April 23, 2011

The Africanness of Europeans (and West Asians)

Sincerely not my ideal of research on population genetics but this new paper still offers some stuff to chew on:


An important bias is the lack of other West Asian or North African references. The paper deals with Europeans, Jews and Palestinians (with emphasis in the comparison with the Yoruba outgroup YRI) and that's about it (we also find Turks in the supplements). This has interest, of course, but also has huge limitations. 

Specially problematic, considering that the paper tries to study an apparent African component in West Eurasian populations, is the lack of North African references. 

The authors fail to see it but for me it is transparent that what they identify as "African admixture" is nothing but North African admixture, surely related to the expansion of Y-DNA E1b (though not strictly so). The origin of this lineage is surely in NE Africa, around modern Sudan, and arrived to West Asia and Europe surely by two main routes:
  1. Via West Asia and Greece and then with the Neolithic flows.
  2. Via West Iberia and then through the Atlantic (mostly) with Dolmenic Megalithism.

In order to illustrate my point, I shall borrow here an image from a previously discussed paper:

The YRI component vanishes when Iberians and NW Africans are directly compared

The authors also pretend to be able to pinpoint the approximate age of such alleged (and ill understood to them) "admixture" but their dates make no sense whatsoever. At least not for me. So I'll ignore that part.

We lack of a good source population: HapMap Yorubas YRI can only be a proxy outgroup and the various Palestinian groups can only be an already much admixed proxy. But estimating from several sources (including this PCA, from this earlier entry) I believe that the "YRI admixture" approximates 1/3 of the true admixture with the original Low Sudanese or Upper Egyptian main true source (?) of West Eurasian Africanness.

So all the results should be multiplied by three to estimate the apportion of "Nubian" input. A "Nubian" true original population is anyhow an approximation, because in the Iberian case most of the Africanness is in fact of Moroccan or otherwise NW African origin, with no immediate ties with Sudan (E1b1b1b1-M81). 

If so, I estimate, based on table 2 (and multiplying by three), that there is roughly the following fractions of Nubianness:
  • Palestinians c. 30%
  • Portuguese c. 6% (notice that the actual amount of North African component in West Iberia is more like 9%, so we are underestimating here the real "Moroccanness" by 50% - this is probably because Moroccans are quite less Yoruba-like than Nubians)
  • South Italians c. 5%
  • Spain c. 3% (4.5% is probably more realistic for "Moroccanness")
  • Sardinians and North Italians c. 0.6%
  • Ashkenazi Jews c. 8%
  • Other Jews c. 12%

Whiter than thou...

And that would be what I'd have to say on this only somewhat interesting paper but I happened to dig into some of the supplementary material and I'd say that fig. S1 has some interest. 

Of course it has some interest for those who enjoy racialist (tending to racist) comparisons. The lack of other outgroups like East or South Asians, allows for a very black and white (literally) reading of this material. That's why the title of this section, because in most racialist forums there is not going to be a blacker than you argument, certainly not involving European populations.

Interestingly there are very few populations whiter than HapMap's CEU sample. As you may know this strange sample that pretends to represent Europeanness in so many simplified genetic studies is made up of Caucasoid people from Utah (USA): it's not even a true European sample. However it seems representative enough for English-plus (most white Utahns claim English or, some, Danish ancestry, according to US census data).

Following fig. S1, most populations are actually blacker than CEU, so to say, i.e. they tend towards YRI at least slightly more than CEU. I could only spot two exceptions: Orcadians and  Basques. So we can well say that CEU are, at least in contrast to Africa, a paradigm of Europeanness (after Orcadians and Basques) - however no judgment can be made in relation to other outgroups (from Asia) based on this paper's data.


Some significant structure within Europe

An interesting element in this supplemental fig. S1 is the second (vertical) dimension. In most tripartite comparisons between CEU, YRI and another European population, CEU and the other Europeans overlap in the second dimension, which only reflects diversity within Yorubas themselves. I understand that if Yorubas are more internally diverse than these two European populations, the potential structure between NW Europeans (CEU) and the compared third population is not-significant, trivial, irrelevant. 

This is the case for: Swedes, Irish, Germans, German-Swiss, Belgians, Scots, Orcadians, Dutch, French-Swiss, French (two different samples), Polish, Austrians, Hungarians, Czechs, one of the two Russian samples (POPRES), Spaniards, Portuguese, Bosnians, Croatians, Romanians, Greeks, one of the two Italian samples (CEPH-HGDP), Tuscans, Italian-Swiss. This was also the case for Turks (not technically European) but not for any of the other West Asian (Palestinian) samples nor the Ashkenazi Jewish one (which is the most Europanized of all Jewish subgroups). 

Among Europeans, a number of populations show more structure in comparison to CEU (and hence the majority of Europeans) than Yorubas show among themselves, defining a second dimension in intra-European terms. These "outlier" European populations are:
  • Basques
  • Russians CEPH-HGDP (which I believe are from the Far North and have probably a strong Finnic component)
  • Italians POPRES (which are partly from the South - North Italians alone did not contrast much with CEU)
  • Sardinians
In addition all the Palestinian groups ("Palestinian", "Druze", "Bedouin") also show such contrast with CEU, as do Ashkenazi Jews but not Turks.

Follow the four relevant eigenvector graphs:

Comparison with Basques

Comparison with POPRES Italians

Comparison with Sardinians

Comparison with CEPH-HGDP Russians

55 comments:

  1. There is no evidence that E1b originated in Sudan. All evidence points to Ethiopia as the origin of E1b1*,E1b1b* and E1b1b1*. (Cruciani '04,'07 ,Trombetta '11)
    Only a limited number of further down stream clades of E1b1b1* are purported to have an origin further north of Ethiopia, namely the Sudanese/Egyptian and/or the Libyan/Egyptian border.

    ReplyDelete
  2. I do not think I claimed that E1b as such originated in Sudan.

    In any case, I am under the strong impression, from both genetics and, specially, archaeology that the influence of E1b1b1 (in its several variants), which is surely associated to the expansion of Afroasiatic languages and possibly also a form of Mesolithic economy that may have been at the ultimate origin of Palestinian (cerealistic) Neolithic, expanded from the area of Nubia in the Sudan/Egypt border. This does not preclude earlier migrations northwards from the Blue Nile area or whatever but I cannot easily document them.

    I'm open to discuss the fine detail of this matter in any case, because I am somewhat unsure.

    I doubt in any case that there is archaeological evidence of any sort suggesting an specifically Ethiopian origin of these Afroasiatic flows, including Capsian culture. The genetics of mtDNA U6 also suggest first a West to East flow (from Morocco towards Nubia), possibly with Oranian, and then (U6a) an East to West backflow, surely with Capsian.

    I would not know how to fit Ethiopia in all that. In any case, the political borders that circumscribe Ethiopia and Sudan have only a few decades or centuries and, even if both areas have overall differences, there is no absolute divide of any sort. In many cases pinpointing the origin of so many clades and maybe even the origin of Humankind itself in that zone, it is almost impossible if we should say Sudan, Ethiopia, Uganda... NE Africa, East Africa, the Horn, the Nile... sometimes it looks a bit a matter of semantic detail.

    But I'm open to be taught a lesson in geographic precision if good. Thanks in advance.

    ReplyDelete
  3. On second thought, I must agree that E1b1b1 looks, per basal diversity (subclades d, f and g specially, plus subclade e in East Africa) like Ethiopian originated. The genetic evidence looks like supporting that (at present level of knowledge).

    But, well, I'm being a bit "eurocentric" on the effects of this expansion, what is excused because of the theme of this entry. The Ethiopian-specific clades are not the ones that affected so markedly the Mediterranean region, so attributing the origin of this flow to "Nubia" is justified, specially as it's not just a matter of this haplogroup but of other evidence as the already discussed archaeological and mtDNA one.

    Still, it is probable that this "Nubian" expansion was only a second stage of one beginning further South at or near Ethiopia. I say this not just because the Y-DNA evidence but also because of the linguistic evidence in regard to Afroasiatic languages.

    ReplyDelete
  4. On third thought, I may be contradicting myself (sorry about that - sometimes I think better while, or rather after, talking... or writing).

    Not here but when discussing Trombetta's paper, I argued that E1b1b1b1-M81 could well be very ancient in North Africa (and hence in Iberia). So I am working in fact on the hypothesis that only E1b1b1a1-M78 is really the lineage associated to Afroasiatic languages' expansion in North Africa and West Asia (together with J1) - and maybe once also in parts of Europe (was the primary Neolithic language in Europe an Afroasiatic language? Maybe not but the high presence and secondary center of this haplogroup in Greece and Albania could suggest it).

    Here I can discern that Ethiopia (but specially Somalia) has highest frequency for this specific haplogroup but I wonder about where the greatest diversity (and hence the likely origin) is. For what I can read all around, Egypt is the locality suggested as origin for this haplogroup. As in the case of Ethiopia, which region within those political borders may be important, and I understand that Upper Egypt (and specifically Nubia within it maybe) is the most likely core.

    However correct me if wrong. Thanks.

    ReplyDelete
  5. "In any case, the political borders that circumscribe Ethiopia and Sudan have only a few decades or centuries and, even if both areas have overall differences, there is no absolute divide of any sort."



    The first attestation of Ethiopia is by the ancient Greeks. We don't know where they got that name from however. Anything south of Egypt was Ethiopia to them, later after the dominance of Nubian civilization in today's Northern Sudan region, the name was restricted to more Northerly Sudan areas often being equated with Kush in the Hebrew texts, even later in and around the 4th century A.D. a certain king of Axum (the ancestral kingdom of modern day Ethiopia) named Ezana, declared all his kingdom Ethiopia and Christianity its religion. Even later, after the rise of Islam, the gravitas of the Kingdom of Axum moved further south and endured an isolation for more than a millennium. So with all due respect, it is fairly simplistic to label Ethiopia as only a few decades or centuries old.



    "it is almost impossible if we should say Sudan, Ethiopia, Uganda... NE Africa, East Africa, the Horn, the Nile... sometimes it looks a bit a matter of semantic detail. "

    Looking at continental Africa as an independent unit and disregarding the narrow arteries that connects it to Eurasia (Sinai,Gibraltar and bab-el-mendal), anything north of the northern border of the D.R.C and East of the Western border of the Central African republic is technically North East African, thus Ethiopia with this defintion would be a North-East-Sub-Saharan African country whereas Northern Sudan and Egypt should be technically North-East-Saharan African countries/regions.


    As far as the Nile, it depends if one would be compelled, like I would be, to name its corresponding region by its source/direction of flow, if so, please do realize that 85% of it's flow comes from Ethiopia. If not, the total span of the river can be technically attributed to both North and South Eastern Africa. (both sub and Saharan in either case)



    "I would not know how to fit Ethiopia in all that."

    Ethiopia can be said to fit genetically between Sub-Saharan East African populations like the Hadza, Sandawe ,Maasai and Saharan Eastern Africans like Egyptians, Northern Sudanese and so forth, not surprisingly, consistent with its North-East-Sub-Saharan African spatial location. The best way to view this would be by looking at an intra-African MDS (or a similar Principal Component Decomposition of Autosomal SNP data), but not with the traditional 2-Dimensional view but rather, an isometric view that includes the plotting of the 3rd eigenvector, like so :

    http://2.bp.blogspot.com/-cM_JnQlOWzs/TbsACPd44tI/AAAAAAAAAHQ/0dpYlX-G_pY/s1600/MDS_ISO.png

    since the traditional 2-Dimensional view, like so:

    http://1.bp.blogspot.com/-uWv_KmvZeiA/TbsAESyBPyI/AAAAAAAAAHU/zfjO6NuEvw4/s1600/MDS_C1C2.png

    obscures the important distinction that is brought forth by including the 3rd dimension.

