maandag 4 juli 2016

Why Basal Eurasian is Still African as of Lazaridis et al 2016

"[The Shuqbah Natufian] type may be described as Mediterranean, but with a distinct bias towards the African variety of that stock represented by the predynastic people of Egypt. [...] The later cave dwellers of Shukbah practiced a rite which is still observed by many negro tribes of Africa. They removed one or both upper central incisors in youth, which resulted in atrophy of the corresponding alveolar part of the upper jaw and in an upgrowth of the unopposed lower incisors." 
Keith 1931 pp. 210-11

The excerpt above (Keith 1931) describes a Natufian sample excavated in the Shuqbah cave's Late Natufian deposits. In comparison to other Natufian sites, the Natufian inhabitants of this cave are known to be particularly African in their morphometric affinities, as shown by the descriptions of various physical anthropologists. The Natufian skull in the image above—complete with the North African practice of upper incisor extraction (Stojanowski et al 2014) and what seems to be a blurred nasal margin—differs substantially from preceding groups in the Levant. Among other things, its cranial length (189 mm) and breadth (128 mm) represent a huge break from the short and broad head trend observed in the preceding Levantine skeletal samples (Stock et al 2005Hershkovitz et al 2005) and are consistent with both North and Sub-Saharan groups.

In my previous post on Basal Eurasian a year ago I made the case that Basal Eurasian is thoroughly admixed with African ancestry. Today, I reject this mixture hypothesis and see it as entirely African (i.e. the modern day African samples with Basal Eurasian are admixed, not the original component itself). After Lazaridis et al 2016's preprint came out, many claimed that the presence of Basal Eurasian in Mesolithic Iran disproves its African affinity. Let's look at what the actual evidence says.

Why my Previous Post Pissed off Some 'Afrocentrics' and Eurocentrics.

A year ago I pissed off some Afrocentrics and Eurocentrics by saying that Basal Eurasian is African. I pissed off some Afrocentrics because many of them can't fathom the idea that indigenous Africans can be distantly related to 'black Africans'. This is why you get confused attempts in some quarters to make Berber and pre-proto-Semitic Eurasian languages. The appearances of people who speak these language today generally don't fit the so-called 'black African' notion and are, at times, deemed a threat to vested interests. To some Afrocentrics, the way most speakers of these tongues look today can be deeply distressing, because their languages are closely related to ancient Egyptian. Not satisfied with ancient Egyptians being African (they have to specifically be 'Sub-Saharan African'), Berber and Semitic identities are sometimes downplayed, ignored and conveniently relegated to 'extra-Africa' in what are supposed to be scientific publications. It is no surprise, then, that these passive aggressive feelings resurface in some quarters when Basal Eurasian reinforces those 'undesirable' linguistic and ethnic connections to ancient Egyptians.

Then there are the Eurocentrics. My post has pissed some of them off as well because the whole notion of Basal Eurasian forces them to rethink their ideas about population affinity. One of Basal Eurasian's implications is that the racial label 'Caucasian' as used by many commentators doesn't just cover West Eurasians. It infringes on indigenous African genotypes and phenotypes. Some Eurocentric commentators marginalize qualms with the label 'Caucasoid'. It is argued that this label only runs into problems in the works of politically correct Africanists and Afrocentrics, because they are in denial. The fact that this label has also run into problems in other parts of the world disproves this sentiment. As shown by the Kennewick Man controversy, the label 'Caucasoid' also infringes on the indigenous phenotypes found in the Pacific. Initially classified as 'Caucasoid', we now have Kennewick Man's genome and he seems no more closer to 'Caucasians' than his immediate ancestors

It's not a coincidence that academics have such a hard time pinning down the genetic and morphological hallmarks of West Eurasians without inadvertently infringing on populations who don't fit for various reasons. Skeletally, what many call 'Caucasoid' are morphological tendencies that can rise and fall in frequency in many distinct human lineages independent of admixture. This is because the first mtDNA M and N carriers were already evolved in this morphological direction, to some extent. (We can see this, for instance, in the general appearance of Ksar Hill I. This fossil is among the oldest we have of Upper Palaeolithic colonists of West Eurasia). And if some of the first mtDNA M and N carriers already show precocious morphological tendencies in this so-called 'Caucasian' direction, the first L3 people must have shared those variations as well. Based on evolutionary principles, one would expect those variations to decrease as we look deeper in the history of Ksar Hill's ancestors. For instance, we might imagine that the mtDNA L3 people had such precocious tendencies more than the L3'4 people, who might have had them more than the L3'4'6 people, and so on.

