A rather interesting question to consider regarding atheism and theism is this one: is either theism or atheism verifiable or falsifiable?
First, let us consider theism. A scientist might claim that theism is not verifiable, on the grounds that there are no experiments or observations that we might make that would show the existence of some type of deity. But this claim is not valid, and is typically based on a presumption of the nonexistence of any deity. When talking about whether something is verifiable, the rules of the game are that we temporarily suppose the thing to be true, and then we imagine whether there are any observations we might make that would verify such a thing is true.
Following such a protocol, we can indeed imagine observations that might confirm the existence of a deity, if a deity exists. For one thing, it is possible that such a deity might reveal itself in some spectacular way to the entire world. An act such as writing a giant glowing message in space or causing the sun to blink out a message in Morse code would presumably cause us to have an observation that would verify the existence of a deity. It is also possible that when we die, we will have some supernatural experience that would qualify as a verification of the existence of a deity. Indeed, some people who have near-death experiences claim to have had such experiences.
So theism is verifiable, in the sense that hypothetically there are observations that we might make that could verify the existence of a deity. I may note here that there is no sound basis for distinguishing between that which is verifiable and that which is “scientifically verifiable.” The term “scientifically verifiable” implies a kind of conformity or similarity to existing methods of scientists, or the current assumptions of scientists. But when talking about what is verifiable or falsifiable, there is no reason why we should make such a restriction. It is all too possible that something might be verified or falsified in some way that does not conform to existing methods or expectations of today's scientists.
Now let's consider: is theism falsifiable? An atheist might argue that theism has already been falsified, because there is so much evil and suffering in the world. But this argument is not valid for two reasons, a simple reason and a complex reason. The simple reason is that evil or suffering could never show the nonexistence or unlikelihood of any deity. At best it might show the nonexistence or unlikelihood of an omnipotent deity, leaving unscathed the possibility of a deity with finite power. The complex reason is that there is always the possibility that earthly experience may be some tiny thread in a vast tapestry, one that might include a happy afterlife for everyone. It could be our earthly experience is some microscopic fraction of some commendable million-year plan in which suffering is necessary for diversity, growth, variety of experience, moral freedom and other good things (and necessary for the avoidance of the evils discussed here). It is possible that if we were to understand this million-year “big picture,” we might conclude that earthly suffering is quite compatible with the existence of an omnipotent deity.
Considering the two reasons listed above, we must conclude that theism is not falsifiable. Even if we imagine the most disastrous observations we can imagine, they would not actually qualify as proof that no deity exists. For example, imagine an asteroid strikes our planet and kills almost everyone, or perhaps every single human. While that might cast much doubt on the idea of an omnipotent deity who rules over our planet with loving care, it would leave unscathed the possibility of a somewhat less than omnipotent deity who might not have the power or inclination to watch over each of billions of planets in the universe. In addition, there would still be the possibility that an omnipotent deity does exist, and that such a disaster is some necessary part of some million-year plan in which devastating setbacks are a vital ingredient. The possibility that everyone will have a happy afterlife despite such a disaster takes the power out of emotional arguments such as “but no deity would allow the little children to die.”
It seems, therefore, that theism is verifiable, but not falsifiable. We can imagine some hypothetical observations that might cause us to say with great conviction, “Now that proves it – there is a deity.” But we cannot imagine some hypothetical observations that might warrant us saying with great conviction, “Now that proves it – there is no deity.”
Now what about atheism: is it verifiable, falsifiable, or neither? The points I have just made about theism not being falsifiable can be used to help establish that atheism is not verifiable. For atheism to be verifiable, we would have to make some observation or observations that show that no deity exists. Such observations could not be made. Even if we were to witness the most horrible tragedy here on our planet, that would still leave undamaged the possibility that the universe was designed or created by a still-existing deity with less than perfect power, who lacks the power to guarantee pain-free conditions on each of the countless planets in the vast universe. Even the possibility of an omnipotent deity would still be standing, because there would still be the possibility that the terrible tragedy was a necessary ingredient in some million-year plan too complex for us to understand (and it would still be possible that all of the loss of life wasn't really a final snuffing out of consciousness, because of the possibility of an afterlife).
So atheism is not verifiable. But is it falsifiable? Yes, it is. We could falsify atheism merely by having some observations that might convince us that some deity exists. It is easy to imagine how such observations could occur – for example, there might appear a giant “God exists” sign in outer space that stretched millions of miles. Or one day the world might be visibly taken over by some divine power. Or after dying you might see some magnificent heavenly scene that would be sufficient evidence of a deity's existence.
Now, an atheist might agree that atheism is falsifiable, and might claim this as a strength. He might argue that this shows that atheism is a really scientific claim. It has been argued that all really scientific claims are falsifiable.
But this would seem to be a case of trying to claim an advantage that isn't really there. For one, thing it is not actually true that all valid scientific claims are falsifiable. If Karl Popper thought this, then he didn't think things through. Here is one example of a major scientific theory that could never be falsified: the theory that extraterrestrial life exists. If we lived in a tiny little universe, we might be able to falsify such a theory, by quickly exploring all planets. But we live instead in a universe of billions of galaxies, a large fraction of which have billions of stars. Even if you imagine a fleet of a million warp-speed spaceships traveling throughout the universe trying to falsify the theory that extraterrestrial life exists, we must imagine that such a search would take many millions of years to check all the countless trillions of planets scattered throughout billions of galaxies. Even if such a search found no extraterrestrial life, it would still not falsify the theory that extraterrestrial life exists. For it would still be all too possible that the search had failed to find extraterrestrial life that was hiding or hard-to-find, and it would still be all too possible that while such a search (spanning millions of years) was going on, some other extraterrestrial life had arisen on the planets that had already been checked.
Too many of these to falsify the theory that extraterrestrial life exists
So the theory that extraterrestrial life exists cannot be falsified. This example shows that it just isn't true that all scientific theories can be falsified. So the fact that atheism could be falsified does not give the atheist anything to brag about.
None of these considerations touch on whether atheism is more plausible than theism. But these considerations may provide a kind of pragmatic reason for not investing too much time in a thesis such as atheism. Our earthly life is short, and it is perhaps best for us to concentrate on claims that might one day be verified through some confirmation event. Just as it may make little sense for a scientist to devote decades on some physical theory that could never be verified, it may (from a purely practical standpoint) be inadvisable for an atheist to devote great energies to a doctrine that can never ultimately result in him being able to say: that proves it, I was right all along.