Saturday, June 19, 2010

KhoiSan: As the "Oldest" Surviving "Tribes"

The reader may recall having heard the claim that the KhoiSan are the "oldest living human populations" at some point or another. The saying may well have taken the character of something of a religious faith within some circles. When the whereabouts of the source for this "understanding" is inquired, common references have been to the National Geographic human genome project web-page and Semino et al.'s (2001) study regarding the "deepest clades" for instance. The former bases this on rather flimsy material that KhoiSan male gene pools comprises of deep root clades like hg A, while the latter has every now and then fallen victim to misunderstanding or misinterpretation of what is being related in the authors' findings within the contexts that they were made. Fact is the KhoiSan are no more representative of the oldest anatomical modern human beings any more than other living populations.

As A. G. Morris (2003)—who wrote on the issue rather elaborately in his "The Myth of the East African 'Bushmen"put it:

"The ancestral allelic states identified from both mitochondrial and Y-chromosome DNA do indeed reflect ancient roots but the total morphological pattern is as derived from the ancestral form as it is in any other human group. The morphology of populations reflects adaptive patterns that are dynamic responses to sequences of adaptive events in the local environment."

Since KhoiSan groups are a fairly conservative cultural bunch who have retained the hunter-gatherer tradition, it is easy to ideologically dismiss them as THE oldest surviving humans; however, level-headed examination of material evidence demonstrates that such ideology does not hold water. It would be more accurate to say that KhoiSan groups have better preserved the lineages directly tied to the deepest clades than many other living populations, but this is a far cry from saying that they are THE OLDEST surviving human populations. As is the case with many hunter-gatherer groups, the San generally comprise of communities of comparatively small effective-population size that are generally isolated from other ethnic groups. This small effective-population size chimes in on the preservation of modal haplotypes, even though it may well work either for or against newly emerged UEPs via the action of random genetic drift, and/or possibly selective pressure. This condition factors into such population-expansion/contraction gauging mathematical models as that of Fu's function Fs (F statistic), where the values reported for hunter-gatherer, nomadic and other groups with small effective-population sizes are generally non-significant, and in the case of Tajima (D), values are expected to approach zero, assuming the population size had stabilized.

While KhoiSan groups may have preserved clades that are understood as those of the same line as the basal clades of human Y-DNA network, even here, new mutations had been attained well after successive bifurcation events that led to other lines of paternal ancestry. It is not as if these groups had remain stunted in their Y-DNA heritage, wherein no additional mutational events had not occurred over the course of their bio-evolutionary history, which raises the question as to when does a UEP truly represent a distinctive genealogical lineage from a preexisting one and hence, considered a bifurcation event? Designated accumulated mutations at designated loci are cross-referenced between distinctive lineages, and herein, apparently the mutations that are shared are considered to be in their ancestral states vis-a-vis the divergent lineages at hand. Henceforth, a series of mutations common to the divergent lineages are treated as belonging to a single line until such point where the UEPs respective to divergent lineages differ, when said UEPs in question are treated as characteristic mutations of the divergent lineages respectively and as departure points from a preexisting lineage. To put it in other words, where divergent lineages depart as far as UEPs are concerned, such departure point is perceived as relic of a bifurcation event, even if the MRCA was technically still living within a preexisting population before a founder effect took hold for the new line of ancestry. With such basics brought to attention, the understanding that hg A diverges into several sub-haplogroups can be recalled. These sub-haplogroups could well be given distinctive names other than the "hg A" designation" as they meet the necessary traits to be deemed as new lineages in their own rights. However, since they are still treated as siblings of a larger hg A family, this cultivates the idea in individuals, that they are the most basal human Y-DNA clades, without putting much mind to the developments wherein these genealogical lines too have accumulated downstream mutations over time. Moreover, these sub-haplogroups are not exclusive to KhoiSan groups either; they have been identified in divergent African samples, from west to east, and north to south. Yes, their frequencies vary, with some places having less distribution, while others having more; hg A is particularly rich in eastern Africa to as far as Sudan, and in southern Africa.

