Doctrine of Creation (Part 12): Arminian And Molinist Accounts of Divine Sovereignty and Human Freedom

August 22, 2018     Time: 42:51

Arminian And Molinist Accounts of Divine Sovereignty and Human Freedom

In our study of divine providence we've looked critically at the Calvinist view of providence, and I offered a five-point critique last Sunday of that reconciliation of divine sovereignty and human freedom. Today we want to turn our attention to the Arminian account of divine sovereignty and human freedom on the basis of God's foreknowledge of the future.

Consider the following biblical passage from Acts 4:27-28:

for truly in this city there were gathered together against thy holy servant Jesus . . . both Herod and Pontius Pilate, with the Gentiles and the peoples of Israel, to do whatever thy hand and thy plan had predestined to take place.

Here we have a staggering assertion of divine sovereignty over the affairs of men. The conspiracy to crucify Jesus (which involved not only the Romans but also the Jews living in Jerusalem at that time, but more particularly by name Pilate and Herod who tried Jesus) is said to have happened by God's plan based upon his foreordination. How can the Arminian make sense of so sweeping a sovereignty as this?

The proponent of God's simple foreknowledge can make no good sense, I think, of God's providential planning of a world of free creatures in the absence of middle knowledge. On such a view, God has (logically prior to his decree to create a world) only knowledge of all possible scenarios that could happen. But he has no knowledge at all of what would happen under any given circumstances. So even given the circumstances of the existence of the Roman Empire and the Jewish Sanhedrin and Pontius Pilate and Herod and Jerusalem in the first century, God has no idea what these free agents would do in these circumstances. So logically posterior to God's decree to create a world involving such persons and circumstances, God must consider himself extraordinarily lucky that all of these people did exactly what he wanted them to. We can imagine God saying to himself, What a break! Pilate and Herod and all those people each did exactly what they were supposed to do! Actually, the situation is even worse than that because, given that the circumstances in which these actions were taken were themselves the result of innumerable prior contingent circumstances, God had no idea whether Herod or Pilate or the Israelite nation or the Roman Empire would even exist posterior to his divine decree to create a world. So God must be astonished to find himself in such a world – a world in which out of all possible worlds that he might have created, the world exists in which mankind falls into sin and God sends his only Son into human history as a substitutionary sacrificial offering to rescue fallen humanity.

Of course, I'm speaking anthropomorphically here in describing God's surprise, but the point remains that without middle knowledge God cannot know prior to his creative decree what the world would be like. If the Arminian goes on to appeal to God's foreknowledge of the future in order to say that God’s foreordination of future events is based upon his simple foreknowledge then this trivializes the doctrine of foreordination or predestination – it makes it a kind of fifth-wheel that bears no weight. For if God knows that something will happen then there's nothing more to foreordain. If it will happen then it will happen. The future by definition is whatever will happen, so if it will happen, it will happen. Foreordination becomes a redundancy, and surely there's much more to the biblical doctrine of foreordination than the triviality that God ordains that what will happen will happen.


Student: If God's foreknowledge is, say, if we compare that with a mathematical equation and where, say, y=mx+b, and if human will are either on the equation or out of the equation and whether they will abide by the equation, doesn't that explain the foreknowledge? God knows the value where human will choose? Like last week, the example where the king throws a banquet and those people that their value system is outside of God's will will not attend the banquet but those that God compelled to come he raised the value so high that those people cannot resist?

Dr. Craig: That would not be the Arminian view because on the Arminian view people have libertarian freedom. This is not causal divine determinism. I think you would be right in saying that you can explain God's foreknowledge of what will happen on the basis that this is what will happen. These events will be freely chosen by these agents, and therefore God foreknows it. But you shouldn't think that in any way because he foreknows it therefore it's determined to happen. They could choose differently in those circumstances. But if they were then God would foreknow something different. So the point of the critique is that this kind of knowledge that we're talking about comes too late for God to do anything with it. If he doesn't have knowledge of what agents would freely do prior to his decree to create the world, but he only knows that after his decree, then it's too late to do anything. It doesn't serve to make a difference.

Student: He has a counter-plan when the thing doesn't happen as he intended – like he intended to throw a banquet but then if they reject it he goes out and finds other people and he still threw his banquet.

Dr. Craig: Right. But remember that illustration we gave last week was an illustration of middle knowledge, not simple foreknowledge. God knew who would refuse to come to the banquet and who would come if invited. But what we're talking about here is simply God's knowledge of who will come to the banquet and who will not come to the banquet. The point of the criticism is that he only gets this kind of knowledge logically posterior to his decree to actualize a world. At that point it is redundant to look into the future, see what will happen, and then declare this is what will happen. Because, by definition, what will happen will happen. There's nothing more for foreordination or predestination to do.

