Are Refutations and Verification Really Symmetrical Within A Theory Comparison?

In my posts on the problems of refutation, I’ve claimed that within a theory-to-theory comparison refutations and verifications/support are symmetrical. I showed that Popper actually intended the asymmetry between refutation and verification to be a matter of logical analysis of universal statements vs existential statements; namely, that scientists care more about universal statements (which can only be refuted) than existential statements (which can only be verified) due to the greater power of universal statements.

I went on to say:

So within a theory-to-theory comparison, support and refutation are logically identical and are merely the inverse of each other. 

From Do Deutsch and Popper Disagree Over Refutation?

Further, I’ve been acting as if ‘support’ and ‘verification’ are synonyms and can be used interchangeably. But that isn’t quite right.

I’ve been (intentionally) using an example in which I knew that it was correct that verification and support more or less meant the same thing. In the case of Newtonian physics vs Einstein’s General Relativity, there is a clear cut case of the Eddington expedition experiment refuting Newtonian physics while simultaneously ‘supporting’ General Relativity (or, in other words, corroborating it) and thus ‘verifying’ that General Relativity is ‘more correct’ than Newtonian physics.

But is this always true? It is not.

Let’s take a different example: Light wave theory vs Light particle theory.

In the contest between those two rival theories, we managed to devise an experiment that corroborated light wave theory while refuting light particle theory. That is the famous experiment where they showed an interference pattern in light. Once it was shown that light has an interference pattern, just like other waves, it seemed that we had corroborated light wave theory and refuted light particle theory, so light particle theory started to disappear.

But then Einstein came up with the photoelectric effect which demonstrated that light comes in quanta and is thus like particles. This observation thus refuted light wave theory and corroborated light particle theory.

So we then had two experiments, each supporting the other theory and refuting its rival.

Now of course we all know how this turned out, really both theories were wrong and we instead came up with quantum mechanics that allows for light (and everything) to be simultaneously (in some sense) both a wave (in some sense) and a particle (in some sense), though not really at the same time.

Or to put this more bluntly, it was not the case that the lightwave interference pattern demonstrated that light particle theory was less true than lightwave theory. Nor was it the case that the photoelectric effect demonstrated that light particle theory was more true than lightwave theory.

I think it probably makes sense to say that each of those experiments ‘supports’ one of the two theories while causing a problem for the other. But I’m not sure it makes sense to say that each of those experiments ‘verifies’ one of the two theories while causing a problem for the other. I’m not even sure it makes sense to say that each of those experiments ‘verifies as more correct’ one of the two theories while causing a problem for the other. As such, I think the language of verification is just not identical to the language of support here.

This is a sense in which I think we can rightly say there is an asymmetry between refutation and verification. Refutations within a theory-to-theory comparison make it clear that one theory is problematic compared to the other but do not necessarily mean that we’ve verified one is more correct than the other. So perhaps the word ‘support’ makes more sense as the inverse of ‘refutation’ (within a theory-to-theory comparison I mean) rather than ‘verification.’

However, it is true that for that particular observation it was demonstrated that one theory was ‘more true’ than the other. But I can see why there might be some hesitance to call this ‘verification’ since you aren’t really ‘verifying’ that one theory is more true than the other. But you are ‘supporing’ the theory compared to the other.

Leave a Reply

Your email address will not be published. Required fields are marked *