Bill Lycan has a new paper on the problem of the Gettier problem. It has become fashionable to denigrate the project of attempting to solve the Gettier problem, and Bill’s paper evaluates a number of attempted explanations of what is wrong with the project, showing why they are not good explanations. Along the way, he also proposes his own account, which is what interests me here.
Lycan’s proposal is the “no-false-assumptions” proposal (in the face of some examples, this proposal becomes the “no-essential-false-assumptions” proposal, but the difference isn’t relevant here so I’ll ignore it), a variant of the “no-false-lemmas” approach suggested by Clark. Lycan thinks that the NFA proposal was not entertained directly, but implicitly by the adoption of the defeasibility approach. In effect, Lycan claims, the NFA approach was bypassed by a theory that attempted to analyze what a false assumption is in terms of the notion of defeat. So what the counterexamples show is the the NFA approach is very hard to characterize in terms of the notion of defeat.
Then Lycan echos a point made here, when he says “Indeed, that difficulty was predictable, because (a) it was almost irresistible to start the further analysis with a subjunctive of some kind, and (b) any time any analysis of anything contains a subjunctive, irrelevant counterexamples will ensue. (b) is worth a paper of its own.” Brian Weatherson joins in the chorus here, saying “I always thought the best simple description of the “Conditional Fallacy” in Shope’s paper of the same name was the fallacy of thinking that the analysis you’re after contains (subjunctive) conditionals. Shope said something more subtle, but that was the easiest takeaway line. I’m glad to see someone else agrees!”
I’m going to post more on Lycan’s paper, but since the conditional fallacy is a pet peeve of mine, I enjoyed seeing some signs that awareness of the problem is becoming more common. If we’re lucky, it will become common enough that no paper with a subjunctive analysis of anything will be sent out for publication without an explicit defense that the analysis avoids the problems that plague nearly every such analysis.