Yes, we can too know things

I like knowing things. One of my favorite things about this universe is that it happens to be the kind of universe that allows me to know true things about it. So I have very little patience for one of the stock philosophy lines.

People often ask rhetorically whether we can ever really know anything. And then everyone reacts as if that person has just said something Deeply Wise, because claiming that nobody can ever know anything is a commonly recognized Philosophy Thing. (Sarcastic capital letters aren’t one, but should be.) And they have half a point, because, as they’ll say if you challenge them, whatever you believe could always be wrong. So you can’t say you know anything. Since I am a person who rather enjoys knowing things, this is my biggest pet peeve that I can think of at the moment. Seriously, it’s even more annoying than small talk and worse than puns, portmanteaus, and punctuation errors combined. But it’s also a bad question. Here’s a better question: Who says I need to be absolutely certain of anything?

I don’t know where this idea comes from, that you can only say you know something if you are absolutely and completely sure of it. That if there’s the slightest chance you might be wrong then you don’t really know. That’s silly. But whenever it comes up, people keep not bothering to differentiate between knowledge and absolute certainty.

This is the type of debate that, if it ever came up on Less Wrong, would be dissolved immediately by people pointing out that the arguers are just using the word “know” differently and if they redefine their terms then they’ll probably end up not disagreeing about anything other than vocabulary. That kind of “tree falling in the forest” debate is a slightly unpleasant waste of time, but it’s not may-its-idiocy-be-stricken-from-the-annals-of-human-thought-forever bad. Where it gets really bad is when people start from this recognized Deeply Wise Philosophy Thing and proceed to say that because nobody can “know” (as in be absolutely and completely certain of, with probability one) anything, they must therefore not be able to “know” (as in, have a reliable degree of certainty about) anything. Or worse, they might not say that, just leave the premise unstated, and let anyone who accuses them of equivocation look like they’re just arguing about definitions.

One of the side effects of believing something to be true with probability 1 is that you can never change your opinion. You can’t see some evidence against it and bring your estimate down to 99% likely. If you’re absolutely certain of something, then you stay absolutely certain no matter what. (This follows trivially from Bayes’ Theorem, but I promised myself I wouldn’t go into the math. If anyone disputes this, the proof is easy.)

So next time you want to claim to be absolutely certain of something, or worse, if you want to say that you need to be absolutely certain before you can “know” anything, ask yourself whether any possible evidence would make you any less sure. If the answer is yes then you don’t have absolute certainty. If the answer should be yes then you shouldn’t require it. The answer should basically always be yes.

(I am willing to concede that there probably exist some exceptions. So in some cases that normative statement doesn’t apply. But those, if they exist, are in the tiny minority and in all or almost all cases it should be possible to be convinced away from your opinion by evidence. For that matter, you don’t even have to admit that it is/should be possible to convince you you’re wrong; you just have to admit that it is/should be possible to make you slightly less certain.)

For a variety of reasons, mostly summed up by Rene Descartes and Thomas Anderson, I think it’s impossible to be absolutely certain of anything. If you disagree, then at least you agree with me on the statement that we can in fact know things. You’ll never say that nobody can ever really know anything for sure, and I count that as a win for me.

If you do agree that infinite certainty is infinitely beyond us, then please don’t accuse people of being unsure as if that’s a bad thing. Never, under any reasonable circumstances, should you seriously say the sentence “we can never truly know anything.” To the extent that that’s true, it’s useless, and to the extent that it’s meaningful, it’s false. (Try it! Substitute your favorite plausible definition for “know” and see if you can get a true sentence that says something that isn’t trivial!)

But sometimes just claiming it’s impossible to know anything isn’t enough. Sometimes people have to claim that there isn’t anything to know. That there is no objective truth or reality at all, just our various subjective experiences. This claim isn’t just wrong, and it’s not just wronger than wrong. It is quite possibly the wrongest it is possible for any statement to be.

I can’t very well argue against the claim that there are no true statements, because I don’t even know what it would mean if it were right. Fortunately, nobody ever claims this outside of philosophy debates. Unfortunately, they sometimes do in them. Which is annoying, because despite the fact that it’s one of the Standard Philosophy Questions, it’s also pretty transparently silly.

I’ve asserted that it’s silly a couple of times, and I can’t really back that up except to say that it is literally logically impossible for it to be true. Like the “this statement is not true” thing, except that it’s exactly like that. Whatever your position on the Epimenides paradox, I highly doubt you would look at the sentence “this sentence is false” and say that it’s a true statement. The claim that there is no objectivity is at least that bad.

There is an objective and knowable universe out there. I know that because I’ve seen it, lived in it, and otherwise personally observed it. I can’t say with probability one that I observed it correctly, of course, but I don’t need to because I don’t believe in absolute certainty. I’ve got more than enough confidence to assume that the world is in fact reliably observable and act accordingly. I consider this to be awesome.

What’s more, you know some things about how the universe works, too. Even if you think we can’t Really Know anything, I guarantee that you act as if you do. You still flip a switch and expect a light to come on. You still eat when you’re hungry and drink coffee when you’re tired. Yes, it’s probably impossible to solve the problem of induction and be completely sure that all of this will still work tomorrow. But even if you can’t be completely sure, you are sure enough to act normally. Who knows? You might catch yourself expecting food to cure hunger and being right. So watch what you’re claiming, because you just might personally have some of this allegedly impossible knowledge. Which would be awesome.


One thought on “Yes, we can too know things

  1. I agree with the points you make in this post, but I’d like to go a step further. The concept of knowledge itself is not terribly useful and actually distracts us from the important epistemological project of assigning the correct probabilities to propositions we encounter.

    One of the conditions of knowledge is actually believing something that is true. But how confident do you have to be in a proposition before you can be said to believe it? Is the magic number any figure greater than 50? This seems so arbitrary as to make the concept of belief irrelevant, and it results in an intractable paradox.

    Suppose three people–Jack, Ben, and Stacy enter a fair lottery. You correctly assign the following probabilities:
    There is a 1/3 chance that Jack will win, and a 2/3 chance that he will lose.
    There is a 1/3 chance that Ben will win, and a 2/3 chance that he will lose.
    There is a 1/3 chance that Stacy will win, and a 2/3 chance that she will lose.
    It follows that you believe that Jack will lose, that Ben will lose, and that Stacy will lose. However, there is (just about) a 100% chance that one of them will win. It seems that your beliefs must conflict with one another despite the fact that you are assigning the appropriate probabilities.

    Furthermore, if the concept of belief is bankrupt and useless, than so is the concept of knowledge, since true belief is a condition of knowledge.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s