Mayan Chultuns and Experimental Archaeology

Posted on December 5, 2011 by


It’s been a while since I’ve contributed to The Love Machine, with Cait providing all the content for a while now. I’ve been busy with grad school work, but that’s not really a good excuse – I could have found the time if I wanted to. I’m re-affirming my presence on this blog and will be posting frequently, if not exactly regularly. But I’ll probably be talking less about video games and Harry Potter (though I’m sure I’ll still touch on that stuff from time to time) and more about the topics that are actually taking up my time these days, art history and archaeology.

While doing research for a paper on experimental archaeology, I came across a super-cool study that won’t really fit into my paper well, but is, like, the coolest thing ever. I thought I’d share:

Experimental archaeology is a method in which experiments are carried out to attempt to answer questions about the past that cannot be answered by traditional archaeological analysis. There’s plenty to hate on about experimental archaeology. If the archaeologist isn’t careful, he or she could end up sabotaging their experiments in any number of ways. And unlike the hard sciences or even social sciences, there aren’t really agreed-upon standards for how archaeological experiments should be set up. The more rigorous ones borrow pretty strictly from traditional experimental design, but it’s easy to delve into Mythbusters territory and tweak the experiment to get the kind of results you want (I do love Mythbusters though). A lot of popular history and popular archaeology books and TV programs also make use of experiment, not always to academic standards, and for this reason also experiment is treated warily by many in the archaeological (and especially the art historical) world.

Beyond poor experiment design lies incorrect interpretation of experiments. It’s tempting, upon completing a successful experiment, to say that you’ve “proven that this is how it was.” But experiments can’t prove that something DID happen, only that it COULD have happened. In what is probably the most famous experiment in archaeological history, Thor Heyerdahl sailed a wooden raft, the Kon Tiki, from the coast of Peru to Polynesia. His intent was to prove that Polynesia was originally settled by South Americans and not, as was traditionally held, Asians. While the Kon Tiki experiment made for a great story, the scholarly community did not accept Heyerdahl’s theory. While Heyerdahl had shown that South Americans COULD have sailed to Polynesia, they felt that there wasn’t sufficient evidence that they DID in fact do this.

“Big” experiments like Heyerdahl’s are filled with possibilities for great error, and should perhaps be avoided. But even in smaller experiments, the temptation to interpret a conclusive finding rather than a positivistic one remains strong (as it does in the traditional sciences, one might add). The important thing to remember is that an experiment doesn’t prove any hypothesis, it only eliminates other hypotheses. That brings me to the experiments performed by Dennis E. Puleston in Guatemala in the 1960s.

Most of the time, experiments attempt to test two hypotheses: either something happened, or it didn’t happen. Either this, or not this. But really interesting experiments, like. Puleston’s, start with a whole bunch of hypotheses, any or none of which may be correct, then narrows them down.

In southern Mexico and Guatemala, man-made underground pits are often found during excavations of pre-Colombian sites. These were first found in the mid-19th century, when they were described as “chultuns.” The name was taken from superficially similar subterranean pits found in the northern Yucatan that were lined with plaster and had been used as cisterns. In 1930, one archaeologist re-plastered one of these Yucatan chultuns, filled it up, and used it to supply water to his dig crew. But the “chultuns” in southern Mexico and Guatemala were smaller, differently-shaped, and had no plaster on their insides. “Cistern” could not comfortably describe their function. In the 100 years between their discovery and the experiment, about a dozen different hypotheses about the southern “chultuns” function were proposed. Along with the original label as cistern, others pronounced them food storage pits, sweat bathing facilities, refuse dumps, tombs, latrines, mine entrances, or drains. One archaeologist even suggested that the chultuns were not man-made at all, but were natural cavities made by uprooted palm trees that the ancient Maya occasionally modified.

All of these uses were plausible on paper, but none had been tested. To test the cistern hypothesis, 400 gallons of water were poured into a chultun. Within 8 hours, all of the water had drained into the surrounding rock. So much for the cistern hypothesis. To test the sweat bathing hypothesis, crew members crouched in the cramped chultuns for various stretches of time. After complaining of back aches and general unpleasantness, this falsified the sweat bathing hypothesis, along with the fact that no stones for making steam were found.

