On Philosophy

May 10, 2007

The Basis Of Knowledge

Filed under: Epistemology — Peter @ 12:00 am

To call some statement knowledge is to recommend it, which is to say that holding that statement to be true is in some way better or more useful, in general, than holding it to be false or being undecided. (Note: it is useful or beneficial because of its content.) I say in general because there may be some specific circumstances, such as in a completely accurate lie-detector test, where the opposite may be true. For example if you are a murderer it is beneficial to know who you murdered, in order to avoid being caught for those crimes. But when subjected to such a lie detector test it would be beneficial if you believed that you hadn’t killed anyone, even if you had. Obviously there is a parallel with ethics here. To say that an action is ethical is to say that it is recommended for everyone, always. But we also recognize that there are rare occasions in which other, more pressing, constraints give individuals reason to act unethically. Likewise, we can hold that knowledge is the beliefs that are recommended for everyone, always, and still admit that there are rare cases in which it may be better not to have the beliefs we call knowledge, because of other unusual factors.

This definition of knowledge is a flexible one; it even categorizes some false statements as knowledge (and is probably a better definition of knowledge because of that). For example, our current scientific theories are probably at best incomplete, and thus in a sense false. However they are also the best approximations to natural law that we have, and so we would call them knowledge, because it is beneficial (useful) to hold them as true. But despite its flexibility we can show, along with some relatively uncontroversial premises, that all knowledge must have an ultimate basis (reasons to believe it) in objective facts.

Our first additional premise is that if something is knowledge additional facts / consequences can be derived from it. We defend this premise simply by noting that if nothing can be derived from some statement then it can’t be useful, and hence not knowledge. For example, consider the belief that the cat is in the box. From this it follows that the cat is not outside the box, that the box is not empty, that to empty the box I must at least remove the cat (and possibly other things), and so on. Our second additional premise is that knowledge is transferable. Which is to say that if we know some person A to be genuinely trustworthy and reliable, and they tell us they know some fact S, and they tell us the grounds on which they know S, then we know S. This seems reasonable because without some kind of transferability we would have to learn everything from scratch, and because we don’t learn everything from scratch it would follow that we know nearly nothing. And that clearly is absurd. The third, and last, additional premise is that if some statement is knowledge then there is some more general rule that leads us to think that it is knowledge. This premise is really more definitional than anything else, it is simply a statement to the effect that all knowledge has some kind of basis and that the way particular knowledge follows from its basis (experience, other knowledge, or whatever) isn’t arbitrary. Again, this “premise” isn’t really involved in the argument, it is more of a way of framing the question.

So suppose that some kind of subjective facts were a valid basis for knowledge. Now assume that a known trustworthy and reliable person, A, experiences some subjective facts and correctly concludes S. Person A thus knows S. Person A then tells you of this fact, including about the subjective facts that makes them believe S to be true. You thus know S. Now consider an equally trustworthy and reliable person B. B has incorrectly derived ~S because they are mistaken about some subjective facts (they actually experienced subjective facts C, but have come to falsely remember the experience as containing subjective facts D). Person B tells you ~S and about the subjective facts that make them think that it is true. You thus know ~S. But this is a contradiction, because you can’t know S and ~S. Now we don’t have to assume that the rules of deduction when it comes to knowledge are as fragile as those of classical logic, such that they explode in the face of a contradiction. Instead we should assume (as in paraconsistent logic) that nothing can be derived from either S or ~S until the contradiction is resolved (more precisely: that you can derive facts from knowing S or ~S if you know that it is impossible for both S and ~S to have a valid basis). Now if the basis of these statements were objective facts then we would resolve the contradiction by looking to the reasons persons A and B claim to know S and ~S and using that to resolve the disagreement. (This is the reason that we specified that grounds had to be provided for knowledge to be transferred.) But in the case of knowledge grounded in subjective facts contradictions can’t be resolved in this way because we don’t have access to the subjective facts that were the basis of the beliefs of A and B. This is why we call them subjective instead of objective. But this rules out subjective knowledge being transferable in general, even if no one has come forth to make the opposite claim on subjective grounds yet, because we simply can’t have any confidence in those claims. Without an objective basis they simply aren’t very resilient; we can’t have any confidence that they aren’t in error; even knowledge based on subjective facts that had managed to endure for thousands of years would have to be discarded overnight on the basis of testimony of a single person. (More precisely: we can never rule out the possibility that both S and ~S might have an equally valid basis, even if we have only been presented with a basis for one of them, given that they are subjective. And thus we can never deduce anything from them.)

So knowledge based on subjective facts isn’t transferable. But couldn’t I believe things on the basis of subjective facts revealed to me? Well given that it isn’t transferable we can conclude that the content of these statements must have no objective consequences. (Note: content, not the belief that they are true; the belief has obvious consequences, the first of which being that I will have the disposition to affirm it as true.) If it did have objective consequences then these could be appealed to in order to transfer that knowledge from one person to another. (Perhaps this is just a way of saying that subjective knowledge is subjective.) It is hard to think of beliefs that don’t have objective consequences, but let us assume that there are some. Given that they have no objective consequences it is hard to say how holding them to be true could possibly be better or more useful, which was our initial criterion of what counted as knowledge. Perhaps we could argue that some of these beliefs make people feel better. But if they know that what they believe has no objective (read: real) consequences it is hard to see how it could make them feel better. On the other hand, if they believe it does have some objective consequences then they will be mislead by it in certain situations, and hence it will actually be to their disadvantage.

The obvious consequence of all this is that there is no such thing as revealed knowledge. (see also) And thus that faith can be thrown out the window as the source of knowledge. But we all knew that already. The more interesting consequence is that we can argue that if even for something as seemingly fundamentally subjective as consciousness that if there is some knowable fact of the matter as to whether people (even ourselves) are conscious then consciousness must have a objective (physical) basis.


Blog at WordPress.com.

%d bloggers like this: