Monday, 15 October 2012

Oldowan [from Melka] Kunture is Not a Star Wars Character

A couple of weeks ago I clicked through to Erin Wayman's Smithsonian blog, Hominid Hunting. The post I was curious about is part of a series called 'Becoming Human' and the article in question was 'The Origin of Stone Tools.' I was struck by the graphic accompaniment to Erin's article, reproduced here, and especially the caption from Erin's blog, which is given verbatim below it.
'Oldowan choppers are among the oldest-known type of stone tools. Image: Didier Descouens/Wikicomons'
The image depicts a chipped stone artifact from Melka Kunture, in Ethiopia. Its age and the conventions of African archaeology demand that it be referred to as an Oldowan tool, after Olduvai Gorge in Tanzania where the famous pair of Mary Leakey and Louis B. Leakey toiled for many decades. The artifacts pictured above could be as old as 1.7 Ma. However, artifacts just like these have been securely dated to 2.6 Ma elsewhere in Ethiopia, at Kada Gona.
     The caption reads 'Oldowan "choppers".' 'Chopper' is one of the names that Louis Leakey gave to these early artifacts, and so, it would seem, has everyone else since the 1950s. The trouble is...where's the evidence that this object was the desired end product such that it represents a distinct 'type'? The evidence, I'd say, existed only in the mind of L. S. B. Leakey, himself. And the trouble with that? Well, many, many archaeologists have used and continue to use Leakey's classification and, implicitly, it's underlying lines of reasoning. And, after a time, use becomes convention, convention becomes unexamined 'truth,' and pretty soon it's 2012 and the whole world thinks these objects are the typological equivalent of arrowheads--in other words the desired end product. Damage to the discipline? You betcha! Retaining such labels is counterproductive--it serves only to perpetuate and further solidify the reification that what you see in the illustration are truly the desired end products in the minds of the creatures that made them.
     So, Gargett, if they're not choppers [or picks, or discoids] what are they? It's not rocket science. I have to give the crafty bipedal apes credit for coming up with the 'idea' of banging one rock against another to remove a sharp edged fragment. Surely this was a stunning accomplishment for mammaldom. But really, aside from the energy required to remove said fragment, there's nothing unique about such an act in the wide world of animals. We all remember that a species of Galapagos finch snaps off cactus needles to use as probes to extract insects from their hiding places. How far apart are the two behaviors? In fact I'd say they were pretty much identical, and probably used about the same cognitive structures, one being the avian version and the other, the ape. Both behaviours are reductive--one species bangs a rock against another rock to remove a piece to use to cut something; the other snaps off a twig to use as a probe. Both use the portion removed to effect their purpose, not the thing that remains after the removal. We wouldn't dream of looking at the cactus, seeing that it's missing a needle, and give it a special name that implies that it was the end product 'desired' by the finch. Why then should we conceive of the rock, lighter by one shard, as being conceptually different than the cactus? The only answer to that, as far as I can see, is that there is no good reason other than the misguided mindset of numerous archaeologists over the past century or so.
     Having said that, do I still need to explain why the artifact shown above isn't what Leakey said it was? Oh, all right, in the view on the upper left, the 'chopper' has a relatively straight break margin. Thinking that this artifact was the end product, and your purpose is to ascribe a function to it, you might interpret such a shape in terms of a technology that you know modern humans have used for tens of thousands of years, and continue to use to this day. You're going to think that the object looks [very roughly] like an axe or a meat chopper.
     In a similar vein, the view on the upper right shows the artifact having a bit of a point. Using the same level of imagination that made the first view into a chopper, the one on the upper right resembles a pick more than an axe. And thus, the artifact could just as well have been classified as an Oldowan 'pick.'
     But why should we think that such artifacts were the intended end product? Isn't it much more likely that the flakes struck off were the 'desired end product,' and the remainder just the 'core?' At least if we adopt a stance like this we won't be privileging one set of behaviours over another. A number of archaeologists working today accept the essential 'core'-ness of Oldowan artifacts. However, many of them believe that we can conserve these anachronistic classifications because they allow us to quickly characterize an assemblage. My credulity is strained to the limit by such arguments. To what end? What, I ask, is the point of knowing how many lumps of rock were discarded after one, two, three, or more flake removals?
     I'll admit, it's easy to pick on the Oldowan artifact classification and see it for what it is. It's straightforward [as I see it]. Unfortunately, as I've stated before, the question of what was really the intention of the rock-knocker becomes crucial to the archaeological narrative as we get closer in time to the present. As you've seen before on the Subversive Archaeologist, it's this notion of the finished artifact fallacy that's at the heart of my critique of the so-called Levallois technique in the Mousterian [and, of course, the critiques of Iain Davidson and Bill Noble].
     I'll save another examination of the Levallois technique for another time. I hope you had a fine weekend. Back to work tomorrow!

[This article has been updated to remove text referring to the illustration as three different Oldowan artifacts. It's not. A tip o' the hat to the ever-perspicacious Marco Langbroek for breaking the news to me in one of his comments.]

Thanks for dropping by! If you like what you see, follow me on Google Friend Connect or Twitter, friend me on Facebook, check out my publications at Academia, or connect on Linkedin. You can also subscribe to receive new posts by email or RSS [scroll to the top and look on the left]. And please don't forget. Oh, and you can always put me on your blogroll! By the way, I get a small commission for anything you purchase from if you go there by clicking through from this site. 


  1. Given that Leaky was educated in the mid 1920's his categorisation of stone tools is not surprising and on par which what was going at the time. The real issue is the failure to update or modify the classification to reflect modern understandings.