It is the just-mentioned kind of thinking that leads to the overlooking of messages being related in such studies as that of Semino et al.'s (2001) "Ethiopians and Khoisan Share the Deepest Clades of the Human Y-Chromosome Phylogeny". In that study, the reader is informed that the KhoiSan samples shared deep root clades with Ethiopian samples; what is overlooked by some, is the observation related therein about the divergence of these shared deep root clades respective to the said groups. By divergence, apparently reference is being made to downstream mutations accumulated over time, after the ancestors of recent KhoiSan groups separated from the ancestors of recent Ethiopian groups. The level-headed reader would instantly read from this that if the KhoiSan diversity was accumulated after its divergence from the other mentioned group, then this diversity cannot possibly represent the "oldest" genealogy of humanity. The other group(s), which in this case are the Ethiopians, implicated in deep root clades is however dismissed on the supposed qualification that it is not a "pristine" autochthonous African group, having received "foreign" gene flow into its gene pool. How "foreign" gene flow is supposed to disqualify the likes of Ethiopians from being considered the "oldest" surviving group even though they carry the very same clades that are supposed to make the KhoiSan the "oldest" surviving groups, is anyone's guess. Rather, the KhoiSan are treated here as "pristine" autochthonous Africans with little to no "foreign" gene flow while sporting the supposed "greatest diversity" than any other known group, but exactly how the KhoiSan attained such diversity without exposure to exogenous genetic intrusion into the preexisting gene pool is not clarified.

Conventional wisdom elicits that diversity requires heterogeneitythe coming together of two different entities; otherwise, from a mutational standpoint, one is talking about the mutations accumulated over time from a preexisting lineage. However, internal diversity within a clade is a whole different entity from diversity born out of divergent clades. Internal diversity of a clade is insufficient to make the KhoiSan into a group with THE "greatest diversity", and so, if this diversity was attained was attained from gene flow from an outsider group, then the supposed qualification of the KhoiSan as a "pristine" group with no "foreign" admixture flies out the window. It matters not if the exogenous genetic intrusion would have come from another autochonous African group; if the lineage at hand was not part of a preexisting KhoiSan gene pool, then it would be "foreign" to their gene pool. On the question of diversity, one study cited by advocates of "KhoiSan are the oldest surviving tribe", besides the obvious reason [the repetition of the claim about the "oldest known lineage"], comes from Schuster et al. (2010):

The genetic structure of the indigenous hunter-gatherer peoples of southern Africa, the oldest known lineage of modern human, is important for understanding human diversity. Studies based on mitochondrial and small sets of nuclear markers have shown that these hunter-gatherers, known as Khoisan, San, or Bushmen, are genetically divergent from other humans. However, until now, fully sequenced human genomes have been limited to recently diverged populations. Here we present the complete genome sequences of an indigenous hunter-gatherer from the Kalahari Desert and a Bantu from southern Africa, as well as protein-coding regions from an additional three hunter-gatherers from disparate regions of the Kalahari. We characterize the extent of whole-genome and exome diversity among the five men, reporting 1.3 million novel DNA differences genome-wide, including 13,146 novel amino acid variants. In terms of nucleotide substitutions, the Bushmen seem to be, on average, more different from each other than, for example, a European and an Asian. Observed genomic differences between the hunter-gatherers and others may help to pinpoint genetic adaptations to an agricultural lifestyle. Adding the described variants to current databases will facilitate inclusion of southern Africans in medical research efforts, particularly when family and medical histories can be correlated with genome-wide data. - Schuster et al., Complete Khoisan and Bantu genomes from southern Africa.

