The Error in Russell’s Teapot

Posted on 26 November 2006


The philosopher Bertrand Russell has a famous argument called “Russell’s teapot” which is related to the maxim: “Absence of evidence is not evidence of absence,” i.e., the lack of evidence for something does not prove that it does not exist. For example, when the belief in God is attacked on the basis that there is no evidence for God, believers sometimes respond with this maxim.

Russell, however, argued that, in the absence of evidence, the burden of proof rests on the person asserting existence (e.g, of God) rather than the person denying existence (e.g., of God). Russell used the following example to argue his point: Consider the claim that there is a teapot in orbit around the sun, so far away that no telescope can detect it. It is clearly more reasonable, he argued, to suppose that such a teapot does not exist than to suppose that it does exist. Russell used this argument to conclude that, absent any evidence, it is more reasonable to believe God does not exist than to believe God does exist.

Russell’s argument is often cited as if it were an established truth that settles the issue once and for all. However, I think there are some serious problems with Russell’s argument. In fact, I think it is invalid, for at least two reasons:

First, teapots are not analogous to God: Absence of physical evidence for a physical existent is not comparable to absence of physical evidence for a metaphysical existent.

Second, another problem with Russell’s argument is that his choice of the teapot as an example biased his conclusion. By picking another example, we can arrive at the opposite conclusion: Consider the interior of the sun. Just like the teapot, we can not see the interior of the sun with our telescopes. Yet, we would normally consider it more reasonable to suppose that the interior of the sun exists than to suppose it does not. So, in this case, the burden of proof would rest with the person who would deny that the interior of the sun exists. Thus, Russell’s argument only appears true since he deliberately picked an example of something that is ridiculous to suppose would exist.

The upshot is that Russell’s argument can not be relied upon as a general justification for asserting that, absent evidence for X, it is more reasonable to deny existence of X rather than asserting the existence of X. But this doesn’t imply the opposite, that it is more reasonable to assert the existence of X. All else being equal, absent any evidence, both existence and non-existence stand on the same footing. Based on logic alone, we don’t know enough to assert or deny the existence of anything.

The reasonableness of a belief in the existence or non-existence of X depends on the worldview we take for granted and how X fits into that worldview. Thus, the existence of the center of the sun is reasonable within a scientific worldview since it makes it more coherent and consistent, while the existence of a teapot in orbit does not. This observation makes it clear that what is reasonable is dependent upon the worldview that is taken for granted. As the worldview changes, so will the judgements about what is and is not reasonable to suppose exists without evidence. For example, in the traditional chinese worldview, it is perfectly reasonable to believe in the existence of chi, but in a materialist worldview it is not so reasonable. Neither the existence or non-existence of chi is reasonable independent of a worldview. We have to first make assumptions about our worldview to make such claims of reasonableness.

So, the existence of certain facts about the world is not so black-and-white as it might appear. Or, to put it another way, if they do appear black-and-white to us, then that’s a sure sign that there are some hidden assumptions at work in us that we are not seeing. In one worldview, it may not be reasonable to believe in a certain type of God. But that does not mean that there is not another worldview in which it is reasonable. Reason, after all, depends completely on the assumptions we start with, and reason alone can not choose those assumptions for us. Our discomfort with this fundamental existential ambiguity is one factor that I suspect drives us to cling so much to our way of seeing things, and fight against others who believe otherwise.

Posted in: Philosophy