Grotte Chauvet’s Aurignacian dates strongly questioned

The famous rock art of the Cave of Lions (Grotte Chauvet, Ardèche) seems now not to be of such an early date as was claimed by Valladas et al. in 2001 but rather from the Gravettian and Solutrean periods, with more solid dates between 26,000 to 18,000 BP.

Jean Combier & Guy Jouve, New investigations into the cultural and stylistic identity of the Chauvet cave and its radiocarbon dating. L’Anthropologie 2014. Pay per view → LINK [doi:10.1016/j.anthro.2013.12.001]

Abstract


The discovery of Chauvet cave, at Vallon-Pont-d’Arc (Ardèche), in 1994, was an important event for our knowledge of palaeolithic parietal art as a whole. Its painted and engraved figures, thanks to their number (425 graphic units), and their excellent state of preservation, provide a documentary thesaurus comparable to that of the greatest sites known, and far beyond what had already been found in the group of Rhône valley caves (Ardèche and Gard). But its study – when one places it in its natural regional, cultural and thematic framework – makes it impossible to see it as an isolated entity of astonishing precocity. This needs to be reconsidered, and the affinities that our research has brought to light are clearly incompatible with the very early age which has been attributed to it. And if one extends this examination to the whole of the Franco-Cantabrian domain, the conclusion is inescapable: although Chauvet cave displays some unique characteristics (like every decorated cave), it belongs to an evolved phase of parietal art that is far removed from the motifs of its origins (known from art on blocks and on shelter walls dated by stratigraphy to the Aurignacian, in France and Cantabrian Spain). The majority of its works are therefore to be placed, quite normally, within the framework of the well-defined artistic creations of the Gravettian and Solutrean. Moreover, this phase of the Middle Upper Palaeolithic (26,000–18,000) coincides with a particularly intensive and diversified local human occupation, unknown in earlier periods and far less dense afterwards in the Magdalenian. A detailed critique of the treatment of the samples subjected to AMS radiocarbon dating makes it impossible to retain the very early age (36,000 cal BP) attributed by some authors to the painted and engraved figures of Chauvet cave.

Other sources: Global Palaeo News, Pileta de Prehistoria[es/en], Paleoantropología Hoy[es].

 

Oldest Okinawan Paleolithic evidence of human presence

A human tooth accompanying a hoard of modified shells shaped as tools have been found in the Sakitari-do cave (Nanjo, Okinawa). They are dated to c. 20-23,000 years ago. They seem to be the first known evidence of human presence in the East Asian archipelago.

The Sakitari-do cave is just 1.5 km away from where the Minatogawa human remains were found, which are however of a somewhat more recent date (c. 18-16 Ka ago). 

Minatogawa 1 (source)

Source: The Asahi Shimbun (via Pileta).

 

The Mal’ta aDNA findings

The recent sequencing of ancient DNA from the remains of a Central Siberian young boy, corresponding to the Gravettian site of Mal’ta, West of Lake Baikal, dated to c. 24,000 years calBP, has caught the interest of many anthropology enthusiasts. During my hiatus of more than two months, most people who asked me to retake blogging with an specific request, talked of these findings. Let’s see:

Maanasa Raghavan et al., Upper Palaeolithic Siberian genome reveals dual ancestry of Native Americans. Nature 2013. Pay per view → LINK [doi:10.1038/nature12736]

