Jeffrey Rose and colleagues gift us with a beautifully written and delightfully detailed open access study on a culture of the Middle Paleolithic of Arabia: the Nubian techno-complex of Dhofar:
Jeffrey I. Rose et al., The Nubian Complex of Dhofar, Oman: An African Middle Stone Age Industry in Southern Arabia. PLoS ONE 2011. Open access.
I strongly recommend reading this paper in full: it really deserves your attention.
The Nubian Complex: extension and origins
The Nubian techno-complex is a facies of the pan-African Middle Stone Age macro-culture (MSA for short), which is roughly equivalent in timeline to the Middle Paleolithic of Europe (and, as techno-culture, to Mousterian in this other context). A facies that is mostly concentrated in North Sudan and Upper Egypt (with the occasional Ethiopian site) and, now we get to know, in Dhofar (Oman).
| Fig. 1 Nubian Complex occurrences |
In Africa:
Late Nubian Complex assemblages have been found in stratigraphic succession overlying early Nubian Complex horizons at Sodmein Cave [11] and Taramsa Hill 1 [21] in Egypt; in both cases separated by a chronological hiatus. The early Nubian Complex roughly corresponds to early MIS 5, while numerical ages for the late Nubian Complex in northeast Africa fall in the latter half of MIS 5.
In Arabia:
For the time being, the apparent distribution of Nubian Levallois technology in Arabia is limited to the Nejd plateau and, perhaps, Hadramaut valley (Fig. 1). Archaeological surveys in central/northern Oman have not produced any evidence of Nubian Complex occupation [66], [68], nor have Nubian Complex occurrences yet been found in eastern [22], [69]–[71], central, or northern Arabia [72]–[74].
| Fig. 10 Dhofar Nubian Complex' points |
Note that the authors' concept of Nedj plateau does not correspond with that of Wikipedia, as they are obviously talking of the sites in highland Dhofar and not anywhere in Saudi Arabia (see map below).
The authors express their expectation that eventually other sites will be found within drainage systems along the western coast and hinterlands of central Arabia, linking Nubia with South Arabia. However it is also possible, I'd say, that the actual link is via the Horn of Africa, specially as Arabia has been quite extensively combed in recent years.
The Nubian techno-complex in Sudan appears to have evolved locally:
Taking into account its distinct, regionally-specific characteristics, Marks [2] notes that the Nubian Complex has no exogenous source and, therefore, probably derives from a local Nilotic tradition rooted in the late Middle Pleistocene (~200–128 ka). This supposition is supported by the early Nubian Complex assemblage at Sai Island, northern Sudan, which overlies a Lupemban occupation layer dated to between ~180 and 150 ka.
The oldest known Lupemban culture is dated to c. 300 Ka ago in Kenya and Tanzania.
The authors reject the presence of Nubian Complex tools claimed in the past for the Levant (Levantine Mousterian) and Persian Gulf (Jebel Barakah).
Previously to this work:
The first hint of the Nubian Complex extending into southern Arabia was documented by Inizan and Ortlieb [31], who illustrate three cores from Wadi Muqqah in western Hadramaut, Yemen, with Nubian Type 1 and Type 2 technological features. More recently, Crassard [32] presents a handful of Levallois point cores exhibiting Nubian Type 1 preparation from Wadi Wa'shah, central Hadramaut, Yemen.
Time frame and ecology: the wet MIS 5
The chronological reference of Marine Isotope Stage 5, time frame of the Nubian Complex, corresponds to a warm period between c. 130 and 74 thousand years ago, and corresponds very roughly with the Abbassia Pluvial, when the arid region of the Sahara and Arabia was quite more welcoming.
| Fig. 3 Dhofar ecological zones and place names mentioned in text. |
MIS 5 is divided in the following substages (figures are Ka ago and may vary a bit depending on source):
- MIS 5a - 84-74 (wet)
- MIS 5b - 92-84 (?)
- MIS 5c - 105-92 (wet)
- MIS 5d - 115-105 (?)
- MIS 5e - 130-115 (very wet and warm: Eemian interglacial)
MIS 5 was followed by MIS 4, a cold and dry period triggered by the Toba caldera explosion (supervolcano).
In what regards to Dhofar:
... the monsoon increased in intensity during three intervals within MIS 5. Among these humid episodes, the last interglacial (sub-stage 5e; 128–120 ka) appears to represent the most significant wet phase within the entire Late Pleistocene, with rainfall surpassing all subsequent pluvials [42], [43]. Later, less substantial humid episodes associated with sub-stages 5c (110–100 ka) and 5a (90–74 ka) are also attested to in the palaeoenvironmental record. Uncertainties remain concerning the extent to which the climate deteriorated in the intervening sub-stages 5d (120–110 ka) and 5b (100–90 ka).
The increased humidity provided water security to all the region and is also correlated with plant and animal migration from Africa, what the authors think should almost forcibly make humans participant in this overall biological outpouring.
Out of Africa: the alternative routes
 |
| Dhofar mountains in monsoon season |
They acknowledge the conceptual debt to population genetics for unveiling the probable Arabian route Out of Africa, with particular mention to Behar 2008, who points to the possibility (that I have re-elaborated myself on my own means but on his data) of mtDNA L3'4'6 (and I'd say also L0) having left very indicative remnants in Arabia Peninsula. However they make unnecessary conceptual contortions in order to adapt archaeological knowledge to the molecular clock pseudo-science when it must be the other way around, if anything. No need.
In any case, and this is very important, they describe two different cultural groups in interglacial Arabia:
... we surmise that at least two technologically (hence culturally) differentiated groups were present at this time: Nubian Levallois in southern Arabia and centripetal preferential Levallois with bifacial tools in northern/eastern Arabia.
They also suggest that, after the arid MIS 4 parenthesis, South Arabia experienced another mildly wet period with the MIS 3 (since c. 60 Ka ago), which would have enabled:
... north-south demographic exchange between ~60–50 ka. South Arabian populations may have spread to the north at this time, taking with them a Nubian-derived Levallois technology based on elongated point production struck from bidirectional Levallois cores, which is notably the hallmark of the Middle-Upper Palaeolithic transition in the Levant [105], [106].
But the whole Persian Gulf and Arabian Sea area, not to mention East Asia, remains to be fit in (archaeologically speaking) if we are to understand this period's colonization of West Asia from the East (according to the genetic data).
See also:
In this blog:
- Some key archaeological papers on the 'coastal route' (on Fields 2007, Bailey 2009 and Rose 2010)
- Coastal route through Arabia 130,000 years ago confirmed? (on Armitage 2011)
- Middle Paleolithic of Nefud (Arabia) (on Petraglia 2011)
- The various options for the migration out of Africa (a review of all the former)
In external sites:
- R.C. Walter 2000, Early human occupation of the Red Sea coast of Eritrea during the last interglacial (Nature - ppv)
- P. van Peer 2003, The Early to Middle Stone Age Transition and the Emergence of Modern Human Behaviour at site 8-B-11, Sai Island, Sudan (PDF)
- Ghanim Wahida 2009, A Middle Paleolithic Assemblage from Jebel Barakah, Coastal Abu Dhabi Emirate (PDF)
- M. Petraglia 2010, Out of Africa: new hypothesis and evidence for the dispersal of Homo sapiens along the Indian Ocean rim (scribd)
Update (Jan 11): I have received a copy of a related paper dealing with the relations of Hadramaut tools in the context of global Levallois technique. It is however too technical and inconclusive for me to discuss separately. Yet I do not see it being published anywhere online (PPV or open source or whatever), so I am just uploading it online (for a year) so you can download and read it yourself:
Very interesting. The dates make Stephen Oppenheimer's theory that HS was already in SE Asia at the time of Toba a bit more credible. Yet I still have my doubts, the Jwalapuram artifacts that Petraglia et al found below the Toba ash layer in S. India are very crude compared to the Nubian tools in Oman. Stone technology went in reverse for 30 kyrs? I am not convinced that the Jwalapuram people were HS.
ReplyDeleteOpenheimer's elaboration was strongly based on some results from molecular clock estimates and I doubt even himself would stand for that today.
ReplyDeleteAnyhow IMO M is somewhat older than N (something I think Oppenheimer admitted to in a later paper - ?) and that, together with the huge size of the M star-like structure, really implies that M is the first clear signal of human expansion in Eurasia (in South Asia in fact but with projection to the East). After M, I find the following sequence of "stars": (1) M4"64 (South Asia), (2) N (SE Asia probably) and (3) R (South Asia again but as part of an N backflow from the East).
Whatever the case, whoever is more correct, the role of Toba in the Eurasian colonization is still a bit of a mystery: we can identify survival before and after Toba in India and we can identify a single quite massive and sudden demographic explosion which is more consistent with many and healthy children for many successive generations (a happy age) after arrival to South Asia that with a the pruning we would expect from such a catastrophe like Toba.
In fact we don't see any such pruning, so either the expansion happened only after Toba or, much more likely, Toba wasn't such a huge big deal after all.
I think that we can detect a slow down in population expansion at (coding region) mutational step #9 counting from L3 (#6 from M, #4 from N). However I'd say that this corresponds with a later period, i.e. with the colonization of West Eurasia after 50 Ka ago, so my general impression is that either Toba was neutral or the Eurasian expansion (altogether) only began after Toba. I don't see any break in between.
"... the Jwalapuram artifacts that Petraglia et al found below the Toba ash layer in S. India are very crude compared to the Nubian tools in Oman".
ReplyDeleteSorry I missed this last part.
How so? I have seen the toolkit and they seem at least as good as these (most of which are cores anyhow, not points or tools). The identification with MSA (of South African facies) was pretty solid and the time frame is wholly consistent with a migration through Arabia in the MIS 5, specially in the late periods of this climatic window. It would not be Nubian facies anyhow but something different (yet of African origin as well).
"Stone technology went in reverse for 30 kyrs?"
They are not even directly related in principle and I really do not agree with your appreciation at all: the Jawlapuram toolkit is as good as this one as far as I can discern, albeit somewhat different.
In any case, stone tech can indeed go apparently "backwards", a lot depending of viewpoint: the macrolithic techno-cultures of Iberia are generally seen at first sight as shockingly "backwards" but they may be not and are inserted between microlithic technologies, appearing much more "modern". This kind of counter-ntuitive evolution does happen in the archaeological record even if I do not think is the case at all in Jawlapuram.
"I am not convinced that the Jwalapuram people were HS".
I am, Petraglia is and really after the various Arabian inroads from Africa we have no other candidates for continuity than the various South Asian late MP groups, including Jawlapuram, which are in fact precursors of the later West Eurasian Upper Paleolithic, having the first stone blades (even in some cases in the early MP, typically considered made by the Hathnora hominins and not Homo sapiens), which become more and more common in some sites.
"The increased humidity provided water security to all the region and is also correlated with plant and animal migration from Africa, what the authors think should almost forcibly make humans participant in this overall biological outpouring".
ReplyDeletePresumably all those the plants and animals accompanied the humans in their boats.
@ Terry - most mammals can swim, the straits weren't very wide
ReplyDelete@ Maju - I admit to not being a very good judge of stone tools. Ok, if all the experts are satisfied that the sub-Toba tools are HS, I have to take it on faith. In that case, there is no major barrier between S India and SE Asia, so HS was there very early too. So Toba not being a big deal seems as good an answer as any for the lack of a genetic bottleneck signal.
In principle Petraglia (in 2007) described not just the site and findings but compared them with an array of other traditions and they clustered almost perfectly with those of Southern African MSA tradition (and strangely not the Eastern African one).
ReplyDeleteThe paper was easy to find full text in the first months /and I read it several times back in the day) but then it became strictly PPV. In any case it is: "Middle Paleolithic assemblages from the Indian subcontinent before and after the Toba super-eruption, Science 2007".
Regarding the Toba eruption it is a piece not yet well fit. In the past I generally liked to think that the M demographic explosion happened right after Toba but the archaeology may support a pre-Toba explosion as well.
ReplyDeleteMy issue is that we should detect a demic explosion upon the arrival to South Asia and then a pruning and a second explosion over the ashes, even if this one might have several or many different origins (as people was already scattered). The problem is that I don't see any obvious signal of pruning or bottlenecking (but it may be not easy to detect) nor of a span of time of contraction (it may have been too short to notice within the quite slow mtDNA coding region mutation rate).
Maybe Toba killed, say, 50% of the people and the other 50% just bounced back in few generations leaving no obvious scar for us to see? 50% mortality is a very mild bottleneck in fact: in the day it'd be seen as a huge catastrophe but the overall long-run effects are almost nil (unless it repeats once and again).
The Toba effect that I am inclined to credit is that it may have greatly weakened populations of archaic humans in Eastern India and Southeast Asia, and that the existence of those populations in an unweakened state had discouraged AMHs in India from moving further into Asia, without necessarily all that much mortality among the AMHs themselves (in the long term demographic sense).
ReplyDeleteThis would parallel AMH expansion into Europe only after Neanderthals who were "holding" that territory were weakened by the volcanic explosions there and subsequent climate shift ca. 40 kya.