    P.S. I have not yet read your other posts, please do allow some time for me to read them and respond (if there is anything I can add that is)

    ReplyDelete
  6. Is it true that the study lacks North African references? How can they claim Europeans are more related to Yorubas than East Asians, if they're completely ignoring North Africans?

    It is quite obvious North Africans are related to Europeans and to East Africans at the same time.

    I don't see why should Yorubas be considerably akin to most Europeans. Historically there were very few contacts between these two regions.

    ReplyDelete
  7. Don't get me wrong, Etyopis, please. What I said is that the borders of the modern polities are that: modern. I am thinking particularly in the tribal province of SW Ethiopia, so rich in cultural and genetic diversity but hardly the core of Ethiopia (Axum, Abyssinia). How fair and how clarifying is it, if we find a trait or a diversity pool in that province, to attribute it to "Ethiopia" in general?

    Or whatever other province or ethnicity. I think it's most fair to say Ethiopia without further clarification if the trait is widespread through the country but not when it's found only in a particular region or ethnic group.

    That's my point: it's not exactly the same an Amhara-specific trait than a SW tribal one (or an Oromo one, a Somali one, etc.) - how reliable is to use a generic tag such as "Ethiopia" for all the populations within that state?

    Similarly in regard to Sudan, the second largest African state and almost as large as the whole European Union, being too generic may be misleading.

    On the other hand, sometimes, for lack of enough data or inference power, you have no option but being generic. My apologies if I happen to place this generality in the wrong side of the border. But I have the strong feeling that which side of the border is not as important as which tribe or ethnicity (and occasionally it is a trans-border thing anyhow).

    I'm aware on the general genetic position of the country and region, as well as of its historical outline (I'm a bit of an "africanist" not your usual ignorant white guy - mind you), though I do appreciate those graphs, specially the tridimensional one (it's hard to discern but curious also).

    In any case what I meant when I wrote that I do not know how Ethiopia fits in, I was all the time thinking on how fits in the Afroasiatic-E1b1b1a1 expansion.

    My impression is that it may well be a source but that's only in the genetic and linguistic plane. But in the archaeological plane I totally lack of the material connection that could help us say: look this Nubia-centered expansion is in fact a second phase of a process that began further south, in Ethiopia (or somewhere over there).

    ReplyDelete
  8. "How can they claim Europeans are more related to Yorubas than East Asians, if they're completely ignoring North Africans?"

    Because it's convenient within their conceptual time-frame, which is judeocentric, and also because, working for a Zionist institution, they seem to lack access to other samples of the region. I was going to mention that in the original entry but I decided it could be seen as unnecessarily political. But it's there anyhow (science does not happen in a bubble).

    ReplyDelete
  9. Back to Etyopis and a bit off topic (but not too much):

    Recently it has been claimed often that the Maasai have important West Eurasian admixture but I have many doubts. In this sense your 3D graph is very illustrative, because the Maasai and Ethiopians seem to define a fourth genetic pole (and hence probably pool) in Africa, the other three being: West (Niger-Congo peoples) South (Khoisan) and North. However, does the existence of this East African pole dismantle the idea of West Eurasian admixture or do both co-exist?

    In a different case, what about the Fulani? I am quite certain that the Fulani make on their own a distinct genetic pool within Africa (based on Bryc 2009 and on Cerný 2006), however they do appear in your PC graphs to be intermediate between West Africans and North Africans.

    I strongly suspect that, regardless of some West Eurasian or North African admixture, there is also a lack of good description of many different (but small) African unique components. What do you think?

    ReplyDelete
  10. And also, Etyopis: are you planning to continue your blog? I see that you write only very rarely, maybe once a year - a little low threshold to add it to my blogroll list. But I feel that African (African-oriented, African-created) anthropological and genetic blogs and sites are lacking... so what are your plans for your blog?

    ReplyDelete
  11. Some great points and questions raised Maju, I will try and address some of them as much as I can, I will go from last to first.

    “And also, Etyopis: are you planning to continue your blog? I see that you write only very rarely, maybe once a year - a little low threshold to add it to my blogroll list. But I feel that African (African-oriented, African-created) anthropological and genetic blogs and sites are lacking... so what are your plans for your blog? “

    Thanks for considering adding my blog to your list, unfortunately my blog is really less of a blog and a personal diary for myself where I store stuff every now and then, I would NOT recommend you adding it as I very sporadically will update it due to the limited amount of time I have to spend on this type of stuff.

    “Recently it has been claimed often that the Maasai have important West Eurasian admixture but I have many doubts. In this sense your 3D graph is very illustrative, because the Maasai and Ethiopians seem to define a fourth genetic pole (and hence probably pool) in Africa, the other three being: West (Niger-Congo peoples) South (Khoisan) and North. However, does the existence of this East African pole dismantle the idea of West Eurasian admixture or do both co-exist? “

    I think it is only fair that first you should know the steps I took to construct those graphs.

    1) I downloaded a global dataset from this link here :
    http://blogs.discovermagazine.com/gnxp/2011/03/analyzing-ancestry-with-admixture-step-by-step/

    2) This global Dataset included 3,970 individuals typed at approximately 27,000 SNPs, with a total of 172 uniquely grouped or entered populations/ethnicities/nationalities.

    3) I then performed an MDS analysis on the Dataset to verify that the Dataset's variation structure is by and large consistent with peer reviewed PCA results on the global level. Which it was, see here:
    Isometric view:
    http://4.bp.blogspot.com/-fXn58sgadX0/TbxFklYYM4I/AAAAAAAAAHc/NzgvlQRdS0I/s1600/Global_MDS_Average.png
    C1_C2 View:
    http://4.bp.blogspot.com/-xwtq-29Crfk/TbxFlCdjC2I/AAAAAAAAAHg/RtkskIdOA4s/s1600/Global_MDS_Average_C1C2.png

    Note that here, I labeled the MDS points by the AVERAGE of the coordinates for each uniquely entered populations, where as sometimes I use the MEDIAN position for each population to label them, there is a difference sometimes but most the time it is pretty close.I also split the Ethiopian data from the original data set into the respective ethnic groups from Behar et. Al 2010, I.e EtO = Ethiopian Oromo, EtT = Ethiopian Tigrayan and EtA = Ethiopian Amhara and EtJ=Ethiopian jews, it is also important to note that the Ethiopian Oromos in this dataset are from the Southern border of Ethiopia with Kenya, so likely from the Moyale area. This means that instead of 172 uniquely entered populations I now had 174.

    4) Finally, I split the dataset by taking all non-continental Africans out of it, leaving with a total sampling of 1151 individuals and 48 uniquely entered populations from Africa. Then, I performed a new MDS analysis to get the results that I posted yesterday on your blog.

    ReplyDelete
  12. To answer your question whether the Eurasian 'Admixture 'exists in the Massai, right now I think it is very hard to say with the tools geneticists have, what is not hard to say is that genes are clinal in nature, the 'Eurasian' element present in the maasai could just simply be explained as Ancestral variation of Eurasians present in the Maasai, as Henn et.al Notes:

    At k = 4, we see a western African/ Bantu-speaking cluster, an eastern African cluster, a cluster representing Europeans that likely also signifies ancestral variation maintained in eastern Africa (e.g., Maasai and Sandawe populations), and finally, a cluster that links all our HG populations. Populations in eastern African have the highest proportion of the European ancestral cluster (represented by the Italian Tuscans, HapMap3), supporting prior models of the Out of Africa (OOA) migration originating from a population of eastern African ancestry (1).


    While some of the Eurasian element in the Maasai could be also from some very ancient back migrations and/or interactions of ancestral populations of the Maasai with Eurasians. For this, we can take a look at the uni parental results for the maasai:

    YDNA from wood et. Al 2005:-
    A-M13 = 27%
    B-M152 = 8%
    E- P1* (x M191) = 12%
    E-M191 = 4%
    E-M35* (x M78, M81) = 35%
    E-M78 = 15%

    mtDNA from Castri et. Al 2008
    L4g ~12%
    M1a ~12%
    L5 ~8%
    L2b ~8%
    L3* ~8%
    L0f ~5%
    L2a ~5%
    L3x ~5%
    L3f ~4%
    L3d ~3%
    L3i ~3%
    R0 ~3%
    L0* ~2%
    L0d ~2%
    L3b ~2%
    L3e ~2%
    L3h ~2%
    L3w ~2%
    N1a ~2%

    To be continued...............

    ReplyDelete
  13. Based on what you mention re. Maasai mtDNA, I will have to accept that they probably have some West Eurasian admixture of about maybe 15%, which is roughly what I can appreciate in the autosomal Admixture analysis in this other post by Razib.

    The question of whether the Fulani (particularly the original Fula people of Westernmost Africa) have West Eurasian admixture seems to remain open however.

    As for your blog, all blogs are open diaries - just that the word diary comes from "day" (a notebook you write more or less daily), not from "year". That would be an annuary ;)

    ReplyDelete
  14. Well it can be more complex than that however since on an autosomal(microsatellite) basis Tishkoff et al 2009's supervised STRUCTURE run did not show show much amount of an element in the Maasai that would be read as West Eurasian Admixture:

    48.45% Cushitic Cluster,
    18.31% Nilo-Saharan Cluster,
    14.81% Niger Kordofanian cluster,
    5.81% Eastern Bantu cluster
    4.33% Chadic Saharan cluster.


    The one very important type of population we are missing with all the Autosomal ADMIXTURE runs done by the bloggers are Nilo-Saharans, especially from southern Sudan like the Dinka, Neuer, Shilluk, etc,,,This may very well change the results of the Maasai and many East African populations, especially for the Maasai since some of their ancestors are thought (conventionally) to be Cushitic speakers that have migrated from further up North in Ethiopia and mingled with incoming Nhio-Saharans.

    Additionally, Cushitic- speakers from the Ethiopian highlands migrated northward to the Red Sea coast of Sudan (where their modern-day descendants, the Beja, presently live) as well as southward into Kenya and Tanzania [6].
    ^Campbell, Tishkoff 2009

    ReplyDelete
  15. Damn, Blogger sabotaged my post. Lost all content. :(

    I was saying, that, following what you say, there is no WEA admixture but it'd be an illusion mediated by the Nile buffer zone, much as happens with the YRI component in Europe or NW Africa.

    I do not have direct access to Tishkoff 2009, so I will have to rely on Dienekes (who has his own questionable opinions). Following the Admixture panel there, 2/3 Maasai subgroups show up as having (visually) some 10% West Eurasian (blue) component, distinct from the East African (purple) and Nilotic (red) components.

    Still I fail to see what logic lays behind this WEA genetic infiltration (other than mtDNA haplogroup M1, which is just too old in the region to be able to explain anything). So I'm kind of not fully believing it... but unsure.

    ReplyDelete
  16. You can find the Tishkoff et. al 2009 paper here for free:

    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2947357/

    The supplementary material, which has over 100 pages can also be downloaded from that same link.

    There are also 2 additional papers that are linked to the Tishkoff paper--

    1)Working toward a synthesis of archaeological,linguistic, and genetic data for inferring African population history:

    http://www.pnas.org/content/107/suppl.2/8931.short

    and

    2)The Evolution of Human Genetic and Phenotypic Variation in Africa:

    http://www.cell.com/current-biology/abstract/S0960-9822%2809%2902065-X

    In my opinion, the Tishkoff study is the most important study on African genetics to date and well worth your time if interested.

