An earlier initial Upper Paleolithic at Kostenki

A paper by Anikovich and colleagues in Science describes revisions to the Upper Paleolithic chronology of Kostenki, Russia.

Here’s what I think about this paper:

  1. The issue of redating for the Kostenki chronology is covered better in a Quaternary International paper by Sinitsyn and Hoffecker last year. This new paper in Science basically takes that earlier paper and cuts out most of the details – both for and against their preferred chronology. The new elements are all hidden away in the supplementary information, but they only include a new stratigraphic description and a series of OSL dates.

  2. If you know a little about Kostenki, the magnitude of the redating – around 10,000 years earlier – may be surprising. If you don’t know anything about Kostenki, well, consider that Kostenki is not a site, but a village with around 20 separate Paleolithic localities around it. Many of these localities already have long series of radiocarbon dates, so that the chronology of the entire array of localities has been based on hundreds of radiocarbon dates. This paper isn’t discarding all those dates, but it is proposing that the older ones should be recalibrated much earlier, and that still doesn’t make them old enough to match the OSL and other kinds of dates.

If you know a lot about Kostenki, then there’s no surprise here; the earlier dates follow directly from accepting that the ash layer is actually 40,000 instead of much younger. That has been known for a few years. It’s really not very novel.

It is interesting that much of the way toward the older date on the radiocarbon dates comes from calibrating them. I’ve written about the problems of radiocarbon calibrations before; this paper doesn’t mention them. The calibration here is enough to make a 37,000 14C date into a 42,000 year calibrated date.

  1. The paper says this about the initial Upper Paleolithic at Kostenki:
The artifact assemblages below the CI tephra do not represent an Upper Paleolithic industry that is "transitional" from the local Middle Paleolithic, but rather an abrupt departure from the latter. Prismatic blade technology is predominant and Middle Paleolithic artifact types are rare. Most of the stone used for artifact production was imported 100 to 150 km from its sources (9), and the perforated shells (Columbellidae) in the lowermost level at Kostenki 14 (Fig. 4G) apparently are derived from a source no closer than the Black Sea (i.e., transported >500 km) (8). Other raw materials include bone, antler, and ivory. Most noteworthy is the carved ivory piece that may represent an example of figurative art. Novel technologies include the rotary drill and - by implication - devices for harvesting small game (26). Although taxonomic assignment of the associated human teeth is tentative, the contents of this Upper Paleolithic industry suggest that it was probably manufactured by modern humans.
Deposits below the CI tephra at Kostenki also yielded several artifact assemblages that primarily contain typical Middle Paleolithic tool forms (e.g., side-scrapers, bifaces) manufactured on flakes (7). They lack imported raw materials, bone-antler-ivory artifacts, and art. The faunal remains are confined to large mammals (30). These assemblages, which are assigned to the local Strelets culture, are analogous to the "transitional" Upper Paleolithic industries of western and central Europe (especially the Szeletian), at least some of which apparently were produced by local Neandertals (1, 26). The Strelets artifacts are not associated with any human skeletal remains and their makers are unknown. They may represent an activity variant of the other Kostenki industry (i.e., probably produced by modern humans) related to the butchering of large mammals. Younger Strelets assemblages are found above the CI tephra (7, 12) (Anikovich et al. 2007:225).

Of course, these paragraphs directly contradict each other. If the assemblages below the ash layer are an “abrupt departure” from the Middle Paleolithic, then they shouldn’t “primarily contain typical Middle Paleolithic tool forms.”

The resolution of this contradiction is that there are two distinct industries represented, one at Kostenki 14/IVb, and one at Kostenki 12/III. And as discussed by Vishnyatsky and Nehoroshev (2004), the Kostenki 14/IVb assemblage may represent something different than the “advanced” industry from Kostenki 17/III and possibly Kostenki 12/II.

There may be a reason for the current paper to gloss over these distinctions (even omitting names for the industries, Streletskian and Spitsynian) – there is currently no reason to think one of them is older than the other. Anikovich and colleagues suggest they may be different use facies of a single industry. The weakness of this explanation may be the long duration of the Streletskian (the one with Middle Paleolithic elements); it would seem to render the more “advanced” boneworking and ornament-making industries as apparently more ephemeral and special-use, because they are not found as widely or as long. Whether elements of them may be mixed together in different proportions at different sites is a good question that somebody should examine – an increased emphasis of ornaments and bone in the later Streletskian may signal this.

At present, there is not really any convincing case for an intrusive origin of the initial EUP at Kostenki. For more information, I have put together a long post on the archaeology of the initial Upper Paleolithic</a> at Kostenki, reflecting on a paper by Vishnyatsky and Nehoroshev (2004). An earlier date makes an intrusive origin more problematic, because it greatly narrows the possible locations for such an originating population. From an archaeological perspective, it is simpler to argue that the Russian Plain itself is the origination point for the advanced boneworking industries of the initial EUP. Absent the need for a migration of prismatic core-knapping and bone carving people into the area, it is not clear whether archaeology is really telling us about the movement of modern humans into this region. I would guess that the important factor is the occupation itself; Neandertals may not have been able to use the Russian Plain effectively, as reflected by a rarity of Middle Paleolithic sites.

  1. The carved ivory “head” is not very persuasive. There is no suggestion of features. Looks like it might be some kind of toggle instead.

  2. If the initial UP at Kostenki can be redated 10,000 years earlier, and if dozens of radiocarbon dates earlier than 32,000 years can unilaterally have 5000 or more years added to them, this inspires little confidence in the existing radiocarbon chronology of Europe. Of course, we’ve been seeing changes in radiocarbon chronology for many years now. Still, the scale of this change is very impressive.

If I had a very important specimen that was supported by a single radiocarbon date, I would be very nervous. Something like Vindija 80…


Anikovich MV and 14 others. 2007. Early Upper Paleolithic in Eastern Europe and implications for the dispersal of modern humans. Science 315:223-226. doi:10.1126/science.1133376

Sinitsyn AA, Hoffecker JF. 2006. Radiocarbon dating and chronology of the Early Upper Paleolithic at Kostenki. Quaternary International 152-153:175-185. doi:10.1016/j.quaint.2005.12.007

Vishnyatsky LB, Nehoroshev PE. 2004. The beginning of the Upper Paleolithic on the Russian Plain. Pp. 80-96 in Brantingham PJ, Kuhn SL, Kerry KW, eds, The Early Upper Paleolithic beyond Western Europe. University of California Press, Berkeley CA.