Abstract
The view that in radical interpretation, the interpreter should aim at optimizing the rationality of agents is defended. A distinction and a parallel is drawn between linguistic interpretation and psychological interpretation. Both can be taken to be governed, in part, and in somewhat different ways, by a principle of rationality. Such approaches have been criticised on the ground that they make it impossible for a speaker or an agent to have wildly irrational or false beliefs. It is argued that the so-called principle of simulation isn't a serious altemative, and that the counter-intuitiveness of the optimization-based approaches has been exaggerated.