    With regards to the West Eurasian Admixture and/or common ancestry present in East Africa, I am of the opinion that at the end of the day, it all boils down to personal philosophy, if one is to believe that there was ever a time in Human history that 'pure' races or populations ever existed, as is the current status quo, then the genetic evidence in East Africa would be the perfect place where such a notion would be relegated to the dust bins of history for good. So from my POV, it is good that you are unsure, if there is West Eurasian Admixture or not in populations like the Maasai, we need more uncertainty in this issue than certainty.

    ReplyDelete
  17. Thanks for the PubMed link. I did not expect it to be freely available there :)

    It is, as you say, a most important paper, indeed.

    On the matter of Eurasian admixture, I do not see that as a matter of "races" because, following genetics there's no such thing as an Eurasian, much less West Eurasian, "race" - at least not a genetically acceptable such concept: Eurasians aren't but a subset (with whatever Neanderthal, etc. minor admixture) of Eastern Africans, which in turn aren't but a subset of the generic African-3 (i.e. without Pygmies nor Bushmen) set, etc. This is easy to see in fig. 1.

    But the components we can distill with Admixture and similar programs are too shallow for that. But in that sense they are informative in a less profound (deep) way: they may tell us of flows that happened since the Holocene or Late Pleistocene.

    And that is interesting to know about. Once we make sure that mirages are not happening (because of using the wrong reference populations, sample sizes, etc.) the results of this kind of analysis are interesting.

    And rare stuff is interesting, so if Finns and some neighbors are the only ones in Europe displaying noticeable East Asian (or Siberian) admixture, that is interesting to know and see if it is confirmed. It is as far as I can tell.

    Similarly it is interesting to find that some particular populations in Tropical Africa have West Eurasian admixture. What I want to know is whether this is real or is a mirage. And in the case of the Maasai it seems real.

    On the contrary, we do not see such admixture among the Fulani, another group claimed to have West Eurasian influences. And that is interesting to know as well.

    It's not (at least not for me) a matter of "race" (this if anything would be anecdotal) but a matter of reconstructing human history by means of archaeo-genetics (complementarily to other disciplines). So, when we see that Finns have Y-DNA and autosomal DNA and even some less important mtDNA from Siberia/NE Asia, we can infer that a migration happened through the arctic circle, even if we only know of its latest episode (Volga to Finland, Combed Pottery culture) by means of archaeology.

    Similarly by identifying this rather unexpected WEA influx into parts of East Africa we can infer some sort of migration, maybe related to Neolithic flows or whatever else (I can't really say at this point). And this is interesting.

    That's my POV on this matter, really.

    ReplyDelete
  18. Now, looking at the S1 material, in the African-only Admixture run, the Maasai do not look anymore so much West Eurasian, instead the Oromo and Rendille (Cushitic peoples) do. This component persist being the dominant one in Mozabites (though there's some weirdness depending on the K number) and one of the main ones among Colored South Africans (CMA), so it's surely the same blue "WEA" component as in the global run.

    In any case it's interesting to look at Africa at a distinct self-contained entity and it's a pity that this graph is hidden in the supplemental material and not first page.

    ReplyDelete
  19. "Because it's convenient within their conceptual time-frame, which is judeocentric, and also because, working for a Zionist institution, they seem to lack access to other samples of the region. I was going to mention that in the original entry but I decided it could be seen as unnecessarily political. But it's there anyhow (science does not happen in a bubble)."

    that is right Maju ,I was all along suspicious of Behar's work too (he inflated the sub-saharan admix in Moroccans to almost Ethiopian level)...
    I posted some pics of the people from Amizmiz and Asni of the high Atlas on anthroscape forums:

    http://s1.zetaboards.com/anthroscape/topic/4260982/1/#new

    http://s1.zetaboards.com/anthroscape/topic/4262536/1/#new

    ReplyDelete
  20. From your more recent blog:

    "E1b1b1a1 (M78) seems to represent (together with J1 and maybe other more localized lineages like R1b1a) the flow of Afroasiatic languages from the area of Egypt/Sudan".

    Actually the whole of E1b1b1-M35 could be at the base of Afro-Asiatic language spread. E1b1b1c-M123 is at least as 'Jewish' as E1b1b1a1-M78. Certainly E1b1b1 is very much present in the region where Austro-Asiatic languages are spoken today. The regions where the haplogroup and the languages don't both exist are easily explained way by later language replacement.

    "if we are to make any association between AA languages and E1b, it must be largely pre-Neolithic".

    OK. E1b's expansion is obviously 'largely pre-Neolithic'. But E1b1b1-M35's is later than that of E1b-P2's, and is probably associated with Austro-Asiatic.

    "No and I don't see how you can reach to any such conclusion".

    We need to get this straight then. So you accept that basal haplogroup diversity indicates region of origin, except when that would conflict with what you alreay believe. Is that correct? Yet at this blog you wrote:

    "I must agree that E1b1b1 looks, per basal diversity (subclades d, f and g specially, plus subclade e in East Africa) like Ethiopian originated. The genetic evidence looks like supporting that (at present level of knowledge)".

    And E1b1b1-M35's expansion is 'starlike'. When considering other haplogroups you have always claimed that this is an indicator of rapid expansion. Do you always alter your interpretation of the evidence to suit your pre-existing belief? I suspect that a starlike pattern of haplogroup phylogeny can mean a population growth rather than necessarily expansion. But Population expansion is usually preceded by a growth in population, and I'm sure that is what we see in the case of E1b1b1-M35's starlike pattern. The same with O3a-M324.

    (continued)

    ReplyDelete
  21. "for me Sardinian E1b is derived from their surroundings and a 'recent' arrival".

    I agree with the recent arrival bit. But the only E1b1b-M215 haplogroups in North Africa outside the Northeast are E1b1b1a1 and E1b1b1b. Both haplogroups are also found scattered through the Mediterranean, and neither have close relations in North Africa. So it is most parsimonious to conclude that their presence in North Africa is a product of their presence in the Mediterranean, rather than the other way round.

    "If you want Sardinia to play any role, the less you can do is document your claims (link preferably)".

    Good old Wiki again:

    http://en.wikipedia.org/wiki/Haplogroup_E1b1b_(Y-DNA)

    Quote:

    "E1b1b1a (E-V68), is dominated by its longer-known sub-clade E-M78 (E1b1b1a1). Three 'E-V68*' individuals who are in E-V68 but not E-M78 have been reported in Sardinia, by Trombetta et al. (2011), when announcing its discovery. The authors noted that because E-V68* was not found in the Middle Eastern samples, this appears to be evidence of maritime migration from Africa to southwestern Europe".

    And:

    "E1b1b1b is dominated by its dominant sub-clade E1b1b1b1 (E-M81) ... V257's discovery was announced in Trombetta et al. (2011) ... They found 6 'E-V257*' individuals in their samples who were E-V257, but not E-M81. A Borana from Kenya, a Marrakesh Berber, a Corsican, a Sardinian, a southern Spaniard and a Cantabrian".

    Surely the presence of these two widespread basal haplogroups suggests strongly that E1b1b1-M35's expansion was rapid, and included the Mediterranean islands. And here's the article the haplogroup diagram in Wiki is based on:

    http://www.plosone.org/article/info:doi/10.1371/journal.pone.0016073

    An earlier comment from Etyopis:

    "All evidence points to Ethiopia as the origin of E1b1*,E1b1b* and E1b1b1*. (Cruciani '04,'07 ,Trombetta '11)"

    Probably more specifically the Horn but I basically agree.

    "when discussing Trombetta's paper, I argued that E1b1b1b1-M81 could well be very ancient in North Africa (and hence in Iberia)".

    As I pointed out above: unlikely.

    ReplyDelete
  22. It can be interpreted as you say, specially if you argue for an origin for Afroasiatic (not "Austro-Asiatic"!) in the tribal zones of Ethiopia (and/or neighboring Upper Nile areas) but it does not need to be that way, specially if you don't mind an origin of Afroasiatic in the Nubia area, which has been argued for as well with legitimacy.

    This is easier to favor if you consider Omotic non-AA or even if you consider Omotic so extremely upstream in the AA tree that it's not really true AA but proto-AA, belonging therefore to an older process.

    So focusing on E1b1b1, it probably spread from the Upper Nile area. When? I'd say that long ago because there is little else where it is, so it has required long time and low densities to reinforce such almost absolute dominance by means of drift.

    Another reason I have to support an older than AA first spread for E1b1b1 is the presence of E1b1b1b1 in West Iberia in frequencies and locations that cannot really be explained well by Neolithic, much less post-Neolithic flows.

    North African "Capsian" Neolithic was no source for Iberian one, which belongs to the Cardium Pottery culture initially, original from the West Balcans. So explaining E1b1b1b1 (and mtDNA U6, etc.) in West Iberia in the large amounts it's found, would require an extreme coincidence of founder effect by means of picking people at Tangiers and then making them colonize West Iberia massively.

    By comparison, East Mediterranean lineages such as J2, E1b1b1a1, G2a... are not so prominent in that region, so it really demands not just quite radical colonization but also a rather unbelievable unlikely coincidence of this one being performed by people picked in North Africa, where Cardium Pottery is extremely rare.

    Later it looks even less likely. In the Megalithic period there could be flows from Iberia to North Africa (not detected in the gene pool) but the opposite is less likely and in other periods like Roman or Cordobese Muslim no reason is apparent for such a flow specifically to West Iberia (and not South or East).

    At first I also thought that it was a Neolithic founder effect but the more I look at it the less likely it appears. Instead there is a mechanism in the Oranian-Solutrean period, when Asturias received settlers from Iberia proper (via probably the West) in a context of intense connections with north Africa (Oranian genesis and related technological flows and backflows acorss Gibraltar Strait).

    So E1b1b1b1 was surely in NW Africa some 20 Ka ago, so the "starlike" expansion of E1b1b1 you believe to see, happened before that date (IMO in the Aterian genesis, roughly the same dates as the OoA). Did they speak proto-Afroasiatic? Maybe but it's highly speculative.

    ReplyDelete
  23. Otherwise, it is very interesting the presence of paragroups E1b1b1b* and E1b1b1a* in SW Europe. However I do not really feel safe about the commonality or rarity of these two para-clades in North Africa and therefore of the meaning of these findings.

    It is probable that the Western Mediterranean E1b1b1b* is a single lineage at least in Europe and that it indicates an old flow reaching out to the Western Mediterranean islands somehow. But without clear data, specially on North Africa and haplotype structure it is impossible to tell.

    It does not demonstrate anything of what you want it to demonstrate. For example the Sardinian E1b1b1a* could have arrived from North Africa in the Phoenician period or whatever else. But Sardinia was colonized first before Neolithic anyhow.

    Similarly it is very possible that E1b1b1a1 arrived to Greece in the Mesolithic period as well (and maybe by crossing the Mediterranean on ships).

    ReplyDelete
  24. Thanks for the very reasoned response. Different from your usual.

    "It can be interpreted as you say, specially if you argue for an origin for Afroasiatic (not 'Austro-Asiatic'!)"

    Sorry. You are obviously correct.

    "the tribal zones of Ethiopia (and/or neighboring Upper Nile areas) but it does not need to be that way, specially if you don't mind an origin of Afroasiatic in the Nubia area, which has been argued for as well with legitimacy".