That's exactly what Petraglia suggested in a letter in 2007: that Toba might have sped up the demise of archaics hominins rather than (presumably more resourceful) Homo sapiens.
ReplyDeleteThe "death" of the Toba bottleneck hypothesis was also hailed by Hawks that same year as a good thing.
In Europe we also have a mini-Toba (the Campanian Ignimbrite Eurption) that is related to the demise of Neanderthals (or whoever made the Chatelperronian), at least in the Franco-Cantabrian region. But then again Italian Uluzzian and Greek Aurignacian peoples lived before and after the C.I. event.
The main effect apparently was to cool the climate for a long time in both cases: MI4 in the Toba case and the shorter HE4 in Europe later on. In both cases, I think we can relate a gigantic star-like mtDNA node: M in South Asia and H in West Eurasia, however it is unclear if that was triggered by the volcanoes or would have happened without them (I'm rather inclined towards the latter option nowadays and would rather associate smaller "stars" like H1 to the volcanoes "clearing" effect if anything at all).
"Terry - most mammals can swim, the straits weren't very wide"
ReplyDeleteIn which case humans probably swam the strait as well.
"After M, I find the following sequence of 'stars': (1) M4"64 (South Asia), (2) N (SE Asia probably) and (3) R (South Asia again but as part of an N backflow from the East)".
So you consider that M4''64 (now M4''67) expanded later than M as a whole. It has but a single mutation from basal M. The same relationship that R has to basal N. So you should now agree that R's expansion was later than was N's.
I agree that M4''67's expansion is later than M's. M4''67 coalesced from a haplogroup that had taken part in the 'original' M expansion. The M haplogroups were already in place by the time of M4''67's expansion.
I also agree that R's expansion ter than N's. R coalesced from a haplogroup that had taken part in the 'original' N expansion. The N haplogroups were already in place by the time of R's expansion. That does not fit with your rider to number 3: 'as part of an N backflow from the East'. Haplogroup N and R expanded independently of each other, N first then R.
"So you consider that M4''64 (now M4''67) expanded later than M as a whole".
ReplyDeleteIt shows up as one mutation downstream and that means that it coalesced after M. Could it be just one or a few generations after M? Maybe but my best estimate is that the average coding mutation rate is several thousand years so in principle not too likely. On the other hand, M has so many daughters that maybe the odds in favor of an early mutation increase, can't say: I'm not such a good statistician.
I just stated now and before how the mutation sequence goes, the actual sequence of events can't be discerned so easily, except for the compulsory "before" and "after" order that the phylogenetic hierarchy demands. The best we can say beyond that is "looks like" it has this or that rhythm, but it's just an appearance whose truthfulness we can discern beyond reasonable doubt.
"So you should now agree that R's expansion was later than was N's".
Looks like... :D
"The M haplogroups were already in place by the time of M4''67's expansion".
Do not run so fast. Many basal M have more than one stem mutation (so they had not yet coalesced, according to the statistical approximation), while the ones that have one defining mutation could be younger or older (very difficult that they coalesced all at the very same day and hour, not even century) - and considering the success of M4'67, it may well be argued that it was the elder of all M subclades, although this is not necessary either.
Here is where we come to the possibility that the M4'67 expansion was tightly inserted inside the M one (but how tightly, if at all?), much as can be said that the R expansion was inserted into the N one (however here we have the issue of different apparent geographies for the focus of those expansions, a difficulty we don't have with M and M4'67).
We cannot in any case reach to any firm conclusion only from this data, just like you can't solve a single quadratic equation, we need more data, such as archaeological one.
...
...
ReplyDelete"R coalesced from a haplogroup that had taken part in the 'original' N expansion".
At risk of being accused of "splitting hairs", I must insist that there is no "haplogroup" between N and R, not even a haplotype if we only consider the coding region (as we are doing here). At most there was a real matriline or matrilineage made of a chain of fleshy mother-daughter sequence. That is not a "haplogroup".
Those women who migrated from point 1, where N coalesced, to point 2, where R did, were all underived N, until one, a quite successful one, had the transition at locus 12705 that defines R. Only since then we can begin thinking in terms of R and a "haplogroup" distinct from N.
Assuming that each mutation took some 2500 years to take place (a half-decent hunch), there were some 100 generations in that line (at 25 yrs. per generation). And that applies to all the other sublineages derived by just one mutation from the basal node (200 generations for 2 mutations, etc.) But these are just statistical approximations and the corresponding mutations may happen as early as one generation downstream and as late as never.
So there could well be a small group of N women migrating slowly but steadily from Burma, coalescing into R in, say Bengal (so they could back-migrate to SE Asia easily, as we know they did) at generation #25 and arriving to Iran at generation #100 when, by the magical power of the fairy of statistics and random events (alias Chaos), all the other N (and R) sublineages popped around naturally.
The notion we get as conclusion is that the migrating population was initially small enough to have R soon as dominant (per drift)... but retained some underived N women, who also benefited from the success of R.
Whatever the case all I say here and elsewhere is just a reconstruction and actual events may well have deviated from the narrative I propose and argue for, although probably not by too much: the overall narration should still be recognizable, because all I do is to reconstruct the most plausible and parsimonious steps explaining the origin of our reality, letting the data do most of the job on its own merit and only adding the narration to explain the fine detail that some seem to find so difficult to understand.
"In which case humans probably swam the strait as well".
ReplyDeleteWith their families? Humans work in groups, you know. If Sam crossed the strait swimming, what is plausible but unnecessarily dangerous, he would not be able to reproduce and would probably find his new solitary life in Yemen meaningless.
So Sam would cross back and build or maybe borrow a boat, so he could cross again with more people, not necessarily all so good swimmers and risk-takers.
Animals are different: they often only gather for mating. And they are often excellent innate swimmers. It depends on the species anyhow.
"Do not run so fast. Many basal M have more than one stem mutation (so they had not yet coalesced, according to the statistical approximation)"
ReplyDeletePerhaps they 'had not yet coalesced', but they must have already separated from basal M. The tail is as likely (in fact more likely) to be the result of drift in a small isolated population rather than being a function of a period of migration.
"Here is where we come to the possibility that the M4'67 expansion was tightly inserted inside the M one"
I agree it's a possibility. In fact I'd be fairly sure that haplogroup R11'B6/B4'5's expansion is 'tightly inserted inside' the R one. Such an interpretation would eliminate India as being the origin of R.
"much as can be said that the R expansion was inserted into the N one (however here we have the issue of different apparent geographies for the focus of those expansions, a difficulty we don't have with M and M4'67)"
As you say, it is less likely because the distribution of the haplogroups is so different. The distribution of the M4''67 is reasonably close to that of M as a whole, especially within India.
"At risk of being accused of 'splitting hairs', I must insist that there is no 'haplogroup' between N and R, not even a haplotype if we only consider the coding region (as we are doing here). At most there was a real matriline or matrilineage made of a chain of fleshy mother-daughter sequence. That is not a 'haplogroup'".
It is splitting hairs really. I'll call it pre-R to satisfy you.
"Assuming that each mutation took some 2500 years to take place (a half-decent hunch), there were some 100 generations in that line (at 25 yrs. per generation). And that applies to all the other sublineages derived by just one mutation from the basal node (200 generations for 2 mutations, etc.)"
Sounds reasonable.
"So there could well be a small group of N women migrating slowly but steadily from Burma, coalescing into R in, say Bengal"
As I said, if we consider R11/B's expansion as being embedded in R's we should place R somewhere around the South China Sea.
"by the magical power of the fairy of statistics and random events (alias Chaos), all the other N (and R) sublineages popped around naturally".
Surely the other N haplogroups were already widely distributed before R coalesced. They were not associated with R's exapnsion in any way.
"With their families? Humans work in groups, you know".
If the claim is made that other animals swam the strait then there is no reason at all why humans could not have done so. I consider it far more likely that the strait had actually dissappeared at the time and both humans and animals could simply walk across.
"If Sam crossed the strait swimming, what is plausible but unnecessarily dangerous, he would not be able to reproduce and would probably find his new solitary life in Yemen meaningless".
The same would apply to an individual of any species.
"Animals are different: they often only gather for mating".
Humans often 'gather for mating' too, even today. Well, for the preliminaries.
"It depends on the species anyhow".
So we really need to know what species were involved with that migration from Africa.
"... they must have already separated from basal M".
ReplyDelete"I'll call it pre-R to satisfy you".
The key misunderstanding in these ideas is that there is a period T of time (many many generations) in which M (or N) is just M (or N) even in the direct line or women leading to M4'67 (or R). Only after the first woman developed the corresponding mutation we can stop talking of undifferentiated M (or N). Until that happened we do not know which of the many M (or N) women would coalesce (if at all) into M4'67 (or R).
There is a real mother-daughter line that we can infer a posteriori but there is no haplogroup or haplotype or any other thing we can discern in the mtDNA, even if we used a time machine and tested them in situ generation after generation.
So you can't never be talking of "pre-R" but always of undifferentiated N (even if you mean in the flesh lineage that would eventually mutate into R). This is NOT the same as L3 leading to M (or N) because in this case there are indeed pre-M (or pre-N) haplotype stages (although originally it was also undifferentiated L3 for many generations).
...
"As I said, if we consider R11/B's expansion as being embedded in R's we should place R somewhere around the South China Sea".
ReplyDeleteWhy do you just take that only haplogroups as indicators? There are some 18 basally derived haplogroups (19 if we take only the coding region mutations as valid) under R and all must be weighted to estimate the origin of R as a whole.
The overall weighting produces South Asia as center, at least it did last time I tried.
What you say is like claiming:
"if we consider P's expansion as being embedded in R's we should place R somewhere around New Guinea"
or
"if we consider U's expansion as being embedded in R's we should place R somewhere around the Caspian Sea"
Etcetera. We must consider all clades because the origin can't be in all places at the same time in the same manner: the common origin should be near the geometric center of all 18/19 basal subclades (and that is somewhere in South Asia, probably the lower Ganges).
When I explained why an expansion may ("may" not "must") be embedded in the other, I pointed to possible time-frame implications and the fact that daughter and mother clades may have shared part (not all) of the mother's pattern of expansion... but never anything that may suggest that the daughter clade biases the geometric centroid calculated using all basal subclades. No, I do not think that is possible at all.
On a very different reasoning, the direction of the ancestral clade of the mother node (i.e. where is L3's origin located in relation to M and N or where are M and N ones in relation to their daughters) could be used to infer a more realistic corrected centroid by pulling the purely geometric one towards the ancestor's location (1/4?, 1/3?) and that way representing better the directionality of expansions.
So if we get the purely geometric diversity centroid of M around Bihar, this correction could pull it towards Gujarat instead, what is surely quite more realistic, etc. But this is very different of what you say (and at the moment it is just tentative and secondary in my reconstructing logic).
"The distribution of the M4''67 is reasonably close to that of M as a whole, especially within India".
ReplyDeleteONLY in South Asia in fact. So it's not really close at all, as far as I can discern because M is a pan-Eurasian macro-haplogroup found in all the Great Continent, as well as Oceania, America and even much of Africa.
"Surely the other N haplogroups were already widely distributed before R coalesced".
You always say that but it sounds meaningless to me. "Surely" is evidence of nothing and you do not elaborate (nor I think you can).
Per the statistical method there were only a few N subclades we can discern leaving any signal at all of their existence at mutational step N+1: N1'5 (in, say, Afghanistan), N9 (in, say, East China), N11 (in South China), S (in Australia) and R (in India), we can add O (Australia) at N+2. All the rest of mother-daughter were in undetermined locations and NOT yet (most likely) in their final destinations. They were obviously very thin and fragile before they finally expanded (some are still that way, not showing any sign of expansion/consolidation ever: N13, N14, N21, N22 - located in Australia and Malaysia).
"Humans often 'gather for mating' too"...
Not ONLY, which was my point: humans gather for nearly everything. I'm not the best to tell but people are most of their time with others of the same species, even in this individualist and gadget-mediated society we live in today.
Lone wolves are dead wolves, even more true for lone humans.
"So we really need to know what species were involved with that migration from Africa".
They are listed in the paper, which is open access, take a look yourself (essentially small carnivores like the genet and trees like the acacia, if I recall correctly). But I don't "need" anything: it's you and only you the one trying to find some unlikely twist you could use as weak pseudo-argument.
"Why do you just take that only haplogroups as indicators?"
ReplyDeleteBecause it has zero mutations from basal R before it diversifies. If we're going to claim any 'daughter' haplogrouops exapanded at the same time as their 'parent' haplogroup we have no better possibility than R11/B. Of course I am much more inclined to the idea that in general daughter haplogroups expand later than their parent haplogroup.
"the common origin should be near the geometric center of all 18/19 basal subclades"
Not necessarily so. If the expansion was basically unidirectional the 'origin' could quite easily lie some distance from the 'geometric center'.
