ABSTRACT
Many of us who were involved in the LSE modelling project with Nancy Cartwright share more or less similar views regarding the position models occupy within the theoretical hierarchy and the role they play within that framework. Taking the notion of a model as ‘mediator’ between theory and the world as a starting point, several of us have tried to articulate specific details of our own views about various aspects of modelling in the natural and social sciences. One of the things I want to do in this chapter is flesh out, in a bit more detail, ideas presented in Morrison (1999) and Morrison and Morgan (1999) about the relation between a model that represents a physical system (a representative model) and its role as a mediator.1 There are many different ways that models can function as a mediator. It can mediate between theory and the world in the sense of being an abstract representation of a physical system governed by one or more theories, or it can be a concrete representation of some feature of an abstract theory. The pendulum is an example that covers both of these cases. In the case of theory application we have the ideal pendulum which represents harmonic motion, and we also have the physical pendulum which is modelled by making various corrections to the ideal case.
