Paleolithic research in the Armenian Highlands and Anatolia

I returned safe and sound from Anakara a couple of weeks ago (an unexpected overnight stay in Terminal 5 at JFK not withstanding) and am happy to report that the trip was well worth it. It was an intimate event, with 14 papers and perhaps 20-30 audience members. Some highlights:

• The first paper was by Amilié Vialet and her colleague M. Cihat Alçiçek, who discussed the Kocabaş skullcap. I was particularly interested to hear this talk, as I knew next to nothing about this important hominin fossil. The remains consist of the frontal and parietal of a hominin that were recovered in 2002 during quarrying activities in the travertines of the famous Pammukale (Turkish: "Cotton Castle") hot springs (and UNESCO World Heritage Site) in western Turkey. (I actually visited this area, which also includes impressive ruins from Greek and Roman times, on a vacation way back in 2001.) The top of the skullcap was literally shaved off by heavy equipment (unfortunately, the rest of the skull, nor any other part of the skeleton, have ever been recovered) but, thankfully, the recovered bones were brought to the attention of Alçiçek, who recognized their importance. John Kappelman and his colleagues had previously published on the remains and suggested that (1) they dated to ~500,000 years ago, (2) provisionally, they represented H. erectus; and (3) the individual suffered from the earliest known case of tuberculosis. Vialet (who, apart from being a world-class paleoanthropologist, is a genuinely nice person) summarized more recent work on the fossil. Perhaps most importantly, they push back the age of the remains using a variety of dating techniques to ~1.4 million years ago, which makes them among the oldest in all of Eurasia. An audience member asked Vialet about tuberculosis, and she vacillated, saying that a confident diagnosis from their team would need to await further analyses. Their 3D reconstruction of the cranium, however, suggested a closer affinity of Kocabaş to Homo ergaster fossils from Africa rather than early Homo (including the crania from Dmanisi) fossils or later, classic Homo erectus populations from Asia. Vialet suggested, then, that Kocabaş represents an "expansion [from Africa that is] different from that represented by the Dmanisi fossils". It is becoming increasingly clear that the movement out of Africa by hominins ca. 2 million years ago was a complex series of events (emphasis on the plural) that involved many populations, some of which were able to establish long-term occupations and others that quickly went extinct. The new information from Kocabaş certainly reinforces this idea.
• What followed was a series of papers by our Turkish colleagues that summarized the Lower and Middle Paleolithic of Anatolia. What struck me is that apart from a few notable exceptions (Kaletepe Deresi 3, Karaïn Cave, Yarımburgaz Cave, and Üçazğlı Cave), much of what is known is based almost exclusively on surface assemblages that lack key contextual information (chronometric dates, paleoenvironmental reconstructions, etc.). For example, Kadriye Özçelik reported on surface surveys that documented for the first time the presence of Paleolithic material in the Denizli Basin of southwestern Anatolia (this is the same region in which the Kocabaş fossil was uncovered). It is quite amazing, if you think about it, that it took until 2014 to officially recognize the presence of Paleolithic tools in this region, especially since, as I later learned, handaxes have been known in Turkey since 1896. While those of us working in Armenia faced a similar situation−that is, a dependence largely on surface finds−a great deal of progress has been made on this front in the last 10 years, due largely, I think, to the support of the Armenian Academy of Sciences, which has done much to nurture Paleolithic research in Armenia. My conversations with Turkish researchers revealed that it is extremely difficult to acquire permits to even examine and collect surface assemblages in Turkey, much less excavate stratified sites. Until that changes, I fear that our knowledge of Paleolithic settlement in modern Turkey will continue to lag behind that of other nearby countries.
• Our paper summarized work on Bagratashen 1, and we officially announced OSL dates from the Middle Paleolithic horizon of around 35,000 years ago. While Middle Paleolithic sites of this age are not unusual, the types of stone tools uncovered from the deposit look like artifacts from other parts of the southern Caucasus and the Levant that are much older, somewhere around 150,000 to 200,000 years ago (see my previous post on Bagratashen 1 here). So, either our dates are way out of whack (which is possible), or we have an example of technological convergence where culturally and temporally unrelated hominin groups independently converged on similar tool types. There is a long history in archaeology of viewing stone tools to track human populations. That is, particular cultures made particular types of tools, so if you find similar tools, you have the same people. However, folks have long pointed out that, as complex a technology that lithic manufacture can be, there are only so many ways that one can reduce a chunk of rock into different shapes. Dan Adler and his colleagues made precisely this point in their paper about their excavations at Nor Geghi 1 in Armenia. What they found is that at about 330,000 years ago, folks in Armenia were creating both bifacial artifacts typically associated with the Acheulean and Middle Paleolithic Levallois products at the same time. This shows definitively that, at least at Nor Geghi 1, making Acheulean bifaces has little to do with being part of an Acheulean "culture" that represents a particular group or species of hominins. Likewise, it seems that the appearance of Levallois technology, again at least in Armenia, need not necessarily represent a new group or species of hominin. Perhaps it is time, as Dan simulated with a humorous slide, to flush many of these stone tool "cultures" down the toilet.