"So if we get the purely geometric diversity centroid of M around Bihar"
That centre would certainly hold for M4''67. However I am reasonably sure that M4''74's expansion is later than that of M as a whole. Just as I'm reasonably sure that R's expansion is later,than that of N as a whole.
"All the rest of mother-daughter were in undetermined locations and NOT yet (most likely) in their final destinations".
How can you claim that with such certainty? That is equivalent to claiming that Indonesian M26 (11 mutations), N21 (5 mutations) and R23 (13 mutations), Indian M10 (8 mutations) and R7 (8 mutations), South Chinese N8 (13 mutations) were not 'in their final destinations'. Unlikely to be the case.
"They were obviously very thin and fragile before they finally expanded (some are still that way, not showing any sign of expansion/consolidation ever: N13, N14, N21, N22 - located in Australia and Malaysia)".
'Thin and fragile' they may be but those haplogroups are likely to have coalesced from basl N where they are now found rather than being the product of a long period of wandering as thin and fragile lines before they wound up where they are now found. Their thinness and fragility are most probably the product of a long period of drift acting in the small region in which they are now found.
"essentially small carnivores like the genet and trees like the acacia"
Both of which are know to swim huge distances.
"Because it [R11'B] has zero mutations from basal R before it diversifies".
ReplyDeleteNot valid: if we follow my method (of just counting coding region mutations), R11'B counts as two haplogroups. Incidentally we know (or think) that they are related because both share a single HVS-I transition but, similarly, any two other unrelated R subclades could be equally related in real prehistory, but, by mere chance, no such HVS-I mutation happened, so we won't ever know. It's just a fluke: phylogenetically informative but statistically meaningless.
"Not necessarily so. If the expansion was basically unidirectional the 'origin' could quite easily lie some distance from the 'geometric center'".
That I have already proposed above as a possible correction to introduce by moving the centroid some 1/4 (?) the distance between the raw geometric and the previous node's center (L3 for M and N). It would yield intersting origins for M (Gujarat, Mumbai) and N (Bengal) and would require no back-migration for R, allowing both undifferentiated N and R to migrate simultaneously in both West and East directions (this is because the mainly eastward vector of N and the mainly westward vector of R would support overlapping of both original nodes after correction: in Bengal).
But it's a complicated matter with some degree of speculation.
"That centre would certainly hold for M4''67".
Have you studied this clade in such detail (locations of each sublineage and the resulting geometry)? If so I'd like to know the data you manage. Otherwise what you say may just be the last of your happy hunches made doctrine. I have certainly never imagined that M4"67 would expand from so far East.
"M4''74"
Does not exist.
"Their thinness and fragility are most probably the product of a long period of drift acting in the small region in which they are now found".
Maybe. Or maybe they are just an illusion caused by lack of genetic research among Australian Aborigines for a long while. (a case similarly outrageous as that of France, even if the logic behind is different).
Whatever the case these lineages did not coalesce in all their length in Australia because it is extremely unlikely that undifferentiated N reached Australia as such. It is not impossible though and that would mean that undifferentiated N would have traveled all the way from the origin of the expansion (in Thailand, Burma or Bengal) to those areas. But if there is any evidence supporting such idea it'd be rather S (hanging from N by just one mutation) than the marginal and mysterious subclades you mention with they "infinite" stems that inform us of NOTHING.
"That is equivalent to claiming that Indonesian M26 (11 mutations), N21 (5 mutations) and R23 (13 mutations), Indian M10 (8 mutations) and R7 (8 mutations), South Chinese N8 (13 mutations) were not 'in their final destinations'".
They are not in the ancestral location of N, M or R by my estimates (excepting maybe M10 and R7) and in any case they could not all be in the place where the first M or N woman lived (she could not have traveled so much in her life possibly). So taking the case of N only (for simplicity), there was a growing cloud of generic N (either still undifferentiated N or random derived subclades or already coalescing major subclades like R) which expanded from the N origin in Bengal or Thailand (but not both) towards the corresponding destinations.
Even if those migrations included some undifferentiated N (and R) which was producing more and more local derived mutations, which in some cases coalesced into recognizable modern haplogroups or whatever, all that happened "on the march", because what we obviously see with N and R (and M and other clades as well) is a shared origin and very scattered destinations, so the differentiation happened at least in great part "on the march".
PS- And that march happened in a T-shaped expansion (as did also that of M, with some lesser differences) with the bottom by the West and the top by the East, near the Pacific Ocean. And that is very clear.
ReplyDelete"allowing both undifferentiated N and R to migrate simultaneously in both West and East directions"
ReplyDeleteInteresting. You use different criteria for each haplogroup. M4''67 migrated after M because it was separated by just a single step and the two R11'B6/B4'5 haplogroups migrated after R even though they are not separated form it by even a single step. But R must have migrated along with N because it is separated from N by a single step? It is obvious that you are not prepared to consider the evidence as a whole, objectively and from a consistent perspective. You have decided in advance what you want the evidence to show and you are now prepared to doctor your interpretation to fit.
"Whatever the case these lineages did not coalesce in all their length in Australia because it is extremely unlikely that undifferentiated N reached Australia as such".
Or undifferentiated M. There goes any idea that there was any sort of 'rapid' migration to Australia. As I pointed out elsewhere humans may have taked 50,000 years to move from Africa to Australia.
"But if there is any evidence supporting such idea it'd be rather S (hanging from N by just one mutation) than the marginal and mysterious subclades you mention with they 'infinite' stems that inform us of NOTHING".
You're talking rubbish again Maju. Not one of the haplogroups I listed is 'Australian' or even 'Melanesia'. The nearest they come to there is island SE Asia. I deliberately avoided using the more distant haplogroups as examples because I'm quite sure the migration to that region was far from rapid.
"what we obviously see with N and R (and M and other clades as well) is a shared origin"
Only in your imagination.
"the differentiation happened at least in great part 'on the march'".
I agree completely. That's why the distribution of the three haplogroups gives us a very good idea of the route they took from their region of origin.
"And that march happened in a T-shaped expansion (as did also that of M, with some lesser differences) with the bottom by the West and the top by the East, near the Pacific Ocean. And that is very clear".
And originating just outside Africa.
"You use different criteria for each haplogroup".
ReplyDeleteI don't think so.
"M4''67 migrated after M because it was separated by just a single step"...
I believe that I have said that this lineage's expansion could well be "inserted" with that of M as a whole - but only in certain areas (Indian subcontinent) and not others (East Asia, Oceania, West Eurasia and Africa) where M4"67 does not exist.
This distribution pattern is what suggests that the M4"67 could be a little more delayed in relation to that of M.
"... and the two R11'B6/B4'5 haplogroups migrated after R even though they are not separated form it by even a single step".
They are a major part of R expansion but only in certain areas, namely East Asia. What I say is that I treat R11'B6 and B4'5 as two distinct haplogroups for the purposes of my analysis.
"But R must have migrated along with N because it is separated from N by a single step?"
Again the step is not that important: it does establish that there is minimal separation but we can't know exactly if this minimal divergence took one generation or thousands in happening. There is no "molecular clock" except as an approximation: it is not precise at all.
In this case, anyhow, unlike the other two, the distributions overlap a lot. R is together with N in all subregions of Greater Eurasia: South Asia, East Asia, Australasia, West Eurasia and America. It is this overlapping pattern what suggest that R's expansion is tightly and specially related to that of N.
"It is obvious that you are not prepared to consider the evidence as a whole, objectively and from a consistent perspective. You have decided in advance what you want the evidence to show and you are now prepared to doctor your interpretation to fit".
It's obvious that you are not listening to what I say nor looking at the geographic evidence with calm and neutrality.
"Not one of the haplogroups I listed is 'Australian' or even 'Melanesia'. (...) I deliberately avoided using the more distant haplogroups as examples because I'm quite sure the migration to that region was far from rapid".
I call that, in your own words, "doctoring the evidence".
You chose to ignore the very important Australian lineages because you are "quite sure" of whatever, i.e. because you have prejudices that distort your judgment.
In fact N-derived S, R-derived P and other lineages all look like arriving to Australasia very early on (just one mutation under their parents). There is also a Melanesian lineage (M29'Q) that is just one mutation under M, making it possibly older than N itself.
There is always uncertainty in measuring time from mutations but here we have many lineages in this situation of having very short stems: Australasia was colonized very fast without doubt, barely after South and SE Asia were.
"I don't think so".
ReplyDeleteBecause you refuse to see it as such:
"I believe that I have said that this lineage's expansion could well be 'inserted' with that of M as a whole"
Possible. But by no means certain. So I agree with, 'the M4"67 could be a little more delayed in relation to that of M'. But then:
"the step is not that important: it does establish that there is minimal separation but we can't know exactly if this minimal divergence took one generation or thousands in happening. There is no 'molecular clock' except as an approximation: it is not precise at all".
Especially not precise where it fails to fit with a particular belief. Surely if the conclusion is that M4''67 expanded later than did M it is reasonable to accept that R expanded later than did N. Why opposing interpretaions of similar evidence?
"It is this overlapping pattern what suggest that R's expansion is tightly and specially related to that of N".
It's not really particularly 'overlapping'. Like M4''67, R11'B6 and R4'5 N does not seem to have been carried along during the later stages. Europe and the Pacific for example. I don't think any 'overlapping' is evidence of 'together', especially when you've rejected any overlapping in the haplogroups M4''67, R11'B6 and B4'5 with their parents in the early stages of the respective migrations.
"It's obvious that you are not listening to what I say nor looking at the geographic evidence with calm and neutrality".
As far as I'm aware you don't look at the geographic evidence at all.
"I call that, in your own words, 'doctoring the evidence' ... You chose to ignore the very important Australian lineages because you are 'quite sure' of whatever, i.e. because you have prejudices that distort your judgment".
OK. We'll include them. We have S in Australia with one mutation as well as N13 (6 mutations) and N14 (10 mutations). Are you going to claim that N13 and N14 gathered that tail before they entered Australia? or did they enter with S and grow the tail within Australia? My bet is the latter, especially when we notice that O with 2 mutations is also Australian. Turning to New Guinea we have M29'Q with one mutation along with M27 (3 mutations) and M28 (5 mutations). Did these latter two enter New Guinea with N29'Q and develop their tail there? or did they develop the tail before they entered that region? I plump for the first option.
"In fact N-derived S, R-derived P and other lineages all look like arriving to Australasia very early on (just one mutation under their parents)".
There you are. So the N haplogroups with just one mutation, including N1'5, all look like arriving near the extremes of their modern distribution very early on. R expanded through India after N1'5 had already become established in Northwest India/SW Asia. Turning to the N haplogroups with tails: because N was so widely distributed early on it's most likely that those tails developed once they had reached somewhere near where they are found today.
"Australasia was colonized very fast without doubt, barely after South and SE Asia were".
Not 'Australasia'. Just Australia/New Guinea. New Zealand is part of Australasia and it wasn't first settled until about 800 years ago.
Did you read my previous answer before posting all those baseless accusations and already answered questions? For example:
ReplyDelete"Why opposing interpretaions of similar evidence?"
They are not "opposing", just slightly different. And the reason is their difference in geographic extension and overlapping with their respective parent haplogroups.
"I don't think any 'overlapping' is evidence of 'together'"...
Strong indication it is indeed. You are in denial. Why would there be two so closely related waves one after the other in so many different places? And if they were, which is the practical difference between "together" and "right behind"? Splitting hairs!
"... especially when you've rejected any overlapping in the haplogroups M4''67, R11'B6 and B4'5 with their parents in the early stages of the respective migrations".
Because we do not see any such overlapping in ALL the Eurasian geography (nor most of it) but just in specific regions. Are you being intentionally obtuse or what?!
"Are you going to claim that N13 and N14 gathered that tail before they entered Australia?"
In truth we do not know. Their long tail and lack of intermediate relatives only tell us one NOTHING. With NO information we can't say much about them before their final coalescence into a well-characterized haplogroup.
But, let's assume that you are right: that N13 and N14 entered Australia with S and O in the early Eurasian expansion, that M27 and M28, arrived to New Guinea with M29'Q... how does that compare with the relation of A and N9 (and Eastern R) or with that of N1'5 or N2 with Western R? Shouldn't we expect all the long-stemmed haplogroups to have been living at the shadow of larger, short stemmed, relatives (or even less related clades) until they found a niche for expansion (or were drifted out, as undoubtedly happened with many others)?
"R expanded through India after N1'5 had already become established in Northwest India/SW Asia".
That we do not know. If R coalesced right after N (a couple of centuries later, let's say, instead of the average 2-5 millennia), then the whole R tree has to be overlapped with that of N because, incidentally, R and N would be practical contemporaries. That would make N1'5, N9, N11 and S effectively contemporary of R0, R2'JT, R6, R11'B6, B4'5, R30, R31 and P.
Am I wrong and there is a difference of a few thousand years? Does it really matter? Is it such a difference? Can we even know at all if that difference is for real in all cases: not only R could be premature, N-derived lineages could be late in some or many cases - no molecular clock exists but as generic reference: just that it looks like it ticks that way it doesn't mean that it actually ticked as we perceive it, much less in all cases.
