Thomas Metcalf

I ran across an interesting argument about God's maximal perfection versus His free will. I think it's somewhere in Rowe and also in Quinn, but I might be wrong. Anyway.

Suppose that there is a greatest possible world, a world better than any other. A maximally perfect being would have no libertarian choice about whether to create this world; there would be no chance of It creating any other world. So this maximally perfect being would lack libertarian freedom.

Suppose that there is an infinite number of possible worlds, and for any possible world, there is a greater possible world. In this case, it appears that a maximally perfect being is impossible, because for any being B that created world W, there would be a possible being B* that created W* where W* is a better world than W, and therefore B* would be better.

Suppose, finally, that there is more than one greatest possible world. Again, it seems as if God has no choice except within these possible worlds. He can only create a world of this sort; there is no possible world in which He chooses to create a world that's not within this set.

The conclusion, I think, is that a maximally perfect being lacks libertarian freedom. I don't know how willing contemporary apologists are to give up Divine libertarian freedom, but I think there are other reasons for them to do so.

Philip Osborne

Suppose that there is an infinite number of possible worlds, and for any possible world, there is a greater possible world. In this case, it appears that a maximally perfect being is impossible, because for any being B that created world W, there would be a possible being B* that created W* where W* is a better world than W, and therefore B* would be better.

I have defended the thesis that there is no "best possible world" on the basis that the qualities that would make a world great do not seem to admit of intrinsic maximums. By "maximal perfection," I take it that you mean "maximal moral perfection," where a being is maximally morally perfect if and only if no being in any possible world is morally more perfect than it, and it possesses the same maximal degree of perfection in every world that it exists in. Since the thesis that there is no best possible world seems plausibly true, the best solution for theists seems to be to reject this definition, or at the very least, the latter part of it. Maximal moral perfection might be construed as necessarily never making errors in moral judgements and actions. This limits God to the set of actions which are not immoral, but most other definitions of moral perfection are more limiting than this anyway, and God might be said to have Libertarian free will to choose within this set of actions. Granted, this means that God may not necessarily be the most moral being that exists, but that is only because the property of being the most moral is an impossible one.

Alternatively, theists might adopt compatibalist free will for God; few theists have done this, because it is inconsistent with the assumptions of the free will defense. Although I would myself be skeptical of this solution, it does have a few advantages: it appears to solve the problem of divine foreknowledge and human freedom, and lends itself to versions of the cosmological argument which are forced to construe God's act of creation as a necessary event.

-Philip

Thomas Metcalf

Good to hear from you on this topic. I recall you mentioning having done some work in this area.

"Maximal moral perfection might be construed as necessarily never making errors in moral judgements and actions."

If there are two beings, and each actualizes a different world, one that it is morally better to actualize than the other, then it seems that they've made morally different decisions. And depending upon how we take "error," I think a case could be made that one has made an error whenever one has performed an action that is not as good as some other action, especially if we're talking about perfection here.

"Granted, this means that God may not necessarily be the most moral being that exists, but that is only because the property of being the most moral is an impossible one."

I suspect I'd be content to say that no being can be maximally perfect, because that would require maximal moral perfection, and there is no such thing. I don't know offhand whether this is the same as the fact that there can be no most moral being, but they look very similar.

"Although I would myself be skeptical of this solution, it does have a few advantages: it appears to solve the problem of divine foreknowledge and human freedom, and lends itself to versions of the cosmological argument which are forced to construe God's act of creation as a necessary event."

As I see it, compatibilism makes theism far more defensible. I believe divine foreknowledge is inconsistent with libertarian free will, and I believe this is a serious problem for libertarian theists.

Philip Osborne

If there are two beings, and each actualizes a different world, one that it is morally better to actualize than the other, then it seems that they've made morally different decisions. And depending upon how we take "error," I think a case could be made that one has made an error whenever one has performed an action that is not as good as some other action, especially if we're talking about perfection here.

The sense of "error" one would have to use to make such a case seems counter-intuitive. Suppose John gives $20.00 to a person without a home or money, while Joe gives $30.00. Joe has made a morally better decision, but I do not think we would say that John has therefore made an error in moral judgement. At least, we would not see it that way in practical, everyday terms. Also, since we are defining "perfection" in terms of "error," we would not want to define error in terms of perfection.

Although it seems God cannot be maximally perfect, He might at least be maximally perfect,* where maximal perfection* is defined as being maximally perfect in all non-moral senses, and maximally morally perfect in the sense I used above. It seems plausible to suppose that a maximally perfect* being would be a suitable object of most theistic religions.

Most theists would at least want to conceive of God Himself as having libertarian free will, although compatibalism does provide a solution to God knowing His own future actions. Compatibalism might also be preferable to those of a Calvinist bent, since it better preserved divine sovereignity. The primary reason it is not accepted among most theists, even if it is an unsound reason (which it may very well be), is most likely that it does not cohere with standard formulations of the free will defense.

