The paper I discussed earlier on the connection between plow-based agriculture and highly inegalitarian gender roles was based on a theory proposed by Ester Boserup. Boserup was a Danish economist who had a lot of interesting ideas about the relationship between population growth and agricultural intensification. She’s best known for arguing that intensification of agricultural production is a response to population growth, rather than a cause of it as Malthus proposed. The basic idea is that the default mode of agricultural production is extensive, i.e., requiring a lot of land but relatively little labor. As population grows, however, the amount of land available per person declines, so people need to get more production out of each unit of land. This generally requires more work, and the amount of production per unit of work declines. That is, although the overall amount of food produced increases, the amount of food produced for the amount of work expended to grow it declines. Thus, people will only intensify agricultural production when they are forced to do so by increasing population.
This model has been very influential, but it has also been criticized on a number of fronts. Many of these relate to the underlying assumptions, which Boserup didn’t really make explicit. The model assumes that the amount of land available is fixed, and doesn’t make allowances for other responses to increased population growth such as trade, migration, and conquest. There have also been a number of cases in which the model doesn’t seem to apply, either because increased intensification does not in fact require more work or because putting more work into intensified agriculture does not in fact increase crop yields. (Note that these objections come largely from cultural anthropologists based on ethnographic data.)
A 1999 paper by Glenn Davis Stone and Christian Downum (available here) tries to incorporate the criticisms of Boserup’s model into a recasting of the model that sees it as applicable only under certain circumstances, namely where increased labor is both necessary and sufficient to raise production. When this is the case, one option for coping with increased population pressure is what they term “Boserupian intensification,” which is basically the process Boserup described in which people work harder and get higher total yields but lower yields per unit of work on a fixed amount of land. There are other options, however, including migration and trade, for dealing with population growth in this context, and in other contexts where Boserupian intensification is not an option because of ecological conditions those other options comprise the whole set of possible responses. In conditions where intensification is possible without harder work, due to new technology or innovative techniques, population pressure ends up not being much of a problem. Examples given in the paper include raised-field agriculture in the Andes and rice paddies in East Asia. In other conditions, however, such as arid environments where the weather is very unpredictable, intensification through increased labor just doesn’t work to increase yields reliably, and population pressure becomes a very big problem that must be addressed through other solutions.
Stone and Downum illustrate their proposal through an examination of Wupatki. This is a very arid part of northern Arizona with similar climatic conditions to Chaco Canyon. Like Chaco, it was also (rather mysteriously) a major population center in prehistoric times. Wupatki’s heyday came mainly in the twelfth and early thirteenth centuries AD, a little later than Chaco’s in the eleventh and early twelfth. Many years ago Harold Colton of the Museum of Northern Arizona proposed that the rise of Wupatki was a response to the eruption of Sunset Crater Volcano in 1064, which covered the area with volcanic ash that served as a natural mulch and greatly increased local fertility, leading to a massive influx of population. This theory has been challenged more recently but it seems to still have a large number of adherents, and Stone and Downum seem to be among them. They take the idea of increased fertility from the volcanic ash for granted and look at what happened once people started to move into the Wupatki area. There have been proposals that the increased population pressure from immigration led to intensification of agriculture at Wupatki in a classic example of Boserupian intensification, but Stone and Downum look carefully at the evidence and conclude that there is very little evidence of any intensification. Instead, they argue that the vast majority of the Wupatki area is totally unsuitable for intensification; there are very few places where any kind of irrigation or floodwater farming can be practiced, and for the most part the only way to grow anything is to extensively dry-farm the open spaces. Thus, intensification was not an option, and other options for dealing with population pressure must have been pursued.
The strategy they see as most likely is political consolidation to secure claims to land. Since the necessity of extensive agriculture made holding on to as much land as possible a necessity, coming up with effective means of demonstrating and enforcing land rights was a high priority. They see the most likely way this would have happened is the formation of political units organized along ethnic lines. It is always difficult to recognize ethnicity in the archaeological record, but Wupatki has long been known as a “frontier” area occupied by people of three distinct archaeological “cultures”: Kayenta Anasazi, Sinagua, and Cohonina. These presumably don’t correspond exactly to “ethnicities” in the modern sense, but there are notable differences in material culture among the three, and Stone and Downum identify two main clusters of settlements including large, impressive pueblos. One of these, centered on Wupatki Pueblo, seems to show Sinagua affiliations, while the other, centered on the Citadel, shows more Cohonina affiliations. The place of the Kayenta Anasazi in all this is unclear. They don’t seem to have a cluster of their own, and Kayenta-affiliated sites are spread all over the area.
