Agnoticism

Gary Gutting responds to shaunphilly's query, saying,

Of course, you can use the terms the way you think best. But your way of putting things ignores two importantly different ways of not believing that God exists. You might not believe in the sense that you withhold judgment as to whether God exists OR in the sense that you believe that God does not exist. In ordinary usage, the first sense of not believing in God is called “agnosticism” and the second is called “atheism”. It seems to me that this is a useful distinction, and I don’t see what you gain by eliminating it.

But I think Gutting himself is erasing a distinction between two or three different kinds of agnosticism.

Atheism is primarily a social and political term. Millions of people who are not professional philosophers or theologians, who don't care what distinctions philosophers and theologians care to make, self-identify as atheists primarily because they're convinced that all these priests, bishops, popes, rabbis, imams, gurus, and other assorted spokesmodels for God are completely full of shit, and they aren't afraid of saying so. But that's beside the point.

The point is that philosophically aware atheists, do not, as Gutting asserts, want to erase the distinction between atheism and agnosticism. Rather, they want to draw the distinction between not knowing because we haven't yet looked and "not knowing" because the proposition is not knowable.

In The God Delusion, Richard Dawkins distinguishes between Temporary Agnosticism in Practice (TAP) and Permanent Agnosticism in Principle (PAP). Under TAP, the speaker believes that the information to answer the question is available; she just doesn't actually have it. For example, there might or might not be life on Europa or Titan. I believe that it's a question we could answer, the information is "out there", but we don't yet actually have the information to draw a firm conclusion. Under PAP, the information to decide one way or another is just not logically possible to have. There is no way of knowing one way or another, for example, whether or not there's a ninja hiding in my bedroom. (This principle is purely epistemic. The ontological status of the assertion is not really relevant: there's nothing ontologically exceptional about a ninja hiding in my bedroom.)

(There's also the sense of "agnosticism" as not making specifically a priori assumptions about some proposition, either because one checks a posteriori or because the proposition is irrelevant. Notepad (the simple text editor that comes with Windows) is agnostic about ASCII and Unicode: it will check and try to determine the encoding of a file. A program that does a binary comparison of two files is also agnostic about ASCII and Unicode; in this case, the encoding is irrelevant.)

There's asymmetric agnosticism: where we could in principle know only the truth of some proposition, but we cannot in principle know its falsity (or it would be many orders of magnitude more difficult to know the falsity or the truth). If a ninja stepped out of the closet and said, "Hello!" I could easily know it's true that there's a ninja in my bedroom, but I cannot ever know it's false; the ninja might be hiding.

It's important to draw this distinction between kinds of agnosticism because it's a routine apologetic strategy to draw the distinction between atheism and (some form of) agnosticism and then equivocate between the different forms of agnosticism: since you're an agnostic, maybe you should actually look before you draw a conclusion. Anyone who's following my The Stupid! It Burns! series will see examples of this kind of thinking. Indeed, by chance the most recent example shows this thinking: "Absolute knowlege [sic] is required to make such a claim that 'God does not exist'."

It is, I think, subtly but deeply misleading to say that one does not know the truth of a proposition that one cannot know. To say that one does not know implies to the ordinary human mind that one might know. At a more philosophical level, it's been asserted by actual canonical philosophers (e.g. William James' "Leap of Faith") that one is actually in a sense justified in having any position on a proposition one cannot know.

But this attitude leads to a severe problem when the proposition in question is about the real world. We intuitively believe that propositions about the world are inherently truth-apt: all (definite and fully-qualified) statements about the world are either true or false. If one is actually justified in believing both the truth and falsity of a truth-apt proposition, then we have a logical contradiction. If I'm justified in believing that "God exists" (for an unknowable God), a statement about the real world, then in a sense "God exists" is true; if I'm equally justified in believing "God does not exist", then "God exists" is false. The only way out is to hold that both statements are actually meaningless, albeit "meaningless" in a more abstract sense than "goo goo ga ga". And why should we require atheists to have any position on a meaningless proposition? When the theist defines God as unknowable, he has not challenged the atheist's position, he has merely lapsed into incoherent babble.

Why should a professional philosopher — in a profession that claims to be about clarity and precision of thought — endorse incoherent babble?