportion are immigrants.

“The southern Han populations were recent northern immigrants because of the expansion of Han culture in the past several thousand years”

But surely the Han probably simply represent a more recent strand of what has been a reasonably long-term southward movement, going back to the Yangtze/Yellow River Neolithic. We’ve just seen two others.

“To remove the influence of relatively recent population admixture, we constructed the STR network excluding the Tibeto-Burman, Altaic, Hmong-Mien, and southern Han populations”

So, to make sure they procured as representative sample of Y-hap O3s as possible the authors carefully eliminate the populations they don’t like, the ones that have definitely moved from north to south. Wouldn’t that rather skew their results in a particular direction?

“the Hmong-Mien populations were clustered closely with Han populations, which reflects the recorded history of admixture”

Or a common origin. But the authors have already eliminated the Hmong-Mien anyway. And they even shoot themselves in the foot:

“recent gene flows due to the expansion of Han culture could also have contributed partly to the homogeneity of the O3-M122 lineage, although Daic and Austro-Asiatic populations had much less influence from Han populations than did the Tibeto-Burman and Hmong-Mien populations”

But there is a good possibility that the Daic and Austro-Asiatic people originate from yet another early pre-Han movement from the Yangtze/yellow River Neolithic. And yet the authors are still really certain that the assortment of Y-hap O3s that are today spread all through East Asia cannot possibly have moved south, only north. Isn’t that perspective ideologically rather than evidentially based?

(cont.)

terryt said...

“we argue that the time of the initial northward migration should be close to the estimated divergence times of the O3-M122 sublineages”

Wouldn’t that have led fairly rapidly to separate northern and southern haplogroups? Rather, what we see is evidence for a more recent common origin for the combined northern and southern O3 population. Because:

“In general, the distribution of the O3-M122 haplotypes did not show distinctive divergence between southern and northern populations, with all the major subhaplogroups shared between them”

And:

“Using the STR data, we calculated the gene diversities; no significant differences were observed between SEAS and NEAS or among different language groups (data not shown). The AMOVA analysis did not show significant between-group divergence either”

How could that be if the south to north migration occurred at least 30,000 years ago? Surely the fact that all O3 haplogroups are widespread shows that any Y-hap O3 expansion, from anywhere, in any direction, must be more recent than 30,000 years ago. Or 50,000 years ago.

“many shared STR haplotypes were observed between northern and southern populations, again confirmation of the recent common ancestry of the M122 lineage in East Asia. It has been well documented that the Tibeto-Burman populations living in southwestern China were originally, during the late Neolithic period, from the north, but they have been under extensive influence from the southern ethnic groups, including Daic- and Austro-Asiatic–speaking populations”

They’re really bending over backwards to make the evidence fit their theory here, aren’t they. But if we accept that the expansion was relatively recent, and southwards, all these convolutions become unnecessary. The authors even admit:

“On the other hand, in the past 5,000 years, the major population migration in East Asia is from north to south, because of the expansion of Han culture”

As with older expansions from the Yangtze/Yellow river Neolithic.

“the MDS analysis showed that the NEAS are closely related by clustering together, whereas the SEAS showed relatively loose connections with larger variance, indicating that SEAS are genetically more polymorphic than are NEAS”

Is that because an originally diverse O3 population has become separated into subunits as they have moved south through the mountains and up and down the rivers?

“therefore supporting the hypothesized southern origin of modern humans in East Asia”

Ahh. So that was their objective all along.

Maju said...

"I seem to remember you maintaining that this method under-estimates age".

Possibly but it is still much better than the brute-force use of pedigree rates. I just contend that there is a lot of uncertainty in these methods of genetic age estimate based because they need to assume a lot of preliminary parameters that are in no way clearly known, not just mutation rate but population sizes and their dynamics, the divergence time for Pan-Homo, etc.

In general I think the Zhivotovski method still falls somewhat short but this may vary depending on the specific parameters and samples the researchers use. It's not rocket science, genetics may be close to that but age estimation based on the molecular clock is not even remotely precise.

"So can I presume you would place the beginning of O3’s diversity at more like 50,000 years?"

In this particular case I have no particular reason but yes, it's possible. But "at least 30 Ka" is enough for our discussion.

"Hmmm. Does that suggest that the various Y-hap O3s actually expanded from the Han homeland, or does it suggest that O3 is an immigrant, from the south?"

Southern Han are not true Han, not more than Andalusians are Castilian or Occitans are French: they have been assimilated. Still Karafet, for instance seems to suggest a SE Asian origin for O3. This is somewhat supported by the high diversity of O3 in general in (mainland) SE Asia (0.511) and the even higher diversity of O in Indonesia (0.597 and 0.547) and Indochina itself (0.550).

In any case there's no haplogroup I know of that can reflect a possible demic expansion of Han southwards. This haplogroup(s) would need to be almost restricted to nominally Han peoples (Sinitic speakers) and show much greater diversity in the Yellow River area. No haplogroup does that, AFAIK, so there's no clear signature of any Han demic expansion (what does not mean there were not localized colonies here and there but always as minority among other peoples, whom they gradually assimilated, within the context of Chinese imperial expansion).

All haplogroups cross the Han linguistic/ethnic barrier and all haplogroups, including O3 seem more diverse towards the South.

...

Maju said...

"So the Tibeto-Burman-speaking people cannot be used to represent the original SE Asian population, although we must concede that the language could have spread without any haplogroup movement".

Hmmm. It's "doctrine" in the Chinese Academy (and used to be in Western circles too) that TB "peoples" (languages) expanded from the North, because they are supposedly related with Sinitic. However there's some controversy on whether Sinitic and TB are in fact related or not. Whatever the case, I'd say that the core of TB peoples (and maybe pre-Sinitic too) was at Sichuan, what is more like West China or SW China than properly North.

"So it seems that the Hmong-Mien-speaking people cannot be used to represent the original SE Asian population either".

Their exclusive autosomal component are certainly quite unique. Other sources I have read (which I doubt but just for the record) consider them as another Sino-Tibetan offshot. In general I do consider actual SE Asian peoples, including southern Sinitic family speakers as pretty much descendants of the ancestral inhabitants... unless it can be demonstrated otherwise. There's a lot of speculation and controversy in these speculative origins.