That San hunter-gatherer bands from disparate regions of the Kalahari are more divergent from one another than that between non-African groups like Europeans where nucleotide sequences are concerned, is hardly surprising, when the aforementioned issue of the hunter-gatherer lifestyle and group 'isolation" is considered. Furthermore, the so-called "Out of Africa" living non-African groups are supposed to have emanated from a subset of African gene pool, and so, their between-group divergence is expected to be comparatively limited when pitted against the potential of that of disparate African samples. For instance, the oldest common recent paternal ancestor of most non-African groups goes back to the F-M89 mutational event. The study implicates a Bantu sample, which is expected to show phylogenetically shallower lineages and reduced genetic diversity since it is a common understanding that the Bantu are a comparatively recent agricultural-oriented offshoot of preexisting Niger-Congo speaking groups in the vicinity of the region straddling southern Nigeria and northwestern Cameroon. Speaking of which, Semino et al.'s 2001 study compares groups picked up from the much more ethnically diverse populations like Ethiopians, against the lone western African sample of the Senelagese as the representative candidate for all "Niger-Congo" speaking populations. It is widely understood that "Niger-Congo" speaking groups are a relatively recent bunch phylogenetically speaking, although preexisting groups of the territories occupied by these Niger-Congo speakers have been integrated into the latter to some degree or another. So of course, these Niger-Congo speaking populations are expected to report lower incidences of deep root clades like hg A or hg B than the remnant hunter-gatherer groups of that very region and southern Africa, as well as the more ethnically diverse areas of Eastern Africa.

While Semino et al.'s (2001) study represented Ethiopian samples, which is predictably rich in deep root clades like hg A, it did not study the other eastern African populations also rich in such lineage, like say, the Sudanese. Had this been done, involving sufficiently large-sized and representative [from a geographically-comprehensive sense] sample, then it would have served as a reality check to those who look to their study as the source of the two-subject ideology of "Ethiopians and KhoiSans are the oldest populations", the alternative viewpoint out there about the supposed oldest surviving human populations. It is this mentality that leads certain individuals to put forward the notion that "Ethiopian and the KhoiSan" or just the KhoiSan serve as the best representations of how the oldest modern human populations looked like phenotypically, particularly where skin tones are concerned. For instance, the somewhat relaxed skin melanin content of KhoiSan groups of the southern Africa is professed to be the default condition of modern humanity, based on the ideology that they represent the "oldest" human groups. Naturally, this train of thought ignores the issue of genetic mutations within KhoiSan groups, even if one were to buy into the questionable proposition that the KhoiSan are "the" oldest surviving groups, as opposed to the more objectively accurate viewpoint of being one of the oldest surviving groups, where genetic structure is concerned.

It was noted on this site earlier that the San populations reportedly carry the "derived" variant of the OCA2 allele:

The lightly pigmented hunter-gatherer San populations of Southern Africa is exceptional in having a high frequency of the derived allele relative to geographically proximate and more darkly pigmented African populations (Jablonski and Chaplin 2000), further supporting the importance of OCA2 in regulating normal variation in pigmentation. The widespread distribution of the derived allele in the CEPH-Diversity Panel suggests that it is not necessarily a new mutation, nor has it been restricted to a specific geographic area.

In relation to this, in the original blog posting the present author wrote: While it seems plausible that the “derived” OCA2 gene came into being before the out-of-Africa migration that give rise to modern Eurasians, it doesn’t appear that this derived allele was necessarily widespread, and may well have been later on selected for in European and East Asians [as the following distribution pattern comparisons between "Native Americans" and "East Asian populations" serve as possible telltale sign for],…

Interestingly, derived allele frequencies at this locus are quite different between Native American (15%) and East Asian populations (45%), suggesting that perhaps the derived allele at this locus did not reach very high frequencies in East Asians until after the colonization of the Americas

Similarly, in the original blog posting, it was noted with regards to the “derived” SLC24 A5 allele:

In contrast, the SLC24 A5 11*A-derived allele is found at low frequencies in several sub-Saharan populations including the West African Mandinka and Yoruba, the Southern African San, and South West Bantu. *The presence of the derived allele (albeit at low frequencies) in some sub-Saharan populations may be due to recent gene flow from European and Central Asian populations...

Once again, the San are implicated in a derived allele. The proposition that the derived allele was attained in the above-mentioned darkly pigmented groups through "recent gene flow from European and Central Asian populations", could just as well have been through third party introduction from coastal northwest Africans in the case of the western African groups, but its presence notably extends to the Southern African San. It is not clear here whether the San are of the so-called "Bushmen" residing in the Kalahari desert, who are noticeably more isolated from other ethnic groups, including non-Africans like Europeans than the Khoi of the southern tip of Africa, who are relatively more integrated into the "exotic" groups of that region. Either way, both KhoiSan groups are generally lightly pigmented in comparison to their darkly pigmented neighbors. Alternatively, the derived SLC24 A5 allele could have been around some time before the major "Out of Africa" migratory event that led to recent non-Africans in low frequencies, to be followed up with loss of the marker in the ancestors of east Asians but with a retention in the ancestors of Europeans; this is one scenario of a two-scenario case...

