A Critical Assessment of Levins's The Strategy of Model Building in Population Biology (1966)

Richard Levins's 1966 article "The strategy of model building in population biology" is an extremely influential analysis of the nature of scientific model building. His claims that model building involves a necessary trade-off among generality, realism and precision, and that truths about nature can be revealed by finding "robust theorems" are important and deserve careful scrutiny. We clarify the concepts of generality, realism and precision and argue that there is no necessary conflict among them. We also examine the idea of robustness and conclude that it lacks proper definition and that its bearing on the question of whether a proposition is true is highly problematic. Accordingly, we believe that neither of Levins's claims should be accepted.