"Turning to the N haplogroups with tails: because N was so widely distributed early on it's most likely that those tails developed once they had reached somewhere near where they are found today".
Maybe but they still had to emigrate from the geographical origin of N to their destinations. And the geographical origin of N must be in SE Asia (raw estimate) or Bengal (corrected estimate).
"Not 'Australasia'. Just Australia/New Guinea. New Zealand is part of Australasia"...
Why are you nitty-picky about this? When we talk of the colonization of Europe we certainly do not mean Norway either, much less remote Iceland. Greenland is part of America and was not colonized until the Dorset culture, even later than New Zealand. Yet nobody rises these issues about marginal remote islands and such... only you.
"They are not 'opposing', just slightly different".
ReplyDeleteThey're completely different.
"And the reason is their difference in geographic extension and overlapping with their respective parent haplogroups".
The real reason is that unless you interpret the sets of data differently they would fail to fit your belief.
"Why would there be two so closely related waves one after the other in so many different places?"
Such as M4'67 with M, and R11'B6 and R4'5 with R for example? If N's expansion is related to that of R surely these two examples hold also. You can't just alter your interpretation to make the evidence fit what you want it to fit.
"Because we do not see any such overlapping in ALL the Eurasian geography (nor most of it) but just in specific regions. Are you being intentionally obtuse or what?!"
It is you who is being obtuse. N and R display no greater overlap with each other than do the other examples I mentioned.
"Their long tail and lack of intermediate relatives only tell us one NOTHING".
Perhaps. But the presence of closely realted haplogroups with short tails suggests a long association with the region where they are now found. In fact there are no examples of haplogroups with long tails whose region of coalescence can unequivocally be placed anywhere other than where they are now found. Unless you're utterly committed to some previously held belief.
"how does that compare with the relation of A and N9 (and Eastern R) or with that of N1'5 or N2 with Western R? Shouldn't we expect all the long-stemmed haplogroups to have been living at the shadow of larger, short stemmed, relatives"
If the only exceptions you can come up with involve those few members of haplogroup N surely that should raise alarm bells for your belief. Presumably the haplogroups you mention are the survivors of a long period of drift in a relatively small population. There is no reason at all why they cannot fit the same sort of scenario as that N13, N14, M27 and M28. Not to mention M26, N21, R23, M10, R7 and N8, all of which have a long tail but most likely coalesced where they are found today.
"If R coalesced right after N (a couple of centuries later, let's say, instead of the average 2-5 millennia), then the whole R tree has to be overlapped with that of N"
Again you're forced to be inconsistent because of what you want to believe. What if M4''67 'coalesced right after' M? that would make them 'practical contemporaries'. We can be even more sure that R11'B6 and R4'5 are practical contemporaries with R, and coalesced right along with it, but you are desperate to separate those expansions because associating them conflicts with your belief.
"Am I wrong and there is a difference of a few thousand years? Does it really matter? Is it such a difference?"
A bit of consistency wouldn't go amiss.
"no molecular clock exists but as generic reference: just that it looks like it ticks that way it doesn't mean that it actually ticked as we perceive it, much less in all cases".
Exactly. We can't even take for granted that M's expansion is older than that of N.
"Why are you nitty-picky about this?"
Well, look who's talking. But it's time you got your terminology correct on the subject.
"Such as M4'67 with M, and R11'B6 and R4'5 with R for example? If N's expansion is related to that of R surely these two examples hold also".
ReplyDeleteThey do but unlike R, which has a similar span as its parent N, these do not and are not essentially different from other subclades of the same type.
So, if we are to insert M4'67 in the overall M expansion, we have to do it the same as other locally fruitful clades with similar stems, such as M1'2'51, M3a, M3c, M5, M9, M12'G, M13'46'61, M23'75, M'29'Q, M32'56, M33, M34'57, M35, M39'70, M40'62, M42'74, M49 and M73'79. However all these clades have only localized, regional at best, spans, so they CAN'T have expanded simultaneously with the main M pulse, which is multi-regional.
Similarly, under N, we have R (which is as multi-regional as N) but then also regional clades one mutation under the root: N1'5, N9, N11 and S. Same as before: they CAN'T be simultaneous to the main pulse of N because they are restricted to certain regions. However R does not fit this profile and overlaps with N almost 100%.
Again, as you insist, within R we have no multi-regional, pan-Eurasian subclades either, but rather localized ones - one step under the root: R0, R2'JT, R6, R11'B6, B4'5, R30, R31 and P. As you can see, the two lineages that you had singled out are not special at all.
So the only and quite obvious exception to the apparent sequentiality of the mutational tic-tac is R: it is the only one which is found in all continental sub-regions, closely following the steps of N, suggesting that, to a large extent, "mother" and "daughter" were almost synchronous and took part in roughly the same demographic expansion process.
However, said that, we can even qualify more in detail and we may find that in Eastern Eurasia (East Asia and Australasia) N might have gone a bit ahead or at least a bit more distinctly in relation to R, while in South and West Eurasia that is clearly not the case and they must have migrated together in an R-dominated wave, that retained some lesser N within it.
But I bet you'll make a lot of noise on this, after your due sequence of intentional misunderstandings of meanings and intent, so I wonder why did I open my mouth and even why do I bother.
"N and R display no greater overlap with each other than do the other examples I mentioned".
N and R overlap in all Eurasia and are both found in all continental regions. WTF?!
You are in denial but you can't really dare to deny something so obvious, can you?
"... all of which have a long tail but most likely coalesced where they are found today".
Maybe, may... but in any case they must have arrived from the origin of N and that is a dotted line from the Gulf of Bengal no matter where you make them coalesce.
Unless you can challenge the coalescence of N in SE Asia or Bengal, and you positively cannot, you still have them migrating from the N urheimat to their destinations. Get the ruler, get the map and draw the lines!
"We can't even take for granted that M's expansion is older than that of N".
I think it all indicates it is. Otherwise N subclades would be much more robust and widespread, specially in East Asia. It might be that the N node is roughly simultaneous to the expansion of M in East Asia (or it may be an effect of Toba later on) but the size of the N star and the fact that it is N the one that has to carve its niche in an otherwise M-dominated landscape strongly supports the M-before-N chronology we can infer from just counting the mutations downstream of L3.
"But I bet you'll make a lot of noise on this, after your due sequence of intentional misunderstandings of meanings and intent, so I wonder why did I open my mouth and even why do I bother".
ReplyDeleteI certainly wonder why I bother. It is you who are making the noise and manipulating the data here.
"I think it all indicates it is. Otherwise N subclades would be much more robust and widespread, specially in East Asia".
It makes more sense that they have been outnumbered by a later expansion of M haplogroups rather than that they are the product of minor haplogroups that expanded along with R, or even along with M.
"it is N the one that has to carve its niche in an otherwise M-dominated landscape strongly supports the M-before-N chronology"
It would be very difficult for a later expanding haplogroup to establish itself within the geographic region of a previous haplogroup, unless the incoming haplogroup has some technological, cultural or ecological advantage over the earlier one. This is unlikely to be the case for N as against M.
"So, if we are to insert M4'67 in the overall M expansion, we have to do it the same as other locally fruitful clades with similar stems, such as M1'2'51, M3a, M3c, M5, M9, M12'G, M13'46'61, M23'75, M'29'Q, M32'56, M33, M34'57, M35, M39'70, M40'62, M42'74, M49 and M73'79".
True. And such is actually quite possible. You seem to believe that M expanded from somewhere in India. M1'20'51 is spread from Africa through India to East Asia. M3a and M3c are both widespread in India. M5 is found in Southern India and Orissa. M9 is found from Northeast India to Laos and East and SE Asia, and so on. So all those haplogroups you list could well have taken part in the 'original' M explosion. I'm not claiming that they did so. I'm merely pointing out that to claim an embedded flow for just one haplogroup simply because it suits your purpose to do so does not demonstrate a consistent view of the evidence.
"However all these clades have only localized, regional at best, spans, so they CAN'T have expanded simultaneously with the main M pulse, which is multi-regional".
I agree that those haplogroups probably coalesced locally from a widely spread basal M haplogroup. Some were able to expand later independently. However I also suggest that the various N haplogroups coalesced locally from a widely spread basal N haplogroup. Some (including R) expanded later independently.
"Unless you can challenge the coalescence of N in SE Asia or Bengal, and you positively cannot, you still have them migrating from the N urheimat to their destinations".
I certainly 'challenge the coalescence of N in SE Asia or Bengal' because it is absolutely unlikely N reached that region by air. As for 'you still have them migrating from the N urheimat to their destinations', isn't that the most logical explanation?
"regional clades one mutation under the root: N1'5, N9, N11 and S. Same as before: they CAN'T be simultaneous to the main pulse of N because they are restricted to certain regions".
Why can't they have coalesced from basal N in the region where they are now found? Just like the M clades seem to have done? Besides which I'd be reluctant to extract the N haplogroups with long tails and demand special consideration for them. Presumably they too coalesced from basal N in the region where they are now found.
"within R we have no multi-regional, pan-Eurasian subclades either, but rather localized ones - one step under the root: R0, R2'JT, R6, R11'B6, B4'5, R30, R31 and P".
ReplyDeleteYes. And as you say for haplogroups M and N: They 'CAN'T be simultaneous to the main pulse of R because they are restricted to certain regions'. They all coalesced from basal R in the region where they are now found, including those haplogroups with long tails. And including haplogroups R11'B6 and R4'5. But that spread cannot tell us where R coalesced originally. However it must have been somewhere within N's original geographic spread.
"N and R overlap in all Eurasia and are both found in all continental regions. WTF?!"
But, in contradiction, you do admit:
"we may find that in Eastern Eurasia (East Asia and Australasia) N might have gone a bit ahead or at least a bit more distinctly in relation to R, while in South and West Eurasia that is clearly not the case and they must have migrated together in an R-dominated wave, that retained some lesser N within it".
I think the most logical explanation in the east is not that 'N might have gone a bit ahead or at least a bit more distinctly in relation to R', but that it was already there by the time R arrived. In South and West Eurasia R moved far beyond N which, again, suggests that R and N haplogroups cannot have 'migrated together in an R-dominated wave'. R bypassed the N haplogroups already in SW Asia. Besides which its difficult to envisage any 'minor' haplogroup surviving any long distance migration embedded within a major haplogroup yet still being able to later expand rather than being eliminated along the way.
Concerning your use of the term 'Australasia', earlier you said:
"Yet nobody rises these issues about marginal remote islands and such... only you".
You don't try to unravel the threads of a garment by starting in the middle. The edges can give us valuable information regarding the spread of haplogroups if we're prepared to actually look. For example Dienekes has put up a blog concerning Y-DNA O3 which is revealing regarding the timing of O3a2c*'s expansion, which in turn is revealing concerning O3a2c1's expansion through China, Tibet and Japan. It cannot be anything other than Neolithic.
"N and R overlap in all Eurasia and are both found in all continental regions. WTF?!"
ReplyDeleteInteresting comment. What about beyond Eurasia? As far as I'm aware no N haplogroups made it into New Guinea, especially not into Oceania, yet members of R certainly did. And as far as I'm aware no N haplogroups made it into Africa, yet members of R certainly did.
"It makes more sense that they have been outnumbered by a later expansion of M haplogroups"...
ReplyDeleteWHY?
Just because you say "it makes more sense..." it does not make more sense automatically. You need reasons, good ones.
It may make sense to you because you have already certain scheme in your mind but I see no logic, I even struggle to understand what are you talking about altogether and how does this half of sentence fit with what follows.
I blame that in part in that you don't seem to feel the need to justify your ideas and that you have a limited grasp of how to expose a point in written form.
"It would be very difficult for a later expanding haplogroup to establish itself within the geographic region of a previous haplogroup"...
Exactly my point. That's why I think that M expanded first (among other reasons). And that's why N had such limited impact in South Asia (but also in SE Asia, even if it is its urheimat).
"... unless the incoming haplogroup has some technological, cultural or ecological advantage over the earlier one".
It's possible that this happened with N (I have already mentioned dogs) but it seems more tightly associated with the strong R "aftershock" than with the rather weak N "front wave".
"You seem to believe that M expanded from somewhere in India".
You sound hypocritically "surprised" about this.
"So all those haplogroups you list could well have taken part in the 'original' M explosion".
None did (in the strict sense you suggest) because none is scattered by all (not even most) of M scatter area (not even M's shortest stem scatter area). You can see that easily: I have sometimes compared M4'67 with R and they have some parallels but there is one big difference: R is everywhere while M4'67 (or whatever other subclade mentioned) is not. So R is very exceptional in that aspect and that demands an exceptional explanation.
...
...
ReplyDelete"... because it suits your purpose to do so"...
No. Because it asks for a special explanation. My "purpose" (which is nothing but discerning human prehistory) is not affected at all. My theories (which are not "purposes" but means) adapt to such new findings... they normally have no problem doing that (sign that they are good enough).
