A hole in the Calvinist logic regarding free will?

According to Roger Olson, there’s a glaring hole in the Calvinist logic regarding free will. And, it’s a hole that Calvinist’s generally refuse to acknowledge.

To see why, Olson points out that many Calvinists contend that incompatibilism as a view of free will is simply incoherent. (There are many different kinds of incompatibilism, but in a nutshell it’s the idea that my having true free will in a given instance is not compatible with the idea that my action in that instance could be caused by some prior event or state of affairs.) The Calvinist contends that if an actually is truly “uncaused,” then it is irrational or random. And, if our actions are irrational and/or random, then they do not come from our choices and they they are not the kinds of actions for which we can be held responsible – i.e. they are not “free.” Consequently, there is no such thing as incompatibilist free will. For the Calvinist, according to Olson, that is an oxymoron; it is incoherent.

But, Olson goes on to argue that this raises a problem for our understanding of God’s free will. If the very notion of incompatibilist free will is incoherent, then God himself cannot have incompatibilist free will. This, in turn, would mean that God’s own actions are caused by some event or state of affairs. And, many Calvinists will agree here, contending that God’s actions are “caused” by his nature. He does the things that he does because he is perfectly the kind of God that he is. But, and here is the real nexus of Olson’s argument, this would seem to mean that all of God’s actions are necessary. He created the universe because he had to; it was an expression of his perfect and immutable character. There’s really nothing else he could have done.

And the problem for Olson is that this account of God’s creative act was clearly rejected and declared heretical by quite a number of early theologians. Most of these thinkers insisted that God’s creative act had to be understood as a free act of his will. God was free to do otherwise, though it was perfectly fitting for him to choose to create.

So, the tension that Olson sets up is this. If you are going to claim that incompatibilist free will is incoherent, then you must also affirm that God’s actions are all necessary consequences of his character. Conversely, if you are going to claim that God has incompatibilist free will, then you cannot also claim that incompatibilist free will is incoherent.

I’m sure that Olson is perfectly aware that none of this actually serves as an argument for maintaining that humans actually have incompatibilistically free wills. But, he maintains that it does place the Calvinist in quite the quandry, and he argues that most Calvinists are unwilling to face it head on.

What do you think? Is this truly a hole in the Calvinist logic regarding free will?


About Marc Cortez

Theology Prof and Dean at Western Seminary, husband, father, & blogger, who loves theology, church history, ministry, pop culture, books, and life in general.

Posted on August 26, 2010, in Anthropology, Philosophical Theology and tagged , , , , , , , , . Bookmark the permalink. 12 Comments.

  1. This would be an issue if not for two issues:

    1) God’s aseity and transcendence
    2) Definition of freedom

    God’s aseity because it points our our non aseity, we are by definition of being creatures, dependent upon a creator, and thus our definition of freedom must be tweaked. God, in his aseity, is “other” and not limited by such dependency.

    Definition of freedom because the Calvinist working of definition is not hinged upon whether something is necessary or not — Freedom in the traditional Calvinist perspective hinges on desire. Desire and necessity are not incompatible, and can most often be seen in relationship. For example, the old “what if your mugged at gun point” question on freedom – My desire is to live, and it is necessary for me to hand over my wallet to do so. Therefore, the mugger gets my wallet.

    • Good thoughts. But I don’t think that your first response will work here. Olson’s argument does not seem to be affected by the Creator/creature distinction because it involves the question of whether the incompatibilism is incoherent. If it is, then an incompatibilistically free will cannot exist any more than a square circle can exist. So, you could affirm that God as Creator has a kind of free will that is not available to humans as his creatures, but you cannot say that the kind of free will he has is logically impossible.

      You’re right on the issue of “necessity.” That is a notoriously difficult term to use/understand and I probably should have avoided it in my summary. (Actually, all the terms in this discussion are notoriously difficult.) Maybe it would be better to talk about whether choices are “determined.” In this sense, the connection between freedom and desire might be more fruitful, and traditional Calvinism would indeed say that choices are determined by desire (compatibilism). But, of course, once we say that God’s choices are determined by his desires, we have to ask, where do his desires come from? And, if his desires are determined by his character, and his choices are determined by his desires, aren’t we still back at saying that God’s choices are determined? We’ve gained the ability to say that this is okay because God chose the things that he wanted to choose. But, we still have to say that God’s choices were compatibilistically free at best. That is, he chose to create because he wanted to create, but his act of creation was fully determined all along (by his character).

      So, we still seem to have the same decision to make. Either God’s will is incompatibilistically free and, therefore, incompatibilism is not incoherent, or God’s will not not incompatibilistically free and, therefore, God’s will is (compatibilistically) determined. And, for Olson, the latter is deeply suspect because the early church clearly rejected the idea that God’s act of creation was determined.

