Using two decades worth of experimental philosophy (aka x-phi), Edouard Machery argues in Philosophy within its Proper Bounds (OUP, 2017) that philosophers’ use of the “method of cases” is unreliable because it has a strong tendency to elicit different intuitive responses from non-philosophers. And because, as Machery argues, appealing to such cases is usually the only way for philosophers to acquire the kind of knowledge they seek, an extensive philosophical skepticism follows. I argue that Machery’s “Unreliability” argument fails because, once (...) its premises are percisified, they are either self-defeating or without justification. This is a significant result because Machery’s arguments are the most widely cited and discussed x-phi arguments for philosophical skepticism and many hold that Machery provides the most empirically informed, convincing, and thus best case for this kind of skepticism. So, if my arguments are sound, then the best x-phi argument for philosophical skepticism fails. I further argue that this result provides strong reason to believe the more general conclusion that “negative” x-phi is likely doomed: x-phi likely can never support a substantive philosophical skepticism. Ultimately, I argue for the broad conclusion that all empirically minded arguments for philosophical skepticism are likely to fail for the same reasons that Machery’s does, i.e. they are (likely) self-defeating. (shrink)

Many philosophical thought experiments and arguments involve unusual cases. We present empirical reasons to doubt the reliability of intuitive judgments and conclusions about such cases. Inferences and intuitions prompted by verbal case descriptions are influenced by routine comprehension processes which invoke stereotypes. We build on psycholinguistic findings to determine conditions under which the stereotype associated with the most salient sense of a word predictably supports inappropriate inferences from descriptions of unusual (stereotype-divergent) cases. We conduct an experiment that combines plausibility ratings (...) with pupillometry to document this “salience bias.” We find that under certain conditions, competent speakers automatically make stereotypical inferences they know to be inappropriate. (shrink)