November 18, 2011

Age of mtDNA haplogroup L3: about 70 thousand years

There are two aspects to this paper: first, it appears to be a solid attempt at inferring the age of mtDNA haplogroup L3. This haplogroup contains several subclades, including M and N, the two macrohaplogroups of the vast majority of Eurasians.

I am usually skeptical of very tight age estimates, but there appear to be no obvious flaws in the paper, and alternative mutation rates are used to derive the 70ka bound. Moreover, the 70ka age is consistent with what appears to be no longer in doubt, namely the arrival of fully anatomically and behaviorally modern humans all over the Old World, starting from the 50-40ka period.

The second aspect of this paper is its claim that pre-70ka dispersals are irrelevant to modern human origins. Indeed, if the early anatomically modern humans from the Levant (Qafzeh/Skhul) or the pre-Toba layers in Asia were ascribed to Out-of-Africa humans, then we would expect their genetic differentiation with East African mtDNA to trace back to Marine Isotope Stage 5 (~130-75ka), and indeed to its early stages, to account for the Mount Carmel hominins.

So, have we solved the Out-of-Africa riddle? Did the Out-of-Africa expansion take place after 70ka? I don't think so, not because there is anything wrong with the mtDNA age, but because the competing hypothesis, that is rarely, if ever discussed, is that there was an Into-Africa event post-70ka.

mtDNA furnishes the best evidence that humans trace their ultimate origins to Africa, since L3, of which M and N are subclades, is a young twig of the mtDNA phylogeny. As the authors of the current paper note:
Although the tree is highly starlike at shallower time depths, suggesting numerous episodes of rapid growth in the human population in the more recent past, it is only at a third of the time depth of the entire tree with the emergence of the L3 haplogroup that the first multifurcating them all the ancient diversity observed outside Africa) (Behar et al. 2008; Torroni et al. 2006; Watson et al. 1997).
Whatever humans were doing between ~200ka (when the first anatomically modern specimen is found in Ethiopia, and when the mtDNA phylogeny coalesces) and ~70ka (when the L3 node does), they were certainly not yet in the overdrive mode we find them c. 50ka when they begin making their grand entrance all over the surface of the planet.

So, while the ultimate roots of modern mankind are in Africa, there is no clear picture -yet- whether the post-70ka major expansion of humans originated in Africa. Certainly, it cannot have originated too far from it, because non M and N mtDNA is virtually absent throughout most of the world. But, it is not possible, yet, to exclude a Near Eastern post-70ka expansion that would make the ~100ka Levantine hominins ancestral to most modern humans, rather than irrelevant sidebranches.

There are several reasons why this may be the case:
  1. East African L3 subclades are found in Arabia, where one finds a rich assortment of basal N subclades, as well as a not insignificant amount of M. These are often dismissed as the result of recent introgression, but they could in fact, and in part, be remnants of an older population, perhaps associated with the Persian Gulf Oasis hypothesis, and certainly absorbed by J1-bearing Arabian ancestors from further north.
  2. The Y-chromosome phylogeny has no clear signal of Out-of-Africa ~70ka. On the contrary, Eurasia possesses DE*, D and E haplogroups, as well as CF, the major human lineage, with C being totally Asian. While Africa possesses the oldest Y-chromosome lineages (basal to CT), the evidence tilts towards Asia being the homeland of CT, which has the closest parallels to a post-70ka event.
  3. Finally, Africa, including East Africa, shows, at present no sign for the presence of fully modern humans at the crucial time period. We do have, of course, Omo ~195ka, crucial anatomically modern humans in Ethiopia, but no clear sign of a bubbling volcano of a population c.70ka ready to errupt onto the Eurasian landmass.
At present, I consider the possibility that the recent post-70ka expansion of modern humans was initiated in the Near East as a possibility that cannot be dismissed. The evidence seems ambiguous, at present, since Eurasia may have a better case for such an expansion in Y-chromosomes, while Africa may have a better case in mtDNA (since it has more basal L3 clades than Eurasia).