      • Hey Marc,

        I wouldn’t say that God’s will could be logically impossible, as I agree that God’s omnipotence is limited by that which is logically able to be done. God cannot make a square circle, a rock to big to lift, a burrito to hot, etc. and certainly a logically impossible will would fit into that. I would say that his argument falls because 1) compatibalism isn’t logically impossible, and 2) the distinction allows for the possibility that God exhibits a will different than ours which would necessarily also be logically possible, but not “trapped” by compatibalism in himself. As it is, Olsen has presented a deductive argument, but deductive arguments only need possibility to be defeated, not outright proofs of reality. For me, the aseity/transcendence of God allows for the possiblity and seems to be a defeater argument. If Olsen put forward an inductive argument, then we need to go with something more like defining “necessity” as it relates to will.

        Also, I take arguments from the fathers to heart. It’s not a good spot to be in when you disagree with historical theology! But, there are a couple places where I do, in fact, disagree — as a baptist, infant baptism is one. Another area where I disagree is with a platonic view of impassibility, which the fathers tended to hold to as well. So, if thinking through this I end up disagreeing with the fathers, I would be weary, but able, to disagree.

        That being said, the tension of incompatibalism being coherent in regards to God but incoherent in regards to humans is one that I’m fine with. This is mostly due to the fact that would seem to be the tension held in the Bible–we being limited by our sinful natures to be slaves to sin, and God being limited only by that which is logically impossible.

        It’s late, so I hope some of that rambling was sensible.

      • I don’t think Olson’s argument requires that compatibilism itself be logically impossible. He just wants to establish that if the Calvinist wants to hold that incompatibilism is logically impossible, this commits him to certain beliefs about the nature of God’s will that he doesn’t think most Calvinists are willing to hold. As I’ve reflected on Olson’s argument a bit more, I’m inclined to think that the best Calvinist responses would be one of the following:

        1. Affirm that incompatibilism is not incoherent. This response would go after one horn of Olson’s dilemma by rejecting the incoherence argument so often used by Calvinist’s (and compatibilists of all stripes). This would allow the Calvinist to affirm that God has an incompatibilistically free will, even if he still chooses to deny that humans also have such a free will. Of course, he’ll have to come up with some other reason for denying that humans have incompatibilistically free will, but the incoherence argument is not the only one in the Calvinist arsenal.

        2. Affirm that God does indeed have a compatibilistically free will. This response would go after the other horn by rejecting that a compatibilist view of God’s will is a problem. This one’s a bit more complicated because it would have to go into a long discussion of “necessity” as it relates to free will and whether this is a problem for God. And, of course, you’d have to engage the objections of the early church fathers. I’m not entirely sure that this would be an insurmountable problem, since I’m not convinced that the arguments the fathers developed against the “necessity” of creation in response to their gnostic interlocutors would necessarily apply to the a Calvinist concept of a God who always acts in ways consistent with his good and perfect character. But, that’s worth a post or two in its own right.

      • Whoops, the first paragraph was in relation to God’s aseity/transcendence, but I forgot to mention that at the start.

      • I think what you said in your latest post is what I was trying to get at (eventually). At least, I would be comfortable affirming what you said in points 1 and 2.

        I was trying to say that God’s aseity/transcendence defeats his argument in terms of allowing the possibility of not having a 1:1 correspondence between our will and God. So his deduction is that it is inconsistent for 1) Calvinists to say incompatibalist free will in man is incoherent, and 2) God’s incompatibalist free will is coherent. But the aseity/transcendence of God allows us to say there is a possibility that that is so, therefore the deductive argument doesn’t work because a logical inconsistency cannot be shown beyond the realm of possibility.

        But defeating a deductive argument is not the same as presenting a defense of your own view, and so to positively state the case of the calvinist, then (s)he would have to then affirm either your point 1 or 2. At that point, I’m comfortable with either argument, and both together is, I think, even better.

  2. Bryan,

    I don’t see the work that your appeal to transcendence is doing. Why would God being transcendent of itself preclude a judgment as to whether he fulfills the conditions on a libertarian conception or a soft determinist one?

    2nd, if it were to do so, then the Reformed position would entail excluding both relative to God.

    3rd. If a concept is incoherent, then it is something that is not true of God either anymore than a morally perfect being performing an evil act is coherent and could be attribted to God.


    If the Reformed concede that incompatibilism/libertarianism is not incoherent, then the arguments against it on that score in anthropology/soteriology go out the window.

    2nd, if God only fulfulls the conditions on a soft determinist conception of freedom, then God doesn’t choose between creating and not creating or saving or not saving. It is hard to see how either creation or salvation are gratuitous then if they are determined actions by God.

    Further, some Reformed writers like Hodge do not in fact see to attribute a libertarian conception of freedom to God relative to creation.

    As an aside, the desire material is irrelevant. Desires are dispositions and not causes of actions. Decisions execute plans of actions (intentions) relative to some desired end.

    And a more thorny problem is whether Christ has free will in the incarnation or not. An answer there would seem to be paradigmatical for how the divine and human wills ought to relate.

    • Perry, good thoughts. Indeed, there’s so much in your comment, that I’m not even going to try to interact adequately with all of it. I may need to post again later on some of this. But, here are some quick thoughts.