Stone and Downum see the construction of large, aggregated pueblos as a sign of group power intended to impress others with the legitimacy of the group’s claims to land and other resources, and they suggest that relatively few people might have actually lived in the pueblos themselves despite being affiliated with them politically. For this system to work, the groups’ land claims had to be backed with the credible threat of coercive force, and there is indeed some evidence of violence at Wupatki that may reflect occasional instances when this force needed to be shown. (There is an interesting parallel here to a more recent article on a different part of the Southwest which I will discuss at some point.)
To illustrate the plausibility of their interpretation of Wupatki, Stone and Downum rely on analogy to contemporary ethnic groups in central Nigeria. This is an area where Boserupian intensification is in fact possible and some groups have dealt with population pressure through intensification. Other groups, however, have responded instead by organizing along ethnic lines to defend their land claims through the threat and occasional application of violence. Stone is a sociocultural anthropologist and this part of the paper is based on his fieldwork in the area. This is a good example of what (sociocultural) anthropology can contribute to interdisciplinary scholarship. Ethnography produces an enormously rich, textured body of qualitative data that can be used to test hypotheses and models to explain social phenomena. Those models can come from anthropology itself, of course, but they can also come from other disciplines, such as economics in this case. Downum is a Southwestern archaeologist, and his role in this paper seems to be in providing the data about Wupatki. Much of the data comes from an extensive survey of sites within Wupatki National Monument which has greatly increased the amount of information available about this interesting but poorly understood area.
I think the arguments in this paper are pretty plausible. There are some questions about the application of the theory to Wupatki, but it seems to fit as well as any other explanation I’ve seen for the processes of aggregation and abandonment that marked this period of Southwestern prehistory. The basic idea is that aggregation occurred because of political consolidation, which came about to secure land claims in the face of population pressure and inability to intensify production. Along with this consolidation came increased conflict, and ultimately that conflict and poor climatic conditions led to abandonment. This is similar to the model I have proposed for the role of warfare in spurring aggregation and abandonment, although there I focused more on warfare specifically as a response to resource scarcity rather than political consolidation as the response and warfare as the result of that. This is really a difference of emphasis, however, and the basic idea is very similar. What Stone and Downum’s theory doesn’t explain, however, is the widespread nature of the aggregation and abandonment processes across the northern Southwest, even in places with much better ecological conditions such as Mesa Verde where intensification would presumably have been an option in a way it wasn’t at Wupatki. They acknowledge this, and make no claim to explain anything beyond the specific local situation, but it’s an issue that is worth thinking about in evaluating theories like this.
Since I’ve been talking a lot lately about disciplinary issues in academia, it’s worth noting that this paper seems to be to be pretty much entirely a cultural anthropology paper. It uses archaeological data, and one of the authors is an archaeologist, but the overall analysis lies squarely within the realm of (sociocultural) anthropology. It’s well-done, too, and quite serious and empirical. I suspect its authors are probably among those who consider themselves “scientific anthropologists” and are outraged by the American Anthropological Association’s moves toward removing “science” from the definition of the field. I wouldn’t call this science, though. There’s no hypothesis testing or statistical analysis, and the analysis is basically comparative and qualitative. It doesn’t go so far in that direction as historical papers like Robin Ganev’s, which I discussed in the previous post, but it’s nowhere near as scientific as the economics paper on plowing that I discussed in the post before that. What I take from this is that scholarship doesn’t need to be scientific to be serious. Indeed, in a paper like this one of the main advantages is to take a more “scientific” theory like Boserup’s and evaluate it from a more qualitative perspective to define the unstated assumptions behind it and the conditions under which it applies. I still maintain that anthropology is not a science, but that doesn’t mean it isn’t worthwhile or doesn’t count as serious scholarship.
Stone, G., & Downum, C. (1999). Non-Boserupian Ecology and Agricultural Risk: Ethnic Politics and Land Control in the Arid Southwest American Anthropologist, 101 (1), 113-128 DOI: 10.1525/aa.1918.104.22.168