A better characterization of Near Eastern mtDNA, especially from Arabia, as well as increased archaeological/palaeoanthropological investigations in East Africa/the Near East/South Asia is needed to finally uncover the material counterpart of the major human expansion that is written in our genes.

A third aspect of the paper is that the human expansion was linked to climate and not on the emergence of symbolic behavior. I have my own reservations on the whole concept of "symbolic behavior". We do see early evidence of such behavior in Africa, such as Blombos Cave in South Africa and North Africa. The authors of the current paper write:
There is an intriguing possible rider to this conclusion. North Africa has been entirely depopulated and repopulated, at least with respect to mtDNA variation (Pereira et al. 2010), since the time of the Aterian industry, where modern symbolic behavior is attested very early, similar to Southern Africa, and in contrast to Eastern Africa (Barton et al. 2009). We might therefore contemplate a possible North Africa ancestry for L3, with its rapid radiation corresponding to an early range expansion into Eastern Africa. However, any potential dispersal between the Mediterranean and the Horn of Africa around the time of the MIS4/3 transition would face severe environmental difficulties, unlike the “green Sahara” conditions of MIS5 and the early Holocene (Drake et al. 2010). We therefore conclude that an indigenous origin for L3 in Eastern Africa remains by far the most likely scenario.

As Mellars (2006) has argued, the early evidence for symbolically mediated behavior in both North and Southern Africa rules out any simple direct link for the expansion of L3 to (Ambrose 1998; Watson et al. 1997). Evidence of engraved ochre now extends back to at least 100 ka (Henshilwood et al. 2009), Nassarius marine shell beads were evidently present across the range of early modern humans from Southern Africa to North Africa and the Levant before 80 ka – possibly tens of thousands of years earlier (Barton et al. 2009; Bouzouggar et al. 2007; d'Errico et al. 2009; Mellars 2006; Vanhaeren et al. 2006) – and evidence for burial ritual is found in early modern humans in the Levant dating to 90–110 ka (Mellars 2006; Shea 2008). Thus, as suggested by Basell (2008) the demographic expansionsthat led to the first successful dispersal out of Africa seem better explained by the play of palaeoenvironmental forces than by recourse to the advantages of “modernity”.
The absence of markers of behavioral modernity in East Africa at the crucial time seems puzzling. Climate may have caused Out-of-East-Africa, but why would Out-of-East-Africans without clear signs of behavioral modernity be able to outcompete the "behaviorally modern" people of North/South Africa and the Levant? This observation, coupled with the absence of any clear identifiable palaeoanthropological population in East Africa at the time in question raises my unease about this scenario.

Moreover, while we can definitely ascribe symbolic thinking to the cases mentioned in the quoted text, but these may represent precursors, and not the full "package" of behaviors that allowed (or even prompted) our ancestors to spread around the planet around the middle of the last 100,000 years.

Mol Biol Evol (2011) doi: 10.1093/molbev/msr245

The expansion of mtDNA haplogroup L3 within and out of Africa

Pedro Soares et al.

Although fossil remains show that anatomically modern humans dispersed out of Africa into the Near East ∼100–130 ka, genetic evidence from extant populations has suggested that non-Africans descend primarily from a single successful later migration. Within the human mtDNA tree, haplogroup L3 encompasses not only many sub-Saharan Africans but also all ancient non-African lineages, and its age therefore provides an upper bound for the dispersal out of Africa. An analysis of 369 complete African L3 sequences places this maximum at ∼70 ka, virtually ruling out a successful exit before 74 ka, the date of the Toba volcanic super-eruption in Sumatra. The similarity of the age of L3 to its two non-African daughter haplogroups, M and N, suggests that the same process was likely responsible for both the L3 expansion in Eastern Africa and the dispersal of a small group of modern humans out of Africa to settle the rest of the world. The timing of the expansion of L3 suggests a link to improved climatic conditions after ∼70 ka in Eastern and Central Africa, rather than to symbolically mediated behavior, which evidently arose considerably earlier. The L3 mtDNA pool within Africa suggests a migration from Eastern Africa to Central Africa ∼60–35 ka, and major migrations in the immediate postglacial, again linked to climate. The largest population size increase seen in the L3 data is 3–4 ka in Central Africa, corresponding to Bantu expansions, leading diverse L3 lineages to spread into Eastern and Southern Africa in the last 3–2 ka.



