This is an issue I see a lot in apologetics and in arguments about theism vs. atheism. Certain theists (who I will just refer to as “theists,” even though I’m not referring to all theists) want to be able to put forward the existence of god as a hypothesis that can be argued for or corroborated by evidence. Belief in god is a positive belief, it does not exist by default in humans. It is something that must be taught or learned. This means that the null hypothesis is the hypothesis that god does not exist. In order to show that the god hypothesis is true, there must be a reason to reject the null hypothesis. In order to do that, the theist must understand the null hypothesis.

When comparing two hypotheses we can distinguish between them by their consequences. If such-and-such is true, then thus-and-so must also be true, where thus-and-so is something we can observe or test to check whether it actually is true. In case it’s false, the hypothesis is falsified. In case it’s true, the hypothesis is corroborated. To compare a hypothesis against its null hypothesis, we need to identify predictions or logical consequences that differ between the two and which can be empirically checked. If we cannot identify anything different, then the hypothesis is unfalsifiable.

Crucially, what is the difference between a world where god exists and a world where god does not exist? The answer cannot be that “there would be no world in the first place without god,” since the question of whether god created the universe is separate from the question of whether god exists. For example, it’s possible that god and the universe have both always existed. Additionally, theists must be able to accept the possibility (as in be able to consider it for the sake of argument) that space and time “began” through some naturalistic mechanism not yet known to science.

Let’s think about possible differences a theist might point to, for example “there would be no reports of miracles if god did not exist.” This kind of reasoning doesn’t work well, because based on our understand of the world and of human psychology, we would expect naturalistic phenomena to sometimes be interpreted as miraculous and for stories about miracles to be present in our culture. What about something science has no explanation for? This gets into the “god of the gaps” issue, but let’s set that aside. In order to use such a phenomenon as a test for the god hypothesis, we have to have some reason of expecting it to be different under the null hypothesis. This would work if, for example, prayer was highly effective and science could not determine why; the problem is, prayer is not more effective than we would expect it to be under the null hypothesis.

The upshot of all this is that the claim that god exists is unfalsifiable. The positive “strong atheism” claim that there is no god is also unfalsifiable. Either view is consistent with the world as we actually experience it. This is not to say that theism or strong atheism are bad positions to hold nor that people shouldn’t be religious. What it does say is that many arguments in apologetics are dead in the water. At best they can be interpreted as rhetorical arguments aimed at convincing people god exists.