Tuesday, April 10, 2012

The Free Will Solution

The Free Will Solution (FWS) is extremely popular among contemporary theists.  Many believe it has permanently solved the problem of evil.  It is based on an insight that can be helpful.  This insight is that evil is permitted for a higher good.  The FWS says this higher good is the actuality of "significantly free acts."  If the solution has failed to correctly identify the higher good, then it fails as a solution. 

The FWS says: natural evil is due to moral evil; 2. Moral evil is due to free will; 3.  Free will is necessary for human dignity.

Some historical context:  J.L. Mackie’s 1955 article on the problem of evil argued that the problem is a logical problem which proves that God does not exist.  This view was generally taken, at the time, as the correct conclusion.

Alvin Plantinga responded in books like "God, Freedom and Evil" (1977) by arguing that there is not a logical problem because for all we know there is a greater good served by evil.  That greater good is the existence of significantly free choices.  By permitting this kind of freedom God leaves open the possibility of evil.  According to this view, this kind of freedom is necessary for real relationships.  Many philosophers of religion continue to believe that Plantinga has demonstrated that there is not a logical problem (there remains the evidential problem).  However, this solution is only as good as the identification of the “higher good”.  It is true that if evil serves a higher good then there is no problem, but has this good been correctly identified.  There are important reasons to think it has not, including that this kind of freedom is not possible or necessary. 

Before we go there, let me introduce Molinism, which is a solution to both the problem of evil and divine sovereignty in salvation.  This solution says that God looks at the possibilities and actualizes the one that most fits his good purpose.  For Molinists like Plantinga and William Lane Craig, this happens to be the one that actualizes free choices.  In this case, to preserve open possibilities, God only actualizes the general framework of the world while leaving free choices open.  There are some aspects of the future that God doesn’t know until they happen.  Some versions of Molinism suggest that God knows even the free choices people will make (if put in this situation, P will choose X), but this threatens the idea of freedom as the possibility to do otherwise.  It also reduces God’s knowledge to fore-seeing, rather than creative determination.  If God is merely the fore-seer,  he is not the sovereign determiner.  

As noted in the previous post "The Dilemma for Libertarianism", this view of freedom requires uncaused events.  Uncaused events undermine moral responsibility, they do not serve as a basis for choice, if granted as possible then they cannot be consistently denied elsewhere, and they are not logically possible (involving a confusion of non-being and being). 

The alternative is to define freedom as doing what one wants.  This is compatible with predetermination, and indeed freedom requires predetermination to uphold the causal relationship between one's thoughts, desires and choices.  

The Molinist solution focuses on “being saved,” on who will make a choice for God.  This is called soteriology.  In some versions, God sees who would choose him and actualizes that world, picking people based on the merit of their choice.  In other versions, God actualizes conditions under which people choose him, thus picking people before their “merit” of choosing God.  The first case still involves uncaused events as well as a limited view of God's knowledge--it is either explicitly open theism or consistently held becomes open theism..  The second case could simply be predetermination or it could be a scenario where God’s predetermination is reduced to “foreseeing.”  

"Being saved" cannot be the higher good that explains evil.  Being saved means being saved from evil to do good.  It still remains to be explained just what is the highest good.


  1. Owen,

    Regarding Plantinga's FWS you wrote: "However, this solution is only as good as the identification of the “higher good".

    It seems to me you were referring here to the FWS with respect to solving the logical problem. My take is that very little is in fact, required to alleviate the pressure of the logical problem of evil. This is because the logical problem presupposes that God does not have morally sufficient reasons for allowing natural evil, but I don't think such a thing can be proved even in principle. Provided that the idea that "God has morally sufficient reasons for allowing evil (whatever these reasons might be)" is not itself incoherent, I don't see why Plantinga or any other theist/theist sympathizer would need anything further to level the logical problem of evil, but perhaps I'm missing something?

    Of course, whether such a non specific, bare bones response solves the evidential problem is another matter and your point would be well taken there.


  2. I think you're right in your assessment, but that in the 20th century analytic tradition it was Plantinga who wrote such an article first. However, everything Plantinga says is anticipated in Hume's "Dialogues," so that nothing new has been said and Hume has not been answered. A student of mine just presented a paper about this at the Pacific APA which we wrote together.

  3. Owen,

    I'm not sure I'm following your comments regarding Hume's challenges. Perhaps you mean that an appeal to ignorance is not enough? If so, again I'd ask insufficient for what, the logical or evidential problem?

    If the logical problem fails to be a problem on grounds that it presupposes that God does not have morally sufficient reasons for allowing evil (whatever they may be) then I'm not sure what of Hume's account have been left unanswered.


  4. This comment has been removed by the author.

    1. I'm thinking Hume sufficiently defends the problem of evil from charges that there is no logical problem, however I don't know if I want to spoil the suspense of having to read our paper :).

      I guess the direction I'd go is the same as Hume in that claims such as "God's goodness is unknowable" or "incomprehensible" or "for all we know . . . " end all religion (prayer and praise). If we live this out consistently we have a serious problem about God's existence and why we should believe in theism. Its like saying "there is no problem because one of the terms in the problem is empty/meaningless." Does that really relieve the problem if denial of the term threatens theism itself.