Other hypotheses were ruled out by standard archaeological evidence. The tomb hypothesis was rejected as a main use of the chultuns, although about one in six did show evidence of burial. These however, clearly had served some other function first, and were only later used for burial purposes. The latrine and refuse dump hypotheses were then rejected on the grounds that humans were unlikely to be interred in former latrines or dumps, and because the lateral shape of the chultuns made them unsuitable for waste disposal use. The mine entrance hypothesis was rejected because it could not account for the existence of floor sills and covering stones on the chultuns, and because other mining sources were discovered apart from the chultuns.

This left the food storage hypothesis. Food storage made sense as a use from the chultuns, as they would keep grain pests away from their crops. In addition, the design of the chultuns was appropriate for food storage; people don’t need to be in the chultun for long periods of time, but they would require rain protection and a covering stone. But there were problems: For one, chultuns were not found in later Mayan sites. If the chultuns stored food, why did their use cease? Additionally, the chultuns, being quite damp, had already proved a problematic environment for storage. One archaeologist had found this the hard way by sealing various metal tools inside a chultun at the end of a dig season. When he returned for the next dig a few months later, he found the tools rusted, with metal parts falling off their handles. Puleston began a round of experiments to determine if and how chultuns could store food.

After crafting their own chultun with period tools to use as a controlled environment, Puleston’s team put an entire spectrum of vegetable crops used by the Maya into a chultun (as well as potatoes, which were not eaten by the Maya but used as a control). Then, these same vegetables were also placed in two different outdoor storage facilities, where they would be more susceptible to vermin. They tracked the vegetable for eleven weeks. Between fungus in the chultun and rodents above ground, neither group lasted very long. For instance, in the chultun, most of the maize decomposed quickly “after sprouting vigorously” with mold, and was later covered in mites. Puleston reports that after the eleven week period, “only 10% of the dried maize on the cob was judged edible.” Based on his gruesome descriptions of the rotting vegetables, I would argue that 10% is a pretty impressive rate.

But while nearly all of the vegetables in the chultun decomposed, there were virtually no instances of insect or rodent vermin in the chultun. Above ground, these animals carried the vegetables away very quickly. But the chultun wasn’t doing its job either, so Puleston concluded that none of the vegetables would have been stored in the chultun for any length of time.

Then, later that year, Puleston noticed a high correlation between the distribution of the ramon tree and Mayan houses. The seeds of the ramon tree were found to be plentiful and nutritional, and could have been a staple crop. The next year, Puleston ran the vegetable tests again, using all of the previous vegetables, new additions like avocados and zapote, and, most importantly, ramon seeds. The avocados, zapote, and the other vegetable decayed quickly, but after nine weeks (and later 13 months!) the ramon seeds remain preserved and edible.

These experiments by Puleston are great because they start with a large number of hypotheses and narrow them down one by one. It doesn’t attempt to prove any particular point, but ends up making a strong argument anyway. The ramon seeds account for most of the problems with the food storage hypothesis. They can keep because they have an extremely low water content, allowing them to withstand the chultuns’ high humidity. They also explain the disappearance of chultuns after 900AD, by which time maize had replaced ramon seeds as the staple crop of the Mayans. And it accounts for the burials in some chultuns, who would have been buried after the Maya stopped cultivating the ramon tree.

The experiments that Puleston and his team conducted were not difficult to execute in the course of their standard archeological excavations. As more and more hypotheses were disproved, Puleston built a more and more specific case for the function of the chultuns, despite not having any strictly archaeological indication of their function. It’s important to remember that Puleston did not prove that the chultuns were used to store the seeds of the ramon tree. His experiments are instead important because he ruled out so many functions that had been, for various good reasons, proposed for the chultuns. When Puleston’s findings were published in the anthology Experimental Archaeology, its editors commented that “In retrospect, it is surprising that no one attempted such simple tests before and published the results; they seem so obvious after the fact.”

As for the author of the experiments, Dennis E. Puleston? He died a few years after these experiments at the age of 38. While watching a thunderstorm on top of one of the pyramids of Chichen Itza, he was struck by lightening.

If you’ve finished all that (congratulations) and want to read the actual report by Puleston (it’s pretty cool), look for it here:

Puleston, Dennis E. “An Experimental Approach to the Functions of Classic Maya Chultuns” in Experimental Archaeology, ed. Daniel Ingersoll, John E. Yellen and William Macdonald (New York: Columbia University Press: 1977), 79-102.

Which was originally published in American Antiquity Vol. 36 no.3, 1971.

And can be found on JSTOR here:

Posted in: Archaeology