Colloquium - Alex Worsnip / Higher-Order Evidence as Undercutting Defeat
Colloquium - Alex Worsnip / Higher-Order Evidence as Undercutting Defeat
Friday October 2, we welcome Alex Worsnip from UNC Chapel Hill, who will share his work on higher-order evidence. The talk is abstracted below.
Suppose that you get some evidence that some belief of yours – say, your belief that you’re an above-average teacher – is irrational. Presumably, this evidence should make you more confident in the higher-order proposition that your belief that you're an above-average teacher is irrational. But should it make you (significantly) less confident in the first-order proposition that you're an above-average teacher? It’s very natural to think that it should. Yet - as has gradually emerged in the literature – there’s a puzzle about what could explain this. This is because the fact that you irrationally believe p is not generally very strong evidence (and maybe not evidence at all) that p is false. Moreover, while there’s a well-recognized phenomenon that decreases rational credence in a proposition without itself constituting evidence against that proposition -- namely, "undercutting defeat" -- the consensus among epistemologists is that evidence of irrationality has special properties that mean that it cannot be assimilated to undercutting defeat. This leads them to either conclude that evidence of irrationality doesn't affect rational credence in first-order propositions after all, or to embrace increasingly complicated and unorthodox explanations of its rational significance. The aim of this talk is to revive the possibility of assimilating evidence of irrationality to undercutting defeat after all. The central idea is that once we accept a modest kind of holism about evidential support, the undercutting defeat explanation can be applied more widely than it first seems. Along the way, I'll distinguish evidence of irrationality from other things that often go under the label "higher-order evidence", and explain why the former poses special challenges that the latter don't. And I'll illustrate the relevance of this debate for socio-politically relevant cases involving "suspiciously convenient beliefs".