Late Preclassic Period platforms were surveyed and test-excavated in a seasonal swamp or akalche, at the large ancient lowland Maya site of El Mirador in Peten, Guatemala. Pedological, hydrological, and archaeological evidence suggest that the climatic and hydrological regimes were drier than they are today to permit habitation and warrant the investment in labor. Nevertheless, observations of present conditions, including the clayey soils, indicate that the akalche was almost as inhospitable in antiquity as it is today, that it had the same extremely poor agricultural potentials and that intensive wetland cultivation would have necessitated sizeable investments in dams to dampen the extreme seasonal fluctuations in water levels. Dams are not observed, and evidence is elicited to indicate that they would be readily observable today had they been emplaced here in antiquity. We offer evidence suggesting that these platforms had domestic functions, but we are forced to conclude that intensive wetland cultivation did not sustain them nor this large regional center as a whole. We argue, instead, that these habitations constitute evidence for population pressures on urban upland resources, especially building sites and perhaps agricultural land. Akalche households appear to have sustained themselves, at least in part, by foraging for animal and plant foods, cordage products, building materials, and firewood. The data are inadequate to decide if these structures housed full or part time specialists, or if they were self-sufficient in subsistence. 0 1994

Based on published lexicostatistical dates, two intervals in the prehistory of southern Mesoamerica stand out as fertile periods in terms of the generation of new languages: the Terminal Preclassic/early Early Classic Periods, and the Early Postclassic Period. After comparing archaeological evidence with language distributions within the subregions of southern Mesoamerica during the first of these periods, we conclude that the cultural processes during both periods had the same potential for producing rapid rates of linguistic divergences. Just as rapid proliferation of linguistic divisions was symptomatic of the well-known collapse of Late Classic Maya civilization, so it can be taken as a sign of a collapse of Terminal Preclassic civilization. Both collapses were characterized by severe population reductions, site abandonments, an increasing balkanization in material culture, and disruption of interregional communication networks, conditions that were contributory to the kind of linguistic isolation that allows language divergences. Unlike in the Terminal Classic collapse episode, small refuge zones persisted in the Early Classic Period that served as sources of an evolving classicism; these refuge zones were exceptions, however, not the rule. Although the collapse of each site had its own proximate cause, we suggest that the enormous geographical range covered by these Early Classic Period site failures points to a single ultimate cause affecting the area as a whole, such as the onset of a prolonged and devastating climatic change.