—1)The relatively high frequencies of the derived allele in Central Asian, Middle Eastern, and North Africa seem likely to be due to recent gene flow with European populations.

—2)Alternatively, the derived allele may have lost in the ancestors of modern East Asians but retained in the ancestral European populations. The allele then rose to high frequency in Europeans following the divergence of Europeans and East Asian ancestral groups.

In the latter scenario, the scattered low incidences in the traditionally dark African populations, including the San, can be seen as a relic of the low distribution of the marker some time prior to the aforementioned "Out of Africa" migratory event, and before its subsequent founder effect events outside of the African territory. Whatever the case may be, nucleotide distribution patterns point to the SLC24 A5 allele in the lightly-pigmented groups to be the more recent [and hence, derived] variant when compared to the variant that predominates the darkly-pigmented groups of Africa and Melanesia. This development would stand in contrast to the ideology of those who fall back on the claim that "KhoiSan are the oldest surviving tribes" to bolster their case, as it would maintain that the alleles found in the San are the derived version, and hence, reinforcing the point that the San are no more physiologically representative of the earliest anatomically modern humans than other recent human groups. Furthermore, from just the genetic standpoint, this finding suggests that Sans are not genetically representative of the earliest human populations either. Still, some insist that the San's largely nominal role as the "oldest surviving tribes" bolsters their ideology against scientific observations to the contrary. Such ideological standing cannot be held tenable any longer once the implications of the distribution patterns set in.

It is for instance, observed that ancestral variants of the pigmentation-affiliated alleles mentioned in Norton et al.'s (2006) study predominated in the darkly pigmented African and Island Melanesians than other groups. One would expect there to be an inclination for alleles linked to dark pigmentation for Africans and the Island Melanesians respectively to be distinct, even if they end up serving the same function, were they the product of convergent but independent evolution and re-adaptation. However, the darkly pigmented sub-Saharan Africans and Melanesians share alleles that are determined to be in the ancestral state, keeping in mind that the classical uniparental markers point to Melanesians as being one of non-Africans preserving the deepest-rooted clades of those implicated in the so-called major Out-of-Africa migratory event [note: the deep clades from the African context should be distinguished here from that used in the non-African context]. This kind of intuition into the distribution patterns involved renders it very difficult to hold it tenable, that the variant of OCA2 for instance, as that described for the San, is the "pristine" or ancestral state of the allele, all in the name of the KhoiSan being the "oldest surviving tribes".

Image Caption: Even though some epidermal melanin content relaxation likely took place at some point in San micro-evolutionary biohistory, they are pretty much still considerably pigmented folks, and could just as well fit in with the rest of the folks who may socially be referred to as "black people". This couldn't be more clearer when a Kalahari San man is standing right next to fairly low pigmented European figures like Spencer Wells of the National Geographic Genome Project, in the photo above.

The evolutionary turn towards considerable skin eumelanin content in tropical Africa, the scientifically understood regional home to modern humanity, would have occurred well prior to the emergence of anatomically modern human; it would have been put in place by the time of anatomically modern humans' immediate ancestors—which would be the African Homo Erectus (Homo Ergaster, per some sources), according to the leading scientific understandingwho also did away with fur. So hypothetically, if the evolutionary trend was towards skin tone lightening or skin eumelanin content relaxation as opposed to intensification against the backdrop of the intense UV radiation environment of the tropics, then most populations today in that same environment would have continued to be lightly pigment in the epidermal layer of the skin. Alas, this is not the case.Instead, skin melanin content relaxation noticeably increases in frequency only as one moves into extra-tropical regions of the planet; taking recent migratory events out of the equation (and extraordinary cases like that of circumpolar hunter groups such as the Inuit, i.e. when compared to their lower latitudinal neighbors), the further the region from the tropics, the greater the trend for loss of epidermal melanin content. Whether living pantroglodytes or else remote ancestors of hominids had fur, and have lightly pigmented skin underneath the fur, is therefore immaterial to the default state of modern humanity, as some imagine to be otherwise.