Abstract

The origins of the First Americans remain contentious. Although Native Americans seem to be genetically most closely related to east Asians^{1, 2, 3}, there is no consensus with regard to which specific Old World populations they are closest to^{4, 5, 6, 7, 8}. Here we sequence the draft genome of an approximately 24,000-year-old individual (MA-1), from Mal’ta in south-central Siberia⁹, to an average depth of 1×. To our knowledge this is the oldest anatomically modern human genome reported to date. The MA-1 mitochondrial genome belongs to haplogroup U, which has also been found at high frequency among Upper Palaeolithic and Mesolithic European hunter-gatherers^{10, 11, 12}, and the Y chromosome of MA-1 is basal to modern-day western Eurasians and near the root of most Native American lineages⁵. Similarly, we find autosomal evidence that MA-1 is basal to modern-day western Eurasians and genetically closely related to modern-day Native Americans, with no close affinity to east Asians. This suggests that populations related to contemporary western Eurasians had a more north-easterly distribution 24,000 years ago than commonly thought. Furthermore, we estimate that 14 to 38% of Native American ancestry may originate through gene flow from this ancient population. This is likely to have occurred after the divergence of Native American ancestors from east Asian ancestors, but before the diversification of Native American populations in the New World. Gene flow from the MA-1 lineage into Native American ancestors could explain why several crania from the First Americans have been reported as bearing morphological characteristics that do not resemble those of east Asians^{2, 13}. Sequencing of another south-central Siberian, Afontova Gora-2 dating to approximately 17,000 years ago¹⁴, revealed similar autosomal genetic signatures as MA-1, suggesting that the region was continuously occupied by humans throughout the Last Glacial Maximum. Our findings reveal that western Eurasian genetic signatures in modern-day Native Americans derive not only from post-Columbian admixture, as commonly thought, but also from a mixed ancestry of the First Americans.

Haploid lineages

The Mal’ta boy, MA-1, carried distinct yDNA R* and mtDNA U* lineages. While both are clearly related to those dominant in Europe and parts of Asia (West, South) nowadays, they are also distinct from any specific dominant lineage today.

R* (yDNA) is neither R1 nor R2 but another distinct branch of R. This kind of R(xR1, R2) is most rare today and found mostly in and around NW South Asia. Following Wikipedia, this “other R” is found in:

• 10.3% among the Burusho
• 6.8% among the Kalash
• 3.4% among the Gujarati

However I must say that I recall from old discussions that some R(xR1) is also found among Mongols and some North American Natives. I would have to find the relevant studies though (maybe in an update).

U* (mtDNA) is also quite rare today but has been found in Swabian Magdalenian hunter-gatherers, as well as in some Neolithic samples, although it may well be a totally different kind of U* (I could not discern the specific markers in the paper nor the supplementary materials and it must be reminded that the asterisk only means “others”).

Autosomal DNA

The study also shows some statistical inferences from the autosomal (or nuclear) DNA of the Mal’ta boy:

Figure 1 [b & c] b, PCA (PC1 versus PC2) of MA-1 and worldwide human populations for which genomic tracts from recent European admixture in American and Siberian populations have been excluded19. c, Heat map of the statistic f3(Yoruba; MA-1, X) where X is one of 147 worldwide non-African populations (standard errors shown in Supplementary Fig. 21). The graded heat key represents the magnitude of the computed f3 statistics.

Here we can appreciate that MA-1 is closest to Native Americans but still rather intermediate between them and South and West Eurasians. Interestingly East Asians are quite distant instead, suggesting that MA-1 was still not too much admixed with that continental population, unlike what happens with Native Americans, who are essentially East Asian in the autosomal and mtDNA aspects. So this kid appears to be some sort of a “missing link” in the Paleolithic ethnogenesis of Native Americans.

Figure 2 | Admixture graph for MA-1 and 16 complete genomes. An admixture graph with two migration edges (depicted by arrows) was fitted using TreeMix21 to relate MA-1 to 11 modern genomes from worldwide populations22, 4 modern genomes produced in this study (Avar, Mari, Indian and Tajik), and the Denisova genome22. Trees without migration, graphs with different number of migration edges, and residual matrices are shown in Supplementary Information, section 11. The drift parameter is proportional to 2Ne generations, whereNe is the effective population size. The migration weight represents the fraction of ancestry derived from the migration edge. The scale bar shows ten times the average standard error (s.e.) of the entries in the sample covariance matrix. Note that the length of the branch leading toMA-1 is affected by this ancient genome being represented by haploid genotypes.

Even if I am not too keen of TreeMix, in this case the results seem consistent.

We can appreciate here that a sample of Native Americans (the Karitiana, maybe not as “pure” as the Xavantes but still very much so) show up in a different branch from MA-1, reflecting their overwhelmingly East Asian ancestry, mostly by the maternal side (mtDNA). MA-1 instead hangs from the South-West Eurasian branch, soon after the split between South Asians and West Eurasians. Both have extremely drifted branches, surely indicating the small size of their founder populations, typical of the Far North. 

