Not Even Wrong

Following my deconstruction of Einstein’s “spooky action at a distance” a couple of weeks ago, I thought I’d have a go at another hackneyed quote from the world of physics. “Not even wrong” was coined by Wolfgang Pauli in the 1950s, and it’s widely used by scientists to describe theories that are fundamentally non-scientific. The phrase may sound like a contradiction in terms, but actually it has a precise and literal meaning. I still get annoyed when people use it, though. That’s not because it’s used incorrectly, but because the people who use it think it means “worse than wrong”, which it doesn’t. Even Pauli himself seems to have used the phrase as a pejorative (the correct equation was photoshopped onto the blackboard by me, as if you hadn’t guessed).

There is no such thing as “worse than wrong”. Right and wrong are extremes, and anything that is not right, and not wrong, must lie somewhere in the indeterminate region between the two extremes. A theory that is “not even wrong” is one that is permanently stuck in a state of indeterminacy. There is simply no way of telling whether it's right or wrong.

When the scientific method was first developed in the 16th and 17th centuries, it was understood that it was applicable only to a subset of human experience – namely to phenomena that take place in the physical world. The kind of aggressive, materialistic atheism associated with science today was virtually unknown in those days. Even the greatest scientists accepted that the physical world—and hence the applicability of the scientific method—was only part of the sum total of human experience. In those days the distinction was between “earth” and “heaven”, although today we are more likely to say “physics” and “metaphysics”.

The physical world is the realm of experiment and objective analysis. The metaphysical world is the realm of belief and subjective experience. Materialistic scientists, from Wolfgang Pauli to Richard Dawkins, are perfectly entitled to hold the belief that the metaphysical world is non-existent. That belief is as valid as any other. What they aren’t entitled to do—and yet they insist on doing—is to attempt to disprove metaphysical theories through the application of the scientific method. You simply can’t do that. You can’t apply a tool that was specifically designed for the physical world to something that is non-physical.

One of the fundamental principles of the scientific method is the idea of the testable hypothesis. This is a brilliant concept when it’s taken in its proper context. The defining characteristic of the physical world is that it’s amenable to repeatable experiments. For a hypothesis to be “scientific” it needs to make predictions that can be tested by experiment. If the results of the experiment agree with the predictions, then the hypothesis may be right. If they disagree, then it must be wrong.

By definition, any meaningful hypothesis about the physical world must be testable.

Equally by definition, hypotheses about metaphysics aren’t going to be testable. Metaphysics is all about belief and subjectivity. It encompasses everything within the scope of human experience that isn’t part of the physical world. So a metaphysical hypothesis can’t be proved right, and it can’t be proved wrong. It may be right, and it may be wrong – it’s a matter of individual belief. That’s what people are referring to when they say “not even wrong”.

What I object to is the fallacious assumption that a non-testable hypothesis, on a metaphysical subject, is wrong by definition. It isn't. It’s indeterminate – it can never be proved one way or the other. If someone chooses to believe in such a hypothesis, then you’re free to disagree with them. But you can’t disprove their theory, and it’s a mistake to think you can.

The error comes about from half-understanding something that everyone is taught in university science classes. You’re taught that a scientific theory, about the physical world, is badly formulated if it isn’t testable. That’s true. But it doesn’t mean that a non-scientific theory, about the metaphysical world, has to be testable. In fact the opposite is true, because if it was testable then it wouldn’t be metaphysics.

The fault doesn't lie entirely with the scientists. There's a growing tendency for metaphysical theories to be presented as if they were scientific, in a misguided attempt to give them more weight. It’s perfectly valid to use the non-testable, “not even wrong” criterion to demonstrate that such theories are non-scientific. But that's as far as you can go. You can't say “also the theory is wrong”... the most you can say is “also I don’t believe your theory”.