Tuesday, November 27, 2012

Skepticism, defining knowledge and Gettier cases

In this post, I am going to look at two things. I am going to summarize my position on the debate about skepticism that was raised in the previous posts and then turn to attempts to define knowledge, particularly as they pertain to the "Gettier cases" (a phrase I'll explain in a bit).

Basically, I think that Descartes' argument is either incoherent or could be readily refuted common-sensically. It is not clear if Descartes or subsequent philosophers have intended 'know' and 'knowledge' to be taken in any special senses or if they are just supposed to be commonsense terms. If merely commonsense terms, perhaps some support would have to be given that the entailments that they, the philosophers, claim exist between their arguments' premises actually exist relative to 'know,' 'knowing,' 'knowledge,' etc. This would need to come from ethnolinguistics or corpus linguistics, perhaps, and would require the documentation of usage where such entailments have been shown to exist.

For example, in the English language as I speak it, in my understanding of how I use the word 'know,' none of the sentences I say about knowing something entail that I'm not dreaming, a brain-in-a-vat, etc. When I say "I know my friend pretty well" or when I'm working on a project and I say "I know what I'm doing," imagine someone replying, "Yeah, but if you know that, you must know you're not dreaming right now, huh?" Think about it: Why must I know that to know that I know what I'm doing at a given moment? So I would actually go beyond Moore's argument and not even allow this entailment.

But for the sake of argument, assume that Descartes is right and that I must know I'm not dreaming to know anything else, like that I'm wearing slippers right now. Just like G.E. Moore, I could claim that since I know I'm working slippers right now, I must know I'm not dreaming or in some other kind of weird scenario.

I will turn to the possibility that maybe philosophers intend some technical notion when they talk about 'knowing' or 'knowledge,' but before that I want to look again, very briefly, at the arguments that P.F. Strawson and Peter Unger gave regarding skepticism. I think Strawson's argument is basically right, that even if we could not answer the skeptical challenge that we have to assume knowledge as part of the basic framework of understanding the world or attempting to understand the world. But he is wrong that we can't answer the skeptic either way. We can, if we can understand what he's saying. And if he's talking about knowledge in the ordinary sense, at least as it's used in English, then yeah we can give an argument like the above: the entailment relations don't hold or even if they do ordinary sensory proof suffices for us to assign the word 'knowledge' to what we're talking about, and anything beyond that is demanding the impossible.

Unger on the other hand is worried about dogmatism, and he thinks that a person falls under one of two categories: one is either a dogmatist or a skeptic. He thinks a dogmatist is someone who claims to be certain of at least one thing and a skeptic is someone who claims to be certain of nothing. Unger combines knowledge and certainty as if they were two things. They're not. Typically, when we say that we know something we're not also saying that we are certain of it. He thinks we are. Claiming that we know something seems to me to be more like applying honorific terms: knowledge is just something we seek to attain, and when we seem to have enough descriptive or explanatory adequacy with regard that thing we say that we know it. The ideal is certainty but our best methods will never get us there. What I've just said here is basically a paraphrase of Chomsky in the previous post I made. But I think this is basically right regarding how we ordinarily use knowledge.

Now, if Descartes or subsequent philosophers mean some more technical sense of 'know,' 'knowledge,' etc., then it must be given definition and explained. Barring this, there's no reason for me to accept what anyone has said or to take the skeptical doubt seriously. If it's a technical notion, and the technical notion has not been given at least some rough definition, it is simply incoherent.

Luckily or unluckily for us, the next posts on this blog we'll be me looking at philosophers' attempts to define knowledge. We will entertain the notion, then, that philosophers are using some more technical sense of the word, and if they are we will look at their attempts to give this word fuller definition.

Plato was the first to try to give definition to what knowledge is in his dialogue Meno. This formulation of knowledge has persisted among philosophers even to this day, or at least some modified form of it has. The best definition the character Socrates and his interlocutor could come to in the Meno was something like this: knowledge is "true belief with a rational explanation." In some sense, this provisional definition seems to match up with our intuitions about what knowledge is. Think about it this way. When we say we know something, we believe it. We also think that what we say we both know and believe is true. And we think we could explain why it's true. So it doesn't look to be too bad.

