One thing that often happens when I’m out giving a tour of one of the great houses at Chaco, usually around the middle of the tour, is that I’ll ask if anyone has any questions and one of the visitors will say, with a knowing tone, “Slavery?”
There are variations, and sometimes it’s more of an honest question than a desire for confirmation, but the topic comes up quite frequently. My usual answer is that there’s just no way to tell. The Chacoans didn’t have a written language, so all the evidence we have for understanding the nature of their society comes from its material remains, which can be surprisingly ambiguous. (There is of course also ethnographic analogy, with the modern pueblos providing the most obvious place to look, but this is problematic for a number of reasons and unlikely to be much help in this particular case.)
Certainly the construction of the great houses and other monumental aspects of the Chacoan system required a vast amount of labor. I’ve noted my skepticism about estimates of construction labor; while this certainly seems like a good way to quantify just how much effort went into construction, in my view it suffers from a couple of serious flaws. For one thing, there needs to be a source for the estimates, which is generally measurements of effort expended on modern (or ethnographically documented) projects using what are assumed to be comparable techniques and technology. While this is plausible enough in theory, in practice it’s not really possible to tell how comparable the modern work is to the ancient work. This makes any numbers dubious to start with.
Secondly, and perhaps more fundamentally, once the calculations have been done the result is a series of estimates generally expressed as person-hours or some other similar unit. From one perspective, having numbers is better than not having numbers, but from another perspective it’s not at all clear how useful these numbers are. After all, the very nature of a measurement in person-hours is that there are two variables that can be manipulated. The same amount of work can be done by a few people over a longer period of time or by many people in a shorter time. In the case of Chacoan sites, tree-ring dates can provide some limitations on the amount of time necessary, but usually not enough to make a definite determination on how to interpret the labor estimates.
The same numbers, then, can be used to argue that the sites were built by small work crews over a considerable amount of time, which implies that the necessary labor could have been freely offered or even done by skilled workers in exchange for something akin to wages, or by a great many people in a very short amount of time, a context in which some sort of slavery or corvée labor looks a lot more likely. Ultimately this type of evidence leads to a dead end, and it’s more useful for supporting theories derived from other lines of evidence than for deriving theories itself (although even then it’s not all that useful, in my opinion).
It’s interesting to compare this to the situation in Aztec-era central Mexico, where we do have records, in the form of pictorial codices from the precontact era and Spanish accounts from after the Conquest, showing how labor was organized and who built the impressive monumental buildings of the urban centers. In that case, it is clear that labor service was treated as a type of tribute or taxation owed by the common people to their lords or kings. While Chaco is clearly quite a different type of place from these later cities to the south, I think recent discoveries such as chocolate suggesting closer contact between Chaco and Mesoamerica than has previously been acknowledged implies that a close examination of Mesoamerican labor systems would be a fruitful way to address the question of labor requirements and the possible societal implications of them. The question of whether the labor to build the Chaco system was freely offered or was the result of slavery, tax, tribute, corvée, or something else entirely is still an open one, and one way to address it would be to look further afield for parallels than has been done so far.