Autonomous villages, competitive ceremonialism

Population continued to grow in the Late Chiripa phase, though at a slower rate than in the Early Formative. However several very interesting changes took place with regards to settlement organization. First, the rate at which new sites were founded decreased dramatically. The population was growing, but the old pattern of village growth and fissioning, resulting in the foundation of new villages, seems to have become untenable in the Middle Formative. This suggests that the set of social forces and processes brought together by Carneiro under the rubric of ``circumscription'' had intensified further in this period ([Carneiro 1970,Carneiro 1988]). This is only to say , as I did in the previous section, that the costs of village fissioning increased to a point where alternative mechanisms of conflict resolution and social integration were sought.

I am most decidedly not suggesting that the subsistence potential of the Taraco Peninsula had been exhausted, or that the communities of the peninsula had begun to feud over a scarce or strained resource base. Rather, I believe the principal factors to have been scale-related conflicts and stresses which challenged the viability of communities exceeding a population index value of a few hundred. Carneiro's concept of circumscription accords explanatory primacy to conflict between groups, and the rapid and competitive increase in the size of political groups as existing communities merge either through conquest or alliance. I am suggesting, on the contrary, that the important dynamic in this period of early villages was conflict within villages and communities which intensified as the size of the villages increased. While the landscape was largely empty, it was a simple matter for villages to fission, and to relocate to a nearby area which was similar in terms of the physical environment to that of the parent community. Several such examples were described in my discussion of the EF.

However, relocation costs increased as the landscape of the Taraco Peninsula filled up with villages. In order to found a new village it would not be necessary either to appropriate lands already claimed by another village - which would presumably involve fighting - or to move off of the peninsula itself into the adjacent Katari and Tiwanaku Valleys, which were still lightly-settled. However, the agricultural potential of these inland areas was not so great as on the peninsula, and communities in these areas would have no direct access to the lacustrine zone, which, I would argue, was always crucial to the subsistence strategies of the Taraco Peninsula communities. Thus village fissioning became an increasingly difficult and unattractive solution to scale-related stresses. This being the case, there was no doubt experimentation with various techniques and strategies for resolving conflicts and increasing social solidarity. As I said, I believe the appearance of the first sunken courts and decorated ceramics in the EF2 to represent exactly this type of experimentation.

Early in the MF, however, some stable solution was apparently produced. From this point on village fissioning ceased on the Taraco Peninsula, and villages grew steadily to sizes many times what had been the fissioning threshold in the Early Formative. Of course, we have no clear idea what the exact nature of this solution might have been. Archaeologically, it is evident as the Yaya-Mama Religious Tradition. Its principal features include:

1. sunken courts and associated architecture,
2. decorated serving ceramics,
3. ceramic ``trumpets'',
4. ceramic burning bowls, almost certainly used in ritual burning activity, and
5. a distinctive style of stone sculpture.

Of course, the Yaya-Mama Religious Tradition was actually produced and reproduced by the actions and routinized practices of the members of the communities involved. The material cultural features by which we define the Tradition are only the byproducts of the actions of these people. The material features themselves, however, allow us to hazard some guesses as to the nature of some of these practices. In particular, the presence of a complex of decorated serving ceramics, which are moreover associated almost exclusively with public architectural contexts, suggests the serving of food and drink in a public context; in a word, feasting; in two, commensal politics. The sunken courts were probably employed periodically for community-wide events involving the serving of food and drink. We may imagine that a system of competitive generosity developed which allowed competition to take place in a way which actually increased the solidarity of the community as a whole.

This process may have occurred more or less simultaneously in many parts of the Titicaca Basin. Only one other settlement dataset has adequate temporal resolution in these early phases to permit useful comparison. Carlos Lémuz, in his signal study of a portion of the Santiago de Huatta Peninsula, documents a 30% decrease in the number of sites in the Late Chiripa phase, together with a 60% increase in occupied site area ([Lémuz Aguirre 2001]: 196). That is, average site size increased, and population became increasingly concentrated in fewer sites, exactly the same process we have observed on the Taraco Peninsula. I have yet to inspect Lémuz's data carefully for evidence of site fissioning in the Early Formative, but I would predict its presence.

It is particularly interesting that, after an early experimental phase in the EF2, the complex of features which made up the Yaya-Mama Religious Tradition stabilized to a remarkable degree and was adopted by communities throughout the Titicaca Basin. The Yaya-Mama Religious Tradition, as a relatively stable configuration of practices, seems to have been very portable, a sort of ``cultural portmanteau.'' It was adopted by many communities, presumably for many different reasons. In this way, it was similar to early horizon styles such as Chavín in the Central Andes and Olmec in Mesoamerica, though with a much more restricted distribution.