    Either could be the case, but I think most argue for Ethiopia/Horn.

    "So focusing on E1b1b1, it probably spread from the Upper Nile area. When? I'd say that long ago because there is little else where it is, so it has required long time and low densities to reinforce such almost absolute dominance by means of drift".

    I'd agree that it was probably in 'the Upper Nile area ... long ago', but that in no way means it 'spread' from there long ago. Several indigenous Upper Nile haplogroups, such as E1b1b1f and E1b1b1g, have obviously been drifted to low levels.

    "it is very interesting the presence of paragroups E1b1b1b* and E1b1b1a* in SW Europe. However I do not really feel safe about the commonality or rarity of these two para-clades in North Africa and therefore of the meaning of these findings".

    'Commonality or rarity' is fairly irrelevant. It is their near relations' absence in western North Africa that is the telling factor. Both haplogroups could easily have arrived along the western end of North Africa at the same time as they arrived in SW Europe.

    "It is probable that the Western Mediterranean E1b1b1b* is a single lineage at least in Europe"

    But E1b1b1b1 is also found in Europe, so E1b1b1b cannot be 'a single lineage' in Europe. There are at least two, given that E1b1b1b* may be a 'single' lineage.

    "it indicates an old flow reaching out to the Western Mediterranean islands somehow".

    I don't think we could argue for 'an old flow reaching out to the Western Mediterranean islands', especially seeing it seems to be so widespread. That argues fairly efficient boating.

    "But without clear data, specially on North Africa and haplotype structure it is impossible to tell".

    The new data seems fairly clear as to North Africa. Maghreb E1b1 haplogroups are solely the two Mediterranean ones E1b1b1a1 and E1b1b1b1.

    "For example the Sardinian E1b1b1a* could have arrived from North Africa in the Phoenician period or whatever else".

    Possible.

    "But Sardinia was colonized first before Neolithic anyhow".

    But perhaps not by members of Y-hap E.

    "Similarly it is very possible that E1b1b1a1 arrived to Greece in the Mesolithic period as well (and maybe by crossing the Mediterranean on ships)".

    I agree they must have crossed the Mediterranean (or gone around the coast) on ships. But I think you'd have a difficult job claiming ships were capable of journeying from one end to the other through the Mediterranean during the Mesolithic.

    "the presence of E1b1b1b1 in West Iberia in frequencies and locations that cannot really be explained well by Neolithic, much less post-Neolithic flows".

    It can be explained as such if its presence there is a product of the spread of E1b1b1 through the Mediterranean. It is simply the haplogroup that made it furthest to the west.

    (Continued)

    ReplyDelete
  25. "North African 'Capsian' Neolithic was no source for Iberian one, which belongs to the Cardium Pottery culture initially, original from the West Balcans".

    I'd agree that E's spread through the Mediterranean is probably pre-Cardial. But perhaps not by much. Cardial most likely involved J2 and G2a. But the technology spread beyond the haplogroups' spread.

    "explaining E1b1b1b1 (and mtDNA U6, etc.) in West Iberia in the large amounts it's found, would require an extreme coincidence of founder effect by means of picking people at Tangiers and then making them colonize West Iberia massively".

    Which suggests that E1b1b1b1 didn't originate in Tangiers, but further east.

    "By comparison, East Mediterranean lineages such as J2, E1b1b1a1, G2a... are not so prominent in that region"

    But E1b1b1a1 is prominent in the Central Mediterranean. And E1b1b1c1 tends to be Eastern Mediterranean. To me it seems quite possible that the three haplogroups (E1b1b1b, E1b1b1a and E1b1b1c) coalesced separately somewhere in each of the three regions in the Med, from west to east respectivley.

    "In the Megalithic period there could be flows from Iberia to North Africa (not detected in the gene pool) but the opposite is less likely".

    So perhaps neither scenario is the case.

    "the 'starlike' expansion of E1b1b1 you believe to see, happened before that date (IMO in the Aterian genesis, roughly the same dates as the OoA)".

    I can't see how you reach that conclusion.

    "Did they speak proto-Afroasiatic? Maybe but it's highly speculative?"

    If the expansion was early Neolithic it is quite possible that they did so.

    ReplyDelete
  26. Are you suggesting that E1b1b1b and E1b1b1a originated in Sardinia (or otherwise Europe)?

    I think that the key is that we really do not have enough data about these para-groups in Africa, because we tend (very strongly) to oversample Europe and undersample Africa.

    But I do not see any reason to expect E1b1b1b1 (which is almost exclusive of NW Africa) to have arrived to SW Europe "at the same time" as its cousin E1b1b1a1 (much more widespread) arrived to SE Europe. There are several reasons, like SW Europe having high densities long before SE Europe or like the archaeological evidence of Iberia-NW Africa strong interactions (only) in the Oranian genesis (and Solutrean scatter within Iberia itself).

    Nothing like that happened in Greece c. 20 Ka ago and also Greece is a lot more distant from Africa than Iberia is, even if we consider Crete or West Asia (neither of which seems to be the bridge, at least not for E1b1b1a*).

    So I think that by factoring "Greek" E1b1b1a1b (V13) as if it'd be the core of E1b1b1a and not just another branch, you totally miss the big picture. It is E1b1b1a1b which arrived to Greece at the origins of European Neolithic, not all E1b1b1, some of which was here earlier (as it seems obvious).

    It's fair to wonder if E1b1b1a1b could have arrived to Greece via Europe instead of West Asia, maybe carried by the same people who might have brought rock art expression to Turkey via Dalmatia.

    It looks in any case as a strong founder effect. But I see no particular reason to expect E1b1b1-whatever to be the carrier of such a West European artistic expression (rather the much more common R1b1b2a1). So probably West Asia, I guess.

    (cont.)

    ReplyDelete
  27. (...)

    "I think you'd have a difficult job claiming ships were capable of journeying from one end to the other through the Mediterranean during the Mesolithic".

    Bah: they did colonize Cyprus and Sardinia in that period and longboats from Denmark are known to have existed then as well, so I guess navigation truly expanded its horizons (literally) in that period.

    If you can go to Sardinia, you can reach any other random point in the Mediterranean. Not that they did necessarily but it was obviously possible already.

    I'd even bet that this unique sailor ability of some pre-Neolithic peoples would be why Cardium Pottery existed at all, at least with that extension.

    "... the spread of E1b1b1 through the Mediterranean. It is simply the haplogroup that made it furthest to the west"

    Not sure what you mean here. E1b1b1b1 is a Western haplogroup by origin, IMO. Talking of E1b1b1 as a catch-all term, we miss the palette of the painting and hence the whole message conveyed by it.

    If you mean E1b1b1a1, specially the Balcanic variant E1b1ba1b, indeed. And that I explain by the following archaeologically traceable processes:

    1. Capsian culture and other flows related to Afroasiatic expansion from Nubia.

    2. European Neolithic (and to some extent also post-Neolithic) demic flows from the Balcans.

    "E's spread through the Mediterranean is probably pre-Cardial. But perhaps not by much".

    It's not a matter of time but a matter of logic of related archaeological stuff we can say: "this is it" with some confidence.

    "Which suggests that E1b1b1b1 didn't originate in Tangiers, but further east".

    What?! I refuse to fall into this sickening pseudologic.

    E1b1b1b1 is not really found in Europe but in Iberia (and some, mostly Atlantic, offshoots) so WTF "East"? What East?

    I insist that European Neolithic is 99% a European process, at most with a West Asian ultimate origin... but there is no E1b1b1b1 in West Asia (or otherwise Eastern Mediterranean but in Egypt at low doses).

    So where on Earth do you place that "East"? In Low Egypt? Not credible.

    "But E1b1b1a1 is prominent in the Central Mediterranean".

    ...a1 is mostly irrelevant in order to explain ...b1. Specially when it is not even all ...a1 but just ...a1b, one (and only one) among several subclades of ...a1.

    "I can't see how you reach that conclusion".

    Because I know of no other clear demic flow from the Upper Nile to NW Africa.

    I do not say there is not but show me when if you have a better suggestion...

    Aterian dates are still under revision but overall the picture seems to be one of continuity between c. 100 Ka (very roughly) to the Oranian period. Correct me if I am wrong in this of course.

    There are the "Aurignacoid" Dabban industries in Libya, which may be attributed with brining mtDNA U6 to the area, but so far I know of no such archaeological findings further West.

    The Dabban period could hypothetically be another moment for E->W genetic flow in North Africa but it's poorly documented.

    "If the expansion was early Neolithic"..

    It was not. Quit trying to fit reality into your preconceptions.

    ReplyDelete
  28. "Quit trying to fit reality into your preconceptions".

    I have no preconceptions at all concerning Y-hap E. I'm merely looking at the evidence. Which, by the way, shows that you have an incredible number of preconceptions.

    "Talking of E1b1b1 as a catch-all term, we miss the palette of the painting and hence the whole message conveyed by it".

    I am in fact looking at 'the palette of the painting'. But I get the distinct impression that you are ignoring it.

    "Are you suggesting that E1b1b1b and E1b1b1a originated in Sardinia (or otherwise Europe)?"

    Most certainly not. The closest relations of both are found in the Horn of Africa and through the Sudan and Egypt.

    "So I think that by factoring 'Greek' E1b1b1a1b (V13) as if it'd be the core of E1b1b1a and not just another branch, you totally miss the big picture".

    There is no way that E-V13 is the core of E-V68. E-V13 is a downstream mutation from E1b1b1a-M68. E1b1b1a1b-V13's closest relations are Egyptian and/or Sudanese (E-1b1b1a1a-V12 and E1b1b1a1c-V22), Greek (E1b1b1a1e-M251) and Maghreb, Italy and Spain (E1b1b1a1d-V657). So E-V13 probaby spread from Egypt/Sudan, according to your concept that basal haplogroup diversity indicates origin.

    "they did colonize Cyprus and Sardinia in that period and longboats from Denmark are known to have existed then as well"

    Really?

    "E1b1b1b1 is a Western haplogroup by origin, IMO".

    Quite probably so in the particular case of the derived E1b1b1b1-M81, but its ancestor E1b1b1b-V257's closest relations are found in the Horn or in Ethiopia (E1b1b1g, E1b1b1f, E1b1b1d), The Horn and the Middle East(E1b1b1c), Tanzania/Southern Africa (E1b1b1e), with just one other: E1b1b1a, whose distribution I've just covered above. So, again according to your theory, E1b1b1-M34 also spread from Egypt/Sudan. Unless there was considerable back movement into Upper Egypt I'd guess that the few members of Y-hap E1b1b1-M34 that emerged from Egypt/Sudan did so along with the several members of E1b1b1a that also emerged from Egypt/Sudan.

    "...a1 is mostly irrelevant in order to explain ...b1. Specially when it is not even all ...a1 but just ...a1b, one (and only one) among several subclades of ...a1".

    Surely you can see that I totally accept that statement.

    "Because I know of no other clear demic flow from the Upper Nile to NW Africa".

    So what has that got to do with E1b? It clearly wasn't involved in any 'demic flow from the Upper Nile to NW Africa'. The evidence of haplogroup distribution shows it entered NW Africa from the Mediterranean.

    "There are several reasons, like SW Europe having high densities long before SE Europe or like the archaeological evidence of Iberia-NW Africa strong interactions (only) in the Oranian genesis (and Solutrean scatter within Iberia itself)".

