stone tools

UPDATE (2011-09-06) Note: The conclusions of the research were later critiqued, I posted on that criticism after this post.

Shannon McPherron, Zeresenay Alemseged and colleagues working at the Dikika field site in Ethiopia have found evidence of stone tool use 3.39 million years ago [1]. That's 800,000 years earlier than the previous first-known tool use, and occurs during the existence of Australopithecus afarensis.

The evidence is a series of cutmarks and one percussion mark on two bovid bones. One is a piece of rib from a large "cow-sized" animal, the other a femur fragment from a smaller "goat-sized" bovid. The analysis goes through several microscopic comparisons to rule out alternative causes for the cutmarks, such as trampling. The key paragraph of the results:

The cut marks demonstrate hominin use of sharp-edged stone to remove flesh from the femur and rib. The location and density of the marks on the femur indicate that flesh was rather widely spread on the surface, although it is possible that there could have been isolated patches of flesh. The percussion marks on the femur demonstrate hominin use of a blunt stone to strike the bone, probably to gain access to the marrow. The external surfaces of ribs have thin sheaths of flesh, so the scraping marks on the fossil rib suggest stripping off of these sheaths.

I have some lingering doubts, none of which are very serious, but that point out the need to look harder at other sites. It sure would have been nice if they'd found an anomalous sharp-edged rock nearby.

The two bones are compelling, but the study does not give much indication of how representative they are. How many similar-sized bone fragments were left at the site? How many were collected? What fraction of "cutmarked" bones does that make? What fraction show signs of trampling and various kinds of post-depositional damage?

Those questions are essential to answer the "green car" problem. If you don't know this one, it's fairly simple -- a witness reports a green car leaving the scene, and green cars are very rare -- the police think this is a great lead. But blue cars are very common in the city, and there is a small chance that the witness mistook a blue one for a green one. Whether it actually was a green car depends on the actual proportion of green to blue cars, and the actual probability that the witness was wrong.

In this case, I think there is a very small chance that the marks on these bones could have been produced by processes other than deliberate cutting by a stone tool. But in a sample of hundreds or thousands of bone fragments, a small chance might well happen a couple of times. It's very difficult to quantify this, because archaeologists don't collect every bone fragment. The only real way to address the problem is to find more cutmarks and do other statistics on them -- do they occur where flesh is attached to bone, etc.

It does seem odd that nobody's identified clear stone tools, which are in later sites a lot more common than cutmarked bones. A tool-user will make many artifacts during her life. (Why "her"? Well, in chimpanzees, it's the females who dominate technology transmission...) We have a lot of australopithecine bones. If this was a long-lasting tradition, we should have found a lot of stone tools by now.

Maybe it wasn't a long-lasting tradition. Chimpanzee technology is significantly clustered geographically, some of the most interesting behaviors have been observed only at a single field site. If Australopithecus had a similar pattern of cultural diversity, tool use may have been innovated many times without "catching on" over a wide geographic or temporal extent. Here's what McPherron and colleagues conclude along similar lines:

Whether A. afarensis also produced stone tools remains to be demonstrated, but the DIK-55 finds may fit with the view that stone tool production pre-dates the earliest known archaeological sites and was initially of low intensity (one-to-a-few flakes removed per nodule) and distributed in extremely low density scatters across the landscape such that its archaeological visibility is quite low (16).

Or maybe we just haven't noticed. Fluvial contexts may have been bad places for Australopithecus to hang out. McPherron and colleagues allude to this explanation for the local absence of tools at Dikika:

However, stone tool production and consequently archaeological accumulations are not expected at this locality given the sedimentary environment characterized by the palaeo-Awash River emptying into a nearby lake (3, 4). In this relatively low-energy depositional environment, clasts suitable for stone tool production are not present (few particles larger than fine gravel, 8 mm diameter). Within the exposed SH Member, the distance from DIK-55 to cobble-sized raw materials (>64 mm) is ~6 km (at Gorgore; Fig. 1). Thus, in this instance the absence of evidence for stone tool production in the immediate vicinity of the cut-marked bones may reflect landscape-level raw material constraints.

