Thursday, January 08, 2015

Do we have a priori knowledge?

By the law excluded middle, an item of knowledge is either inferred from prior items of knowledge or it is not inferred from prior items of knowledge.  Items of knowledge that are inferred are called a posteriori, and items of knowledge that are not inferred are called a priori.  The only way to escape this dichotomy is to deny that there's any such thing as knowledge at all.

The denial of a priori knowledge leads to absurdity.  If you deny that it's possible to know something without inferring it from something else, then you would have to say that for everything you know, there is a prior item of knowledge from which you inferred it.

For example, let's say you know that Socrates is mortal.  Perhaps you inferred that item of knowledge from these two items of knowledge:

1.  All men are mortal.
2.  Socrates is a man.

But if each item of knowledge you possess must be inferred from something prior, then there must be further items of knowledge from which you infer 1 and 2.  Otherwise, you couldn't know that Socrates is mortal.  And then there must be further items of knowledge from which you infer those as well.

If you continue to think this through, you'll quickly see that it leads to an infinite regress.  If all of our items of knowledge must be inferred from prior items of knowledge, then the only way it's possible to know anything is if there is a beginningless line of reasoning from prior premises that leads up to your present state of knowledge.

But that is impossible for three reasons.  First, none of us know an infinite number of things.  Second, because nobody has been around long enough to make an infinite number of inferences.  Third, because if there were no beginning, there would be no way to even get started making inferences.  You couldn't very well start at the beginning because there's no beginning to get started from!

So it is impossible that all of our knowledge is a posteriori.  If we know anything at all, then at least some of our knowledge must be a priori.  That means that unless there are some things we know immediately without inferring them from anything else, then knowledge is impossible.

Some people may be willing to bite the bullet and say, "Well, okay, then nobody really knows anything at all, including me."  This view is called global skepticism.  It is a self-refuting point of view for a couple of reasons.

First, it's self-refuting because if nobody knows anything, then we wouldn't know it.  A person who claims not to know anything is stating something he doesn't know to be true.

Second, it's self-refuting because if nobody knows anything, then nobody knows that any of the premises that lead to the conclusion are true, nor that the conclusion follows from those premises.

Besides that, there are plenty of things we obviously know.  We each know that we exist.  We know that two contradictory statements can't both be true at the same time and in the same sense.  I know that I have a sister named Jennifer and a brother named James.  There are all kinds of things we know.  So there must be a priori knowledge.

1 comment:

K Y Lai said...

[I'm Diqiucun_Cunmin on Debate.org]

I agree with you completely. I think a priori knowledge, as you philosphers call it, is the stuff inherent in the faculty of our brains. Chomskyian linguistics, for example, posits the existence of a univeral grammar in our language faculty. The guy modifies his theories every decade or so, but the gist of it is the same. There is bound to be some language knowledge that we, as humans, know congenitally.

An example of this is the locality principle, a principle in Chomsky's principle and parameters theory. For example, the first sentence is right and the second wrong:

Will you go to the zoo?
*Go you will to the zoo?

To explain this, we need to look at the D-structure of the sentence:
You will go to the zoo?

To decide whether we move the modal or the lexical verb to the beginning, we need to consider which movement is shorter. The modal moves one place to the beginning, while the lexical verb moves two places, so the modal wins.

It is important to note that babies do not learn the locality principle during language acquisition. Babies do not produce sentences that violate the principle, so adults cannot correct them for it. Locality principle-violating sentences only occur in linguistics books.*

*Stimulus-based language acquisition has been debunked. In fact, input from adult language is not the only thing that affects what a child says. Children do production of their own during L1 acquisition. This is known as the poverty-of-the-stimulus argument (which has a formal definition, but I'm putting it in informal terms here).

(I might be telling you stuff you already know, but I decided to put this here in case you don't, just to add some scientific basis to your philosophical thoughts ^^)