Ksar Akil I Juvenile (Bergman and Stringer 1989) is one example in a long line of fossils that refute the fallacy that 'Caucasian' features = European, or even West Eurasian.

These apomorphic variations predate 'Caucasians' and explain why some Africans are phenotypically and genetically intermediate. Note that I'm not saying that Eurasian admixture here and there doesn't contribute to this intermediate position of some African populations. What I'm saying is that if you strip those 'intermediate' Africans of their Eurasian contributions, you might get a genetic profile similar to, say, Mota (who is intermediate), as opposed to one that you'd expect to find more likely in equatorial inner Africa. In other words, a Berber speaker stripped of all Eurasian ancestry would still look somewhat like Berbers, albeit with darker skin on par with equatorial Africans. So, the sentiment out there that living North Africans would necessarily blend in with a crowd of 'black Africans' (as lay people often put it) when stripped of their Eurasian ancestry is a myth. The same applies to East Africa. Mota's lack of Eurasian ancestry doesn't stop him from being genetically (and likely also morphologically) intermediate. In this article I will make the case that 'Basal Eurasian' is a later departure from the Y DNA CT / mtDNA L3'4'6 people than Mota. Moreover, I will make the case that they departed from the L3'4'6 stem before the M and N people did.

Why Basal Eurasian Can't Represent pre-Toba OOA Migrants Like Skhul and Qafzeh

Lazaridis et al seem to mostly agree with this assessment about Basal Eurasian's phylogenetic positioning. In fact, Their models bear this out as they have Basal Eurasian sandwiched in between Mota and individual Eurasians fossils like Ust-'Ishim. Unfortunately the authors don't carry their own models fully and unconditionally to their logical conclusion when it comes to the identity of Basal Eurasian. While they agree that this sandwiched position is consistent with Basal Eurasian being African, they maintain that it's also consistent with earlier, pre-Toba, OOA migrants. In other words, they're open to the possibility that Basal Eurasian represents Middle Palaeolithic OOA migrants in the Middle East. This pre-Toba migration is thought to be distinct from the post-Toba OOA migration that living Eurasians inherit most of their DNA from.

But here is the thing. Lazaridis et al identify Ust'-Ishim and his immediate ancestors with the post-Toba OOA event. But Ust'-Ishim is only 45ky old. This means that Ust'-Ishim's population can hardly be cognate with Middle Palaeolithic OOA populations who seem to have split off from living humans at least 120kya. And even this 120kya date is a conservative estimate. These humans are already found in North Africa and adjacent regions (Levant, Arabia) at a time when the populations we consider to be our most distant living relatives (e.g. Khoisan and Pygmies) all seem to have split off closer to southern Africa than North Africa and the Levant. Since our ancestors couldn't have been in two places at once 120kya, the fact that these living hunter gatherer relatives live on one end of the continent (south/central) whereas Aterians and Nubian Complex people live on the other side of the continent (north) indicates that our ancestors would likely have been nowhere near these Middle Palaeolithic OOA humans around 120kya. This, in turn, indicates that pre-Toba OOA migrants aren't cognate with our human lineage until very early, likely before the mtDNA L0'6 node that contains all of us. This would make their human tree cognate with our human tree, as opposed to them being small branches on our mtDNA L0'6 tree.

But there is more. It's not just this sandwiched position that points to Basal Eurasian's African origin. Closer inspection reveals that all the basic properties that make Basal Eurasian what it is, undermine its identification with pre-Toba OOA migrants. For instance, Basal Eurasian carries little to no Neanderthal DNA. A remarkable finding under the scenario that Basal Eurasian represents pre-Toba OOA ancestry. Those familiar with the Multiregional Evolution (MRE) vs Recent Africa Origin (RAO) debate will remember that continuity (both archaeological and morphological) between archaic humans and pre-Toba humans in several regions was a cornerstone of MRE. For instance, Multiregionalists maintained that anatomically modern humans in various regions used tools that were used by preceding archaic humans. This was interpreted as evidence that regional anatomically modern humans 'evolved' more or less from regional archaic humans. Today, we know that's not true. However, the curious finding that some pre-Toba migrants abandoned their tools in favor of tools made by archaic humans has relevance here. Why? Because if Basal Eurasian represents pre-Toba people, one would not expect this much interaction between pre-Toba and archaic humans to result in a (near) lack of archaic human DNA in Basal Eurasian.