The aforementioned Semino et al. (2001) study was referenced by A. G. Morris (2003), who was mentioned here earlier, for its reliance on certain authors to bolster the claim that KhoiSan territory likely extended above the equator at some point in remote antiquity, and this is what he had to say about it:

Our understanding of the origin of African populations has been advanced considerably by research into mitochondrial and Y-chromosome DNA. One outcome of this research has been the recognition that living KhoiSan populations of southern Africa demonstrate mitochondrial DNA haplotypes that have extremely ancient origins (Vigilant et al. 1991; Soodyall & Jenkins 1992; Cavalli-Sforza et al. 1994). This, by implication, suggests that KhoiSan populations represent a very ancient form of humanity (Soodyall & Jenkins 1992). Perhaps more surprising is the Y-chromosome genetic link between the KhoiSan and some living Ethiopians (Cruciani et al.2002; Semino et al. 2002). Both papers describe a shared 'ancient paternity' between the groups and suggest that KhoiSan territory had once extended above the equator into Ethiopia and Sudan. They support their theory of linkage between Ethiopians and the KhoiSan with archaeological information that indicates a prehistoric presence of KhoiSan people in the northeast region. The authorities they quote are Nurse et al. (1985) and Tobias (1978). Cavalli-Sforza et al...

The evidence from osteology, serogenetics and anthropometry provides no satisfactory biological support for the presence of KhoiSan peoples in East Africa during prehistoric times. Nearly all of the evidence in support of this contention was collected during the period of racial typological assessment in physical anthropology and the conclusions of these studies are considered suspect by today's methodological techniques. Studies that have re-assessed prehistoric crania according to multivariate techniques have consistently rejected any relationships between these crania and the skulls of modern or prehistoric KhoiSan populations. Likewise, anthropometric and serogenetic studies have shown no relationship between living East Africans and their South African counterparts. Linguistic connections between the East African 'KhoiSan' and the South African KhoiSan languages are not rejected but the presence of clicks in these languages must not be considered proof of the biological unity of the people who speak the languages. Like the mitochondrial and Y-chromosome DNA haplotypes shared by modern Ethiopians and Khoisan, these should be taken as signs of survivorship of ancient genetic polymorphisms and not survivorship of ancient people. The history recorded by the presence of clicks and ancient gene lineages underlie more recent population events (Underhill et al. 2001) but it is the more recent population events that have defined and shaped the KhoiSan and modern East Africans

With the demise of the East African KhoiSan myth, other models of origin for the populations of sub-Saharan Africa must be considered...This paper favours a distinct regionalisation of prehistoric populations where living KhoiSan groups are seen as native to the southern African sub-region and their specific biological features represent their specific genetic history in that region alone. - A. G. Morris (2003), The Myth of the East African 'Bushmen.

Of course, it is not inconceivable that the ancestors of KhoiSan groups were originally situated in territory above southern Africa, in the tropics. In fact, their DNA and aspects of their phenotype point to this. However, these ancestors would not have been fully ethnically shaped as the living KhoiSan as we know them. Even the KhoiSan mtDNA structure shed some light to gradual [and relatively recent, vis-a-vis the age of anatomical modern humans] ethnogenesis of KhoiSan groups as we know them today over time. Advocates of KhoiSan as the "oldest surviving tribes" generally point to modal hg L0d and L0k clades associated with KhoiSan samples.