The research article is accompanied by an essay by David Braun reviewing the find [2]. He stretches a bit, but I think the interpretations he suggests are worth airing. One -- why are there cutmarked bones 6 km from any good source of stone raw material?

The meat and marrow of large animals must have been a valued resource, because McPherron et al. conclude that the tool users incurred the cost of transporting stones 6 kilometres from where they occurred naturally to the site where the butchery took place. Further costs that were associated with the consumption of carrion, and were apparently worth the risk, include exposure to parasites and competition with large carnivores.

Two -- what about the "meat-brain" connection?

This provides exciting evidence of how A. afarensis behaved. At one time, the species was considered to be a relatively primitive hominin, but this perception is being redefined. For example, it now seems that Lucy's kin had body proportions that were more similar to those of humans than of apes (6). Analyses of the hand of A. afarensis show that it had relatively short fingers that would allow the kind of fine-scale manipulation necessary for tool use (7). A recently discovered skeleton from the Woranso–Mille area of Ethiopia suggests that A. afarensis did not have the ape-like, 'funnel-shaped' thorax usually associated with a large digestive tract and low-quality diet (8). Perhaps the findings that these hominins used tools and had a carnivorous component to their diet should not have been so unexpected.

A 2.6-million-year-old butchery tradition should already have refuted the hypothesis that meat-eating caused the expansion of brain size in Homo. But it was still possible to maintain that the initial Oldowan was insufficiently dedicated, or that the anatomical specializations (e.g., small guts) allowing brain expansion took time to develop, or that as-yet-undiscovered large-brained hominins would be found. Any of these are still possible, but the observations Braun points out pretty much demolish the 15-year-old story of "expensive tissue." Australopithecus seems to have had a small gut, and a bigger brain than chimpanzees. If there was a tradeoff, A. afarensis had already made it.

Braun didn't mention A. sediba, which adds another wrinkle. A late species of Australopithecus with human-sized teeth. Or (as some prefer), a pre-habilis species of Homo with an Australopithecus-sized brain. What was its diet like? I have a feeling we'll know before too long.

Meanwhile, I'll be floating for the rest of the year, since I included this as the far-out "bonus" entry in my 2010 New Year predictions! You know, the one that's so bizarre that it seems like it'll never happen. Heh.

UPDATE (2010-08-11): John Noble Wilford got ahold of some skeptics for his NY Times story on the discovery:

Still, the discoverers are already being pressed to defend their interpretation that the cut marks on the bones are evidence of stone-tool butchery. Tim D. White of the University of California, Berkeley, one of the foremost investigators of early human origins, said flatly that their “claims greatly outstrip the evidence,” and noted, “We have been working sites in this area for 40 years, and not a single stone tool has been found in deposits of this antiquity.”

Sileshi Semaw, a paleoanthropologist at Indiana University who was a discoverer of the oldest confirmed stone tools, from 2.6 million years ago, noted in an e-mail message from Ethiopia that researchers had often been misled by bone markings left by trampling animals and other natural causes. “I am not convinced of the new discovery,” he said.

UPDATE (2010-08-12): Maybe some are looking for more about australopithecine diets. My post from 2005, "Chemistry and early hominid diets" has a good compilation of stable isotope observations and what may explain them. With the cutmark evidence, you can read through the discussion of C₄ plant contributions, and think about the grazers that A. africanus may have been eating.

UPDATE (2010-08-16): Science Friday with Ira Flatow covered this story last week, including commentary by Alemseged and David DeGusta, who suggests that the marks may be crocodile bite marks. Doesn't look like it to me, but as I wrote above, I'd like to see statistics on a few hundred damaged bones to see the probability that an arbitrary one will look like stone cutmarks.

References

A new paper is pushing back the time of initial occupation of Flores by hominins to at least 1.0 million years ago. Adam Brumm and colleagues (2010) are reporting that they've found stone tools in a site from the Soa Basin of Flores, the same geological region as the previous site of Mata Menge.