The possible discovery of  the source of pre-Toba OOA migrant aDNA; does it match Basal Eurasian?

I was recently alerted to a promising finding that looks like it sheds light on the relationship of Basal Eurasian to the population source of pre-Toba OOA migrants. You can find the abstract here. The gratuitous references to slave trade aside, and assuming that this aDNA is authentic and not a total dud, the abstract describes Moroccan aDNA that is unrelated to any modern sample. This is a curious finding, to put it mildly, because we have a pretty representative array of genomes. We might not have sampled all living populations yet, but filling in the minor gaps here and there is not going to be enough to explain this discrepancy at this point. Why? Because even the most genetically distant living humans can be modeled with 'inappropriate' samples if crisp precision is not a priority. However, from the abstract it looks like the enigmatic aspects of these genomes can't be modeled meaningfully with modern references. The one result they do report should be looked at with a grain of salt and seems to be driven in part by the authors' conformation bias of invoking slavery where it doesn't need to be. The fact that living North Africans, who have ancestry from both sides of the Mediterranean, are distantly related to these genomes proves these cave-dwellers can't be a simple mixture of groups from both sides of the Mediterranean. After all, living North Africans themselves are a mix involving these same sources of ancestry and they're poor candidates.

The archaeological record of North Africa is unique in the region as far as relic Mousterian-like1 populations surviving late into the Late Palaeolithic (Smith 1966Vermeersch 2009). Even in Morocco, as of yet unknown MSA industries appear in the Late Palaeolithic (Barton et al 2013). These potential links with this aDNA are wailing to be explored and make fixation on slavery a costly and missed opportunity. But that doesn't mean we as bloggers and readers can't hypothesize about connectable dots, while remaining careful about overcommitting to small shreds of information from abstracts. If aspects of these cave-dwellers' genomes do, in fact, turn out to lack continuity with living populations, this can only mean that this DNA comes from populations who are 'older' than living populations. (Older, in the sense of separating earlier from the AMH tree). This impression seems to be validated strongly by the "South African" affinities mentioned in the abstract. Humans older than the Khoisan might harbor some ancestry that's reminiscent of Khoisan due to fact that they both split off from the mtDNA tree rather early, and, therefore, are expected to lack the derived mutations that accumulated in the L1'6 people. In terms of genetic distance, such a separation-based relationship with similar distances to L1'6 people could express itself as an affinity relationship, even though they're not meaningfully related and have accumulated derived mutations in their own right. One paper that shows such a result is Tishkoff et al 2009, where Pygmies and Khoisan share a 'hunter gatherer' component. This component can't possibly be taken as a sign of close relationships. Instead, it's likely a reflection of the similar timing of their split from the human tree.

One thing to keep in mind is that more answers in regards to these cave dwellers' affinities won't necessarily prove that their ancestors are (related to) the authors of the North African 'Mousterian'. In order to justify the connection of these genomes with the North African 'Mousterians', transparent analyses have to first confirm that 1) parts of these genomes occupy an outgroup position relative to both L0 and L0'6 people, 2) frequency and diversity of this ancestry decrease with distance from North Africa and 3) that the haplotype blocks that make up this ancestry have the right LD properties. Because skeletal remains of modern humans associated with the North African and Levantine 'Mousterian' so far seem to form a morphological clade, the results—whatever they turn out to have in store—should be generalizable to modern human makers of the 'Mousterian' in general. If they're not generalizable in terms of proving that these people belong to the same genetic meta population, then at least we should be able to generalize that they all share their 'olderness' relative to L0'6 people. The reason for this is that their suite of morphological characteristics clearly 1) make them morphometrically intermediate between L1'6 modern humans and archaic humans and 2) tend to put them on a separate branch from L0'6 carriers (Harvati and Hublin 2012, fig 12b). The archaeological evidence is consistent with the idea that the Middle Palaeolithic modern humans all over North Africa represent a metapopulation (Scerri et al 2014). So, based on everything I've just mentioned, the modern human makers of the 'Mousterian' either all have to be L0'6 people, or they all have to be older than the L0'6 people; they can't be both at the same time. And if they're pre-L0'6 people, which is clearly supported by most evidence, they would not have mtDNA L0'6 lineages before they came in contact with the ancestors of all living humans.