      1. I agree that transcendence doesn’t seem to help much as long as the issue is whether incompatibilism is incoherent. If it is, then God can’t have it either. Transcendence would only come into play if we determined that incompatibilism was not incoherent. Then someone might use transcendence might to argue God alone has that attribute.

      2. Yes, if Reformed theologians conceded that incompatibilism was not incoherent, they would no longer be able to use the incoherence argument. But, that is not the only argument available against incompatibilism, though it is one of the more prevalent. So, they’re task would become harder, but not necessarily impossible.

      3. The question of grace does seem to be the tricky problem if you affirm a compatibilist understanding of divine volition. The best (only?) option that I can see here is to argue that grace is an essential attribute of God’s character and that his volitions flow from this character. So, his choices (understood as compatibilistically free determinations of his divine nature) are gracious in that they are expressions of his gracious character. This is not going to satisfy anyone who affirms incompatibilism, but it seems like the best option available for a Calvinist with a compatibilist understanding of God’s will.

      4. Yes, there are quite a number of Reformed theologians who openly affirm a compatibilistic understanding of God’s will. I was a little surprised to see Olson say that he has a hard time finding anyone who’s willing to go down this road.

      5. Whether desires can function in a causal account of volitions is disputed. So, I don’t think they can be set aside quite that easily.

      6. Good call! I’d love to see Christology brought into discussions of human volition more often. That may be my next post on the subject.

  3. A point of clarification. The taxonomy is compatibilism or incimpatibilism. Either option only tells us about what is logically possible or impossible, and not whether such is true in fact or not.

    From there, we have Soft Determinism on the Compatibilist fork and Hard Determinism or Libertarianism on the Incompatibilist fork. So what God has if he has anything is that he fulfills the conditions on Soft Determinism, Hard Determinis or Libertarianism.

    1. Still needs some clarification. We’d need a specific notion of transcendence that would rule out that kind of knowledge as available to us. Just any notion of transcendence won’t do that or not obviously so.

    2. I agree, but it is an awfully big loss in the Reformed arsenal. Much of Edwards goes down the toilet.

    3. By attributes we need to be clear. Attributes are not properties, at least not intrinsic properties or qualities, sans a Reformed gloss on divine simplicity. They are the things that “get said” about God or are predicated of God, they may or may not have grounding in God (depending on a given tradition and how it glosses that relation.).

    Grace here would be a determined way God is disposed, which seems to fly in the face of something he could refrain from in the sense of gratuity. Are things of nature grauituous or no? What is the relation between nature and grace? That is an awfully big hornets nest.

    God’s being gracious in disposition doesn’t seem like omniscience in that he could not choose to cease to be all knowing but he could refrain from having saved Joe and saved Sally instead. It seems like a matter of will and a matter of will where he could have willed otherwise. Here I don’t think the escape hatch of conditional analysis will help as it is defunct.

    Further, we’d have to say that God couldn’t in fact save the reprobate because it flowed from his nature/character that they be damned. The salvation of the elect would then be on a par with not only the existence of the world, but God’s own existence. This brings us back to the same problem faced by Origen.

    What is worse, if the world is determined to exist due to God’s own determined actions, why isn’t the world eternal like God? What room is there for a difference between an act in eternity and an eternal act given an eternally determined being and an eternally determined act? Given simplicity how can there be a difference in God (not in predication) between the two?

    4. I think contemporary Reformed aren’t as clear in the doctrine of God. Older writers more or less followed Aquinas or Scotus in a more or less modified form concerning God’s freedom relative to creation.

    5. I can grant that some dispute that desires aren’t causes, but they do so at the cost of leaving deliberation and execution qua decisions as explanatory danglers. And if they are disputed then soft determinists can’t appeal to them in an uncontroversial way as doing the kind of work they think they do.

    6. As far as Christology goes, this was hashed out in the Sixth council against Monothelitism and Monoenergism, of late much has been written. I do not think Soft Determinism is compatible (PUN!) with that conciliar resolution, but I’ll leave that for your future posts on the topic.

    • I think your third point was particularly helpful in pointing out the challenge of affirming a compatibilist view of divine volition. You’ve done a nice job summarizing the issues that anyone faces if they choose to take this approach to answering Olson’s argument. And, it’s for reasons like these that I’ve always been more inclined toward understanding God’s volition in incompatibilist (libertarian) ways.

      I also appreciated your comment that the disputed role of desires in causal accounts of volition means that soft determinists can’t simply appeal to them either. That’s absolutely right. I’m often frustrated by compatibilists who appeal to desires as though that settled the matter of what accounts for free will and responsibility. It’s going to take a lot more than that.

      I definitely agree that the 6th council and Maximus’ theology in particular have a lot to say about this. At times it feels like wading through one mystery (the incarnation) to understand another (free will), but I’m firmly convinced that theological anthropology (in all its aspects) must be centered in Christology.

  1. Pingback: Jonathan Edwards on the Freedom of the Will « scientia et sapientia

  2. Pingback: 5 Arguments that Calvinists Should Stop Using « scientia et sapientia

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: