A Brief Response to William Lane Craig on Molinism

A few months back I wrote a post entitled “The Fallible God of Molinism” which was prompted by an exchange between William Lane Craig and Paul Helm. Some folk alerted me to the fact that Dr. Craig briefly responded to my argument in a recent podcast. This is what he said:

I think he’s made a misstep in his argument here. It certainly is true that in any particular freedom-permitting circumstances an agent is free to do other than as God knows he will do. But in that other world in which the agent does something different, God’s plans wouldn’t be the same. What this fellow doesn’t seem to remember is that in that world God would have different plans. God would know in that world that S would do something different in C and so in that world he would have plans for that to happen. So what he’s tried to do is keep God’s plans firm and fixed from world to world, but then vary the value of the counterfactuals, and you can’t do that. If you switch to a world in which S does not do A in C then you can’t say, “Well, in that world God’s plans are that S would do A in C.” No, no, in that world God would have different plans.

So when you switch the truth-value of the counterfactuals, you’ve got to switch the providential plans as well, because the providential plans are based upon the counterfactuals that are true in those worlds. So given that God’s plans are based upon what he knows the free agents would do, the plans will change from world to world along with the decisions of the agents. So there’s just not any problem. God’s plans never fail, and he’s not fallible.

Let me say first of all that I’m honored Dr. Craig considered it worthy of comment! But I do have a few things to say in response.

Craig is correct that I made a misstep in the original argument, as I acknowledged in reply to a comment by Greg Welty. I said at first it wasn’t relevant whether God’s decree (or “providential plan,” to use Craig’s terminology) is included in C, i.e., the circumstances in which God places the free agent S, knowing that S will do A in C. But that’s mistaken, because Molinists will say that C doesn’t include God’s decree, such that in those possible worlds in which S does not do A in C, while C is the same, God’s decree is different. (As Craig puts it, God’s providential plan is not “firm and fixed” across worlds in which S chooses in C, even though C is fixed.) The counterfactuals are different in those alternate worlds and therefore God’s decree will also be different in those worlds, since it is based on those counterfactuals.

Fair enough, but as I argued in the comments to that original post, I don’t think that observation gets the Molinist out of the woods. In the first place, it strikes me as special pleading to exclude God’s decree — which is an active decision with causal consequences, not a mere passive knowledge — from the circumstances in which S makes his free choice. The Molinist has to say, in effect, “Don’t worry, it’s possible for S not to do A in C provided we don’t include God’s decree in C, for then it would be possible for S to do other than what God has decreed.” In other words, the Molinist has to restrict C to intramundane circumstances.

But why should the Molinist be permitted to make that move? After all, the circumstances are causally connected to God’s decree (and necessarily so in the Molinist system). The circumstances obtain only because God decided that they should obtain and caused them to obtain. If we can draw the boundary line of C so as to exclude God’s decree and his actions to implement that decree (i.e., his manipulation of intramundane circumstances), why not draw the boundary line even more narrowly, so as to exclude, say, all circumstances more than five years prior to S’s choice, or all circumstances more than fifty miles away from S’s location? Is there any principled reason to draw the boundary line of C where the Molinist wants to draw it other than to save the system? Why think that it being hot and sunny in Charlotte today is a relevant element of my circumstances but God’s making it hot and sunny in Charlotte (partly in order to direct the choices  of Charlotteans) is not a relevant element?

On a related point, I think the Molinist preserves divine infallibility only by compromising his commitment to libertarian free will (LFW). One of the core intuitions behind LFW is that S’s choices are free only if they aren’t determined by factors (especially the actions of other agents) beyond S’s control. As I pointed out in my original post and in the subsequent comments, there’s a clear sense in which, on the Molinist view, God determines S’s choices. It isn’t a causal determinism, but it’s still determinism in the sense that God’s decree that S will do A (which is fixed prior to any of S’s choices) guarantees that S will do A. Given that God has decreed that S will do A, S cannot do otherwise than A.

So the Molinist has to say, “Given only that S is in C (which excludes God’s decree) S can do otherwise than A, but given that God has arranged C precisely in order that S do A, S cannot do otherwise than A.” Now why shouldn’t we consider that a gerrymandered notion of LFW? Why would someone who is committed to the idea of LFW be content with the view that his choices are only free with respect to intramundane circumstances and not free with respect to extramundane circumstances (i.e., God’s decree and active manipulation of the intramundane circumstances)? If you can live with such a diluted form of LFW, why insist on it at all?

I think this is one of the reasons why many Arminians (never mind open theists) won’t get on the Molinist train. They sense that Molinism is still a form of determinism and isn’t fully committed to LFW. In short, it still gives God too much control (even though, from the Calvinist’s perspective, it doesn’t give God enough control!).

So I stand by the remarks I made at the end of my original post, which I reproduce here:

As I see it, this tension at the center of Molinism arises because it aspires to be a deterministic indeterminism. ‘Indeterminism’ because of its commitment to libertarian free will. ‘Deterministic’ because God’s decree somehow (we know not how) determines in advance that his creatures will make certain choices. It may not be a causal determinism, but it’s deterministic nonetheless (as many non-Molinist Arminians, such as Roger Olson, can clearly see). If preordaining that S chooses A doesn’t mean predetermining that S chooses A then what on earth does it mean?

