Bead-making in the early Upper Paleolithic

Mary Stiner and her colleagues (2013) recently published a really fun analysis of shell beads from the Paleolithic site of Üçağızlı Cave I in Turkey. The levels at the site date to the early Upper Paleolithic, between about 41,000 and 29,000 radiocarbon years ago (calibrated dates place the occupations between about 45,000 and 35,000 years ago).

Examples of perforated shells from Üçağızlı Cave I. From Stiner et al. (2013: Figure 5). 

After demonstrating (convincingly) that the modified shells are, in fact, artificially punctured beads, Stiner et al. (2013) show that the types of shells brought back to the cave, and the types that were chosen as beads, change over time. They provide four hypotheses to explain the change in shell type frequencies: (1) environmental changes determined what sorts of shelled animals were living near the cave, and thus what was available for bead exploitation; (2) the distance from raw materials (mediated by sea level fluctuation) changed over time, in which case people would accumulate more "junk" when raw materials were close and/or convenient and be very selective (i.e., only take the best stuff) when raw materials were distant and/or inconvenient; (3) the diversity of shell types is a function of the number and heterogeneity of shell collectors and users (that is, the more folks, and the more types of folks, exploiting shells will result in greater shell diversity); (4) cultural "conformism," where norms determine the types of shells used during any one time period.

The first hypothesis was not supported, as habitats don't appear to have changed significantly over time. The second hypothesis found some support in the fact that the diversity of ornament shells increased in those levels with evidence of frequent shellfish consumption. The latter would, theoretically, signal that the coast, and thus the raw materials for beads, were relatively close by and people became less picky about what they brought back to the cave for ornamental use. However, Stiner et al. (2013) question whether the distances to, and the weight of, the shells would have played much of a role in mediating what was, or was not, transported to the cave.

In the end, they feel that group size and composition and conformist behavior best explain the patterning in shell usage at the site. This is based on two observations. First, an inverse relationship exists between the diversity of species brought back to the site and the frequency of shells that were actually perforated for use as beads. That is, during time periods where many different types of shell were collected, only a few of those types appear to have been utilized for bead use. In the earliest levels, on the other hand, people transported only a few types of shells back to the cave, but most of them were subsequently perforated. Second, the size and shape of perforated specimens is relatively consistent regardless of time period. Stiner et al. (2013: 396) therefore conclude that:

Several factors contributed to the variation in the UcI ornament assemblages. Sample size effects were addressed early in the analysis and do not explain the patterns that concern us. The taxonomic diversity of the assemblages as a whole stems first from the natural biotic diversity of the marine environment. However, changes in littoral habitats to not seem to account for the differences between assemblages or changes over time. Many other aspects of the variation developed as the material was filtered by human choices. The narrow range of shapes and shell sizes is clearly attributable to human preferences, as are the bias toward whole specimens and the methods of modifying the shells for suspension. The extent of 'noise' or variance in the ornament assemblages is a product of the relationship between raw material collection activities and ornament manufacture and use. 

The fact that the taxonomic diversity of the UcI shell ornaments increased with time while the percentage of holed specimens decreased demonstrates that variation in the assemblages must be considered according to two distinct technological stages: raw material collection and artifact manufacture. The more playful and less constrained stage of artifact life history is raw material collection. Variation in overall raw material diversity among the assemblages related in part to how convenient access was to the shoreline and/or the overall importance of marine foraging from UcI. Variation in raw material also correlates with increasing occupation intensity and dietary breadth; the least variable assemblages formed during the lightest occupations. Changes in the size and complexity of the social groups that occupied UcI therefore may have played a role in shaping variation in the ornament assemblages. Unfortunately, the predictions for neutral social effects (more people on site or present for longer periods) and high-grading effects (raw material access) are difficult to distinguish in this particular case. What we can be certain about is that more 'junk' found its way into the shell raw material brought to the site at some intervals and that very different criteria took over at the manufacture stage. As the times collected were streamed into more formal uses, their characteristics narrowed, and many of the shells initially collected were rejected before further modification. It is for these reasons that we conclude that post-manufacture characteristics of the ornament assemblages most clearly reflect the influence of shared esthetic and visual norms. 

I actually ran into this article while I was mining the latest publications on modern human origins, which we are currently covering in my Paleolithic Archaeology class. One of the traits that many archaeologists associate with "modern" behavior is the use of personal adornment, since it is extremely likely that these items were used by ancient people as symbols to transmit information about themselves (just as jewelry does today). Stiner and her colleagues do not involve themselves in the modern human origins debate (as they note, the beads are not among the oldest in the world, or even in the Levant), but there analysis got me thinking about the importance of group size and complexity in the production of symbols. There is, in fact, quite a bit of recent work in Paleolithic archaeology that invokes the importance of demographic factors (mainly population size and density), rather than cognitive factors, to explain the appearance (and disappearance) of symbolic objects among pre-Upper Paleolithic cultures. Think about it: if you interacted with other groups only very rarely, or you ran into the same groups over and over again, there is very little incentive, even if you were cognitively able, to create a ton information transmitters (i.e., symbols) of any kind, much less those that would preserve in the archaeological record.

References:

Stiner, MC, Kuhn, SL, Güleç, E (2013). Early Upper Paleolithic shell beads at Üçağızlı Cave I (Turkey): Technology and the socioeconomic context of ornament life-histories. Journal of Human Evolution 64, 380-398.