In spite of the shortcomings of the hit movie, The Matrix, the first film in the trilogy raised important questions about human knowledge.
For example, how did Neo know when he was in the Matrix and when he was in the real world? If his experiences before taking the red pill were produced by the Matrix, how could he be sure his experiences after taking the red pill weren't also produced by the Matrix? In other words, how could Neo be certain he ever left the Matrix? How could he be certain that his experience of leaving the Matrix wasn't itself produced by the Matrix?
Much more importantly, if Neo could never be sure that he wasn't in the Matrix, how can you and I be sure that we're not in the Matrix? And if we can't be sure — if we can't offer an airtight argument for the conclusion that we're not — how can we claim to know anything?
Let's call this line of questioning "skepticism." In this article, I'll critique skepticism by showing that it makes indefensible assumptions about knowledge.
We can distinguish two different questions relevant to the human quest for knowledge.
Question (1) asks about the specific items of knowledge you possess; answering it gives you a picture of the extent and limits of your knowledge. Question (2) asks about the specific circumstances in which a belief would count as knowledge — it asks what all instances of knowledge have in common with each other.
Now suppose you want to appraise the intellectual respectability of your worldview and, in order to do so, decide to sort all of the beliefs that compose it into two groups: those that count as knowledge and those that do not. How should you proceed? Can you sort your beliefs with any accuracy if you don't know what the criteria for knowledge are? If not, then it seems that you should start by answering question (2). But the only way to answer question (2) is to look at instances of knowledge and see, first, what they all have in common and, second, how they differ from beliefs that don't count as knowledge. But you can only do this if you've already answered question (1).
So you can't answer question (1) if you don't first have an answer to question (2), and you can't answer question (2) if you don't first have an answer to question (1). And if you can't answer either (1) or (2), then how can you know anything at all? This predicament is what philosophers call the problem of the criterion. 1
There are three main responses to the problem of the criterion. First, there is "skepticism." The skeptic claims that no good solution to the problem exists and that we therefore can't know anything. Second, there is "methodism" (no connection to the denomination). According to the methodist, we can know things, and the solution to the problem starts with an answer to question (2). Moreover, methodists claim that a belief can only count as knowledge if you first know (a) what the criteria for knowledge are and (b) that the belief in question meets these criteria.
Unfortunately, methodism leads to serious problems. If knowledge of any belief requires prior knowledge of (a) and (b) — as methodism claims that it does — then the skeptic can ask the methodist how he knows (a) and (b). And since (a) and (b) are themselves beliefs, answering the skeptic will require the methodist to make further knowledge claims. But defending these knowledge claims will require him to make even further knowledge claims, and these will need defense as well, which will require yet more knowledge claims ... and so on and so forth to infinity. 2It would seem, then, that methodism is in trouble.
The third response to the problem of the criterion is called "particularism." According to particularists, you know many things without being able to prove that you do and without understanding how you know them. Thus you can answer question (1) without having to possess or apply any criteria for knowledge — i.e. without first answering question (2). Moreover, reflecting on your answers to question (1) will allow you to develop criteria for knowledge consistent with them. This criteria can then be used to make judgments in borderline cases of knowledge. 3But note that these criteria are justified by their congruence with specific instances of knowledge rather than the other way around, as in methodism.
Of course, the skeptic objects to particularism as well. First, he'll claim that particularism begs the question 4by simply assuming the very thing in need of proof — that some of our beliefs count as knowledge. Second, he'll try to force the particularist into methodism — and all of its problems — by asking him such questions as "How do you know you've picked out the right beliefs as instances of knowledge? Isn't it possible that you're wrong? And if it's possible that you're wrong, shouldn't you have to prove that you're not?"
Fortunately, particularism offers a good response to both of these objections. Regarding the charge that particularism begs the question, the skeptic claims the particularist must prove that some of our beliefs count as knowledge. But if the skeptic doesn't offer any reason for thinking this, his skepticism can be dismissed as arbitrary, rooted in personal preference rather than a substantive position or argument. If, on the other hand, his skepticism is the result of an argument, this argument must be reasonable to be taken seriously. But how can an argument be reasonable unless its premises count as knowledge? 5And if its premises do count as knowledge, then it can't reasonably conclude that none of our beliefs count as knowledge. Unbridled skepticism is not a rationally defensible position; it cannot be rationally asserted and defended without presupposing that some of our beliefs count as knowledge.
Regarding the skeptic's attempt to push the particularist into methodism, the particularist can resist by reaffirming that he can know things without being able to prove that he knows them. For example, the particularist could say "I know that mercy is a virtue even though I can't prove that I know it. Moreover, why think I have to prove I know it before I can know it?"
Further, the particularist argues that, just because it's possible that he's mistaken about the beliefs he counts as knowledge, that doesn't mean he is mistaken; nor does it mean he has good reason to think he's mistaken. For example, suppose I claim to know that I first visited Disneyland in 1985 and a skeptic points out that it's possible that I'm mistaken. He's right; it is possible. But it doesn't follow that I first visited Disneyland in some year other than 1985, or that I never visited Disneyland at all. Unless the skeptic gives me good reasons to think that I didn't first visit Disneyland in '85, the bare possibility that I might not have isn't sufficient to call into question my claim to know I did.
The particularist and skeptic have very different approaches to knowing. Of the two main tasks in the quest for knowledge (obtaining true beliefs and avoiding false ones), the skeptic elevates the latter over the former, whereas the particularist thinks obtaining true beliefs is at least as important as avoiding false ones. For the skeptic, the burden of proof is on anyone who claims to know something. For the particularist, the burden of proof is on the skeptic; he requires the skeptic to make a good argument for skepticism before he allows the skeptic to bother him about knowledge. And given the fact that the skeptic can only make a good argument for his skepticism by assuming the very opposite of his skepticism — that he actually knows things — the particularist sees no reason to deny what's obvious to all of us: that we know many things after all.