The Wolo Sege excavation yielded no faunal remains, but 45 in situ stone artefacts were recovered from the conglomerate and two fine-grained metavolcanic flakes were excavated from the lower tuffaceous siltstone layer ~15–20 cm above the Ola Kile Formation (Fig. 3e, f). A single volcanic flake was also recovered from the upper overbank deposit during extraction of sediment for dating. The Wolo Sege stone artefacts are predominantly small and morphologically undifferentiated flakes struck from cobbles by direct hard-hammer percussion (Fig. 3; see also Supplementary Fig. 2), but include a bifacially and centripetally worked ‘radial’ core, similar to those characteristic of the Mata Menge assemblage of stone artefacts.

The radial core is not illustrated, but several of the flakes are figured in the paper. The conglomerate in question is overlain by a layer with a minimum date of 1.02 million years.

A date of 880,000 years ago for human occupation made for a convenient explanation of faunal turnover on the island, which happened around that time. The turnover included the extinction of the small pygmy stegodont species Stegodon sondaari, which was replaced in later faunal assemblages by the Java-derived Stegodon florensis. It also included the extinction of giant tortoises.

This suggests that the non-selective, mass death of S. sondaari and giant tortoise, associated with stratigraphic evidence for a major volcanic eruption at Tangi Talo ~0.9 Myr ago10, could represent a localized or regional extinction, and that the faunal turnover may have been a result of climate change, volcanic activity or some other natural process or event (Fig. 5).

I discussed this with my graduate seminar yesterday. The long persistence of this toolmaking culture, in what must have been a rather small human population, is weighing on my mind. Were there recurrent contacts from Java, keeping the population going? How dependent were these people on their tools?

Hominin predators can lead to unstable dynamics -- most predators will undergo predator-prey cycles, but humans can switch to other resources and continue to press a small prey species to extinction. The long persistence of tasty animals on Flores in the presence of hominins suggests that the subsistence practices of these hominins were different in some ways from later humans.

This finding doesn't really help us to resolve the issues of the later Flores record, including the relationships of the skeletal individuals with other hominins. There's been some press about the hobbits lately, but it's all paleoanthropological tree-marking -- except for the news that they'll be reopening excavations at Liang Bua.

References:

Brumm A, Jensen GM, van den Bergh GD, Morwood MJ, Kurniawan I, Aziz F, Storey M. 2010. Hominins on Flores, Indonesia, by one million years ago. Nature (advance online) doi:10.1038/nature08844

Michael Balter writes about the work of Liverpool archaeologist Natalie Uomini, who is studying the evolution of handedness by experiment and attempting to find signs of hand dominance in finished stone tools.

Uomini argues that these findings [in human experiments] are similar to those found in studies of nonhuman primates, in which the degree of handedness varies depending on the complexity of the task they are facing and the level of skill required to perform it. The nut-cracking task required five basic actions, Uomini says—“grasp nut, place nut on anvil, grasp hammer, hit nut, and eat nut”—whereas the puzzle required at most three actions (grasp flake, grasp core, fit flake to core.) And along with other researchers who have offered a similar hypothesis, she suggests that the increasing sophistication of hominin toolmaking technologies over time may have selected for a greater degree of handedness. Indeed, Uomini says, the “technology-dense lifestyles” of early hominins might have required our ancestors to more or less make up their minds about what hands they were going to use to perform complex tasks. Moreover, such hand bias could have aided the learning process as hominins taught each other toolmaking and other skills; a number of studies have shown that people learn manually difficult tasks, such as knot-tying, more easily when they use the same left- and right-hand movements as their teachers.

In other words, once a procedure becomes sufficiently complex, it is likely to require asymmetric spatial operations that will be learned in a biased way from models who have mastered the procedure. But if a model bias is likely to exist, that's not sufficient to induce selection in favor of added bias. Random changes that reinforced the bias wouldn't be selected against, but that still leaves us wanting an explanation for the frequency of lefties -- as Balter and Uomini both point out.

I'm not sure that this technological explanation actually explains anything. Possibly the causation is in the opposite direction: right versus left handedness tends toward a given frequency for other reasons (maybe pleiotropy, maybe chance) and the use of hands in technology accommodates to the prevailing population mean -- to a greater or lesser extent for more or less complex procedural operations.

Still, it's a very interesting article about some clever ways of applying behavioral science in archaeology.