Wednesday, June 01, 2011

The Arabian Middle Paleolithic and the southern route of human dispersal

In a comment on my last post, Maju who's a regular commenter on this blog, pointed out that recent finds in the Arabian Peninsula and the Persian Gulf suggest that modern humans might have been present in the Middle East by the time Shanidar 3 was killed. Some of the specific evidence in support of this that has come out in the past year include that presented by Armitage et al. (2011) and Rose (2011), and a more recent paper by Petraglia et al. (in press), which he briefly discussed in a post of his own. Here, I just want to provide some additional thoughts about this series of papers, and what they tell us about the modern human population dynamics in the region.

Armitage et al. (2011) report a series of three stratified assemblages from the site of FAY-NE1, in the Jebel Faya in SW Arabia, near the Straits of Hormuz. The interesting thing about these assemblages is their age, and the fact that their typology suggests affinities - or lack thereof - to assemblages from other regions. The oldest assemblage dates to ca. 125kya (dated through single-grain OSL), and the authors argue it shows affinities to assemblages of similar age in East Africa. On that basis, they argue for an early dispersal of modern humans out of Africa along the so-called 'southern route', which comprises the southern coast of the Arabian Peninsula and Indian subcontinent. They link this, in turn, to a potential early population of modern humans in Indian prior to the Toba super-eruption ca. 74kya (Petraglia et al. 2011 and references therein). The later assemblages from FAY-NE1 are older than ca. 40kya, and show little to no affinities to either the MSA/LSA or the Levantine Paleolithic. Armitage and co. interpret this as evidence of their not having been made by Neanderthals, but don't really explore this issue in great depth. They conclude that human occupation of FAY-NE1 and Arabia more broadly would have been tethered to humid periods that made it hospitable to humans, who would have disappeared from the interior of the peninsula during arid periods.

Rose (2010), on the other hand, argues that one area that humans might have found refuge in during these hyper-arid periods would have been the Persian Gulf. Because it is comparatively shallow, arid periods (which correspond with colder periods associated with decreased sea level) would have effectively exposed much of the gulf, which would have been associated with perennial sources of water, which would have acted as a drawing force for human populations from surrounding areas. In a nutshell, the 'Gulf Oasis' would have provided something of a safe haven for humans at times where both Arabia and parts of Iran would have been too inhospitable for humans to occupy. If modern humans were present in the region by 125kya, it stands to reason that the people who would have congregated in the 'Gulf Oasis' would have been modern humans, who could have in turn recolonized the areas to the E and W of the Gulf during wetter periods. The prolonged isolation of people in the Gulf Oasis during prolonged (multi-millennial) episodes of dessication would have lead to cultural drift, perhaps explaining the unique configuration of Assemblages A and B from FAY-NE1 (Armitage et al. 2011).

While these new data are very interesting, they still concern the potential presence of humans mainly along the coasts. What this means is that they tell us little about whether or not humans ventured far inland and/or northwards, and what this implies about their interactions with other, putatively 'archaic' human populations in those areas. A new paper by Petraglia et al. (in press), however, helps shed some light on this situation. These authors present a preliminary report on the site of Jebel Qattar 1, located near Jubbah, Saudi Arabia, on the shores of a paleo-lake. This places the site smack-dab in the center of northern Saudi Arabia, hundreds of kilometers from any coastal area. The site dates to ca. 75kya (OSL), which associates it with comparatively moist conditions that would have made the region habitable by humans.

Overall, the presence of JQ1 agrees with the scenario for human occupation of inland Arabia proposed in the previous two papers. Namely, it shows that people would have ventured far inland only during comparatively moist periods. What is really interesting here, however, is that the assemblage found at JQ1, in contrast to those from FAY-NE1 (Armitage et al. 2011), fits well within what Petraglia and Alsharekh (2003) have described as the Arabian Middle Paleolithic, which shows some affinities to the Levantine Mousterian. However, as the authors state

"Given the current absence of pre-Holocene hominin fossils in Arabia, and the fact that Levantine Mousterian assemblages are associated with both early modern humans and Neanderthals, caution is warranted in attributing a maker to the JQ1 and other Arabian Middle Paleolithic assemblages." (Petraglia et al. 2011: 4)

Here, I just want to point the contrast between this assemblage, and those from FAY-NE1, who show affinities either to the MSA or to no other known industries to the West. What this means for the presence of modern humans in the northern Zagros, close to Shanidar, remains an open question. It may well be, however, that what was happening along the coast of the Indian Ocean during the Late Pleistocene may have been quite different from what was happening in the interior of the landmasses it borders on, with attendant implications for scenarios about modern human dispersals. I close with words by Petraglia et al. (2011:4, references excised) to that effect:

"If modern humans were responsible for the early Arabian toolkit, then our findings contradict the argument that the dispersal of Homo sapiens out of Africa was accompanied by a microblade technology 60 ka ago. Furthermore, the presence of JQ1 in the interior of northern Arabia, 500 km from the nearest coast, indicates that an exclusive coastal corridor for hominin expansion out of Africa can no longer be assumed."