In addition to this basic tree, two admixture events are signaled: one is the already known Denisovan (H. erectus?) weak one into Australasian Natives (represented by Papuans) and the other one, quite more intense, is the one hanging from upstream of MA-1 to Native Americans (Karitiana), reflecting the partial South-West Eurasian ancestry of Native Americans (noticeable also in their dominant paternal ancestry: haplogroup Q). 

The fact that the admixture signal stems from quite upstream of MA-1 indicates that this boy (or rather his relatives) were not direct ancestors of Native Americans in any significant way but rather a different branch from the same trunk. Probably proto-Amerindians were already in this period at the North Pacific coasts, not sure if in Beringia or around Okhotsk or what but certainly they had already separated from the Mal’ta population.

What did we know of Native American genesis before this finding?

There are three principal lines of evidence:

1. Y-DNA, which among Native Americans is essentially haplogroup Q (plus some C3, which is from NE Asia). By phylogenetically hierarchical diversity, haplogroup Q must have coalesced in West or Central Asia (or maybe South Asia?), very possibly in or near Iran. The NE Asian and Native American branches are clearly derived, even if more important numerically today.
2. mtDNA, which among Native Americans is essentially from NE Asia (A, C, D), middle East Asia (B) but also in a small amount from West Asia (X2). 
3. Archaeology: we can track, more or less directly, the proto-NAs by means of following the Upper Paleolithic sequence in Siberia and nearby areas. 
1. C. 47,000 years ago (calBP) H. sapiens with Aurignacoid technology (i.e. linked to West Eurasian earliest Upper Paleolithic) reached Altai, displacing the Neanderthals to the Northern fringes of the district.
2. C. 30,000 years ago, Upper Paleolithic (“mode 4”) technology with roots in Altai reached other parts of Siberia, Mongolia and North China, from where it expanded eastwards and southwards gradually in a process of, probably, cultural diffusion. 
3. By c. 17,000 years ago they were already in North America and c. 15,000 years ago in South America. In the LGM they were probably in Beringia already (but this is only indirectly attested so far). 

So we already had a good idea about the origins of Native Americans: their ultimate roots, at least patrilineally, seem to be in Altai (where they were part of the wider West Eurasian colonization at the expense of Neanderthals with Aurignacian-like technology and dogs). Then, probably around 30,000 years ago they expanded eastwards through Siberia and maybe nearby areas, entering in intense and intimate contact with the already existent East Asian populations, with whom they admixed once and again, mostly by the female side. 

It would seem therefore that their society was already patrilocal because otherwise their patrilineages would have just got dissolved among the locals and would have never reached Beringia nor America in such dominant position.

Overall this is the quite clear notion that I have on Native American earliest genesis and for me there is no reasonable doubt about this narrative (except maybe in the fine details). However I must reckon that some individuals have reacted very negatively against it. But no matter how much they yell, I fail to see their arguments. 

How does this new finding affects this narrative?

It simply confirms it with further evidence. By 24,000 calBP the proto-NAs were surely already, as I said before, in NE Asia close to the Pacific coasts, so this Mal’ta population is a branch left behind in their migration (plus whatever new inflows from the West, which we can’t evaluate). The very low affinity level with East Asians, in spite of its quite Eastern location, shows that early East Asians had not yet reached, at least in significant numbers, so far North. If they had, they probably did only at more eastern longitudes, probably near the sea, where resources were more plentiful.

In other words: the first Central Siberians were of South+West Eurasian stock and the current East Asian genetic and phenotype hegemony in that area reflects post-LGM flows, mostly lead by yDNA N1. 

Early Native Americans were the product of admixture of these earliest Siberians with NE Asians, admixture that surely happened East of Lake Baikal, although the exact details are still unclear. 

What does MA-1 say about the West?

His mtDNA is generally consistent with other common U-derived lineages found in West Eurasian Upper Paleolithic, so not much other than he was somehow related, what is confirmed by autosomal analysis. 

His yDNA is more interesting maybe, nonetheless because it is probably the oldest sequence of this kind but also because it belongs to haplogroup R. It certainly discards whatever “molecular clock” guesstimates for R that are shorter than this site’s age but on its own it is not able to set a real age other than a bare minimum. 