Indeed, as analyzed on this site earlier, hg L0d and L0k have roots that are situated near the root of the human mtDNA tree, but even these haplogroups have sub-clades that are as recent, if not more recent in some cases, as sub-clades of phylogenetically-younger macrohaplogroups. These clades are not restricted to southern African KhoiSan, but have been identified in eastern African "KhoiSan" and non-KhoiSan speaking groups as well, although some studies have attempted to explain off their presence in non-KhoiSan speaking groups as possible gene flow from the geographically proximate "KhoiSan speaking" groups [e.g., Tishkoff et al. (2007)]. Even so, as implied in A. G. Morris' piece, the linguistic assignment of certain eastern African "KhoiSans-speaking" groups in the same language family as the southern African counterparts should not be considered as proof of biological unity, although the presence of hg L0d on the other hand, may suggest just that. It should be noted that the placement of languages of these geographical discrete groups is primarily based on the phonological similarities expressed in "click" sounds, rather than full genetic, phonetic and grammatical structural analysis. And even in the case of shared clades such as hg L0d, the coalescence estimations point to chronologically variant time frames of expansions, thereby suggesting shared distant "common" origins, rather than gene flow from the southern African based KhoiSan speaking groups into the eastern African based ones. Tishkoff et al. (2007) for example allude to this, when they note...

Although the presence of L0d in the Sandawe and SAK establishes a unique connection between these populations, it is possible that L0d could be a shared ancestral trait, or symplesiomorphy, rather than a shared, derived character.

They go onto add...

The absence of the L0d haplogroup in the Hadza suggests a lack of contact and gene flow between the Hadza and the SAK-speaking populations; however, the absence of L0d may also reflect a recent population bottleneck in the Hadza suggested by demographic data (Blurton Jones et al. 1992). Because of their antiquity, these haplogroups (mtDNA L0d and Y chromosome A) provide no evidence of recent exchange or recent common ancestry (prior to ~ 35 kya) of these southern and eastern African populations.

Our results (specifically mtDNA MDIV analyses, L4g- and E3b1-M35*-based date estimates [delta * mu]2, and simulated likelihood analyses for the Y chromosome) are consistent with the possibility that the SAK speakers diverged from a population ancestral to both Hadza and Sandawe populations over 35 kya, and then the latter population split to form the Hadza and Sandawe population lineages ~15–21 kya, with little subsequent gene flow (fig. 9a)...

For example, the 3 populations may have diverged from one another >35 ky (where population divergences occurred so close to one another that the events can be summarized as a trifurcation), then the Hadza and Sandawe came into contact with one another and exchanged alleles roughly 15–20 kya, with little subsequent gene flow (fig. 9b). In both these scenarios, the mtDNA L0d lineages would have existed in the populations ancestral to all 3 groups and were lost subsequently in the Hadza. The latter lineage loss is quite likely given the estimated time depth of L0d common ancestry between the Sandawe and SAK (.58 kya) and the likelihood of a recent bottleneck in the Hadza population (Blurton Jones et al. 1992).

This next piece is an interesting one...

Additionally, the observation in a Turkana population from northern Kenya of an L0d lineage that is phylogenetically close to the Tanzanian L0d lineages (fig. 5b) suggests that at one time the L0d haplogroup may have been more widespread across eastern Africa (Watson et al. 1997). it flies in the face of the implication that certain groups like say, the Burangi sample, contain the L0d markers from possible gene flow from "KhoiSan" groups due to "geographic proximity", as the authors seem to insinuate in an earlier passage. The L0d could just as well be serving as a relic of this earlier distribution pattern now being acknowledge by the authors. Continuing with the study, the authors note...

A third possibility (fig. 9c; consistent with the sharing of the L0d haplogroup by the Sandawe and SAK speakers and the IM analyses for mtDNA) is that the SAK-speaking and Sandawe populations diverged from one another more recently, although still >35 kya, than either split from the Hadza population (.55 kya).

As in model 9b, our estimates of common ancestry of the Hadza and Sandawe at 15–20 kya may reflect longer term isolation of these 2 populations with high levels of interaction (gene flow) around 15–20 kya, and little subsequent gene flow. Under this scenario, either the Sandawe and SAK speakers share the L0d haplogroup through a common ancestral population >35 kya (after separation from the Hadza) or they share the L0d haplogroup through a population ancestral to all click-speaking populations and the L0d haplogroups was lost subsequently in the Hadza population. These data provide no insight into whether populations ancestral to present day click-speaking populations originated in eastern Africa and migrated south, or vice versa. Although these 3 scenarios (a–c) differ, particularly in terms of the nature of the relationship between the Hadza and Sandawe, under any of the scenarios divergence of these 3 click speaking populations occurred very deep in time.

