Wednesday, 2 May 2012


All right. I'm working on the dregs of a halfway decent boxed chardonnay. I'm waiting for the washing machine to be free. I'm ready. It's time to turn up the heat on the claims emanating from Wonderwerk Cave, Northern Cape Province, South Africa. There's a fatal flaw in the argument that there's evidence for the controlled use of fire in the Acheulean (i.e. Lower Palaeolithic) levels, at about 1,000,000 years BCE [and I'm not talkin' 'bout Brigitte Bardot].
     Wonderwerk Cave's located where there's a diamond labelled WK in the heart of the photomap below.
After Henshilwood and Dubreuil. 2011. The Still Bay and Howiesons Poort, 77–59 ka: Symbolic Material Culture and the Evolution of the Mind during the African Middle Stone Age. Current Anthropology, 52, 361-400.
I'm seeing all these news reports! This story is everywhere you look. If you watch the news ticker up top for a few minutes, you'll see not only the English-speaking media reports, but also those from areas where they speak Spanish, French, Portuguese, German, Scandihoovian and some I don't even recognize. In words of one syllable this is big news to a lot of folks. And well it should be, if it were true. As you now know, I have have reason to counter the claim.
     I may not be a geochemist or a micromorphologist, but I am sentient. And when I see an extraordinary claim... Well, you know how it goes. I do what I can to learn enough to achieve a minimal level of critical ability and then assess the extent to which the claimants attempt--successfully--to rule out natural processes. In the case of this Lower Palaeolithic Prometheus, the chemistry and the technology are my stretch domains. However, logical argumentation is my comfort zone. So, with a little Googling and some gumption, that's where I'm headed today. 
     I'm talking about a paper, recently published in PNAS.
Berna, et al. 2012. Microstratigraphic evidence of in situ fire in the Acheulean strata of Wonderwerk Cave, Northern Cape province, South Africa. PNAS April 2, 2012.
 Published online before print April 2, 2012, doi: 10.1073/pnas.1117620109
[Hey Rocky! Watch me pull a(nother) rabbit outa my hat!]
The authors contend that their findings provide 'the earliest secure evidence for burning in an archaeological context.' [At this point all of you readers will, automaton-like, voice the pre-eminent subversive question: 'Did they rule out natural causes before imputing their discoveries to humans or other hominids?' Pat, pat. Good little Subversive! Here's a biscuit.]
Paul Goldberg and Francesco Berna