I should also point out that Phil was able to bring together Turkish and Armenian researchers to discuss Paleolithic archaeology. This shows the power of science in general, and archaeology in particular, to transcend political and ethnic enmity and bring together people with common goals. I was honored to be part of such a momentous event.

Paleoanthropology and Paleolithic archaeology in the southern Caucasus (oh yea, and the new cranium from Dmanisi)

So, a bit more detail on the workshop in Tbilisi:

• The southern Caucasus is quickly becoming a hotspot for paleoanthropology and Paleolithic archaeology. Of course we have Dmanisi, but the entire region is cholk-full of Paleolithic sites. Western researchers are not as familiar with it, nor do they use it as much, as they should, mainly because most information is published in Georgian, Armenian, and Russian, and the data has generally been of relatively low quality. We are, however, rapidly moving towards the production of a very rich dataset for the region.  
• Tbilisi is a pretty, vibrant city, and the food is really good, particularly for a carnivore such as myself. The Georgian National Museum is also a beautiful, state-of-the-art facility.

Angela Bruch giving introductory remarks in the Georgian National Museum's lecture hall.

• Michael Maerker, a geographer at Tuebingen, gave a talk about using GIS predictive modelling to identify the location of paleo-lakeshores. This is really interesting work given how closely tied early hominins appear to be associated with these sorts of habitats. (He's also a really nice guy and fun to be around.)
• I presented our work on a couple Middle Paleolithic sites that we've started excavating in northern Armenia. One of the complaints I had was that we hadn't found any faunal material yet. Reid Ferring said that he had found several MP sites just north of the border in Georgia with bones, so there's hope I suppose...
• This was not a coincidence, I suspect, but the official announcement of the most recently discovered cranium (D4500) was made during the last day of the workshop. So, that brings the count of hominin crania from the site up to five now. This was particularly interesting since the newest cranium, found in 2005, apparently matches up with the mandible (D2600) that was discovered way back in 2000. David Lordkipanidze, the leader of the Dmanisi team, made the official announcement at the Georgian National Museum (it was all in Georgian, of course). They're basically calling all the specimens Homo erectus and, in fact, are suggesting that all early Homo specimens from this time period should also be referred to that species.  

David Lordkipanidze speaking to the press about the new Dmanisi cranium.

• The highlight of the workshop was, naturally, the visit to Dmanisi. The best part was that we were given a personal tour by Reid Ferring, who's been studying the archaeology and geology of the site for 20 years or so. gave us a fantastic tour of the site and a fantastic explanation of the geology. I have to admit that the site's stratigraphy had always been somewhat of a mystery to me, but Reid's on-site explanation really cleared things up (we had actually visited the site in 2012, but the excavation team was not there that day).

A shot looking down into what's referred to as the M5 Trench at Dmanisi. For scale, the total depth is over 6 meters.

Overall, a really wonderful experience.