    Well, according to Wiki:

    "E1b1b1b1 (E-M81), formerly E1b1b1b, E3b1b, and E3b2, is the most common Y chromosome haplogroup in the Maghreb, dominated by its sub-clade E-M183. It is thought to have originated in the area of North Africa 5,600 years ago"

    Note: '5,600 years ago'. I agree that E's expansion is probably not strictly 'Neolithic', but it is almost certainly no earlier than Holocene.

    ReplyDelete
  29. Yes, Terry, really! The oldest boat known in Europe and maybe in the World is a Danish longboat from the Maglemosian period (Epipaleolithic, c. 11,000-7000 BP).

    This does not mean that the proto-Danes invented boats (because a lot of chance and conservation circumstances went on that - peat bogs are incredible!) but it means that there were longboats already at the very beginning of Epipaleolithic in West Eurasia.

    Something that, no doubt, has something to do with the (arguable?) early colonization of Cyprus, Sardinia and Corsica.

    You don't need to be able to grow wheat in order to tame a goat nor you need any of these land abilities in order to build a boat and use it for fishing and traveling.

    ReplyDelete
  30. "The evidence of haplogroup distribution shows it entered NW Africa from the Mediterranean".

    Uh?

    I cannot agree with that. At least for E1b1b1b1, we are before a clear case of NW African haplogroup which have expanded from there.

    Even in the case of E1b1b1a1, it clearly arrived from Egypt (along with J1) via inland routes, not the sea. This is coincident with the archaeological pattern of Capsian culture - in a time when the Sahara was much more fertile than it is now.

    "according to Wiki"...

    According to Wiki you give me a TMRCA hunch and expect me to swallow? Please!

    I'm not sure what to think of E1b1b1b* and specially E1b1b1a* but I would think that, as they are not found in West Asia or the Balcans, expanded via Iberia (or maybe, as you seem to believe, across the Med directly from Tunisia into Italy).

    I doubt that they can demonstrate anything, unless you'd like to claim a (highly dubious) European origin for E1b1b1a, as it'd be in Sardinia (according to your data) where the highest basal diversity is. I think this is just a problem of undersampling in Africa and oversampling in Europe, specially in Sardinia.

    ReplyDelete
  31. "I think this is just a problem of undersampling in Africa and oversampling in Europe, specially in Sardinia".

    Possibly, but we have to work with the data we have. And E is well sampled compared to C and O.

    "it means that there were longboats already at the very beginning of Epipaleolithic in West Eurasia".

    Maju. Not 'longboats'. The link specifies that the 'plank boat' (long boat) 'appeared in the Middle Bronze Age'. And the link specifies that it was in the Neolithic that people used dugout canoes. And the date 'c. 11,000-7000 BP' is quite within the time frame I'd accept for Y-hap E having reached the Mediterranean islands.

    "I cannot agree with that. At least for E1b1b1b1, we are before a clear case of NW African haplogroup which have expanded from there".

    I wouldn't claim it is actually so 'clear'.

    "Even in the case of E1b1b1a1, it clearly arrived from Egypt (along with J1) via inland routes, not the sea".

    I'd agree except for one very inconvenient fact. E1b1b1a1-M78 is derived from E1b1b1a-V68. Regarding that haplogroup: 'E1b1b1a* Found in individuals in Sardinia'. So it could be that Sardinia was a staging post on the expansion into the Mediterranean, but it's hard to argue that E1b1b1a1 derives via an 'inland route' from Egypt.

    "as they are not found in West Asia or the Balcans, expanded via Iberia"

    Y-hap E1b1b1c is certainly found in West Asia. So an expansion through the Mediterranean is quite likely. If a relatively small group expanded through the Mediterranean it would hardly be surprising if they broke into discrete related haplogroups wherever they set up their homes.

    "Something that, no doubt, has something to do with the (arguable?) early colonization of Cyprus, Sardinia and Corsica".

    Did you actually read the first chapter of the Cyprus link, 'The First Humans and the Last Pygmy Hippopotami on Cyprus'? Quote: 'Despite such claims, Cherry has convincingly demonstrated that there are few, if any, compelling arguments for people having been on most of the Mediterranean islands during the Late Pleistocene or early Holocene'.

    And the Sardinia link: 'A much more likely starting moment for this extinction, the arrival of humans, has been well argued by several other scholars'. Further: 'The chronological position of the Corsican [and by association Sardina] Mesolithic site, occupied by permanent hunter-gatherer groups, covers the Middle and second half of the 9th Millenium'. And the Corsican link dates human arrival there to the '7th Millenium BC'.

    So there you have the date for E's arrival in Sardinia.

    "as you seem to believe, across the Med directly from Tunisia into Italy"

    I would have favoured that view in the past, but it now looks unlikely.

    "This is coincident with the archaeological pattern of Capsian culture - in a time when the Sahara was much more fertile than it is now".

    Just 'wishing it so' does not 'make it so'.

    ReplyDelete
  32. "... we have to work with the data we have".

    We have no choice but we should not overstretch our conclusions when the extant data is feeble or otherwise inconclusive.

    "And E is well sampled compared to C and O".

    Africa is not better sampled than East Asia, not at all.

    "Maju. Not 'longboats'".

    Log boat. My bad.

    A quite decent ship in any case able to be equipped with a sail.

    Still I do not see a clear difference between a log and a long boat: both would accommodate several oarsmen, right? Same for leather canoes. All are about the same thing to me, just different construction techniques but similar result.

    "E1b1b1a1-M78 is derived from E1b1b1a-V68. Regarding that haplogroup: 'E1b1b1a* Found in individuals in Sardinia'".

    We have previously found Sardinian Y-DNA lineages deep in Africa. Sample properly and you'll doubt should be resolved.

    Otherwise you're basically saying that E1b1b1a coalesced in Sardinia and that is extremely implausible. It could make for a great "Sea Peoples" fantasy story though :)

    As for Epipaleolithic colonizations in Cyprus and Sardinia/Corsica, I already said it's arguable but, on second read, the Corsican case is not: people lived in the Mediterranean island (and hence in nearby Sardinia surely too, only separated by a narrow strait) at least 1500 years before the arrival of Neolithic and high-seas navigation to the region (c. 8500 BP vs. c. 7000 BP).

    On the other side, I'm uncertain on being glad to have persuaded you by such indirect and unwilling means that the hippos of Cyprus died of "natural causes", without human intervention. Actually, in this particular case, I do think that human intervention was decisive in the extinction of dwarf hippopotamuses (we are discussing an island after all) but you must claim that the hippopotamuses died of natural causes if you are to deny Mesolithic human colonization of Cyprus - and I doubt you can do that without a major logical short-circuit happening in your brain.

    "And the Corsican link dates human arrival there to the '7th Millenium BC'".

    Yes, that's a lot earlier than Neolithic, which only arrived in the late 6th or early 5th millennium BCE. Some 1500 years earlier.

    Further to the west Neolithic would take still many more millennia to arrive (probably because a barrier of humidity and oceanic cloud cover, as well as disinterest by the natives). Initial European Neolithic does not have a single date but a range of more than four millennia.

    I say because I know you're thinking in terms of Eastern Mediterranean Neolithic here, which was indeed advanced for the 7th millennium but at that time farmers were still reaching out to the Balcans.

    So you have the earliest European Neolithic in Thessaly in the 8th millennium and some of the latest European Neolithic in most of Atlantic Europe not earlier than the 4th millennium. It's double the time between Caesar and us, about as much as what separates us from Minoan, Indus Valley or Olmec earliest civilizational impetus.

    A whole History! Many eras!

    West Mediterranean neolithic is closer to that of the East than to that of the Atlantic but still some 1500 years younger. It's not that they were in any rush, the colonization (and assimilation) was very slow and gradual.

    ReplyDelete
  33. “Still I do not see a clear difference between a log and a long boat”

    Really? Next time try thinking before you tap the computer keys. Why would people bother to go to the trouble of hollowing the log out?

    “I do think that human intervention was decisive in the extinction of dwarf hippopotamuses”

    I'm certainly glad to have persuaded you that the hippos and elephants, along with myotragus and various species of giant dormice and dwarf deer, died out in the Mediterranean islands as a result of human intervention.

    “but you must claim that the hippopotamuses died of natural causes if you are to deny Mesolithic human colonization of Cyprus - and I doubt you can do that without a major logical short-circuit happening in your brain”.

    You obviously don’t bother to read what I write in my comments. I’ve known for years that the extinctions on the Mediterranean islands occurred some time before the Neolithic reached them. I recently wrote:

    “I agree that E's expansion is probably not strictly 'Neolithic', but it is almost certainly no earlier than Holocene”.

    So you see: we agree.

    “people lived in the Mediterranean island (and hence in nearby Sardinia surely too, only separated by a narrow strait) at least 1500 years before the arrival of Neolithic and high-seas navigation to the region (c. 8500 BP vs. c. 7000 BP).”

    So at last we agree that people reached the islands no earlier than around 8,500 years ago and caused the extinctions on them. We just don’t agree yet that these people carried Y-hap E.

    “Yes, that's a lot earlier than Neolithic, which only arrived in the late 6th or early 5th millennium BCE. Some 1500 years earlier … Further to the west Neolithic would take still many more millennia to arrive (probably because a barrier of humidity and oceanic cloud cover, as well as disinterest by the natives). Initial European Neolithic does not have a single date but a range of more than four millennia”.

    Perhaps so. But we’re not actually talking ‘Neolithic’ here, especially not ‘European Neolithic’. So that takes care of the remainder of your comments. But back to some of your earlier ones:

    “Otherwise you're basically saying that E1b1b1a coalesced in Sardinia and that is extremely implausible”.

    Not impossible, but it requires substantial back migration of its related haplogroups up the Nile to Ethiopia and Kenya from the Mediterranean. E1b1b1a’s immediate ancestry appears to be in the Horn of Africa, or Ethiopia. Of its 6 ‘brother clans’ (E1b1b1b’’g) 3 are found in the Horn/Ethiopia, one in Tanzania and two in the Mediterranean. So presumably E1b1b1a simply accompanied his ‘sons’ and other relations into the Mediterranean.

    (continued)

    ReplyDelete
  34. “We have previously found Sardinian Y-DNA lineages deep in Africa”.

    That’s not what you wrote recently:

    http://forwhattheywereweare.blogspot.com/search/label/Y-DNA

    Quote:

    “E-V68* (E1b1b1a*) is found only in Sardinians”.

    The derived haplogroup E1b1b1a1-M78 is African. And have a look again at the map of E-M78’s distribution at your other blog. Three centres: The Horn, the Balkans and the Strait of Gibraltar. And we see much the same for E-M35. And E-M81 is in the western Mediterranean while E-M123 is in the east. And a quote from that blog:

    “I would think that this maritime migration proposed for these two paragroups, is highly suggestive of a flow into Iberia in the context of the Oranian (or Iberomaurusian) culture genesis in North Africa and the related changes (back-tip and wings in arrow heads notably) in the Iberian Gravetto-Solutrean complex, which influenced West Iberia up to Asturias but not for example the Basque area, which remained closer to the Aquitanian and Occitan Solutrean (a purer variant of this techno-culture, not influenced by Gravettian, which appears intrusive in any case, nor the Aterian-Dabban North African substrate)”.

