VARIABLES VERSUS MODELS OF EARLY PLEISTOCENE HOMINID LAND-USE

Alternative reconstructions, or models, have been advanced to explain the archeological behavior patterns of early toolmakers in East Africa. Current models include central place foraging (home bases), multiple place foraging at stone caches, routed foraging, and riparian woodland scavenging. These models imply that certain socioecological variables affected hominid attraction to specific places. Such factors include social cohesion, predator avoidance, costs of stone transport, habitat patch choice, and tethering of toolmakers to fixed resources. When viewed as alternative hypotheses, however, these models become static and miss possible simultaneous effects of variables on early hominid behaviors. A paleolandscape study of excavated artefacts and habitat indicators in Member 1 (ca. 0.99 Ma) of the Olorgesailie Formation, Kenya, illustrates the interplay of these variables in comparison with other Plio-Pleistocene contexts: (1) resource tethering was important for some raw materials and possibly handaxes, but not for other components of the artefact assemblage; (2) stone transport varied greatly among different basin contexts; (3) overlap with predators was minimal at Olorgesailie but large in Bed I Olduvai; (4) no specific correlation between hominid traces and microhabitat is evident at Olorgesailie; (5) no clear spatial indications of human social aggregation (e.g., shelters, hearths, activity areas) have been found. Diverse sedimentary basins can be expected to differ in the strength of these factors. Growing awareness of variability in Plio-Pleistocene contexts changes the pursuit of archeology away from the single best reconstruction toward an understanding of environment-behavior covariation. Geographic and temporal variations in hominid land use suggest that natural selection may have entailed responses to habitat and resource variability, not merely consistent selection pressure or adaptation to a model environment.