Scientific models almost always contain idealizations. This raises a number of fundamental philosophical questions: Is idealization inevitable regardless of the amount of data and computational power at our disposal? Does idealization have any nonpragmatic value or justification? Will idealization persist as science progresses? Much of the recent philosophical literature correctly suggest that idealization has epistemic value and is likely to persist as science progress. This has been defended in different ways, including by demonstrating how idealized models can identify core causal phenomena. However, one important defense of idealization has not yet played a central role in philosophical discussions: the existence of tradeoffs. Modelers have multiple theoretical desiderata including precision, generality, and accuracy. Some desiderata cannot be simultaneously maximized because they constrain and trade off against one another. This necessitates the construction of multiple idealized models for a given phenomenon. Intellectual Merit: Tradeoffs and Idealization in Biological and Chemical Modeling will defend the existence of tradeoffs among the desiderata of models and modeling and argue that these tradeoffs are part of the explanation and justification of idealization's ubiquity. The analysis will focus on classic models of predation, competition, and molecular structure, as these exemplify theoretical practice in evolutionary ecology and quantum chemistry. This project can be divided into three components. The first part consists of identifying and characterizing the most important desiderata for theoretical modeling. Examining universal desiderata of models as well as specific properties of ecological and chemical models, this stage of research will generate a set of definitions that are both analytically rigorous and consistent with scientific practice. In the second part, tradeoffs and other constraints among these desiderata will be identified. This will consist of showing, by both mathematical and empirical means, that increasing the magnitude of some desiderata will necessarily affect the magnitude of others. This analysis will yield a list of the most important tradeoffs. Finally, these analyses will be used to argue that the existence of tradeoffs is part of the explanation and justification for idealization. This project is primarily a contribution to the scientific and philosophical literature about modeling and idealization. Broader Impacts: It will also have a very direct impact on the debate between biologist Richard Levins and his critics over the existence of tradeoffs. While Levins has long maintained that simultaneously maximizing all theoretical desiderata was impossible, his lack of formal arguments for this thesis have left his insights vulnerable to critical attack. This project will yield arguments to defend Levins' position. Another wider impact concerns scientific pedagogy. Ideas from this project will be reformulated for use in secondary and higher education with the primary goal of making citizens more informed consumers of science, especially in fields such as climate science where critics prey on public misunderstanding of modeling and idealization. A secondary goal is to improve advanced textbook discussions of the justification for modeling and theorizing. Finally, the arguments of this project may be useful in adjudicating disputes between different groups of theorists. Some theorists believe that idealization should ultimately be eliminated from their models; others believe that idealized models will always be important in scientific inquiry, despite their inaccuracies. Rational resolution of these disputes might be aided by fully accounting of the rationals for idealization, including the existence of tradeoffs.