The modified form, as this definition has been received to philosophers nowadays, is that knowledge is "justified true belief." (We could argue if this is the same thing Plato meant, but at least in the subsequent writings, I'll take it as equivalent, at least for the sake of argument.) This position among philosophers stood the test of time, so the history of ideas goes, until a man named Edmund Gettier, desperately looking for tenure track and not to be fired for not having published, actually did publish. He published a paper that was about two or three pages long. And this paper demolished this definition with two clear examples. These two examples have since been called "Gettier cases."

What the "Gettier cases" point out is that you can have justified true belief but not have knowledge. To get the demolition going, he makes two assumptions, and he hopes you'll accept them too. The first is that you could have good reasons or justification for believing something that's false. The second is that if you have good reasons for believing one thing, and you know that if that one thing is true then another thing is true, then you have good reasons for believing that other thing.

I don't know if I've expressed Gettier's two assumptions clearly enough but if not I hope the examples will flesh them out. I'm going to paraphrase the examples a bit to make them as clear as possible. We'll first look at the job interview scenario.

Suppose you and your buddy Bob are up for a job interview. You're pretty sure your buddy Bob is going to get the job, no matter how the interview goes, because you know the interviewer, and the interviewer said he thinks Bob is really the better applicant, and so the interview for you is really just a formality. Suppose also you and Bob were talking in the waiting room too and he showed you the two dollar bill he got in change. And then you saw him put the two dollar bill in his wallet. When Bob goes in there for the interview, and you're outside waiting, you're thinking: "Bob's going to get the job, and he's got a lucky two dollar bill in his wallet." You're also thinking: "The man with the lucky two dollar bill in his wallet is going to get the job." Maybe you even say this second sentence out loud. Now you have good reasons for knowing the first thing you think. Again, the interviewer told you he'd give the job to Bob. Also you saw Bob put the two-dollar bill in his wallet so you know he's going in there with a two-dollar bill. But if that first way of thinking is true, you're pretty sure that the man with the lucky two-dollar bill is going to get the job. Because of your first line of thinking, you think this, and so you've got good reason for thinking it, right?

But then there's a surprise. Bob leaves, you do your interview, and after you're finished, the interviewer tells you his first impression was wrong, Bob's the wrong man for the job, and you're perfect for the job, so he's going to give it to you. You get home, tickled pink, and later that night you examine the contents of your wallet. You realize that you two had a two-dollar bill stuck in there from change you got (maybe there's a lot of two-dollar bills out there this season). So here's something: You were pretty sure the man with the two-dollar bill was going to get the job, you were right that the man with the two-dollar bill did get the job, but you didn't really know he would because you didn't know that man was you. Does that make sense? Did I explain this correctly?

The next case goes something like this. Again, I'll change the example a bit. Your buddy Bob (who didn't get the job) has always owned a Ford truck and today he drove you to the job interview in the Ford truck. (Let's assume you took a taxi back and that you didn't want to tell him about your good news and his bad news yet.) Because he's always owned a Ford truck, and because he picked you up in one, you think this truck that he's driving is his. In fact, later you get into a spat with your best friend about this. Your best friend tells you Sam actually owns a Hyundai. You tell your friend, "Well, either he owns a Ford truck or I'm the product of incest." Your best friend proceeds to explain to you that the truck was Sam's brother's, and Sam was driving it today because Sam's new Hyundai already broke down and was in the shop for repairs. Then your friend tells you further that actually you are the product of incest. You investigate and find out that you are. So what you said was true, but not because you know you were the product of incest. In fact, you thought it was true because you were pretty sure you knew Sam owned a Ford truck. Again, a case of believing something to be true, something being true, and you having good reasons for assuming that it's true, but not really knowing it's true.

Is this clear so far?

And anyway, how do we answer this challenge?

No comments: