Sunday, October 21, 2012

Is God free?

I’m going to comment on a post by Roger Olson. The original post doesn’t seem to be directly available, but you can find it here:

Many (not all) Calvinists argue that libertarian free will or, the power of contrary choice, is an incoherent concept. (E.g., Jonathan Edwards, Lorraine Boettner, R. C. Sproul, John Frame, John Piper, et al.)

i) It isn’t just “many Calvinists” who take that position. There are non-Calvinist philosophers who also take that position.

ii) Olson has a bad habit of citing random Calvinists. But if you’re going to attack a belief-system, then you need to attack its best representatives. That means you need to distinguish between popularizers and high-level thinkers. You also need to differentiate between different specializations. Are we dealing with philosophical theology? Exegetical theology? You need to target the best exponents of the relevant discipline. 

The reason is, they argue, that it amounts to belief in uncaused effects. They argue that people act according to their strongest motive.

Olson doesn’t stop to ask if that’s an accurate characterization or mischaracterization of libertarian freedom. Does he disagree with Calvinists who say libertarian freedom amounts to uncaused effects? Or does he agree with that characterization, but denies the incoherence of uncaused effects?

What I’ve often wondered is whether Calvinists who argue this believe God has power of contrary choice. If God has power of contrary choice, then it cannot be a strictly incoherent concept.

Well, that’s simplistic. They may think uncaused effects are incoherent in the case of contingent, timebound creatures. It wouldn’t follow that that’s incoherent in the case of a timeless, self-subsistent agent (i.e. God).

But to say God does NOT have power of contrary choice seems to make God a prisoner of creation; without power of contrary choice God’s decision to create would be necessary and that would make creation less than gracious and, in fact, a part of God’s own life – not a free act as if God could have done otherwise.

Yet Olson has also said:

We have run around this bush numerous times here and I tire of it (no offense intended). From an Arminian perspective, God knows because something happens; it doesn’t happen because God knows it. God’s foreknowledge corresponds to what happens; it does not cause it or even render it certain.

That makes God’s knowledge of the future dependent on the future itself. For instance, God’s knowledge of what human agents will do is caused by what they will “freely” do (in the libertarian sense), apart from divinely agency. Doesn’t that make God a “prisoner of creation”? His foreknowledge is contingent on the independent choices of his creatures. If his knowledge of the future is the effect of what they freely choose, if he is dependent on them for that information, then doesn’t that make him a prisoner of creation?

The way Jonathan Edwards attempted to get around this in The Freedom of the Will was to say that "God always does the wisest thing." Contemporary Calvinists who follow him closely agree. In other words, according to Edwards, God could have done otherwise than create the world, but he created the world because it was “most fitting” to do so.

My question is how this gets around the problem. To me it seems like a dodge; that is, it seems to attempt to answer the challenge without answering it. It seems like saying both at the same time – that God could have not created and that God could not have not created.

The question is simply this: Is it logically conceivable that God might not have created the world? Is it conceivable that God might have decided against this creation or any creation?

Edwards’ answer seems to say yes and no at the same time. That’s against the laws of logic UNLESS he can explain how the “yes” and the “no” are referring to different things. But in his explanation, they aren’t.

The question is: Is God the prisoner of his own wisdom (or of anything)?

i) Well, that’s a very different question. What’s the alternative? Is Olson suggesting that for God to have libertarian freedom, he must be free to think or act foolishly? Is Olson saying the God of Arminian theism is a fool?

ii) There’s a difference between saying God always acts wisely and saying God always does the wisest thing. We can affirm the former without affirming the latter.

IF one says that God “always does the wisest thing” WITH THE ASSUMPTION that there is always only ONE “wisest thing,” then how is one not making creation necessary and therefore not gracious? (A basic principle of theology is that what is by nature cannot be by grace. If I HAVE to rescue you, it’s not an act of mercy or grace.)

The upshot is, of course, that IF the creation and redemption of the world by God is truly gracious and not automatic, then God must possess libertarian free will, power of contrary choice. And if God possesses such, it cannot be an incoherent concept.

i) Olson is equivocating. Does he mean “gracious” in the sense of “gratuitous” or “gracious” in the sense of “merciful”?

ii) Moreover, since he seems to be using “gracious” in the soteriological sense, his objection undercuts a key plank of Arminian theology. Arminians routinely contend that God would be morally defective unless he made salvation available to everyone. But on Olson’s own definition, unless God is at liberty not to be gracious or merciful to sinner, then salvation isn’t really an act of mercy or grace.

It seems to me that to say “God always does the wisest thing,” implying by that that God must do such-and-such (e.g., create the world), is the same as to say that God is a machine and that the creation and redemption of the world is not by grace but by nature. Only if God really could have done otherwise than create can creation be by grace only. Grace cannot be compelled and still be grace.

