Kinds of models

Knowledge of chimpanzees has been explicitly built into models of human evolution for at least 30 years, since the emergence of sometimes startling findings from modern field-studies (Kortlandt & Kooij, 1963; Goodall & Hamburg, 1974; McGrew, 1979, 1981a; Tanner, 1981, 1987; Ghiglieri, 1987; Wrangham, 1987). Other species of African primates have also been cited in reconstructions of hominisation: savanna baboon (Washburn & DeVore, 1961), gelada (Jolly, 1970), bonobo (Zihlman et al., 1978). Such models abound. Foley & Lee (1989) listed nine published between 1963 and 1987. This modelling has ranged from speculative outlines (scenarios, just-so stories, evolutionarios?) to systematic, point-by-point formulations (Wrangham, 1987; Wynn & McGrew, 1989). Further, seemingly countless numbers of articles on primate and especially chimpanzee natural history have ended with an apparently obligatory final paragraph on the implications for human evolution. How can we make sense of and choose between these many options?

One starting point is to distinguish between two main types of model: referential (Tooby & DeVore, 1987) or analogous (Dunbar, 1989) versus conceptual or more specifically strategic (Tooby & DeVore, 1987). The former make use of a known phenomenon such as the living chimpanzee as a referent for an unknown phenomenon such as an extinct proto-hominid. The latter use basic evolutionary and ecological theory as developed from studies of all living organisms to construct a tailored set of principles to elucidate the absent proto-hominid. Each type of model has its advantages and disadvantages (cf. Tooby & DeVore, 1987), but when used thoughtfully both yield testable hypotheses, in the form of predictions or post-dictions to explain the data.