Kirby asked:

Does knowledge that something exists prove its causal efficacy?

Does mere existence imply causal efficacy?

Answer by Geoffrey Klempner

Why would anyone ask these questions? Without a context, it is hard to see the point. But I’m guessing that this is about a version of mind-body dualism known as ‘epiphenomenalism’.

Epiphenomenalism is a theory about the nature of the inner, a theory for which the philosopher who holds the theory claims to offer proof. What is odd about this view, however, is that it implies that we know that something exists — non-physical thoughts, feelings, experiences, consciousness — which according to the theory have no causal effect on anything physical.

Physical states of the brain cause these non-physical mental states, but the mental states are themselves causally inert. They just exist. They do not possess ‘causal efficacy’.

This would be an very interesting discovery, if it were true, because it would imply that we can know of the existence of items of a certain kind, in the absence of any causal connection between those items and our knowledge.

There is another class of objects for which this claim has been made: numbers. A natural number, according to set theory, is a pure abstract object, made up of sets, which are themselves made up of sets, etc. etc., while the only set which is not made up of sets is the null set. We know that numbers exist, because we know that arithmetical statements such as 2+2=4 are true.

There has been much debate on this question in the Philosophy of Mathematics. (See Putnam and Benacerraf Readings in the Philosophy of Mathematics Cambridge University Press, for a classic survey.) It is difficult to see, however, how these considerations can help with the mind-body problem.

The key difference is in the argument for epiphenomenalism (epiphenomenal dualism) which is allegedly based on direct experience, not an inference as in the case of numbers, where we argue from the truth of mathematical statements to the existence of mathematical objects.

The argument goes like this. I can imagine a being who is my perfect physical double. No-one else can tell us apart. The only difference between me and my perfect physical double, however, is that my double has no mental states, only physical states. I know I have mental states because I can ‘see’ them directly, by looking inside myself. But for my zombie double, all is darkness inside.

The problem with this argument is that, by hypothesis, my zombie double is also writing this answer (on twin earth), utters the words, ‘for my zombie double, all is darkness inside’ with the same tone of sincerity as I utter those words, because he is, like me, an ‘epiphenomenalist’.

— There’s something not quite right about the argument I have just given, in effect, a reductio ad absurdum of epiphenomenal dualism, but what exactly? Is it because I am simply begging the question? Can it not be the case that when I say, ‘For my zombie double all is darkness inside,’ I am really saying something, uttering English words with a meaning, while when my double on twin earth utters the words, ‘For my zombie double all is darkness inside,’ those are just meaningless sounds which a (non-zombie) hearer mistakenly takes to have meaning?