Andrew Oh-Willeke said...

Even if the mtDNA data are precise enough to be accurate within +/-4%, or so, as indicated, it is quite another thing to say that the mtDNA data are accurately calibrated at that time depth. (Carbon dating, e.g., can be accurate to +/- a century or two but requires considerable calibration despite involving a system simpler than mtDNA mutation rates.)

It matters because it is much easier to find a scenario compatible with the data in which, for example, some subset of modern humans (carrying M and N or at least L3 with a small effective population) migrates pre-Toba to Southern India (where we find their material culture on either side of the Toba ash fall), ca. 77 kya, and the folks left behind in the Levant then die off ca. 75 kya +/-3 ky.

This casts doubt on the classification of one outlier set of remains in China, but leaves the rest of data everywhere else in the world in tact.

Moreover, I simply don't see any case where the link between a specific date and mtDNA patterns is so definitive that calibration tighter than +/- 10% is actually possible.

This looks to me like yet another case of scientists coming close to the right mean result, but materially underestimating the size of their error bars due to systemic as opposed to statistical considerations.

Azerty said...

modern arabian peninsulars are majoritarly on J1c3d Y-DNA "frequencies 25% to 67%" all most on J1c3d2-I222.2 at 70% , estimated around 4000BC to 7000BC. and probably so Younger than 4000BC ... cuz all were positive on I222.2+, They're new comers from somewhere in North Eastern Africa since we have noticed a strongest presence of J1c3d among Nubians and Sudanese peoples frequencies 45% to 74% with high Genetic diversity on J1c3d in general" attested by FTDNA Arab Project", Tofanelli et al. (2009) take a strongest position for J1c3d people in africa as native people of neolithic origins,he reject totally any correlation between the Islamic expansion and J1 motifs.

so according to these Genetic studies and observations I think that we have here an interesting story about the presence of J1c3d carriers in arabia which can help us to understanding the Genetic situation of pre-historic east africa and arabia , In the Res Gestae Divi Augusti, Augustus claims that:

"By my command and under my auspices two armies were led at about the same time into Ethiopia and into Arabia, which is called the Blessed . Great forces of each enemy people were slain in battle and several towns captured. In Ethiopia the advance reached the town of Nabata, which is close to Meroe; in Arabia the army penetrated as far as the territory of the Sabaeans and the town of Mariba."

the Roman emperor Augustus have noticed the Nubian expansions into arabia from Meroe and Napta to Mariba in Yemen ! and this is really a real Joke ...

Even more if you look at the official results of arab dna project , we find a lot of samples on C5 and C* Y-DNA !
actually more than 10 results mostly from UAE, KSA and Oman
Haplogroup C is believed originally from southern coastal arabia along Southern Asia around 60.000BC, into Southeast Asia and Australia, and up the Asian coast. It is believed to have migrated to the Americas some 6,000-8,000 years before present, and was carried by Na-Dené-speaking peoples into the northwest Pacific coast of North America

look here :

there is also another interesting fact about the pre-historic genetic mtDNA structure of arabian peninsulars. this one :

Internal diversification of mitochondrial haplogroup R0a reveals post-Last Glacial Maximum demographic expansions in South Arabia.

Carleton Coon and Ernest Hotoon have always declared that pre-histroric peoples of arabia are somewhat like modern australiods , they have proved that by the presence of some Extinct Yemenites who look like Australiods with the presence of dravidians and Veddids in arabia.

clearly there is a recent repopulation of arabia from east africa which must be looked for serious, and thinking that J1 or J2 are arab or semitic genes thats completly a waste of time .

terryt said...

