Who really killed Shanidar 3?

Fun with footnotes, today at AVRPI!! You'll remember that a couple of summers ago, a study by Churchill et al. (2009) tried to argue that the cut marks on a rib from the Shanidar 3 Neanderthal were the result of a wound inflicted by a modern human on that poor sap. Naturally, the science press had a field day with this, although several commentators argued that the evidence presented by Churchill and co. had been stretched way too thin, and that there really was no way to know who (what?) had killed Shanidar 3.

I just finished reading a paper by Trinkaus and Buzhilova (in press) on the death of the Sunghir 1. Sunghir 1 was an old man dating to the Early Gravettian (somewhere b/w 28-24kya) who was buried with an extremely lavish set of grave goods, including ornaments comprising several thousands of mammoth ivory beads, ivory bracelets, a schist pendant and a lot of red ochre spread over the entire burial. There is no question that this is one of the most remarkable Upper Paleolithic burials know.

How does that relate to Shanidar 3, you ask? Well, beyond the fact that Shanidar 3 has been argued to be an intentional burial, Trinkaus and Buzhilova report that, during a renewed inventory of the Sunghir 1 remains undertaken in 2009 which included a careful cleaning of the remains, they identified "an oblique defect, medial-caudal to lateral-cranial, on the left ventral corner of the body" of the Sunghir 1 first thoracic vertebra (T1). I'll return to the details of the Sunghir 1 injury in an upcoming post, but suffice it to say here that they are able to show that the lesion is indicative of a wound inflicted by a sharp implement and that would have been lethal to Sunghir 1.

OK, so returning to our dead homies Neanderthals, this study provides the third most ancient case of a weapon-inflicted wound leading to the death of a Paleolithic forager, the other two being Saint-Cesaire 1 and Shanidar 3, in decreasing order of age. Both of these have been described by Churchill et al. (2009) has resulting from inter-specific violence by modern humans on Neanderthals. However, Trinkaus and Buzhilova argue that "[i]n neither of them is there sufficient evidence, given current geochronology and available technology to meaningfully hypothesize intergroup aggression.". Their justification for this different assessment is then provided in one of the most epic and detailed footnotes I've come across, which I quote in full here:

Churchill et al. (2009) have argued that the injuries to Shanidar 3 and Saint-Césaire 1 are likely to have been perpetrated on these Neandertals by early modern humans. They argue for probable regional sympatry of Shanidar 3 with early modern humans in southwest Asia and certain sympatry for Saint- Césaire 1 in western Europe and superior projectile technology among early modern humans with respect to the Middle Paleolithic Shanidar 3. However, their argument, despite caveats, requires distortions of the relevant geochronology, misrepresentation of the available technology, and special pleading. In western Europe, there is no evidence for (presumably modern human associated) Aurignacian levels stratified below those of the (Neandertal associated) Châtelperronian (Bordes, 2003; Zilhão et al., 2006), as it is at Saint-Césaire (Lévêque et al., 1993), and all of the reliable dates place the Châtelperronian prior to the Aurignacian (Zilhão & d’Errico, 2003). The one radiometric date for the Saint-Césaire Châtelperronian level, a TL date (36 300±2700 cal BP) (Mercier et al., 1991), has a sufficiently large standard error to make it inappropriate to date the burial relative to Châtelperronian or Aurigiacian levels in the region. There is therefore no evidence, either paleontological or by assuming that the earliest Aurignacian was made by modern humans, that there were modern humans in western Europe at the time of Saint-Césaire 1. Churchill et al.’s assessment of the relative ages of Shanidar 3 and early modern humans in southwest Asia confuses radiocarbon and calendar years and makes unwarranted assumptions of who was responsible for which technocomplex; a reassessment of the available dates for diagnostic human remains, plus the stratigraphic position of Shanidar 3, clarifies the chronology. Shanidar 3 derives from near the top of Level D of Shanidar Cave, but stratigraphically well below the radiocarbon dates of ~47 and ~51 ka 14C BP (~51 and ~56ka cal BP) (Trinkaus, 1983). The youngest Middle Paleolithic modern humans within southwest Asia (at Qafzeh and Skhul) are MIS 5c in age (~90–100 ka cal BP) (Valladas et al., 1988; Stringer et al., 1989), and hence much older. Modern human remains do not reappear in southwest Asia until at least 35 ka 14C BP (~40 ka cal BP) (Bergman & Stringer, 1989), ~15 000 years later. In the Zagros the Baradostian technocomplex, the more recent phases of which are associated with modern humans (Scott & Marean, 2009), is dated to ~36 ka 14C BP (~41 ka cal BP) (Otte & Kozlowski, 2007). In addition, contra Shea & Sisk (2010), there are no diagnostic human remains associated with the eastern Mediterranean littoral IUP, that is ~35 ka 14C BP (~40 ka cal BP); Qafzeh 1 and 2 are undated, Ksar Akil 1 is younger, and the few IUP Üçağızlı teeth may well be Neandertals (Neuville, 1951; Bergman & Stringer, 1989; Gulec et al., 2007). One must go to equatorial Africa to find roughly contemporaneous modern humans (Haile-Selassie et al., 2004). With respect to technology, either Châtelperronian or Aurignacian lithics could have inflicted the frontal wound on Saint-Césaire 1. Although Middle Paleolithic spears appear to have mostly had relatively thick lithic points (Shea, 2006), thinner tools capable of producing the Shanidar 3 injury are represented in the Shanidar (and southwest Asian) Middle Paleolithic (Skinner, 1965). Moreover, one has to go to southern Africa to find evidence for contemporaneous ‘advanced’ projectile weaponry (Shea, 2006; Lombard & Phillipson, 2010; but see comments and caveats in Villa & Soriano (2010) and Lombard & Phillipson (2010)). Therefore, contra Churchill et al. (2009), the Shanidar and Saint Césaire Neandertals had the technology available to inflict their respective wounds, and there is no evidence (direct or indirect) for synchronous and sympatric modern humans. It is inappropriate to infer that individuals responsible for the Shanidar 3 and Saint-Césaire1 injuries were other than Neandertals." (Trinkaus and Buzhilova, in press: 7).