One such example of contact between the makers of the North African 'Mousterian' and L0'6 people could be represented by Iwo Eleru, who is found in association with LSA artefacts that are more typical of L0'6 people, but who has morphological affinities with 'Mousterian' Afro-Eurasian modern humans (Stringer 1974Harvati et al 2011). Note, BTW, that Iwo Eleru's affinities are just another example of getting the impression that makers of the 'Mousterian' were not on the L0'6 tree. What I mean is the following. Iwo Eleru's morphological relatedness to these 'Mousterians' seems to be mirrored by the presence of A1b (known as A0, today) in both regions. A1b so far seems to have a distribution between Northwest and West/Central Africa (Scozzari et al 2012, fig 2 and 3). It is noteworthy that North African A1b is not derived from the West/Central African versions discovered so far (Cruciani et al 2011b), consistent with (but not proof of) a north to south movement. Interestingly, there is more to the correspondence between Iwo Eleru's morphometric affinities and the agreeing distribution and structure of A1b. For one, A1b is older than the L0'6 people (consistent with everything I've been saying), making it a good candidate lineage of at least some makers of the North African 'Mousterian'. Secondly, the West/Central African versions of A1b so far are consistent with an arrival in the region sometime in between ~40-25kya (Scozzari et al 2012, fig 2). To anyone familiar with North Africa's archaeological record, that last date range should immediately get their attention. ~40kya marks the end of the Aterian and the onset of arid conditions in the Sahara. The dating and structure of West African A1b therefore fits the idea of Saharan migration towards hospitable areas like West Africa.

As you can see, the data doesn't fit well with the idea that people related to Skhul and/or the Nubian complex people were the Basal Eurasians.

Why it's not a 'Crushing Blow' that Some Natufian Samples don't Have Much Basal Eurasian Ancestry as Neighboring Samples

Let's move on to the other arguments against an African origin of Basal Eurasian. Not necessarily Lazaridis et al, but various commentators have claimed that the higher proportion of Basal Eurasian in Hotu III (~60%) than in the sampled Natufians (~40%) necessarily makes Basal Eurasian inconsistent with an African origin. This is a fallacy, because from the skeletal remains this was already self-evident. For instance, in my previous post on Basal Eurasians I made it clear that Natufian sites have yielded skeletal remains with some appearing more tropically adapted than others. Therefore, to those who subscribe to the view that Basal Eurasian is African, it's not an earth-shattering revelation that some Natufian samples don't have as much Basal Eurasian as neighboring sites. In order to falsify that Basal Eurasian is African, skeletal remains from the Shuqbah and Athlit caves analyzed by Keith have to be sampled. This is what Keith 1931 has to say about these Natufians in particular:

In physical type the late cave people of Athlit were the same as the Natufians of Shuqbah. An elaborate comparison of the human remains found by Miss Garrod in the Capsian deposits of Shuqbah with those from the Aurignacian and mesolithic deposits at Athlit has led me to the conclusion that all represent the same racial stock. The skull from Shuqbah, depicted in figs. 67, 68, may be accepted as a type. Especially remarkable was the nasal development of this cave people—often almost negro-like in the flattening of the nasal bridge and in the width of the inter-orbital septum. The nasal bones, although their transverse arch is depressed, still have remarkable dimensions—such as might herald the pronounced nasal development of later Semitic races (figs. 67, 68)." 
Keith 1931 pp. 221-22
Clearly, not all Natufian sites are interchangeable proxies when it comes to testing whether the Levant was the main corridor through which Basal Eurasian left Africa and reached other places in the Middle East. We've seen this sort of Eurocentric contentment happen before in the blogs when, for instance, hotchpotches of assorted Natufian skeletal remains were indiscriminately pooled in a single sample and didn't come out as 'African' as they have in the past. See, for instance, Arensburg & Lahr 1995 where the measurements, including stature, betray that some Natufians range from somewhat, to completely unlike Keith's Natufian sample (the stature of the latter sample fell in the range of 1525-1600 mm., which is a far cry from the Natufians at Mallaha, whose mean stature was 1741 mm.). A similar outcome happened in Holliday 2013, which saw the Natufians (this time, exclusively from the El Wad site) group with somewhat cold adapted groups rather than with Africans. It's not uncommon to see partisans on the blogs uphold such results as the 'ultimate evidence' that earlier analyses on the Natufians are now somehow refuted.

