Abstract

ABSTRACT Broome criticizes reasons-responsiveness conceptions of rationality by arguing that while rationality supervenes on non-factive mental states, reasons-responsiveness in the relevant sense does not. I give a limited defence of reasons-responsiveness conceptions of rationality against Broome’s criticisms. I argue that Broome fails to show that reasons-responsiveness conceptions of rationality are barred from regarding non-factive mental duplicates as equally rational in the sorts of ‘New Evil Demon’ scenarios that tend to motivate the intuition that rationality supervenes on non-factive mental states. Still, while reasons-responsiveness conceptions may be able to save our intuitions in such global deception scenarios, I also argue that reasons-responsiveness conceptions have to make surprisingly large concessions in cases of more local deception. Indeed, supervenience failures in local deception cases look rampant. Nonetheless, it is not clear why reasons-responsiveness conceptions should be concerned about this. This is because it is not clear why the idea that rationality supervenes on non-factive mental states should be the kind of non-negotiable starting point for theorizing about rationality that Broome thinks it is. For example, Errol Lord, a recent proponent of a reasons-responsiveness conception of rationality, self-consciously rejects a perfectly general supervenience thesis, regarding it as more important that our conception of rationality satisfies different desiderata: namely, that rationality is normatively significant, and that evaluations of rationality or irrationality imply person-level praise or criticism. Interestingly, Broome regards these latter desiderata as less important, and they are also desiderata that his own conception of rationality has more trouble satisfying. Ultimately, then, disputes between different conceptions of rationality turn in part on disputes concerning which desiderata for a conception of rationality are most important and why.