      In other words, behind the "for all we know" answer is that the highest good is not currently knowable. If that's true, that's a problem that precedes the problem of evil and is a serious problem for Christian theism.

  5. I guess I'll have to read the paper now since you've piqued my interest. One point of clarification, you wrote: "In other words, behind the "for all we know" answer is that the highest good is not currently knowable. If that's true, that's a problem that precedes the problem of evil and is a serious problem for Christian theism."

    Did you mean to say "not currently knowable" or something stronger like "theoretically not knowable"? It seems to me anyway that this proposed new problem (if thought of as a logical problem) would only go through if it is the latter, which again would be one of those tricky universal negatives needing to be proved. Perhaps the theist need only respond to your new problem thusly: "look you can't show that any morally sufficient reasons for God's permitting natural evil are principally unknowable on the basis of my (or any other person for that matter) not knowing what they are."


  6. Hume discussed this in his dialogue as well by saying "for all we know there is an easy solution." I think there is an easy solution that Hume overlooked due to his empiricism and academic skepticism.

    The problem isn't whether there is a "good moral reason," the problem is knowing what is "the good." If the theist doesn't know the good then that's a prior problem that needs to be addressed before we can move to the problem of evil.

  7. It's not so obvious to me that "if the theist doesn't know the good then that's a prior problem that needs to be addressed before we can move to the problem of evil." It may certainly be a philosophical problem to consider and an important one at that since it would seem to have implications in ethics and the philosophy of action, but I'm not seeing why it is a logically prior problem to the problem of evil? Perhaps this is addressed in your paper?

    Moreover, a person's actual failure to know the Good for man doesn't entail that it is not possibly known. So again if there is a logical problem threatening Theism as you say, it appears that it would need to rule out the metaphysical possibility of man coming to know the Good (which once again means proving a universal negative). This is how I am imagining the dialogue occurring:

    Humean: P of E.

    Theist: I don't know precisely why God permits natural evil in the world, but you can't rule out the possibility that God has morally sufficient reasons for doing so.

    Humean: Well, you don't know what those morally sufficient reasons are and worse, you don't even know what the good for you is and that seems like a prior problem you need to address if you're going to answer the problem of evil in the way that you are attempting to.

    Theist: Look, for whatever reason I haven't come to know what is the Good for me, but how does this render the concept of God incoherent? I may someday come to know it after all, it isn't as if you can rule out the very possibility that I would. Thoughts?

  8. There can't be a problem of evil if we don't know what is good or evil. Once we are able to define these then the problem of evil arises in relation to theism. Going back at that point and either saying we don't know what one of the terms in the problem means, or changing the definition of God from theism to a finite and changeable deity doesn't solve the problem. The solution comes in the face of recognizing the apparent tension that nevertheless more greatly reveals the good.

    So I'm in agreement with the idea that there is a good reason for permitting evil, I just don't agree that it is unknowable or that we can have the problem evil before we know the good. If the theist doesn't know what "good" means, then the problem is evil is: if God is all powerful and all blick, why is there non-blick? That's a meaningless question, not a logical problem.

  9. So you aren't saying that the question of the knowability of the Good is a prior problem that threatens the coherence of theism, but rather that it is a prior problem in more general terms in that conceptually the problem of evil presupposes some account of good vs. evil? If this is what you mean I think I agree, in fact that hasn't been the thesis I have been arguing against.

    In effect it seems like you are saying that unless the Humean were to give a satisfactory account of good vs. evil, the traditional problem of evil argument is a non-starter. If so, then I don't see how the FWS would fail, in fact it would neither succeed nor fail since as you say there would be no problem of evil in the first place. Interestingly, I have been arguing that it takes very little to brush off the logical problem of evil also which is why I haven't been able to understand why you are pointing to the FWS as insufficient.

    Perhaps your point is just that belief in the God of Christian Theism which includes that God is wholly good requires theists to identify what is meant by 'good'. That seems right to me, but I'm failing to see the import of that into this discussion regarding the problem of evil and whether FWS succeeds or fails (or neither). I've been questioning the following lines considered together from your original post: "It is true that if evil serves a higher good then there is no problem, but has this good been correctly identified." And "If the solution has failed to correctly identify the higher good, then it fails as a solution."

    As I've hinted above, my comments about the knowability of the good (in terms of actuality vs. possibility) were directed at the idea that there is an analogous problem threatening the coherence of Theism based on a theist who responds to the P of E with, "I don't know what the Good is, but there isn't a logical problem threatening the coherence of Theism". Which of course, if this wasn't your point to begin with, those comments were entirely misdirected.

  10. I suggested that the idea that says there is a good reason to permit evil is correct, but that the FWS does not identify this, and mistakenly claims it is libertarian free will (uncaused events). To say that it isn't libertarian free will, but it is something else that we can know but at this time do not, is not the FWS. I would approach that differently than I did the FWS, more like what I've done in these comments, about whether it is clear what is good and what the implications are for Christian theism. I've put up a post today about the good and plan to go in that direction with subsequent posts.