Nefud For Thought: Scerri et al.'s Arabian Levallois Artifacts Give Us Pause To Reconsider Middle Paleolithic Typology

Far be it for me to downplay discoveries that have the potential to overturn long-standing misunderstandings about what went on in the past. On the other hand, I can't help but be excited when I see images of artifacts such as these, for what they imply about the bipedal apes that made 'em.

From Scerri et al. "Middle to Late Pleistocene human habitation in the western Nefud Desert, Saudi Arabia," Quaternary International, available online 8 October 2014.

I call your attention to object number 1. As you can see, it's described as a "Centripetally prepared preferential Levallois core," as is object 2. Object 3 is a "recurrent centripetal Levallois core," 4 is a "Single platform core," 5 a "bidirectional Levallois point core, 6 and 7 "discoidal cores," 8 another "single platform core," 9 a "multiple platform core," 10 and 11 are "bifaces." [BT dub, gotta give the authors credit for not referring to 10 and 11 as hand axes, since that's what they, evidently, are, and would normally be called that in run-of-the-mill Middle Paleolithic scholarship. O' course, they do identify similar objects as hand axes elsewhere in their paper—I guess maybe these two weren't sufficiently 'well made' or pointy enough to make it into that category. In that case, one wonders why they weren't treated as hand axe 'rough-outs' as others have done. Somebody stop me!]

What I'm seeing here could imply much more than what Scerri et al. are reporting—that these Arabian Peninsula Middle Pleistocene assemblages are evidence for multiple demes spanning tens of thousands of years in MIS 5, the last interglacial—including anatomically modern Homo sapiens from northeastern Africa and Homo neanderthalensis from southwestern Asia.

I think the authors are missing something about the size of these artifacts. Here's a close-up of number 1, with some everyday objects for comparison. Yep. Those are a One Pound coin, a Tooney, a Sacagawea dollar, and a 2 Euro coin.

Yet, despite its size, Scerri et al. describe this fragment of rock—barely larger than pocket change—as a "Centripetally prepared preferential Levallois core." It sounds impressive! It is impressive! If it's what the authors' claim, this little beauty could be evidence for a group of bipedal apes with hands smaller than a five-year-old's! Either that, or this so-called Levallois industry is just that, a three dressed up as a nine. After all, could any of these pieces actually be Levallois cores [which, forgive me, perhaps I've missed something all along, but, I thought a Levallois core was something that was being prepared for the final removal of a flake of predetermined shape]? Give me a break!


Object number 3 is an even more impressive piece when you consider that it's a "recurrent centripetal Levallois core." As so many of you already know, I'm not a lithic analyst. But I think that means—unlike number 1, this has more than one flake scar, implying more than one flake removal in preparation for that magical, mystical flake of predetermined shape, which, equally mysteriously, was never removed. So. When is a Levallois core not a Levallois core? And why were all of these not taken to the next stage, according to [the conventionally wise construct of the] plan?

For my part, I don't know why we just can't use the old Bordesian typology, illustrated below. In that case number 1 wouldn't even make it as a Levallois core, and, at best, number 3 would be a "Classic." A classic it be, for sure! Either way, given their size, any flake of predetermined shape that a bipedal ape hoped to remove at this point in the reduction sequence would either be doomed to failure, or so small as to have been useless for any function more adaptive than cleaning one's fingernails. The same goes for the rest of 'em. And for any inferences that they can milk out of these bit of rock, however they want to describe 'em, and regardless of how many others in our field accept their inferences. [But then, that's why I'm here, in my metaphorical pajamas in my metaphorical (dead) mother's basement, bashing out this piece of uninformed criticism.]

Nevertheless, if these little bits of stone can be described in this fashion, and yet be, to all intents and purposes, useless as Levallois cores—in the original sense of the type—what does it say about all of the grown-up 'cores' that we're told were being shaped for the purpose of removing a flake of predetermined shape?

Let's face it, it's easy to recognize a Levallois core after that flake's been removed. Number 5, above, is one such example—mostly because the central flake had a point. It takes a real expert to recognize one that never got that far! Unless, of course, those final flake removals weren't, after all, the predetermined end product of the presumed process.

Many times before I've trotted out the Levallois cores from Douara Cave as evidence for the fallaciousness of the presumed Levallois technology. But it won't hurt to do it again.

Levallois cores from Douara Cave (Credit Akazawa in Suzuki and Takai 1974).

I've outlined in red the final flake removal on these Levallois cores. Can we not agree that if their final shape was 'predetermined' there's nothing here that would fit, without considerable massaging of reality, in Bordes's original—and highly idealized—classification of Levallois flakes, which you see above.

I'd be very happy to throw out the whole Levallois technique baby with this Arabian Peninsula bath water.

I'll say it again, I'm not a lithic analyst. If I were, and if these rocks from the Arabian Peninsula were what the authors said they were, and if I believed them, I'd consider jumping on their bandwagon of desert archaeology and riding it as far as I could on the road to academic fame and fortune. But I'm not. And I don't. And I won't be joining them on their—obviously very prestigious considering the number of papers they're publishing and in what august scholarly journals they're publishing them—bandwagon.

I'm outa here.

Taxi!