"North Africa has been entirely depopulated and repopulated, at least with respect to mtDNA variation (Pereira et al. 2010), since the time of the Aterian industry, where modern symbolic behavior is attested very early, similar to Southern Africa, and in contrast to Eastern Africa"

Perhaps that explains, 'The absence of markers of behavioral modernity in East Africa at the crucial time'. They didn't go via the Bab al Mandab. The expansion left the western margin of the Ethiopian Highlands, including the Sudan, and moved north on down the Nile and out into The Levant . The barren Red Sea coast had nothing to do with the expansion.

Dienekes said...


I don't see how J1 can be migrants from Northeast Africa. The opposite is true, J1 are migrants from the Near East to North Africa. And, within the Near East itself, they are migrants from the north (where their J2 cousins are) into Arabia, with a later expansion of J-P58 bearing Proto-Semites from the southern Levant/north Arabia general area.

eurologist said...

I agree with Andrew, here - uncertainties in time estimates can very easily be 10%-20% - if not close to a factor of two, that long ago. There are so many different "lifestyle" operations over that time frame, it is silly to try to model that with some ad hoc populations sizes. Too many isolated groups, too many bottlenecks, and we know that only a few out of many mtDNA groups tend to survive locally (by chance or selection), anyway (that is, the total human population size in an area can easily be an order of magnitude or two larger than that of a specific non-autosomal parental carrier that later dominates).

Even the latest autosomal results appear to point to a pre-Toba SE Asian population; and then there is y-DNA D, which looks like a truly first and old entry, there.

We definitely need more skeletal finds from N and NE Africa, Arabia, and also all of West Asia. But from my gut feeling (and common sense), the green Sahara period was the one with the largest and geographically most wide-spread population explosion in and for a long time - and also one of the only periods during which water, plants, and animals where available during travel East. If it quacks like a duck...

The main question is, how far east did these people initially get (beyond the Levant) before being caught in deserts in a swiftly declining climate, and where did they survive (if not in the Levant). Perhaps a small group made it through all the way to E Asia, but wasn't that successful, perhaps a second group (yet still pre-Toba) did better.

ktwop said...

As an engineer come lately to this fascinating subject, my apologies if my comments / questions are naive or stupid.
I understand that it is pretty clear from current evidence that most (or is it all?) non-Africans most likely (or is it certainly?) derive (exclusively?) from mtDna haplogroup L3. I also understand that the location for the emergence of L3 is either N. Africa, East Africa or possibly the Arabian peninsula where the latter two would imply a southern route for Out of Africa.
If so and if the earliest age for the emergence of L3 is 70kya, then does that necessarily mean that all earlier dispersals – if any – died out completely? Could it not be that the L3 dispersal (post Toba) met up and mixed genetically with some few survivors of earlier pre-Toba dispersals of AMH, or does the genetic evidence preclude that as a possibility?

Dienekes said...

If so and if the earliest age for the emergence of L3 is 70kya, then does that necessarily mean that all earlier dispersals – if any – died out completely?

If L3 originated in Africa, then yes, it would imply that pre-70ka AMH outside Africa die out, at least in terms of their mtDNA. I don't believe that is the case, and the more evidence turns up for AMH outside Africa prior to 70ka, the more difficult that is to believe.

If L3 originated in Arabia or the "Gulf Oasis" as I suggest in my post, then there is no mystery about what happened to the pre-70ka people who lived outside Africa.

DocG said...

I find it interesting that social scientists place so much emphasis on "hard data" from either genetics or archaeology, while physicists place far more emphasis on wide-ranging patterns, from which very ambitious theories can be developed. The theories are often developed first and then tested against the evidence. And if certain evidence appears to contradict the theory, as was the case, for example with Einstein's formulation of special relativity, as often as not it is the "evidence" that is eventually proven wrong, not the theory.

The key to developing a successful theory is not necessarily its conformity to all the evidence (especially when the evidence is ambiguous), but its explanatory power. Since Einstein's theories explain so much, physicists are reluctant to abandon them even when experimental results might appear from time to time that seem contradictory. (I'm thinking now of the apparently "faster than light" neutrinos that have apparently been observed.)