Hat tip: For what they were... we are.

PS: I should also mention here that Michael Petraglia and his team have a blog, Ancient Indian Ocean Corridors, where they post about the Indian and Arabian Paleolithic , issues related to their ongoing research in those areas.

References

Armitage, S., Jasim, S., Marks, A., Parker, A., Usik, V., & Uerpmann, H. (2011). The Southern Route "Out of Africa": Evidence for an Early Expansion of Modern Humans into Arabia Science, 331 (6016), 453-456 DOI: 10.1126/science.1199113

Petraglia, Michael D., & Alsharekh, Abdullah (2003). The Middle Palaeolithic of Arabia: Implications for modern human origins, behaviour and dispersals Antiquity, 77 (298), 671-684

Petraglia, M., Alsharekh, A., Crassard, R., Drake, N., Groucutt, H., Parker, A., & Roberts, R. (2011). Middle Paleolithic occupation on a Marine Isotope Stage 5 lakeshore in the Nefud Desert, Saudi Arabia Quaternary Science Reviews DOI: 10.1016/j.quascirev.2011.04.006

Rose, J. (2010). New Light on Human Prehistory in the Arabo-Persian Gulf Oasis Current Anthropology, 51 (6), 849-883 DOI: 10.1086/657397


11 comments:

Maju said...

Thanks for the credit and link, Julien. :)

I'd say that the Nefud finding is not that important because it could well be a Neanderthal site and, would it be sapiens, it does not seem related to either what is in Africa nor in South Asia.

However Michael Petraglia has argued, even directly with me at his blog, that the evidence is not strong enough for a purely coastal migration, that most of the sites he and related teams are studying are inland:

"All I can tell you is that in Arabia and South Asia, there are a tremendous number of inland archaeological sites. I would argue that populations were using river valleys and inland basins as they slowly spread across Asia".

But it does support a "coastal" migration in the loose sense of "along the geography north of the Indian Ocean", which may well include strictly coastal journeys and (IMO) exploitation of coastal resources such as fish, shellfish, marine birds, etc.

Julien Riel-Salvatore said...

Hi Maju -
thanks for your comment. I'm not sure I agree with you about the 'coastal migration' loosely defined... JQ1 is located right in the middle of Saudi Arabia. If it's part of this migration, I'd say that at least one of the paths taken by humans out of Africa was pretty far removed from any coast at all.

Re: strictly coastal journeys, it would make sense, but it's also interesting to note that there is almost no evidence for these kinds of adaptations with the non-African sites used to support the southern route hypothesis. In fact, none of these sites are located on the coast - it would seem like a logical next step in that research program to go looking for coastal sites associated with very steep coastal shelves, so that their position relative to the sea would not have changed significantly since the Late Pleistocene... then, we could get some information about whether or not these behaviors were involved in a migration along the northern coats of the Indian Ocean.

JRS

Maju said...

You are not going to find many strictly coastal sites unless you begin making underwater archeology massively, what is not likely at all.

If you want to discuss whether people exploited the coast or not in those times the answer is surely yes, because they used sea mollusks for ornaments and because there are at least some coastal sites known (an island off Eritrea's coast whose name I can't recall for example).

But that's not my point when I say "roughyl coastal": I mean around the Indian Ocean's shores and not across inland Asia, as in Khazakstan, Altai or Mongolia. It's "coastal" or "tropical" as opposed to "continental" or "subarctic".


Genetics also cannot describe the exact routes taken but can make this distinction.

terryt said...

"the presence of JQ1 in the interior of northern Arabia, 500 km from the nearest coast, indicates that an exclusive coastal corridor for hominin expansion out of Africa can no longer be assumed."

To my mind the theory has never been able to hold water, so to speak.

"our findings contradict the argument that the dispersal of Homo sapiens out of Africa was accompanied by a microblade technology 60 ka ago".

I have tried numerous times to persuade Maju that such is the case.

"the fact that Levantine Mousterian assemblages are associated with both early modern humans and Neanderthals"

There we go again. Australia? Altai?

"But it does support a 'coastal' migration in the loose sense of 'along the geography north of the Indian Ocean'"

How far north are you prepared to stretch the definition of 'geography north of the Indian Ocean'? The Subarctic?

"In fact, none of these sites are located on the coast - it would seem like a logical next step in that research program to go looking for coastal sites associated with very steep coastal shelves, so that their position relative to the sea would not have changed significantly since the Late Pleistocene..."

The problem is that 'very steep coastal shelves' are seldom easy of access from inland. And they're certainly not easy to travel along. So I think we can rule them out as part of any 'coastal migration'.