So for example Eupedia‘s estimate of 29 Ka for R as such could still be valid, although I would say that extremely unlikely. 

Indirectly however it does say something by confirming the overall narrative of Native American origins as above and that means that Eupedia’s estimate of a mere 24 Ka age for haplogroup Q is almost certainly wrong by a lot. 

Using that tree, we would have to at least double the age of Q in order to fit with the Altai narrative (which begins at c. 47 Ka ago), what, extrapolating, implies an age for R of at least 58 Ka. I have estimated some 48 Ka of age for R1 and 68 Ka for P, so it makes good sense after these so necessary corrections. The exact ages we may never know but the approximate ages should be something like these. 

And that’s about all I can say. More in comments (and/or updates) if need be.

Update (Dec 6): R* and P* (and other rare clades) among Central Asians

A reader sent me copy of the study by Wei-Hua Shou et al. (2010) titled Y-chromosome distributions among populations in Northwest China identify significant contribution from Central Asian pastoralists and lesser influence of western Eurasians, published by Nature (doi:10.1038/jhg.2010.30).

While it is not the bit of info I was recalling above, it does add some information about unmistakable R(xR1,R2) and P(xQ,R) among Central Asian populations (from P.R. China territory). In detail:

• R* is found in 5/31 Tayiks, 1/41 Kazakhs and 1/50 Uyghurs.
• P* is found in 1/31 Tayiks and 1/43 Kirgizes. 

Also of interest should be the presence of:

• Q(xQ1) in  8/35 Dongxiang (a Mongol ethnicity), 1/45 Kirgizes and 1/50 Tu (another Mongol ethnicity).
• F(xG,H,I,J,K) in 2/32 Yugu (Yugurs, a distinct Uyghur sub-ethnicity), 2/41 Kazakh, 1/31 Tayiks and 1/50 Tu.
• K(xN,O,P) in  32/533 total (i.e. 6% in Easternmost Central Asia), among which are most notable: 9/50 Uyghurs, 6/23 Uzbeks, 6/27 Bao’an (another small Mongol ethnicity), 3/32 Xibo (a Tungusic ethnicity), 2/32 Yugu and 2/5 Mongols. I guess that it is possible that this is a distinct K subclade, although it can well be either part of MNOPS (NO*?) or also belong to LT (L?).
• R2 in 1/31 Tayiks and 2/27 Bao’an.

 

Echoes from the past (August-28-2013)

Oh, yeah, I admit it: I have been procrastinating a lot. Result: an extremely long “to do” list. Naturally, I can’t make up for all the past laziness, so here goes a synthesis of what would otherwise be left unattended, take your time, please. 

Middle Paleolithic:

Atapuerca holds “uninterrupted” sequence of European humans since 1.2 million years ago. Soon-to-be-published theory of continuity from H. erectus to Neanderthals in the Castilian site → Paleorama[es], EFE Futuro[es].

More Neanderthal evidence for symbolism found in Fumane cave (Veneto, Italy): polished and ochre-painted shells (pictured) → PLoS ONE (open access), El neandertal tonto ¡qué timo![es].

Upper Paleolithic:

Epigravettian burial, dated to ~14,000 BP, found in Cuges-les-Pins (Provence). The Epigravettian (and not the more widespread Magdalenian) culture of this site indicates a direct link to Italy → INRAP[fr], La Provence[fr], Los Andes[es].

Oldest modern human remain of Galicia found at Valdavara cave (Becerreá, Lugo province). The milk tooth is 17,000 years old, 7000 years older than any other such finding in the NW Iberian country → Pileta[es], IPHES[cat].

Epipaleolithic:

Thousands of engravings, dated to c. 6000 years ago, found in Coahuila (Mexico) → RTVE[es].

“World’s oldest calendar” found in Scotland → BBC.

Female burial found at Poças de São Bento (Sado basin, Portugal) → Paleorama[es].

Neolithic:

First farmers were inbred (at least in Southern Jordan) → Science Magazine.

Qatar Neolithic dig shows the peninsula was in contact with early Sumerian civilization (Eridu or Ubaid period, the first empire ever) → The Archaeology News Network.

Manure was already used by Europe’s first farmers → Science.

Chalcolithic:

Haryana (India) town is oldest large IVC settlement → Business Standard.

