December 15, 2013

Arabian origin of the Upper Paleolithic in the Levant

This is a very useful review of research on the origin of the Upper Paleolithic (Emiran) in the Levant, arguing against a recent (c. 50kya ) African origin and in favor of an Arabian one. The argument is mainly archaeological, although it is informed by genetic evidence. From the chapter:
After a century of research, the origins of the Levantine UP still remain an enigma. At this point, at least one thing is clear: the Emiran has no African progenitor. As such, there is a disconnect between the archaeological database and the Replacement paradigm, which necessitates that the earliest Levantine Upper Paleolithic must have come fully developed from northeast Africa. The Replacement model should have been a parsimonious prism through which to view the transition from the MP to the UP in the Levant. It was not.
The recent acceptance of: (i) a slower autosomal mutation rate, and (ii) evidence for interbreeding with Neandertals largely predating the c. 50kya mark, and (iii) coalescence of Eurasian mtDNA haplogroup N well before that time, have all but killed, in my opinion the idea of a 50kya spread of modern humans from Africa. Modern humans must have lived in Eurasia much earlier than that time, and what remains is to figure out how much earlier.

A century of research into the origins of the Upper Palaeolithic in the Levant

Anthony E. Marks and Jeffrey I. Rose

Link

14 comments:

Maju said...

"... we suggest a local sequence where the Emiran develops, primarily, out of a Tabun D technological system in combination with external stimulus from residents of Arabia. A strong wet phase across central and eastern portions of the Peninsula between 55 and 50 ka (...) might have facilitated range expan-sions and demographic exchange in both directions".

All this is quite speculative and ignores the factoid that by that time West Asia and surrounding regions were almost certainly experiencing the most relevant back-migration of Asian H. sapiens from further East, which founded the Western Eurasian macro-population as we know it. It's true that we do not have any other candidates for the origin of the "mode 4" and it is still quite possible that this one evolved in West Asia (the Levant rather than Arabia, I'd say on light of what Marks and Rose discuss) but all this is far from being able to claim that the UP evolved in Arabia.

Also they do not even discuss the LSA, whose earliest ages are very similar to those of Emirian, cf. Ambrose 1998: "the Later Stone Age began substantially earlier than 46,000 years ago".

The question of the exact origins of "mode 4" or UP blade technology therefore I think must remain open, as this paper does not really provide any single clear evidence of where its origins may lay. What seems clear to me in any case is that, regardless of the actual "inventors", its spread was accompanied by very dynamic human expansion from Asia to the West (a very broad area including West Asia, Central Asia, Europe and parts of Africa).

bicicleur said...

I agree with the 1st comment made by Maju
'For thirty years, researchers have failed to identify an African industry that was the source of, or influenced the development of, the Levantine UP. Rather than searching for an external stimulus on the Emiran in Africa, the answer may lie somewhere in Arabia'
why Arabia? I think the arguments in the paper for this are very speculative.

mousterian said...

Let me begin in saying that this paper was originally meant to be published in French, and was supposed to be simply testing the waters ahead of a more rigorous technological analysis of the Emiran in relation to Tabun D, Tabun B, the Taramsan, the Arabian Nubian, and the Mudayyan lithic industries. The editor chose to keep it in English; hence, the chapter has received far more attention that I ever expected. We’re are a bit ahead of ourselves in the conclusion, and so my apologies that it comes off as speculation. At the moment, Tony and I are finishing up the “real” paper that explains this hypothesis in some depth.

The essential point is that, from a technological standpoint, the Arabian Nubian Complex provides the missing link to the Levantine MP-UP transition, exhibited at Boker Tachtit and ‘Ain Difla. In both cases, the traditional Tabun D-type reduction strategy is suddenly replaced by bidirectional preparation - the hallmark of the Arabian Nubian. Moreover, in the past few months, a number of Nubian and late Nubian (Mudayyan) sites have been discovered in central and northern Saudi. Once these data are published, we’ll be able to finalize our paper and present a more complete picture of this proposition. At the very least, I’m happy to have stirred the pot a bit and provided a counterpoint to the absolutely ridiculous “coasting out of Africa” paradigm.

terryt said...

"why Arabia? I think the arguments in the paper for this are very speculative".

Agree the UP may not have actually originated in Arabia, but it is obvious it did not originate in Africa.

"there is a disconnect between the archaeological database and the Replacement paradigm, which necessitates that the earliest Levantine Upper Paleolithic must have come fully developed from northeast Africa. The Replacement model should have been a parsimonious prism through which to view the transition from the MP to the UP in the Levant. It was not".

It is also obvious that so-called 'modern' humans were widespread by the time the Upper Paleolithic expanded. Therefore the presence of UP can no longer be used as a marker for the first presence of modern humans, a position I have long claimed against Maju's belief.