“The southern Han populations were recent northern immigrants because of the expansion of Han culture in the past several thousand years”

That's not demonstrated at all. In fact there's a lot of data suggesting the opposite. This is nothing but traditional Chinese national doctrine, you can skip these opinions and go to the hard data... and try to interpret it without any of those prejudices.

"So, to make sure they procured as representative sample of Y-hap O3s as possible the authors carefully eliminate the populations they don’t like"...

That's arguably a valid criticism on your side. Notice anyhow that by doing that, the authors are only measuring age of O3 among Northern Han, which must be more recent than overall.

"But there is a good possibility that the Daic and Austro-Asiatic people originate from yet another early pre-Han movement from the Yangtze/yellow River Neolithic".

Wild meaningless speculations on your side. You and the Chinese Academia seem to have decided that everything is North to South and are totally ignoring the data and repeating old clichés which are false with all likelihood.

Maju said...

"Wouldn’t that have led fairly rapidly to separate northern and southern haplogroups?"

I'd say the pattern represents a continuous flow from SE Asia towards Central East Asia (North China, Korea, Japan) but not further north, where other lineages dominate instead. Maybe with some backflow added. It's similar to what we see in East Asian mtDNA R (B, F and related lineages) and that's why I make an association between the expansion of Y-DNA NO or O and that of mtDNA R in East Asia. Hence this probably happened in the late sub-phase of the Great Eurasian Expansion (c. 60-40 Ka ago possibly).

"How could that be if the south to north migration occurred at least 30,000 years ago?"

Because the expansion of O3 (not of O) seems to represent, together with that of O2b the northern frontier of that expansion. You have to look at the whole picture and, anyhow, if it doesn't necessarily implies a S->N distribution, it does not imply a N->S one either.

"They’re really bending over backwards to make the evidence fit their theory here, aren’t they".

Yeah, ignore those rantings about populations migrating southwards, they are nothing but mere speculations for the greatest part. Classical speculations but a-scientific ones anyhow.

I think that you are confused in part because you give too much importance to the "literature" within these kind of papers and books, instead of focusing on the hard data and the solid conclusions where these exist. In fact I decided to learn more (and more and more) about Prehistory for a similar reason: I read opinions which were often contradictory, so I felt a real urge to gather the facts and make up my mind.

After all the average IQ is barely above 100 if you know what I mean, and that includes academics. So it's not worth for me to accept others' opinions at face value. So I tend to ignore all that literary and speculative ranting.

terryt said...

You're certainly not one to let the evidence interfere with your prejudices.

"ignore those rantings about populations migrating southwards, they are nothing but mere speculations for the greatest part".

So you're extremely happy to accept the parts of the paper you agree with, but your pre-existing belief necessitates your ignoring the bits that don't fit. Regarding the southern Han populations being recent northern immigrants you exclaim:

"That's not demonstrated at all".

You have got to be joking.

"You and the Chinese Academia seem to have decided that everything is North to South"

Maju. Do you really have an IQ of more than 50? Did you not notice that all the authors but one are Chinese? Most from K'unming in fact. South China. Surprise, surprise. I'll grant their perspective has influenced their interpretaion of the data, but they certainly argue for a south to north movement.

"That's arguably a valid criticism on your side. Notice anyhow that by doing that, the authors are only measuring age of O3 among Northern Han, which must be more recent than overall".

No they're not. They're interested in the age of O3 without them. They are eliminating them so they don't interfere with their southern origin theory.

"you give too much importance to the 'literature' within these kind of papers and books, instead of focusing on the hard data".

So once again you are the only person in the world who knows the truth. And you are in fact able to convince yourself of this by ignoring 'the hard data'.

"Because the expansion of O3 (not of O) seems to represent, together with that of O2b the northern frontier of that expansion".

So why on earth would they have stopped at that frontier? Is it perhaps that that is the northern limit for their early agriculture?

"Southern Han are not true Han"

But they have almost exactly the same haplogroups as the northern Han. So in what way are they not 'true Han'?

"In any case there's no haplogroup I know of that can reflect a possible demic expansion of Han southwards".

Of course not, because you refuse to see it.

"All haplogroups cross the Han linguistic/ethnic barrier"

Yes. Isn't that an amazing fact. The Han movement is well-attested, yet you see no evidence for it.

"In general I do consider actual SE Asian peoples, including southern Sinitic family speakers as pretty much descendants of the ancestral inhabitants... unless it can be demonstrated otherwise".

OK. Let's see if we can find any typical southern Y-hap O3s. I presume you'd agree with the authors' statement, 'Daic and Austro-Asiatic populations had much less influence from Han populations than did the Tibeto-Burman and Hmong-Mien populations', or any other Chinese population you care to nominate. So it should be a simple matter to look at the Daic and Austro-Asiatic haplogroups, and determine which ones are different from the Han (northern Han if you insist). From that it should then be a simple matter to calclate the level of Han introgression into the Hmong-Mien, not to mention any other groups. Why didn't they do it?

"Hence this probably happened in the late sub-phase of the Great Eurasian Expansion (c. 60-40 Ka ago possibly)".

There's actually a huge problem if you're going to argue that the Y-hap O3 movement from SE Asia dates back to any more than something like 10,000 years ago. The authors even express their own surprise that there is no evidence of bottlencks, which there surely would be if the scenario you (and they) propose is correct. Virtually all O3 haplogroups are spread widely. Surely that must indicate a relatively recent, and widespread, expansion. Not some Paleolithic one.

Maju said...

"Did you not notice that all the authors but one are Chinese?"

That's precisely what I meant: they are echoing the unproven (and unlikely) but traditional beliefs of Chinese historiography, which in turn is somewhat based on Western historiography and anthropology from a century ago or so.

"No they're not. They're interested in the age of O3 without [Northern Han]".

That's what you said, quoting them:

“To remove the influence of relatively recent population admixture, we constructed the STR network excluding the Tibeto-Burman, Altaic, Hmong-Mien, and southern Han populations”

Excepting Altaic all these are Southern populations! But they seem to have included some other Southern groups? I see now that there are also some Daic and Austroasiatic samples. They did consider Northern Han in any case.

Anyhow, take a look at (phylogenetically sorted) haplotype structure in fig. 6. O3a* (M324*) is the only paragroup that may have a "Han" root, which shows as Southern Han when all populations are considered and as Northern Han with the restricted set. O3* looks rooted among Southern Han in the all-samples tree and between Southern and Northern populations in the restricted sample.

So guess it's pretty safe to conclude that O3 originated in South China and it being the northernmost of all three O basic subclades.

