What attributes are important for the measurement of assemblage reduction intensity? Results from an experimental stone artefact assemblage with relevance to the Hoabinhian of mainland Southeast Asia

Citation: Ben Marwick (2008) What attributes are important for the measurement of assemblage reduction intensity? Results from an experimental stone artefact assemblage with relevance to the Hoabinhian of mainland Southeast Asia. Journal of Archaeological Science 35: 1189-1200 (RSS)
Internet Archive Scholar (search for fulltext): What attributes are important for the measurement of assemblage reduction intensity? Results from an experimental stone artefact assemblage with relevance to the Hoabinhian of mainland Southeast Asia
Download: http://www.sciencedirect.com/science/article/pii/S0305440307001653
Tagged: Anthropology (RSS) reduction intensity (RSS), attributes (RSS), Hoabinhian (RSS), stone (RSS), artefact (RSS), assemblage (RSS), Southeast Asia (RSS), experiment (RSS), dorsal cortex location (RSS)

Summary

Marwick analyzes what attributes one can look for to help make sense of Hoabinhian lithic assemblages. In particular, he focuses on how the attributes relate to the intensity of reduction. He then goes on to consider how these may relate to issues surrounding mobility, resource availability, and other such questions. He analyzes what methods might be useful for acquiring data that relates to the larger, overarching questions. Examples are how Hoabinhian assemblages contribute to our understanding of foraging in a tropical forest, how geographically spread the technology was, or how related to the domestication process and associated changes it is.

In order to address these questions, Marwick takes an experimental approach where he attempts to recreate Hoabinhian-like assemblages to take measurements on throughout the reduction process. This approach was undertaken due to a lack of useful methods for analyzing the Hoabinhian assemblages, which tend to have few retouched flakes and little to no unique artifact forms. He advocates for a new way of analyzing these assemblages, as previous studies have focused on typology rather than interpretation. The goals of the analysis are to discover how flake variables are affected through the process of core reduction and then to determine what are the best variables to use for answering archaeological questions.

The experimental method involved recording 28 variables from flakes that were obtained by hard hammered percussion from 30 river cobbles. There was an attempt made to simulate a ‘typical’ Hoabinhian assemblage consisting of sumatraliths, short axes, choppers, and other unifacially flaked cobbles. The flakes were given a ranked intensity of reduction on a scale of one to ten, with one being the first reduction phase. In total, there were 625 flakes and 159 non-flakes.

The results indicate that there is not a significant change in flake mass associated with the level of reduction. Additionally, overhang removal was strongly correlated with an increased level of core reduction. The interior platform angle and level of reduction were also strongly correlated. The overhang removal and interior platform angle seem to be connected. The percentage of dorsal cortex is also strongly linked to the level of cobble reduction. It works best as a sign of early reduction. The amount of dorsal flake scars is significantly related to the amount of reduction, but can be less relied upon as a sign of reduction level than the other listed variables. A new method, which analyzes the dorsal cortex location is proposed. The dorsal cortex location is directly linked to intensity of reduction. It appears to be independent of flake size. There are four classifications that are identified. These are the primary, crescent, distal and tertiary. With this approach, the inter-observer error has been shown to be low and may approach zero when trained lithic analysts use the method. The distal cortex class can be further analyzed to provide even more resolution on the reduction process, especially in the later stages. When analyzing a core after the reduction process has been completed, the number of flake scars does not relate to the number of flakes removed. Cortex ratio seems to be a good indication of reduction intensity.

Theoretical and Practical Relevance

This experiment is likely to have application to non Hoabinhian assemblages. More experiments should be undertaken to see how these methods, in particular the new method of dorsal cortex location, and variables compare to other types of assemblages. Further studies should be conducted to explore how these experimental findings can be applied to the interpretation of actual Hoabinhian assemblages, especially when considering 'big picture' questions.