I think Edwards skirted the issue and so do his followers who repeat his argument in one form or another. To say “God always does the wisest thing” is either to imply that God is an automaton, in which case creation and redemption are automatic and not gracious, or to imply that God COULD do that which is something other than “the wisest thing.”

Assuming, for the sake of argument, that there is a wisest thing to do, always choosing to do the wisest thing wouldn’t make God an “automaton” or a “machine.” Rather, it would mean God is benevolent and rational. God always does the wisest thing because that’s the most reasonable thing to do, and a good God is a reasonable God. Automata don’t act for reasons.

The alternative is for God to be unreasonable. Once again, is Olson admitting that the God of Arminian theism is unhinged?

Why assume that there is always only ONE “wise thing” to do – even for God? Why couldn’t it have been wise to create but also wise not to create? Of course, as any rationalist will ask, then why did God create? Was it simply an arbitrary choice – like throwing the dice?

However, I prefer to argue that for God, as for us, there are moments when two alternative options are equally wise and no controlling, determining factor interior (such as motive) or otherwise determines which option one must choose to be right.

I reject the notion that “God always does the wisest thing,” not because I think God is anything less than absolutely wise but because I don’t believe there is always only one “wisest thing” in every situation of choice between options. To avoid making creation and redemption other than gracious, we have to suppose that God really could have chosen not to create. To say “God always does the wisest thing” is to imply that God really could not have done otherwise.

i) Because Roger Olson inhabits an Arminian bubble, where he defines his position exclusively in reaction to Calvinism, he’s oblivious to the fact that the question of divine freedom is hardly confined to Calvinism. That’s an issue which cuts across various schools of thought in historical and philosophical theology. Arminianism is by no means exempt from the same considerations:

William Rowe, Can God Be Free? (Oxford University Press, 2004).

Arminians must also wrestle with the question of how or whether God is free.

ii) Historically, the question is bound up with two interrelated issues: (a) Is there a best possible world? (b) The principle of sufficient reason.

iii) There are different ways of engaging the argument. You can deny the existence of a best possible world. You can affirm the existence of a best possible world, but deny that God is obligated to instantiate the best possible world. You can deny the PSR, although that’s a very costly denial. You can also argue that the question poses a false dichotomy.

iv) Speaking for myself, I doubt the existence of a best possible world. I think there are better and worse possible worlds, but it’s not obvious to me that there’s a best possible world.

Among the better possible worlds, we have tradeoffs between incommensurable goods. These are incommensurable inasmuch as not all possible goods are compossible goods. One possible world encapsulates some goods to the exclusion of other goods. These can’t both be realized within the same timeline. Rather, they represent alternate timelines.

Let’s compare two possible worlds. In 1.0, Ethan marries Effie. They have a happy marriage. They have three kids, who turn out well.

In 2.0, Ethan marries Effie. They have two kids before Effie dies of cervical cancer. Ethan then marries Gwen, by whom he has two more children.

His son by Gwen betrays his (son's) best friend. Later, his son becomes a Christian, repents of his perfidy, and is reconciled to his old friend.

His daughter by Effie is so mad at God for letting her mother die that the daughter becomes a bitter atheist who goes to hell when she dies.

Which possible world is the better possible world? Well, 1.0 is better in the sense that it generally avoids the evils of 2.0. However, it avoids the evils of 2.0 by eliminating the goods of 1.0.

For one thing, Ethan has three kids by his second wife. They don’t exist in 1.0. In addition, Ethan’s son in 2.0 experiences redemption.

Although 2.0 has certain evils not found in 1.0, those are evened out by certain goods not found in 1.0.

So it’s hard for us to say which possible world is better overall.

v) But let’s assume, for the sake of argument, that 1.0 is somewhat better than 2.0. Even so, is there some compelling reason why God should choose 1.0 over 2.0? I don’t see why.

“Better” for whom? There are people who go to heaven in 2.0 who don’t exist in 1.0. So 1.0 isn’t better for them! And there’s no corresponding good in 1.0, for they have no counterparts in 1.0.

Likewise, even if 1.0 is better overall, 2.0 may have a distinctive good which is better in itself. What if it’s a tradeoff between a possible world where the whole is greater than the parts, and a possible world where the parts are greater than the whole?

Is a possible world with some unique goods which are individually better than anything in another possible world less preferable? Is a more uniformly good world preferable to a world with higher highs and lower lows? Hard to say how we’d make that calculation.

But even if a more equitably good world is better overall, why assume that’s preferable to the alternative? What’s superior in one respect (whole greater than parts) may be inferior in another respect (parts greater than whole).

vi) For all we know, this is a false dichotomy. What if God doesn’t have to choose between instantiating one possible world rather than another? Perhaps God created a multiverse in which different forking paths actually play out in parallel worlds. (Which doesn’t mean God instantiates every alternative, just some.)

vii) But the issue is also bound up with the PSR. Alexander Pruss has offered a sophisticated exposition and defense of this principle:

The Principle of Sufficient Reason: A Reassessment (Cambridge University Press, 2006)

“Leibnizian cosmological arguments,” W. L. Craig & J. P. Moreland, eds. Blackwell Companion to Natural Theology (Oxford: Blackwell, 2009)

viii) There are stronger and weaker versions of the PSR. Let’s define the PSR thusly: Every contingent fact has an explanation. There’s a reason for every event.