"So why on earth would they have stopped at that frontier?"

It's a good question. While I don't have a definitive answer I do guess it has something to do with ecosystems and cultural adaptation to such climatic constraints. Beijing was the southernmost reach of permafrost in East Asia (at the LGM) and we know by the genetic signatures (such as the star-like structure of mtDNA D4 and A) that the peoples of NE Asia experimented a sudden expansion at some point. This left NE Asia pretty much in the hands of Y-DNA haplogroups C3, N and Q. C3 is surely an older arrival than the O3 expansion and may have been pushed by the pressure of the "O clan", N is more like the northern branch of O (NO in fact) and Q is the intruder from the West probably. All these "clans" three must have got adapted to the special conditions of the Far North, an area that the "O clan" was not even interested in (but its "brother" N and "cousin" Q were).

"But [Southern Han] have almost exactly the same haplogroups as the northern Han".

Do they? Don't they actually have more like the same haplogroups as their Daic, Austroasiatic, TB and Hmong-Mien neighbors, more or less?

For instance in Fig. 6 it's clear that haplogroup O3a3b (M7) is shared by Southern Han and other SE Asians but not a single Northern Han shows up. So O3a3b is a clearly a non-Han (in the sense of non-Northerner) clade.

Other papers such as the HUGO research also showed that Southern Han cluster much more closely with other Southern peoples, specially the Kadai (Kradai, Daic). They do seem to have some less important amount of Northerner blood but that's about it.

"Yes. Isn't that an amazing fact. The Han movement is well-attested, yet you see no evidence for it".

The "Han movement" is not well attested. We know that the Chinese Empire expanded from, essentially, North to South and that such expansion involved instances of colonization but also of assimilation. These were probably dominant on light of the genetic evidence.

In the past there was a tendency to equate linguistic flow with genetic flow but that was only because little was known of genetics and people get weird simplistic ideas of that kind.

"OK. Let's see if we can find any typical southern Y-hap O3s".

O3a3b (M7).

terryt said...

"For instance in Fig. 6 it's clear that haplogroup O3a3b (M7) is shared by Southern Han and other SE Asians but not a single Northern Han shows up. So O3a3b is a clearly a non-Han (in the sense of non-Northerner) clade".

But even you must admit that O3a3b is several steps removed from O3*, and could well have coalesced in the south.

"To remove the influence of relatively recent population admixture, we constructed the STR network excluding the Tibeto-Burman, Altaic, Hmong-Mien, and southern Han populations"

Yes, they EXCLUDED them.

"I do guess it has something to do with ecosystems and cultural adaptation to such climatic constraints".

I'm certain that it does.

"All these 'clans' three must have got adapted to the special conditions of the Far North, an area that the 'O clan' was not even interested in (but its 'brother' N and 'cousin' Q were)".

And don't forget C. Surely the presence of O's brother N to the Far North to the exclusion of O suggests (although I admit not necessarily so) that the four haplogroups (N, O1, O2 and O3) originated from somewhere near each other.

"Do they? Don't they actually have more like the same haplogroups as their Daic, Austroasiatic, TB and Hmong-Mien neighbors, more or less?"

I found an intersting article that I'll post separately.

terryt said...

"The 'Han movement' is not well attested".

That's not what these people say ('Genetic evidence supports demic diffusion of Han culture'):

http://159.226.149.45/compgenegroup/paper/wenbo%20han%20culture%20paper%20(2004).pdf

Admittedly written by a group of Chinese, so I've no doubt you'll disagree. But some quotes:

"The Han people, like East Asians are divided into two genetically differentiated groups, northern Han and southern Han, separated approximately by the Yangtze river"

Surprisingly near to part of the Yangtze/Yellow river Neolithic.

"However, the substantial sharing of Y-chromosome and mitochondrial lineages between the two groups and the historical records describing the expansion of Han people contradict the cultural diffusion model hypothesis of the Han expansion"

From here on I'll confine the extracts to Y-haps, which is what we have been disagreeing over:

"In addition, haplogroups O1b-M110, O2a1-M88 and O3d-M7, which are prevalent in southern natives, were only observed in some southern Hans (4% on average), but not in northern Hans"

So they have at least endevoured to separate 'native' SE Asian haplogroups from northern ones. Note, however, they don't consider Y-hap O2b, which is a northern version of O2. And remember the authors are considering only Han movement, no consideration of possible pre-Han movement and the general similarities of northern and southern Y-haps.

"The results of analysis of molecular variance (AMOVA) further indicate that northern and southern Hans are not significantly different in their Y-haplogroups (Fst=0.006, P>0.05) demonstrating that southern Hans bear a high resemblance to northern Hans in their male lineages"

"For the Y-chromosomes, all southern Hans showed a high proportion of northern Han contribution"

"We provide two lines of evidence supporting the demic diffusion hypothesis for the expansion of Han culture. First, almost all Han populations bear a high resemblance in Y-chromosome haplogroup distribution, and the result of principal component analysis indicated that almost all Han populations form a tight cluster in their Y-chromosomes. Second, the estimated contribution of northern Hans to southern Hans is substantial in both paternal and maternal lineages and a geographic cline exists for mtDNA"

Note, no cline for Y-haps.

"A sex-biased admixture pattern was also observed in Tibeto-Burman-speaking populations"

So there goes your theory. Finally:

"Our genetic observation is thus in line with the historical accounts"

So they interpret the similarity of northern and southern Han Y-haplogroups as being the result of recent (in the last 2000 years) movement. How is it possible to reconcile the wider similarity between northern and southern Y-haps with a Paleolithic expansion?

Maju said...

"But even you must admit that O3a3b is several steps removed from O3*, and could well have coalesced in the south".

Absolutely. But as far as I can see there is no lineage that must have coalesced in the North. This paper anyhow fails to provide enough resolution for the rest of the haplogroup O3, only the two mentioned, which are quite large, are dealt with any detail and both look southerner.

"Yes, they EXCLUDED them".

But they included Northern Han!

And anyhow, you still have the full STR network side by side in the same graph 6. They just double-checked.

Can you bother looking at fig. 6?!

"And don't forget C".

I did not forget C3, which is the only C haplogroup in mainland NE Asia. I explicitly mentioned it in my previous post, saying roughly that I am under the impression that C3 is somewhat of an older layer (together with unmentioned D and C1) that was displaced in the late Eurasian Expansion by "NO clan" people (which should be essentially the same as the mtDNA "B+F clan", plus some other mtDNA haplos like maybe N9).

