Topics Knowledge

Question How do we justify our knowledge of the external world? Knowledge of the external world seems to be fallible in any case if we put the threshold of success at the highest level, namely 100% certainty. But this still raises a question: if we want to avoid complete skepticism, how can we be certain that our knowledge is at least likely to be true? In order to create a probability about the validity of our knowledge of the external world we need to start from perception. The problem is that we can be certain of the existence of perception but not the source of it (the matrix/the real world), and that is essential for the knowledge of the external world. In order to calculate our probability we then need the number of possible events E and the one favourable event F we're looking for: E = 2 possible events are external source or non-external source (matrix, hallucination, dream etc.) F = 1 favourable event i.e. external source P(F) = F/E = 1/2 = 50% It seems to me that both possibilities are equally likely. Why should I believe one over the other? A lot of people answers this question by saying that the simulation hypothesis is too convoluted and not as simple as common sense realism. But Doesn't that depend on the context? For example: If the reality in which I'm simulated there's an infinite amount of simulations the hypothesis would not be that convoluted. If I have no access to the external reality how am I supposed to establish what is convoluted and what is not? My question is: What makes external world realism more plausible?

Accepted:
August 20, 2019

Comments

Setting external world

Setting external world skepticism aside for a moment, suppose I'm about to roll a die. Now there are two possibilities: it will come up 1 or it won't. If I reason as you did, I will conclude that the probability is 1/2 that the die will come up 1.

Something has gone wrong here. For one thing, we can't get the answers to probability questions just by counting. There are many ways to slice up the space of possibilities, and if we use your rule, the answer we get will depend on how we do the slicing. This is a well-known problem, and there is no simple fix. But there's another problem: the probabilities here aren't chances. They are degrees of belief. Even if we thought (though we shouldn't) that the right way to slice things up is that our experience has an external source or it doesn't, without adding anything more fine-grained, we don't have to agree that the two possibilities are equally probable. You say "it seems to me that both possibilities are equally likely." It's worth wondering whether you really know your own mind on this question, but even if you do, that carries no weight for the rest of us.

Some people call these degrees of belief personal probabilities, so let's follow suit. My personal probability for the proposition that none of my experiences have an external source is infinitesimal. I give that possibility essentially no credence at all. That's not a justification for rejecting external-world skepticism. It's a confession of my incredulity.

Now you might think that any philosopher who admits this should turn in their union card. Interestingly, this would mean that many professional epistemologists would have to look for a new profession. Near as I can tell, most philosophers who work in theory of knowledge aren't much tempted by external world skepticism. They assume that there's an external world and that we know things about it. The goal is to understand the structure of that knowledge.

You write "If I have no access to the external reality, how am I supposed to establish what is convoluted and what is not?" Leave aside the question of whether this is the right way to think about judgments of what's convoluted. If there is an external world of more or less the sort we usually think, then you do have access to it. You see it, hear it, touch it, taste it, smell it. The fact that you doubt whether your experiences are veridical doesn't show that they aren't caused by something outside your mind.

One of the problems with radical skepticism is that the skeptic typically is looking for something we couldn't have (certainty about contingent matters) and sees the fact that can't have it as a reason to think the skeptical hypothesis is credible. But even if we talk in terms of probability (not that I actually think it will help us much to do that here), not having probability one is not the same thing as not knowing anything. I know things about the world if I believe them, they're true, and I'm actually in contact with things in the right sort of way. I can believe things even while admitting I could be mistaken, the things I believe could be true, and what's responsible for my beliefs could be the external world, more or less as I believe it to be. Given the notion of proof the skeptic usually has in mind, I can't prove that I have any such knowledge. But (to repeat my confession) I don't see this as anything I should lose sleep over. You think solipsism or something like it is as likely as not. I don't have the slightest inclination to agree. And though this is hardly a proof of my view, my view has the advantage of freeing us up to think about questions that seem to me, at least, to be a lot more interesting and productive.