East China engravings show first Oriental writing (~5000 years’ old, just slightly younger aged as Sumerian cuneiform writing but much more recent than the controversial Tărtăria tablets of Bulgaria) → The Guardian, English People.

North American natives caused lead pollution in Lake Michigan (oldest recorded) → PPV paper (ER&T),  University of Pittsburg.

Perdigões enclosure and collective burial was pilgrimage center. Antonio Valera (often so scholarly cryptic at his blog) loosens up when interviewed by a Portuguese publication, giving meaning to the archeology he’s working with → Super Interesante[por].

Bronze Age:

Cypriot harbor city dug: Hala Sultan Tekke, near modern Larnaka, had 25-50 Ha. and was active between the 16th and 12th centuries BCE → The Archaeology News Network.

Also from Cyprus: large settlement dug out near Nicosia (Cyprus), dated to 2000-1500 BCE → The Archaeology News Network.

Human evolution:

Modern human shoulder much more efficient than chimpanzees’ at throwing… but also than H. erectus’ → BBC.

Neanderthals did speak (of course) → Science Daily, open access paper (Frontiers in Psychology).

Note: their unfounded insistence on most unlikely H. heidelbergensis shared origins of Neanderthals and us casts some doubt on elements of their reasoning however.

Genetics:

Record ancient DNA: ~700,000 years’ old horse sequenced → Nature Communications (PPV).

Experimental archaeology:

How to carve your own stone tools and weapons out of modern materials: very interesting videos (in English) at Paleorama[es] (scroll down). 

More tomorrow (nope my “to do” list is not at all finished with this entry).

 

Indian Microlithic industry almost contemporary of Western initial UP and LSA

Mehtakheri toolkit

That is what a new study has found, albeit on just one date. Based on that they argue that the recent claim by Mellars et al. (see also here) about an extremely late date for the migration out of Africa (OOA) becomes more plausible.

Sheila Mishra et al., Continuity of Microblade Technology in the Indian Subcontinent Since 45 ka: Implications for the Dispersal of Modern Humans. PLoS ONE 2013. Open access → LINK [doi:10.1371/journal.pone.0069280]

However considering the pivotal role played by South Asia in the genetics of Humankind after the OOA it is still impossible that this microlithic industry corresponds with that process, because the migration and successive Eurasian expansion must:

1. Have minimal dates of well before 60-55 Ka ago, time when the presence of H. sapiens becomes undeniable from Palestine to SE Asia and Australia. 
2. Go at least largely through South Asia; because the distribution and basal diversity of mtDNA M and R, as well Y-DNA F demand it without any reasonable alternatives. 

The authors themselves acknowledge that the finding is inconclusive in this debate but they choose to lean for a revised Mellars-style interpretation on their own subjectivity.

Their hypothesis is not exactly like Mellars et al. These proposed an extremely late (c. 40-35 Ka BP) OoA, which would imply also extremely late colonization of East Asia and Australasia by Homo sapiens (via South Asia). In order to “explain” the lack East Asian blade-like technologies (necessary for the old professor’s ideas about “modern human behavior”) they proposed that the Eastern colonization was led by small populations who somehow lost the technology. But well, as I discussed back in the day, the hypothesis does not stand.

Mishra’s revised hypothesis is somewhat more coherent (but still very unlikely): she proposes that East Asia and Australia were actually colonized with Middle Paleolithic technology (neither blades nor microblades) in the time demanded by archaeological data and that South Asia instead was not colonized by our species until c. 45,000 BP, possibly because there was some kind of intelligent archaic hominin (Hathnora?), which blocked the expansion of our species initially.