"Surely the presence of O's brother N to the Far North to the exclusion of O suggests (although I admit not necessarily so) that the four haplogroups (N, O1, O2 and O3) originated from somewhere near each other".

I'd try to be faithful to the phylogenetic hierarchy, hence N compares with O as a whole and not with its subclades. The NO bifurcation happened surely in SE Asia (possibly in southernmost China but unsure) because N is also most diverse towards the South, AFAIK, even if it's relatively rare over there.

So both O and N look southern East Asian by origin, with N taking the primary role in expanding northwards within this haplogroup NO and O instead being the leader locally at the original zone.

I'm unsure if O1, O2 and O3 can be compared directly with anything under N, as the phylogenies are not really looking parallel at all and possibly their timings were also somewhat different after they parted ways. So I'll focus on O.

Maybe in parallel to N migrating northwards, O split into its three branches: O1 spread through SE Asia, O2 along the coast and O3 more generically (i.e. probably at the interior). Without a good archaeological record for Chinese and SE Asian Paleolithic we can't really reconstruct how or when these events happened. But for me it's pretty clear that O replaced pre-existent "clans" (D, C1 and C3 in the Y-DNA record), what implies a major demic replacement that could only happen in a Paleolithic context, specially considering the TMRCA produced by this paper (and others, I believe).

Alternatively one could imagine that C (C1 and C3) spread more or less along the NO expansion but what about D? Maybe they are just random founder effects at peripheral areas, where the dominant position of O would not drift them out but the pattern is suggestive of at least two waves in the MP-UP Paleolithic timeline of Eurasian Colonization.

This double wave can also be perceived to some extent in mtDNA, I believe.

If so, the expansion of D and C would correlate best with the earliest expansion starred by mtDNA M subclades (M7 and D specially), while the NO expansion would correlate well with that of R derivatives: R11'B7, B4'5 and R9 (pre-F) - and surely also with N9.

Curiously the same pattern is found in Sahul, where Australia, that has no mtDNA R also has virtually no Y-DNA MNOPS (or K), while Melanesia, that has strong mtDNA R (P) also has abundant Y-DNA MNOPS in the form of M and S.

And the timing must be almost the same: one coding mutation downstream of the mtDNA R node. You will never claim that Papuan Y-DNA M and S (and other MNOPS*) are "Neolithic", then why dare you claim that the parallel expansion of NO in East Asia is?

Look at the whole picture, please.

Maju said...

"written by a group of Chinese, so I've no doubt you'll disagree".

I have absolutely no reason to disagree with researchers just because they are Chinese. I do disregard some ideas that some of them (and not just Chinese, you are a clear example) declare thoughtlessly based only on old school historiography. Luckily not all Chinese authors fall on that trap, specially not as the dates of publication approach the present, because it's becoming more and more obvious that the overall pattern of genetic spread in East Asia is from South to North and largely (not only) along the coast, what clearly dismantles the old held beliefs to a large extent.

I am not going to discuss all the sentences because it's tiresome and mostly pointless. What I want you to do is to look at the PC graphs (almost the only relevant info in that letter, not really a paper) and realize that almost 50% of the haploid variation goes along the horizontal axis (48% for Y-DNA) and only 13-23% goes in the vertical axis.

So half of the variation does not align across the Han/non-Han axis but across some other axis that could well be said to run between Hmong/Austroasiatic peoples and Daic ones with Han, and specially Southern Han, aligning more with the first group than the later one. The clustering of some South Han with North Han happens indeed but one would need to know which are the samples because some South Han are more like North Han than others on light of what I have seen so far. Some other South Han instead cluster with Hmong-Mien and look not much related even by Y-DNA to North Han.

The situation is less clear for mtDNA. Here there's no clear structure in PC1, which is polarized by two South Han individuals (and then by two Hmong-Mien ones). PC2 shows (only 13%) a Northern Han pole but still most South Han cluster with South ethnics and those who do not are intermediate.

Anyhow I'm not sure if PC analysis has any real use in haploid genetics because you have to assign arbitrary values to haplogroups, often regardless of their phylogeny (but based on geography and such). It'd be much more useful to make a good comprehensive analysis of all Y-DNA and mtDNA haplogroups at depths as good as possible through all East Asia, from Wallacea to Chukotka. That would really help.

"So they interpret the similarity of northern and southern Han Y-haplogroups as being the result of recent (in the last 2000 years) movement. How is it possible to reconcile the wider similarity between northern and southern Y-haps with a Paleolithic expansion?"

First, this is not really clear from that letter, which lacks depth of analysis and can only be considered an exercise that must also be considered along with its 2004 timestamp (i.e. pretty old for this rapidly evolving field).

Second, I am not denying that the Chinese expansion southwards carried some colonization with it. I am just relativizing it to its proper dimension.

Third, the whole letter is written and designed in order not to explore the facts as much as to support a theory. Obviously they were already in 2004 facing more and more authors suggesting the south to north main flow pattern and minimizing the N->S Han one but this paradigm has grown in strength since 2006 and later, as the "coastal" migration model became dominant in our understanding of the peopling of Eurasia.

I'm sure that there are papers on these matters with more recent timestamps, and therefore more productive, hopefully.

terryt said...

"the whole letter is written and designed in order not to explore the facts as much as to support a theory".

That is exactly what you have been doing with every comment yuou've made on this post.

"I'm sure that there are papers on these matters with more recent timestamps, and therefore more productive, hopefully".

I invite you to try and find one, preferably one that doesn't manipulate the evidence in order to 'prove' a southern orign for Y-hap O3. I haven't been able to find one.

Maju said...

Look, I just stumble on that kind of data all the time. I was just reading this new paper on altitude adaptation of Tibetans and Andeans and, guess what, there's a structure graph of East Asian autosomal genetics in fig. 2 and again Han (not sure what kind of Han right now) cluster with SE Asians (all them) rather than with NE Asians, at least at K=3 (fig.2-D). And in the PC analysis (fig.2-B) again they cluster better with SE Asians than with NE Asians.

Han are, even the Northern ones, largely of SE Asian origin and affinity (even in Dienekes' craniometrics!)when compared with other populations of NE Asia. So you should not be surprised at all that they have a haplogroup that expanded from SE Asia somewhat (but not too much) more recently than the typical NE Asian lineages like D, C3 and C1.

