Fire in the hole: earliest evidence of human cooking.

April 4, 2012 • 5:59 am

You may be familiar with Harvard anthropologist Richard Wrangham’s book Catching Fire: How Cooking Made us Human, in which he posits all sorts of features of modern humans, including not only morphology (teeth and brain size), but also sociality, intersexual relationships, division of labor, and other forms of behavior were impelled by the discovery that we could cook food on fires.  I haven’t read this book but I have read about it, including an interview with Wrangham at Edge.  Based on this admittedly cursory reading, I think his hypothesis is a little simplistic (i.e., maybe easily digestible protein allows us more energy to evolved a larger brain, but there still must be selection pressures for that larger brain), but there may be something to it.  However, there are many other alternative theories for bigger brains (my favorite is social interaction and language), several of them may have operated simultaneously, and it’s hard to discriminate among them.

One piece of evidence against Wrangham’s hypothesis was, until now, the finding that enlargement of human brains appeared to have begun well before hominins started cooking.  Brain size enlargement in humans is traditionally throught to have begun its rapid phase with Homo erectus, about 1.5 million years ago, although there may have been an evolutionary speedup about 300,000 years ago. (Wrangham has also noted that human molar teeth underwent a relative reduction in size at that time, which one might expect if we were processing cooked rather than raw foods.)  But the first evidence for controlled use of fire was about 400,000 years ago.  There had been earlier reports of burned bones and vegetation associated with human presence from 1.5 million years ago, but those might have been due to wildfires rather than fires controlled by humans.  (Controlling is important if cooking is to be frequent enough to affect our evolution.)

Now, however,  a new paper in PNAS by Berna et al. (free download!) gives a pretty unequivocal report of controlled fire use about 1 million years ago, in “Wonderwerk Cave,” in northern South Africa:

The evidence for controlled fire use includes the fact that there are multiple findings well inside the cave of charred bone and burnt plant material.  The source of the bones wasn’t identified, but presumably they’re animals rather than human (no cannibalism!), and reflectance spectra of the bones that match the spectra of newly heated bones and differ from those of unheated bones.

The burnt plant material was not wood but grasses, leaves, or brushes, indicating that humans hadn’t yet perfected the use of cooking over logs (this of course meant that they had to feed the fire constantly).  The reflectance spectra suggests that bones were heated to a temperature between 400 and 500 degrees C, which is hot enough to cook stuff but not hot enough to cook a steak well done (but why would they want to?).  The highest temperature suggested by bone and plant scans is about 700 degrees C—not hot enough to be produced by wood.

Here are a couple of pieces of charred bone for your inspection (these are from the paper):

And here’s a bit of “ashed plant material” along with the bone fragments (don’t ask me which is which):

The authors conclude this:

Thus, our data, although they do not show evidence of constructed combustion features, as listed by Roebroek and Villa as a criterion of controlled burning (3), demonstrate a very close association between hominin occupation and the presence of fire deep inside Wonderwerk Cave during the Early Acheulean. This association strongly suggests that hominins at this site had knowledge of fire 1.0 Ma. This is the most compelling evidence to date offering some support for the cooking hypothesis of Wrangham(1).

Note how carefully the authors hedge their conclusion; this is good science, showing that conclusions are tentative.


Berna, F., et al. 2012. Microstrategraphic evidence of in situ fire in the Archeulean strata of Wonderwerk Cave, Northern Cape province, South Africa. Proc. Nat. Acad. Sci. USA: published ahead of print April 2, 2012, doi:10.1073/pnas.1117620109

For the earlier suggestion that human morphology of a million years ago had already been evolutionarily molded by the use of fire, see:
Organ C, Nunn CL, Machanda Z, Wrangham RW. 2011. Phylogenetic rate shifts in feeding time during the evolution of Homo. Proc Natl Acad Sci U S A. 2011 Aug 30;108(35):14555-9. Epub 2011 Aug 22.

31 thoughts on “Fire in the hole: earliest evidence of human cooking.

  1. Amusingly enough, “wonderwerk” translates roughly as “miracle” which is an unintentionally ironic name for a site that has been so valuable to our knowledge of early hominins.