  2. I actually think the photograph depicts *only one tool* seen from different sides. I.e. the two top images are two side views and the third is the top view. Of one and the same artifact.

  3. I agree, by the way, that names like "chopper" etcetera are meaningless in a functional sense. I alos agree, and I think many palaeolithic archaeologists would agree, that the class called "choppers" in most cases are cores.

    I hate typology, for the reasons you write: people take it too literally. It also "hides" the variation present within one typological class, but can also put superficial boundaries in what in some cases is a continuum.

    At the same time, I do think iyt serves a function: when you say "chopper" all archaeologist will know what you are talking about.

    You do it yourself too. You use the class name "core", which is a typological unit with a functional implication itself and your use of the word hence is not much different from the use of the word "chopper"

  4. What I see as the biggest drawback of typologies, is the distinction between formal "tools", and "waste" (unmodified flakes).

    Modified artifacts falling into a formal "tool" category get all the attention (e.g. in spatial analysis) while unmodified flakes are treated as "waste", as by-product, as a somewhat uninteresting leftover category.

    In reality, it are these un-modified flakes which are in many cases the desired "tools" and should be the focus of attention.

  5. @Marco: I'm. Such. An. Idiot. I've ... er ... adjusted the text accordingly and acknowledged your timely and helpful [and gentle] observation. A bit reckless of me. Thank you very much.

  6. Marco was/is right (again) (and so is Iain). But 1) typological is much more pervasive than just in the stone tools. The splitting of the hominin fossil record into named species is another example. With the definition of a new species known only from a tooth, a bone and its DNA,a whole new option has opened up. Maybe all (or some of) those Chinese hominins that did not fit neatly into the Eurocentric classification (typology) of fossils were actually Denisovans. We will not know until a skeleton is found with enough Denisovan aDNA. Of course it is also true that the Skhul and Qafzeh skellies are only modern human through such a process and I might be quite wrong about the destruction of the equation M=N (Mousterian equals Neandertal).
    2) Iain is so right, but it is not just stone tools but the assemblages of them. Try to unpick the complexities of the Aurignacian or the Magdalenian in Europe and then persuade any European gatekeepers that the concepts should be abandoned.
    Just sayin'
    Iain Davidson

  7. Iain. Mousterian equals Qafzeh at 100 Ka and Kebara at 60 Ka. That's when the suppose interbreeding took place, not at 40 or 35 in Europe when there were Mousterians and Us.
    Just sayin' ;-)

  8. I agree that the current interpretations should force us to postulate an interbreeding between 100 000 and 65 000. Positively Tantric.
    But if S/Q is moderns then Mousterian can be moderns earlier, later, there, elsewhere. And the whole game changes in ways I have only just begun to write about.

  9. @Iain D.: I think that *in Europe* when predating 45 Ka, Mousterian IS Neandertals. For the simple reason that we have no convincing fossils of H. sapiens there before 41 ka while we have sh*tloads of Neandertal fossils from that period (mind you, in my opinion the 44 ka date for Kent's Cavern KC4 "modern" maxilla is an overinterpretation not properly taking into account taphonomy and stratigraphic vagaries at the site. I think it is likely =<41 Ka).

    Post 41 ka, in Europe too "Mousterian" indeed need not equal "Neandertals" and that is the main point of contention for example regarding Slimak et al's "Arctic Neandertals" (Slimak et al, Science 332, p. 841 (2011). See for example the critique by Zwyns et al (Science 335, p.167 (2012)).

    Outside Europe it is a whole different matter.

    But it also works the other way around. There is still a tendency to regard Eurasian laminar technologies as "entering H. sapiens", even though we now know that Neandertals produced laminar assemblages from at least OIS 7 onwards. There currently is a tendency for example to see the Bohunician at ~47-40 ka as "the first H. sapiens entering Europe" although there is not a single sapiens fossil dating to this period (but several Neandertal fossils that do). I am convinced it are actually Neandertals re-colonizing previously abandoned territories in west Europe at ~50 ka from refugia elsewhere (= central Asia).

    @Iain D.: indeed many (including me) have harbored the suspicion that "Denisovans" equal the so-called "archaic Homo sapiens" of East Asia (i.e. the East Asian "Homo heidelbergensis" variant) right from the beginning. I know that Stringer for example is certainly positive towards that idea.

  10. I have the working hypothesis that when it comes to fossils the archaeological trained are splitters cause every variation has meaning and the biologically trained tend to be lumpers because they are used to variation in species like dogs or gum trees.

    As for traditional tool classification schemes didnt Brian Hayden have some sort of traumatic experience when a Western Desert traditional Aborigine used a scrapper to chop!

  11. @Marco
    Thank you for your thoughtful and detailed comment. Of course I am familiar with these arguments. I have never seen a map which shows exactly where the Neandertal fossils have been found at different dates, and where the Mousterian industries are at those same time ranges. I could probably do it myself, but I would lack some of the literature (and quite a lot of the inclination). It is still the case, though, that even the presence of Neandertal skeletal material does not prove the point. Maybe neandertals consistently died in caves and moderns did not. But even if that were not the case, the fact that M=N does not work somewhere means that we should be suspicious that it does not work anywhere.
    What proportion of Mousterian sites have skeletal remains? What proportion of non-Mousterian sites do not have skeletal remains? these are empirical questions that I would like to see (someone else produce) the answers to.
    I am also not a believer in the spiritual significance of blades. Let us not get into that.
    Glad we agree on Denisovans.
    I am very much enjoying your QI paper. The first part needed to be said and needs more publicity.


Thanks for visiting!

Note: only a member of this blog may post a comment.