Michael Huemer really knows how to goad a reader into responding to his work. In Knowledge, Reality, and Value, after giving an argument that belief in determinism is self-refuting, he says:
In my experience, almost all philosophers who hear this argument hate it. They immediately feel as if it’s just some sort of cute trick, and therefore they assume there’s something wrong with it even if they can’t say what it is, and they aren’t very interested in examining it. (p. 193)
- About the free will issue, we should believe only what is most likely to be true. (Premise – presupposed by reasoning)
- In general, if S cannot do A, then it is not the case that S should do A. (Premise)
- Therefore, with respect to the free will issue, we can believe only what is most likely to be true. (From 1, 2)
- If determinism is true, then there is only ever one thing that a person can do; we never have alternative possibilities. (Def. of determinism)
- Therefore, if determinism is true, then we believe only what is most likely to be true. (From 3, 4)
- I believe in free will. (Premise based on introspection)
- Therefore, if determinism is true, then the belief in free will is what is most likely to be true. (From 5, 6) (p. 193)
I found step 3 confusing at first because of the word “only”. If you interpret it as
The only option available to us is: believe whatever is most likely to be true
…then step 3 doesn’t follow from 1 and 2. I assume it should be interpreted like this instead:
One of the options available to us is: believe whatever is most likely to be true and don’t simultaneously believe any less-likely-to-be-true thing
…which does indeed follow from 1 and 2.
Premise (1) is much less plausible than it sounds at first. One way to see this is to realize there’s an independent reason to think proposition (3) is false, and if it is, you can reverse the argument of steps (1)-(3) to yield an argument that (1) is false.
According to (3), “with respect to the free will issue, we can believe only what is most likely to be true.”
Huemer doesn’t say who “we” refers to—does he mean all rational beings, or all adult humans of at least average intelligence, or everyone reading this book, or what? In general it doesn’t seem safe to assume that all people have the same level of ability or obligation to “believe only what is most likely to be true” about all issues. But since (1) is supposed to be “presupposed by reasoning” I assume Huemer must at least mean everyone capable of reasoning about the question of free will.
Lots of smart people have been thinking hard about free will for a long time. Yet there is no consensus among them; they reach mutually incompatible conclusions. So at least some of them—many of them—are wrong. It’s implausible to suggest that they freely chose to be wrong. It’s more plausible that they were trying to “believe only what is most likely to be true”, but failed. So we can make the following argument:
~3. With respect to the free will issue, at least some of us
can’t believe only what is most likely to be true. (Premise -
inferred from observation)
2. In general, if S cannot do A, then it is not the case that S should
do A. (Premise)
~1. About the free will issue, for at least some of us, it is not the
case that we should believe only what is most likely to be true. (From
~3, 2)
Compared to Huemer’s steps (1)-(3), this is much more in line with how premise (2) is normally used in ordinary life. The idea that “should implies can” is usually used to defeat an initial intuition that we should do something, by pointing out that it’s not actually achievable. For example, consider this conversation:
John: Everyone should save money for
retirement.
Robin: Some people hardly have enough
money for food and housing.
John: Well, yeah. But everyone who
can should.
Think how weird it would be for the conversation to go like this instead:
John: Everyone should save money for
retirement.
Robin: Some people hardly have enough
money for food and housing.
John: No, I trust my intuition that
everyone should save money for retirement, so I’m compelled to
believe they actually have the ability to.
That’s basically the move Huemer is trying to pull in steps (1)-(3). Premise (1) sounds plausible mainly because it’s desirable to “believe only what is most likely to be true”, just like it’s desirable to save money for retirement. When something is desirable, and we aren’t aware of—or just aren’t currently thinking about—any reasons it’s unachievable, it’s natural to say we “should” do it. But this way of speaking doesn’t reflect any special direct knowledge that the thing is actually possible. When confronted with evidence that it’s not possible, we adjust our language to say something like “should try” or “should if possible” instead of “should”. That’s what we should (try to :P) do with Huemer’s premise (1), too:
About the free will issue, we should try to believe only what is most likely to be true.
Unfortunately, this more plausible revised premise would not support the rest of Huemer’s argument.
Huemer didn’t just present premise (1) as intuitively plausible; he claimed it was “presupposed by reasoning”.
Maybe some people do assume “I should believe only what is most likely to be true” whenever they engage in reasoning. So a conversation about their motivations might go like this:
M: Why are you analyzing arguments
about free will?
J: I believe that I should believe only
what is most likely to be true about it.
That’s reasonable. But for the self-refutation argument to work, this has to be the only available presupposition that would justify engaging in reasoning. It’s not. The following is also perfectly reasonable:
M: Why are you analyzing arguments
about free will?
J: I believe that I should
try to to believe only what is most likely to be true
about it.
So is this:
M: Why are you analyzing arguments
about free will?
J: I believe that I should believe only
what is most likely to be true about it if I can, and
I don’t know that I can’t.
It would not be reasonable to try something if you knew for sure that you couldn’t do it. But it’s reasonable and common to try to do things that, as far as you know, you might not be able to do. When you try to run an ultramarathon or climb a mountain, you may not know for sure that your body has what it takes. When you try to solve a math problem, you may not know whether you’re smart enough to find the answer, or even smart enough to avoid being fooled by subtly invalid proofs. It’s still reasonable to try.
The same applies to thinking about contentious philosophical issues like free will. As I argued above, many of us probably can’t get the right answer. It would really be quite arrogant for any of us to take for granted that we’re definitely one of the ones who can. There’s nothing self-contradictory about trying anyway in the hope that we are.
Posted 2025-07-12 by Jacob Williams on brokensandals.net. Feel free to argue with me by email: jacob@brokensandals.net.