As I see it, the Toba model, especially as developed by Oppenheimer, explains so much about the large-scale distribution of human genetics, morphology and culture that it, rather than all the various bits of "evidence" that are continually being offered and then refuted and then recast ad infinitum, should be used as the basis for estimates regarding the date of the Out of Africa exodus. Since Toba blew roughly 72,000 years ago, then as I see it, this very solid date, rather than the very shaky estimates derived from purely genetic models, is the most likely candidate for a reliable "standard candle."

For more on the potential significance of Toba, and the possibility of alternative scenarios, see Chapters Nine, Ten and Eleven of my blog-book, Sounding the Depths:

Azerty said...

according to Tofanelli 2009

J1 STR motifs previously used to trace Arab or Jewish ancestries were shown unsuitable as diagnostic markers for ethnicity.

and not Even related to the spread of Islam or modern arabians. They find that coalescence times for lineages point to the more distant past: the period between 5500 -7200 BP(3500-5200 BC) Neolithic Time

modern sudanese and Nubian population with their Genetic diversity have worlds highest frequencies of J1c3d -Y-DNA "74% to 45%" they have recieved this Near Eastern Gene around 9000BC at the beginning of pastoralism in Africa. Indeed Nubia represents the Homeland of Pastoralism and Bedouinism , remember that Nubians were Nilo-saharan speaking people ...hence southern arabians like Yemenites who have 67% of J1c3d are southern arabian semitic speaking people who have absorbed a lot of Nubian words . Even arabic languages containts a lot of Nubian words. accroding to experts of Linguistics themselves.
the way why modern arabians admit it. here an arabic Video which explains how the nubian words was Introduced into the arabic language

notice that arabic language was appeared just around 5BC in Yemen !
The wet period as the Ice Age glaciers melted saw deserts bloom. Odd as it may seem today, the Eastern Sahara was the cradle of pastoralism in Africa. From a beginning there about 8,500 BC, the herding lifestyle gradually spread. Semi-nomadic tribes roamed the Sahara and presumably Arabia. Then a shift to a cooler, drier climate from around 5,500 BC encouraged pastoralists to seek better-watered havens in the highlands, such as the Atlas range, the Tassili massif and southern Arabia. The cave paintings of Tassili, Southern Algeria, capture the pattern of their lives. Tofanelli et al feel that Y-DNA J1 is the genetic legacy of the dispersal of pastoralists in the Sudanese Sahara, North Africa and Arabia.

Amharas and other east african groups which have more than 36% of J1 shows different J1 motifs than arabians mostly on J1b ; J1d ; J1a wich were distinct variety of J1c3d accroding to their different clusters :

YCAII=22-22 and DYS388≥15

hence modern arabians didn't have any Genetic diversity on J1c3d nor in J1* clusters.
In Ethiopia, all Oromo and Cushitics are approximately 29% of J1 chromosomes on J1* (meaning that they are J1 but not in J-P58 = J1c3d ). directly related directly to In the Caucasus and Eastern Turkey.

I didn't see any supposed semitic migration into East Africa but Probably an Exchange of culture and languages with strongest male East African expansion into arabia. moreover J1c3 or J1e-P58 is not semitic marker , this one is originally from eastern turkey before 10.000BC related to the Zarzian culture in Taurus mountains ...while proto-semitic language is believed to be associated with the caucasian peoples of North Africa "rosa et al 2008" before 15.000BC mainly with Natufian and Harifian cultures.
Nop ! wrong : J2 derivations in arabia are typically matched with the Persian and kurdish clusters on J2a4! hence Chiaroni et al 2010 admitted the fact that worlds diversity of J2a4 was Iran !

so this is why J1 and J2 are not even the same peoples nor were cousins by languages or cultures in their spread of origins. both they're founded among afro-asiatics , nilo-saharans , indu-europeans , and caucasians.

Anonymous said...

Don't tire yourself Dienekes, Azerty is known throughout the net for his theories about J1.

He spends most of his time on Youtube (known as samysamy25 or arabiannight100 among many other fake accounts) creating videos about how black and similar to Indians Saudis are and how white North africans are.
He believes that Humanity arose in Northwest Africa and that E1b1b1 is the original marker of Cro-Magnon before then associating it with Afroasiatic speakers in general.