"regardless of the actual 'inventors' its spread was accompanied by very dynamic human expansion from Asia to the West (a very broad area including West Asia, Central Asia, Europe and parts of Africa)".

But by that time humans had long before reached Australia and East Asia.

Fiend of 9 worlds said...

Since India seems to have been inhabited by moderns 100k+ years ago I don't think 50k years ago is even a remote possibility any more, even to the most diehard OoA supporters.

Since peking man is now dated to 750k BC we should be talking about ino africa.

bicicleur said...

'At the moment, Tony and I are finishing up the “real” paper that explains this hypothesis in some depth.'

ok, keep us informed

terryt said...

"At the very least, I’m happy to have stirred the pot a bit and provided a counterpoint to the absolutely ridiculous 'coasting out of Africa' paradigm'.

Thank you. That idea has always seemed totally ridiculous to me. And I have been criticised many times for saying so.

"ok, keep us informed"

Yes, it promises so much.

mousterian said...

@terryt

Thankfully, we seem to be approaching a critical mass and will soon be able to put all this beach combing nonsense to rest. I’ve said this before here, but let me reiterate - I’ve explored a good portion of the South Arabian coastline from Mukalla in Yemen to Ras Al Hadd in Oman, and not once have I seen anything remotely resembling a Middle or Upper Palaeolithic artifact. Plenty of Lower Palaeolithic and Neolithic, but nothing in between. And there is virtually no continental shelf in this area, so the excuse that they’re submerged cannot invoked. On the other hand, the moment you drive up onto the limestone plateau in the South Arabian hinterlands, MP and UP abound, it’s hard to stop the car without finding a site.

@maju

Those dates you reference in Ambrose 1998 are obsidian hydration, a technique that has subsequently been demonstrated to be ineffective. Not to mention that the lithic technology at Enkipune Ya Muto is entirely unrelated to anything found outside of Africa.

I do find the idea of a back migration into western Eurasia and maybe even Africa to be quite appealing. In the Levantine UP, there are two quite distinct traditions - the Ahmarian and the Aurignacian. While we have argued that the Ahmarian is a combination of local development with an influx from Arabia, the Aurignacian is clearly intrusive from somewhere else. Otte et al. 2007 have presented a compelling case for its origins in highland Iran (maybe further east if we ever get to look?), with subsequent dispersal into the Levant sometime after 45 ka BP.

As for East Asia, South Asia, and Australia, clearly AMH have been there well before 50 ka BP and have no bearing on this discussion of the origins of the Emiran. Perhaps I should make it clearer in this next manuscript, that what we’re talking about here is essentially the N-lineage, unrelated with anything further to the east.

Maju said...

Thanks for the clarification re. the dates, Mousterian. Do you know which are the most realistic ones?

In any case I would think logical, considering the overall demic/genetic flow that seems associated to the Early UP that African LSA would be partly derived from West Eurasian UP. However, for the same reason I would expect precursors of the West Eurasian UP in the Indian subcontinent - but AFAIK nothing yet other some a bit imprecise references to "blades" and "bladelets", mostly from rather late dates, which could also be Levallois "blades" (not "mode 4" but "mode 3") unless clarified otherwise.

"... the Aurignacian is clearly intrusive from somewhere else. Otte et al. 2007 have presented a compelling case for its origins in highland Iran (maybe further east if we ever get to look?)"...

It seems to make sense. At least I would think so. It would certainly fit well with the genetic evidence as I understand it.

"... with subsequent dispersal into the Levant sometime after 45 ka BP".

What about Emirian? Emirian is said to be "Aurignacoid" (depends who you read but for Hoffecker, for example, it is at the origin of European (proto-)Aurignacian or close to that origin) and seems dated to c. 55 Ka BP (Taramsa-1, OSL date).

As for the 45 Ka BP, I presume you're using uncalibrated C-14 dates, right? Otherwise the latest dates for Europe and Altai Aurignacian or other Aurignacoid would be earlier than that date. If I'm correct (from memory without using the calibration table) 45 Ka BP would be something like 50 Ka calBP or even a bit earlier, right? I ask just for "syncronizing" chronologies (I really got used to "calBP" as of late because of the difficulties to conciliate the various methods of age estimation otherwise and also because it's supposed to be closer to "real years", although I am well aware of the uncertainty and CI issues).

"Perhaps I should make it clearer in this next manuscript, that what we’re talking about here is essentially the N-lineage, unrelated with anything further to the east".

Perhaps not (I must warn you). MtDNA N's basal diversity is concentrated around the Pacific Ocean and only three minor sublineages of it (out of 16) correspond to the West, plus other 3 more important basal sublineages of R (the main sublineage of N by size), again with most of them (total 18) concentrated further East, in this case more commonly in South Asia but also several in the Far East.

