Yes Virginia, there is free will.

I've been thinking lately about the following question:

    the Existence Question: does free will exist?

Of course, one's answer to the Existence Question will depend greatly on what one thinks free will is.  But there seem to be two different general strategies for trying to answer the Existence Question in the affirmative; these are what I'm calling for now the indirect and the direct approaches.

Direct approaches work as follows.  First, one specifies what free will is (e.g., free will is xyz) and then shows how that thing exists (e.g., "Hey look, there's xyz!").  This is how I think of, for example, John Fischer's view.  Free will is certain kind of control (guidance control), which we then find in actually existing people.  (I know that Fischer's semicompatibilism per se doesn't commit him to the existence of free will, just as it doesn't commit him to the truth of determinism; but it certainly seems to me that Fischer thinks there is guidance control.  Please correct me if I'm wrong.)

Indirent approaches are different.  After specifying what free will is, they'll argue that it is necessary for some further thing, and then show that that further thing exists.  Here are two examples.  Van Inwagen argues that free will is necessary for moral responsibility, the existence of which he thinks is evident.  Second, a number of scholars argue that free will is necessary for rational deliberation, a process which (though they're not always explicit about this) they think people atleast sometimes engage in.

Any thoughts on the distinction, or the relative merits of these two approaches?

(I'm also thinking of general strategies for answering the Existence Question in the negative, which may also come in direct and indirect varieties.  But I need to think about this more before posting something even as preliminary as this.)

Comments

[James Gibson]

I'm not clear about the distinction between direct and indirect approaches. For the direct approach, you say one first specifies what free will is and then "shows" how that thing exists. But what exactly does this "showing" come to? To understand my question, of the indirect approach, one also starts with specifying what free will is. Following this, one "argues" how this is necessary for some further thing. In your examples of the indirect approach, it is taken that this further thing is evidently had. So given that this further thing is granted by various parties involved in the Existence dispute, why doesn't the "arguing" in the indirect approach amount to "showing" free will exists? And if it does, then I don't see the difference between the approaches as you've identified them.

[James Gibson]

I think I just realized the difference, the indirect approach being that one shows not that free will exists, but some further thing for which free will is a necessary condition. One of the confusions I had was, I think, due to the fact that you say one shows the further thing exists, but the examples you use don't involve showing this. At least it's not clear that they are more like "showing" the further thing rather than "assuming" the further thing.

[Bob Doyle]

Hi Kevin,

One way to consider the Existence Question is to examine the Standard Argument against the Existence of Free Will.

Either determinism is true or indeterminism is true. These exhaust the logical possibilities according to J. J. C. ("Jack") Smart.

If determinism is true, we are not free. If indeterminism is true, our actions are random and our will lacks the control to be morally responsible. This is the standard argument against free will.
http://en.wikipedia.org/wiki/Standard_argument_against_free_will

How might we get around it? Traditional views have argued for a tertium quid, something that is neither chance (indeterminism) nor necessity (determinism). These are the agent-causal libertarians, some advocating a dualist non-physical substance, others that it is a mysterious gift of God, others that no cause at all is involved. We are simply free.

Or, since William James in 1884, several philosophers and scientists have argued for an artful combination of chance and necessity. Those describing a "two-stage model of free will include Henri Poincaré, Arthur Holly Compton, Karl Popper, Daniel Dennett, Henry Margenau, Robert Kane, Alfred Mele, myself, and most recently Martin Heisenberg, son of Werner Heisenberg. I have written individual web pages with critical analysis of the work of each of these thinkers on Information Philosopher

If we can regard a combination of indeterminism and determinism as the tertium quid, we can define a new "conservative libertarianism" view.

Event-causal or "causal indeterminist" libertarians (Kane, Ekstrom, Balaguer, van Inwagen) should be called "radical libertarians" because they insist that chance is involved directly in the decision, which makes no sense except for "liberty of indifference" situations.

Conservative libertarians limit chance to the generation of alternative possibilities for deliberation and evaluation. This is the "free" stage of free will.

At this moment I am the only conservative libertarian - since Al Mele, who calls the two-stage model "modest libertarianism," is agnostic and Dennett, who calls it "Valerian," is a compatibilist.

Since indeterminism is true (because of quantum indeterminacy), there is no strict causal determinism in the universe. So the second "will" stage in conservative libertarianism is an act of "determination."

I think this model, which provides all the determination of the will that compatibilist philosophers really want and need, may prove acceptable to many thinkers.

Just because our will is "determined" by our motives and reasons, our character and values, and our feelings and desires, that in no way implies or entails that it is predetermined by a causal chain going back to the beginning of the universe, as van Inwagen's Consequence Argument would have it. (Philippa Foot first pointed out this obvious fact in an article called "Free Will as Involving Determinism, Philosophical Review, vol LXVI, (1957), p.439)
http://www.informationphilosopher.com/solutions/philosophers/foot/

The famous article by R. E. Hobart in Mind 1934 with a similar title also advocated "determination," not determinism.

To sum up, in Conservative (or "Adequate" or "Valerian" or "Modest") Libertarianism the will itself is not "free" in the sense of random, because it is "adequately" determined in selecting an action from the alternative possibilities generated in part by chance.

But we are free from predetermination and forknowledge.

Our thoughts come to us freely.

Our actions come from us with determination.

[Carlos Patarroyo]

Hi Kevin:
I have to say that the “indirect” approach reminds me a lot of Kant’s “Transcendental Arguments” (think, for example, of Kant’s answer to Hume concerning the principle of causality, that shows that in order for something like objective succession to be possible, the category of causality must be present and acting. Otherwise objective succession just wouldn’t be possible). The problem is the status of that thing that you use as a point of departure (in Kant’s case, objective succession). Is it real? Is it an illusion? Just as Hume could answer to Kant that he has not proved that the causal power has real metaphysical existence (and of course, Kant wasn’t trying to), if one departs from saying that moral responsibility is only possible if free will exist, someone can question if moral responsibility is an illusion or not. If it were, what does it matter then that this illusion is only possible if some other illusion allows it? I think the same happens with rational deliberation.
My point is that in the indirect approach one has to be able to find -as a point of departure- something that cannot easily be labeled “illusory” o “mistaken” and, at least for me, nothing comes to mind that could have this strength. But, of course, I could be wrong about this last part, and about the whole thing.
Best regards…

[Tamler Sommers]

I think the indirect approach is fine. It just hasn't (in my view) been employed successfully yet. For the approach to work, we have to be convinced of two things:

1. FW is necessary for the further thing.

2. We have that further thing.

Van Inwagen's "free will is necessary for MR" strategy doesn't provide sufficient justification for (2). He gives virtually no defense for the belief that we are MR aside from an ad hominem anecdote about an MR skeptic who complains about book theft. (See Smilansky's fantastic Analysis paper ""Van Inwagen on the 'Obviousness' of Libertarian Moral Responsibility" for an argument along these lines.)

And the FW and deliberation strategy (am I wrong or is it only libertarians that attempt this move?) fails to provide sufficient justification for (1). ) If I were convinced that a belief in libertarian free will was required for rational deliberation, I might be pulled in the libertarian direction. My belief that I can deliberate is close to non-negotiable. But deterministic accounts of deliberation seem perfectly reasonable, to me anyway. (Pereboom's 2008 JOE paper "A Compatibilist Account of the Epistemic Conditions on Rational Deliberation" gives a compelling defense for this claim.)

But again, the approach itself seems unobjectionable.

[Fabio Fang]

Hi,
I disagree with Carlos (Hi!). As Tamler wrote, there are two steps in the indirect approach and the conclusion depends on both of them. If you proved that FW is necessary for a further thing, you still have to prove that the further thing exists in order to prove that FW exists. Carlos' problem seems to be that we can't prove that the further thing exists (whatever this turn out to be?). Of course, if the further thing can be prove to exist, then the argument fails; but that depends on both, what counts as "proved" and what is it that we say that needs FW, i.e., what is the further thing here.
Maybe Carlos' point can be rephrased in this way: the indirect approach is useless when the further thing is something like MR, because the existence of MR is (almost) as problematic as the existence of FW. Or: *If* the existence of MR is (almost) as problematic as the existence of FW, then to prove the existence of FW from the existence of MR is not good. I agree with this last thesis.

Leaving that aside, note that the indirect approach is not valid for any skeptic conclusion: even if FW is necessary for some further thing (say, MR) and that further thing doesn't exist, that does not imply that FW doesn't exist. Oxygen is necessary for this room to be on fire, but the lack of the fire doesn't imply the lack of oxygen.

One last thing: I think there are two other ways people have attempted to address the Existence Question:
- You can try to prove that FW exists by showing that it is (logically?) possible,
- And you can try to prove that FW does not exist by showing that it is (logically?) impossible.
I think that only the second approach is, by itself, a valid argument: if something is impossible, then that something doesn't exist; but if something is possible, it may not exists. The first approach works only as a response to the second one.

Thank you.

[Kevin Timpe]

Thanks for the comments so far; I’m sorry that it’s taken me a bit to respond, but I was out of town over the weekend and without internet access. (Gotta love the mountains!) A few follow-up comments…

James, van Inwagen does gesture at a proof for the existence of moral responsibility:

"There are, moreover, seemingly unanswerable arguments that, if they are correct, demonstrate that the existence of moral responsibility entails the existence of free will, and, therefore, if free will does not exist, moral responsibility does not exist either. It is, however, evident that moral responsibility does exist: if there were no such thing as moral responsibility nothing would be anyone’s fault, and it is evident that there are states of affairs to which one can point and say, correctly, to certain people: That’s your fault (“How to Think about the Problem of Free Will”)."

Now, you may think that he needs a further argument for this (and like Carlos and Tamler say, in order to think that this is a good strategy, one would need to be convinced of this ‘further thing’, here the existence of MR). But I think that he thinks the existence of this further thing (or this further, further thing) is obvious. But of course not all agree, so not all will be persuaded by his indirect proof.

Bob,
I hope to address general strategies for answering the Existence Question in the negative in a future post. One way will be ‘direct’, in showing that free will is such-and-such-a-thing, and then showing that that thing doesn’t exist. This seems to be the strategy of the argument that you mention. But merely pointing out that a direct argument for the non-existence of free will is a bad argument will not show that free will does exist.

Fabio,
I know no one who thinks that proving free will is possible is enough to prove that it exists. Of course, if one can prove that it is impossible, this will be one way of ‘directly’ proving that free will doesn’t exist.

[backoffscience.wordpress.com]

Hi there,
I've been trying to figure out where my thoughts about free will comes in the spectrum of isms, can you help?

It seems obvious to me that there is no neutral argument position from which to mount an attack on freedom. Whether we like it or not, we have to tell some story or other about our being free, and the effect of that story on our lives trumps all other arguments.

After all, whatever we think about free will, we still have to make decisions. And if free will and choice aren't connected, then what the hell are we talking about?

I'd also say that free will is not a question for science at all, but rather one about how we use language (which isn't something I think science can investigate successfully).

It seems to me that the prerequisites for free will are 1) being able to imagine different courses of action 2) being able to decide between them and 3) trying. Its essentially an question about our ability to do certain things with language, and whether we have the courage to act as we aim to.

But as that argument has to be seen in light of the first, I suppose its kind of ignorable.

Anyhow, I'm a bit of a newbie to the current debate, so any guidance on where I might fit would be appreciated.

[Tamler]

Kevin, how does PVI distinguish being moral responsible for something and something being our fault. If there's no distinction, then that wouldn't be a proof for MR, it would just be a tautology. But it sounds like moral responsibility is a necessary condition for faultness. Is the only distinction that we can MR for praiseworthy acts too? That would seem unsatisfying...

[Saul Smilansky]

Thanks, Tamler. I don't recall that this paper has ever been cited. It's my first published paper! Here it is, in case someone wants to read it:

http://philo.haifa.ac.il/faculty_pages/smilansky/van.pdf

[Kevin Timpe]

Thanks, Saul. I'd just requested it via interlibary loan. I shall remedy the oversight in a paper that I'm working on.

Tamler, I think you're going to be unsatisfied.

[Alan]

Ok please forgive me. I have always been proud that my first Analysis piece followed Saul's in that very issue and that it was also on PVI--though I always have had some regret about my note's somewhat haughty tone. I tried (and I think accomplished) in that piece to point out that it was false that PVI's earlier Analysis claim that the Ethics and Mind arguments' conclusions were inconsistent if one took them to be respectively about the separate components of what classic compatibilist FW means, and not an equivocation on FW as PVI's proof required. And like Saul--still up till now--I think no one has cited it. I am thus in good company in that minimalist sense, but proud to have been so happily juxtapositioned with Saul's first publication. So after all these years--hi Saul!

[Saul Smilansky]

Hi Alan. Proud to have shared a semi-detached with you in van Inwagen lane back in 1990. & Cool that we both had our first papers on PVI-on-free will, and in the same issue. One can sort of tell it's our first, in that we are really aggressive. And neither papers cited. Possible theories explaining that:

a. Our papers are bad.

b. Who's van Inwagen anyway?

c. Which papers get cited doesn't make too much sense.

I tend to favor theory "c". Worth discussing some time in more detail. But then how can we complain, when we haven't even cited ourselves...

[Alan]

Hah! I can't argue with you Saul. Still, I think my note couldn't compete with your more substantial PVI piece--it deserves citation. Right after we published our pieces I had the good fortune to meet Gary Watson (a wonderful kind man) and I sent him a copy of mine. I think recent bibliographical history would bear out my impression that I didn't exactly influence Watson with that note! Thanks for the kind reply.