Given the location of the base of the L0 clade near the root of the human mtDNA network, the scenario entailing the loss of hg L0d in the Hazda is the more likely, but in any case, as the authors themselves acknowledge, the divergences of the groups that carry this marker is "very deep in time", even if the TMRCA ages of sub-clades respective to the groups chronologically differ. The keyword here is distant shared common ancestry, which would mean that the common ancestor was not ethnically identifiable as either recent group involved, including the southern African KhoiSan groups. Tishkoff et al. (2007) do report relatively older coalescent ages for the southern African KhoiSan groups compared to the eastern African groups, but this as they themselves acknowledge, simply means that latter expanded relatively later or diverged from the very same common ancestral source that the southern African groups diverged from at a later time. This situation does not make southern African KhoiSan THE ancestral group or the "oldest surviving tribe".

Furthermore, markers in other ethnic non-KhoiSan speaking groups have reported coalescent ages greater than that of the southern African KhoiSan groups by some research estimates. Gonder et al. (2006) for example, found that the coalescent age of hg L5 in their sample was greater than the hg L0 counterparts, including hg L0d and L0k, in their samples, due to the nucleotide separation between the L5 marker of a single Tanzanian Mbugwe individual and those of the three Tanzanian Sandawe individuals which differed from one another by very few sequences. The MRCA age reported for hg L5 was 129 +/- 22.1 kya compared to the MRCA age of 106 +/- 20.6 kya reported for hg L0d as a whole; note that the MRCA age of just the southern African KhoiSan hg L0d by themselves was  90.4 +/- 18.9 kya, while TMRCA age of the southern African KhoiSan hg L0k was 70.9 +/- 19.7 kya. The TMRCA age of hg L0f, which the authors say was restricted to eastern Africa only, was also older than that reported for southern African KhoiSan hg L0 markers (both L0d and L0k), having turned up as being 94.9 +/- 9.4 kya.

Another source that advocates of "KhoiSan are the oldest surviving tribes" draw inspiration from, is Behar et al.'s 2008 study of "The Dawn of Human Matrilineal Diversity", which was discussed here 2 years ago. The authors estimate that mtDNA TMRCA ages of their KhoiSan samples suggest that the divergence of these hunter-gatherer folks could not have occurred no later than ca. 90,000 years ago, and they place the upper bound of this divergence within the range of 140,000 to 210,000 years ago. Their 90 kya age is consistent with Gonder et al.'s (2006) estimates noted above for southern African KhoiSan hg L0d mtDNA markers, although Gonder et al.'s age estimations for southern African KhoiSan hg L0k does leave the door open for a possible divergence of said KhoiSan maternal ancestors at a time frame later than 90 kya.

In the earlier ATI blog review, the problem with linking the ethnogenesis of the southern African KhoiSanas we know themto such remote time frames had been covered via examination of several classical uniparental paternal markers that make up living KhoiSan gene pool, along with examination of potential archeological indicators of southern African KhoiSan presence in southern Africa in antiquity. First of all, considering the distinct TMRCA ages for the two characteristic markers of hg L0d and hg L0k, where would one place the attainment of the KhoiSan ethnic identity as it is understood today? Did it come with the arrival of hg L0d or that of hg L0k, assuming that they arrived at chronologically different time frames [although, they almost always occur together in the southern African KhoiSan groups, but not necessarily in the eastern African branches]? To complicate matters, as noted in the earlier blog analysis, the ages of  modal KhoiSan paternal markers, while they too consist of basal clades, date well later than the 90 kya mark. Did the KhoiSan identity then come with the paternal lineage, as opposed to the maternal ones? It is hard to imagine southern African KhoiSan as ethnic groups without the paternal component. If there had been a paternal component that is no longer genetically represented in contemporary southern African KhoiSan, but which was supposed to have arrived with the mtDNA markers at the estimated 90 kya migratory wave, well then, such a development only reinforces the earlier-mentioned point that the said KhoiSan group as they are today, is a relatively more recent development than the time frame at hand, and that they would hence, not represent the survival of that earlier component that drifted out of existence. On the other hand, if one is to assume that the ethnogenesis of the KhoiSan identity came about with one of the deep root male clades that survive in their gene pool today, then that would place such development well beyond the 90 ky ago period at a later time frame. Furthermore, contemporary southern African KhoiSan paternal gene pool is even more heterogeneous than the maternal counterpart. This would imply that multiple waves of migration cemented the contemporary KhoiSan ethnic identity, as the clades range from deep root clades like hg A and hg B to the more recent ones like hg E1b1a and hg E1b1b1, unless one maintains that the KhoiSan ethnic genesis was put in place with the arrival of one deep root clade or the fusion of groups with distinct deep root clades.

