A well-known trilemma faces the interpretation of Kant’s theory of affection, namely whether the objects that affect us are empirical, noumenal, or both. I argue that according to Kant, the things that affect us and cause representations in us are not empirical objects. I articulate what I call the Causal Power Argument, according to which empirical objects cannot affect us because they do not have the right kind of power to cause representations. All the causal powers that empirical objects have (...) are moving powers, and such powers can only have spatial effects. According to Kant, however, the representations that arise in us as a result of the affection of our sensibility are non-spatial. I show that this argument is put forward by Kant in a number of passages, and figures as a decisive reason for rejecting empirical affection and instead endorsing affection by the things in themselves. (shrink)

According to Kant each person has an empirical character, which is ultimately grounded in one’s free choice. The popular Causal Laws interpretation of empirical character holds that it consists of the causal laws governing our psychology. I argue that this reading has difficulties explaining moral change, the ‘gradual reformation’ of our empirical character: Causal laws cannot change and hence cannot be gradually reformed. I propose an alternative Causal Powers interpretation of empirical character, where our empirical character consists of our mind’s (...) causal powers. The resulting picture of empirical character allows for moral change and Kantian weakness of will. (shrink)