Obviously, the discrepancies in the various reports are due to the varying degrees of the Natufians' aboriginal Middle Eastern (i.e. WHG-like) ancestry. In Stock 2013, for instance, we seem to be dealing with a Natufian sample with less WHG-like ancestry, as their post-cranial affinities seem to differ from the affinities reported in Holiday 2013 (the Natufians from Stock 2013 cluster with Predynastic Egyptians and the Masai as opposed to their Epipalaeolithic predecessors). This doesn't mean that either Stock or Holiday is 'right' and that one undermines the other; they're both 'right'. Now that we've established that an African origin of Basal Eurasian doesn't require all Natufians to necessarily surpass neighboring samples in their amount of Basal Eurasian, we can move on to the next objection.

Why More Basal Eurasian in Mesolithic Iran Doesn't Rule Out an African Origin

Various commentators online point out that there are no African haplogroups so far in the Iranian samples, yet, they have more Basal Eurasian than Natufians. Therefore, these proponents argue, Basal Eurasian must not be African. Similar questions have been raised regarding the apparent sparseness of E-V13 among European farmers. Such arguments arise out of ignorance in how haplogroups propagate over time under evolutionary forces. I've already explained this in my previous post on Basal Eurasian, which you can find here. What I want to deal with right now is the fact that the features that make Keith's Natufians stand out, can be found in the same Iranian sites that we're told are supposed to be free of African admixture. We'll get to the features of the oldest individuals in the Hotu cave (Hotu II and Hotu III) in a minute. But first, some context is required. A year ago I had the following to say about the African contingent among the Natufians:
Descriptions of Natufian and PPN remains consistently report post bregmatic depression, among other features consistent with Sub Saharan African ancestry [Meiklejohn et al 1992; Agelarakis 1993; Hershkovitz et al 1994; Bocquentin 2003]. Strangely, despite proximity to Africa, the prospect of these being a marker of recent African ancestry is typically not seriously considered, as in [Meiklejohn et al 1992], where it is treated as an artificial deformation.

A Natufian skull pictured in Bocquentin 2003 exhibiting many features consistent with recent African ancestry, including post-bregmatic depression (see the slight depression along the length of the vault of this skull).

As was pointed out in this quote, post-bregmatic (or post-coronal) depression (PBD) is a cranial trait that is often used in forensic analyses as a possible indication of Sub-Saharan African ancestry. What the quoted excerpt doesn't explicitly state is that the data in several of these sources indicate that PBD is found in the wider Middle East, seemingly from the late Epipalaeolithic/Mesolithic onwards. For instance, it is attested in the proto-neolithic in northern Iraq (Shanidar cave), the neolithic in Iran (Ganj Dareh) and yes, even in the Mesolithic of Iran (Hotu cave). However, one problem is that cranial deformation practices (including some that, perhaps, induced PBD) became frequent all over the Middle East after the Mesolithic and Epipalaeolithic. This makes the use of PBD to diagnose African ancestry after the Epipalaeolithic and Mesolithic somewhat unreliable. This is exacerbated by the fact that the people involved with studying these practices make no serious effort to distinguish genetic from man-made causes of PBD.

Meiklejohn et al 1992 don't even consider that at least some of the PBD in neolithic Iranian and other sites could have a genetic cause. They attribute all incidences of PBD to artificial cranial deformation. Granted, the clear evidence of various forms of cranial deformation (other than PBD) in these sites indicate that artificial deformation could explain at least a portion of the sky-rocketed frequencies of PBD during the neolithic. However, I'm not saying this because the authors' arguments were so compelling. In fact, it strikes me as remarkable that the authors don't take a step back from their conclusions given the discrepancies in their data. According to their own data, most of cranial deformation-inducing practices left behind cranial 'scars' reasonably consistently. Only one type of supposed 'deformation' was not associated with any scarring pattern. Curiously, this exception was PBD. For instance, the bandaging practices that supposedly induced cranial deformation artificially, left behind horizontal and diagonal groove patterns on most of the skulls. However, the authors report no vertical grooves that would be consistent with PBD-inducing vertical bandage practices (see image to the left). Instead, we're left with the following substitute for actual evidence:
Of the features noted by Lambert, Post-Coronal Depression and Parietal Bulging are hard to explain other than by deformationMeiklejohn et al 1992 
But PBD was the most frequently observed 'deformation'. Is it merely a coincidence that no scarring pattern was found for PBD? Remarkably, the authors seem to be unaware of the fact that PBD is a non-metric trait. Either that, or they refuse to consider that genetic causes plays a role here, in which case "hard to explain", as they put it, simply means 'unwilling to consider'.

