Ideal patterns and non-factive understanding

This chapter begins with the observation of a Levins style trade-off in models of complex occurrences, between predictive accuracy of the model, on the one hand, and intelligibility (the capacity of the model to provide scientific understanding), on the other hand. The existence of this trade-off lends support to non-factivist accounts of scientific understanding. However, non-factivism faces an important objection, concerning the hypothetical availability of de-idealised models that afford factive understanding. I reply to these objections by arguing that a weakness in the non-factivist account presented by Potochnik in Idealization and the Aims of Science comes from an ontological commitment to “real causal patterns.” I show that the account can be strengthened by replacing this with the notion of an “ideal pattern,” where the targets of model building are not patterns that are simply “out there” in nature, but they are, to some extent, dependent on the methods of data collection and processing chosen by the researcher. This proposal is then related to earlier accounts of scientific phenomena.