This SEA origin and affinity of even Northern Han can confuse things but you just have to sample some other NE Asians to see their real place in East Asia overall.

Maju said...

And as for the paper, Hong Xi 2005, the paper you mentioned above is the only one I know that deals with the origins of O3. It concludes that is from the South and it has not bee contended AFAIK until today.

You may however be interested in visiting Quetzacoatl Anthropology Forums where there is some very knowledgeable people, most of them Asians. For example, I stumbled, searching for more stuff, with Ibra who is quite an expert, IMO, and someone who taught me some good stuff) HERE clearly arguing in favor of the "slow boat" hypothesis for the Austronesian migration to the Pacific islands, meaning that there was an episode of admixture at the Lapita Culture, when they got the C2a lineage that made this thread diverge so much from Gibraltar.

So yea, let's get focused: there are other threads about East Eurasia. This one is about West Eurasia, so please no more off-topic posts.

terryt said...

"You will never claim that Papuan Y-DNA M and S (and other MNOPS*) are 'Neolithic', then why dare you claim that the parallel expansion of NO in East Asia is?"

I don't. I claim it is Paleolithic. Our disagreement is where NO expanded from. I agree that MNOPS* is SE Asian, but it looks to me very much as if NO moved north and then N went north (with some small amount also moving south) and O moved south.

"Can you bother looking at fig. 6?!"

I have. And I see virtually no geographic differentiation. That surely fails to provide evidence for an ancient O3 expansion. In fact it supports the concept of a Neolithic expansion.

"I did not forget C3, which is the only C haplogroup in mainland NE Asia".

From here on you post is sprinkled liberally with 'could have' and 'maybe'. I agree that C and D represent an older layer. To me the haplogroup evidence for the whole of East Asia, from Japan to Australia, is most parsimoniously explained if we postulate that C and D were originally northern haplogroups, C1 and D reaching Japan quite early, leaving what became C3 behind somewhere. What became C2 and C4 moved down the eastern Eurasian coastline and crossed Wallace's line to reach Australia, missing New Guinea. And D moved south by a more inland route. Near Wallacea somewhere C met up with MNOPS*. From there MNOPS*'s descendants expanded. M, S and some K crossed Wallace's line in turn and colonised New Guinea and a few islands beyond. NO moved north, reaching what is today the border between China and Inner Mongolia. N moved into the higher country beyond, and O diversified in the region to the south, later moving further south again once the Neolithic had developed in the region. That leaves Y-hap P. To me it seems most likely that P moved west, into India, where it gave rise to R, and then P and R moved on north into south Central Asia, where P gave rise to Q and R1b developed. Q then moved east, north of Mongolia, eventually reaching America.

"This double wave can also be perceived to some extent in mtDNA, I believe".

So do I. Y-haps C and D arrived in Japan with what became N9 and carried the mtDNA haplogroup south to Australia, where it became mtDNA S. Along the way mtDNA R formed, which then expanded with various derivatives of Y-hap MNOPS*. MNOPS* had already carried several mtDNA M lines into the region. MtDNA P crossed Wallace's line with Y-haps K, M and S, as did mtDNA M derivatives, including Q. Other mtDNA R lines entered India, presumably with Y-hap P, where they became a significant element of the population, even though mtDNA M was already present there. Y-hap NO had carried several of these mtDNA M lines north with them into China.

terryt said...

"Han (not sure what kind of Han right now) cluster with SE Asians (all them) rather than with NE Asians"

So? Surely you realise that NE Asians are predominantly Y-haps N and C3, not O.

"so please no more off-topic posts".

OK. Last comment here. From me anyway.

Ebizur said...

Maju said,

"Ebizur: what percentage of that same O* category is found in Mongols and other related NE Asian "micro-Altaics" (specially Tungusic peoples)?

I understand that, if those O* of Central Asia have a Turco-Mongolic origin, they should be found at meaningful levels in Mongols and possibly Tungus, additionally to Koreans."

34/101 = 33.7% O2b-SRY465 Manchu (Katoh et al. 2004) [1/101 O-M175(xO1a-M119, O2b-SRY465, O3-M122), 3/101 O1a-M119, 43/101 O3-M122]

14/45 = 31.1% O-M175(xO1a-M119, O2a-M95, O3-M122)
Korean (Wells et al. 2001)

Manchu (Xue et al. 2006)
1/35 = 2.9% O1a-M119
3/35 = 8.6% O2*-P31(xO2a-M95, O2b-SRY465)
2/35 = 5.7% O2b-SRY465(xO2b1-47z)
6/35 = 17.1% O3-M122(xO3a3a-M159, O3a3b-M7, O3a3c-M134)
2/35 = 5.7% O3a3c-M134(xO3a3c1-M117)
5/35 = 14.3% O3a3c1-M117

5/35 = 14.3% O-M175(xO1a-M119, O2a-M95, O3-M122)
19/35 = 54.3% O-M175 total

Oroqen (Xue et al. 2006)
1/31 = 3.2% O-M175(xO1a-M119, O2-P31, O3-M122)
2/31 = 6.5% O2*-P31(xO2a-M95, O2b-SRY465)
2/31 = 6.5% O3-M122(xO3a3a-M159, O3a3b-M7, O3a3c-M134)
2/31 = 6.5% O3a3b-M7
1/31 = 3.2% O3a3c-M134(xO3a3c1-M117)
1/31 = 3.2% O3a3c1-M117

3/31 = 9.7% O-M175(xO1a-M119, O2a-M95, O3-M122)
9/31 = 29.0% O-M175 total

Daur (Xue et al. 2006)
1/39 = 2.6% O-M175(xO1a-M119, O2-P31, O3-M122)
2/39 = 5.1% O1a-M119
1/39 = 2.6% O2*-P31(xO2a-M95, O2b-SRY465)
1/39 = 2.6% O2b-SRY465(xO2b1-47z)
6/39 = 15.4% O2a-M95(xO2a1-M88)
7/39 = 17.9% O3-M122(xO3a3a-M159, O3a3b-M7, O3a3c-M134)
3/39 = 7.7% O3a3c1-M117