"If you want to discuss whether people exploited the coast or not in those times the answer is surely yes, because they used sea mollusks for ornaments and because there are at least some coastal sites known"

I have no doubt that people who lived near the coast exploited the resources. But it by no means follows that they used to coast, apart from sandy stretches, as some sort of main highway.

Millán Mozota said...

Gulf oasis... theory?
Should a bring a shovel with mel? Are we unearthing Childe's 1928 book for another chronology? ;)

Millán Mozota said...

btw great post

CamArchGrad said...

I think you're being a little unfair to Maju there Terryt.

In North America we have various coastal migration theories, and south of LGM (Around Puget sound), it's assumed that the major river systems (Columbia et al) would have provided access to the interior to the continent.

Likewise, it only took 200 years for Europeans to fully penetrate North & South America using the major river systems.

I can't get the paper (paywall), but I wonder if this paleo lake wasn't connected to a paleo river which drained into the coast. If you have even simple aquatic technology coasts, rivers and lakes are essentially highways.

It's quite easy to imagine people taking advantage of pluvial conditions to penetrate the interior of continents via river systems from the coast, rather then hauling themselves overland.

Also, it may be a mistake to assume marine migration = marine resource exploitation, in cases like the south pacific lapita migration Marine resources were certainly important, but in north and south america (aside from sites in peru) the prey seems to have been megafauna, with marine resources becoming important only after the collapse of megafauna populations after clovis. It's fully possible that groups in the past could have used aquatic technology to broaden their access to megafauna and plant resources, with marine only having minor roles.

Finally, the issue about the lack of sites is a bit of red herring when you see that lakes and rivers are part of an aquatic adaptation. Along the edges of these paleo lakes and paleo rivers their should be abundant sites unaffected by sea level rise. Perhaps Jebel Faya is one of these.

Maju said...

If you haven't done yet, you may want to check the latest entry of Petraglia's blog, including two Scribd links to his latest papers. One is the paper on the Nefud site but the other, titled "Trailblazers across Arabia" is more a short article with a quite illustrative map, discussing the various possible routes taken by our ancestors in route to South Asia, not just his own but also Armitage's.

terryt said...

Thanks for the link.

"the various possible routes taken by our ancestors in route to South Asia, not just his own but also Armitage's".

Obviously I prefer the red or green route over the purple as being the most likely.

"I think you're being a little unfair to Maju there Terryt".

Probably. But he tends to stick obsessively to his current theory, and refuses to consider alternative interpretations of the evidence. Several times the alternatives I have suggested have turned out eventually to have more evidence in their favour. This is after he has insisted that my explanation had been impossible, and stupid.

"In North America we have various coastal migration theories, and south of LGM (Around Puget sound), it's assumed that the major river systems (Columbia et al) would have provided access to the interior to the continent".

I'm quite prepared to accept that to be so. However I doubt that the first people to America traveled all the way to South America via the coast. I'm reasonably comfortabl;e with the route through Beringia for the first lot.

"If you have even simple aquatic technology coasts, rivers and lakes are essentially highways".

I am reasonably certain that early humans did not possess effective aquatic technology. We see evidence for open water crossings with the Aborigines' arrival in Australia, but Mediterranean island hopping does not effectively occur until the Holocene. On the other hand humans have to stick close to rivers and lakes because they need good supplies of fresh water. So yes, they do tend to migrate along the shores of freshwater bodies, but not necessarily along the middle of them.

"Also, it may be a mistake to assume marine migration = marine resource exploitation"

I think it is a mistake to assume 'marine migration' in the first place. The excuse that sea level rise has covered the evidence doesn't hold up to any sort of critical analysis.

Andrew Oh-Willeke said...

Shorter version:

Something on two legs was roaming Arabia in this time period, even inland, but it is hard to distinguish the two most likely candidates: Neanderthals and early Out of Africa anatomically modern humans from each other based on the limited tool kit evidence available so far.

Given that Neanderthals were present in the Levant before modern humans arrived ca. 100,000 years ago, and during some of the period in which modern humans appear to have disappeared, from ca. 75,000 to 50,000 years ago, I'm inclined to associate tool kits consistent with Levantine Mousterian assemblages with Neanderthals rather than modern humans, even if the tool kits could be consistent with both, since this is consistent with what we already know about the hominins in that region in that time period, unless there is some new data to suggest that these were modern human settlements instead (e.g. bone tools or fishing toolings or skeletons consistent with AMHs).

The ideal place to look would probably be coastal underwater caves in the Persian Gulf.

I also wouldn't be surprised if there are people in the oil business who know more than they have let on about the extent of the archaeological record in the area because they don't want to interrupt oil drilling in the interest of archaeology. There has been a lot of digging and geological research done on land and at sea in the area, and geologists hired for jobs like that tend to be pretty clever.

onix said...

mh. i think there is enough of a body of evidence to make interpolations for early hominids range. assuming personally that coastlines would be basically routes. human do have early aquatic options mind you;) they swim.