    1. Well, it’s a afrikaans name, and it does translate to miracle. Wonderwork would be a direct translation I guess, but no-one uses that as far as I know.
      I could have sworn there was another Miracle cave somewhere round here, but that might have just been Wonder cave, which links to the Sterkfontein caves.

  2. Forgive me if I’m being stupid, but 400° C is 752° F, and the last time I checked my oven didn’t go that high. Wouldn’t that be plenty of heat to hockey-puck a steak Dawkins style? Did you mean Kelvin?

    1. Yeah, sorry. You could hockey-puck a steak, but you’d have to leave it on there a while. The lower temperature is simply evidence that they used leaves and grass rather than wood.

  3. I read Richard Wrangham’s book, and it is well written, I feel he overstates his case, due to an excess of speculation.

  4. Nice! I read Wrangham’s book a while ago, and loved it. Well-written, concise, and I found it convincing, despite indeed the lack of evidence for use of fire that early on – until now!

    1. Ditto. Loved it and found it convincing, though having been a cook for most of my life I may be prejudiced. The idea that it all started with cooking resonates with me. Also, the idea that my ancestors, a million years ago, were cooking food and sitting around socializing, makes me happy, for some reason.

  5. but there still must be selection pressures for that larger brain

    I would think that a brain that is complex enough to control fire could only arise within a context of pressure for a more complex brain and larger size allows for that complexity. The cost of a larger brain would attenuate the pressure. Cooking would simply lower the cost.

    It has interested me for some time that for a million years until 250,000 years ago, the human line had a steady increase in brain size of 200 cc. The rate of increase was maintained in both lines after the H. sapiens line and the H. neandertalis line diverged. 250,000 years ago, both lines simultaneously began to increase their brain capacities by 250 cc over the next 200,000 years. 50,000 years ago, with the H. sapiens population increasing and the H. neandertalis population diminishing, the increase in brain size ceased. For the last 20,000 years (after the H. neandertalis line went extinct), the brain size of H. sapiens has been decreasing. Perhaps that is from becoming domesticated.

    Has anyone put forth a theory about the selection forces on human brain sizes for the last quarter of a million years that explains the sudden acceleration of the rate of brain size increase in two different species? I suppose the sudden stoppage of the rate of increase could be because of the birth canal.

    1. In a simplistic sense, clearly selection criteria impelled larger brains didn’t go away when the lineages split (and there were almost certainly other splits we aren’t even aware of yet – look how long it took to find H. Floresiensis).

      I still favor runaway sexual selection, where the fruits of bigger brains in social interaction are increasingly desirable by those with bigger brains (i.e. bigger brains both desire and beget bigger brains).

      Just because Neanderthals didn’t leave as much in the way of durable cultural artifacts doesn’t mean they didn’t have equally sophisticated social lives. After all, they actually had bigger brains than us in the end.

  6. You missed Justin Zimmer’s point. Those temperatures on a conductive surface would burn a steak to carbon in seconds. Were they Kelvin degrees or just Celsius with an extra zero?

    1. No, I didn’t miss his point; I said that those temperatures could cook a steak (though they certainly weren’t on a conductive surface), but they are too low to have been attained by a wood fire. They are temperatures in degrees Celsius, as I said. You can consult the paper if you don’t believe me.

  7. I saw this on Science Daily, and thought it looked interesting, when I read the paper itself, one of the things that stood out to me was that it seems that they were using rocks, whether to contain the fire or to place the food on I could not get exactly from the paper, but that they say that the pot-lid fractures were consistent with temperatures over 500C would indicate that they were in close proximity to the fire rather than being fractured off of the cave walls by the heat, especially given the proximity of the site to the entrance (30m from present day entrance)

    1. I think one of their points was that these fires were not “manufactured” so leaving little evidence and not being compliant with earlier posited definitions of cooking practice.

      I assume the manuport flint was for making stone tools, resulting in a handy extra set of evidences for in-situ burning.

    2. Here: “Thus, our data, although they do not show evidence of con-
      structed combustion features, as listed by Roebroek and Villa as
      a criterion of controlled burning (3), demonstrate a very close as-
      sociation between hominin occupation and the presence of fire
      deep inside Wonderwerk Cave during the Early Acheulean.”