3/39 = 7.7% O-M175(xO1a-M119, O2a-M95, O3-M122)
21/39 = 53.8% O-M175 total

Nanai/PRC (Xue et al. 2006)
1/45 = 2.2% O2*-P31(xO2a-M95, O2b-SRY465)
2/45 = 4.4% O2b-SRY465(xO2b1-47z)
11/45 = 24.4% O3-M122(xO3a3a-M159, O3a3b-M7, O3a3c-M134)
2/45 = 4.4% O3a3c-M134(xO3a3c1-M117)
7/45 = 15.6% O3a3c1-M117

3/45 = 6.7% O-M175(xO1a-M119, O2a-M95, O3-M122)
23/45 = 51.1% O-M175 total

1/16 = 6.3% O-M175(xO1a-M119, O2a-M95, O3-M122)
Tajik/Dushanbe, Tajikistan (Wells et al. 2001)

3/60 = 5.0% O2b-SRY465 Uriankhai (Katoh et al. 2004) [4/60 = 6.7% O3-M122]

2/41 = 4.9% O-M175(xO1a-M119, O2a-M95, O3-M122)
Uighur/Kazakstan (Wells et al. 2001)

1/22 = 4.5% O-M175(xO1a-M119, O2a-M95, O3-M122)
Crimean Tatar/Uzbekistan (Wells et al. 2001)

2/60 = 3.3% O2b-SRY465 Zakhchin (Katoh et al. 2004) [5/60 = 8.3% O-M175(xO1a-M119, O2b-SRY465, O3-M122), 5/60 = 8.3% O3-M122]

1/40 = 2.5% O-M175(xO1a-M119, O2a-M95, O3-M122)
Tajik/Samarkand, Uzbekistan (Wells et al. 2001)

1/54 = 1.9% O-M175(xO1a-M119, O2a-M95, O3-M122)
Kazak/Kazakstan (Wells et al. 2001)

1/68 = 1.5% O-M175(xO1a-M119, O2a-M95, O3-M122)
Uzbek/Surkhandarya, Uzbekistan (Wells et al. 2001)

1/70 = O-M175(xO1a-M119, O2a-M95, O3-M122)
1.4% Uzbek/Khorezm, Uzbekistan (Wells et al. 2001)

0/26 O-M175(xO1a-M119, O2a-M95, O3-M122) Evenk/PRC (Xue et al. 2006)
2/26 O1a-M119
1/26 O2a-M95(xO2a1-M88)
1/26 O3-M122(xO3a3a-M159, O3a3b-M7, O3a3c-M134)
1/26 O3a3c-M134(xO3a3c1-M117)
4/26 O3a3c1-M117
9/26 = 34.6% O-M175 total

0/85 O-M175(xO3-M122) Khalkh (Katoh et al. 2004) [16/85 = 18.8% O3-M122]

Ebizur said...

Both O2b-SRY465 and O2*-P31(xO2a-M95, O2b-SRY465) have been found in Tungusic and Mongolic populations of Greater Manchuria. Even O1a-M119 and O2a-M95(xO2a1-M88) have been found in some of these populations, especially the Mongolic-speaking Daurs. I suppose the O-M175(xO1a-M119, O2a-M95, O3-M122) individuals that Wells et al. have found in their Central Asian samples are probably O2b-SRY465 or O2*-P31, but, as you can see, the combined frequency of these haplogroups is generally not significantly greater in the easterly Mongolic and Tungusic populations than it is in the westerly Iranic and Turkic populations. Only the Koreans and the Manchu sample of Katoh et al. 2004 have much higher frequencies of O-M175(xO1a-M119, O2a-M95, O3-M122) Y-DNA, but that is entirely due to the elevated frequency of O2b-SRY465 among them.

Maju said...

Thanks, Ebizur. It gives the impression of there being a relevant O2(xO2a,O2b) in NE Asia. I'm not sure of the apportions in SE Asia but if nothing like that exists, we should consider if O2 did expand from North to South (after a first South to North migration if we accept that the origin of MNOPS, NO and O are in the South).

This can be very interesting because it could make O2 people the carriers of bladelet technology, assuming that this one arrived to East Asia via Altai (40 Ka) and Mongolia (20 Ka) with Son Vi culture known in SE Asia (also 20 Ka).

Ebizur said...

Actually, I should have said that the Manchu sample of Xue et al. 2006 also has a moderately high frequency of {O2b+O2*} (5/35 = 14.3%), but in that case it is mostly due to O2*-P31 (3/35), whereas the Manchu sample of Katoh et al. 2004 has only 1/101 potential O2*-P31 (1/101 O-M175(xO1a-M119, O2b-SRY465, O3-M122)).

In other words, the Manchu sample of Katoh et al. 2004 has a high frequency of O-M175(xO1a-M119, O2a-M95, O3-M122) Y-DNA, but this is, as in the case of the Koreans, almost entirely due to their large apportion of O2b-SRY465. The Manchu sample of Xue et al. 2006 has a moderate frequency of O-M175(xO1a-M119, O2a-M95, O3-M122) Y-DNA, but in that case O2*-P31 makes a much greater contribution to the {O2b+O2*} total.

Maju said...

"as you can see, the combined frequency of these haplogroups is generally not significantly greater in the easterly Mongolic and Tungusic populations than it is in the westerly Iranic and Turkic populations".

But no data from Mongols proper. The Manchu and Oroqen frequencies of O2* are important (6-9%) and larger than those in Central Asians for all O* (which could also be O2b)

Overall, O(xO1a, O2a, O3) the apportions among NE Altaics are rather large: 7-14% , clearly more and a possible source of Central Asian one (1-6%). Founder effects within the Turkic expansion should also be considered as the densities were rather low and the overall Turkic impact major in Central Asia, specially in Y-DNA.

Ebizur said...

No, Maju. I already showed you the relevant data derived from the Khalkh sample of Katoh et al. 2004.

