One important line of evidence in understanding the climatic history of Chaco Canyon, a subject of considerable interest given the harsh aridity of the current climate and the incongruous grandeur of the archaeological remains, has been the study of packrat middens. These are collections made by packrats of materials found near their nesting locations, which they collect and then seal with urine. The key thing about these middens, in addition to the preservation ensured by the protective coating of urine, is that the packrats have fairly small territories within which they collect material, so the contents of the middens, which can easily last for thousands of years, reflect the vegetation of the immediately surrounding area at the time the midden was created. Since they generally consist of organic material that can be radiocarbon dated, the middens can potentially offer enormous insight into climate change over time.
There is some difficulty with this, however, because the proper interpretation of the contents of the middens is not always clear. Interpretations are often based on the visible plant remains (known as “macrofossils”), but these may be biased by behavioral factors owing to the diet and habits of the packrats themselves. Juniper, for example, tends to be overrepresented because it is one of the main components of the packrat diet. One way to avoid this problem is to look not just at the macrofossils but also at the pollen grains contained in the middens. Since these are too small for the packrats to be consciously choosing them, they would have to be deposited by wind or other natural processes, which may make them more reliable clues to the relative abundance of various plant species as opposed to the mere presence or absence of species.
One researcher who has long been associated with this point of view is Steve Hall, who was for many years a professor of geography at the University of Texas. He is now retired and has a consulting firm in Santa Fe specializing in the geology of Southwestern archaeological sites. He has been involved with packrat midden studies at Chaco since the 1980s, and has recently published an important article reporting on a recent reanalysis of some of the middens that resulted in some remarkable conclusions.
What Hall found in looking at the midden samples was corn pollen. This is not surprising in and of itself, since the Chacoans were of course an agricultural people with a corn-based subsistence strategy. The focus of the paper, however, is on radiocarbon dating of the pollen grains, which resulted in some remarkable findings. Conventional radiocarbon dating requires much larger samples than would be possible for pollen grains, but the relatively new accelerator mass spectrometry (AMS) technique requires only very small samples, and also tends to give more precise dates than traditional radiocarbon dating. Hall dated pollen samples from two packrat middens in crevices near Casa Chiquita, referred to as CC-2 and CC-3. These are both right along what is now the “Petroglyph Trail” section of the Peñasco Blanco trail, and fragments of the middens can be seen from the trail, although they are now substantially broken up from sampling for study. CC-3 is right beneath one of the best-known petroglyph panels on the trail. CC-2 is not as directly associated with any particular petroglyph panel, but it is in close proximity to a few.
The results basically showed that the corn pollen was old. Some of the samples were very old indeed, on a par with the oldest known maize in the Southwest. The oldest sample, from CC-3, had a 95% probability of being from between 2567 and 2332 BC. Another sample, from CC-2, had a 95% confidence interval of 1457 to 1254 BC, while the rest of the samples from both middens were later but still pretty old, dating to the 800s through 400s BC. In addition to the AMS dating, Hall measured the abundance of maize pollen in the overall pollen samples from the middens and looked at the size of the pollen grains themselves. He found that maize pollen was quite abundant, and that the grains were consistently bigger than comparison samples of modern and later Chacoan maize pollen. Since maize pollen is pretty heavy in general, and this particular maize pollen was even heavier than usual, he concluded that it couldn’t have traveled far, and that it must have been blown into the crevices from a cornfield directly in front of them, where the packrats later incorporated it into their middens along with various other materials. One thing he points out in the article is that twigs from the same middens dated to much later than the pollen, which casts doubt on the common practice of dating middens only by single macrofossils. It seems that material incorporated into the middens can vary considerably in age, and direct AMS dating of pollen is a better approach to determining its age than dating of associated macrofossils. There’s a bit more to the article, but those are the highlights.
This actually shouldn’t be all that surprising to anyone who has been following recent developments in the archaeology of the Archaic period in the Southwest, but I have not been, and I found it pretty surprising. After reading this paper I looked into earlier research a bit and found that there has been quite a bit of direct dating of maize and associated materials resulting in comparable dates, including some from rockshelters near Chaco. I knew that there was considerable evidence for early maize further south, in places like the Tucson area, but it seems that evidence for Archaic corn agriculture is pretty well established further north as well. The implications of this for Southwestern archaeology in general are therefore fairly limited, and it mostly just adds another data point to the accumulating evidence on Archaic agriculture.
The implications for Chaco specifically, however, are huge. This shows pretty convincingly that agriculture in the canyon goes back to long before the main flourishing of the Chaco System in the eleventh and early twelfth centuries AD, and it suggests that Chaco’s importance may go back to thousands of years before that at least on some level. One interesting thing Hall points out is the proximity of these middens, and by implication the nearby cornfield, to Atlatl Cave, a well-known Archaic site on the mesa top about a kilometer away. The radiocarbon dates for Atlatl Cave match up pretty closely with the later dates on the corn pollen, and Hall suggests that the people who lived there were the same ones growing the corn in the canyon below. It’s a reasonable suggestion, and it puts a lot of interpretations of the early prehistory of Chaco in a new light. I’ll be reading up on the Archaic period and trying to understand what this discovery means for our understanding of Chaco during that time. I should have some more posts on the topic soon. For now, though, I’ll just note that this may indicate more continuity of occupation or at least cultural knowledge of Chaco than is generally assumed.
Hall, Stephen A. (2010). Early maize pollen from Chaco Canyon, New Mexico, USA Palynology, 34 (1), 125-137 : 10.1080/01916121003675746
Thank you for this post! Fascinating stuff. Interesting how again, as so often in American archaeology, deeper study reveals greater time depth of the culture being studied.
Glad you like it. And yeah, it really is interesting how far back some things seem to go once technology reaches the point where it’s possible to see such things.
By the way, since the DOI for this article doesn’t seem to work for some reason, I figure I should include a link to it. The full text is available free online in html format at that link, and from it you can also download a pdf.