    Surely if you’re going to give Y-hap R1b1a2 and its immediate descendants a ‘Gravettian’ entry into Europe it is no problem to assign the entry to Africa of it’s ‘father’s brother, R1b1c-V88, to something at least nearly as ancient. You even make the connection yourself, although for some reason you include E:

    “it is probable that the small presence of Y-DNA R1b and I in North Africa (which must be somewhat old if we have to judge from the Guanche mummies DNA) and the presence of Y-DNA E1b1 in Iberia (with ‘strange’ extension towards the NW specially) dates from this Paleolithic episode”

    I don’t see any reason to include E here. Y-hap R was entering Africa from outside whereas Y-hap E was already there, somewhere.

    “It could make for a great ‘Sea Peoples’ fantasy story though”

    I harbour a strong suspicion that any E1b1b1a1b-V13 today found along the Levantine coast came in a secondary migration, from Albania with the ‘Sea People’.

    ReplyDelete
  35. “Still I do not see a clear difference between a log and a long boat”

    Really? Next time try thinking before you tap the computer keys. Why would people bother to go to the trouble of hollowing the log out?

    “I do think that human intervention was decisive in the extinction of dwarf hippopotamuses”

    I'm certainly glad to have persuaded you that the hippos and elephants, along with myotragus and various species of giant dormice and dwarf deer, died out in the Mediterranean islands as a result of human intervention.

    “but you must claim that the hippopotamuses died of natural causes if you are to deny Mesolithic human colonization of Cyprus - and I doubt you can do that without a major logical short-circuit happening in your brain”.

    You obviously don’t bother to read what I write in my comments. I’ve known for years that the extinctions on the Mediterranean islands occurred some time before the Neolithic reached them. I recently wrote:

    “I agree that E's expansion is probably not strictly 'Neolithic', but it is almost certainly no earlier than Holocene”.

    So you see: we agree.

    “people lived in the Mediterranean island (and hence in nearby Sardinia surely too, only separated by a narrow strait) at least 1500 years before the arrival of Neolithic and high-seas navigation to the region (c. 8500 BP vs. c. 7000 BP).”

    So at last we agree that people reached the islands no earlier than around 8,500 years ago and caused the extinctions on them. We just don’t agree yet that these people carried Y-hap E.

    “Yes, that's a lot earlier than Neolithic, which only arrived in the late 6th or early 5th millennium BCE. Some 1500 years earlier … Further to the west Neolithic would take still many more millennia to arrive (probably because a barrier of humidity and oceanic cloud cover, as well as disinterest by the natives). Initial European Neolithic does not have a single date but a range of more than four millennia”.

    Perhaps so. But we’re not actually talking ‘Neolithic’ here, especially not ‘European Neolithic’. So that takes care of the remainder of your comments. But back to some of your earlier ones:

    “Otherwise you're basically saying that E1b1b1a coalesced in Sardinia and that is extremely implausible”.

    Not impossible, but it requires substantial back migration of its related haplogroups up the Nile to Ethiopia and Kenya from the Mediterranean. E1b1b1a’s immediate ancestry appears to be in the Horn of Africa, or Ethiopia. Of its 6 ‘brother clans’ (E1b1b1b’’g) 3 are found in the Horn/Ethiopia, one in Tanzania and two in the Mediterranean. So presumably E1b1b1a simply accompanied his ‘sons’ and other relations into the Mediterranean.

    ReplyDelete
  36. I was thinking of R1b1a when I wrote “We have previously found Sardinian Y-DNA lineages deep in Africa”.

    "Surely if you’re going to give Y-hap R1b1a2 and its immediate descendants a ‘Gravettian’ entry into Europe it is no problem to assign the entry to Africa of it’s ‘father’s brother, R1b1c-V88, to something at least nearly as ancient".

    No it's no problem to me. Not at all.

    "I don’t see any reason to include E here. Y-hap R was entering Africa from outside whereas Y-hap E was already there, somewhere".

    First I was talking of R1b1b1a2, not R1b1a, when I said "R1b" in that context.

    Hence E (E1b1b1b1) is mentioned as migrating into Iberia in the same bidirectional process at the genesis of Oranian.

    "I harbour a strong suspicion that any E1b1b1a1b-V13 today found along the Levantine coast came in a secondary migration, from Albania with the ‘Sea People’".

    It'd be from Greece if anything (most Sea Peoples were Greek probably) but there is also a Cardium Pottery link with Lebanon (from the Adriatic Sea TO Lebanon, not vice versa as misrepresented in some Wikipedia map) and there are other options IF there must be a flow of this haplogroup from the Balcans to the Levant.

    BUT on what specific data, if any, do you found this speculation? Is there any reason why Levantine E1b1b1a1 should be derived from Balcanic one and not vice versa (or some third option)?

    ReplyDelete
  37. "Hence E (E1b1b1b1) is mentioned as migrating into Iberia in the same bidirectional process at the genesis of Oranian".

    On what grounds?

    "most Sea Peoples were Greek probably"

    I actually suspect a wider source population actually. They seem to be a product of the breakdown of a central authority, probably Crete. So they could have come from anywhere (or everywhere) within the Cretan sphere of influence. Greek certainly, but possibly also from further afield.

    "there is also a Cardium Pottery link with Lebanon (from the Adriatic Sea TO Lebanon, not vice versa as misrepresented in some Wikipedia map)"

    I'd associate that much more with Y-hap J rather than E.

    "Is there any reason why Levantine E1b1b1a1 should be derived from Balcanic one and not vice versa (or some third option)?"

    I was considering specifically E1b1b1a1b-V13 as being from the Balkans, as you would know if you'd actually read what I wrote:

    "I harbour a strong suspicion that any E1b1b1a1b-V13 today found along the Levantine coast came in a secondary migration, from Albania with the ‘Sea People’".

    That haplogroup seems to be centred on the coast of Albania, where the Balkan Peninsular is closest to Southern Italy. Presumably that is not a coincidence. E1b1b1a1-V68 is a step further back, and its downstream haplogroups are spread from Kenya north to the Mediterranean and on into Italy and Spain. Again, you would know that if you'd read what I wrote.

    ReplyDelete
  38. Sea Peoples are not any clear phenomenon but in any case they are surely not the product of any authority breakdown but that of authoritarian imperialism.

    A most clear case of Sea Peoples in action we find at Troy. And, thanks to Homer, we know A LOT about that war. Those Sea Peoples were not ravaging bands of pirates, who would be unable to besiege and defeat a strong city-state like Troy or Ugarit (or Crete or whatever) but they were organized armies under princes of their own states confederated in ample coalitions for plunder. We know that for a fact.

    Similarly Vikings were not the product of any catastrophe but of socially organized war drive, sometimes looser and sometimes more structured. You see that also with Celts and other warmongering peoples (typically but not only Indoeuropean).

    It's not rag tag armies which conquer but solid military and political organizations normally. When the Vikings set siege to Paris in 885 they had a huge army with them, not just a bunch of ill-equipped pirates. They might be "underdeveloped" by comparison to their more "civilized" rivals (be it Rome, Troy, Ugarit, Egypt or the Frankish Empire) but they are no mere bunch of ill-equipped marooned slaves or revolting peasants: these are well organized armies for the standards of the age, led by princes of major cities or states like Mycenae or Denmark and at least partly manned by highly specialized aristocratic warriors.

    "I'd associate [Cardium Pottery] that much more with Y-hap J rather than E".

    What J and what E? J2 was surely involved in that one but there were other haplogroups: E1b1b1a1 for sure, G2a probably too and I2a with some likelihood as well and maybe even some R1b1b(xR1b1b2a1). But other E or J were not implied, except maybe in very localized flows.

    You oversimplify and that is of no help.

    "That haplogroup seems to be centred on the coast of Albania"...

    Precisely why it must have spread with Cardium Pottery. Because Coastal Albania, along with other Adriatic and Ionian Sea regions was the origin of Cardium Pottery culture.

    Poor old forgotten and mistreated Albania! It's always Greece or other larger neighbors who take all the credit.

    ReplyDelete
  39. "J2 was surely involved in that one but there were other haplogroups: E1b1b1a1 for sure, G2a probably too and I2a with some likelihood as well and maybe even some R1b1b(xR1b1b2a1)".

    I agree with most of the haplogroups you list.

    "Precisely why it must have spread with Cardium Pottery. Because Coastal Albania, along with other Adriatic and Ionian Sea regions was the origin of Cardium Pottery culture".

    I was sure you've been insisting that E1b1b1a1 is ancient in Northwest Africa. So how come you now have it moving west with Cardial pottery? And I'm sure we had agreed that E1b1b1a1's movement west through the Mediterranean was pre-Neolithic.

    ReplyDelete
  40. "I was sure you've been insisting that E1b1b1a1 is ancient in Northwest Africa".

    E1b1b1b1 (M81) is the one I say.

    E1b1ba1 and J1 would have arrived, in my theory, with Capsian culture (and proto-Berber language) in the late UP: older than European E1b1b1a1 in any case but not "ancient", at least not as ancient as E1b1b1b1.

    The Cardium Pottery and other European Neolithic E sublineage would be only (as far as I can tell) E1b1b1a1b-V13, only a fraction of E1b1b1a1 (but the most important fraction of all E in Europe).

    ReplyDelete
  41. "E1b1b1b1 (M81) is the one I say".

    I agree completely that E1b1b1b1 is Northwest African. But it's almost certainly a Holocene arrival in Northwest Africa. It derives from E1b1b1b which is part of the E1b1b1 clade. E1b1b1 has its greatest diversity in the Horn/Ethiopia.

    "E1b1ba1 and J1 would have arrived, in my theory, with Capsian culture (and proto-Berber language) in the late UP"

    By boat. Through the Mediterranean. And more like Late Holocene rather than specifically late UP.

    "The Cardium Pottery and other European Neolithic E sublineage would be only (as far as I can tell) E1b1b1a1b-V13"

    Quite likely. And in the case of E1b1b1a1b-V13 from Albania. But its ancestry, in the form of E1b1b1b1 was almost certainly from the Bab all Mandab. Along with the remaindr of the E1b1b1 haplogroups.

    ReplyDelete
  42. "I agree completely that E1b1b1b1 is Northwest African. But it's almost certainly a Holocene arrival in Northwest Africa. It derives from E1b1b1b"...

    Which is also a NW African clade.

    The "Holocene" conjecture can't be defended, IMO, because the lineage is too consolidated as geographically distinct.

    As I said before (the parallel debate here), you do not see such radical founder effects anywhere on Earth where a known colonization has happened: not in the USA, not in Bantu Africa. You see increase or decrease of immigrant clades but not any such radical founder effect - not at all. When you find isolated clades not present in the immigrants' homeland, you typically think that they are native, not miraculous arrivals from outer space.

    "E1b1ba1 and J1 would have arrived, in my theory, with Capsian culture (and proto-Berber language) in the late UP"

    By boat. Through the Mediterranean. And more like Late Holocene rather than specifically late UP
    .

    I refuse to even discuss this knee-jerk claim, which would be laughable would not it be too much to keep faith in this debate going anywhere other than a maelstrom of nonsense.

    If you know nothing about Capsian culture or North African Prehistory, go learn.

    "And in the case of E1b1b1a1b-V13 from Albania. But its ancestry, in the form of E1b1b1b1"...

    Uh? I imagine that you mean E1b1b1.

    "... was almost certainly from the Bab all Mandab".

    More like the Nile. Mind you that while Somalis have extremely high frequencies of certain clades (E1b1b1a1a) they have, like Yemenis across the strait (J1), very low diversity. Both cases are probably Holocene founder effects - and they are good examples of what we can expect with "recent" (Neolithic??) founder effects.

    ReplyDelete
  43. "'It derives from E1b1b1b'...
    Which is also a NW African clade".