Mongol Y-DNA haplogroup O-M175 data:

Daur (Xue et al. 2006)
1/39 = 2.6% O-M175(xO1a-M119, O2-P31, O3-M122)
2/39 = 5.1% O1a-M119
1/39 = 2.6% O2*-P31(xO2a-M95, O2b-SRY465)
1/39 = 2.6% O2b-SRY465(xO2b1-47z)
6/39 = 15.4% O2a-M95(xO2a1-M88)
7/39 = 17.9% O3-M122(xO3a3a-M159, O3a3b-M7, O3a3c-M134)
3/39 = 7.7% O3a3c1-M117
21/39 = 53.8% O-M175 total

Inner Mongolian (Xue et al. 2006)
1/45 = 2.2% O2*-P31(xO2a-M95, O2b-SRY465)
5/45 = 11.1% O3-M122(xO3a3a-M159, O3a3b-M7, O3a3c-M134)
3/45 = 6.7% O3a3c-M134(xO3a3c1-M117)
5/45 = 11.1% O3a3c1-M117
14/45 = 31.1% O-M175 total

Zakhchin (Katoh et al. 2004)
2/60 = 3.3% O2b-SRY465
5/60 = 8.3% O-M175(xO1a-M119, O2b-SRY465, O3-M122)
5/60 = 8.3% O3-M122
12/60 = 20.0% O-M175 total

Khalkh (Katoh et al. 2004)
16/85 = 18.8% O3-M122

Outer Mongolian (Xue et al. 2006)
1/65 = 1.5% O2*-P31(xO2a-M95, O2b-SRY465)
3/65 = 4.6% O3-M122(xO3a3a-M159, O3a3b-M7, O3a3c-M134)
1/65 = 1.5% O3a3c-M134(xO3a3c1-M117)
3/65 = 4.6% O3a3c1-M117
8/65 = 12.3% O-M175 total

Uriankhai (Katoh et al. 2004)
3/60 = 5.0% O2b-SRY465
4/60 = 6.7% O3-M122
7/60 = 11.7% O-M175 total

Khoton (Katoh et al. 2004)
1/40 = 2.5% O3-M122

As you can see, there is 1/65 O2*-P31 in the Outer Mongolian sample of Xue et al. 2006 and 0/85 O-M175(xO3-M122) in the Khalkh sample of Katoh et al. 2004.

However, you are correct that the frequency of O-M175(xO1a-M119, O2a-M95, O3-M122) may be slightly higher in northern Manchuria and Hulunbuir than in Central Asia:

Manchu (Xue et al. 2006)
3/35 = 8.6% O2*-P31(xO2a-M95, O2b-SRY465)
2/35 = 5.7% O2b-SRY465(xO2b1-47z)

5/35 = 14.3% O-M175(xO1a-M119, O2a-M95, O3-M122)

Oroqen (Xue et al. 2006)
1/31 = 3.2% O-M175(xO1a-M119, O2-P31, O3-M122)
2/31 = 6.5% O2*-P31(xO2a-M95, O2b-SRY465)

3/31 = 9.7% O-M175(xO1a-M119, O2a-M95, O3-M122)

Daur (Xue et al. 2006)
1/39 = 2.6% O-M175(xO1a-M119, O2-P31, O3-M122)
1/39 = 2.6% O2*-P31(xO2a-M95, O2b-SRY465)
1/39 = 2.6% O2b-SRY465(xO2b1-47z)

3/39 = 7.7% O-M175(xO1a-M119, O2a-M95, O3-M122)

Nanai/PRC (Xue et al. 2006)
1/45 = 2.2% O2*-P31(xO2a-M95, O2b-SRY465)
2/45 = 4.4% O2b-SRY465(xO2b1-47z)

3/45 = 6.7% O-M175(xO1a-M119, O2a-M95, O3-M122)

Evenk/PRC (Xue et al. 2006)
0/26 O-M175(xO1a-M119, O2a-M95, O3-M122)

The Manchus, Oroqens, Evenks, and Nanais are traditionally Tungusic-speaking populations, though all these groups, and especially the Manchus, have suffered varying degrees of language shift. The Daurs are an eccentric Mongolic-speaking population that lives in close proximity with some Tungusic populations.

Anyway, the presently available samples are probably insufficient to demonstrate a statistically significant cline in the frequency of O-M175(xO1a-M119, O2a-M95, O3-M122) between Central Asia and northern East Asia (excluding southern Manchuria, Korea, and Japan, where O2b-SRY465 is common).

Maju said...

Oops, sorry, I forgot about that.

For me it looks a Tungus area founder effect then but more detailed research would be needed to confirm (i.e. haplotype trees and such). Maybe a Tungusic clan became heavily involved in early Turkic expansion and turkified in the process. Little is known of Turkic origins and early expansion.

I see no reason to think that haplogroup O (any variant) was important in Central Asia before Turkic expansion. I may be wrong but, would this be the case, we should see it also in South Asia and Iran, even if in minor apportions, right?

In any case, thanks for the detailed information.

terryt said...

"OK. Last comment here. From me anyway".

Sorry. Couldn't resist:

"It gives the impression of there being a relevant O2(xO2a,O2b) in NE Asia. I'm not sure of the apportions in SE Asia but if nothing like that exists, we should consider if O2 did expand from North to South"

Yes, we should. And O2 in the south is almost exclusively O2a. As I said, O2 basically divides into northern and southern versions. So why the qualification:

"after a first South to North migration if we accept that the origin of MNOPS, NO and O are in the South"

I mean it's extremely unlikely that it was O that migrated north, it was NO.

"I see no reason to think that haplogroup O (any variant) was important in Central Asia before Turkic expansion".

I think you are correct here. O in the north is a late phenomenon, even later than its expansion south in the early Neolithic.

Maju said...

"Sorry. Couldn't resist".

Guess it's ok because as soon as I told you to stay on topic, Ebizur came with the stuff from days ago...

But it'd be nice if discussions on East Asian genetics would be placed under East Asian genetic topics, for several reasons: (1) a mere matter of some order, (2) a casual reader can come and not find this interesting stuff as it's not where it "should be", (3) a reader may expect to find 100+ comments here on Gibraltar and it happens that almost all are on East Asian and Oceanian genetics what can be frustrating, and (4) any of us may find later hard to find this thread and the references and data thrown in here.

So in the future, let's please all, including myself, try to stay on topic a bit.

Maju said...

"And O2 in the south is almost exclusively O2a".

Well, according to Wikipedia and my memory O2b is also found in SE Asia. I remember the island of Hainan, in southernmost China, from another discussion, but Wikipedia also mentions Thais and Vietnamese. O2a is not found in the north instead. That's a reason to think that O2 as a whole originated towards the South.

We'd need anyhow a better resolution for the whole continental region to judge properly on this matter.

"... it's extremely unlikely that it was O that migrated north, it was NO".

If O1 is southerner and O2 and O3 flowed from South to North, then O as whole must have coalesced in the South.

N anyhow also has a rather southerner coalescence area (south China probably), specially considering its hyper-northerner spread. So it's reasonable to think that the expansion of N southwards was blocked by its "big brother" O.

terryt said...