However the hypothesis is still plagued by problems:

1. As I said above, any model that dictates that South Asia was not central to the expansion of Homo sapiens in Eurasia and surroundings must be wrong: genetics demand otherwise. A settlement of South Asia that is posterior to that of East Asia, Papua and/or West Eurasia (other than the initial Arabian trailblazers or boaters) simply does not make any sense.
2. The African microblade technology is still quite older (70-60 Ka BP) than the South Asian findings and the similitude may well be a mirage or a matter of convergent evolution. Not the only time that people reinvent the same thing separated by time and space: look for example at Neolithic, which was developed at least in four separate regions of the World, maybe more; or look at the Solutrean style of retouch, used in many different Paleolithic cultures separated by time and space (Africa, Europe, America, etc.)
3. It would require that Homo sapiens would travel through Altai and all the evidence in this North Asian keystone region, a necessary corridor for transcontinental travel before the domestication of camels (or at the very least horses), indicates that it was inhabited by “archaic” hominids (Neanderthals, H. erectus/Denisovans) until c. 47 Ka BP, when industries related to those of West Asia and Europe show up (at later dates associated to H. sapiens remains).

The facts:

A C14 date was obtained for the site of Mehtakheri (near Barwah, Nimar region, Madhya Pradesh) annotated as: >42,900 BP, > 46,555 calBP, >45,028 – 48,081 (68% CI range for the calBP date). Another C14 date from the same site is much more recent (34,380 ± 991 calBP).

They also obtained five of OSL dates for section 2 ranging from 41.6(±3.3) to 47.0(±4.9) Ka ago. Another date for this unit of 55.5(±5.8) was not used by the authors because it corresponds to an unstudied layer.

Section 3 has older dates (65-78 Ka) but it corresponds to the Middle Paleolithic.

The microlithic industry seems to continue in South Asia until the Iron Age, suggesting that Neolithic and later developments did not substantially alter the demography of the subcontinent. 

All this is very informative but the conclusions suggested don’t seem to make any sense. It is much more logical to infer that H. sapiens left Africa with an MSA-like Middle Paleolithic toolkit that was not related to the Nubian culture (the dead horse being beaten once and again by both Mellars and Mishra) but to other ill-defined groups of possible South African affinity (as claimed by Petraglia). Insisting on the Nubian techno-complex, when we do not know it reaching beyond Dhofar (i.e. they did not reach the Persian Gulf “oasis”, unlike Petraglia’s trailblazers or Armitage’s Jebel Faya findings) is taking the part for the whole, as if there was not already a much more widespread and diverse African Middle Paleolithic (MSA, Lupenbiense, Aterian) in those times already.

Instead these data may indicate a relation of some sort with West Eurasian Upper Paleolithic and African Late Stone Age, which are of roughly those dates. This tentative relationship does not imply migration but would just need some cultural contact. 

It would be interesting to know more about the MP-UP transition in the area around Arabia Peninsula in order to develop better theories on this tripartite interaction between the West Eurasian early UP, the African early LSA and the South Asian early microlithic industry. These very possible cultural interactions fit well within the wet phase of the Mousterian Pluvial (c. 50-30 Ka ago).

Update (Jul 11): “microliths” that are not microliths

I just looked for the first time at the technical issue of what is a microlith (~1 cm long, ~0.5 cm wide) and the published toolkits only seem to have one microlith senso stricto: the J4 point. All the rest have lengths of 2 cm or larger, often 5 cm or more.

The presence of some microlith-sized pieces (usually points) in early UP cultures is almost standard: Emirian, Chatelperronian, Aurignacian and Gravettian all them have occasional “microliths” (measured by size) an in all cases these are points, exactly as happens in Mehtakheri.

So these toolkits seem to have more relationship, if anything, with Western Eurasian early UP ones, which are roughly contemporary (Emirian is the only clearly older one).

Furthermore, archaeologist Millán Mozota sees even similitudes with Mousterian flaking style (see comments):

Bladelet flaking is a typical flaking strategy for this blank type
(small pebbles). Specially if the raw material itself is of good enough
quality.

It has been documented, for high quality quartz on
Mousterian sites, like in Grotte Breuil and, if i recall correctly,
other sites in that area of the Italian Peninsula.

Being also puzzled because the inventories described suggest a strong blade/bladelet component, instead of microblades. 

 

Mellars challenges the ‘early out of Africa’ model

I do not have yet access to this potentially key paper, so first of all I want to make an appeal here to share a copy with me (→ email address). Thanks in advance. Update: got it (thanks to all who shared, you people are just great!) I will review it again as soon as possible.

Update (Jun 18): complementary review of the full paper now available here.