    Where did you get that idea from?

    http://en.wikipedia.org/wiki/Haplogroup_E1b1b_(Y-DNA)#E1b1b1b_.28E-L19.2FV257.29

    Quote:

    "They found 6 'E-V257*' individuals in their samples who were E-V257, but not E-M81. A Borana from Kenya, a Marrakesh Berber, a Corsican, a Sardinian, a southern Spaniard and a Cantabrian".

    A single 'Marrakesh Berber' makes the haplogroup 'Northwest African'?

    "The 'Holocene' conjecture can't be defended, IMO, because the lineage is too consolidated as geographically distinct".

    Really? Don't you mean 'The Holocene conjecture can't be defended because I refuse to accept it'? And if you take the trouble to compare E1b1b1b with the other E1b1b1 haplogroups you will see that none of them come close to Northwest Africa. So you obviously believe that E1b1b1b has no ancestors. It sprang up out of nothing.

    "you do not see such radical founder effects anywhere on Earth where a known colonization has happened"

    Polynesia springs instantly to mind. America is not a bad example either. Northern Eurasia fits fairly well.

    " You see increase or decrease of immigrant clades"

    And I presume that's what we're seeing in Northwest Africa. Obviously the examples above are all of groups moving into previously uninhabited regions, but perhaps Northwest Afica was sparsely settled before E1b1b1b got there.

    "Uh? I imagine that you mean E1b1b1".

    No. I don't I mean what I said.

    "More like the Nile".

    If you'd read what I said a few days ago you would realise that is exactly what I suggested.

    "Somalis have extremely high frequencies of certain clades (E1b1b1a1a)"

    And they, along with Ethiopians and Egyptians, have E1b1b1a1c, E1b1b1g, E1b1b1f, E1b1b1d and E1b1b1c. So E1b1b1 is obviously a Northwest African clade (a joke, in case you don't get it).

    "they have, like Yemenis across the strait (J1), very low diversity".

    What? You call that 'very low diversity?

    "Both cases are probably Holocene founder effects - and they are good examples of what we can expect with 'recent' (Neolithic??) founder effects".

    You've certainly got it the wrong way round.

    ReplyDelete
  44. A single 'Marrakesh Berber' makes the haplogroup 'Northwest African'?

    Plus 4/6 SW Europeans. Not a single individual from the Eastern Mediterranean (nope: Kenya is much farther south). All (but tha single Kenyan) of the E1b1b1b* is in an arch between Morocco and Sardinia (notice: Sardinia, not mainland Italy!).

    And then there is the other half of the matter (I imagine E1b1b1b* is a single subhaplogroup, excepting maybe the Kenyan?), which is E1b1b1b1, which is clearly of NW African origin.

    So I am all for a NW African origin. And only a SW European origin could compete (in mere theory: the clade is way too rare in this part of the World, IMO and has an issue with immigration routes that NW Africa does not have).

    We need better sampling in Africa and less pointless arguments.

    "... if you take the trouble to compare E1b1b1b with the other E1b1b1 haplogroups you will see that none of them come close to Northwest Africa"...

    Sudan, Egypt and Palestine, where E1b1b1a and E1b1b1c concentrate are close to NW Africa. The other clades are not that far, just upstream the Nile a bit.

    Whatever the case NW Africa is still closer to all them than SW Europe, which is the only possible alternative. There's no reasonable way to reach Iberia or Sardinia from the Nile area without going through NW Africa, at least Tunisia.

    "So you obviously believe that E1b1b1b has no ancestors. It sprang up out of nothing".

    LOL

    It sprang out of whatever. But you're arguing like if the origin of Y-DNA I should be measured by that of its ancestors IJ, IJK and F. It has to be measured by the internal parameters of I first of all. And then maybe a correction based on where other related clades may have originated (i.e. where the shared ancestors may have lived) may be interesting to add - but it is not the quid of the question at all, at most a small correction to be introduced for even greater plausibility when looking at the big picture.

    The key is basal diversity and AFAIK the basal diversity of Y-DNA E1b1b1b is so clearly concentrated in the Western Mediterranean areas that I cannot believe we are wasting our time having this argument.

    "Polynesia springs instantly to mind. America is not a bad example either. Northern Eurasia fits fairly well".

    They all are colonizations of virgin lands, which just confirms my point because I was talking of colonization of already occupied lands. (Try to understand before you argue back, ok?)

    "Obviously the examples above are all of groups moving into previously uninhabited regions, but perhaps Northwest Afica was sparsely settled before E1b1b1b got there".

    Exactly. In the Aterian period or whatever but before the Oranian one in any case. The persistence of a vast majority of mtDNA lineages that date from the Oranian period or earlier (everything but K, JT, X, W and maybe some L(xM,N) clades) attests that when Capsian peoples arrived, they had to face a consolidated previous occupation, which is still extremely influential in the modern genetic makeup of the area.

    ...

    ReplyDelete
  45. ...

    "And they, along with Ethiopians and Egyptians, have E1b1b1a1c, E1b1b1g, E1b1b1f, E1b1b1d and E1b1b1c. So E1b1b1 is obviously a Northwest African clade (a joke, in case you don't get it)".

    Nobody is saying that E1b1b1 is a NW African clade, I'm always talking of E1b1b1b and E1b1b1b1.

    Somalis anyhow do not have all that diversity and Egypt/Sudan must be considered rather than Somalia in the process of scattering of the parent haplogroup E1b1b1, which, by all accounts, have a Nile area origin.

    "You call that 'very low diversity?"

    Please illustrate me about the "high diversity" of Somalis. I think I know something about the matter, as I have learned from educated Horners in the past on this and other issues, but I may be wrong.

    As far as I can tell, Somalis are known to have E1b1b1a1a (E-V12) in extremely high apportions (c. 70%), original from Nubia though. Their particular clade is E1b1b1a1a2 (E-V32), shared with some other Cushitic peoples. In addition, they have some E1b1b1d (E-V6) (permeated from Ethiopia).

    Wikipedia has an interesting footnote you may be interested in reading:

    Prior to Cruciani et al. (2007), Semino et al. (2004) had proposed the Horn of Africa as a possible place of origin of E-M78. This was because of the high frequency and diversity of E-M78 lineages in the region. For example, Sanchez et al. (2005) found that 77.6% of 201 male Somalis tested in Denmark were members of this clade. However, Cruciani et al. (2007) were able to study more data, including populations from North Africa who were not represented in the Semino et al. (2004) study, and found evidence that the E-M78 lineages in the Horn of Africa were relatively recent branches (see E1b1b1a1b (E-V32) below). They concluded that Northeast Africa was the likely place of origin of E-M78 based on "the peripheral geographic distribution of the most derived subhaplogroups with respect to northeastern Africa, as well as the results of quantitative analysis of UEP and microsatellite diversity".

    NE Africa for Cruciani is "Egypt and Libya".

    However the clades discovered by Trombetta this very year, as pointed out by Etyopis earlier, seem to slant the overall origin of the haplogroup E1b1b (M215) and its dominant subclade E1b1b1 (M35) either in Ethiopia or the neighboring area.

    But with so many subclades, it's likely that E1b1b1 will have some intermediate branches, yet to be discovered: some internal structure between the "father" and the "children". I imagine that this will be resolved in the following sense:

    1. E1b1b1 spread from near Ethiopia

    2. A subclade spread from near Nubia into the Mediterranean areas.

    We'll see but it's clear that the migration from the Upper Nile into NW Africa must have included some stop in between and typically this one has been Nubia, attending to what happened with Capsian at a later time and the high diversity of the Nubian area anyhow.

    ReplyDelete
  46. "Plus 4/6 SW Europeans. Not a single individual from the Eastern Mediterranean (nope: Kenya is much farther south)".

    But Kenya is certainly not NW African. So we're left with just 1/6 NW African, yet somehow you still see the clade as 'NW African. How?

    "(I imagine E1b1b1b* is a single subhaplogroup, excepting maybe the Kenyan?)"

    There you go again. Making up things to suit your belief.

    "Sudan, Egypt and Palestine, where E1b1b1a and E1b1b1c concentrate are close to NW Africa".

    Apart from the Sudan' they're a heck of a lot closer to the Mediterranean than to NW Africa.

    "The other clades are not that far, just upstream the Nile a bit".

    Even closer to the Mediterranean, and connected to it by water. So that leaves just Sudan as being any distance from the Mediterranean. And it's basically connected by the Nile.

    "It has to be measured by the internal parameters of I first of all. And then maybe a correction based on where other related clades may have originated"

    That's exactly what I'm doing with Y-hap E1b1b1, and exactly what you're ignoring about the haplogroup.

    "Nobody is saying that E1b1b1 is a NW African clade, I'm always talking of E1b1b1b and E1b1b1b1".

    And completely ignoring E1b1b1c, E1b1b1d, E1b1b1e, E1b1b1f, E1b1b1g and especially E1b1b1a.

    "Please illustrate me about the 'high diversity' of Somalis".

    That list shows 'high diversity' in regions near the Horn at least. Three of them are found in Ethiopia, so perhaps the diversity is centred in that region. As you admit:

    "Egypt/Sudan must be considered rather than Somalia in the process of scattering of the parent haplogroup E1b1b1, which, by all accounts, have a Nile area origin".

    So you're beginning to agree.

    "I imagine that this will be resolved in the following sense: 1. E1b1b1 spread from near Ethiopia 2. A subclade spread from near Nubia into the Mediterranean areas".

    We draw closer by the minute. Except several subclades 'spread from near Nubia into the Mediterranean areas', via the Nile.

    "found evidence that the E-M78 lineages in the Horn of Africa were relatively recent branches"

    Late Holocene? At that time E-V68 remained undiscovered so the comments would likely also apply to the other E1b1b1-M34 clades.

    ReplyDelete
  47. "But Kenya"...

    One single individual! An erratic unless proven otherwise.

    "So we're left with just 1/6 NW African, yet somehow you still see the clade as 'NW African. How?"

    Because otherwise I'd be left only the choice to say: it is SW European.

    Actually I am going to become the Devil's Advocate and I'm going to defend this point until you beg for a more natural NW African origin and acknowledge that we need more African samples before we can fully clarify the matter.

    So from now own (Devil's advocate) E1b1b1b* is from Valencia and E1b1b1b is from Marakesh. The halfway point is in Málaga. So as of now E1b1b1b is original from Málaga.

    Have fun with that slight twist to a NW African origin. Oops Málaga is just near Ceuta... maybe we are going a bit too precise in this calculation? I'd say so but as I'm working for the Devil right now, I will not and claim that E1b1b1b originated exactly in the same house of Marbella as Osama Bin Laden spent his wealthy Arabian childhood summers, and where Hannibal was once visited by the Goddess Astarte herself. Exactly there: nobody knows yet but you, me, the Devil, Astarte and our select and amused public.

    "E1b1b1a and E1b1b1c..."

    Not the matter of discussion. I also have cousins in Brazil, so?

    "And completely ignoring E1b1b1c, E1b1b1d, E1b1b1e, E1b1b1f, E1b1b1g and especially E1b1b1a".

    And I will keep doing so. These haplogroups inform us about their "father" E1b1b1 but say nothing about their "brother" E1b1b1b.

    ReplyDelete
  48. "One single individual! An erratic unless proven otherwise".

    The same holds true for the single Marrakesh Berber. An eratic unless proven otherwise.

    "So from now own (Devil's advocate) E1b1b1b* is from Valencia and E1b1b1b is from Marakesh".

