Thursday, May 17, 2018

Obsidian Sourcing

by Angela Huster


While we have posted here about exporting obsidian samples for chemical sourcing – in order to determine which volcano the obsidian came from – I realized that we never posted anything about the results.

Chemical sourcing techniques measure the amount of different elements in a sample. Since different volcanoes (or even different eruptions from the same volcano) have difference “fingerprints” of rarer elements, archaeological samples can then be matched to their source area this way. We used two different techniques, XRF (by Adrian Burke) and INAA (by the MURR lab), which agreed pretty well for the subset of samples which were analyzed both ways.

Getting from the results of the sourcing analyses to the actual frequencies of obsidian from different sources at the Calixtlahuaca ended up being a multi-step process due to how we selected samples. First, the obsidian that we found included both grey and green pieces. Since green obsidian in Central Mexico is almost always from the Pachuca source in Hidalgo and is easy to visually identify, we only selected a few pieces from each household for analysis, just to confirm that that it came from Pachuca – and it did! This means that the actual percentage of Pachuca obsidian at the site is what we identified visually, not the much lower number in the sourcing results. Second, when we picked the grey obsidian samples, we tried to get pieces representing different production techniques and artifact types for each household, but the resulting samples weren’t necessarily representative of the different artifact types in the parent assemblage. When we got out results back, we realized that most of our grey obsidian blades were from Ucareo in Michoacan, while most our grey obsidian bifaces and flakes were from Otumba, in the Basin of Mexico. As a result, we also had to correct our frequencies to account for the types of lithic technologies in our sourcing samples (using Brad Andrew’s typological lithic analysis data). This is why you should think very carefully about your sampling strategy before you select artifacts; your results are only as representative as your samples were!

Once we did all that, our results showed that most of the obsidian at Calixtlahuaca came from three sources; Ucareo in the Tarascan Empire, and Otumba and Pachuca in the Aztec Empire.  We also had occasional pieces from seven other sources, including four pieces from minor Toluca Valley sources. The proportions of the three major sources change over time, with increasing percentages of material from the Aztec Pachuca and Otumba sources over time (Figure 1). However, for all phases, the percentages of Ucareo obsidian at Calixtlahuaca are much higher than at other sites in Central Mexico (Golitko and Feinman 2015). This might be related to the presence of intrusive sites with Matlatzinca style ceramics near the Ucareo source (Hernandez and Healan 2008).

Figure 1. The percentages of obsidian from difference sources at Calixtlahuaca over time, after various correction factors were applied.


Another way to think about obsidian source frequencies is whether the volume of obsidian at a site is changing over time. One way to do this is to calculate the ratio of obsidian artifacts to ceramic sherds. At Calixtlahuaca, the volume of obsidian reaching the site during the Yata (LPC-B) phase drops, meaning that even though there is a higher percentage of obsidian from Aztec sources, this is because there is less Ucareo obsidian reaching the site, not because there is actually more Aztec obsidian (Figure 2). This is interesting because it suggests that the near-total Aztec monopoly on obsidian at other sites might have been due to cutting off access to other sources, rather than flooding the market with increased supply.

Figure 2. The frequencies of obsidian from difference sources per 1000 sherds over time at Calixtlahuaca





Works Cited:

Golitko, Mark and Gary M. Feinman
                2015       Procurement and Distribution of Pre-Hispanic Mesoamerican Obsidian 900 BC-AD 1520: a Social Network Analysis. Journal of Archaeological Method and Theory 2015(22):206-247.

Hernández, Christine L. and Dan M. Healan
                2008       The Role of Late Pre-Contact Colonial Enclaves in the Development of the Postclassic Ucareo Valley, Michoacan, Mexico. Ancient Mesoamerica 19(2):265-282.