This thesis presents a new framework for the philosophy of physics focused on methodological differences found in the practice of modern theoretical physics. The starting point for this investigation is the longstanding debate over scientific realism. Some philosophers have argued that it is the aim of science to produce an accurate description of the world including explanations for observable phenomena. These scientific realists hold that our best confirmed theories are approximately true and that the entities they propose actually populate the world, whether or not they have been observed. Others have argued that science achieves only frameworks for the prediction and manipulation of observable phenomena. These anti-realists argue that truth is a misleading concept when applied to empirical knowledge. Instead, focus should be on the empirical adequacy of scientific theories. This thesis argues that the fundamental distinction at issue, a division between true scientific theories and ones which are empirically adequate, is best explored in terms of methodological differences.

In analogy with the realism debate, there are at least two methodological strategies. Rather than focusing on scientific theories as wholes, this thesis takes as units of analysis physical principles which are systematic empirical generalizations. The first possible strategy, the conservative, takes the assumption that the empirical adequacy of a theory in one domain serves as good evidence for such adequacy in other domains. This then motivates the application of the principle to new domains. The second strategy, the innovative, assumes that empirical adequacy in one domain does not justify the expectation of adequacy in other domains. New principles are offered as explanations in the new domain.

The final part of the thesis is the application of this framework to two examples. On the first, Lorentz's use of the aether is reconstructed in terms of the conservative strategy with respect to the principles of Galilean relativity. A comparison between the conservative strategy as an application of the conservative strategy and TeVeS as one of the innovative constitutes the second example.