There has been some talk recently about the hard heartedness of libertarians. But libertarians are not hard-hearted; quite the opposite. Indeed, one libertarian has recently (and, I suspect, without realizing it) offered us a nice gift: a defense of the view that free will is compatible with determinism. That libertarian is Mark Balaguer, and he offers us this gift in his ‘A Coherent, Naturalistic, and Plausible Formulation of Libertarian Free Will’ in a recent issue of Noûs.
Balaguer, like Kane, hangs his theory of libertarian free will on what he calls ‘torn decision’ cases. In these cases, an agent has roughly equal reason to perform each of two or more incompatible alternatives. Balaguer argues that if our choices in torn decision cases are undetermined, then we exert appropriate control over our choices, and we exercise L-freedom (libertarian free will). I think he’s wrong; I think we don’t exercise relevant control in these cases. But that’s not the question I want to focus on here. Instead, I want to show how his view entails that free will is compatible with determinism, at least with regard to some choices.
Balaguer, like Kane, thinks that many or even all of our choices are L-free, even when they are determined. Kane’s view is that if we are ultimately responsible for our characters, because we have formed them through indeterministic choices, then we are responsible for those decisions that are determined by our characters. Balaguer’s view is different. Instead, he holds that if an agent is capable of L-free choices in torn decision cases in a deterministic world, then that agent is L-free, and all the choices of an L-free agent are themselves L-free. L-freedom transfers from undetermined torn decision cases to cases in which the agent’s decision is determined by her reasons.
But if it is mere capability for undetermined choices in torn decision cases that makes an agent L-free, then there is no need for her to actually make such choices. If I am capable of such a choice (Balaguer calls a type-1 decision), then I am L-free, whether or not the world I live in is deterministic. So if I am capable of type-1 decisions, then all my type-2 decisions (where my reasons determine my choices) are L-free. Hence, libertarian freedom is possible in a deterministic world.
Of course, Balaguer does not intend to offer us a defence of compatibilism. I'm pretty sure that he means to say that agents who actually exercise libertarian freedom, by making type-1 decisions, have L-freedom in their type-2 decisions. So I’ve caught him in a slip, which he can easily avoid, by altering his definition of an L-free agent. However, there is a deeper problem here. Balaguer owes us an argument as to why we can’t exercise L-freedom in a deterministic world. Presumably, any such argument will turn on the claim that indeterminism is necessary for the existence of L-freedom: he might say, for instance, that it is not possible to transfer a property you don’t have, and that therefore in order for libertarian freedom to be transferred from type-1 decisions to type-2, agents must actually exercise libertarian freedom in type-1 cases. But this won’t work, simply because (by definition) type-2 decisions are determined, whatever kind of world we live in, so whatever the truth of determinism L-freedom is never transferred to type-2 decisions. I conclude that Balaguer has no principled reason to withhold the attribution of L-freedom to agents in a deterministic world. Hence, on his view, free will is compatible with determinism.
Recent Comments