Dawkins is wrong: you can't defend science with circular reasoning

By Aaron Sherman

Prof. Dawkins is a brilliant man, but he tends to take shortcuts in his reasoning when he thinks his audience isn't up for the harder answers. Such was the case recently, when he was asked why we should have faith in the scientific method. His answer was a common bit of circular reasoning that you hear quite often in answer to this question, but it's not any less wrong because it was Dawkins who was saying it.

So, here's the general form of his defense: we know that the scientific method works because we have so many wonderful advances that we've been able to make because of it. We have cars and drugs and exotic new materials and we can see into tiny spaces and gaze at the universe around us. Science just works! (Dawkins humorously punctuates this with "bitches" which I did chuckle at).


However, he's wrong, and it would be unfortunate if people, especially young people, didn't understand how he's using circular reasoning, here. So, let me explain:

We don't know that the scientific method works. The one and only true axiom of science is this: experimentalism combined with rigorous logic is a sufficient test for truth. That axiom must be true, or science is just so much coincidence (or the universe lying to us, see below). Experimentalism is, basically, the scientific method. It's the idea that you can perform a test some "reasonable" number of times and gain confidence that all future tests that use the same starting conditions will yield the same results.

So, when you say, "we know that the scientific method works because we have so much evidence," what you're really saying is, "we know that experimentalism works because we've performed so many experiments on it." That's the very definition of circular reasoning.

It's perfectly valid to say, "the scientific method is sufficient because it seems unlikely that the universe is a massive construct designed to fool us by giving us consistently wrong answers." It's perfectly valid to say, "we can rely on experimentalism because it's the best tool we've ever had, and we're willing to accept the risk that it will simply stop working (perhaps catastrophically) one day."

Dawkins would not be shocked to hear this, and he could have said as much, but I think he's wary of saying something that anti-science types could use as ammunition and, frankly, I don't think he thinks enough of his audience to go into the detail.

Ultimately, though there are two competing world-views: what we feel is sufficient to arrive at truth / what we observe is sufficient to arrive at truth. There's also a variety of middle-grounds between the two, of course, but most world views select one or the other to be dominant. Science selects the latter. So far, it's served us very well, but we should never forget that it's a choice based on assumptions which can never be proven.