Another video from the Climate Modelling conference has been posted on the Institute’s YouTube channel.
Missing the Forest for the Fish
Eric Winsberg
(The University of South Florida, Philosophy Department)
Abstract: In a recent series of papers, Roman Frigg, Leonard Smith, and others
have developed a general epistemological argument designed to cast doubt on
the capacity of a broad range of models to generate “decision relevant predic-
tions. ” According to these authors, the basic reason for doubt is that many
non-linear models are subject to what is deemed the “hawkmoth effect.” In
the hawkmoth effect, small structural differences between the predictive model
and the ‘true’ system can lead to “fundamentally misleading” predictions. They
suggest that the “burden of proof ” falls on the modelers, who must show that
the ‘true’ systems have those desirable structural features which make their
models stable under these perturbations. As a way of bringing out some con-
cerns about the generality of this epistemological challenge, in this paper we
will consider the extent to which the hawkmoth effect should undermine con-
fidence in the centerpiece decision relevant predictions generated in the most
recent IPCC report.