Two Fallacies That Aren't

God of the gaps and appeal to incredulity

1 Why you should commit many logical fallacies

“Some people think that the task of the critical thinker is to have a list of informal fallacies memorized … and just be ready to throw out those term, and even better if they’re in Latin.”

—Tomas Bogardus.

People love accusing others of committing logical fallacies. My friend Emerson Green has a nice article about the absurdity of this. Defenders of the legitimacy of these terms love to play this motte and bailey game, where when defending the legitimacy of the term, they’ll point out cases where the reasoning is actually invalid, but when the term is deployed it is almost always deployed against correct reasoning. Emerson gives the example of the so-called post hoc ergo proctor hoc fallacy, wherein you infer that A is caused by B because A came after B.

Obviously, it is true that because A comes after B, this does not logically require that A was caused by B. But no one in the history of the world has ever suggested it did. Instead, people use the fact that A came after B as evidence that A was caused by B, which is perfectly legitimate reasoning. If I see someone looking stressed right after an exam, it is not fallacious to think that the stress was probably caused by the exam. I submit that nearly every time in the world anyone accuses someone of committing the post hoc ergo proctor hoc fallacy, it is being used against someone making a probabilistic inference, which is perfectly good reasoning. Of course, it’s not always justified to assume that A was caused by B because it came after B, but it can be some evidence that A was caused by B.

There are two fallacies, however, that are very commonly discussed when debating theism that annoy me to no end. So here, I’ll explain why, as standardly formulated, they are perfectly good reasoning and not, in fact, fallacious.

2 Appeal to incredulity

The appeal to incredulity “fallacy” involves one asserting that something doesn’t happen because they don’t see how it could happen. As Shatz explains in an article describing how to spot and avoid the fallacy:

The argument from incredulity is a logical fallacy that occurs when someone concludes that since they can’t believe something is true, then it must be false, and vice versa. For example, someone using the argument from incredulity might claim that since they don’t see how a certain scientific theory could be true, then it must be false.

…

Arguments from incredulity generally have one of two basic forms:

“I can’t imagine how X could be true; therefore, X must be false.”

“I can’t imagine how X could be false; therefore, X must be true.”

Shatz gives an example of this supposedly fallacious inference:

“There is just no way that the concept of evolution is right; it just doesn’t make any sense to me. Creationism is a much better explanation of how we came to be.”

Obviously if stated deductively—I can’t imagine how X happened therefore X must not have happened—the argument is fallacious. But this is true of all forms of probabilistic reasoning. When stated probabilistically—that because something makes no sense it probably didn’t happen—this isn’t fallacious reasoning. Imagine:

John (speaking to his three year old son): who ate the chocolate cake?

Little Timmy (his son): I’m sure it was the dog.

John: Then why do you have chocolate smeared all over your face?

Timmy: The dog eating it caused it to be on my face.

John: What? That makes no sense. Why would the dog eating it get chocolate on your face?

Timmy: That’s just how things work.

John: That makes no sense.

Timmy: You’re committing the appeal to personal incredulity fallacy. Just because something doesn’t make sense to you doesn’t mean it didn’t happen. For many years, a naturalistic explanation of lightning didn’t make sense to people to they inferrred Zeus did it. You’re just like that—just because the view doens’t make sense to you doesn’t mean it’s true.

Clearly, Timmy is reasoning incorrectly here. The fact that something makes no sense, in fact, makes it less likely. The mistake creationists make is that evolution does make sense. There’s a perfectly clear and consistent story of why evolution works. When proving that evolution makes sense, one can point to other data, like the fact that lots of scientists find it coherent, and can ask what about it makes no sense.

Obviously, the fact that something seems incoherent doesn’t automatically mean it is. But it reduces the plausibility of a view if it seems incoherent.

3 God of the gaps

Every time a theist uses God to explain something, atheists invariably claim that they’re committing the God of the gaps fallacy. The claim is that it doesn’t follow from the fact that something is unknown that God did it. Rational Wiki explains the argument form:

P1: Natural phenomenon X seems mysterious.

P2: God works in mysterious ways.

C: Therefore, God causes X!

Has anyone ever used this reasoning? X is mysterious so it must have been caused by God for he too is mysterious? Generally, people will point to events in the natural world and argue that there are not good naturalistic explanations while there are good theistic explanations—e.g. for the fine-tuning of the universe. Here is how the god of the gaps accusation usually plays out:

Person 1: God naturally explains fine-tuning which is otherwise very improbable.

Person 2: You’re just using God to plug in a gap that you don’t know, just like those who appeal to Zeus to explain things.

Here’s a little secret: every time anyone uses some theory to explain any event, they’re making a theory of the gaps argument. There are gaps in our understanding—we don’t know how event X happened—and they argue that some theory explains it. For instance, the evidence for evolution includes a universal genetic code, features like an ape’s thumb, goosebumps, and more. One could make the same accusation against this inference: this is just an evolution of the gaps. You’re saying “we don’t know why apes have shoddily designed thumbs, people have goosebumps, and there’s a universal genetic code therefore it must be evolution. That’s fallacious, like saying Zeus makes lightning.”

Or suppose that you see a dog looking guilty and a pool of pee on the floor. You assume the dog peed. “Fallacious, dog of the gaps reasoning,” cries the proponent of this fallacy. “You’re saying you don’t know why there’s a pool of pee therefore it must be dog. Just like Zeus with lightning.”

This is clearly stupid. If there’s some theory that is pretty inherently likely and explains a lot of things, you should believe it. If a view naturally fills a gap, that’s evidence for it, because it makes those events more likely. The dog peeing hypothesis is a good one, because dogs pee a lot, so it’s pretty intrinsically likely, and it explains the pool of pee. It wouldn’t be rational to infer that, for instance, Zeus caused lightning, because it’s not intrinsically likely that a being like Zeus would exist and cause lightning—there are other better theories.

Now, when people are defending the God of the gaps charge, they’ll point to the history of more things being explained by science, and thus the role of God shrinking. But this point doesn’t merit the charge that using God to fill gaps is fallacious—it just means that we should downplay the evidential force of the gaps because it’s likely, inductively, that there will be a naturalistic explanation of the events.

In addition, while it’s true that science has cleared away some mysteries, it’s made even more. The best arguments for God—from anthropics, fine-tuning, and psychophysical harmony—are things we only know about with modern science and philosophy. Thus, the claim that God’s role gets smaller as we learn more is wrong.

Furthermore, for God to be a good explanation, there must be a reason to think that God explains the thing. This means God isn’t a good explanation of, for instance, lightning. Why would God make lightning? But God is a good explanation of these events. Thus, I think the inference is backward—learning more about science and philosophy has allowed us to uncover more things well-explained by theism. The only argument for God that’s been eliminated has been the argument from biological life, but that’s been supplanted by the much stronger fine-tuning argument.

There are other beliefs that we’re justified in having, even though over time the case for them has weakened. A friend of mine has given the example of belief in the law of non-contradiction. Before discovering the liar paradox, the law of non-contradiction was super obvious—over time, it’s gotten less obvious, with smart philosophers denying it. Yet that doesn’t mean that we should doubt the existence of the law.

The God of the gaps charge is just a fully general objection to using God to explain anything. If applied consistently, it would rule out all inference to the best explanation in any domain and make belief in every scientific theory irrational. It’s thus a terrible argument, only convincing because it gives an easy way out of all theistic arguments.