Student: I'm wondering if the Arminian suffers from a similar problem that the objector to the fine-tuning argument suffers from. Since God wouldn't have a counterfactual knowledge logically prior to the divine creative decree then he wouldn't have counterfactual knowledge of certain quantum scenarios which are indeterministic. Neither would he have, for example, knowledge of what the constants and quantities of the universe would be like. If it's incredibly improbable that the constants and quantities in the universe obtain as they do – and that's to the discredit of atheism – then wouldn't it be also improbable to the discredit of the Arminian to affirm that we live in a world . . .

Dr. Craig: I think that would depend on whether or not you affirm that quantum indeterminacy is ontic, that is to say is it a feature of objective reality or is quantum indeterminacy simply epistemic (it is in your mind)? We cannot determine the precise position and momentum of a subatomic particle, but some theorists would say that doesn't mean that they don't have such quantities. They do even if they are not determinable by us. So if you think that this kind of indeterminacy is not real but just in your mind then it wouldn't be a problem for God or the Arminian to say that God has selected those constants and quantities that are physically determined by him and will produce a universe that's conducive to embodied conscious life like ourselves.

Student: Could the Arminians say (not advocating, just questioning) – and this speech that is quoted in Acts 4 is after Christ's resurrection and after the Ascension – could the Arminians say God did not have to foreknow Pontius Pilate and Herod doing these things in terms of Jesus’ crucifixion in order to predestine or preordain that? Instead, logically prior to creation, God had determined that he would implement a redemptive plan. These circumstances occurred later, and looking back (these people here are looking backwards) now we can say that this was foreordained because God chose those circumstances at that time to implement his plan. He didn't have to foreknow these people logically prior to creation. Could the Arminian say that?

Dr. Craig: Well, it seems to me that he would. It says that they were gathered together to do whatever thy hand and thy plan had predestined to take place. How can people be predestined to do something if God doesn't even know that those people would exist or doesn't know what they would do?

Student: Couldn't you have just predestination of the general contours, not of the specific people?

Dr. Craig: Yes, you could say that God causally determines the sort of broad circumstances and then lets agents run free within that. But that's why a verse like this one is so striking. It is so specific that you have not only the Jewish nation and the Roman Empire but by name Pilate and Herod were all planned out by God and predestined to do what they did in crucifying Jesus. This is not just setting the sort of general contours of history. This is knowledge that is so specific that it cries out for some sort of explanation as to how God could plan something so contingent to happen.

Student: Just to follow up on that question. This is more of a fundamental and basic question, but my understanding of the word that is translated into “predestination” is that it can be literally interpreted as knowing beforehand. Is that completely off-base?

Dr. Craig: Yes! [laughter] The word in the Greek for “foreknow” is proginosko which means to foreknow. But “foreordination” is proorizo which means foreordain. Orizo is to ordain something, so proorizo means to foreordain or sometimes translated predestined. Now, what the Arminian will say is that God’s foreordination is based on his foreknowledge. He looks into the future, sees what will happen, and then foreordains that that will happen. My criticism is that that trivializes the doctrine of foreordination because it just means that God declares that what will be will be, and that's a tautology. Of course what will be will be. So there's got to be something more to this doctrine than that.


Let’s turn to a Molinist account of providence. If we take the biblical word for foreknowledge (prognosis) to encompass middle knowledge then we can make perfect sense of God's providential planning of a world of free agents for via his middle knowledge God knew exactly which persons if members of the Sanhedrin would freely vote for Jesus’ condemnation, which persons if in Jerusalem would freely demand Christ's death and favor the release of Barabbas, what Herod (if king) would freely do in reaction to Jesus, and what Pilate if holding the prefecture of Palestine in AD 30 would freely do under the pressure of the Jewish leaders and the crowd. Knowing all of the possible circumstances and persons involved and the permutations of these, God decreed to create just those circumstances and just those people who would freely do what God willed to happen. Thus, as Luke insists, the whole scenario unfolded according to God's plan.

This is truly mind-boggling if you reflect on it. When you think that the existence of the various circumstances and the persons involved were themselves the result of myriads of prior choices on the part of these and other free agents and that those in turn were predicated upon yet prior contingencies on and on into the past then you quickly see that only an omniscient mind could providentially direct a world of free creatures toward his sovereignly established ends. In fact, Paul reflects, “None of the rulers of this age understood this, for if they had they would not have crucified the Lord of glory” (1 Corinthians 2:8). I'll just read that remarkable verse again: “None of the rulers of this age understood this, for if they had they would not have crucified the Lord of glory.” Once you grasp it, the doctrine of divine middle knowledge issues in adoration and praise of God for so breathtaking a sovereignty.