  8. Non-biologist that I am, I’ll venture a query.

    Could the large brain size have come about only to be maintained easily once our ancestors learned how to cook food?

    The large brain was selected for by some other means–like sexual selection noted above. But populations without cooking simply couldn’t keep up that growth without more protein.
    Weren’t most of our ancestor populations thought to be rather small. Maybe the big brain kept them so until cooking started?

    1. Interesting point, Lynn…I’m really not sure how much the discovery of cooking serves to elucidate things, although untestable narrative-style explanations are probably as good as it gets for events in the remote past.

    2. Interesting!

      I rather fancy the idea that the brain size was secondary to sexual and social selection for flat, large, easy to read faces with a nice show of hair jutting out over the forehead and showing health. (Why else would we retain hair and beards?)

      H. floresiensis may show that brain size isn’t all that critical for humans.

  9. “Note how carefully the authors hedge their conclusion; this is good science, showing that conclusions are tentative.”

    In the past I have occasionally caught myself mildly protesting what I have subjectively and irrationally perceived to be the (over)use of word “suggest(s)” in the concluding paragraph of a study. That is, I want the researcher(s) to be more declarative.

    But, of course, that’s nothing but an unreasonable and entitled “Whaah!” on my part. One can’t meaningfully comment beyond the current state of research. Re: “tentative,” “provisional.”

    The scientific ideal is that one should be rigorously self-critical. If you don’t want to be embarrassed, don’t make someone else have to do it for you. If you don’t, someone else surely will.

    I’m also reminded of Hitchens’s reflection (as others no doubt also have reflected), words to the effect that we are only now getting some rough idea of just how much we don’t know, though having learned quite a bit in the last few hundred years. As Hitchens said, “At least I know that I don’t know.”

    Something is not true simply and solely and merely because someone “says so.”

    Another aspect is not to toot ones own horn much if any. It may be true enough, as an old-timer in the Great Smoky Mountains put it, that “Hit ain’t braggin’ if hits thuh truth!” Still, it looks and sounds better if someone else does the braggin’ and tootin’ on ones behalf.

    Enough. Cheers!

  10. I read Wrangham’s book and it presents compelling evidence and some conjecture, and very well written. Newer evidence of controlled fire corroborates Wrangham’s well-presented thesis. Thanks.

    FWIW (regarding your more post), I come here for the science and not the religion and cat posts. I think you neglect the pervasive human condition that is linked to belief in supernatural phenomena and a more scientific exploration could be interesting. Yeah, I stopped believing in god-tings a long time ago, but I cannot get over how it permeates human culture. It’s the ultimate meme!

  11. A bit of speculation: Could the firing of meat have been a means of sterilizing it, rather than cooking it? As Acheulean hunter, if you wanted to take your slab of meat from the morning hunt with you for the day’s walk to another hunting site, would the meat spoil more slowly if charred on the outside? This might be a viable selection pressure in early cooking practice.

    1. Maybe some or all is a transitional cooking technique. A bit of smoke, charring et cetera adds taste. Just throw the pieces in the burning plants and rake them out after a while.

      That would have been the easier improvement, making the food more digestible and preserved further down the line and ushered in the need for wood, fire containment and steak stones et cetera.

  12. Jerry: “I haven’t read this book but I have read about it, including an interview with Wrangham at Edge. Based on this admittedly cursory reading, I think his hypothesis is a little simplistic (i.e., maybe easily digestible protein allows us more energy to evolved a larger brain, but there still must be selection pressures for that larger brain)”

    All this has been nodded to in comments above, but it bears clarification: Wrangham’s hypothesis is centered on the hypothesis that cooking provided the necessary calories that *allowed* extant selection pressures for larger brain size to work. Cooking didn’t create larger brains, it simply released the caloric energy necessary for it to evolve. As others have felt, I too found it a very persuasive circumstantial hypothesis, and I really think you ought to read it, Jerry, it’s a short and sweet book. I’d love to hear counter-arguments to his energetics argument – as presented, it seems rather killer.

    I wonder about potential genetic evidence for his hypothesis – perhaps an observed correlation with molecular evolution studies around digestive capacities?

Leave a Reply