So, in a nutshell, there is no good reason to assume that the wounds sustained by Shanidar 3 were inflicted by modern humans. In fact, all of the available evidence points to Shanidar 3 having lived at a moment when only Neanderthals were kicking around the Zagros, and that they had access to technology that could well have left the mark found on the Shanidar 3 ribs.

Why does this matter? Well, for one thing, setting the record straight on this destroys any evidence for the interactions between Neanderthals and modern humans having been strictly inimical and violent. For another, it provides up a fascinating and heretofore underappreciated glimpse into the range of interpersonal relations Neanderthals could have had with other Neanderthals. Given the tendency by many to see Neanderthal behavior has homogeneous and monotonous, emphasizing that their interactionswith others of their kind were occasionally violent to the point of being lethal contributes to showing them to have been all too humans in certain respects.

References:

Churchill, S., Franciscus, R., McKean-Peraza, H., Daniel, J., & Warren, B. (2009). Shanidar 3 Neandertal rib puncture wound and paleolithic weaponry Journal of Human Evolution, 57 (2), 163-178 DOI: 10.1016/j.jhevol.2009.05.010

Trinkaus, E., & Buzhilova, A. (2010). The death and burial of sunghir 1 International Journal of Osteoarchaeology DOI: 10.1002/oa.1227

The raw and the cooked, caveman redux

A few months ago, Henry et al. (2011a) published a truly remarkable study that analyzed the phytoliths and starch grains that had gotten encrusted in the dental calculus (i.e., plaque) of three Neanderthal individuals, two from the site of Spy (Belgium), and another from the site of Shanidar (Iraq). Their study provided the first direct evidence that plant foods were an integral part of the Neanderthal diet by identifying phytoliths (the microscopic silicate 'skeletons' of plant parts that are unique to each plant species) of date palms in the calculus of Shanidar III, and starch grains attributed to various species of wild barley and legumes. The Spy I and II specimens yielded no phytoliths, but did yield starch grains that indicate they consumed water lily corms as well as some variety of sorghum and five other types of plants. The fact that these plant microfossils were found on the teeth of both Spy specimens is a strong indicator that these plants were part of the diet of that group as a whole, too. And the fact that Neanderthals from such distinct ecological settings preserve direct evidence of plant consumption suggests that it was a widespread dietary behavior, an observation that contrasts markedly with the 'Neanderthals as super-carnivore' idea that's been growing in the literature.

As if that wasn't enough, however, some of the starch grains found in the calculus of Shanidar III also had a peculiar morphology ("partly gelatinized") that matched that of experimental cooked barley starch grains. Further, Henry et al. (2011:487) observe that "The overall pattern of damage to the starch grains matches most closely with that caused by heating in the presence of water, such as during baking or boiling, rather than “dryer” forms of cooking like parching or popping (38). The finding of cooked Triticeae [barley] starches on the Shanidar teeth reinforces evidence from other studies that suggest that Near Eastern Neanderthals cooked plant foods." In other words, not only can we tell that Neanderthals cooked some of the plants they consumed, we can even get an idea of how they cooked them, most likely through boiling or baking. To me, this is doubly (triply?) neat because it also tells us something about the cooking technology this very likely required, namely some kind of cooking vessel, made out of either leather or wood.

However, Collins and Copeland (2011) now have a letter in press at PNAS that argues that Henry et al. (2011a) did not properly account for alternative manners in which starch grains can become partly gelatinized through various forms of diagenesis, that is post-depositional chemical or physical alteration, or what we commonly think of as decomposition. This diagenesis could have been triggered by both the presence of water in the cave sediments and/or their exposure to high heat.

I'm happy that PNAS actually gave Henry et al. (2011b) a chance to respond to this letter at the same time. In their reply, they point out, first,  that the heat and humidity required to prompt 'spontaneous' gelatinization would be extremely unlikely to be present in cave sediment. They also indicate that complementary analyses of starch grains on stone tools from some of the same contexts from which teeth associated with altered starch grains are not altered, in contrast to what would be expected if these processes were affecting whole archaeological layers. Specifically, it suggests that lithics were used to process uncooked plant matter that was then cooked and later consumed by Neanderthals. Overall, the evidence therefore continues to indicate that Neanderthals not only ate plants, but also cooked them to both facilitate their consumption and increase their nutritiousness.

References

Collins, M., & Copeland, L. (2011). Ancient starch: Cooked or just old? Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.1103241108

Henry, A., Brooks, A., & Piperno, D. (2010). Microfossils in calculus demonstrate consumption of plants and cooked foods in Neanderthal diets (Shanidar III, Iraq; Spy I and II, Belgium) Proceedings of the National Academy of Sciences, 108 (2), 486-491 DOI: 10.1073/pnas.1016868108

Henry, A., Brooks, A., & Piperno, D. (2011). Reply to Collins and Copeland: Spontaneous gelatinization not supported by evidence Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.1104199108