He also proposes this weird theory about J1 being Nubian/Cushitic (J1* could've been involved with the diversification of Cushitic in the Horn of Africa but obviously that's not what he means when he says "Cushitic"), I don't really know what went through him.
He seems to have a problem with J1c3d being associated with the spread of Semitic somehow.

He also misquoted several studies to suit his needs, often taking them out of context.

You better not pay much attention to him and avoid arguing with a fool, since other people who are watching from a distance can't tell who is who.

Azerty said...

whatsoever tofanelli works on, he mayhave demonstrated a New perspective of the Phylogenetic Legacy of J1 motifs, southern Egyptians , Sudanese Copts and Nubians Plus Khartoum Population are all most on J1c3 which was the most common Haplogroup among them "74% to 45%" sources:(Hasan et al 2008 - FTDNA J Project/European journal of Genetics 2009 and Chiaroni et al 2010, Tofaneli et al 2009)

these peoples aren't semitic peoples nor were majoritarly arabian speaking people , you should take that into a serious consideration both all we are aware about the phenomena of Pan-Arabism and Arabization in East/North Africa.

Yeah there is Room for J1c3d formelry L147 in Africa specially for his dominant sub-clade J1c3d2-I222.2, allmost diverse In africa and very Old than his distribution in arabia Peninsula. more Interesting he was the Genetic Marker of Pastoralism which africa was its homeland mainly "Nubia" confirmed by many anthropolgical and archeological facts.

Indeed, J1c3d2 is allmost absent in eastern turkey and caucazus area but frequent among africans and arabian peninsulars.

the presence of semitic languages in east africa can not be explained always by the presence of arabians ! since it was its southern branchs which were very recent "probably due to a Levantine Expansion "such like Arameans or Nabateans" into southern arabia and then to Ethiopia"

I see that the semitic speakers were primalry E-M123/(M34) in Arabia, Cruciani (2004) suggest that E-M34 was likely introduced into Ethiopia from the Near East.

this haplogroup is common among sephardic/ashkenazi jews, founded among Yemenites , saudi arabians , Levantines , and other Khalejis.

East Africans are very different peoples , they're majoritarly North African males admixed with native Negroid females since the paleolithic to the Neolithic times,with very minor asiatic contacts. Hence I consider J1 as Neolithic North African marker just like V32 and others who were expanded from the Egyptian/Lybian area into ethiopia. representing the expansion of pastoralists into east africa.

anyway, arabs suffers so much about the lack of Genetic diversity on J1* or in J1c3d. East Africans are mixed and occupy an intermediate position between North Africans "not arabians" and Sub-Saharan Africans.

East africa is very complexed area attested by the ancient presence of caucasiods since the paleolithic just like the Eburran Mechtoid culture in the ethiopian highlands around 15Kya

Dienekes said...

Hence I consider J1 as Neolithic North African marker just like V32 and others who were expanded from the Egyptian/Lybian area into ethiopia.

Markers don't pop out of thin air, there must be precursor populations. Now, only a small part of J1 is shared between Africa and the Near East, whereas there is J1 in the Near East and the Caucasus that is not found in Africa.

Moreover, there is J* and J2 in the Near East, which is lacking in Africa. And, indeed IJ has a Eurasian distribution, as has IJK. So, it takes a very strange kind of thinking to propose that J1 emerged in Africa, when all the series of its relatives are clearly of Eurasian origin.

Azerty said...

I am not talking about the Parent Clade J1* and his sub-divisions in association with, pls don't confuse the two because they have different stories there, J1* is not like J1c3d-L147 nor J1c3d2-I222.2.
I am talking in this specific marker J1c3d which was founded among non semitic speakers and shared with arabized Population along together with arabian peninsulars. all we are aware that IJK series are Eurasian by origins no Problem with, but J1c3d seems to be radiated in africa in the neolithic than in the Near east its because of total absence of his sub-divisions in the Levant or in Eastern Turkey and caucazus " so now I think you get my point right:) , the presence of J1 in africa formelry J1c3d goes back around 9000BC mainly higher in the Nubian area ... the fact why many Genetists claim Nubians more Eurasians by origins than other north/East Africans, since J1c3 is the pre-historic marker of Nubian Population, and founded among other nilo-saharans , copts and Kushitic Groups why we deny that facts ? and saying that they have semitic near eastern origins ?! its Irrelevant ! which semitic " arabs or jews or what exactly " both these groups are too much recent in our human history.
you forget totally the presence of J1*J1c3 "old parent clade" among Oromo ethiopians at 3% and among Maghrebians 3% ! data of cruciani 2004