It's quite possible (although it'd require a complex argumentation, more than we can afford in a comments section) that the wave that came to West Eurasia some 50+ Ka ago began (earlier) close to or even in SE Asia. Recent evidence of Papuan Y-DNA lineage being closest to Western/South Asian P support this and in any case the two join with East Asian NO (upstream) and the other Papuan lineage S (at unknown position) as well as with some of the K-other.

My (educated) opinion anyhow but if you wish to discuss it in greater depth, feel free to email me at lialdamiz [at] gmail [dot] com. In any case I would not overemphasize N as in the popular (but obsolete) idea of it being a Western lineage because if anything it should be Eastern by origin, at least considering the much greater basal diversity in that area.

John Rudmin said...

Does the Boker Tachtit transition from Middle to Upper Paleolithic have a significance in this?

And as for assertions that there is no technological infusion from Africa at the start of the U.P., I am always left wondering about the Aterian. Doesn't the Aterian predate 50kBP? Were they just really isolated and didn't export any methods?

eurologist said...

Mousterian,

I think many readers here have two severe problems with the interpretation as presented:

(1) There is no precedent of a 50,000+ years cultural continuitiy here nor nearby, during this time. This argument is strengthened by:

(2) There is no evidence that AMHs could survive the extreme draught, and at times rather cold conditions between ~100kya and 50kya from Arabia to Iraq / Iran.

A homeland for West Eurasians should instead be sought in the extreme NW Indian subcontinent, i.e., the generally mild climate isolate of E Afghanistan and NW Pakistan.

terryt said...

"MtDNA N's basal diversity is concentrated around the Pacific Ocean and only three minor sublineages of it (out of 16) correspond to the West, plus other 3 more important basal sublineages of R (the main sublineage of N by size), again with most of them (total 18) concentrated further East, in this case more commonly in South Asia but also several in the Far East".

Using that logic N originated in Australia. It has 4 basal N haplogroups, more than any other single region. And you still have to explain how N made it all the way to SE Asia without leaving any trace of its path through South Asia. There are no exclusively South Asian N haplogroups. I agree that R looks SE Asian as each island from Australia to the Malay Peninsula basically has its own basal R haplogroup. And overall the haplogroup is spread continuously from Australia to Western Europe whereas N breaks up very definitely into a western and an eastern group.

"It's quite possible (although it'd require a complex argumentation, more than we can afford in a comments section) that the wave that came to West Eurasia some 50+ Ka ago began (earlier) close to or even in SE Asia".

I agree completely with that although it only applies to mt-DNA R as it is most likely that the western N haplogroups are a carry over from the original OoA.

"There is no evidence that AMHs could survive the extreme draught, and at times rather cold conditions between ~100kya and 50kya from Arabia to Iraq / Iran".

As far as I'm aware there has always been small pockets where humans could survive quite easily such as the Tigris/Euphrates valley and around the Caspian Sea in parts of Anatolia. The fact that the western N haplogroups seem to have been subject to a long period of isolation supports their survival in some region(s).

terryt said...

"It's quite possible ... that the wave that came to West Eurasia some 50+ Ka ago began (earlier) close to or even in SE Asia".

I've just realised that's an amazing comment coming from you, considering the vehemence with which you attacked me when I first suggested it some years ago. To me the data has always shown such a scenario as being extremely likely, but you were determined to have everything originating in South Asia.

Unknown said...

"In any case I would think logical, considering the overall demic/genetic flow that seems associated to the Early UP that African LSA would be partly derived from West Eurasian UP."

Full UP in Southern Africa although perhaps not the source of Eurasian UP, does seem too early beginning and far to be derived from it

Border Cave and the beginning of the Later Stone Age in South Africa
Paola Villa et al.

http://www.pnas.org/content/109/33/13208.full
.pdf

"Data from Border Cave (KwaZulu-Natal) show a strong pattern of technological change at approximately 44–42 ka cal BP, marked by adoption of techniques and materials that were present but scarcely used in the previous MSA, and some novelties. The agent of change was neither a revolution nor the advent of a new species of human. Although most evident in personal ornaments and symbolic markings, the change from one way of living to an- other was not restricted to aesthetics. Our analysis shows that: (i) at Border Cave two assemblages, dated to 45–49 and >49 ka, show a gradual abandonment of the technology and tool types of the post-Howiesons Poort period and can be considered transitional in- dustries; (ii) the 44–42 ka cal BP assemblages are based on an expe- dient technology dominated by bipolar knapping, with microliths hafted with pitch from Podocarpus bark, worked suid tusks, ostrich eggshell beads, bone arrowheads, engraved bones, bored stones, and digging sticks; (iii) these assemblages mark the beginning of the LSA in South Africa; (iv) the LSA emerged by internal evolution; and (v) the process of change began sometime after 56 ka."