Fishy story from Koobi Fora

I have to credit a reader for that headline, and for forwarding the paper. It's another case of the infamous PNAS release policy. The press that came from the paper's announcement preceded the paper's availability in this case by a week. That approaches the case where a Hollywood studio won't screen a movie for reviewers before it's released. That means no reviews, which in the case of movies can only mean one thing. It's bad.

Scientific papers fortunately don't suffer from this shortfall -- the quality of the paper seems more or less unrelated to the release policy of the journal. In this case, the press went with a story that is interesting, but not necessarily that important in the scheme of things. And I don't get to write about it until two weeks after the news stories hit the presses.

David Braun and colleagues report on the fauna at locality FwJj20 of Koobi Fora, Kenya Braun:aquatic:2010. The archaeological remains here, including stone tools and fauna, date back to 1.95 million years. It's an interesting time because of what may have been going on with hominin anatomical evolution, but does it represent anything new in behavioral evolution?

The authors point out that there are archaeological sites that are much older, going back to 2.6 million years. Some of those earlier localities -- notably, the earliest, Gona OGS 6 and OGS 7 localities -- have hundreds of stone artifacts combined with fauna and hominin-modified bones. FwJj20 stands out in combining a very large number of stone artifacts (2633) with a high proportion of hominin-modified bones (5.9 percent of 405 faunal specimens). Even in later deposits such as Olduvai Gorge that have a high number of localities with some stone tools, it is rare to find localities with evidence of butchery of many animals. Those are the kinds of archaeological debris that would be expected of a real focus of hominin behavior. So every additional site like this adds substantially to our knowledge of hominin behavior at the dawn of hunting and gathering.

Here, one interesting aspect of the faunal exploitation is the small amount of surface modification consistent with bone-smashing. The authors suggest that the site had little marrow extraction than expected based on experimental replication of butchery. There is very little evidence for carnivore activity at the site, and both bones and faunal remains are clustered within a small vertical horizon of around 6 inches in thickness. The presence of small flakes and bone fragments helps to substantiate that the site did not accumulate under the influence of high-velocity water flow, and that it represents a primary activity locus for the hominins who left the tools there.

The faunal assemblage is interesting for the relatively high proportion of aquatic animals preserved, including both turtle and crocodile bone specimens with cut marks, and some fish bones. This is the part of the paper emphasized in the press that described the site, and the paper gives a good summary of the aquatic proportion of the fauna, including the evidence that the animals were actually butchered by the hominins.

The skeletal representation of fish bones [over-abundance of cranial fragments: 64% of fish NISP (28)] and turtle/tortoise bones [over-abundance of carapace and plastron fragments: 90% of turtle/tortoise NISP (29)] corresponds to ethnographic and archaeological distributions associated with hominin foraging. The number and taxonomic diversity of hominin-modified bones imply that hominins used the FwJj20 locality for the acquisition of meat from several different carcasses of terrestrial and aquatic animals as well as marrow from mammalian bones. This provides strong evidence of a diverse animal component in the diets of hominins before the appearance of H. ergaster/erectus (Braun et al. 2010:10004).

But....I think that the relevance of the aquatic animals has been exaggerated. According to the MNI (minimum number of individuals) table in the paper, the turtle and crocodile bones may represent one single turtle and one crocodile. The number of fish bones is also very small -- only 15 total, and the authors do not provide an MNI for fish. Compare these small numbers to a minimum of 11 hippopotamus individuals represented by in situ bone elements, and 17 bovids. One turtle. Seventeen bovids.

MNI is not the best indicator of dietary importance -- for mammals, it is heavily influenced by mandibles and teeth. Humans may drag mandibles back to a central place as part of the head, even if they eat the rest of the animal elsewhere. Being highly diagnostic, we can work out easily when there were lots of individuals from a mandible -- not so for broken turtle carapace pieces. But it's not very meaningful to count every crocodile bone, either. The site really does not provide any evidence that reptiles and fish simply made up a large fraction of the meat consumed there.