actually most of arabian peninsulars of J1 are majoritarly on J1c3d2-I222.2 at 70% , the rest were under down-streams of J1c3c only founded among Amharas, with few cases of J1b , J1d , J1b all were recently Introduced into arabia from Ethiopia "cruciani 2004" or from Iran , Kurdistan and The Levant." frequent in these regions."

the Great problem that most of arabian peninsuars J1c3d2 males have recent Ages according to their TMRCA. estimated around 1800BC To 880AD, attested by sub-divisions of I222.2 known as (L65.2/S159.2), while in arab dna project a sample from a sudanese carrier of J1c3d2 has discovered a TMRCA dated around 4500BC to 7000BC.
or you should Look at the new tree of J1c3d on DNA FORUM.
another Important thing , J2 among east africans is all most J2b , mainly among ethiopians and other east africans , known to be observed in southern europe , India and the Levant. the way why J2b in east africa it mayhave linked to the Hellenic arameans of syria and Roman Nabateans who were expanded into southern arabia "introducing southern semitic languages" around 500BC to 5AD along together with E-M34 and then to Ethiopia.

there is nothing strange from what I say but refuting all these things "you can verify" is like rejecting what Tofanelli and others are so right on J1 being the Genetic marker of pastoralism in Africa and then to arabia since it was pastoralism = bedouinism.

Dienekes said...

J1c3d goes back around 9000BC

It does not, since J-P58 itself goes back to the 3-4 thousand years BC.

Anyway, you are free to hold on to your belief about J1c3d came from Africa, but this is off-topic, and you've made your case clearly enough.

DocG said...

In view of all the above pedantic squabbling: I rest my case. :-)

jes-r said...

Well, anything is possible. I find it remarkable that L6 and L4 (which are older than L3 – but phylogenetically relatively close compared to the other L's) are quite diverse in Yemen/Arabia (albeit at low frequencies). Most of the L6 and L4 clades in Yemen do not show recent matches with East Africans, which suggests an ancient presence of these lineages.

Pascvaks said...

At this point in the discussion it seems worth renoting a statement made at the top of the page: "there appear to be no obvious flaws in the paper, and alternative mutation rates are used to derive the 70ka bound".

terryt said...

"Most of the L6 and L4 clades in Yemen do not show recent matches with East Africans, which suggests an ancient presence of these lineages".

Is that really the case? As far as I'm aware L6 has just two clades, L6a and L6b. Both are found in Yemen but L6a is also present in Egypt and L6b is also present in Ethiopia. L6's diversity in Yemen is therefore most likely to be a product of the two separate sources.

Much the same applies for L4. Both L4a clades are present in Ethiopia with just one clade within it, L4a1, present in Yemen. The other, L4a2, is present in Arabia as well as Ethiopia. L4b1 is actually found in Yemen, but L4b2 is found through Ethiopia and Tanzania. It is therefore quite possible that L4 and L6 are at least post-OoA in Yemen.

Andrew Oh-Willeke said...

"I find it interesting that social scientists place so much emphasis on "hard data" from either genetics or archaeology, while physicists place far more emphasis on wide-ranging patterns, from which very ambitious theories can be developed. The theories are often developed first and then tested against the evidence. . . . Since Einstein's theories explain so much, physicists are reluctant to abandon them even when experimental results might appear from time to time that seem contradictory. (I'm thinking now of the apparently "faster than light" neutrinos that have apparently been observed.)"