"I certainly 'challenge the coalescence of N in SE Asia or Bengal' because it is absolutely unlikely N reached that region by air".
N did not if it coalesced there, pre-N did and did so by either walking, boating or a combo of both methods.
That is what the geometry of its descendants' scatter strongly suggest. If N would have coalesced elsewhere, the descendants would say so by reason of their geography.
If we cannot agree on this, we won't agree on anything else. So it's surely better that we agree to disagree forever and part ways.
You should now focus on demonstrating (if you can) that N coalesced elsewhere. I do not think that the comments section of my blog is appropriate for that endeavor. I think you should do that in your blog - but first of all in your head because I have seen so far no hint that you can demonstrate or even positively argue such conjecture of yours.
It is not just by direct criticism, much less of the kind of "I don't think so" that a theory is demonstrated wrong, you should put forward a better theory instead. I have not seen that theory anywhere except sketched quite poorly here and there in our discussions. You should focus on developing that theory and proving it.
"Why can't they have coalesced from basal N in the region where they are now found?"
In fact it's what I think. But between the first N woman and the first S woman there are n generations (hunch: 100-300) these generations of real women walked/boated from the N urheimat to Australia. And that is my whole point: that the N urheimat, the place where the first N woman, the N matriarch, lived at is not "everywhere" but in a very well localized area that I have estimated could be somewhere between Bangkok and Calcutta. The N matriarch (and her people) did not live at the same time in most Eurasia and Australasia: being a most important ancestor does not make her an omnipresent goddess.
"But that spread cannot tell us where R coalesced originally".
ReplyDeleteYes it can: by the method I have used several times of estimating the centroid of the geometrical scatter (each basal sublineage weights the same). It is in fact the only reliable method of estimating the origin of any haplogroup with any pretense of scientific method.
Said that, it's not rocket science either, so the idea of correcting the result to compensate for possible directionality of the expansive drive by means of moving the corrected result towards the centroid of the ancestral node maybe 1/4 makes sense and is a good idea that I support.
"But, in contradiction"...
For you every subtle shade is a "contradiction". I know that there is apparent "contradiction" but only for the shallowest B&W of possible interpretations, yet you choose to adopt such a shallow approach once and again, not conceding that I may need to shade my statements and conclusions, that a lemon is yellow but also green, that a flavor may be bitter and sweet at the same time.
What I say is that R and N overlap and did much of their journeys together but maybe not strictly so in all cases and all locations. There was maybe an N moment (without R) and an R moment (with less N but still with some N) in that wave but that it is still very difficult if not impossible to separate both in the strict way you'd love to do.
"In South and West Eurasia R moved far beyond N"...
This is madness: if we divide Eurasia in two at the Bay of Bengal (where R surely coalesced, as did N), R seems to be pushing behind N(xR) (yes, it does look like that and not just moving "beyond"), yet one step beyond it does not and lives with it side by side. But I do not see that R moved beyond N(xR) anywhere: you have N1 and I in Europe since ancient times, you have lots of N in Australia...
Something we have to understand also is that R and N(xR) are not fully comparable: R being descendant of N is comparable separately with each of the other N subclades. And then we also have to understand that each local founder effect is to some extent random and that it could have been otherwise (so we can only consider patterns that repeat all around and not just are imagined to happen in Europe or wherever, and nowhere else).
The more you discuss the matter the more clear is to me that R is part of N's expansion to the core.
...
...
ReplyDelete"Besides which its difficult to envisage any 'minor' haplogroup surviving any long distance migration embedded within a major haplogroup yet still being able to later expand rather than being eliminated along the way".
Drift needs time and this expansion was fast enough. Drift also needs contraction or stagnation to be most effective and this was an expansion in any case. Also we do not know how many lineages were pruned anyhow (probably many). U8a has survived in Europe since the Aurignacian, so they say, and is carried by a handful of people only. Shouldn't have drift removed it? Instead of just 2000 years it's been 40,000 and yet...
I have no idea what you say re. O3.
"What about beyond Eurasia? As far as I'm aware no N haplogroups made it into New Guinea, especially not into Oceania, yet members of R certainly did".
Actually N(xR) is dominant in Australia. When I said "Eurasia" I also meant the island-continents that depend on it phylogenetically: Australasia and America. In both cases we see both N(xR) and R. The details may vary a bit but the overall picture does not. America would not even be important for our debate, as it was colonized only later on, yet it does fit in as well.
"And as far as I'm aware no N haplogroups made it into Africa, yet members of R certainly did".
X1 is an essentially African haplogroup. U6 is the only R-derived African-specific haplogroup (HV, K, etc are shared with Europe and West Asia, same from N-derived W, etc.) Even in North Africa the pattern does persist, even if this is just a subset of West Eurasia.
You're beating a dead and rotten horse. And it stinks!
"You're beating a dead and rotten horse. And it stinks!'
ReplyDeleteYes. I agree. You are a dead and rotting horse.
"None did (in the strict sense you suggest) because none is scattered by all (not even most) of M scatter area (not even M's shortest stem scatter area)".
Yopu obviously have no idea how haplogroups become distributed. Haplogroups M1, M2 etc. did not all expand from a single region. Haplogroup M was what expanded. The subclades developed through drift from that 'original' expansion. You seem dimly aware of this as you write, 'Drift needs time and this expansion was fast enough'.
"Drift also needs contraction or stagnation to be most effective and this was an expansion in any case".
But it's virtually impossible that this 'drift' could occurr while the haplogroup is moving. It will only happen once the haplogroup has become isolated.
"But between the first N woman and the first S woman there are n generations (hunch: 100-300) these generations of real women walked/boated from the N urheimat to Australia".
I doubt it. If that was the case S would not be confined to Australia. We should find small pockets of S scattered along the way. It was basal N that got to Australia. S coalesced there.
"No. Because it asks for a special explanation. My 'purpose' (which is nothing but discerning human prehistory) is not affected at all. My theories (which are not 'purposes' but means) adapt to such new findings... they normally have no problem doing that (sign that they are good enough)".
ReplyDeleteI'm pleased that you concerned with 'discerning human prehistory' because you have given no indication that you are really interested in that subject. And yopu are certainly reluctant to 'adapt to such new findings'. Coatal migration, anyone? You have a pre-conceived belief and you make every effort to manipulate the data to fit that belief. As shown by:
"I know that there is apparent 'contradiction' but only for the shallowest B&W of possible interpretations"
Shallowest of contradictions? Total contradiction more like. The only reason you use contradictory criteria when considering the R haplogroups with a tail of just one as against the other N haplogroups with a tail of just one is that you are trying to fit the evidence to your pre-existing belief. That is also the reason why you are forced to adopt contradictory criteria when considering the N haplogroups with long tails as compared to all the M and R haplogroups with long tails. Such inconsistency is necessary because you're committed to maintaining your belief no matter what. Surely it is better to examine the evidence first with an open mind and then come up with an hypothesis that explains that evidence in a consistent manner rather than having to invent reasons why the evidence fails to fit a particular belief.
By your own logic, using mutation tail length, N1'5 would have separated into N1 and N5 in NW India/SW Asia at virtually the same time as R was beginning to expand. Likewise S was diversifying considerably in Australia just as R was beginning to expand (wherever that was from). Likewise with with N9, probably somewhere near North China. So N was already widespread by the time R bagan to expand.
You seem to agree that R's distribution strongly suggests that it's expansion was later than that of M. Except perhaps for the northeastern and southeastern Eurasian extremities where M haplogroups may have expanded along with R haplogroups. Perhaps R replaced various indigenous M haplogroups, but I suspect R largely just added to the existing population numbers in each region.
Similarly M's distribution strongly hints that its expansion was later than that of N. That latter haplogroup survived scattered around M's periphery. But again rather than full scale replacement I suspect M added to the pre-exising population numbers.
So rather than just two major early basal mt-DNA haplogroup expansions we actually have three: M, N and R.
"But it's virtually impossible that this 'drift' could occurr while the haplogroup is moving. It will only happen once the haplogroup has become isolated".
ReplyDeleteI don't know: there are no "haplogroups in movement" but people (most of which is not tested for mtDNA). Recently a Europe-only haplogroup (H20a) was found in an early Neolithic sample in Catalonia (I'm following Argiedude in this). Most H20a is in Iberia, while its predecessor H20(pre-a) (in this case clearly the root, as it's lacking one mutation) is in West Asia. This means that the H20a "grandmother" either:
1. Lived in Iberia (or nearby) and expanded from there in several directions (Bavaria, Greece, etc.), where it's found on occasion.
2. Lived in Greece but had few offspring, as H20a only really shows sign of expansion once in Iberia.
3. Lived in West Asia but all the descendants carrying her lineage today are, because of mere chance (drift+founder effect) in Europe.
There are maybe other options but these are the basic ones. Which one is the correct one? I don't know about you but I can't decide easily (2 is my favorite however but it's just an intuition, not any reasoned decision).
"If that was the case S would not be confined to Australia. We should find small pockets of S scattered along the way".
No because S evolved (possibly) only once in Australia but the 100-300 N-root (pre-S) women still had to travel the journey between Bengal/Cambodia and Australia. Alternatively S surviors in Sundaland and Wallacea may have been "drifted out" by other haplogroups (or if you wish hunted and cannibalized by the carriers of these "winner" lineages).
Notice that I say "pre-S" but they could well also be "pre-O" or whatever because we are talking of undifferentiated N-root haplotype here. It is perfectly possible that there was a shared (N haplotype) "grandmother" of S and O but that this one left no track in the phylogeny.
Similarly we can imagine, as something plausible but indemonstrable, a shared "grandmother" (descendant of the N matriarch and with the same haplotype) for other regional groupings, even if there is no genetic marking.
We can imagine a lot of things where we have no info about, we can't demonstrate them however.
"Shallowest of contradictions?"
ReplyDeleteShallowest of interpretations, which you insist on.
"By your own logic"...
You are not interested in understanding my logic, because it is much richer and somewhat more complex than what you are willing to consider.
Mutation number tells of probability of timing but probability is not fact. Probability says that if I throw a coin twice I should get a head and a tail but, as often as not, this is wrong in real life (also because of probability).
There is no rigid tic-tac, just a statistical likelihood. I have used this probable (but not necessarily correct in all cases) tic-tac to outline a map of the overall human expansion in Eurasia-plus (and separately in Africa as well). Because the number of clades and samples involved is usually high, it's probable that the inference is more or less correct as a whole (irregularities should cancel each other, Chaos allowing).
But this general analysis is not necessarily correct for each singular step: a single derived mutation may have taken one generation or thousands to happen. Even if my low estimate average of 125 generations is correct (what I'm not even really sure about), the actual span of each single mutation may vary a lot.
So singular items should always be considered with precaution and an open mind. Of these singular items three are outstandingly important and get the conventional name of macro-haplogroups: M, N and R.
M and N are not enough in number to be reasonably sure about their timing only based on the 'tic-tac' of the mutational sequence, what forces us (at least me) to consider the whole picture with due care. For example I'm pondering that N may compare better with M+1 than with the M+2 position it mathematically falls in.
But comparatively this is a lesser issue. More puzzling is the role and status of R within the N expansion. Sure, if you choose to adhere to the "tic-tac" approximation rigidly, fanatically (and irrationally), then R is not different and should not go before any other N subclade. But the "clock" can't be adhered so strongly to: it's just an estimate for the whole process that loses much of its relevance when analyzing singular points: the confidence interval expands massively, and so does the uncertainty.
So we have to look at the problem comprehensively and in the R case there are other factors that matter, specially the geographical scatter and relevance.
By the way, while the issue of R is only dealt with secondarily (but also mentioned), I addressed yesterday the matter of the origin of N here (you may have missed it).
"while its predecessor H20(pre-a) (in this case clearly the root, as it's lacking one mutation) is in West Asia. This means that the H20a 'grandmother' either"
ReplyDeleteSurely there is no problem here. H is spread widely through the whole region and H20 most likely entered Iberia by boat through the Mediterranean. The example is completely different from what you're claiming for N.
"S evolved (possibly) only once in Australia but the 100-300 N-root (pre-S) women still had to travel the journey between Bengal/Cambodia and Australia".
So did O, N13 and N14 but you insist that these haplogroups with longer stems somehow behaved differently from those with short stems. On what grounds to you claim this to be so?
"they could well also be "pre-O" or whatever because we are talking of undifferentiated N-root haplotype here".
That's exactly what I said yesterday. The same goes for N1'5, N2 and X. They were undifferentiated N-root haplotype when they arrived in SW Asia. They would not have moved all the way from Bengal/Cambodia already differentiated. In fact it's most unlikely that their ancestors were ever anywhere near Bengal/Cambodia.
"Similarly we can imagine, as something plausible but indemonstrable, a shared 'grandmother' (descendant of the N matriarch and with the same haplotype) for other regional groupings, even if there is no genetic marking".