    I suppose that is actually possible. But you have to get E1b1b1 from the Upper Nile to Valencia. That is where we differ. You have it moving through the arid reaches of the Libyan desert, while I have it moving through the much more manageable Mediterranean.

    "And I will keep doing so. These haplogroups inform us about their 'father' E1b1b1 but say nothing about their 'brother' E1b1b1b".

    They do really:

    http://en.wikipedia.org/wiki/Haplogroup_E1b1b_(Y-DNA)

    From the link:

    "All major sub-branches of E1b1b1 are thought to have originated in the same general area as the parent clade"

    In other words, somewhere round Ethiopia.

    ReplyDelete
  49. I can't understand how you can be so absolutely sure that ancient humans emerged from Africa and traveled along the Arabian coast in boats yet at the same time be so absolutely sure that a group of recent humans was not able to emerge into the Mediterranean from the Nile and travel along the African coast in boats. I just don't get it.

    ReplyDelete
  50. "All major sub-branches of E1b1b1 are thought to have originated in the same general area as the parent clade"

    In other words, somewhere round Ethiopia"
    .

    Ah cheater! The original Wiki-text, which you quoted in the parallel discussion (here) said:

    "All major sub-branches of E1b1b1 are thought to have originated in the same general area as the parent clade: in North Africa"...

    Stop playing games that go nowhere. You have decided the conclusion a priori but even your own data and selective quotations contradict it.

    "I can't understand how you can be so absolutely sure that ancient humans emerged from Africa and traveled along the Arabian coast in boats yet at the same time be so absolutely sure that a group of recent humans was not able to emerge into the Mediterranean from the Nile and travel along the African coast in boats".

    I don't say they were not able, what I say is that you see the evidence nowhere. Only at Gibraltar Strait (Alboran Sea if you wish) we do see clear and rather strong indications of pre-Neolithic exchange between the two shores (before Neolithic times).

    I do not describe these issues in terms of possibility (ultimately prejudice because you cannot predict what is possible at any moment) but in terms of factual evidence (what really we see happening in the past based on the data).

    If a river in some place was a barrier, because the data says so, then I will accept it was (can't recall a single case but just an example), if the evidence would say that Hymalayas or the Atlantic Ocean were crossed in Paleolithic times, I'd have to bow to those facts - even if I do consider both feats practical impossibilities for that period.

    ReplyDelete
  51. "All major sub-branches of E1b1b1 are thought to have originated in the same general area as the parent clade: in North Africa"...

    Again you leave out the part that is inconvenient for your belief. What they actually said was:

    "All major sub-branches of E1b1b1 are thought to have originated in the same general area as the parent clade: in North Africa, East Africa, or nearby areas of the Near East".

    That leaves a fairly wide region, of which 'North Africa' is just one possibility, whatever they mean by that region. Possibly 'eastern North Africa' seeing that the other two regions they suggest are 'East Africa, or nearby areas of the Near East'. And if you're going to accept the link as correct you have this problem:

    "E1b1b1b1 (E-M81), formerly E1b1b1b, E3b1b, and E3b2, is the most common Y chromosome haplogroup in the Maghreb, dominated by its sub-clade E-M183. It is thought to have originated in the area of North Africa 5,600 years ago".

    That is hardly 'Paleolithic'.

    "You have decided the conclusion a priori but even your own data and selective quotations contradict it".

    Excuse me. I think we have very much established that it is you who have 'decided the conclusion a priori' even when the data you quote contradicts that conclusion.

    "I don't say they were not able, what I say is that you see the evidence nowhere".

    Wrong. We see it everywhere. Wiki again:

    "The E1b1b clade is presently found in various forms in the Horn of Africa, North Africa, parts of Eastern, Western, and Southern Africa, West Asia, and Europe (especially the Mediterranean and the Balkans)".

    Note: 'especially the Mediterranean and the Balkans'. And further:

    "E1b1b and E1b1b1 are quite common amongst Afro-Asiatic speakers. The linguistic group and carriers of E1b1b1 lineage have a high probability to have arised and dispersed together from the region of origin of this language family".

    Could you remind us, where is 'the region of origin of this language family'.

    "Only at Gibraltar Strait (Alboran Sea if you wish) we do see clear and rather strong indications of pre-Neolithic exchange between the two shores (before Neolithic times)".

    We see 'rather strong indications of pre-Neolithic' arrival on Corsica and Sardinia, both of which have E1b1b1 haplogroups, and in Cyprus, which perhaps has some Y-hap E1b1b1 clades.

    "in terms of factual evidence (what really we see happening in the past based on the data)".

    And the 'factual evidence' shows that Y-haps E1b1b1a and E1b1b1b arrived on Sardinia very early during the expansion of those two haplogroups. So we can date the expansion fairly accurately.

    "If a river in some place was a barrier, because the data says so, then I will accept it was"

    It looks, from the data, that neither the Nile river nor the Mediterranean were barriers during E1b1b1's expansion. On the contrary, they look as though they were pathways.

    ReplyDelete
  52. "Again you leave out the part that is inconvenient for your belief".

    LOL, that was you. Re-read, please!

    I was quoting you to the letter there. I left nothing out of what you so selectively quoted. And now you accuse me of altering the quote!

    Stop playing games of I will have to renounce to debate with you.

    Either you are straightforward and say the truth or you are not worth debating with. Your choice.

    "That leaves a fairly wide region, of which 'North Africa' is just one possibility"...

    The possibility that matches perfectly E1b1b1b1 (and therefore E1b1b1b as a whole, with all other rare E1b1b1b* lineages found also in or beyond NW Africa).

    "And if you're going to accept the link as correct"...

    I am never and I will never accept "a link as correct" as a whole. Each piece of info has its own analysis to be made about, just because, say, Plato got one thing right once doesn't mean he was right in all he said.

    "That is hardly 'Paleolithic'".

    That's a random Wiki-quote. Pointless even if properly documented (as I do not accept any validity for the "molecular clock" idiocy).

    Then I ask you for E1b1b1b evidence and you come with ""The E1b1b clade"...

    No! I'm not accepting all that mish-mash you do, where the end of the road is the same as the beginning and whatever else.

    I'm fucking tired of this confusionism you are promoting and the waste of time that debating with you is becoming, seriously.

    Afroasiatic languages: I associate the expansion of Afroasiatic languages (possibly without Omotic, which may be an older relative) with E1b1b1a1 and J1. The origin would be around Nubia or otherwise in Sudan/Egypt. Without Omotic the center of gravity of Afroasiatic moves northward quite a bit.

    "And the 'factual evidence' shows that Y-haps E1b1b1a and E1b1b1b arrived on Sardinia very early during the expansion of those two haplogroups".

    No, there's no "factual evidence" saying that. Nothing at all: just two individuals in hyper-sampled populations, surely arrived from Iberia or SE France or even directly North Africa...

    "It looks, from the data, that neither the Nile river nor the Mediterranean were barriers during E1b1b1's expansion"...

    The Mediterranean it was: we do not see the same haplogroups within E1b1b1 homogeneously spread at both sides: even if some of the derived lineages crossed it at narrow points, their frequencies north of the Sea are a lot less than south of it. E1b1b1 (all clades) is like 75% in North Africa, while only 10% in those areas of Europe most influenced, like West Iberia.

    Remember that in West Iberia we see quite a bit of E1b1b1b1 but very little of other E sublineages and almost no J1, which is invariably associated to E1b1b1a1 in North Africa (specially). This indicates that E1b1b1b1 is older in NW Africa than all the rest here considered (other E, J1).

    ReplyDelete
  53. "No, there's no 'factual evidence' saying that. Nothing at all: just two individuals in hyper-sampled populations, surely arrived from Iberia or SE France or even directly North Africa..."

    Only because you 'say so'!!! There are several (probably relevant) facts you are determined to ignore. You are determined to ignore the apparently star-like diversity of E1b1b1. You are determined to ignore the geographic distribution of its clades and subclades. You are determined to ignore the the fact that two basal E1b1b1 clades are found in Sardinia, one of them (E1b1b1a*) apparently exclusively so. You are determined to ignore the the fact that those haplogroups can only have arrived in Sardinia no earlier than around 10,000 years ago.

    You are determined to ignore the above because you have decided a priori that E1b1b1b is Upper Paleolithic in Northwest Africa. But you are unable to provide any actual evidence for this belief, just opinon. The latest evidence of Y-hap A's diversity in Africa actually allows us to see that Y-hap E as a whole may have carried the Oranian or the Capsian west into NW Africa. And Y-hap R1b1c was almost certainly involved in one or the other. But you are determined to ignore these possibilities because you have decided a priori that E remained behind in Africa while haplogroups C, D and F moved out, curiously leaving no descendants behind.

    "The Mediterranean it was: we do not see the same haplogroups within E1b1b1 homogeneously spread at both sides"

    Which probably means members of the E1b1b1 haplogroup hugged the coast, leaving haplotypes behind as it went.

    "some of the derived lineages crossed it at narrow points"

    Especially Sicily/Tunisia.

    "Remember that in West Iberia we see quite a bit of E1b1b1b1 but very little of other E sublineages"

    Surely that is simply a product of founder effect. And drift along the way.

    "and almost no J1, which is invariably associated to E1b1b1a1 in North Africa (specially)".

    Which should tell you that J1 was not involved with E1b1b1b's voyaging. Instead it became associated with the secondary expansion of E1b1b1a1.

    "This indicates that E1b1b1b1 is older in NW Africa than all the rest here considered (other E, J1)".

    Yes. It was the only E1b1b1 haplogroup to originally make it that far.

    ReplyDelete
  54. "You are determined to ignore the apparently star-like diversity of E1b1b1".

    Nonsense, I'm totally fine with that, I agree with an Upper Nile origin for E1b1b1.

    What I do not agree is for such an origin for the next downstream steps in the E1b1b1b1 branch. Quit mixing up and down. Focus on M81.

    "... you have decided a priori that E1b1b1b is Upper Paleolithic in Northwest Africa. But you are unable to provide any actual evidence for this belief, just opinon".

    LOL. The evidence is there fore the taking: you know it well by now: the only alternative to NW Africa is SW Europe!

    And you agree it's not a valid option, so WTF?

    "Y-hap E as a whole"....

    ... is trivial for the debate on E1b1b1b1 and E1b1b1b, the same that Y-DNA F as a whole is trivial for I2a: I2a did not coalesce in India just because F probably did.

    I think you are still able to understand that, right? Or am I wasting my time?

    "And Y-hap R1b1c"...

    Oh, no, ISOGG has revolted around all R1b nomenclature just because of FTDNA rantings! This is clearly the end of the world! (Sucks!)

    I doubt that R1b1c was involved in E1b1b1b moves: R1b1c is a major haplogroup in Sardinia and seems to come from Italy, E1b1b1b* is an erratic and seems to come from Spain, via Corsica probably. These are two different processes.

    "Which should tell you that J1 was not involved with E1b1b1b's voyaging. Instead it became associated with the secondary expansion of E1b1b1a1".

    Bravo! I'm glad you realize that (it was exactly my point).

    "Especially Sicily/Tunisia".

    I know of no such relations: you are making them up as you go. Besides "Sicily/Tunisia" is not any "narrow crossing": that's open sea.

    I'm not taking you seriously unless you begin taking seriously yourself.

    ReplyDelete
  55. "LOL. The evidence is there fore the taking: you know it well by now: the only alternative to NW Africa is SW Europe!"

    Rubbish.

    ReplyDelete

Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).