For more discussion, check out the exchanges in the comments on the original post.

22 Responses to A Brief Response to William Lane Craig on Molinism

  1. Do Calvinists get to make the same move? That is, do we get to exclude God’s decree when considering whether Calvinism allows for the relevant kind of freedom? There’s plenty of worlds where I do otherwise than I do in the actual world, it’s just that those worlds contain a different decree too. Now, perhaps the response will be that we must include all the causally relevant factors, and on Calvinism, the decree is a causally relevant factor. But don’t think this constraint is ultimately right. Here’s two reasons why: (i) I think it’s *false* that we have to only include the causally relevant factors, for there’s factors that are a threat to freedom yet don’t cause our action; and (ii), for all I know, something like Jaegwon Kim’s non-causal determinism is how God determines all that comes to pass, yet I think the non-Calvinist would still want us to include God’s decree even if it determined non-causally.

    • Nice observations, Paul!

    • Godismyjudge

      “Do Calvinists get to make the same move? ” Yes, at least it’s very common for C’s to say that God chooses one world out of the many possible worlds. We do not ask C’s to explain how the other worlds were possible given God’s choice.

      God be with you,
      Dan

  2. James,

    I tend to agree with you on your first point (“it strikes me as special pleading to exclude God’s decree”), but I still find myself balking at the second point (“there’s a clear sense in which, on the Molinist view, God determines S’s choices”). So take my lengthy comments on the second point as a plea for help (since, like you, I’m not a Molinist).

    On your first point, you’re right that Craig’s solution abstracts God’s decree from the agent’s circumstances. He has to, since otherwise the decree does end up being fallible (in that Westminster 5.2 sense :-). And it seems very problematic to characterize the decree in non-causally-relevant terms (regardless of how one might construe foreknowledge, perhaps as O-foreknowledge rather than A-foreknowledge, to use Helm’s categories).

    However, if they are able to solve this first problem, then I don’t see how your second point is a problem. So maybe what I’m saying is that your second problem is parasitic upon the first, and isn’t independent of it.

    Your second point says:

    “but it’s still determinism in the sense that God’s decree that S will do A (which is fixed prior to any of S’s choices) guarantees that S will do A. Given that God has decreed that S will do A, S cannot do otherwise than A.”

    Can’t we distinguish between God’s decreeing A and God’s causing the circumstances C in which S does A? I say this because I think the Molinist can hold that God can accomplish very different decrees by causing the same circumstances. If so, it isn’t clear that God’s causal activity can be characterized as determinism. To illustrate, assume the relevant counterfactuals are:

    CF1 If God were to cause C and put S in C, then S would freely choose A.
    CF2 If God were to cause C and put S in C, then S would freely refrain from choosing A.

    Let’s say that God knows the truth of CF1 (and thus the falsity of CF2). He goes ahead and decrees S’s choosing A. He gets the job done by actualizing C. Thus, in human history, S chooses A.

    You say: “Given that God has decreed that S will do A, S cannot do otherwise than A.” But S can do otherwise. He can refrain from A in C. Of course, if he were to refrain from A in C, that would be in the situation in which CF2 was true and CF1 was false. Notice that if CF2 were true, then God’s causing C would result in S freely refraining from choosing A.

    Given this, it’s a strange sort of ‘determinism’ in which God’s causing C is consistent with S choosing A and with S refraining from A.

    In world 1, God decrees S’s choosing A: CF1 is true, God causes C, and S chooses A.
    In world 2, God decrees S’s refraining from A: CF2 is true, God causes C, and S refrains from A.

    So, S can do otherwise in C: he can choose A, and he can refrain from choosing A. And in both scenarios, God causes C. It follows that God’s causing C doesn’t determine S to choose A. There’s a possible world in which God’s causing C doesn’t have that result. God’s decree that S will do A does not determine S to choose A, because all that God’s decree commits him to, causally speaking, is causing C, and we’ve seen that the causing of C doesn’t determine S to choose A.

    Of course, my assessment of your second point presupposes that the Molinist can make a meaningful distinction between God’s decree that S chooses A, and God’s causing C. In particular, it presupposes that we can somehow abstract God’s decree from the causally relevant circumstances in which S makes his choice. And I tend to agree with your first point that this abstraction seems quite arbitrary.

    [Sorry that my italics got stripped out of this copy-and-paste from my text editor. I'm just too lazy to hardcode the HTML tags. I don't think it affects the sense. Indeed, I tend to overuse italics so it's just as well :-)]

    • Greg,

      Yeah, I overuse italics too. I even used them in the transcript of Craig’s oral response! (But he did emphasize those words.)

      As to your concerns about my second point: I think what the Molinist needs to do is to drive a wedge between God’s acting (rather than decreeing) and God’s causing. When God acts, God acts for reasons and with intentions. It makes no sense to speak of an action in a way abstracted from those reasons and intentions. So God acts to bring about C in order that S does A. But then I don’t think it’s plausible to separate God’s causing C from God’s acting to bring about C. Isn’t the cause of C just God’s action (or God, acting)? What would it mean to say that God’s causing C is consistent with S refraining from A, yet God’s acting to bring about C is not consistent with S refraining from A?