I've been seeing similar claims from Middle Palaeolithic contexts since the late 80s, almost always accompanied by Paul Goldberg's micromorphological analysis and Steve Weiner's or Francesco Berna's Fourier Transform Infrared Spectrometry of the phosphate minerals. I've been skeptical all along, which is my proclivity, nevertheless I've been reluctant to mount a critique. Until now. The stakes have finally grown too great to remain silent. 
     I've known Paul since the mid 80s. We spent time together at Keatley Creek, in B.C. Then after a month-long stint at Kebara Cave in '89 he toured me around the south of Israel, at which time I met Arlene Rosen, and during which I got hotter than I can remember ever being before or since, in the Dead Sea Depression. 
     I had a job once when I was at SFU. I was to impregnate a bunch of micromorphology samples from the Keatley Creek site with fibreglass resin. Unfortunately the vacuum chamber I was using was a little short of breath and some of the samples turned out crap. But that's beside the point. [Really, Rob? What IS the point of all this reminiscing?] The point is that I know Paul. I know his work. He [quite literally] wrote the book on the micromorphological study of archaeological sediments. I accept his findings. Nevertheless, when all's said and done in the authors' analysis of Wonderwerk Cave, it's not Paul's micro-study that's at fault. Nor is it Francesco Berna's FTIS results. It's the way they argue on the basis of their findings at which I baulk.
Wonderwerk Cave [er, the entrance, at any rate]
In case there are one or two of you who're wondering what's so special about fire, I would just remind you that the controlled use of fire has always been considered one of the hallmarks of human achievement [remember Prometheus?]. It's the same in human evolution. Making and using fire is as human as it gets. That there are numerous articles propounding its use in the Middle Palaeolithic is at odds with other evidence of cognitive evolution [there's much subversion yet to be done there]. That's bad enough for this subversive; the claim from Wonderwerk Cave could set back empirical enquiry into the true first use of fire by a hundred years. 
     At Wonderwerk a deeply stratified deposit contains an abundance of what are referred to as 'rubefied' [or reddened] sediments. This should come as no surprise, since the deposits are made up almost exclusively of aeolian sand that's encrusted with iron oxide [=rust]. However, there are also places where the excavators recovered floral ash and burned faunal fragments. Bat caves are capable of building up prodigious deposits of guano--they've been mined, historically, for fertilizer--and those deposits are known to burst into flame spontaneously. So, why couldn't the fire-affected traces be the result of naturally occurring fire(s)? After all, the authors report no features at the site that might be construed as hearths.
     To their credit, Berna et al. have succeeded in mounting a convincing argument that enough heat was generated here and there in the deposits to have altered the chemistry of plant and animal remains, leaving ash and burned bone and plant remains. They infer that temperatures higher than 550 °C were not in evidence. 
     They base their refutation of the spontaneous combustion argument on the work of one geologist from one cave in Romania, Cioclovina. The authors conclude that natural, spontaneous combustion of bat guano had not occurred in Wonderwerk Cave. 
     This is the shape their argument takes
a) When bat guano spontaneously combusts it transforms some of the guano's aluminum phosphates into two very rare forms: Berlinite and hydroxylellestadite.
 b) FTIS found no Berlinite or hydroxylellestadite in the samples from Wonderwerk cave.
c) Therefore no spontaneous combustion of bat guano can explain the fire-affected flora and fauna in the cave.
Here are some brutal facts up against which this argument cannot stand.
1. The discovery of Berlinite in unconsolidated sediments at Cioclovina Cave is unique. All the rest of the Berlinite known in the world occurs in the form of ore. At Cioclovina the Berlinite formed during one or more episodes of very high temperature combustion of bat guano--so hot, in fact, that a stratum of sediment was fused solid (Onac, Effenberger and Breban, High-temperature and “exotic” minerals from the Cioclovina Cave, Romania: a review, Studia Universitatis Babeş-Bolyai, Geologia, 52, 3- 10, 2007).
2. The discovery of hydroxylellestadite at Cioclovina Cave is equally unique. 
3. Berlinite forms at 583 °C ( [I was able to find no empirical data on the temperature at which hydroxylellestadite forms].
4. Bat guano degrades normally to form phosphate minerals including, but not limited to Brushite, ardealite, monetite, Taranakite, variscite, crandalite, sasaite, hydroxylapatite, whitlockite, vivianite, Hopite, sampleite, and churchite (Encyclopedia of Caves, 2nd Edition, by W.B. White and D.C. Culver, Academic Press, 2012).
5. Under great pressure and very high temperatures bat guano will form Arnemhite, Berlinite, and Pyrocoproite [and is now known to form hydroxylellestadite] (Encyclopedia of Caves, 2nd Edition, by W.B. White and D.C. Culver, Academic Press, 2012).
Berna et al. have argued that because no Berlinite was found at Wonderwerk Cave they were able to rule out spontaneous combustion of bat guano at the source of the heat that altered the floral and faunal remains they examined. However, the same authors have determined that at Wonderwerk Cave combustion occurred at temperatures below 550 °C. Mineral science teaches us that Berlinite and hydroxylellestadite form at temperatures at or above 583 °C. No such temperature was reached in Wonderwerk Cave. Ipso facto, prest-o change-o, Berna et al. have erred in presuming that Berlinite would have formed at Wonderwerk Cave.
     Berna et al. state the following
Moreover, micromorphology and mFTIR did not show evidence for remains of guano and/or high-temperature phosphate mineral phases (i.e., berlinite and hydroxylellestadite); these minerals characteristically form during spontaneous combustion of bat guano—a rare event but one documented inside caves. 
I would be verrry interested to see the complete results of the FTIS analysis of the Wonderwerk sediments. Given that there are Chiroptera remains in the sequence it's likely that there would have been at least some guano lying around at some time during the hominid use of the the cave. And, as you prolly know, bats are known to aggregate in the gabillions in some caves. I'd be surprised to find NO evidence of the phosphate minerals that would belie the earlier presence of bat guano--burned or not.
     So, there you have it, Paul and Francesco [we've never actually met, Francesco, but I hope you won't mind me using your first name--they always told me I should act like a grown-up once I got my Ph.D., and so I try now and again to follow their advice]. 
     The gauntlet is thrown down. 
     Gentlemen! Show me the monetite!