According to these authors' source, Lambert, this practice kicks off in the Middle East during the neolithic. This is unlikely, as it occurs in the preceding Hotu sample (Angel 1952). It would be interesting if artificial deformation was an attempt to mimic the natural headshapes of prestigious or mythological groups in the Middle East. Are the artificial examples of PBD a case of 'art' imitating life? Whatever's the case, PBD, along with other features that are consistent with Africans, are found in sites where Basal Eurasian peaks so far. These sites include the Iranian neolithic (Meiklejohn et al 1992) and the Iranian Mesolithic (Angel 1952). Angel has, among other things, the following to say about a specimen from the latter site (Hotu II):

The pentagonoid skull vault is 75 to 100 cc. larger than the 1,325 cc. modern average in capacity, and is long and high with marked post-coronal depression and concave and sinuous temporal planes, as if the infantile sharp curve of the parietal bone had not been fully corrected by later peripheral remodelling.
(Angel 1952).

About Hotu III Angel says, among other things:

The skull has a strikingly capacious vault (1420-1460  cc.),  ovoid,  broad, well-filled  with  wide-set base  (129  mm.)  and  approaching the  "square- head"  minority  among  Upper  Palaeolithic  and later Europeans.  The face has wide cheeks, wide nose,  and  a  protruding  chin,  and  probably  resembled number 2.
 (Angel 1952).

If this Hotu III individual measured by Angel corresponds to "Hotu IIIb" sequenced by Lazaridis et al 2016, he's more than consistent with the Shuqbah Natufians and many Africans he's supposedly unrelated to. His nasal index is estimated at 58.2 (very broad) while his upper facial index is estimated at 44.2 (again, relatively broad). (Estimated, because the Hotu III had to be reconstructed). Hotu III combines these features with contrasting features (including somewhat broad neurocrania [CI of 77.7], a stocky build, and a general Upper Palaeolithic European 'look') that seem like they were inherited from contemporary West Eurasians. In this respect, Hotu III calls to mind Homo I from Atlit Yam, a PPNB individual who combines similar African-like and West Eurasian-like morphometric features. This mosaic of features is consistent with Hotu III's genome, as he's reported to be 40% West Eurasian and 60% Basal Eurasian.

Homo I from Atlit Yam (PPNB). Taken from Hershkovitz & Galili 1991. Note similarities with and differences to the aforementioned descriptions of both Hotu III and Atlit Cave Natufians. Also note the missing upper central incisors (a North African practice [Stojanowski et al 2014]and corresponding atrophy.

As you can see, rather than constituting a 'problem' to the African origin of Basal Eurasian, Hotu III seems to embody quite the opposite. He fits in an emerging picture that shows that Eurasian genomes with high degrees of Basal Eurasian ancestry likely also exhibit 'African' features, whether morphological or otherwise. This tight correlation hasn't been violated yet. Aside from the Mesolithic Hotu individuals, we see it in Stuttgart's cranio-facial appearance and various pooled early neolithic samples. If (and this remains to be seen) Kostenki-14 has Basal Eurasian, as has been suggested by some (Seguin-Orlando et al 2014), some of his cranio-facial features can also be interpreted as standing out in the African direction. Unfortunately, we do not have the skeletal remains of the recently sequenced hunter gatherers from the Caucasus. They have a substantial input of Basal Eurasian (Jones et al 2015), so, in light of what I just said, their skeletal remains would be of interest. Interestingly, the circumstantial evidence so far suggests that this correlation holds for them as well. The reason for this is that their partial descendants, the Corded Ware people, had exceptionally long and narrow heads (Coon 1939), The fact that the height of their neurocrania exceeds their cranial breadth (indicating a acroplatic index near zero) is interesting, as anthropologists from the previous century have often claimed this to be atypical of recent Europeans (Smith 1941).

"The index known as the acroplatic index, which was devised by Carl Pearson, is of special importance. It gives the relationship between height, width and length, and it has been found that Europeans as a rule show an index of something in the vicinity of +6, Negroes as a rule are approximately zerobut Egyptians, strangely enough, show a minus index, and in this particular case, -4.2." (Smith 1941)

Their cranial index (CI) and body measurements also show a peculiar trend, even compared to Upper Palaeolithic Europeans. In these two characters early Upper Palaeolithic Europeans are more 'linear' (Holliday 1997a; Vercellotti et al 2008Churchill et al 1999) than late Upper Palaeolithic Europeans. The Corded Ware people not only reverse this trend to the exclusion of nearby WHG-like groups, but, at least in terms of CI, they come out on the other end. Meaning that their head shape seems to be more 'linear' than their ancestors in both eras. This can't be explained with purely ecological explanations.While many of the Corded Ware peoples' features go in the other (i.e. European) direction, this is still a strange combination of traits for a people who supposedly have been out of the tropics for over 50ky.

