Free Will and the Problem of Evil

I don't really have much to add to Scott's excellent analysis of the problem of evil. I think he gets the explanation of the problem exactly right and does a very nice job of exploring the various bullets that a theist might choose to bite. I think, however, that Scott may be giving the much-vaunted free will defense more credit than it rightly deserves.

The heart of the problem of evil, as Scott nicely illustrates, is that it seems to involve a set of mutually inconsistent claims. But theists face a similar problem when it comes to the topic of free will. Consider the following:

  1. God is omniscient.
  2. God is omnipotent.
  3. God is infallible.
  4. God created the universe.
  5. Humans have free will.

You can see the problem, I'm sure. If God really does know everything, then God is already aware of all the things that I will do. And if God created the universe, then She created a world already knowing that I would do whatever it is that I'm going to do. So we have an obvious problem here: if God already knows what I'm going to do and if God cannot possibly be wrong, then it looks like maybe my will isn't as free as I'd initially thought. I pretty much have to do whatever it is that God knew I would do when She created the universe.

Of course, there are ways around this. The most obvious one would be to deny that "free will" really just means "could have done otherwise." Harry Frankfurt offers what seems to be a pretty good knock-down of the freedom-means-the-ability-to-have-done-otherwise thesis:

Allison is contemplating whether to walk her dog or not. Unbeknown to Allison, her father, Lloyd, wants to insure that that she does decide to walk the dog. He has therefore implanted a computer chip in her head such that if she is about to decide not to walk the dog, the chip will activate and coerce her into deciding to take the dog for a walk. Given the presence of the chip, Allison is unable not to decide to walk her dog, and she lacks the ability to do otherwise. However, Allison does decide to walk the dog on her own.

In other words, even though Allison absolutely must walk the dog, there is a very real sense in which we can say that Allison freely chose to walk the dog in this instance. We can look at the chip and see that it didn't activate. Even if she couldn't have done otherwise, she could have chosen otherwise. So maybe it's possible after all to have free will even in if we couldn't possibly do otherwise. And thus God and free will turn out to be compatible after all. Score one for the theists!

Except, maybe not so much.

See, if we're going to make any sense at all of the idea of free will, then it will have to be the case that, regardless of what I end up actually doing, I will have to have had the ability to have chosen otherwise. So, for instance, when I made a snotty remark to my spouse last night (not the worst form of evil, but still a bad thing, all else being equal), it must be the case that I could have freely chosen not to make that remark. That is to say, it is logically possible that I might have refrained from making the remark.

But that, you see, is something of a problem for our theist. See, God, being omniscient, can envision every single logically possible world. That means that, prior to creating the universe, God already knew about possible world A (this one) and possible world B (the one that is exactly like this one, only I freely chose not to make the snotty remark to my spouse). And God, being omnipotent, had to choose to actualize one of those two possible worlds. Yet God chose to actualize A -- the one in which I freely chose to make a snotty remark to my wife -- rather than B.

The problem, in a nutshell, is that the moment we make free will perfectly compatible with God both knowing what I will do and creating the world in which I do it, the free will defense stops being a defense against the problem of evil. God could just as easily have actualized the possible world in which I (and everyone else) freely chose good all the time. That's because God's knowing that I would choose X and creating the world in which I choose X is not, ex hypothesi, inconsistent with my still having the ability to freely choose X.

For that matter, there's an even simpler response to the free will defense. God could have let me freely choose all I want but still made it the case that my poor choices never ended up causing evil to anyone other than myself. That gets us free will, real responsibility and far less evil in the world. But we don't have that, which leads me to think that the free will defense won't allow the theist to consistently hang on to God's omnipotence, omniscience and omnibenevolence.

Share this

The theist may claim there

The theist may claim there is a logical impossibility involved in the existence of free will and omniscience, in much the same fashion as Lewis claimed a logical impossibility between free will and omnipotence.*

That is to say, if we grant that having one's powers circumscribed to what is logically possible is no bar to omnipotence, perhaps having one's knowledge circumscribed to what is logically possible is no bar to omniscience. If this is true, then there would be a logical impossibility to both 1. granting free will and 2. knowing the results of the choices that result from it. Indeed, you make just that argument.

Perhaps then, God, though omniscient, cannot know the results of our free choices, as such knowledge involves a logical impossibility. His omniscience extends to everything else.

I haven't heard this argument made before, but it seems a reasonable extension of the free will defense. It does seem to need some sort of libertarian free will to succeed, but such a doctrine is not unreasonable.