From my perspective, I think that's just fine. Aquatic animals aren't important because of their sheer numbers, but because they tell us about the flexibility of foraging behavior. Living hunter-gatherers eat turtles and reptiles when they can, and because they are usually small food packages, they often eat them where they find them instead of returning to a base camp first. Hunter-gatherers are flexible in what they eat and where they eat it. FwJj20 is showing at least a substantial taxonomic flexibility in the meat-eating of early Oldowan hunters.

Croc, turtle and fish remains also document that the Oldowan-makers were actively foraging in and around river or lake margins. That may not be earth-shaking, since we are, after all, talking about a water-dependent primate in a hot climate. But sometimes the importance of an archaeological discovery is that it strikes a "couldn't have done it" from the record.

Still, this really isn't a case where anybody could credibly maintain that early hominins were excluded from foraging on lake or river margins. Just last year I discussed two archaeological sites that give evidence for human exploitation of aquatic resources in the Early and Middle Pleistocene. At Trinil, Java, it seems clear that people were exploiting molluscs ("The shells of Trinil"), and the somewhat later Gesher Benot Ya'aqov site in Israel has evidence of systematic fish and crab exploitation ("The fishy spaces of the Middle Pleistocene"). The possible exploitation of papyrus by A. boisei also would show a mastery of shoreline habitats by hominins. It's hard to argue that the threat of the water was lower for robust australopithecines than for Homo.

Finding such repeated evidence of aquatic resource use, extending back near the dawn of stone tool manufacture, ought to prove one thing: The fatty acids in aquatic meat were not the cause of the expansion of brain size in Homo erectus.

Oh, I know, the news stories all said exactly the opposite, claiming that the fatty acids were essential to brain growth, and that this shows that stone tools were important to getting this essential nutrient. Hey, Braun and colleagues started it -- they wrote it right in the last sentences of the paper:

In addition, although animal tissues provide nutrient-rich fuel for a growing brain, aquatic resources (e.g., fish, crocodiles, turtles) are especially rich sources of the long-chain polyunsaturated fatty acids and docosahexaenoic acid that are so critical to human brain growth (2). Therefore, the incorporation of diverse animals, especially those in the lacustrine food chain, provided critical nutritional components to the diets of hominins before the appearance of H. ergaster/erectus that could have fueled the evolution of larger brains in late Pliocene hominins (Braun et al. 2010:10005).

But "fueled" is a metaphor, not a valid evolutionary concept.

I accept that reptile and fish meat may be nutritionally desirable. The question is whether they caused the increase in brain size associated with Homo. One way to read that hypothesis is as Lamarckism, which is simply wrong (Larry Moran has commented on that topic). I don't think that any paleoanthropologists are seriously Lamarckist, but some need to be more careful how they describe the relationship of fitness and diet.

Let me construct a version of the hypothesis consistent with evolutionary biology. Suppose that other factors -- social competition, technological requirements -- induced selection for cognitive skills in early Homo. The response of the population to this selection may have been impeded by selection in favor of smaller brains and/or shorter life histories. That is to say, directional selection on cognition may have been impossible because of stabilizing selection on brain growth. Now diet changes might become relevant, by relaxing the stabilizing selection on brain growth. This scenario might predict an increase in the size of the brain when people began to consistently supply themselves or their children with the right nutrition.

Understand that I don't subscribe to this hypothesis. We have much to learn about what the "right" nutrition might be.

But the hypothesis is testable. The archaeology now suggests that significant meat consumption preceded the expansion of the brain by a half million years or more, and that fish and reptile meat made up a hunter-gatherer-like part of early hominin meat consumption from the start.

Now it could be that later increases in diet quality -- for example, by increasing the total amount of meat, or decreasing nutritional unpredictability -- are what actually caused (or allowed directional selection on) the increase in brain size. That change would be a different hypothesis, however -- the hypothesis that selection against larger brains was relaxed by behavioral innovation. Fish fat could be a correlate of behavioral change in this hypothesiss, but it would not be the cause.