Moving Forward in 2016

It has always been possible to detect clearly intrusive tropically adapted features in many Middle Eastern and Central Asian sites in between the early Holocene and the Bronze Age (Ogihara et al 2009; Dubova 2001, Stock et al 2007). What's new is that it will now, for the first time, not be up to researchers' whim whether they want to 'admit' they're there and show morphological discontinuity with groups in the region today. Did they come from Africa? The Persian Gulf Basin? Saudi Arabia? The possibilities used to be too numerous to get to the bottom of things. Moreover, any progress in this area was doomed to begin with because the select few with access to these remains have no interest in the study of Africans beyond Africa (indeed, sometimes they're not even interested in studying Africans beyond Sub-Saharan Africa, see Harich et al 2010, for instance, who allow for no mtDNA L in North Africa other than through slave trade). In this sense, aDNA can act as watershed moment.that allows for rapid progress in an area that has been both intentionally and unintentionally held back.

One gray area that will inevitably become illuminated going forward is the fact that Afroasiatic has many far flung relationships with extinct and extant pre-Semitic languages in the Middle East. What I'm interested in is not whether Basal Eurasian has something to do with this (those who are not in denial already know that this is likely), but, to what extent Eurasian languages were influenced and how many African migrations were involved:

These arguments speak for the Levantine [origin of Afroasiatic]: 
Distant relationship of Afroasiatic with Kartvelian, Dravidian, Indo-European and other Eurasiatic language families within the framework of the Nostratic hypothesis (Illič-Svityč 1971-84; Blažek 2002; Dolgopolsky 2008; Bomhard 2008). Lexical parallels connecting Afroasiatic with Near Eastern languages which cannot be explained from Semitic: Sumerian-Afroasiatic lexical parallels indicating an Afroasiatic substratum in Sumerian (Militarev 1995). Elamite-Afroasiatic lexical and grammatical cognates explainable as a common heritage (Blažek 1999). North Caucasian-Afroasiatic parallels in cultural lexicon explainable by old neighborhood (Militarev, Starostin 1984). Blažek

As you can see in the excerpt above, this evidence of contact was used by various linguists (as well as bloggers) as evidence that Afroasiatic can't possibly be African. However, this hypothesis against the African origin of Afroasiatic has now completely imploded, as it relies on the mistaken notion that Levantines weren't partially African, genetically and linguistically. Now that African Y chromosomes have been observed among Natufians, linguists who oppose an African origin of Afroasiatic can no longer pretend to be ignorant about the skeletal record and claim that Afroasiatic doesn't become associated with Africans until it arrives in Africa in the early Holocene. 

North Caucasian, Kartvelian and Indo-European languages show evidence of interactions with Afroasiatic. We already know that Caucasus hunter gatherers had Basal Eurasian (Jones et al 2015), so that potentially explains the observed linguistic relationships between Afro-Asiatic on the one hand and North Caucasian and Kartvelian on the other hand. As we've already discussed, the proto-Neolithic individuals in the northern Iraq (Shanidar cave) show the same morphological features that pull the Natufians in an African direction in morphometric analyses. If genetic analyses confirm that neolithic Iraqis have Basal Eurasian and/or other types of African ancestry, this can potentially explain the observed linguistic relationships between Sumerian and Afro-Asiatic. Lastly, Basal Eurasian in Hotu III and early Iranian farmers can potentially explain links between Afroasiatic and Elamite.

Shanidar Cave proto-Neolithic individual (left). Wouldn't look out of place among mid-holocene Nubians and Egyptians as shown by its general resemblance to the dynastic Egyptian skull from Memphis (right). Images taken from Agelarkis 1993 and Martin 1841, respectively.

1 Throughout this article I use 'Mousterian' loosely as a catchall phrase to refer to the modern human makers of Mousterian-like tools in Afro-Eurasia. Terms like like 'Pre-Toba modern human OOA migrants' are too unwieldy to use repeatedly. In this article such a shorthand was useful, but I'm aware of the pitfalls. I would prefer the term MSA on the African continent where relevant

12 opmerkingen:

  1. WOW! This has got to be one of the most well written blogs that I have read this year. In my opinion you should really consider going into this field and writing your own published papers or write a book.