Early hominin meat-eating and expensive tissues

I actually was not aware of its publication until I came across it on John Hawks's weblog, but Joseph Ferraro and his coworkers have come out with a detailed taphonomic analysis of the faunal remains from Kanjera South, an important Oldowan site from Kenya that dates to about 2 million years ago. Their abstract does a nice job of summarizing the significance of the site and the study's findings (I guess this is what abstracts are supposed to do, after all; Ferraro et al., 2013: 1):

The emergence of lithic technology by ~2.6 million years ago (Ma) is often interpreted as a correlate of increasingly recurrent hominin acquisition and consumption of animal remains. Associated faunal evidence, however, is poorly preserved prior to ~1.8 Ma, limiting our understanding of early archaeological (Oldowan) hominin carnivory. Here, we detail three large well-preserved zooarchaeological assemblages from Kanjera South, Kenya. The assemblages date to ~2.0 Ma, pre-dating all previously published archaeofaunas of appreciable size. At Kanjera, there is clear evidence that Oldowan hominins acquired and processed numerous, relatively complete, small ungulate carcasses. Moreover, they had at least occasional access to the fleshed remains of larger, wildebeest-sized animals. The overall record of hominin activities is consistent throughout the stratified sequence - spanning hundreds to thousands of years - and provides the earliest archaeological evidence of sustained hominin involvement with fleshed animal remains (i.e., persistent carnivory), a foraging adaptation central to many models of hominin evolution.

This research team has been working hard out a Kanjera for many years now, and its really nice to see a comprehensive analysis of the faunal material from the site (we'd been getting tantalizing hints in various publications and presentations for some time).

Before we proceed, let me summarize the state of affairs just prior to these latest data. The 1.8 Ma time marker that Ferraro et al. mentions refers to the burst of evidence for meat-eating that emerges almost exclusively from Olduvai Gorge in Tanzania. One site in particular, the very well-known Level 22 at the gorge's FLK locality (also known as the Zinjanthropus Floor), dates to about 1.84 Ma and preserves thousands of fossils, many of which bear clear indications of hominin butchery. Now, up until a few years ago, it was thought that the animal bones from many of the other sites from Beds I and II of the gorge (ca. 1.9-1.2 Ma) were also largely the result of hominin activity. However, my colleagues and I showed that there are really only two sites, the previously mentioned FLK 22 from Bed I, and the site of BK, in upper Bed II (about 1.3 Ma), that are largely the result of hominin butchery (Domínguez-Rodrigo et al., 2007, 2009; Egeland, 2008; Egeland and Domínguez-Rodrigo, 2008). Now, we're not saying that hominins weren't at the sites; they certainly made, used, and left stone tools at these locations, but they were not doing a lot of meat-eating. There are a couple of other Oldowan sites here and there with some evidence for butchery, but if we ignore FLK 22 for the moment, good evidence for lots of meat-eating (or, to use Ferraro et al.'s term, "persistent carnivory") really doesn't pick up until much later, perhaps about 1.5 Ma.

What does all of this have to do with expensive tissues? Well, researchers have come up with several well reasoned, and very popular, human evolutionary models based ultimately on the shift to meat-eating. To start, brains and guts are very expensive tissues: one does a lot of thinking and the other does a lot of digesting, both of which take up good amounts of energy. If you start eating more meat, which is nutrient dense and easy to digest, you can divert energy from the guts to develop bigger noggins. Other possible correlates of a diet based increasingly on meat would be increased range size (carnivores, and other animal that eat high quality, easy to digest foods, tend to have larger ranges) and unique life histories (extracting nutrients using technology, and hunting with technology in particular, are difficult things to learn, and you don't want to die before you learn how to do them well, so perhaps we've evolved extended life spans to fit this need). People have traditionally seen the evolution of Homo erectus, with its bigger brain, long, lanky legs, and ability to leave Africa to colonize parts of Eurasia, around 1.8 Ma as great evidence for these shifts. Ok, all well and good, but, to use an old phrase: where's the beef? In other words, where is the evidence for sustained meat-eating just before and as H. erectus was evolving? Other than a single site, FLK 22, there really wasn't much...until now.

This is what makes the Kanjera evidence so important. I'm not sure it completely quashes my reservations (after all, we still only have two sites with good evidence for regular meat-eating between 2.6 Ma, when stone tools were first invented and used to butcher carcasses, and 1.5 Ma), but it is a good start.

References:

Domínguez-Rodrigo, M, Barba, R, Egeland, CP (2007). Deconstructing Olduvai: A taphonomic study of the Bed I sites. Springer, New York.

