Assume ``point contemporaneity'' of occupation

Large scale artifact scatters can often be thought of as palimpsests. That is, they may actually be composed of a series of smaller overlapping artifact scatters formed at different points in time. A good example of this is the so-called `standard model' of Amazonian settlement (per [Meggers 1991], see [DeBoer et al. 1996,Heckenberger et al. 1996] for refutations). This model holds that the inhabitants of Amazonian villages practice swidden agriculture and are therefore obliged to relocate their villages periodically. Over time, localities preferred for settlement may accumulate a substantial cultural overburden and comprise artifact scatters of considerable extent, even though they in fact represent nothing more than a superposition of many small settlements occupied at different points in time.

Figure 4.3: Scenarios for the formation of an archaeological sector

a) the sector, b) the palimpsest scenario, c) the ``point contemporaneity'' scenario (``local maximum'' at Time 2).

Among truly sedentary societies, such as the Titicaca Basin lacustrine civilizations, large artifact scatters often do in fact reflect large villages, towns or cities. They are not palimpsests in the same sense as the Amazonian sites as understood by Meggers. However, it is undoubtedly the case that the size of these settlements fluctuated over time as the social units that inhabited them grew, fissioned, migrated or were decimated by disease, famine or warfare. Therefore, the corrected size of a sector represents not the occupied area of the settlement for the entire phase in question, but rather its occupied area at the moment of its maximal extent. Thus, ``point contemporaneity.'' Under this assumption sectors are not palimpsests, and the occupied area of the sector directly reflects the number of resident households at the time of the ``local maximum'' (see Figure 4.3 for an illustration of the difference).

Assuming point contemporaneity does seem to be reasonable at least for the phases (pre-Pacajes) characterized by nucleated habitation. In these periods, the villages and towns undoubtedly grew and contracted; however there is no indication that they moved from place to place and back again. In the Pacajes phases, the matter is less certain, since dispersed residences were probably occupied for shorter periods of time and were considerably more mobile. The small size - equivalent to only one or a few households - and low density of most Pacajes sites would seem to indicate very limited within-phase sector abandonment and resettlement, especially since the adobe structures endure no more than a few years if uncared for (see [Bermann 1994] for a discussion of the decay process of adobe residential structures). The point contemporaneity assumption is therefore preferable to the palimpsest assumption for all phases of the region's settlement history.