ABSTRACT
However, this interpretation of anti-adaptationism is inadequate. The defenders of adaptationism treat Stephen Jay Gould and Richard Lewontin's paper The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme' {1979] as the definitive antiadaptationist statement. Gould and Lewontin discuss alternative, nonadaptive forms of evolutionary explanation, but they also raise issues about the testing of adaptive hypotheses and the relative significance of different evolutionary mechanisms. Elliot Sober interprets the biological literature on adaptationism not as a debate over how many traits are adaptations, but as a dispute about the relative importance of adaptive forces and other evolutionary mechanisms in determining the trajectory and destination of organisms in the space of possible designs (Sober [1987, 1993]). In 'Optimality Models and the Test of Adaptationism' Steve Orzack and Sober [1994] distinguish three claims about adaptation. The first is that it is ubiquitous. Most characters are subject to natural selection. The second is that it is important. To define this notion more precisely they introduce the idea of a censored model. This is a model of evolution from which certain mechanisms have been deliberately omitted. Adaptation is important if a model censored of natural selection would significantly rnispredict the actual form of the organism under study. Finally, it might be contended that organisms are optimal. Orzack and Sober argue that an organism is optimal if a model censored of all evolutionary mechanisms except natural selection would accurately predict the form of that organism. They suggest that adaptationism is best construed as the claim that most phenotypic traits are locally optimal.