Paul Mellars et al., Genetic and archaeological perspectives on the initial modern human colonization of southern Asia. PNAS 2013. Pay per view (6-month embargo) → LINK [doi:10.1073/pnas.1306043110]

Abstract

It has been argued recently that the initial dispersal of anatomically modern humans from Africa to southern Asia occurred before the volcanic “supereruption” of the Mount Toba volcano (Sumatra) at ∼74,000 y before present (B.P.)—possibly as early as 120,000 y B.P. We show here that this “pre-Toba” dispersal model is in serious conflict with both the most recent genetic evidence from both Africa and Asia and the archaeological evidence from South Asian sites. We present an alternative model based on a combination of genetic analyses and recent archaeological evidence from South Asia and Africa. These data support a coastally oriented dispersal of modern humans from eastern Africa to southern Asia ∼60–50 thousand years ago (ka). This was associated with distinctively African microlithic and “backed-segment” technologies analogous to the African “Howiesons Poort” and related technologies, together with a range of distinctively “modern” cultural and symbolic features (highly shaped bone tools, personal ornaments, abstract artistic motifs, microblade technology, etc.), similar to those that accompanied the replacement of “archaic” Neanderthal by anatomically modern human populations in other regions of western Eurasia at a broadly similar date.

A review has been published at Live Science.

South Asian artifacts from ~30-50 Ka BP.

By “genetic evidence” they obviously mean “molecular clock” nonsense, so it is not evidence at all but mere speculation. However I am indeed very interested in knowing in detail what they mean by “archaeological evidence”, because they seem to get into direct confrontation with much accumulated evidence, first and foremost all of Petraglia’s research in both India and Arabia but also with the quite strong evidence for pre-60 Ka human presence in Australia and growing evidence for pre-60 Ka modern humans in SE Asia (in some cases even as old as 100 Ka). 

It must be said that Paul Mellars has been criticized before a lot for several reasons but very especially for his adherence to the quite speculative “modern human behavior” conjecture and, relatedly, bigotric attitudes against Neanderthal intellectual capabilities, based on nothing too solid. Therefore I’m generally skeptic about what Mellars has to say on this matter because this kind of conclusion is what one would expect from him. 

However Mellars is certainly a distinguished academic and, even if prejudiced and stuck to his own old-school and somewhat Eurocentric interpretations, he knows his trade as archaeologist and prehistorian. So he may be onto something, even if it is not exactly what he wants us to believe. 

For example, it is not impossible that this research may have, unbeknown to the authors, found evidence of a secondary OoA wave (maybe related to the spread of Y-DNA D and mtDNA N?) or even a distinctive evolution in Southern Asian technology prior to the expansion of Western Eurasia. 

It is interesting that they suggest that the 80-60/50 Ka toolkits of India would have been made by Neanderthals, when they are not describing them at all as Mousterian, the almost exclusively Neanderthal techno-culture, or Mousterian-related.

I have some difficulties judging before reading the whole study. However the supplemental material (quite extensive) is freely accessible and for what I can see there:

1. They dedicate much text to attempt to justify a particular version of mainstream “molecular clock” hypothesis, which are clearly broke in my understanding. The kind of arguments “rebated” are more or less what I have been putting forward since many years ago. Ironically their “molecular clock” estimates make N and R much older than M, what I absolutely oppose (just count mutations downstream of the L3 node).
2. No real attention is given instead to the geographical structure/distribution of major mtDNA haplogroups, only mentioned in relation to “molecular clock” speculations.
3. The criticism of the African affinity of the Jwalapuram (Jurreru Valley) cores (Petraglia 2007) focuses on dismissal of any possibility of comparison, rather than on alternative comparisons. 
4. Another “criticism” is that there is no apparent connection between Jwalapuram and the Nubian Complex (why there should be any?, it is not the only East African techno-culture, nor the only group that shows indications of traveling to Arabia in the Abbassia Pluvial).
5. Also it is “criticized” that the most comparable African culture, Howiesons Poort) is not recorded before c. 71 Ka BP (what IMO may indicate late cultural dispersals to Southern Africa from East Africa, for example, but, hey!, Mellars is fencing off balls like crazy at his conservative goal). 
6. They find clear similitudes between Indian and African microlithic industries (apparently related to the development of “mode 4” in both areas, as well as in West Eurasia). Indian industries are dated to c. 38-40 Ka BP, while African ones are dated to c. 49 Ka BP (Kenya) or later. However West Eurasian ones have dates as old as 55 Ka BP (not for Mellars, who remains stuck in older date references which he describes as ∼40–45 ka [calibrated (cal.) before present (B.P.)]), what really suggest that we are talking here not of the “out of Africa” but of the West Eurasian colonization process (necessarily from further into Asia, genetic phylo-geographic structure demands) with offshoots to the nearby regions. 
7. Another element of late Africa-India “similitude” they find is “the remarkable, double bounded criss-cross design incised on ostrich eggshell”, dated in India (Patne) to at least ∼30 ka (cal. B.P.), much earlier in South Africa. For Mellars this is beyond the range of either pure coincidence or entirely independent and remarkably convergent cultural evolutionary processes. Hmmm, really? Or are we before a clear case of wishful thinking as happens with the Solutrean-Clovis relationship hypothesis? Isn’t it 30 Ka BP anyhow well beyond any reasonable expectations for the OoA time frame, including Mellar’s own conjectures?
8. Mellars accepts the paradox that the geographical limits of these highly distinctive microblade and geometric microlithic technologies are confined to the Indian subcontinent, with no currently documented traces of these technologies in regions farther to the east. And then makes up excuses for it, such as biological and cultural bottlenecks caused by “founder effects”, mysteriously leading to a loss or simplification of cultural and technological know-how, as well as fininding new and contrasting environments (in the same latitudes?!)
9. Even in the case of Arabian colonization, Mellars shows to be in a very defensive attitude, admitting only to the reality of the Palestinian sites with clearly modern skulls, as well as to the area of Nubian Complex colonization (on whose peculiarities he insists a lot, as if it would be the only expression of the wider MSA techno-complex), disdaining all the other MSA colonization areas and, often ill-defined, variants.

In brief, for what I could see in the supplemental material, along with some potentially interesting references to the relative cultural community spanning from East Africa to South Asia at the time of emergence of “mode 4” industries, it seems that Mellars and allies are essentially putting the cart (their models) before the horses (the facts), what is bad science. 

In 2008, Zilhao and d’Errico angrily accused Mellars of being an obsolete armchair prehistorian (different words maybe, same idea). Back in the day I was tempted to support Mellars but nowadays I must agree that he is clearly stuck in a one-sided interpretation of prehistory whose time is long gone. Whatever the case I welcome the debate and can only hope that will help to produce even more evidence to further clarify the actual facts of the Prehistory of Humankind.

 

South Arabian genetic refugium

This is not about the L(xM,N) lineages but about the Eurasian ones like R0a or R2.

Jeffrey I. Rose et al., Tabula rasa or refugia? Using genetic data to assess the peopling of Arabia. Arabian Archaeology and Epigraphy, 2013. Pay per view → LINK [doi:10.1111/aae.12017]

Abstract


This paper provides a broad overview of the current state of archaeogenetic research in Arabia. We summarise recent studies of mitochondrial DNA and lactase persistence allele -13915*G in order to reconstruct the population histories of modern Arabs. These data, in turn, enable us to assess different scenarios for the peopling of the Peninsula over the course of the Late Pleistocene and Early Holocene. The evidence supports the posited existence of Arabian refugia, although it is inconclusive which (e.g. Persian Gulf basin, Yemeni highlands and/or Red Sea basin) was/were responsible for housing ancestral populations during the Last Glacial Maximum. Synthesising genetic and archaeological data sets, we conclude that a substantial portion of the present South Arabian gene pool derives from a deeply rooted population that underwent significant internal growth within Arabia some 12,000 years ago. At the same time, we interpret the disappearance of Nejd Leptolithic archaeological sites in southern Arabia around 8000 years ago to represent the termination of a significant component of the Pleistocene gene pool.

Rose uploaded the full paper at Academia.edu. Very much worth a careful read because it is a rare case of paleogenetics being done by a researcher who is primarily an archaeologist and who knows well the material Prehistory of which he’s talking about, at all moments seeking to reconcile archaeological and genetic evidence and not, as way too often happens, creating genetic-only models with absolutely no material foundations and unavoidably clashing with prehistoric reality.