Click on all images for greater resolution.

Image captions: Note in the Y-DNA maps that the deep root clades hg A and hg B are essentially distributed in every region of the African continent, and not limited to just southern Africa, even if in different frequencies and sub-clades. As for the mtDNA and Y-DNA frequency charts, they just go to show how heterogeneous southern African KhoiSan can be, particularly more so for their Y-DNA gene pool.

It has been determined that clades like hg E1b1b1 likely came relatively more recently into the preexisting southern African KhoiSan gene pool with the arrival of pastoralist groups from eastern Africa, while hg E1b1a was likely attained from the arrival of Bantu-speaking groups. Such scenarios would suggest that the southern African KhoiSan identity was likely in place prior to the introduction of such markers into their gene pool. Still, there remains that gulf between the TMRCA ages of the mtDNA and that of the deep root Y-DNA clades, and hence the question of when contemporary KhoiSan ethnic identity and their present unique physiology had been shaped. There is certainly no reason, nor evidence, to suggest that they've always looked the way they presently do, even sans gene flow from exotic neighboring populations; but rather, that micro-evolutionary events shaped them into their present cultural and physiological identity as any other known living group today. Let's not forget either, given the heterogeneous composition of southern African gene pool, both maternally and paternally, there is no such thing as a genetically "pristine" tribe, since that diversity would have been attained from gene flow from the then "external" or newly arrived out-group that would have been foreign to the ancestors of KhoiSan who came with the earliest wave of deep-rooted clade(s) introduction into southern Africa. It matters not if this then "foreign" group was another autochthonous African group or from elsewhere. The question of the differentiation of languages of living southern KhoiSan falls into similar lines with their cultural and physiological identity vis-a-vis classical uniparental genetic markers, even if these traits and aspects of them were attained gradually and respectively in chronologically distinct time frames. Even in the case of deep root paternal clades, as is the case with the maternal ones, the southern African KhoiSan share them with other autochthonous African groups across the continent, but especially so in eastern Africa. Such state of affairs when understood, make it intellectually untenable to maintain that "KhoiSans are the oldest surviving tribes". It is more objectively accurate to simply describe them as one of the oldest surviving populations of humanity.

Watch out for possible future updates of this post.

Recommended additional reading: 
An Analysis of "The Dawn of Human Matrilineal Diversity"

—A. G. Morris, 2003, The Myth of the East African 'Bushmen.

—Semino et al., 2002, Ethiopians and Khoisan Share the Deepest Clades of the Human Y-Chromosome Phylogeny.

—Schuster et al., 2010, Complete Khoisan and Bantu genomes from southern Africa. 

—Norton et al., 2006, Genetic Evidence for the Convergent Evolution of Light Skin in Europeans and East Asian.

—Gonder et al., 2006, Whole-mtDNA Genome Sequence Analysis of Ancient African Lineages.

—Behar et al., 2008, The Dawn of Human Matrilineal Diversity. 

—Tishkoff et al., 2007, History of Click-Speaking Populations of Africa Inferred from mtDNA and Y Chromosome Genetic Variation.

—Chiaronia et al., 2009, hg A-M91 and hg B-M60 maps of Africa.


Too Tall Jones said...

Excellent roundup, causing me to rethink the standard mantras as regards the KhoiSan. Keep on posting this valuable data.

Mystery Solver said...

I'm glad you enjoy stuff that is posted here. As long as time permits it, I'll keep sharing information here with all those who care to listen.