Possibly false knowledge

Journal of Philosophy 112 (5):225-246 (2015)
Download Edit this record How to cite View on PhilPapers
Many epistemologists call themselves ‘fallibilists’. But many philosophers of language hold that the meaning of epistemic usages of ‘possible’ ensures a close knowledge- possibility link : a subject’s utterance of ‘it’s possible that not-p’ is true only if the subject does not know that p. This seems to suggest that whatever the core insight behind fallibilism is, it can’t be that a subject could have knowledge which is, for them, possibly false. I argue that, on the contrary, subjects can have such possibly false knowledge. My ultimate aim, then, is to vindicate a very robust form of fallibilism. Uniquely, however, the account I offer does this while also allowing that concessive knowledge attributions – sentences of the form “I know that p, but it’s possible that not-p” – are not only infelicitous but actually false whenever uttered. The account predicts this result without conceding KPL. I argue that my account has the resources to explain some related cases for which the KPL account yields the wrong predictions. Taken as a whole, the linguistic data not only do not support the proposal that subjects cannot have possibly false knowledge, but indeed positively favor the proposal that they can.
Reprint years
PhilPapers/Archive ID
Upload history
First archival date: 2015-11-21
Latest version: 2 (2015-11-21)
View other versions
Added to PP index

Total views
774 ( #8,202 of 69,189 )

Recent downloads (6 months)
30 ( #27,723 of 69,189 )

How can I increase my downloads?

Downloads since first upload
This graph includes both downloads from PhilArchive and clicks on external links on PhilPapers.