I'm sure that will turn out to be the case for M at least. I suspect that just a limited number of M haplogroups made it beyond Assam/Burma/South China for example. And possibly even into South India.
"You are not interested in understanding my logic, because it is much richer and somewhat more complex than what you are willing to consider".
I cannot understand your logic because it is confused.
"Mutation number tells of probability of timing but probability is not fact".
And you consistently use mutation number to indicate timing when it suits your purpose and consistently ignore it when it doesn't suit your purpose.
"Even if my low estimate average of 125 generations is correct (what I'm not even really sure about), the actual span of each single mutation may vary a lot".
I agree, but I don't shift position depending on what I'm claiming.
"I'm pondering that N may compare better with M+1 than with the M+2 position it mathematically falls in".
To me it is easier to explain an arrival in Australia that misses New Guinea than it is to explain the opposite. That would mean that N arrived in island SE Asia before M did. However if M is land-based and N coastal-based we can still have N younger than M if N bypassed M. However we cannot have N coastal-based along India. Doesn't fit.
"More puzzling is the role and status of R within the N expansion".
You're making the unjustified assumption that R is embedded in the R expansion. Try separating them, and then examine the haplogroup structure.
"in the R case there are other factors that matter, specially the geographical scatter and relevance".
Especially the lack of any sort of stem within the R11'B6/R4'5 clade. But you're very unwilling to apply your own method in regard to that problem.
"I addressed yesterday the matter of the origin of N here (you may have missed it)".
I had missed it when I commented here yesterday but obviously I have since seen it.
"H20 most likely entered Iberia by boat through the Mediterranean"...
ReplyDeleteWhether it entered by boat is trivial, my whole point is that there can be such coalescences "on route", even if the route is relatively short, and that we can't discern what happens between the ancestral and descendant nodes if no branchings exist.
Why do you keep misunderstanding everything?
"but you insist that these haplogroups with longer stems somehow behaved differently from those with short stems".
A long stem is an state of uncertainty: it is NO evidence of anything. It can (the stem) coalesce in situ (because of pressure by other clades) or it can coalesce prior to migration (the expansion happening upon arrival to a new "virgin" land, which is the case of M and probably also A, X and others) or it can never coalesce as true finished haplogroup, remaining as private, tiny, lineage (what is the case of those Australian lineages you brandish), which is the less informative of all cases and says nothing about themselves after the N node at all.
"The same goes for N1'5, N2 and X. They were undifferentiated N-root haplotype when they arrived in SW Asia".
Yes but only X clearly coalesced in West Asia 'sensu stricto' of all the three. N2, with it's Siberian and Pakistani presence, looks more like Central Asian ("Afghan", even if some W, and even N2, is also found in West Eurasia proper), while N1'5 is so 50-50 that we must also declare it "AfPak".
So they did arrive at West Asia 'senso stricto', most probably, as N* (pre-X), N1'5 (pre-N1), N2a and W.
But they did arrive to South Asia as N*, very possibly as minor private lineages within the "Western R" migration.
"And you consistently use mutation number to indicate timing"...
Not so consistently! If you'd argue that N and M are simultaneous I cannot discuss it only based on mutational stems, because these are uncertain enough to be inconclusive. I must look at other patterns such as the fact that M expanded initially much more strongly (huge star-like structure) than N, what suggests that N was much constrained by M, which must have therefore expanded first.
But there are quite striking patterns, affecting not one but several clades, that are almost compulsory to accept as indicative of a geographic-chronological sequence. Hence if we have 7 N subclades at mutational steps (MS) numbers 1 and 2, then a gap at MS 3 and then three scattered subhaplogroups at MS 4, I see a pattern. More so when this pattern shows a "concentric circles" structure from the purported origin of N.
And this pattern is as close as we can get to evidence within the mtDNA phylogeny (other types of genetic data have other problems and are not better anyhow).
"... the lack of any sort of stem within the R11'B6/R4'5 clade".
I'm not sure what you have in mind here other than nagging but I'm sure that R11'B6 and R4'5 are being considered by me as they deserve and they count not just as one but as two haplogroups. They still can't outweigh the other 17.
"my whole point is that there can be such coalescences 'on route', even if the route is relatively short"
ReplyDeleteand we can easily follow the trail, and perhaps even discern whereabouts along that trail that the haplogroup coalesced. In this case H20a seems most likely to have colalesced in the Balkans, from an incoming U20. The route to Polynesia is full of such coalescences along the way. In all cases we can easily discern where the particular subclade coalesced. The same cannot be said for haplogroup N if you insist on moving it through India. Twice.
"It can (the stem) coalesce in situ (because of pressure by other clades) or it can coalesce prior to migration"
Far more likely to coalesce in situ. Most private lineages would rapidly become extinct if they moved at all far.
"the expansion happening upon arrival to a new 'virgin' land, which is the case of M and probably also A, X"
I agree in the case of M, but it was probably part of a population spread originally from somewhere between Africa and the Iranian Plateau. But both A and X coalesced from basal N somewhere within the region they are now found. They would not have migrated far as 'haplogroup A' or 'haplogroup X'.
"Yes but only X clearly coalesced in West Asia 'sensu stricto' of all the three. N2, with it's Siberian and Pakistani presence, looks more like Central Asian ('Afghan', even if some W, and even N2, is also found in West Eurasia proper)"
Thank you.
"while N1'5 is so 50-50 that we must also declare it 'AfPak'.
And certainly neither pre-N1'5 nor pre-N2 ever moved through 'India'. They coalesced from N somewhere within the region they are now found. As you say, 'they did arrive at West Asia 'senso stricto', most probably, as N* (pre-X), N1'5 (pre-N1), N2a and W'. But where is the evidence for an SE Asian origin for them? Let alone a migration through India from SE Asia.
"But they did arrive to South Asia as N*, very possibly as minor private lineages within the "Western R" migration".
As I pointed out elsewhere, there are no N haplogroups shared between SE or East asia and India. So how do you reach that conclusion?
"I must look at other patterns such as the fact that M expanded initially much more strongly (huge star-like structure) than N, what suggests that N was much constrained by M, which must have therefore expanded first".
ReplyDeleteThere is another, and more obvious, conclusion but you are determined to dismiss it because it conflicts with your version of reality. M and N did not expand together, so N was not 'constrained by M' at all. They didn't cross paths until R had formed from N in SE Asia.
"Hence if we have 7 N subclades at mutational steps (MS) numbers 1 and 2, then a gap at MS 3 and then three scattered subhaplogroups at MS 4, I see a pattern".
But you see a very strange pattern, one that doesn't make sense. You have a series of migrations, depending on the number mutational steps, all starting from a single point. But often later migrations finish up in a region already occupied by a related haplogroup. Surely it's far more likely that these differing mutational steps tell us nothing about timing of migration. Just the timing of eventual expansion, if applicable.
"I see a pattern. More so when this pattern shows a 'concentric circles' structure from the purported origin of N".
Have another look at your own map, especially if you've shifted A and R to their more likely coalescence sites. You don't see 'concentric circles' at all. What you see is a narrow line from Australia to Palestine, missing India completely until R sets out on its migration.
"I'm sure that R11'B6 and R4'5 are being considered by me as they deserve and they count not just as one but as two haplogroups".
Three actually. Don't forget about R24 in the Philippines.
"Whether it entered by boat is trivial, my whole point is that there can be such coalescences 'on route', even if the route is relatively short, and that we can't discern what happens between the ancestral and descendant nodes if no branchings exist".
ReplyDeleteYopur comparison is irelevant anyway. H20a can hardly be said to have a 'long tail'. It's only one control region mutation from H20.
"Far more likely to coalesce in situ. Most private lineages would rapidly become extinct if they moved at all far".
ReplyDeleteThere's no difference: it goes with a private (thin) line of mother-daughter and it makes no difference if some of these migrate or not. What happens with private lineages and long stems in general is that they are constrained and can't expand. Why? Mostly because they are under pressure from their companion lineages, which typically began with greater numbers and hence tend to win the "drift race" once and again. Many, as you say, don't survive, but some do.
This pressure can happen "on the road" through a landscape that does not allow for meaningful expansion or in a constrained eco-social niche (your typical remote New Guinea valley, for instance) but it's almost invariably because of others' pressure. Physical islands are not the rule in the realm of humans but an exception: impossibility of expansion is caused mostly by competition with peers.
"I agree in the case of M"...
You should also agree for all populations showing signs of expansion after similarly rather long stems, such as A, X, N2...
"M and N did not expand together, so N was not 'constrained by M' at all".
Then by whom? Homo erectus?
You are just stating once and again your ideas but NEVER providing any evidence that supports them: you drew your version of prehistory with your imagination and NOT with the scientific method. And now you try to bend reality to fit with your story but it does not fit no matter what you do.
A normal person would surrender but you charge once and again, yet it's always the same song: no evidence, no method, it just "must" be the way you imagined it.
"But you see a very strange pattern, one that doesn't make sense. You have a series of migrations, depending on the number mutational steps, all starting from a single point. But often later migrations finish up in a region already occupied by a related haplogroup".
Those are probably not "later migrations" but lesser lineages in the same migration, which had enough surviving their peers' competence.
Some lesser lineages did not survive in those areas but, before totally dying off, managed to expand in new frontiers, outside the primary "cloud". Others did not survive at all but these left no evidence.
"Have another look at your own map, especially if you've shifted A and R to their more likely coalescence sites. You don't see 'concentric circles' at all".
I see no need to move R to anywhere else and I don't see how the exact location matters at all. You want to think that the origin of N and R is in Burma instead of Bengal? I don't care: it makes no practical difference, and it would make no big difference either if in Cambodia.
"What you see is a narrow line from Australia to Palestine, missing India completely until R sets out on its migration".
If you'd be correct about this, what is not the case (Altai should have its own signature and has none), it'd be, if anything, a migration from SE Asia to West Asia via Altai and not vice versa.
But, considering that we know pretty well that Altai UP is derived from further South/SW and inserted in a process initiated in South Asia, we cannot make any sort of East-to-West connection via the steppe (nor vice versa either, mind you). Y-DNA-wise we also see P migrating through South Asia into West Asia (Q) and then Altai, which must be at the origin of the West Eurasian lineages (Y-DNA Q and mtDNA X2) we find in Siberia and America.
You know all that, why do you insist in confusing what is obvious? Beats me.
"Yopur comparison is irelevant anyway. H20a can hardly be said to have a 'long tail'. It's only one control region mutation from H20".
It applies to all the nonsense you were arguing about short-stemmed clades like S.
"It applies to all the nonsense you were arguing about short-stemmed clades like S".
ReplyDeleteExactly. That's why we see some of S's close relations in Australia too, and why we see others of its relations scattered through Indonesia, Malaysia and South China. But, ominously for your belief, we see no N(xR) haplogroups in India. 0nly R-derived haplogroups. The nearest are N1'5 in Pakistan and N8, N10 and N11 in South China. Surely, if N had moved through India at any time we should expect to see at least one N haplogroup. Especially as around 8 or 9 R haplogroups are present in the subcontinent and so that haplogroup had no trouble surviving.
"What happens with private lineages and long stems in general is that they are constrained and can't expand. Why? Mostly because they are under pressure from their companion lineages"
Possibly. Occasionally. But far more likely is that thye become isolated within a relatively small region and drift leads to the long-term survival of just on haplogroup. The number of haplogroups within any population of fixed size will reduce over time. If a migrating population consists of several haplogroups of varying proportions those with a smaller proportion will usually die out along any route they may take.
"Many, as you say, don't survive, but some do".
There is a saying here, and I presume you have similar ones there, 'apples never fall far from the tree'. Descendant haplogroups are very likely tro be found near their ancestors. You cannot demonstrate a single instance where we can be sure that a haplogroup's long tail developed while it was on the move.
"This pressure can happen "on the road" through a landscape that does not allow for meaningful expansion or in a constrained eco-social niche (your typical remote New Guinea valley, for instance)"
And that is presumably the reason for the surprising diversity of M haplogroups in New Guinea. The Australian?New Guinea haplogroups with long tails did not develop them on the route to Sahul. They developed them once they reached there.
"You should also agree for all populations showing signs of expansion after similarly rather long stems, such as A, X, N2..."
I do agree [the expansion happening upon arrival to a new 'virgin' land] that those haplogroups developed their long stems before they expanded, but they developed them somewhere within the region they are now found, not in Bengal/SE Asia, especially so as no N(xR) haplogroups are found anywhere near Bengal. And R's expansion certainly looks as though it is through 'a new 'virgin' land' for some aspect of its ecology. Which suggests further that it is an immigrant both through India and along the East Eurasian coastline.
"You are just stating once and again your ideas but NEVER providing any evidence that supports them"
ReplyDeleteYou can provide no evidence at all from haplogroup distribution in support of your ideas. No N(xR) in India, and any sort of analysis of the haplogroup data shows that the migration from India to SE Asia cannot have been coastal. South China and Laos share many M haplogroups with Northeast India. And Vietnam, through which we would expect any such migration to have passed, has no basal N or M haplogroups. What it does have is plenty of R haplogroups.