Most of the objections to a Molinist account of providence are really objections directed at the doctrine of middle knowledge itself. We've already talked about those in our study of divine omniscience when we studied the attributes of God together. But what specific objection might be raised, not against middle knowledge, but against the Molinist account of divine providence? Here the Reformed theologian might object that a Molinist theory of providence is too successful in showing how God could sovereignly control a world of free creatures. The Reformed theologian could say it gives God such control that the account ultimately becomes virtually indistinguishable from the Calvinistic view. For given that the circumstances (let's call them C) in which a person finds himself are freedom-permitting circumstances, it must be just a brute fact about how a person (call him P) would choose in circumstances C. But then there must be innumerable other circumstances (let's call them C*) which differ from C in imperceptible ways in which person P would choose differently than he would in C. For example, maybe C* involves a molecule in Alpha Centauri being moved a couple of centimeters in one direction. And in C* the person would choose differently than he would in C. But then it would be plausible that what God could do would be by placing a free agent P in these circumstances he could get the free agent to freely do virtually anything that he wanted him to do without any deleterious impact upon God's providential plan. A causally remote event like a stellar event at Alpha Centauri wouldn't upset the applecart of God's providential plan here on Earth. So by placing P in C* rather than C he can bring it about that P does whatever God wants him to do without upsetting his providential plan and thus God can sovereignly bring about any creaturely decision he wants and therefore any world that he wants.


Student: I guess maybe I'm just not fully grasping this. I'm still a little confused here. How again is it that just because in C* you say you have maybe a molecule that's moved differently – how is that supposed to make it such that somebody . . . ?

Dr. Craig: It's based upon the claim of the Reformed theologian that since these are freedom-permitting circumstances (they're indeterminate, right? They don't determine P to do what he does) that it must be just a brute fact. It's just a brute fact that that's how P would choose in C. But if that's the case then it might just be a brute fact that in C* he would choose differently. I sense your skepticism about that move, and I think it's well-placed, and I'll make that point myself. But in terms of understanding the objection, that's the idea. These are brute facts, and therefore God can fiddle around with the circumstances in imperceptible ways that won't hurt anything in which P would choose differently than he does in C.

Student: I want to say: so what? What I mean is, man still has his free will. He could, but God knows he won't.

Dr. Craig: OK, this class is so perceptive and smart. I mean, someone explains one objection, you nail the other one. I'm going to make that point in the next minute. You are right. So what? Good for you.

Student: Would people believing this theory try to claim something like quantum entanglement is the reason that you can alter an atom in a far off galaxy?

Dr. Craig: No, no. What you are talking about is when you have these experiments in which a couple of photons are sent in opposite directions so that they are so remote from each other no causal influence can go from one to the other given the finite velocity of the speed of light. And if you measure a certain type of measurement on one the other automatically takes on the correlated value even though there's no causality between them. This isn't appealing to anything like that. Indeed what the objector here is proposing is that there is no correlation between the events in C* and the events in C. It's just a brute fact that if the agent P were in C* he would choose differently. It's not because of any quantum entanglement.

Student: Would it be correct to say that this is more or less just Dean Zimmerman's anti-Molinist argument?

Dr. Craig: Yes. What you are asking is is this an objection that has been pressed by Dean Zimmerman, a Christian philosopher at Rutgers. Also by Robin Collins. I'm going to be looking more at Collin's version than Zimmerman's.

Student: It's worth mentioning that there's a book called Molinism and “Divine Voodoo Worlds” by Randy Everist. You'll see him at ETS. He has an objection to this, and he calls it the “so-what” objection.

Dr. Craig: Oh, OK! Is this the book by Lang?

Student: No, this is by Randy Everist. He wrote it for his Masters. He'll be doing a lecture at ETS.

Dr. Craig: OK. And the title of the book?

Student: Molinism and “Divine Voodoo Worlds”

Dr. Craig: Voodoo! Oh, OK. Yeah, Zimmerman uses that word “voodoo” in this respect. OK, good, thank you.


Let's go ahead then and give some assessment of the objection.

I think at the very, very start we would do well to just pause a moment and ask ourselves even if this objection is sound, so what? It would do nothing, I think, to undermine the Molinist account of providence as such. In particular it wouldn't do anything to undermine the freedom of the person in whatever circumstances he finds himself because their choices are in every case causally undetermined. If a choice is freely made in C then that choice will be made in C* as well in which some causally irrelevant event takes place that's not included in C. So if, when God places a person P in circumstances C, P’s freedom is not compromised then it would not be compromised by the mere fact that if God had put P in C* instead then P would have chosen differently.