You're misstating the case in physics. The precision of the tests of the Standard Model and General Relativity are stunningly awesome. We are talking parts per billion accuracy - the equivalent, for example, of dating the Toba explosion to a precision of a fraction of a second. Even the OPERA faster than light neutrino result differs from the predicted result by only part per 100,000 with an error in the variance from the predicted result of +/- 16%. The Standard Model has withstood adjustment other than the addition of neutrino mass for the last 40 years with thousands of physicists working full time in that time frame at big budget experiments looking for deviations.

The total data set for all genetic and archaeological data in the time period from the earliest pre-sapien, post-chimp hominins and every DNA test ever done amounts to perhaps a month of data from the large hadron collider.

Also, there is no good reason for prehistory to fit any theory with the kind of precision seen in physics. In human affairs, weird, unpredictable things can and do happen and we almost certainly underestimate how often they do.

Physicists are able to summarize so much existing data with theories that have such good empirical fits that any outlier result is a big deal because even devising any possible alternative theory that could fit both the old data and the outlier is often very difficult to do in any way that make any sense.

jes-r said...

@ terryt,

The L6 and L4 haplotypes in Yemen do not show recent relationships with those in the Horn region. They branched off from each other tens of thousands of years ago, and can't be explained by recent gene-flow scenarios.

Also, L4b1*, one of the oldest L4 branches (~90 KYA) hasn't been found in East Africa yet.

German Dziebel said...


"You're misstating the case in physics."

And you're misapplying your corrected physics case. If the database for genetics, archaeology, etc. is infinitely smaller than physics databases, then it stands to reason that any inference about time and place of migrations, sources of admixture, order of mutations, etc. is doomed to be futile after just 200 years of data accumulation. We just witnessed the discovery of a new hominid species, which shares genetic material with us, humans, but that we know about only by way of a pinkie and a tooth (Denisova). What are the chances of recovering another case like that? And why hadn't we found it earlier? Maybe because we didn't have enough time. We may need another 400 years at the current pace of data accumulation to actually begin talking about it.

In a word, your case in physics strengthens the point made by whoever you directed your critique against, not weakened it. Theory always beats data in human origins research.

terryt said...

"They branched off from each other tens of thousands of years ago, and can't be explained by recent gene-flow scenarios".

Where did you get that information from? Phylotree has just the two L6 haplogroups, neither specifically 'Yemeni', and just the L4 clades I mentioned. Hardly evidence of a particularly long presence in Yemen.

"Also, L4b1*, one of the oldest L4 branches (~90 KYA) hasn't been found in East Africa yet".

You may be on slightly firmer ground for an ancient Yemeni presence here, but other explanations are possible. L4b2 is certainly present in East Africa, and even in the Hadza and Khoisan, so L4b is presumably East African. Lb4a could easily be a post-OoA immigrant to Yemen whose parent haplogroup has died out in Africa.

jes-r said...

''Where did you get that information from?''

Behar et al. 2008 data:

''Phylotree has just the two L6 haplogroups, neither specifically 'Yemeni'.''

However, both are find in Yemen, see the spread I posted above. It is currently more diverse in Yemen than either Egypt or Ethiopia. In Ethiopia L6 only appeared in Semitic speakers so far (Kivisild et al.) so back-migration may play a factor.

terryt said...

Thanks for the link. I see from it that L6a has two subclades. One a single Egytian and the other three Yemenis. Impossible to tell from that which way the movement was. L6b has 3 subclades. One a sigle Ethiopian Jew, one as single Yemeni Jew and the third an Ethiopian of unknown affinity. So two Ethiopians and one Yemeni. Again it is impossible from that information to discern the direction of movement.

"It is currently more diverse in Yemen than either Egypt or Ethiopia".

Four Yemenis, two Ethiopians and one Egyptian. Sure, more diverse in Yemen. But that diversity could be a product of an original diversification between Egyptian and Ethiopian L6 with subsequent movement to Yemen.

"Ethiopia L6 only appeared in Semitic speakers so far (Kivisild et al.)"

As far as I'm aware the argument as to where Semitic originated has not been resolved yet. If it was in Ethiopia that would explain your observation.