"you drew your version of prehistory with your imagination and NOT with the scientific method".
Incorrect. I drew my version od history using what I already knew about Polynesian origins along with the data provided in the 2005 McDonald maps. I was not diverted by any imaginary 'great southern coastal migration theory'.
"it's always the same song: no evidence, no method, it just 'must' be the way you imagined it".
I see no need to alter my song. It still fits the data perfectly. You, on the other hand, are forced to conjure up disappearing lineages across huge swathes of the earth's surface.
"Those are probably not "later migrations" but lesser lineages in the same migration"
That's exactly what I have claimed all along. It is just that you see R and N as having migrated together whereas I don't see their respective migrations coinciding at all. Especially through India, the region you claim as an origin for both haplogroups.
"I see no need to move R to anywhere else and I don't see how the exact location matters at all".
You are standing on shaky ground when you insist on claiming an Indian origin for R. Shifting its centre and that of A would alter your map considerably.
"If you'd be correct about this, what is not the case (Altai should have its own signature and has none)"
I keep reading about some sort of mysterious 'Ket' N. And members of haplogroup A are certainly spread through much of Central Asia, not just NE Asia as you have it in your map.
"considering that we know pretty well that Altai UP is derived from further South/SW"
On what grounds do you insist on associating the Upper Paleolithic expansion with the fist modern human expansion? most scientists are more than happy to separate them.
"Y-DNA-wise we also see P migrating through South Asia into West Asia (Q) and then Altai, which must be at the origin of the West Eurasian lineages (Y-DNA Q and mtDNA X2) we find in Siberia and America".
Yes. But that migration picked up members of other haplogroups already there, such as A, C and D, and perhaps B, and carried them to Siberia and America too.
N1'5 (N5) is found in India, as is N2 (W), that I have drawn their estimated centroids in the AfPak area is not the same as they being lacking in India republic, much less as I have not even applied the 1/4 (?) directionality correction, which would in any case push them towards Bengal, i.e. they are drawn as raw uncorrected centroids (always estimated).
ReplyDelete"the surprising diversity of M haplogroups in New Guinea"...
It is much lower than in South or East Asia, so no idea why you talk of "surprising diversity". What we find is not "surprising" for me at all.
"I do agree [the expansion happening upon arrival to a new 'virgin' land] that those haplogroups developed their long stems before they expanded, but they developed them somewhere within the region they are now found"...
This sentence is a total oxymoron the first part and the last part are totally contradictory.
"... not in Bengal/SE Asia"...
I did never claim so. I suggested "on the road", i.e. in between origin and destination, loosely.
And as I always say, one could well argue for "flying saucer abductions" as well because we can't discuss what we have no data about. Yet you insist on arguing about this void of data, probably because you can somehow use it as pseudo-evidence by imagining what might have happened - instead of acknowledging that we do not know.
"apples never fall far from the tree".
That's similar to what I said about fleas, however, unlike apples and trees, people do move.
"Descendant haplogroups are very likely to be found near their ancestors".
Exactly my point and the whole point of the other article on the origin of N, which you insist to contest without any ground.
But people do move and they in more they have greater chances to move farther and to accumulate more mutations, so, if any, short-stemmed haplogroups weight more. Although I made all basal haplogroups to weight exactly the same, we get effectively the same results if we only weight the short-stemmed "elder" subhaplogroups.
"You cannot demonstrate a single instance where we can be sure that a haplogroup's long tail developed while it was on the move".
In a sense I have demonstrated it for N long-stemmed subclades, by following your "apples never fall far from the tree" logic: most "apples" (by large) are around SE Asia, not West Asia. And, incidentally, those which are farther from the inferred "tree" have relatively long stems, indicating that some greater time (and hence more events, such as migratory episodes) must have happened in between the parent "tree" and the new "tree" that signals where once an "apple" germinated. These "apples" have legs anyhow, mind you.
So it looks a lot like these mutations evolved "on the move". Positive "sigma 5" demonstration is extremely hard to do but it's a quite solid inference that you cannot prove false either.
"You can provide no evidence at all from haplogroup distribution in support of your ideas".
I contend that haplogroup distribution is the only evidence we have; your fantasies are not evidence.
"No N(xR) in India"
Wrong: N5, N1 (N1d, N1a, I1), W (W1c, W3 and W*) and N2* (N2a?) do exist in India, according to the data of Palanichamy 2004, never mind Pakistan (which is part of broader subcontinental India). The only "Western N" that is not found in the Republic of India is X.
...
...
ReplyDelete"I drew my version od history using what I already knew about Polynesian origins"...
Polynesians are trivial in regards to Paleolithic. Whatever the case, looking only at them has obviously given you an extremely narrow idea.
... "along with the data provided in the 2005 McDonald maps".
We have all used those maps as reference and, curiously, joining the dots on those maps once and again brought me to India, so often that I had problems to write down the letters on that corner of Asia.
The problem is that you do not use centroids. And also that you have come to such rigid conclusions that you can't change your mind even if drowning in the piles of evidence that contradicts them. You have gone in crusader-berseker mode against the coastal migra theory just because you have a priori decided that boats were invented only and never before in Wallacea.
Bullshit!, I say. That's such a low quality manure that is not even worth to use for fertilizer.
"I was not diverted by any imaginary 'great southern coastal migration theory'".
You were indeed, just that by taking a stand of denial, of fundamentalist, extremist, irreducible opposition to this idea. Of course you have been diverted... a lot! You have been diverted to the point of making extremely irrational claims such as people having to climb mountains only to wade a river and such.
Nothing but a pile of manure!
Not significantly better than "out of America".
And you have chosen my blog to defend that idiocy. Day after day, for years! Shut up!
"You are standing on shaky ground when you insist on claiming an Indian origin for R".
You don't even say why you'd claim that. Bengal is more or less the centroid produced by the geometry of R basal subclades. Have you even bothered to estimate it? Nope.
"I keep reading about some sort of mysterious 'Ket' N".
It's N2a.
"And members of haplogroup A are certainly spread through much of Central Asia, not just NE Asia as you have it in your map".
The centroid still falls roughly on that area, there's no particular tendency towards Central Asia other than the general trend of scatter of NE Asian lineages under the recent Turco-Mongol expansion (even if it'd be older, A's diversity is still centered far away from Central Asia.
But you are splitting hairs anyhow, moving A's dot (which is not correct) would not change anything: it's just 1/15 of weight!
"On what grounds do you insist on associating the Upper Paleolithic expansion with the fist modern human expansion?"
In West Eurasia (incl. Central Asia but excluding Arabia/Palestine)? They are totally synonymous. Show me any evidence that says otherwise.
"But that migration picked up members of other haplogroups already there, such as A, C and D, and perhaps B, and carried them to Siberia and America too".
You completely misunderstand and mercilessly twist everything, just like the Russian guy of the "out of America" madness. After arriving to Altai from West Asia (where most internal diversity is), some men with Y-DNA Q migrated to northeasternmost Asia, to Beringia, getting some East Asian mtDNA lineages (i.e. real women in practical terms). MtDNA X2 in America is residue of the West Asian origin of this flow, while Y-DNA C3 attests to the NE Asian incorporation also on the Y-DNA side.
"It's N2a".
ReplyDeleteThanks.
"that I have drawn their estimated centroids in the AfPak area is not the same as they being lacking in India republic"
Just because they are 'found' there today doesn't mean they 'coalesced' there.
"Wrong: N5, N1 (N1d, N1a, I1), W (W1c, W3 and W*) and N2* (N2a?) do exist in India, according to the data of Palanichamy 2004, never mind Pakistan (which is part of broader subcontinental India). The only 'Western N' that is not found in the Republic of India is X".
And no N haplogroups are found in Eastern India, particularly in Bengal.
"Yet you insist on arguing about this void of data"
Maju. You face a yawning chasm when it comes to data for Eastern India, yet you vehemently insist that you can see a pattern that relies on migration through that region.
"Exactly my point and the whole point of the other article on the origin of N, which you insist to contest without any ground".
Yet you insist that many N haplogroups moved huge distances from their ancestral haplogroup's homeland. Surely it was the ancestral haplogroup that moved, not the derived form. The derived form coalesced through drift in the parent population's geographic extent.
"In a sense I have demonstrated it for N long-stemmed subclades"
You have only done so in your imagination. Your demonstration involves a whole swag of special pleading.
"incidentally, those which are farther from the inferred 'tree' have relatively long stems, indicating that some greater time (and hence more events, such as migratory episodes) must have happened in between the parent 'tree' and the new 'tree' that signals where once an 'apple' germinated".
What a lot of rubbish. S has a long tail? And at the opposite end, N1'5 has a long tail? Your statement wallows in special pleading.
"The problem is that you do not use centroids".
I am quite happy to use centroids. In fact I have referenced them occasionally when the evidence fails to fit your belief. At such times you criticise centroids as being of little use. Make up your mind whether you are going to examine all the evidence in a consitent manner instead of altering your point of view depending on what you wish to 'prove'.
"You have gone in crusader-berseker mode against the coastal migra theory just because you have a priori decided that boats were invented only and never before in Wallacea".
ReplyDeleteI have gone on a 'crusader-berseker mode against the coastal migra theory' because it cannot stand any examination, and there has never been any evidence for it. Just wishful thinking. And a whole raft of weaknesses anyway. I didn't decide 'a priori' 'that boats were invented only and never before in Wallacea'. Just that such an hypothesis might explain the sudden expansion from SE Asia. An expansion, incidently, which you have recently begun to agree with in spite of your bitter opposition to the idea originally.
"The centroid still falls roughly on that area"
The 'centroid still falls roughly on that area' simply because it is weighted in favour of the more recent expansion eastward of haplogroup A5 and whatever you like to call the haplogroup that gave rise to American/Beringian A. A's dot should certainly be shifted further south that you have it. It is extremely unlikely it coalesced so far north.
"They are totally synonymous. Show me any evidence that says otherwise".
Any realistic OoA date predates the Upper Paleolithic by many thousands of years, perhaps by 50,000 years. I thought you already accepted that.
"After arriving to Altai from West Asia (where most internal diversity is), some men with Y-DNA Q migrated to northeasternmost Asia, to Beringia, getting some East Asian mtDNA lineages (i.e. real women in practical terms). MtDNA X2 in America is residue of the West Asian origin of this flow, while Y-DNA C3 attests to the NE Asian incorporation also on the Y-DNA side".
That is pretty much exactly what I said. And Y-DNA Q (or C3) also picked up D and B from East asia. But Q didn't go via South China to collect them. So what do you mean by:
"You completely misunderstand and mercilessly twist everything, just like the Russian guy of the 'out of America' madness"?
"You face a yawning chasm when it comes to data for Eastern India, yet you vehemently insist that you can see a pattern that relies on migration through that region".
ReplyDeleteI love the expression "yawning chasm" but the alleged lack of data (the region is poorly sampled) is not by any means worse than that the effective presence of negative data when it comes to your pet corridor of Altai. There is not a single haplogroup, at the basal levels we are dealing with here, which shows any sign of coalescence in Altai or Siberia. And almost not one nearby either. By the "apple tree" logic we'd expect most N subclades, notably the short-stemmed ones to be in the areas closest to the alleged origin, yet there's almost nothing at all.
"... a whole swag of special pleading".
ReplyDeleteQuite "literating" and look at the matter for what it is, look at the fact dispassionately if you can.
"S has a long tail? And at the opposite end, N1'5 has a long tail?"
No! And that's exactly why the origin of N MUST be between them and N9 (as they are extreme in the scatter for such short stems).
... "wallows in special pleading".
For all the "literating" slang, you got redundant, ahem.
"I am quite happy to use centroids. In fact I have referenced them occasionally when the evidence fails to fit your belief. At such times you criticise centroids as being of little use".
I've never criticized a properly estimated centroid, i.e. one based on phylogenetic diversity and not mere amorphous numbers. Emphasizing population (trivial) and ignoring diversity (crucial) is the error of all shallow analysis.
You know that (or should) so this is just another diversion.
"the alleged lack of data (the region is poorly sampled)"
ReplyDeleteNortheast India is 'poorly sampled'? You've got to be kidding! Countless numbers of M haplogroups are confined to the region.
"There is not a single haplogroup, at the basal levels we are dealing with here, which shows any sign of coalescence in Altai or Siberia".
A and X are reasonable possibilities. One to the west and one to the east.
"we'd expect most N subclades, notably the short-stemmed ones to be in the areas closest to the alleged origin"
Why would we expect that? I strongly suspect the opposite. Both M and N themselves have relatively long tails indicating the suffered a period of drift before their respective expansions. Surely we would half expect any members of either haplogroup who remained behind to continue suffering a similar level of drift. The ones who moved would be the ones who expanded and diversified.
"look at the fact dispassionately if you can".
You are certainly not looking at the facts 'dispassionately'. You have made up your mind in advance what you want the 'facts' to show and you are very determined that they will show it. To achieve that you have to use different spectacles to examine each individual morsel of data.
