Is Agnosticism Possible?
To be agnostic on a subject is to say that you do not know what is true, to not take a position one way or the other. Although one can refuse to state a position, however, it is sometimes impossible not to take a position, or even not to believe one position or the other.
Let’s start with an uncolored example. Suppose I say I am agnostic about whether the stock market will either crash this year, or double in value. I can say that, but you can tell my true belief by looking at my investment behavior. If I stay invested in stocks, or I sell all my stocks, you can tell that I am not really agnostic.
On the other hand, suppose I say I am agnostic on whether there is life outside the solar system. That is a subject on which I can truly stay agnostic. It does not affect my behavior in the slightest, so I can truly avoid believing one thing or another. Put differently, there is no way for me or anyone else to know what I really believe. Even if I say I am *not* agnostic, and strongly believe in life in the stars, it is hard for me or others to know how sincerely I believe that.
Come now to belief in God. Can one be agnostic about whether the Christian God exists? I think not. If you believe He exists, you must worship Him, or at least fear Him. If you do not, then you believe He does not exist. There is not a middle ground.
I do not mean to say that there is no middle ground in terms of confidence in one’s belief. It is certainly possible to waver between belief and disbelief. But a behavioral decision must be made. And a decision can be made with just a hairsbreadth margin of confidence.
If you are told that either the floor where you are standing now, or the floor of the room through the door will disappear and leave you to fall into a fiery furnace, you cannot be agnostic. You must either stand still or walk through the door.
March 5th, 2006 at 7:03 pm
Very good post, Professor.
March 6th, 2006 at 4:44 pm
Suspicion is not knowledge. To be agnostic is to be without knowledge (gnosis being Greek for knowledge). You can suspect there is or isn’t a God without knowing. I suspect there is not, but I don’t pretend to know. That’s why I don’t consider myself an atheist. I consider atheism to be an act of faith.
March 7th, 2006 at 5:57 am
I think the logic of (at least) the last example is flawed. Given an event with 50-50 probability of happening, with no way of determining the outcome, the rational choice is to be agnostic and do nothing. If I am sitting in this room and you tell me the floor of either this room or the next will open up, since I have no reason to select one over the other — I’ll just save my energy and stay where I am.
April 5th, 2006 at 4:30 pm
This comment is along the same lines as those left by “Mark” and “Doug” above.
I have always considered God’s existence to be a fact which can neither be proven nor disproven. I do not rule it out as impossible but I do assign a very low probability to it subjectively. In fact when it comes to the existence of a particular god – say the particular form envisioned by the Lutherns or the Mormons, etc – I think there is a good argument to be that such an event has measure zero. This is because there are potentially a continuum of possible gods – each of which is equally unverifiable. (Akin to saying that the probably of uniform random variable on (0,1) turning out to be exactly .3 is zero – which is not to say that .3 is impossible.)
Does it affect my behaviour? Well yes, I suppose it does. Given my beliefs (low probability of gods existence in general and zero probability on the existence of a particular god), my behaviour (when I am acting rationally) is observationally equivalent to that of an aetheist. This equivalence can essentially be explained by information costs, transaction costs and time preference.
November 26th, 2006 at 2:00 pm
God’s existence is hard to prove or disprove, but that’s true of lots of things– whether Homer existed, why Louis XIV didn’t shoot at protesters, whether the current global warming is caused by carbon dioxide, etc. The difference is that God’s existence is a more important question, and isn’t usefully attacked by accumulating more facts.
The idea that any particular God has measure zero is interesting, but we can use more probability theory to get round it. A completely precise type of God has zero probability, but is near very similar types— so similar, in fact, that our senses cannot tell the difference. THus, the probability of any observable type of God is positive.