Saturday, 11 February 2012

Asses Wild: Böeda et al.'s (1999) Middle Palaeolithic Card Game

This lame duck has been waiting to be put out of its misery now for 13 years. It's called 'A Levallois point embedded in the vertebra of a wild ass (Equus africanus): hafting, projectiles and Mousterian hunting weapons,' and its thesis is that a Levallois point was driven into an ass's neck with such force that it breached the pedicle of its 3rd cervical vertebra and became lodged there for eternity (or close enough). 

Unfortunately, instead of its being a complete Levallois point, all that was left inside the vert was a small medial fragment of a larger flake. It's shown in two views above, one by itself, the other incorporated into the hypothesized Point. The two are drawings from the Böeda et al. publication. Notice that the flake's fairly small, and appears to be coming to a point--just what you'd expect from a Levallois Point. I'll get to the issue of hafting in a moment. There are a few matters to get out of the way first. This article requires me to take little bites. I think I'll call them Stages. For the first you'll need to refer to the two drawings and this, the only, photo of the flake that shows its true morphology [albeit from a lousy angle and badly lit].
From Böeda et al. (1999) 


The first thing you'll notice is that the two line drawings of the little fragment are different, so it's difficult to know which of them follows most closely the actual specimen's morphology. This isn't a major imperfection. But to my mind inconsistencies like this don't inspire confidence [I know. That's and argument ad hominem, and I don't hafta lower myself to that level--there are enough juicy, unwarranted inferences in this paper to preclude my needing to use fallacious argument]. For now, let's go with the two dissimilar renditions above, and the shape of the inferred original flake, shown the other way around in the illustration below. As it happens, the two inferred flake outlines aren't identical either, and in the figure below we see yet another version of the actual flake's marginal morphology. 
     Anyway, when I finally bothered to pick up this article the other day, after having been challenged to see what I could say about the 'evidence' for Middle Palaeolithic hafting, the first thing that crossed my mind was a question. WTF? Or words to that effect. I wasn't expecting there to be any possibility of doubt that the flake was a Levallois Point. Silly me.
From Böeda et al. (1999) 
How did Böeda et al. infer, from a small, medial fragment, that it had at one time been a Levallois point, and a hafted one, at that? With inabashed alacrity, I think you might say. The authors state, with what sounds like supreme confidence, that it's a Point, and that it must have been hafted, otherwise the Neanderthal that wanted to remove it from the bone couldn't have pried on the proximal portion of the flake with enough force to snap it in two. There's so much that's wrong with that line of thinking that I can hardly imagine how it was ever published in a refereed journal, much less the august, stiff-upper-lip British stalwart, Antiquity. But, Hey. That's the archaeology of the Neanderthals for you.
     So... Why do they think it must have been a Levallois Point? Böeda et al. point out [cough] that the dorsal flake scars are just like those on a Bordesian Levallois Point. That's true. But that's it? That's the only flake shape that could ever, in the whole long tenure of the Mousterian, have evinced a dorsal morphology such as this, anywhere along its length? Well, no. There is abundant evidence to contradict Böeda et al.'s assertion that their tiny ass pain of a flake fragment is, unequivocally, that of a Levallois Point. 
      To back up my assertion, I've put together a bunch of images from the 1980s excavations at Douara cave to show how precariously Boeda et al.'s interpretation clings to reality. [Profound gratitude needs to be extended to the Japanese team that worked at Douara. They illustrated (seemingly) everything they dug out of the ground, and not just the pretty bits--which is what one usually gets in a site report.] 
     Have a look at the various Levallois flakes shown below, none of which could reasonably be called a classic Levallois Point. [By showing you these images all in the same place, I know I risk spawning another round of raving about specialized Levallois blade industries. Please, Dear Reader, curb your urges in that direction.] 
     All of the Levallois flakes shown below are longer than they are wide, and yet a portion of each is analogous to the flake fragment that was allegedly such a pain in the ass. To help you see what I see, I've even outlined the relevant portion of each flake. Flakes such as these, that are longer than Points, would not be as difficult to snap in two as Böeda et al. aver would have been the case with a Point. 
All whole flake images are borrowed from Y. Nishiaki, 'CHAPTER 5: Middle Paleolithic Assemblages from the Douara Cave, 1984 Excavations,' in PALEOLITHIC SITE OF DOUARA CAVE AND PALEOGEOGRAPHY OF PALMYRA BASIN IN SYRIA, edited by T. Akazawa and Y. Sakaguchi, Bulletin No. 29, The University Museum, The University of Tokyo, 1987. 
After seeing the above I think you'll agree that there's literally no--NO--reason why anyone should accept Böeda et al.'s assertion that the fragment in the ass's vertebra was part of a Levallois Point.       None. Nada. Nil. Zip. Zilch. Squat.  The best part about each and every one of the examples above is that any of them could have been snapped off by the bare hands of even a septuagenarian Neanderthal without any need for hafting to increase leverage. 