    1. Glad to know you found it useful. On paper it's of more use than in my head, where it benefits no one other than myself. I reserve posts on this blog for groundbreaking stuff that warrants in-depth analysis, so when papers like Lazaridis come out and I read the comments online, I know it's time to start writing.

    2. I see you touch on admixture just read this piece .. What do you think Clyde winters disagrees

      Dr. Clyde WintersSunday, June 26, 2016 2:49:00 pm
      This paper is nonsense. It is found on hypothesis which do not reflect the African reality. Firstly,there is no Afro-Asiatic language family and the Bantu speakers did not originate in West-Central Africa. In addition, there is no discussion of archaeological evidence in support of any of the authors propositions, statistics should not be enough to support the hypotheses discussed in this paper.

      Reading this paper is like reading any other racist Eurocentric article written at the turn of the 20th Century perpetuating the Hamitic myth.THE Hamitic myth states that everything of value ever found in Africa was brought there by the Hamites, allegedly a branch of the Caucasian race. Seligman formulated this hypothesis which led researchers to declare that the Fulani and Afro-Asiatic speakers were Hamites. This racist theory was abandon, but appears to be coming back into vogue among geneticists who lack knowledge about African history.

      As a result, when this study declares that the Fulani, who are not of Eurasian origin, and the Afro-Asiatic speakers have a high frequency of Eurasian (white) admixture, this paper is just reinforcing a hypothesis that lacks credibility. The results of this paper only perpetuates the Hamitic myth, many researchers had thought was abandoned--but has remained constant by geneticist who dress the hypothesis up in new clothes based on statistics, instead of actual archeaogenetics evidence.

      The authors assume that the Bantu migrated out of Cameroon 2,5kya. This is ludicrous because the Bantu had been living in the Nile Valley long before 500BC.

      In summary this paper is maintaining the status quo dogma that the Bantu and the rest of the Niger-Congo speakers are true Negroes, and the Afro-Asiatic speakers and Fulani are Hamites, i.e., dark skinned Caucasians. This paper offers nothing new in relation to African genetics, it is a throwback back to the 1930's racist antropological studies.

    3. When you mention admixture in regards to my post, are you commenting on the part where I say that Basal Eurasian component in Africa is not completely pure?

      This is not a controversial idea. When you look at Ethio-Semitic speakers or even Lowland East Cushitic speakers and you compare them to closely related groups further inland, like the Borana or Rendille, you can tell they look different. For instance, they tend to have lighter skin.

      They all have Basal Eurasian but Ethio-Semitic speakers especially are good examples of Africans whose Basal Eurasian backmigrated to a large extent. Maybe to the tune of 50% (or more) of their Basal Eurasian is due to Eurasians who brought with them certain haplogroups while in the others the percentage of indigenous Basal Eurasian is higher.

  2. It's a great write-up for the anthropologically savvy. My only criticism is that the academic prose, while appropriate for your intended audience, might drive off some laypeople who might otherwise find your arguments of interest. It's not like I can link to this on Facebook and expect all my friends to understand what you're saying since they wouldn't understand the technical language. Do you think it possible to break down this information into something lay readers (e.g. people who read publications like Scientific American or LiveScience) could understand?

    1. Thanks for your comment. I agree with what you're saying. I'll think about it. I could use some help though as I wasn't planning on doing another article until something important comes out. If you're serious about this I can give you admin access to this blog (with your own blogger account) with the click of a button. We can then work on a draft together over time.

    2. Sure, I'll see if I have enough time over the weekend to write something up. I think I should.

  3. Deze reactie is verwijderd door de auteur.

  4. Deze reactie is verwijderd door de auteur.

  5. Have you read the recent North African Neolithic paper (Fregel et al.)? It seems that in their analysis, Natufian has a higher affinity to Sub-Saharan Africa than almost all contemporary Middle Easterners. And sure enough, the geospatially more remote Iranian Mesolithic/Early Neolithic genomes show the same African affinity. So pretty much what you predicted in this post.

    It's in Figure 6.3 in the supp. material:

  6. your blog is very useful, i hope you post something again, it would be very appreciated

  7. Africans everywhere for as far as the eye can see. Nile Valley, Levant, Tigris-Euphrates Valley, Persia/Iran, and Harrappa. Of course this stuff is controversial. It completely upends "standard" history and leaves room for the so called Afrocentrics to say:
    "see ..... we told you so"