Hacker News Reader

Comment by nkurz | Hacker News Reader

I've often been intrigued by cases where erroneous logic coincidentally leads to a correct conclusion. In philosophy, these are known as "Gettier Problems": https://en.wikipedia.org/wiki/Gettier_problem

They are named after the American philosopher Edmund Gettier, who in 1963 published what is probably the most influential 3 page paper ever written by a philosopher: https://fitelson.org/proseminar/gettier.pdf

Searching once again to find the name for this phenomenon, I was surprised to see that Gettier himself was still alive until earlier this year. The submission is a summary of his career by one of his colleagues at UMass.

View on HN

Gettier cases are not cases where erroneous logic leads to a correct conclusion. The reasoning in a Gettier case can be perfectly fine; this is why the belief of Gettierized subjects remains justified. The issue is, rather, that because of "luck" there is a risk for reasoning/belief formation mechanism to go wrong, at the same time that it turns out to have produced a correct/true belief. The issue derives from there being a gap between a proposition being justified and it being true (Linda Zagzebski has a great paper that shows that whenever that is the case, you will end up with an analogous problem).

Interesting story: Gettier only published his paper on the subject because of pressure to publish. Apparently he wasn’t especially prolific and this led him to publish something, anything, to get people off his back.

I recall thinking that the famous paper of his was a little forced. As I understood it, it seemed to conflate the question of whether JTB is a good definition of knowledge — I think it absolutely is — and whether some given proposition is actually known. In the latter case, you would need to know that proposition p is true, that your justification is true, and that you believe it. This becomes a bit more murky IMO, but it doesn’t undermine JTB I would say.

Reply on news.ycombinator.com