Rather (and here's, I think, the answer to the so-what question), what the objection threatens to undermine, I think, is the theological utility of the doctrine of middle knowledge. If the objection is correct then the distinction between broadly logically possible worlds and worlds that are feasible for God becomes inconsequential because God can bring about whatever creaturely free choices he wants to without detriment to his providential plan. Therefore the Molinist account of providence would be useless in explaining why apparently less than optimal states of affairs obtain in the world (for example, evil and sin in the world). The Molinist would have to say (like the Calvinist) that these less than optimal states of affairs are just God's perfect will – that it is his will that a world of sin and evil and suffering exists. So the Molinist account then would not be useful in explaining why these less than optimal states of affairs obtain. Like the Calvinist, he'd have to just say it's God's will. But the Molinist account would still enjoy the considerable advantage of making room for creaturely freedom. It just wouldn't be of much help in explaining, for example, why evil and suffering exist. But it would still provide a reconciliation of divine sovereignty with genuine human freedom.

But is the objection successful? Well, I think not because it's predicated upon a number of questionable assumptions. Let me identify some of these assumptions.

First, the objection seems to assume that we're dealing here with events that are distributed randomly by pure chance across the sets of circumstances. But it is not by chance that P would choose some action A in circumstances C. Rather, P acts for reasons in those circumstances. So we shouldn't think of P’s choices as randomly distributed across the possible sets of circumstances. On the contrary, free choices are indeterminate events which are done for reasons. I think that gives good grounds for thinking that P’s choices in C would not vary wildly if he were placed in C* instead. This is the point that someone earlier was just making. I think the reason would be is that for P circumstances C* are indistinguishable from C, and therefore his reasons for choosing A in C would also apply for choosing A in C* as well. Here's empirical evidence for that claim. Just ask the relevant person: Would you have chosen to do differently if a molecule in Alpha Centauri had been moved slightly to the left? I'm sure he will say, No, I would have chosen exactly the same way. I think that the objection fails because it assumes that our choices are like random events that do not occur for reasons when in fact free choices, though indeterminate, are done for reasons – reasons that if they hold in C would also hold in C*.

Secondly, the objection assumes that the circumstances are unlimited. But if you think about it, this is far from obvious. It's universally agreed, for example, that events which are later than P’s time of choosing ought not to be considered in the circumstances. When we say that P would choose something in circumstances C we mean at a time T. And events which are in the future aren’t considered to be part of those circumstances. Why not? Simply because events which are future at the time of P’s choice can have no influence on P’s situation at T, and therefore they are just irrelevant to P’s decision. But if you think about it, events which are sufficiently distant from P are just as irrelevant as future events if those events are simultaneous with or even earlier than time T.

If we imagine the event of P’s choice, from P the fastest causal signals that can go out into the future will be a light signal. That's a limiting velocity for physical influences. The events which can causally impact P’s choice will lie within a cone of light signals from prior events. Events which have what's called a space-like separation from P cannot have any impact on P's choice because they cannot send an influence that would affect P's choice. Any event that can affect P's choice will have to lie either inside or on this past light cone of P's decision at that point. But events which are space-like separated from P are just completely irrelevant to P’s choice. They are as irrelevant as events that lie in P's absolute future. In fact, according to the special theory of relativity, for some observers at this point (those moving at near-light speeds), events in this space-like region are future for the observer at the place of P's choice. I think this suggests that if we do not consider future events to be part of the circumstances in which P decides, neither should we consider events having a space-like separation from P. Rather, the only events that should be included in C are events which are in or on P's past light cone. In that case the substitution of circumstances C* for C will not affect the counterfactual in question since only the circumstances C in this light cone are relevant. What happens over here in some distant event in Alpha Centauri is simply irrelevant. It's not part of the circumstances C. So the circumstances in which a free agent makes a choice are a lot more limited than what the objector seems to assume. They're really limited only to those that are in this past light cone, and when you think that the universe had a beginning this past light cone is not infinite. It's only about 13 or 14 billion years old. So this is a finite region that should be considered as part of those circumstances.

The third point is the objection assumes that imperceptible events included in P's past light cone can be altered without significant effects upon P's situation at T. The assumption would have to be that you could alter certain events in this past light cone and it really wouldn't make much difference (if any) to P's decision at time T. But the lesson of both quantum physics and of chaos theory have taught us that that assumption is false. Both of these show how imperceptible tiny changes can make enormous differences on the macro level. For example, the imperceptible quantum indeterminacy and the position of a cue ball on a billiard table is such that after only a dozen shots that indeterminacy will be magnified such that the ball could be anywhere on the table. Only a dozen shots are required to magnify that imperceptible indeterminacy in the ball’s location to be as large as the whole billiard table. You wouldn't have any idea where it would be. Similarly, chaotic systems vary unpredictably with the smallest perturbation. You alter just a little thing (a butterfly's wings) and it can upset macroscopic events like hurricanes over the southern United States. Now, certainly it's true that some events in the past light cone – events that are very far away and very recent – might be alterable without a great impact upon P's decision at time T. But these will be, after all, finite in number, and it will be pure speculation as to whether or not manipulation of those events would lead P to freely choose differently than he will in fact choose. The available alterations may fall far short of what would be necessary for God to bring about any desired free choice on P’s part.

Finally, the fourth objection to this view is that the objection assumes that God's concern is with P's choice alone – just with this one choice. But God's concern is with the whole history of free creatures on into eternity future. Even if substituting C* for C were sufficient for bringing about a different free choice of P at T, that says nothing about the feasibility of actualizing a whole world of free creatures on into the infinite future – a task which I think plausibly would involve infinite complexity.

So it's not at all implausible that the difference between broadly logically possible worlds and worlds that are feasible for God would then become very significant and very dramatic.


Student: What sort of defenses do the people who give this kind of objection . . . surely they identified these sorts of assumptions and then give defenses for them, or do they?

Dr. Craig: I haven't seen . . . well, Zimmerman has responded to me. But his view is based upon a really peculiar understanding of human freedom that I think is just incorrect. As long as we think of libertarian freedom in terms of being free from external causal constraints, I think on this account a person's decisions are free in both C* and in C and therefore the claim by someone like Zimmerman that this destroys human freedom I think fails.

Student: It is also worth mentioning that in that book that I mentioned earlier, Randy also mentions the agent-causal-libertarian account, and I think Zimmerman has even conceded at this point that his account doesn't work.

Dr. Craig: OK. Good. Thank you.


This is by far the most sophisticated objection to a divine middle knowledge or Molinist account of providence, and I'm persuaded that it doesn't work. Therefore, I think that the Molinist reconciliation of divine sovereignty and human freedom is the superior view of the three alternatives.

Not only does Molinism enjoy these theoretical advantages, but this past week it's come to my attention that Molinism also has significant practical advantages. I was sent a blog by one of our Reasonable Faith chapter directors entitled “Molinism Saves Marriages.”[1] He tells this story.

. . . my wife and I used to fight about a certain subject all too often. . . . This issue that caused such tension between the two of us was regarding a theological issue: God’s sovereignty and human freedom. That is to say, my wife held to a theological view known as Arminianism . . . — and I was a staunch Calvinist. . . .

Churches split over this debate and our fledgling relationship was on the rocks because of it too. This topic would arise constantly and it seemed as if every night we would go get our Bibles, not to grow closer to God, but to use them as weapons against each other! She had her proof texts and I had mine. . . .

We would have yelling fights that would lead to tears. In retrospect, I was a total jerk, but I figured that God had causally determined me to be a jerk and that He was forcing me to act that way for some reason, unbeknownst to me. . . .

We eventually decided to sweep this problem under the rug and we got married despite being unequally yoked with theological differences. However, the problem remained. . . .

Deep down, however, I actually questioned her salvation, as I could not understand why God would not force her to believe the truth, as I was confident He did with me. . . .

After several years of not talking about this issue God finally intervened, at the same time I started studying Christian apologetics. Ironically, a Calvinistic pastoral colleague of mine introduced me to the work of William Lane Craig. . . .

Pretty soon, I stumbled upon the doctrines of “middle knowledge” and “Molinism.” . . .

It soon occurred to me that the dichotomy between Calvinism and Arminianism is a false one. That is to say, there is at least one other possible option to consider: Molinism. . . .

Molinism “clicked” and made sense to me. . . .

It has been several years since I parted ways with Calvin. My wife and I are both advocates of Molinism today and our marriage is flourishing. Rarely a week goes by, however, that my wife does not lovingly remind me that she was right and I was wrong about the faults of TULIP Calvinism. . . .

Bottom line: Molinism saves marriages!

Well, we can be very grateful for that practical advantage. I would just remind the author of the blog here that his wife shouldn't gloat over her triumph because she was equally wrong about Arminianism, and it was only by making a mutual compromising coming to the center that they arrived at a compatible view on which they could be happily married. So I rejoice with them and that practical outcome of Molinism.[2]


[2]                      Total Running Time: 42:51 (Copyright © 2018 William Lane Craig)