What’s striking about the PSR is how closely that principle corresponds to predestination. According to predestination, everything happens for a reason. Indeed, there’s a good reason for whatever happens.

So it’s hard to attack predestination without attacking the principle of sufficient reason. But if Arminians attack the PSR, they will pay a very high price. That throws into question the rationality or intelligibility of the universe. If you question the PSR, you question our ability to explain anything. Where do we draw the line?

ix) However, commitment to the PSR doesn’t commit us to the proposition that God wasn’t free to choose otherwise. Rather, it simply means God had a good reason for choosing to make this world.

Here I’m tempted to throw back at the Calvinist his or her own argument that God’s choice of "some to save" and "others to damn" is not arbitrary without any hint at what might explain it. In other words, if it’s fair for the Calvinist to argue that divine selection is not based on anything God “sees” in the elect or the damned (that differentiates them) and yet is not arbitrary, then why couldn’t the person who believes in God’s power of contrary choice argue that God’s choice to create is not arbitrary even though no specific reason for it can be given?

This is one of Olson’s stock objections to Calvinism, as if that hasn’t been dealt with. Olson has a dishonest habit of repeating the same objections while ignoring the answers.

i) As I recently observed, in 1 Cor 1-3, Paul talks about God disproportionately electing or reprobating certain social classes. One might be tempted to say that makes election conditional, but that’s ambiguous–for in this case, God creates the distinguishing conditions. God determines when, where, and to whom you will be born. So God isn’t electing or reprobating individuals on account of their social class, as if that’s an independent variable. It’s not “conditional” in that sense.

Likewise, both Calvinists and Arminians say faith is a necessary condition of salvation. But in the case of Calvinism, faith is not an external factor which affects or effects the divine response. Rather, faith is a divine gift. That’s a condition which God himself supplies and satisfies.

In both cases, the condition is ultimately contingent or dependent on divine agency. Not something God responds to. Rather, our responsiveness, or lack thereof, is the result of divine agency.

Likewise, God can have reasons for electing one sinner and reprobating another sinners. But these are his reasons. They don’t take their source of origin in the creature. If God differentiates one creature from another, God made them different in the first place.

ii) Moreover, it’s not just a question of the individual, but his life-story. If God elected the same individual rather than reprobating said individual, that would result in a different life-story. Conversely, if God reprobated the same individual rather than electing said individual, that would result in a different life-story. And when you combine different life-stories, that, in turn, generates an alternate world history.

So God can have a reason for electing one person and reprobating another: he prefers one world history over another world history. Consider the chain-reaction triggered by God calling Abraham out of Ur. That sets in motion a long-range series of nested events, none of which would take place if God reprobated Abraham.

iii) Furthermore, Arminians don’t posit libertarian freedom for its own sake. Rather, they claim that’s a necessary condition for praise and blame. However, even if God had libertarian freedom, that doesn’t mean libertarian freedom is a necessary condition for praise or blame.

Now, it’s another thing entirely to argue that God possesses power of contrary choice but humans don’t. That’s a different argument. The natural answer is “Why?” If God possesses it, why couldn’t he give it to humans? There doesn’t seem to be anything about power of contrary choice that requires deity. It’s not like omnipotence, for example.

For one thing, that’s giving some humans godlike power over the fate of other humans. For instance, the past choices or choices of past libertarian agents create our present. The alternate possibilities they select become our realties. They shuffle the deck. We must play the hand they dealt.

Consider how Arminians construe Rom 14:15 & 1 Cor 8:11. They actually imagine that God has delegated to mere human beings the power to effectively damn their fellow human beings. To seal their eternal demise through the choices they made for them. Sinful, shortsighted men reprobating their fellow man.


  1. One wonders why Olson does not simply claim he is an open theist. He continues to flirt with open theism and I say the time has come for him to marry it.

  2. Most if not all choices have a moral element and God knows exhaustively each choice that is made. Obviously He does not have the power of contrary choice with regard to moral choices. He cannot choose to be wicked, which would be contrary to his character.

    Also the concept of freedom and bondage of the will is a discussion that usually centers around salvation, not choices of an insignificant nature. In this sense, even Olson, who believes in prevenient grace, does not believe man has contrary choice apart from grace. So the question is not whether man's will is free or not. This has already been established. The question is whether God's grace is effectual (as the Bible affirms) or simply places man is a neutral place outside his natural depravity (as Arminians believe apart from ANY scriptural evidence).

    John Hendryx