Domínguez-Rodrigo, M, Mabulla, AZ, Bunn, HT, Barba, R, Diez-Martín, F, Egeland, CP, Egeland, AG, Yravedra, J, Sánchez, P (2009). Unraveling hominin behavior at another anthropogenic site from Olduvai Gorge (Tanzania): New archaeological and taphonomic research at BK, Upper Bed II. Journal of Human Evolution 57, 260-283.

Egeland, CP, Domínguez-Rodrigo, M (2008). Taphonomic perspectives on hominid site use and foraging strategies during Bed II times at Olduvai Gorge, Tanzania. Journal of Human Evolution 55, 1031-1052.

Ferraro, JV, Plummer, TW, Pobiner, BL, Oliver, JS, Bishop, LC, Braun, DR, Ditchfield, PW, Seaman III, JW, Binetti, KM, Seaman Jr, JW, Hertel, F, Potts, R (2013). Earliest archaeological evidence of persistent hominin carnivory. PLoS ONE 8, e62174.

How pathological is the Nariokotome Boy?

Regula Schiess and Martin Haeusler (2013) recently published a paper in the American Journal of Physical Anthropology that critically examines the issue of pathology for KNM-WT 15000 (the "Nariokotome Boy"). This analysis comes on the heels of other recent re-examinations of this important 1.5 million year old skeleton. For example, it was thought by many that the Nariokotome Boy, who died between the ages of 8-11 (based on tooth eruption and microanatomy) and 12-15 (based on bone development), would have attained an adult stature of perhaps 6'1" (185 cm). This suggested to many that a linear, heat-adapted physique was characteristic of Homo erectus. Graves and colleagues (2010) questioned this reconstruction, arguing that the individual would have largely finished growing, and that an adult stature of about 5'4" (163 cm) was closer to the mark (John Hawks, on his weblog, provides additional commentary here).

Several authors have suggested that Nariokotome suffered from growth pathologies that, if corroborated, would preclude the use of this particular skeleton as a reference for Homo erectus skeletal biology (things like maturation rate or stature). Schiess and Haeusler (2013) set out to test whether the skeleton possesses evidence for:

• Unusually small, platyspondylic (that is, flattened) vertebrae.
• Spina bifida, which refers to a series of conditions characterized by the failure of one or more neural arches to close and thus fully encircle the spinal cord.
• Condylus tertius, which is an additional condyle found on the anterior margin of the foramen magnum, right between the two occipital condyles.
• Particularly small spinal canals (spinal stenosis).

If present, this complex of symptoms could be linked to a rare growth anomaly that goes by the fun name of spondyloepiphyseal dysplasia tarda. To carry out the analysis, they compare KNM-WT 15000 to a reference sample of 63 modern humans of various ages (I suspect that since these are archaeological specimens, their ages were estimated, rather than known. The researchers get around this by lumping their sample into broad categories, juveniles, adolescents, and adults, rather than specific ages). So, they found that:

• KNM-WT 15000's vertebral heights are more-or-less what you'd expect for his age group (i.e., they are not pathologically flattened relative to modern adolescent humans).
• The skeleton's vertebrae (as measured by superior surface area) do in fact appear to be small.
• The non-sacral neural arches that are preserved show no evidence for non-fusion. The 3rd through 5th sacral elements are unfused (see below).

Sacral elements of KNM-WT 15000. From Schiess and Haeusler (2013: Figure 4).

• A condylus tertius is definitely present.
• KNM-WT-15000's spinal canals are very small compared to the modern reference sample.

Schiess and Haeusler (2013: 372) therefore conclude:

[T]he apparent flatness of the vertebrae is typical for the immature age of the Nariokotome boy. There is no indication for spina bifida sensu stricto. An extension of the sacral hiatus from S5 up to S3 is normal in humans of the same age as the Nariokotome boy. The condylus tertius is a developmental anomaly that is relatively common in the normal population unaffected by spondyloepiphyseal dysplasia tarda. It has no clinical relevance. Both the relative smallness of the vertebral bodies and a narrow spinal canal are not classic features of spondyloepiphyseal dyspasia tarda. The criteria for spinal stenosis are met in the cervical, but not in the thoracolumbar region. Moreover, pedicle and canal shape are not typical of congenital spinal stenosis. Rather, a comparison with other fossils suggests that a narrow spinal canal and small vertebrae might be specific to early hominins.

