Taking the piss out of epistemology

I've got a degree in maths & philosophy, and one of the subjects I studied in philosophy was epistemology: the study of knowledge. Among the basic concepts in the area are truth and knowledge. What are they?

Defining truth normally proceeds along the lines of: the statement "P" is true if and only if P. For example, if P is "I have blue eyes", then we have: the statement "I have blue eyes" is true if and only if I have blue eyes. (As I recall, you can reach some thorny problems about whether or not this is really a properly-formed statement, as P is both referred to and dereferenced. But that's not relevant to where I'm going.)

Knowledge is supposed to be a justified true belief. This isn't quite enough; the justification needs to be of the right sort, and you can then tie yourself into knots about how and what that should be.

But I want to challenge the most basic assumption: that knowledge must be true.

I know that there are no aliens on the planet. I know the sun will rise tomorrow. I know the curtains in this room are blue.

Some of these statements may not actually be true. Perhaps there are aliens on the planet. Perhaps the sun will disappear tonight and not rise tomorrow. Perhaps the curtains in this room are white but lit in a special way to make them look blue. Perhaps I'm a brain in a jar, and all of this is illusory.

I say: it doesn't matter, and my justifications for my beliefs are what make them knowledge, to a greater or lesser extent. Truth isn't needed.

So my justifications would be something like this:

• I believe there are no aliens on the planet because I've never seen any plausible evidence of them. Given the other things I know about the world, it's an extraordinary claim that there are, and I demand extraordinary evidence for extraordinary claims.
• I believe the sun will rise tomorrow because it has risen every other day and I have no reason to believe it won't tomorrow.
• I believe the curtains in this room are blue because they look that way in the variety of lighting conditions I've seen them in, and I've no reason to believe
• I believe I'm not a brain in a jar because there's no evidence of it, and it's an extraordinary claim. (Note that the whole point of the brain-in-a-jar hypothesis is that you can't possibly know whether or not it's true; but that's fine, it doesn't challenge my definition of knowledge.)
  

My knowledge is relative to, and only as good as, my justifications. Those justifications may change and can be challenged: if I meet an alien on Earth, or read a credible newspaper report of one, I'll know they do exist; but my justifications for believing it will be different in the two cases. The justifications may come in any form; I may know some things from personal experience, some from reading about them, some from talking to other people.
 
Truth can also be made relative to scope. From my viewpoint, I know the sun will rise tomorrow. I may observe it, and so it turns out to actually be true. But perhaps there is a strange creature somewhere who is in a position to know that I am a brain in a jar, and there simply is no sun; but my statement can be true (in my world) and false (in its). With the right justification or explanation, the same statement, referring to the same thing, can switch from being true to false, or vice versa.

It's perhaps a more scientific than epistemological approach to knowledge. If my justification is challenged, then my knowledge is challenged. Things I know may turn out to be wrong, or right for the wrong reasons, in which case I update my justifications. But that's fine, that's how life goes. The traditional, philosophical view of knowledge doesn't allow that, because it demands knowledge must, above all, be true.

This thought came to me while considering relativity and Gödel's theorems. It seemed to me that the goal of traditional epistemology is similar to Hilbert's goal to prove completeness and consistency of maths pre-Gödel, or attempts in physics pre-Einstein to describe universal frames of reference.

Maybe there's a greater philosophical step to take from here. I'm making knowledge and truth relative to their justifications... in some ways relative to the language they're expressed in. What else is relative to language in a similar way? Well, science is a linguistic game played in a particular way: it demands particular conventions and relationships to reality, i.e. falsifiable hypotheses with instructions which, when followed, test and fail to falsify the hypotheses. Philosophy is a linguistic game where we demand arguments are coherent, logical, rational: what we have before we can formalize a set of rules to carve out a sub-part as its own subject. Perhaps literature is a superset of that, where the only rules are those of the language itself, and even those are there to be bent. Each area has its own quirks of language, its own rules and conventions, and what it can do is relative to them.