On the Modal Analysis of Libertarian Freedom in Plantinga's Free Will Defense

Central to Plantinga's free will defense is his notion of significant freedom, which is, roughly, libertarian freedom with respect to moral actions. Plantinga attempts to capture a necessary condition of significant freedom in terms of the actions of a given agent at different possible worlds. Very roughly, an agent S is significantly free with respect to a moral action A only if there is a metaphysically possible world W at which S performs A and there is a distinct metaphysically possible world W' at which S does not perform A.

I worry that possible-worlds analyses of libertarian freedom -- like possible-worlds analyses of most things -- are too coarse-grained to capture the essence of the notion. So, for example, I'm inclined to think that an omnipotent and morally perfect being could be significantly free even if there is no metaphysically possible world at which such a being performs a wrong action. It seems to me that such a being would have the power or the ability to do wrong; it's just that such a being would never exercise that power, in virtue of their moral nature. To use a term common in the literature on dispositional properties, the ability to do wrong is a "masked" disposition in God.

Similarly, it seems to me that a morally virtuous created being could be significantly free even they never perform a wrong action in some metaphysically possible world. Like God, such a being would have the power or ability to do wrong; it's just that their moral nature is such that they would never exercise that power. As with God, then, the ability to do wrong in such creatures is a "masked" disposition.

But suppose someone is unsatisfied about all this without the prospect of a "worlds" analysis of freedom. Fine. Hyperintensional analyses are all the rage these days, and so one could appeal to impossible worlds and non-trivial counterpossible conditionals to account for the sense in which such beings have such an ability.

Is there a punchline? It will take some time to flesh out the point properly,  but intuitively, the worry is that if modal analyses of signifcant or libertarian freedom are inadequate, then this will cause problems for Plantinga's free will defense, and for his account of transworld depravity.


8 comments:

mpg said...

I have a problem with the idea that an Omni-being could have the power to do X, but there is no possible world in which an Omni-being actually does X, due to its moral nature. Surely to say that an Omni-being has to power to do X, implies that there is a possible world where it does X? If, in the language of possible worlds, it is impossible for the being to do X, in what meaningful sense does a being actually have the power (or potential) to do X. Seems contradictory to me.

exapologist said...

Hi mpg,

I'm taking significant/libertarian free will to be reducible to an ability, where the ability is a dispositional property. Furthermore, I think that there are dispositions whose manifestations are metaphysically impossible. Daniel Nolan and Carrie Jenkins develop and defend an account of such dispositions in their paper, "Disposition Impossible", Nous (forthcoming). An earlier draft of the paper is available here.

Dr. Rizz said...

Ex,

Your possible worlds analysis of libertarian freedom does not seem to get close enough to the essence of libertarian agency to ground a critique of a possible worlds analysis of that notion. LF involves being able to do X or refrain from doing X at a single world, right? Why is it even relevant to the question "did I freely lie to the judge?" that I did not lie in another possible world? Hence, Plantinga can (and does on my reading) say that an agent is trans-world depraved only if they always choose to do evil at least once in all worlds in which they are LF to do so. I don't see the difficulty yet.

exapologist said...

Hi Dr. Rizz,

Right. Plantinga wants to give a possible-worlds analysis of LF, and I want to say that, from an orthodox Christian theistic perspective, there's pressure to say that any such account is inadequate. For I take it that God is supposed to have LF, and yet (in virtue of his moral perfection) there is no metaphysically possible world at which God performs at least one morally wrong action. I'm also inclined to think that modal analyses of virtually anything are inherently inadequate (I think we need to bring in a hyperintensional semantics for essences and analyses).

Dr. Rizz said...

Does he want to give a possible world's analysis of freedom? I have never interpreted him thus in regards to this 3 major presentations of his free-will defense. I thought that he is just trying to show that there is nothing incoherent about the idea that all free creatures do at least one evil act in all of the possible worlds that God can weakly actualize. I take him to be shooting for the claim that this is epistemically possible.

Given his distinction between possibility and actualizability, he could argue that there is no actualizable world in which God commits evil whilst insisting that God is free to do evil in all worlds.

exapologist said...

Hi Dr. Rizz,

I thought that he is just trying to show that there is nothing incoherent about the idea that all free creatures do at least one evil act in all of the possible worlds that God can weakly actualize. I take him to be shooting for the claim that this is epistemically possible.

Right. As I tried to sketch in the previous post on this topic (sorry, I should've linked to that in the current post), I argued that it's not clear that he's right about that. That is, I argued that it's not clear that Plantinga's thesis of possible universal transworld depravity (PUTWD) is even epistemically possible. My worry is roughly the same one raised by Joshua Rasmussen ("On Creating Worlds Without Evil - Given Divine Counterfactual Knowledge", Religious Studies 2004), although Rasmussen does a much better job of developing and defending it.

Given his distinction between possibility and actualizability, he could argue that there is no actualizable world in which God commits evil whilst insisting that God is free to do evil in all worlds.

So are you suggesting that Plantinga could say that there are possible but unactualizable worlds at which God freely does wrong, and that this is compatible with God's moral perfection? That's interesting. However, I'd like to probe this proposal a bit more if you don't mind.

Now I take it that we want to say that it's true not just at the actual world that God (could but) wouldn't freely do wrong -- i.e., that God is merely contingently morally perfect (in the proposed sense) --, but rather that no actualizable world is such that God does, or would do, something morally wrong. But on those assumptions, there is no metaphysically possible world at which every essence suffers from transworld depravity (since God's essence doesn't suffer from it), i.e., PUTWD is false. Furthermore, once we allow that God's essence is immune to TWD, I think the burden is on the defender of Plantinga's FWD to say why the same can't be true of at least one creaturely essence (I recommend Rasmussen's paper for a case that really pushes this point). This is one sort of worry that pushes me to the sort of account of LF I proposed.

Dr. Rizz said...

Hi EA,

You state:

Now I take it that we want to say that it's true not just at the actual world that God (could but) wouldn't freely do wrong -- i.e., that God is merely contingently morally perfect (in the proposed sense) --, but rather that no actualizable world is such that God does, or would do, something morally wrong.

I actually don't think that God is free to choose evil. I just commented on your most recent post as to why I don't think this raises a problem for the value of freedom. I am sure that similar thoughts are in the literature.

Yet, if God is free to choose evil, it is hard for me to see how there couldn't be a possible world in which he makes such a choice. Hence, I suppose I prefer to hold that none of those worlds are actualizable even though this wreaks havoc with the claim that God is essentially morally perfect. Maybe, on this view God is not, because there is a potential to do evil within God. Yet, it would still be significant if that potential were not actualizable. This would leave some very hard questions on the table such as what grounds the fact that God would never commit evil even though he could? This is why I prefer to deny that God is libertarian free to do evil.

best,

Dr. Rizz

Dr. Rizz said...

Hi EA,

Just read your responses to MPG. I haven't read the disposition impossible paper, hence I take back my assumption that a real potentiality to do evil entails that there is a possible world in which that disposition is actualized.

Ricki Bliss's Cambridge Element on Grounding, Fundamentality, and Ultimate Explanations

 ...is now out , and available for free download for a limited time. Required reading.