Around 450 B.C., after the Yaya-Mama Religious Tradition had apparently become established throughout the Titicaca Basin, the lake level dropped to about 16 meters below its modern level. Since the Lago Wiñaymarka is very shallow - almost all of it is less than 20 meters deep - it became almost completely dry at this time. The lake we see today did not exist. In its place was an immense grassy plain, crossed by small, meandering rivers and dotted with marshes. The Desaguadero River was dry; the lake had no outlet.

There can be no doubt that this event had a significant impact on the Taraco Peninsula communities. It is almost certain that the drying of the lake - and the resulting disappearance of lacustrine resources such as fish and birds - forced a reorientation of the economies of these communities, which were probably predominantly lake-oriented beforehand. Pastoralism no doubt became more common, since the former lake bed would have rapidly become a vast pampa, ideal for grazing animals. Agriculture also probably became more important. As important as these shifts in the subsistence economy, however, was a change in the manner in which the Taraco Peninsula communities were integrated with the remainder of the Titicaca Basin.

Since the little lake was mostly dry, it was possible to walk directly from the Taraco Peninsula to the Copacabana/Yunguyu area, and from there to the important population centers of the western Basin (Figure 6.8). At this time trade with the western basin expanded dramatically. The evidence for this is the large quantities of a particular lithic raw material which is found almost exclusively in sites of this time period. This material, a kind of olivine basalt, may come from outcroppings near Incatunahuiri, north of modern Chucuito. It is certainly exotic to the Taraco Peninsula. It was used for the manufacture of agricultural implements, principally hoes, and is much superior to the materials available locally. This material appeared for the first time in Late Chiripa sites and virtually disappeared when the lake level rose again, around 250 B.C. This stone was being imported in very substantial quantities. In Chapter 6 I calculated that perhaps as many as a quarter million hoes of this material were imported to the Taraco Peninsula during the MF. And it is virtually absent from the surface of any site from an earlier or later period.

It is clear that this rock is present on the Taraco Peninsula as a result of external exchange, possibly with the western basin. Probably, though, it was only a part of a much more elaborate trading system, other items of which have not been preserved in the Titicaca Basin environment. These could have included products of the lowland yungas and selva, including coca, ají, cotton, feathers, pelts, and hallucinogenic substances. Thus in the late MF the Taraco Peninsula communities may suddenly have found themselves located on a major trade route.

I would like to suggest that local leaders incorporated the acquisition and distribution of desirable exotic goods - the aforementioned hoes being only the most visible example - into the existing system of status competition and competitive generosity. That is to say that a wealth finance strategy (see [D'Altroy and Earle 1985,Earle 1987b]) was added to the existing commensal system. Not coincidentally, it is also in this period - from 450 to 250 B.C. - that we see the first indications of wealthy graves at Chiripa (T-1). And finally, it is precisely in this period of lowered lake levels that the well-known Upper House Level complex was constructed at Chiripa. These are the famous double-walled structures first discovered by Bennett, located on a monumental artificial terrace. This complex was clearly the most elaborate public architecture yet seen on the peninsula; the houses were constructed around 380 B.C., shortly after the drying of the little lake. Similar terraces were found at a number of contemporary sites, though these have yet to be excavated.

Also during the MF, perhaps in the latter half, a two-tiered site size hierarchy emerged on the Taraco Peninsula. At the top were four large villages or small towns, each with population index values between 350 and 450. These were the sites of Kala Uyuni (T-232/T-225), Janko Kala (T-394), Yanapata (T-130) and Chiripa (T-1). What is interesting is the fact that this site size hierarchy was produced by a difference in the growth rates of different sites.

The four sites that by the end of the MF came to make up the first tier of the settlement hierarchy grew at an average rate of 0.13% annually during the MF period. By contrast, the smaller sites grew at an average rate of only 0.09% annually. I have argued that these differential rates of population growth can be explained by small differentials in residential choices over time. These residential decisions in turn were affected by the distribution of both staple goods - through public feasts - and imported materials - through public prestations probably in ritual contexts. Essentially, I am suggesting that the inhabitants of the Taraco Peninsula took into account the relative lavishness of each village's system of competitive generosity when deciding where to locate their residences. Over time, this produced higher growth rates in communities with more lavish ceremonies. The lavishness of ceremonies in turn would be correlated with the success of each village's leaders in external exchange relations and in the negotiation of internal community politics. Assuming all of this to be true, then by this standard the most successful villages in the MF were Kala Uyuni (0.15% annual growth; see Table 6.4), Yanapata (0.14%), and Chiripa (0.13%). Janko Kala in fact grew at the very slow rate of 0.09% annually, and is in the first tier of the MF settlement hierarchy only because it was the second-largest village in the EF2 (the largest, T-271, fissioned near the EF2/MF boundary). The fact that Janko Kala is also the only one of these four large sites which lacks unambiguous evidence of MF public architecture would seem to support my interpretation. Also, Kala Uyuni's MF exceptionalism may in part explain its success in the following Late Formative 1 period.