How the flake fragment ended up inside the 3rd cervical vertebra of the ass seems to be a foregone conclusion to Böeda et al. Yet, even an ageing out-of-work archaeologist with an embittered heart can see, from the visual evidence they present, that Böeda et al. are missing some fairly important evidence on the bone that surrounded the little tyke of a flake.
     I can't forbear letting the authors speak for themselves. I couldn't possibly paraphrase their argument. Every line is packed with ore with which to forge a reasonable come-back to their facile fantasy. In their own words
...after having traversed the vertebra through the ... pedicle (the bony wall that connects the vertebral body to the neural arch) from the right side. It entered at a slightly oblique angle, from high to low ... 
Push the STOP button. At this point I must interject (something I can't imagine this flake/point ever did). Have a good look at the illustration below. 
From Böeda et al. (1999) 
In the right-hand image of the call-out we see a flake that's not the shape of the flake shown in the first three illustrations. [Now a total of four unique representations of the same flake in one brief article.] On the left, we see that the flake is oriented with the thickest (proximal) part to the left and the thinnest (distal) part to the right, just as you'd expect if the imagined whole flake had penetrated from the right hand side of the animal, as described by Böeda et al. Look again at the right side of the call-out. In this view the front of the horse is at the top. Thus, in this view the flake would have penetrated from the right. However, the marginal morphology given in this illustration makes it darned difficult to figure out the direction from which this version of the flake was pointing, and thus, from which side it would have entered. This really changes nothing. And, once again, how the authors depict the object is not as important as the reality. I'm just being picky, and losing confidence by the minute.
     The authors continue
... its presence in the bone cannot be the result of butchery activity. ... a considerable force was required ... to traverse the ... pedicle of the 3rd cervical vertebra of a wild ass (Equus africanus).
They'll get no argument from me on that [clears throat] point. Let's keep going, shall we?
The left edge of the lithic fragment presents traces of alteration. The most marked are observable at the base and present a scarring of the edge due to a strong pressure when the proximal end broke.
Press STOP again. I'm not familiar with the term scarring. It's probably one of those specialized lithic analyst words that we benighted non-flint-knappers don't know. So, we'll have to take their word for it that the proximal break indicates pressure exerted dorsally, something that isn't obvious in any of the four drawings or the one photo. You'd think, wouldn't you, that it would be a simple matter to determine in which direction the flake was bent with sufficient force to break it, based on the morphology of the break itself? In that case it does seem odd that the authors saw fit to mention this 'scarring' almost as if they wanted it to be evidence of the direction the force was applied. In a minute you'll read that the 'point' penetrated the vertebra without its point. Strange that there wouldn't have been more extensive marginal modification in that case. Oh well. I'm not a flintknapper. And Böeda most certainly is. I'll just take his word for it. To continue

Indeed, the base broke off when the fragmentary point was solidly embedded into the vertebra and could no longer move. This fracture was accompanied by a scarring of part of the edge. These accidents, the mesial/proximal fracture and the scarring of part of the fracture plane and the left edge, indicate that the non-embedded portion of the Levallois point exerted a strong, asymmetrical force on part of the fracture plane. However, for this to occur, the proximal part of the Levallois point must have been firmly hafted.
I'm afraid they've lost me here. So, the proximal break is 'accompanied' by the marginal 'scarring' on the 'fracture plane,' which I take to mean the flake's ventral surface, and on the edge. I wish they'd been a bit more specific about the 'scarring' on the ventral surface. Think about it. Pressure exerted upward proximally would produce pressure downward distally. The smooth ventral surface, in that case, would have been pressed against the inside of the vertebral foramen, also a smooth surface. One wonders what kind of damage that might inflict on a big, strong, rock like Levallois Point, and how one might describe it. Was the scarring some form of abrasion, which left fine striations? Why didn't they say so? Surely they don't mean that the rock was dented by the bone. I'm probably just dull-witted, but why didn't they come out and describe the morphology of this 'scarring,' to remove any possibility of ambiguity? I think you know what I'm thinking.
Other traces of alteration are observable on the dorsal face of the left edge. These consist of micro-denticulations formed by small, adjacent removals (1 to 2 mm in depth) which were detached following a violent contact of the edge with bone.
Thrilling stuff, eh wot? Such violent contact should have left some mark on the bone. Don't you think? Again, I'm sure Böeda et al. know best. Onward and ever forward.
The missing distal extremity of the point was not found in the medullary canal [sic] [vertebrae don't have medullae], and the reasons for this can have important implications. Did the point initially penetrate the vertebra and then break off in the medullary canal? Or, alternatively, was the point already broken before it penetrated ... the vertebral body? Several indices favour the second hypothesis. In effect, according to the reconstitution of the missing distal extremity, its length is estimated at 2.4 cm; however, the distance separating the distal extremity fixed in the bone and the opposing wall of the foramen is only 1 cm. Thus, if the distal extremity of the point had been in place at the time of penetration, there would forcibly have been a violent contact with the bone situated in its trajectory and there would be evidence in the form of impact traces. However, there are none. It is thus probable that the distal end of the Levallois point was broken off upon initial contact with soft tissue, or with bone, and only the mesial part penetrated into the vertebra.
OK. I think Böeda et al. just shot themselves in their collective foot. So, the pointy part of the putative Point broke off either when it contacted the firm, juddering flesh of the ass's neck or broke off when it contacted the bone, leaving the distinctly un-pointy remainder to crash through the vertebral pedicle and become lodged inside. Chew on that for a while. Seems a bit of a reach, to me. And where, pray tell, is the portion of the vertebra that was displaced when the putative point putatively penetrated the vertebral pedicle? Was it differentially preserved? Unfortunately bioturbated in the direction of elsewhere? Again, hardly seems likely. But we'll take them at their word. Böeda et al. are winding up for the big finish here.
In order for such an object to penetrate ... a vertebral body, a strong force is necessary, thus requiring a very strong grip. Such a grip cannot be achieved by holding the object directly in the hand, even with some sort of protection. Moreover, independent of the fact that a considerable force was necessary to penetrate the bone, the object also needed to traverse the intermediate soft tissues. It thus seems evident that this Levallois point was hafted onto a shaft that extended the long axis of the triangular object.