"And that's exactly why the origin of N MUST be between them and N9 (as they are extreme in the scatter for such short stems)".
But where 'between them' is the question. Unlikely to be Bengal. Besides which the haplogroup could have expanded from either end. It need not have expanded from the middle.
"I've never criticized a properly estimated centroid"
A 'properly estimated centroid' is easily defined. It is one that you agree with.
"Northeast India is 'poorly sampled'?"
ReplyDeleteAnother annoying misinterpretation: South Asia as a whole! It has improved though but every other semester something new is found, so there is a lot of top-level genetic diversity there.
"A and X are reasonable possibilities [for coalescence in Siberia]"
They are not. If you locate their basal descendants and estimate the phylogenetic centroid on the map, they belong to NE Asia and SW Asia respectively, and very clearly so.
Even if there would be two such lineages (which they are not), they could hardly compare with Australia or East Asia and hence the resulting centroid would always fall towards SE Asia. Never mind that the two lineages have "long" stems and hence belong to a period quite late in relation to the main N expansion.
"Both M and N themselves have relatively long tails indicating the suffered a period of drift before their respective expansions. Surely we would half expect any members of either haplogroup who remained behind to continue suffering a similar level of drift".
They suffered drift BEFORE arrival to their destinations. M did not suffer almost any drift once it arrived to South Asia, but a massive explosion instead (all drift happened in Arabia, where other "African" clades apparently survived instead). Your ideas hold no ground and make no sense.
"Unlikely to be Bengal".
Burma, Cambodia...
"Besides which the haplogroup could have expanded from either end. It need not have expanded from the middle".
"Granny N" was a normal human being not an all-pervading divine force: she must have lived somewhere, even if she was a great sturdy walker, she surely only moved traveled a few hundred kilometers between first and last pregnancy. Most likely she lived all her live in a very specific area we can't actually name out of uncertainty.
But well...
I repeat here what I said in the other post which you hijacked with debate on N even after I posted this entry for the purpose:
ReplyDeleteMan, it's been months, or rather years of this endless discussion in circles, almost every day. Even if now and then you make a criticism worth considering or introduce some information I may have missed, most of what you say is waste of letters. Mind you: reading your one-liners and finding the context and trying to understand and, normally, debunk them takes much of my time - too much. It's a discussion for the sake of discussion, which you may be fond of but I can only find tiresome and pointless.
I know your hypothesis (even if you have bothered explaining it formally nowhere) and I know that it does not hold at all. There is no "bread crumbs" trail from the origin of L3 but actually an explosion from a shared center after a "private level" (small) migration.
You may still disagree but you cannot persuade me unless you'd find more, many more, basal N subhaplogroups in West Eurasia and Africa. Considering that there are 11 of those along the Pacific Ocean, such a turnover is effectively impossible.
Do you want for N to migrate back to West Asia via Altai (the only possible alternative route to South Asia)? That might be a reasonable claim where we could disagree very reasonably and gentlemanly... but you want to revert the arrow of that flow, and that is extremely unreasonable, considering the evidence and the fundamental logic of the "apples" (with long legs) that requires that most lineages remain as close as possible to where the real origin (the "tree") was.
I know that you are bent onto torturing me and making me waste my time but I can only beg you to desist for the sake of not making me go moderator-bersek.
Thanks.
"I know that you are bent onto torturing me and making me waste my time but I can only beg you to desist for the sake of not making me go moderator-bersek".
ReplyDeleteI'm trying to help you by pointing out where your reasoning is incorrect so you can cease talking nonsense.
"Another annoying misinterpretation: South Asia as a whole! It has improved though but every other semester something new is found, so there is a lot of top-level genetic diversity there".
South Asia, including Northeast India, has been well sampled. Nearly 50 basal M haplogroups have been defined and about one third of them are found only in India. Something like 12 basal M haplogroups are found primarily in, or confined to, Northeast India. If any N haplogroups were present there someone would have found them. So stop making things up.
"Never mind that the two lineages have 'long' stems and hence belong to a period quite late in relation to the main N expansion".
Surely you cannot still believe that a 'long stem' equates with 'recent arrival'. Both haplogroups arrived within the region where they are now found with the original N expansion.
"They suffered drift BEFORE arrival to their destinations".
No. They suffered drift before they were able to expand from their region of coalescence.
"M did not suffer almost any drift once it arrived to South Asia, but a massive explosion instead (all drift happened in Arabia"
Well, if you ignore haplogroups M10, M11, M41, M53 and, to a lesser extent, M6 and M31 you could believe that. How do you explain those haplogroups' long stems?
"Burma, Cambodia..."
Also unlikely. As far as I'm aware no basal N haplogroups are found in either place. Just some derived clades and representatives of R.
"she must have lived somewhere, even if she was a great sturdy walker, she surely only moved traveled a few hundred kilometers between first and last pregnancy".
You're being deliberately idiotic again, and again I hope it doesn't represent your natural state. 'Haplogroup N' was not a single woman, although obviously the final mutation occurred in a single woman in a single place. Her descendants, still 'N', spread from there. They remained 'N' for something like 2500 years by your own reckoning. So no need to postulate a single woman walking all the way from Africa to Australia (or from India to both Australia and Africa, as you would have us believe).
"South Asia as a whole! It has improved though but every other semester something new is found, so there is a lot of top-level genetic diversity there".
ReplyDeleteTo make up for my troubling you so consistently you may find this interesting:
http://www.biomedcentral.com/1471-2148/8/227
Deals specifically with R7 but:
"The inclusion of our 35 novel sequences (Table 1) into the phylogeny of haplogroup R allows the recognition of eight new subclades within six haplogroup R branches unique to the Indian subcontinent (Fig. 1, [see Additional file 1]). We refine here the internal topology of haplogroups R5, R6 and R8, and describe two novel sub-clades of hg R7, to be discussed below in detail. Subclade R5a is defined by a deletion at nucleotide positions (np) 522–523 and one control region mutation at np16266. R6a is defined by two control region substitutions (at sites 16129 and 16266). In haplogroup R7, two new subclades R7a and R7b can be identified (for details see further down). A new subclade of R8, called R8a, is defined by a single coding region substitution at np 5510. Haplogroup R30 splits into two subclades R30a and R30b, the former supported by ten coding region substitutions and the latter by 24 coding and control region mutations. Similarly, in haplogroup R31 a new subclade R31a can be distinguished by 17 control and coding region mutations. Coalescent estimates suggest an ancient branching pattern in hgs R30 and R31, dating back almost to the earliest diversification of the superhaplogroup R itself. This most probably occurred soon after the out of Africa dispersals into the Indian subcontinent"
"I'm trying to help you by pointing out where your reasoning is incorrect"...
ReplyDeleteI wish. You point out how its should be "corrected" in order to fit with your fantasy.
"South Asia, including Northeast India, has been well sampled."
You are referring to a very recent study (a year old?). Most such studies acknowledge once and again that they are surprised of the diversity found. It can't be compared with the simplicity and relatively throughout certainty we find in Europe in any case, where sampling is indeed very extensive and detailed.
Notably we do not know enough about the basal diversity of N2 and N1'5 in South Asia, which seems to be quite more than just a rare erratic in any case, according to Palanichamy.
But let's assume for a moment that you'd be right re. South Asia... that would only (very potentially) transfer the centrality towards SE Asia, "next best" in basal diversity, yet you insist in making it West Asian and Altaian! That's what I cannot accept.
"Surely you cannot still believe that a 'long stem' equates with 'recent arrival'".
At this moment sentences like the above one can only irritate me the outmost, notably your insistence on manipulating the objective notion of stem length.
I DO NOT "BELIEVE" IN ANY KIND OF NECESSARY CORRELATION BETWEEN STEM LENGTH AND "ARRIVAL" TIME. It's impossible to know anything for sure about what happened in the long stem state, we can maybe infer from each particular context but as itself it only indicates a relatively long period of low numbers - wherever. This I describe as the UFO model: between the L3 node and the N node and between the N node and the, say, A node, the carriers were abducted by ETs (for all practical purposes they were "nowhere").
In real terms? Did they stay put, did they move, did they form a separate but tiny population or were part of a larger one... all that belongs to the realm of conjecture and speculation. Feel free to "believe" what you wish, I choose to remain skeptic unless contextual evidence is rather clear and that only applies in a case by case basis.
In general though, at last for the relatively simple case of N, where all subclades belong to three basic evolutionary groups, those clades in the intermediate class of 4 mutations (X, N2 and A) appear to have expanded upon arrival to some sort of land of opportunities, so they surely did not coalesce in situ (what would imply a super-abnormal kind of evolution) but at least some hundred kilometers away.
...
...
ReplyDelete"How do you explain those haplogroups' long stems?"
I haven't considered but I imagine that at least some were constrained by larger local clades that impeded their growth. In some other cases they may have made remote valleys as their homes as well.
Each case should be considered on its own merit. Sadly it's difficult for me to evaluate the genetic geography of South Asia in the detailed manner it deserves, no doubt.
I trust (?) that you will surprise us with such a detailed map in your blog.
"As far as I'm aware no basal N haplogroups are found in either place".
Certainly there's N* in Indonesia, as well as M*. Whatever it is, I can't say. Burma has not been sampled (the rare sample is probably from refugees, the state keeps totalitarian control on everything that happens inside its borders). Do you know of any good study on Indochina at all? All the info I have is scattered and extrapolations (not necessarily correct but not necessarily wrong either) from South China and Malaysia/Indonesia.
What is indubitable is that it is Indochina what falls in the middle of the N explosion, and also of that of Y-DNA C (approx.) being also attributed (though maybe it's more like Yunnan) with being the homeland of Y-DNA D, while MNOPS may have coalesced in Indonesia. Obviously the region can't be a mere passage from the "demographic pump" of Altai (sarcasm intended) but rather the other way around: Altai would be if anything a mere passage for the demographic pump of SE Asia. Even if, IMO, that happened only after SE Asians migrated back westwards through South Asia, this is less important: what matters is where the origin lies and what must be therefore interpreted as destinations.
This is where we deeply disagree and no rants about long, short or mid-sized stems will change a comma about this.
"To make up for my troubling you so consistently you may find this interesting:
ReplyDeletehttp://www.biomedcentral.com/1471-2148/8/227
Deals specifically with R7"...
Thanks.
"I trust (?) that you will surprise us with such a detailed map in your blog".
ReplyDeleteI actually sent you a copy of the M haplogroups' distribution. Not much has changed except for the addition of some newly discovered haplogroups.
"Do you know of any good study on Indochina at all? All the info I have is scattered and extrapolations (not necessarily correct but not necessarily wrong either) from South China and Malaysia/Indonesia"
I'm in much the same position, with the possible advantage that I have been studying the topic more closely and for much longer than you have. No single study, or at least one that would include modern genetic studies. I have worked out distributions from a number of studies. Some haplogroups I have not been able to find anything about. But feel free to ask about anything. I'll tell you what I know.
Sorry. I missed the earlier comment.
ReplyDelete"Notably we do not know enough about the basal diversity of N2 and N1'5 in South Asia"
We know the two haplogroups are fairly much confined to the northwest, no matter how diversified. I've put up a comment regarding N2 at another of your posts. It certainly looks far from 'South Asia'. Perhaps 'North Pakistan' at most, but probably west of even that region.
"It's impossible to know anything for sure about what happened in the long stem state, we can maybe infer from each particular context"
Why do we need to have different interpretations for 'each particular context'? Surely one thing we do know is that the haplogroup has undergone a period of drift. Drift usually occurrs in a population of limited size, which usually correlates with a population of limited geographic extent. We can easily see periods of migration and expansion in the haplogroup phylogenies. Long stems almost certainly represent periods spent at the opposite extreme: complete lack of migration and expansion.
"This I describe as the UFO model: between the L3 node and the N node and between the N node and the, say, A node, the carriers were abducted by ETs (for all practical purposes they were 'nowhere')".
Yes. That describes your belief concerning those haplogroups completely. L3 or N were 'abducted by ETs' and carried to SE Asia. As a result they were unable to leave any descendants behind. Then A took a UFO from SE Asia leaving no evidence of its passing.
"those clades in the intermediate class of 4 mutations (X, N2 and A) appear to have expanded upon arrival to some sort of land of opportunities, so they surely did not coalesce in situ (what would imply a super-abnormal kind of evolution) but at least some hundred kilometers away".
Im completely disagree. The 'intermediate class of 4 mutations (X, N2 and A)' all arrived as part of N's expansion. In fact they did 'coalesce in situ'. Presumably a deterioration in climate held their subsequent exansion, but 4 mutations later an improvement in climate or technology allowed a second expansion.
Those with marked interest for the techno-cultural aspects of this culture and facies may want to check the update, which is a link to an (unpublished??) related paper on nearby Hadramaut findings. It's very technical and inconclusive but you may learn a bit on the different types of Levallois debitage and what not.
ReplyDelete