It would have been nice to have included individuals with these skeletal abnormalities in the reference sample; surely there is variation in how these conditions manifest (I'm not sure how difficult it is to get a hold of such specimens). Overall, however, they make a convincing case. All of this is not to say that the Nariokotome Boy was perfectly healthy. Schiess and Haeusler do note that the last two lumbar vertebrae are asymmetrical, likely the result of a severely herniated disc, which in turn probably caused a significant limp. Nevertheless, these findings show that the incongruities of the skeletal and dental age-at-death estimates are not the result of a developmental disorder and, further, that this wonderfully complete skeleton can still serve as a useful reference for Homo erectus skeletal biology.

References:

Graves, RR, Lupo, AC, McCarthy, RC, Wescott, DJ, Cunningham, DL (2010). Just how strapping was KNM-WT 15000? Journal of Human Evolution 59, 542-554.

Schiess, R, Haeusler, M (2013). No skeletal dysplasia in the Nariokotome Boy KNM-WT 15000 (Homo erectus) - A reassessment of congential pathologies of the vertebral column. American Journal of Physical Anthropology 150, 367-374.

Reconstructing subsistence during the Plio-Pleistocene

I just finished going through an interesting analysis by Michael Pante of the large mammal faunal assemblage from the site of JK2, which is in Bed III of Olduvai Gorge and dates to approximately 1.0 million years ago. The study in important for a number of reasons, not the least of which is the fact that we have so few decently preserved faunal assemblages that date to this time period. What is more, JK2 preserves butchery marks, another rarity among sites of this age, which show definitely that hominins (in this case, probably Homo erectus) were cutting flesh from carcasses and breaking open bones for marrow.

Ultimately, Pante uses the frequency and anatomical placement of the butchery marks and the carnivore tooth marks to argue that Homo erectus was gaining early access to carcasses (that is, before other carnivores had a chance to consume the carcass). This is potentially important, because there has been a lot of discussion about the importance of meat in the diets of early hominins. If meat was a staple of the diet, it is possible that it partly drove other evolutionary changes such as increased brain size (meat is easy to digest, so if you could free up energy that is usually channeled to the guts to process food, it can be diverted to other important organs, namely the brain). Pante compares the frequencies of butchery and tooth marks in the fossil assemblage to experimental assemblages of bones exposed to various processes:

• Hammerstone-only, in which humans cut the flesh from bones and then broke them open with stones to access the marrow
• Carnivore-only, in which carnivores (mainly hyenas) consumed carcasses
• Hammerstone-to-carnivore, in which humans cut the flesh from bones, broke them open for the marrow and, afterwards, carnivores scavenged the remains
• Whole bone-to-carnivore, in which humans cut the flesh from the bones but left everything else (flesh scraps and marrow) for carnivores to scavenge
• Vulture-to-hominin-to-carnivore, in which vultures ate some of the flesh, humans broke open the bones for marrow, and then carnivores scavenged the leftovers

The important observation here is that each of these experimental scenarios results in different frequencies and locations of butchery and tooth marks. For example, when humans remove the flesh and the marrow, there are very few tooth marks, since carnivores have little reason to gnaw on bones that are devoid of edible tissue.

While I agree that Homo erectus probably gained early access to carcasses, what struck me is the fact that the JK2 assemblage really does not match up very well with any of the experiments. I have run across this myself in analyses of other assemblages, and I can't help but wonder that the experimental scenarios that we've come up with so far, while extremely useful, simply are not comprehensive enough to model the complexities we're seeing at these Plio-Pleistocene sites, a point that Pante concedes in the paper. Regardless, this paper provides additional data on a critical, and currently poorly sampled, time period.

In my mind, two of the most important things that Plio-Pleistocene taphonomists need to work out are (1) reaching consensus on exactly how we identify marks on bones and (2) producing experimental bone assemblages that can test a wider variety of potential behavioral scenarios.

References:

Pante, MC (2013). The larger mammal fossil assemblage from JK2, Bed III, Olduvai Gorge, Tanzania: implications for the feeding behavior of Homo erectus. Journal of Human Evolution 64, 68-82.