Dissolving logical fatalism

Why the law of excluded middle is irrelevant to free will

Logical fatalists argue that if statements must be true or false, then we don’t have free will. They would say that if “I will have cereal for breakfast tomorrow” is true, then I can’t decide not to eat cereal tomorrow as that would be logically inconsistent.

While this would indeed be logically inconsistent, the conclusion is wrong. It is me having breakfast tomorrow that forces my statement about cereal to be true, not the statement being true that forces me have breakfast. This point might be clearer if I made a statement about the stock market in America instead — it’s going to do whatever it is going to do and my statement in no way influences or binds it.

We are now at the stage where we have one argument for and one against. But what we really want to do is to dissolve the question. According to (some versions of) free will “I will have cereal for breakfast tomorrow” or “I won’t have cereal breakfast tomorrow” don’t cover all the possibilities. There is also the situation “I might have cereal for breakfast tomorrow”. Here “might” doesn’t just mean that I haven’t made up my mind yet, but that it is fundamentally unknowable or indeterminate whether I will have cereal or not, even if someone had perfect knowledge of the past and present.

Framed this way, there is no conflict with classical logic’s assertion that each statement must be true or false. Here “I will have cereal for breakfast tomorrow” is interpreted as “It is determined that I will have cereal for breakfast tomorrow” and “I will not have cereal for breakfast tomorrow” as “It is determined that I will not have breakfast tomorrow”. So, if the actual situation is a “might”, then “I will have cereal for breakfast tomorrow” is false since the situation is undetermined, even if I do in fact end up having cereal tomorrow.

There might be some convenience in talking about it being “both true and false” for particular purposes, but classical logic already handles this case perfectly well. Arguing against logic fatalism does not require an extra truth value. It is much simpler to use three possibility descriptions instead — (determinate) “will”, (determinate) “won’t” and (indeterminate) “might”. There are some valid reasons for considering non-classical logics, but there is simply no need to do so here.