Models as fictions A pretence theory of scientific modelling

Abstract

Scientific models are often not physical objects. They are simplifications of the real world, they are inaccurate, inconsistent and incomplete. This raises questions: what sort of entities are models, how do they represent their target system, and how can we learn from them? The aim of this thesis is to investigate the idea that scientific models share a lot of aspects with fictions. An account of scientific modelling that views models as fictions could well provide answers to the questions above. In particular, Kendall Walton's pretence theory will be explored in order to formulate a theory of scientific modelling in analogy with his theory of fiction. The main idea of Walton's theory is that objects (props) can induce and coordinate our imaginings. If we use props this way, we are engaging in a game of make-believe. What is true in a game of make-believe is a fictional truth. Fictional truths do not depend on whether or not they are imagined, it is the prop together with appropriate principles of generation that prescribe what is true in the game. By carefully constructing a model, we can use it as a prop in a game of make-believe. The model and the principles of generation then determine what is true in the world of the model. I will argue that this is a viable account of scientific modelling, but that there are some major challenges that are in need of more attention.