"So in the future, let's please all, including myself, try to stay on topic a bit".

I agree, so I won't comment on your next entry although I see some huge holes and unwarranted assumptions in your reasoning.

Ebizur said...

Xue et al. 2006

O2a-M95(xO2a1-M88)
20/34 = 58.8% Li
13/35 = 37.1% Buyi
6/39 = 15.4% Daur
3/33 = 9.1% Qiang
3/34 = 8.8% She
2/34 = 5.9% Hani
2/35 = 5.7% Yao/Liannan, Guangdong
2/47 = 4.3% Japanese
1/26 = 3.8% Evenk/PRC
1/30 = 3.3% Han/Lanzhou, Gansu
1/32 = 3.1% Han/Yili, Xinjiang
1/34 = 2.9% Han/Chengdu, Sichuan
1/35 = 2.9% Yao/Bama, Guangxi
0/25 Korean/PRC
0/31 Oroqen
0/31 Uygur/Urumqi, Xinjiang
0/35 Hui/PRC
0/35 Manchu
0/35 Han/Meixian, Guangdong
0/35 Han/Harbin, Heilongjiang
0/35 Tibetan
0/39 Uygur/Yili, Xinjiang
0/41 Xibe
0/43 Korean/Korea
0/45 Nanai/PRC
0/45 Inner Mongolian
0/65 Outer Mongolian

O2a1-M88
15/34 = 44.1% Hani
6/35 = 17.1% Buyi
4/34 = 11.8% Han/Chengdu, Sichuan
1/33 = 3.0% Qiang
1/34 = 2.9% Li
1/35 = 2.9% Yao/Liannan, Guangdong
0/25 Korean/PRC
0/26 Evenk/PRC
0/30 Han/Lanzhou, Gansu
0/31 Oroqen
0/31 Uygur/Urumqi, Xinjiang
0/32 Han/Yili, Xinjiang
0/34 She
0/35 Tibetan
0/35 Yao/Bama, Guangxi
0/35 Han/Meixian, Guangdong
0/35 Han/Harbin, Heilongjiang
0/35 Hui/PRC
0/35 Manchu
0/39 Daur
0/39 Uygur/Yili, Xinjiang
0/41 Xibe
0/43 Korean/Korea
0/45 Nanai/PRC
0/45 Inner Mongolian
0/47 Japanese
0/65 Outer Mongolian

O2*-P31(xO2a-M95, O2b-SRY465)
13/35 = 37.1% Yao/Bama, Guangxi
5/35 = 14.3% Han/Meixian, Guangdong
3/32 = 9.4% Han/Yili, Xinjiang
3/35 = 8.6% Han/Harbin, Heilongjiang
3/35 = 8.6% Manchu
2/31 = 6.5% Oroqen
1/25 = 4.0% Korean/PRC
1/33 = 3.0% Qiang
1/34 = 2.9% Han/Chengdu, Sichuan
1/34 = 2.9% Li
1/35 = 2.9% Hui/PRC
1/39 = 2.6% Daur
1/43 = 2.3% Korean/Korea
1/45 = 2.2% Nanai/PRC
1/45 = 2.2% Inner Mongolian
1/65 = 1.5% Outer Mongolian
0/26 Evenk/PRC
0/30 Han/Lanzhou, Gansu
0/31 Uygur/Urumqi, Xinjiang
0/34 Hani
0/34 She
0/35 Buyi
0/35 Tibetan
0/35 Yao/Liannan, Guangdong
0/39 Uygur/Yili, Xinjiang
0/41 Xibe
0/47 Japanese

Ebizur said...

The high frequency of O2*-P31 in the Yao people of Bama, Guangxi is due to a fairly recent founder effect (as I recall, they all share an identical or almost identical haplotype). Note that O2*-P31 has not been found in the other Hmong-Mien samples in this study (Yao from Liannan, Guangdong and She).

The most clearly latitudinally biased subclades of O-M175 are the two common subclades of O2-P31, namely O2a-M95 and O2b-SRY465; the former has been found mainly in Southeast Asia and South Asia, while the latter has been found mainly in the vicinity of the Sea of Japan. However, both subclades have been observed occasionally outside of their typical zones of occurrence (O2a-M95 in Daurs, Japanese, Evenks, Uzbeks, etc. and O2b-SRY465 in Thais, Vietnamese, etc.). The subclade O2a1-M88 has not been found further north than the Qiang in East Asia as far as I can recall, but it has been found among Indo-Iranian speakers in South Asia (Pashtuns, Tharus, etc.).

terryt said...

"However, both subclades have been observed occasionally outside of their typical zones of occurrence"

To me that is yet more evidence for a relatively recent expansion, not the Paleolithic one that Maju is so keen on.

"The subclade O2a1-M88 has not been found further north than the Qiang in East Asia as far as I can recall, but it has been found among Indo-Iranian speakers in South Asia (Pashtuns, Tharus, etc.)".

To me O2a1-M88 is the only possible candidate for being any male element having carried the SE Asian-related Munda languages into India. These languages, too, are much more likely to be a Neolithic introduction rather than Paleolithic.

terryt said...

By the way Ebizur. Do you have any information for various populations in Kashmir? Obviously this is not the place to post the information, but if you have we'll come up with something.

Ebizur said...

I have never seen any report of any O2a1-M88 being found in a Munda-speaking population. O2a1-M88 has been found in various populations of southern China (especially the southwestern parts of that country, such as Sichuan and Yunnan, but not in Xizang), Indochina, and Borneo and in Indo-Iranian-speaking populations of South Asia. Only O2a*-M95 has been reported in samples of Munda-speaking populations.

As for Kashmiris, they seem to belong mainly to Y-DNA haplogroup R, with a quantum of at least Y-DNA haplogroup J in addition to R:

Qamar et al. 2002
Kashmiri
3/12 P-92R7(xR1a1-SRY1532.2, R1b1b2a1a2c-SRY2627)
7/12 R1a1a-M17
2/12 J-12f2

Two of the P(xR1a1) Kashmiris in this small sample appear to share an identical haplotype that belongs to haplogroup R1b. The other P(xR1a1) Kashmiri belongs to some unusual sort of P-92R7, probably the recently identified R2*-M479(xR2a-M124).

terryt said...

"Only O2a*-M95 has been reported in samples of Munda-speaking populations".

Thanks for that clarification. And for the information regarding Kashmir.