-Philip

Kenny

Actually, here is one theist that does take the compatiblist route. You are correct that many Calvinist theologians adopt a compatiblist viewpoint. I am a Calvinist, but my conversion to a compatiblist perspective of free will preceded my conversion to a Calvinsit sorteriology. I became convinced that libertarian conceptions of freewill are incoherent.

I don’t adhere to the free will defense, but I think that it could be effectively reformulated along compatiblist lines. On a compatiblist account of free will (roughly) an action, A, upon the part of a volitional agent, S, is free if A is the intentional result of a volitional act on the part of S and the cause of said volitional act is such that it flows inevitably from S’s character in conjunction with the set of circumstances, X, in which S finds herself, but where X alone (independent of S’s character) does not wholly determine that S should volitionally intend that A. Given this definition of freedom, it is still conceivable (perhaps even more plausible) that a particular free agent or group of free agents may suffer from what Plantinga termed “transworld depravity” (the property of making at least one evil choice in all possible worlds in which one exists).

Presumably, S’s character is shaped by S’s essence in conjunction with the circumstances of S’s life up to any given point in time in which S is alive. It is conceivable that S’s essence is such that, for all possible characters that S might have and all possible circumstances in which S finds herself, there is always at least one evil choice that S freely makes, provided that her free volition is not interfered with. If such were the case, then there would be no possible worlds in which God creates S and S does not make an evil choice, provided that God leaves S’s free volition in tact.

Consequently, I don’t see why it is preferable to hang on to libertarian freedom for the sake of the free will defense.

In Christ,
Kenny

Keith Russell

Kenny said:
Presumably, S’s character is shaped by S’s essence in conjunction with the circumstances of S’s life up to any given point in time in which S is alive. It is conceivable that S’s essence is such that, for all possible characters that S might have and all possible circumstances in which S finds herself, there is always at least one evil choice that S freely makes, provided that her free volition is not interfered with. If such were the case, then there would be no possible worlds in which God creates S and S does not make an evil choice, provided that God leaves S’s free volition in tact.

Kenny, don't you mean to say "there would be no possible worlds in which God creates S and S does not have the option of making an evil choice, provided that God leaves S' free volition intact.

If S makes an evil choice in all possible worlds, that can't be called 'free will', in any sense of the term...

Keith.

Philip Osborne

If S makes an evil choice in all possible worlds, that can't be called 'free will', in any sense of the term...

The whole notion of compatibalism, if my understanding is correct, is that the possibility of choosing another option is not necessary to a workable concept of "free will." As Peter Van Inwagen states it (although he is not a compatibalist himself), the idea is that freedom is a negative concept: it is not some special property with ontological status that we all have, but merely refers to the absence of certain conditions. For instance, being free to type this paper just means that there are no obstacles or constraints to me typing this paper. This does not mean that I have the option to choose other than to type this paper, and that somehow I have the particular power of choosing which option I will act upon.

Kenny

No. I meant just what I said. S does have the option of not making an evil choice, on this account, if her choice is free (since there would be no external constraints upon her character and volition forcing her to make an evil choice), but she simply chooses to make at least one evil choice in every possible set of total life circumstances in which she might find herself.

Why not? The fact that S makes at least one evil choice in all possible circumstances, provided her free volition is not tampered with, means that her choices in this respect reflect something deep about S’s identity rather than S’s choices merely being a function of, and therefore solely determined by, the circumstances in which S finds herself. If S is choosing her course of action, out of her own character, without being wholly determined in her choices by factors external to herself, then I do not see how that could be classified as anything but free will. Certainly, S’s choosing a particular type of action in all possible worlds does not violate the definition of free will I gave in my above post. Do you have a problem with that definition?

God Bless,
Kenny

Kenny

I’m not sure I agree with Peter Van Inwagen’s characterization, at least not with respect to all compatiblist views. Some versions of compatiblist freedom (such as that of David Hume) are purely negative, where freedom is merely characterized by the absence of external constraints upon one’s choices. However, in my definition, notice that I included positive characterizations – namely that there must be some activity of character and volition on the part of a personal agent for her actions to be free. If someone were to be hypnotized or brainwashed, for example, then their actions might cease to be free, even if external constraints are lacking, on my account, because those actions may no longer flow out of that person’s character. The term compatiblism itself merely refers to the view that free will is in some sense compatible with determinism and that the counter-causal characterizations of libertarian freewill are not necessary for (perhaps even contrary to) an adequate account of free will.

God Bless,
Kenny

Thomas Metcalf

Originally posted by Philip Osborne :

"The sense of 'error' one would have to use to make such a case seems counter-intuitive."

Well, yes, it does. But I think if we're going by our intuitions, we'd still say that the person who donated $30.00, all else equal, is morally better. I would say moral perfection, most intuitively, is simply a result of there not being a possible morally better decision for any decision.

"It seems plausible to suppose that a maximally perfect* being would be a suitable object of most theistic religions."

I agree. At this point, I'm content to replace omnipotence with omnipotence* and maximal perfection with maximal perfection*.

I would be interested to see how the best formulations of the free will defense require libertarian free will.