As to the final paragraph, I imagine the theist could argue, plausibly, that people need to interact, that such interaction is our best chance for happiness, or holiness, et al. Denying that interaction would thus violate the omnibenevolence of God, and, perhaps, the only way to allow such interaction is to allow their choices to affect one another. Again, this seems reasonable.

Thank you for the response.

*I am still unsure Lewis is the source of this. I have heard it only secondhand.

Good points

You're right that exempting the results of free choice from the realm of things that God could know in advance eliminates much of the problem. And I think you're exactly right to head in this direction: omnipotence doesn't mean "can do anything" nor does omniscience mean "knows everything." God, for instance, probably doesn't know how to ride a bike, what with the lacking of a physical body and all.

That said, making a move like this creates a brand-new set of tensions. Traditionally, God is also taken to be immutable. There's a fair amount of textual evidence for that claim in both the Hebrew and the Christian bible.

But for God to be immutable, it seems to me, God could not gain new knowledge. And yet, if God doesn't know in advance the results of my actions, then She must constantly be learning new things. It's hard to see how that's particularly consistent with being immutable.

Lack of immutability raises some worries about God's perfection and about the necessity of Her attributes. It probably also undermines any hope of making some version of an ontological argument work. Still, it might be the least-bad option for a theist to give up on immutability if it means salvaging the free will defense and thus resolving the problem of evil. It's not very attractive, though, to have to sacrifice your only decent argument for some proposition in order to refute an argument that your proposition is incoherent.

I'm pretty sure Christian

I'm pretty sure Christian believe Jesus could ride a bike.

On free will and foreknowledge

I kinda like this solution to the free will/omniscience dilemma, but it does imply that God lacks a good deal of FOREKNOWLEDGE. I sense many theists would find that troubling.

“What’s going to happen to us?”

“God knows.”

“Uh, I hate to break it to you, but I was just reading Distributed Republic and ....”

Whether or not you postulate the existence of a God with foreknowledge, it’s hard to build a model providing a mechanism for free will. And yet, I regard the operation of free will to be one of those rare conclusion I can embrace by reasoning backwards from the conclusion to the premise, reductio ad absurdum.

Basically, I cannot avoid believing in free will because, if free will does not operate, I cannot believe anything. Because I cannot do anything. Absent free will, I have no agency. I do not do, except in the sense that a boulder rolling down a hill does. Now, I can’t deny the possibility that free will is an illusion, just as I can’t deny the possibility that all my sensory perceptions are illusions and all my thoughts are misguided. But these lines of thinking seem to lead to a dead end. They don’t lend themselves to further analysis, I cannot conceive of any advantage in believing them, and I cannot conceive of any harm in denying them.

Good comments

Basically, I cannot avoid believing in free will because, if free will does not operate, I cannot believe anything. Because I cannot do anything. Absent free will, I have no agency. I do not do, except in the sense that a boulder rolling down a hill does. Now, I can’t deny the possibility that free will is an illusion, just as I can’t deny the possibility that all my sensory perceptions are illusions and all my thoughts are misguided.

This strikes me as pretty plausible. It's actually a bit reminiscent of Kant's argument for free will.

What relevance does it have for the "problem of evil" if free will has no relationship to doing? As C.S. Lewis remarked in Screwtape Letters XXIX, "the act of cowardice is all that matters; the emotion of fear is, in itself, no sin...." So if no one is is ever harmed by "evil" -- defined as a willful choice to do wrong -- because that choice has no relationship to the wrongful conduct, is it still evil?

I don't entirely disagree with you here. It seems to me that there has to be a pretty strong connection between free will and actual consequences to get the free will defense of the ground. Of course, the argument here isn't supposed to be that there isn't a link between the two; it's just that free will doesn't mean the ability to have done otherwise. It has to mean something else, something like my making a choice. The resulting freedom would end up being something like Kant's understanding of autonomy as the ability to give laws to myself.

Truth ... without consequences?

Even if she couldn't have done otherwise, she could have chosen otherwise. So maybe it's possible after all to have free will even in if we couldn't possibly do otherwise. And thus God and free will turn out to be compatible after all. Score one for the theists!

Of course, this thread is entitled "Free Will and the Problem of Evil." What relevance does it have for the "problem of evil" if free will has no relationship to doing? As C.S. Lewis remarked in Screwtape Letters XXIX, "the act of cowardice is all that matters; the emotion of fear is, in itself, no sin...." So if no one is is ever harmed by "evil" -- defined as a willful choice to do wrong -- because that choice has no relationship to the wrongful conduct, is it still evil? And to the extent that I feel guilt for my choosing to engage in harmful conduct, wouldn't that guilt be entirely unwarranted because the same harm would result regardless of my choices?

Incompatibilist

So we have an obvious problem here: if God already knows what I'm going to do and if God cannot possibly be wrong, then it looks like maybe my will isn't as free as I'd initially thought. I pretty much have to do whatever it is that God knew I would do when She created the universe.

Among other things, that seems to presuppose an incompatibilist position on free will. In fact a lot of the commentary here seems to presuppose incompatibilism.

I'm a compatibilist, have been since I read Dennett's Elbow Room (and probably before, but that book brought the issue to my attention). As a compatibilist, I accept that it is in principle possible for a being such as Laplace's demon to know in advance what a being will freely decide.

Harry Frankfurt offers what seems to be a pretty good knock-down of the freedom-means-the-ability-to-have-done-otherwise thesis

I'm not so sure about that. The problem is here:

if she is about to decide not to walk the dog, the chip will activate and coerce her into deciding to take the dog for a walk

For the chip to work it is necessary that she be free to be about to decide not to walk the dog. Now, you may think, "but that refers to a future event which is prevented by the chip." Well, the chip isn't a time traveler! (If it is we can discuss that but let us suppose that it is not.) The chip therefore is reading the current state of her brain, and therefore there is such a state as the state that Frankfurt has labeled being about to decide not to walk the dog. But what is it about that state of her brain that makes it the state of being about to decide, and not the state of already having made the decision? If you look at a brain operate, what you see is a sequence of states followed by an action. Which specific state is the state of deciding to walk the dog? It seems entirely natural and reasonable to hold that the brain-states corresponding to deciding and having decided to walk the dog are none other than the states at which it becomes a done deal that the dog will be walked. And this will include the state that Frankfurt has sought to label being "about to decide".

Thus, properly understood, she is free to decide and has the ability to decide either way, at which point, depending on her decision, the chip may reverse her decision.

edit:

Even if she couldn't have done otherwise, she could have chosen otherwise.

Ah. So you seem to get that, even though Frankfurt's presentation denies that she could have chosen otherwise (he allows only that she could have been about to choose otherwise). But I would like to point out here that, evidently, Frankfurt thinks the issue of whether she could have chosen otherwise is the key - hence his presentation. You think that the key is whether she could have done otherwise. That, however, is not Frankfurt's focus, nor is it mine. So Frankfurt and I seem to agree that the question of whether she could have chosen is key.

I'm not sure we disagree

Perhaps I wasn't as clear as I meant to be in the post, but my intention was to say something similar to what you seem to be saying here. I started out with an incompatibilist view because, it seems to me, it's the most natural starting position. That is, IMO, our untutored beliefs about free will tend to be that free will is the ability to have done otherwise. That sort of position, however, is pretty clearly not one that a theist can hold.

As for Frankfurt, you might be right about his thought experiment. I think that we could probably rework the experiment in order to get around the problem (perhaps by specifying that we've put the chip somewhere in the middle of the process, such that as Allison is picking up the remote to switch on the television, the chip kicks in and has her put it back down). I'd need to think more about whether such a move would work, so I'm not really committed to rescuing Frankfurt's actual analogy. Nor do I think it necessarily matters. All that I was trying to do is to set up some support for the idea that there are compatibilist ways of thinking about free will.

My larger point is that I don't think that a compatibilist understanding of free will is sufficient to rescue the theist from the problem of evil. If my choice is severed from the consequences of my action in some important way, then there's a real sense in which -- metaphysically, anyway -- I'm not responsible for the actual bad stuff that happened. I'm certainly responsible for having chosen evil, and I think that there is still plenty of room for moral agency and personal culpability within compatibilism. But the actual bad stuff that happens, on a compatibilist account, isn't something that I have directly caused.

These sorts of discussions sometimes distinguish between moral evil (which contains a whole spectrum of evils ranging from Nazis to my snotty remark to my spouse) and natural evil, which is basically stuff like floods that kill thousands of people, diseases that kill and maim, etc. The free will defense is usually thought of as a way of solving the problem of moral evil. But compatibilism, it seems to me, simply moves whole classes of moral evil over to the realm of natural evil.

I read Dennett's Elbow

I read Dennett's Elbow Room

If you liked Elbow Room, then you may like this.

Free Will and God's Omniscience

God is omniscient. He knows what you will do before you do it. This does not stop you from making a free choice, since *you* do not know what you will choose before you choose it.

Problem of evil

For further reflection:

http://thinkpoint.wordpress.com/category/evil-in-the-world/

SC