I recently received an email from an agnostic named Dan who was, ironically enough, quite militant about his agnosticism! According to Dan, “one cannot logically be an atheist” because “a negative can never be proven.” Notice, however, the statement “a negative can never be proven” is itself a negative statement. Either the statement “a negative can never be proven” can itself be proven, in which case the argument is self-refuting, or it can’t be proven, in which case it doesn’t provide a reason to reject atheism. I make this and other points in my essay, “Is a Sound Argument for the Non-Existence of a God Even Possible?“

When I referred Dan to the article, I explained that there are actually two ways to prove the nonexistence of something. One way is to prove that it cannot exist because its very concept is self-contradictory (e.g., square circles, married bachelors, etc.). The other way is by carefully looking and seeing. Both of these methods can and have been used to disprove various conceptions of God.

Dan took issue with the second method because it “presumes your senses and abilities are w/o bound.” Not really; the fact that we are finite beings does not prevent us from legitimately concluding that, given some body of evidence, a particular hypothesis is more probable than another hypothesis. This goes for conclusions about God’s existence just as it does for other areas where inductive logic is used, such as weather forecasting and criminal forensic investigations.

I suspect that Dan is holding both theism and atheism to a much higher evidential standard than we apply to other empirical questions. The fact that we might discover some new item of evidence in the future that supports a contradictory conclusion in no way undermines the fact that the evidence we have today supports an explanatory hypothesis. For example, the probability that the Declaration of Independence was signed on July 4, 1776 is extremely high, but it is not 100%. I suppose it is logically possible that in the future someone might discover historical evidence demonstrating the Declaration of Independence was signed on February 2, 1775. But the fact that such a thing is logically possible is irrelevant to the inductive (probabilistic) conclusion that the Declaration was indeed signed on July 4, 1776. In other words, a conclusion can be highly probable even if it is possible that the conclusion is false.

As a nontheist, I don’t demand that someone prove to me that the existence of God is absolutely certain (i.e., has a 100% probability). I would settle for an argument showing that the total relevant evidence merely makes the existence of God highly probable. Similarly, we don’t need absolutely certainty (i.e., 100% probability) in order to know that there is no God. We can use inductive arguments to show that God’s nonexistence is more likely, even much more likely, than God’s existence. And, indeed, I think there are inductively correct arguments that show religiously significant conceptions of God–such as the God of Judaism, Christianity, and Islam–do not exist.

After corresponding with presuppositionalists who make the ridiculous claim that there are no atheists, I thought I had heard everything. I was wrong. After reading my defense of the second method of proving the nonexistence of something, Dan then proceeded to tell me that I am “really” an agnostic but I just refuse to admit it. This remarkable conclusion is supposed to follow from the fact that I don’t claim to be able to prove with absolute certainty that God doesn’t exist. The fact of the matter, however, is that ordinary usage of the word “atheist” does not support that conclusion. There is no requirement that one has to be absolutely certain that there is no God in order to qualify as an atheist. I have never claimed to be absolutely certain that atheism is true, but it doesn’t follow that I merely lack belief in God or that I am not an atheist. Rather, I believe God’s existence is very improbable.

To make an analogy, I don’t know with absolute certainty that the sun will rise tomorrow, but I am highly confident that it will (i.e., it has an epistemic probability for me of >99.999%). That hardly means I am “agnostic” about whether the sun will rise. (In other words, one can hold a belief without assigning an epistemic probability of 100%.)

I’m reminded of Taner Edis’s recent discussion about individuals who claim that everyone is born ‘my’ way (i.e., theists who claim that everyone is born theist, atheists who claim that everyone is born atheist). Granted, Dan’s claim (that I and at least some other atheists are really agnostics) is not quite in the same category as the claims that Taner was discussing. But it is interesting to me that so many people’s argumentative strategy includes defining one’s ‘opposition’ out of existence!