  1. A rather tight composition I think. I admire your temper and civility. I don't know Drs. Goldberg or Berna. I'm not an academic but I have almost 4 decades of field experience and research, along with what some today consider the lowly Masters degree (although it wasn't so lowly in 1977 when I undertook the effort). I definitely follow your argument and after a little fact checking, I think you're more than justified in asking for further proofs especially on a subject that is as important as the origins of human behavior. I don't always agree with Sagan's comment: Any argument worth a shit merits reasonable proofs. If someone is going to make an argument for controlled use of fire a million years ago...well, then they had better have a mountain of evidence to support it; absent such, I cannot imagine how this article made it past peer review but then at the risk of sounding discourteous, an awful lot of what's published by PNAS seems a little contrived. Your criticisms should have come from a pre-publication peer review, which begs the question who did they have peer review the article? Obviously, we'll probably never know but damn...the author's a + b = c summation simply doesn't follow. They must either produce more evidence or retract their statement. Controlled fire in the Lower Stone Age? A million years ago? Really? Did they find evidence for alien intervention too? Oh there goes my big mouth. Shut up Roger.

    Love your blog by the way!

  2. Thank you, Roger. I really enjoy getting comments from the aether. I especially like 'em when they are from people who say I'm probably right! I fear you're right about PNAS. Not quite as sad a case as PLoS one, but definitely on the continuum. Thanks for dropping by.

  3. I would also point out that this story is absolutely shrouded in local politics - one mans lives work being systematically and swiftly stolen by another man. Having seen the "hearth" first hand, i am quite convinced it is genuine, but i too would like to see further proofs that it is controlled. ahhh the endemic issues of south african archaeology :-)

  4. I've really enjoyed your comments on the politics of South African archaeology today, Dave.

  5. Hi Rob,

    With all due respect, having read the article I think your critique is really just a torching of a strawman. The authors are actually more cautious in their conclusions than you suggest; their exact conclusion being: "This
    association strongly suggests that hominins at this site had knowledge
    of fire 1.0 Ma." Albeit, the "strength' of their "suggestion" is debatable (but that is the nature of archaeological discourse, see your own blog for definition of these probability qualifiers). Furthermore, I don't see any evidence that the authors are oblivious to the temperature at which berlinite forms; it seems to me that their point is that the lack of high-temperature phosphates supports their argument that this was not a bat guano fire (which typically burn at much higher temperatures than exhibited here). The absence of bat guano in the deposits adds further to this argument. I'm not suggesting that they have proven anything, but proof is a very very rare thing in archaeology. We construct arguments and engage in debate.

  6. Thanks, Bob. Please. Where do you get the idea that guano combustion 'typically' produces a certain temperature. Please. I'm dyin' for a reference. As for the authors' 'cautious' inference...they don't seem to be making much of a fuss about the media frenzy that fastened on the inference, regardless of it's understated elegance. As for the presumed 'absence' of guano in the deposits, they give no substantive evidence for that conclusion other than the absence of Berlinite. With all due respect, W.D., I think you're being too charitable. Oh, and, reference, please ;-)

  7. I have no wish to be recognized as an expert in bat guano. I'll admit my understanding of bat guano, and Berlinite for that matter, is pretty much limited to the Onac et al. article you yourself cite above. There and/or in other reports by the same lot, it is stated that bat guano burns at temperatures in excess of 500 C (they cite Martini 1994?, a report that I have no access to) and furthermore that the Berlinite they observed formed at 550 C.

    A little research into Berlinite also reveals that it can form (albeit, in an artificial environment) at temperatures as low as 130 C.

    I will acknowledge that having now looked at the literature on Berlinite, it seems silly that the archaeologists would employ it in their argument, and what I called your strawman may well instead be their house of cards.

  8. That's the Bob I grew to know and like! I thought as much about the Onac business. That guano burned at extremely high temp because it was compressed. If you look at his paper it looks fused. Not your average guano fire (I am, now, officially, an expert, 'cause I think I've read everything there is on the matter). As for the low temp. I had that up my sleeve for a later round if this bout went on for much longer. Thanks for taking the time to do what was necessary to realize I was right all along! ;-) You had me going there!

  9. Well, I don't recall ever saying you are right ;p I still think your reaction is a bit overblown. Their claim that H. erectus had "knowledge of fire" is not something that keeps me awake at night. My dog has knowledge of fire. She knows that a wiener that falls into the fire can only be retrieved by barking at me incessantly for several minutes.

    But the more I think about the Berlinite nonsense, the more I wonder why they even mention it in their article. Given that the discovery of natural Berlinite is itself still contested and further the assertion that it was a "bat Guano fire" that created it seems to be completely unsubstantiated. For all they know it may have been a group of Nenaderthals engaged in some sort of elaborate group mortuary ritual that got out of hand resulting in the cave fire. Or maybe the Neanderthals were trying to forge Berlinite. Hmm... I think there may be an honours thesis in there somewhere.


Thanks for visiting!

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