Knowing and Believing

When I say I know X, I mean -- and I think most people would agree -- that I have a fair degree of confidence that X is the case and that this conviction is justified.* (It's always fair to ask why someone is confident.)

But what does it mean to say, "I believe (in) X"? That seems to be a mushier claim. I'd like to try to sort this out.

Let's start by acknowledging that many words have multiple senses. There is no platonic realm where we can look for the true definition of words. Language is a living social institution, and words "move" over time. (I recommend linguist John McWhorter's enjoyable book Words on the Move: Why English Won't - and Can't - Sit Still [Like, Literally].) From the time of Samuel Johnson, dictionaries were descriptive, not prescriptive. They showed how words were commonly used, but not how they ought to be used. (Of course, if you want to communicate with other people, you'll want to infer prescription from description.) When dictionaries are revised, they not only add new words; they also take account of how old words are being used differently since the last edition.

That any particular word has more than one sense does not normally create misunderstanding and chaos. Context speaks volumes, but it is no guarantee of clarity. So writers and speakers should exercise care.

As to the word believe, I can't escape the idea that before one can believe X, one must be convinced that X is the case. (Whether one has strong or weak reasons for being convinced is another matter.) Wittgenstein said that we wouldn't understand someone who said, "It's raining outside, but I don't believe it." We might wonder if that person was fluent in English. Of course, if he said, "I don't buh-LIEVE it," we'd understand him perfectly. We often use sarcasm to mean what the words don't literally say. Most of the time that's not a problem.

This raises the question: can you believe X without being persuaded (to some degree) that X is the case? I don't see how. What would that mean? Maybe you'd like to believe X. Maybe you plan to act as though (that is, pretend that) X is the case. But I don't think you actually believe X if you remain unconvinced. For one thing, genuine belief would be likely be accompanied by certain expectations. What would it mean to will an expectation into existence? If I'm not convinced you're going to visit me today, what would it mean to say I nevertheless expect you to visit? (I'm influenced here by Gilbert Ryle's The Concept of Mind.) 

Note this too: when you become convinced that X is the case, you don't take a second step called believing. Believing is baked into being convinced. When I understand the solution to an algebra problem, X=5, I don't then go on to believe it. Understanding is believing. This is not only true of mathematics and logic. Seeing that it's raining is believing that it's raining. It's only one thing. (That doesn't mean I can't be wrong; maybe someone is on my roof with a garden hose.)

That brings us to agnosticism. It's often been pointed out that agnosticism is not some middle ground between theism and atheism. This is true. Agnosticism refers to (the lack of) knowledge, while theism/atheism refer to belief. It's been said that one can be either an agnostic atheist or an agnostic theist. But to understand this, we need further clarification. 

We can make a another distinction between agnostics. Person A could say, "I don't know if God exists," while Person B could say, "God exists, but I know nothing about God's nature." Some heavyweight theologians in the Abrahamic tradition have said that God's existence is undeniable, but that we can't say anything more because God is ineffable -- indescribable. This raises a question: if something were really ineffable, how would you know it existed? At this point we'll be told that God is pure being, that his essence is existence, that he is necessary, etc. None of this helps one bit.

In my view, B is an agnostic theist, but I don't think A is because he's unconvinced that God exists. He would be like the person who says, "I don't know if it is raining, but I believe it is." If all that person means is that he is going to act as though it is raining, say, by taking an umbrella when he leaves the house, then I think it would be more accurate for him to say, "I don't know if it is raining, but I am going to assume that it is." Nothing wrong with saying that, and nothing wrong with the original formulation, as long as we understand that he cannot actually believe it is raining if he is not convinced that it's raining.

*The commonly accepted idea that knowledge is "justified true belief" was shown to be inadequate by Edmund Gettier in 1963. It's known as the Gettier Problem. Here's how I think about it: imagine two people, A and B. At 2 p.m. Person A looks at his (accurate) watch, sees that it reads 2 o'clock, and confidently says to himself, "It's 2 o'clock." We can reasonably say that Person A knows that it is 2 o'clock.

At the very same time his neighbor, Person B, looks at his watch and also sees that it reads 2 o'clock. He confidently says to himself, "It's 2 o'clock." But does Person B really know that it's 2 o'clock? Not if his watch has been stuck at 2 o'clock since yesterday and would have read 2 o'clock even if he had checked the time an hour earlier or an hour later. He indeed has the true belief that it's 2 o'clock, but however justified he thinks he is, he really isn't justified in believing he knows the time.

What counts is not apparently justified true belief but actually justified true belief. Antony Flew refers to this as "the crucial distinction between believing truly and knowing."