Wednesday, August 3, 2011

Libertarian Free Will?

Putting together a few thoughts on free will which I've been toying with lately, I've come to consider two possible views of libertarian free will. One might be possible, I think, but the other doesn't seem to work. (I should mention at the outset that I'm not considering any version of metaphysical libertarianism that postulates supernatural entities.)

First, some general observations about free will and what it means to make a decision.

To be an alternative is to be represented as an alternative. To have an option is to have a representation of something as an option. Whether or not that representation corresponds to a physical possibility, or whether or not the choosing of that option is physically possible, is irrelevant. Viable options must be logically possible, not physically possible. All that matters for people to have options is for them to have processes of a particular sort which regard options as such.

If somebody snaps their finger to my left, and I turn my head to look, I have not necessarily made a choice. I only chose to turn my head if part of my behavior involved regarding turning my head as an option. This need not have been a conscious act. I don't see the harm in supposing that we can make unconscious choices. However, I think it is evident that we do make conscious choices. Sometimes turning our heads to look at something is a conscious choice, and sometimes it isn't. To say it is a conscious choice is only to postulate a certain amount of reflective awareness (or control) over the deliberating process.

When we make a choice, we utilize a certain sort of process which can be wholly deterministic. However, I'm not sure it has to be deterministic for our choices to be valuable to us.

Libertarian free will is, on some accounts, the ability to make choices which are in line with our beliefs and desires, but which are not determined by our causal histories. This sort of free will is commonly rejected on the grounds that, if the choice is not based on our causal histories, then it cannot be based on our beliefs and desires--it can't be our choice.

Yet, if the generation of options (which, remember, are options by virtue of the fact that they are represented as options in our decision-making processes) utilizes a purely random process (such as quantum physics might allow), then we might have a decision-making process that uses our beliefs/desires to choose between alternatives which are not fully determined by our causal histories. I wouldn't assume that we have any such random-option generators, but I see no reason to discount the possibility. So this sort of libertarian free will may be worth having.

Interestingly, even if we don't have this sort of free will now, we might be able to develop it. Imagine a computer which, given a particular problem, produced possible solutions by utilizing a purely random process. If we acted on an option generated by the computer, then we will have, in a very real sense, chosen an alternative that was not determined by our causal histories, and yet which was based on our beliefs and desires. This sort of libertarian free will is thus a theoretical possibility for our futures, even if it is not a fact about our present.

What must be clear, however, is that we don't need libertarian free will to make decisions. What makes a behavior a decision is the choosing between alternatives. It doesn't matter how the alternatives were generated.

The final point is that we don't need libertarian free will to have genuine responsibility. Freedom from causal history cannot make us more responsible for our actions. At least, not in a way that matters. When we judge people by their decisions, our judgment does not depend on their alternatives having been detached from their causal histories. The judgment is about what they chose to do, and the fact that they had other options to choose from. If I'm guilty of X-ing and not Y-ing or Z-ing, then it doesn't matter if I represented any of those options via some quantum indeterminacy. I'm guilty because I intentionally X-ed when I shouldn't have. Maybe I represented X, Y and Z to myself thanks to a quantum generator. Maybe I didn't. It doesn't matter.

Consider the paradox that would result if we took the opposite view, and said that we can only be responsible for choices if we were not determined to have them as choices. In that case, I could not be responsible for using quantum indeterminacy. But, then, how could I be responsible for the outcome of my use of quantum indeterminacy, if I wasn't responsible for the use of it in the first place? I certainly can't be responsible for the outcome of the randomizing process alone, since that was random. And yet I can't be held responsible for the utilization of the process. So there is nothing left for me to be responsible for.

We are responsible for our actions, not the processes which ultimately make them possible.

If libertarian free will is the idea that responsibility requires that we make decisions which are not determined by our causal histories, then I don't think it is a viable option. If, however, libertarian free will is just the idea that we can make choices in accordance with our beliefs and desires, and yet which were not determined by the past or the laws of nature, then I think it's a legitimate possibility. However, I don't see any legal or social matters riding on it. It's an interesting question, and it would be a great discovery about how our brains work, but it won't change how we think about responsibility.