This last bit is so full of unwarranted assumptions as to be, honestly, laughable. To summarize. A blunt flake of some indeterminate length was somehow thrust through several centimetres of soft tissue and horse hide, broke through solid bone without shattering, and left no trace of the bone from the part of the vertebra it blasted through. Then an individual with two hands and two feet grabbed the near end of the flake/hafted flake and snapped it in two just inside the vertebra, causing tiny pressure flakes to be removed on the distal left flake margin. This is really starting to hurt my brain. 
     But we're not done here.


Look closely at the (really rather nice) line drawing of the vertebra of interest, below. See all those Swiss cheesy areas? The authors, curiously, have nothing to say about them. Perhaps they weren't aware that they weren't looking at an unaltered skeletal element (except for the part where they say the point penetrated). In fact, this piece of bone has been quite thoroughly gnawed by (most likely) a canid. Hyaenas would have made short work of breaking up such an object, and would simply have swallowed and digested the bone (which is, as many of you know, what Hyaenas are particularly adept at doing). The horizontal Swiss-cheesy area in the middle is what's left of the spinous process. Based on this and the photographic illustrations it appears that the cranial end of the centrum has been removed. Likewise the other breached areas of this specimen. Note also the four missing zygapophyses. 
From Böeda et al. (1999) 
Now have a look at this photograph taken from the side where the flake is said to have broken through. I've circled the area in question. The horizontal, midline breaks that enabled the archaeologists to take the vert apart are way-post-mortem. They are columnar, which cannot occur in green (i.e. fresh) bone. 
From Böeda et al. (1999) 
For comparison, I include below an illustration of a complete equid 3rd cervical vertebra. A sort of After and Before. Both the photo above and the drawing below are viewed from the same side and are oriented with the head end to the right.
Now that you know what's missing from the archaeological specimen, in the closeup below have a look at the area described by the yellow ellipse in the photo above.
Click on this to see it larger. From Böeda et al. (1999) 
Look at the margin of the cortical (smooth) bone visible in the foreground at the bottom of this view. Do you see how rounded and shiny the edge looks? That's not what bone looks like when it's freshly broken, and it doesn't happen naturally while buried. In the absence of other rounding and polishing that might indicate rolling in a flowing stream, a break margin will only begin to look like that (and in short order) when a canid works away at the piece of bone with its rasping tongue, trying to lap up the yummy fat contained in those bubbles you see there, which have been exposed, almost certainly, by the same canid when it levered off the piece of cortical bone that previously encased all that lovely, fat-filled, bubbly bone.
     Look, too, and the break margin above the area where the flake is supposed to have penetrated this specimen (i.e. the plano-convex really dark area in the lower centre of the image. It's nicely rounded too. And, is it just me? The concave portion of the opening--the opening that was putatively produced by a flake with a trapezoidal cross section--isn't trapezoidal. It's distinctly arcuate. I suppose it's possible that the canid simple enlarged the hole that the forcefully driven flake made, and in so doing transformed the opening from quasi-trapezoidal to smoothy concave. But that seems like a leap, considering what canids are capable of doing to skeletal elements of this size, and the manner in which the lower breach would have been achieved by the carnivore's attentions. So, how did that flake get in there?
From Böeda et al. (1999) 
In fact, any number of peri- or post-depositional processes could have created the juxtaposition of the flake and the bone. But what the heck, it's at least theoretically possible that an elongated Levallois flake penetrated the vertebra. Maybe a steel-gripped Neanderthal snapped said putative flake in two. Only then, maybe, did the canid wander by to have a nosh. Either way, Böeda et al.'s argument for hafting in the Middle Palaeolithic has gone, in my estimation, from a Neanderthals-R-Us Hero to a Zero (or close enough). 


I've been told by reputable, flint-knapping, lithic analyzing archaeologists that if a body doesn't know how to replicate stone tools they have no business doing archaeology.

I hope you'll forgive me for proposing this equally arrogant corollary, only half-facetiously. Any archaeologist that can't recognize bone modification caused by non-human actors has no right pontificating on the cognitive or technical abilities of the Neanderthals.

I hope I've made my point. *cough...cough*        
     Have a nice day.