Saturday, May 30, 2015

What Does it Mean to Know?

Ernest Sosa, a philosopher at Rutgers, offers us a quick lesson in basic epistemology for the layman at The Opinionator. Epistemology is the (surprisingly, perhaps) fascinating study of the nature of knowledge and belief. Most of us think we know something if we are certain that it's true, but things are much more complicated than that. The classical definition of knowledge is that whatever it is we claim to know must be true, we must believe it's true and we must have good reasons for believing it's true.

Unfortunately, there are difficulties with this definition lurking in the philosophical weeds and Sosa points out the biggest of them, something called the Gettier Problem. He writes:
What is it to truly know something? In our daily lives, we might not give this much thought — most of us rely on what we consider to be fair judgment and common sense in establishing knowledge. But the task of clearly defining true knowledge is trickier than it may first seem, and it is a problem philosophers have been wrestling with since Socrates.

In the complacent 1950s, it was received wisdom that we know a given proposition to be true if, and only if, it is true, we believe it to be true, and we are justified in so believing. This consensus was exploded in a brief 1963 note by Edmund Gettier in the journal Analysis.

Here is an example of the sort used by Gettier to refute that theory. Suppose you have every reason to believe that you own a Bentley, since you have had it in your possession for many years, and you parked it that morning at its usual spot. However, it has just been destroyed by a bomb, so that you own no Bentley, despite your well justified belief that you do. As you sit in a cafe having your morning latte, you muse that someone in that cafe owns a Bentley (since after all you do). And it turns out you are right, but only because the other person in the cafe, the barista, owns a Bentley, which you have no reason to suspect. So you here have a well justified true belief that is not knowledge.
Sosa then goes on to discuss a modern formulation of the definition of knowledge, which, for what it's worth, I don't think is much of an improvement, but read the article and judge for yourself.