      In any event, I just can’t see how one can plausibly say that God infallibly brings about that S does A (which the Molinist does want to say) while denying that in any sense God determines that S does A. As I noted in my post, Craig is happy to use the language of ‘preordaining’. How is that different from ‘predetermining’?

  3. Pingback: James Anderson Responds to William Lane Craig on Molinism | Remington's

  4. James,

    You don’t seem to be disputing that God’s causing C isn’t God’s determining S to choose A. Rather, you’re saying that God’s acting to bring about C comes packaged with God’s reasons for and intentions in doing this: he wants to ensure that S chooses A. So even if the mere causing of C doesn’t determine S to choose A (God’s causing C is consistent with S refraining from A), once you add divine reasons and intentions you do get the equivalent of such determinism.

    Couldn’t the Molinist wonder, though, what is causally relevant about reasons and intentions? Are these causes, over and above God’s causing C? And if not, how can adding a couple of non-causal factors suddenly transform a non-deterministic sequence into a deterministic one?

    (Still, Paul’s point might apply here: the sequence could rise to the threshold of non-causal determinism, even if it’s not causal determinism. And, of course, I’m quite partial to analogies to causal determinism which are such that the Molinist cannot meaningfully distinguish between acceptable Molinist ‘determinism’ and unacceptable Calvinist determinism!)

    • Greg,

      No, that’s not quite right. Sorry if I haven’t been sufficiently clear. I accept that on the Molinist view, God’s causing C doesn’t causally determine S to choose A. After all, C doesn’t cause S to choose A, so simply adding God as the cause of C isn’t going to generate a causal determinism.

      But I still maintain that there’s a non-causal determinism at work here, because God’s acting to bring about C so that S chooses A entails that S chooses A (at least in conjunction with divine infallibility). While C alone may be consistent with (a) S choosing A or (b) S not choosing A, God’s action is only consistent with (a). And I’ve suggested that the Molinist cannot cogently tease apart God’s causing C and God’s acting to bring about C in the way that he needs to.

      What is causally relevant about reasons and intentions? They’re intrinsic to God’s actions, which just are the cause of C (or, if you prefer, God himself is the cause of C in so acting).

      Another way to put my point: LFW ought to mean that both S choosing A and S not choosing A are consistent with the totality of actions (of any agent) that precede S’s choice. But on Molinism, that isn’t the case.

      • Godismyjudge

        Dr. Anderson,

        You said: But I still maintain that there’s a non-causal determinism at work here, because God’s acting to bring about C so that S chooses A entails that S chooses A (at least in conjunction with divine infallibility). While C alone may be consistent with (a) S choosing A or (b) S not choosing A, God’s action is only consistent with (a).

        I do think this is the same as the standard foreknowledge argument. It’s no less true that: God’s foreknowing that S chooses A in C entails that S chooses A (at least in conjunction with divine infallibility). An it meets the same standard answer – i.e. there’s a difference between the necessity of the consequence and the necessity of the thing consequent. Put anther way, Molinists are OK with the idea that the proposition “S will A in C” is a logically necessary conclussion given God’s decree and foreknowledge, so long as S has the ability to act in such a way such that if he did, he would be the basis of truth of the proposition S nonAed in C.

        God be with you,
        Dan

        • Good to hear from you, Dan. Thanks for the comments.

          Since I’m not arguing that Molinism renders our choices logically necessary, I fail to see the relevance of your point about distinguishing the necessity of the consequence from the necessity of the consequent. Moreover, I’m not aware of any philosophers of religion today who think that the freedom-foreknowledge dilemma is resolved simply by observing that distinction. See, e.g., Zagzebski’s article in SEP.

          In any event, I don’t think the argument I’m making is equivalent to the foreknowledge argument. As I’ve pointed out several times now, the significant difference is that on Molinism God is acting so as to guarantee that S will do A in C. That’s a greater challenge to full-blooded LFW than God merely passively (fore)knowing that S will do A in C. Surely if S has LFW, S is free either to do A in C or to refrain from A in C regardless of how any other agent has acted in the past (or in eternity). (I’m ignoring here complexities due to will-setting, etc.) But that’s not the case on Molinism. Pointing out that there are possible worlds in which S refrains from A in C, but in which God has acted otherwise than in this world, does nothing to distinguish Molinism from Calvinism, because (as Paul has pointed out many times) Calvinists can say the very same — and not many folk think Calvinism is consistent with LFW. :)

          I don’t know what you mean by saying that “S would be the basis of truth of the proposition S nonAed in C.” It sounds like you’re saying that everything is fine so long as S is the truth-maker for counterfactuals about S (remember, we’re talking about S having LFW in the actual world, the world in which S does A in C). But I’m sure you’re familiar with the objections to the notion that creatures are truth-makers for the counterfactuals of freedom which God knows prior to creation.

          • Godismyjudge

            Dr. Anderson,

            Thank you for the kind response.

            You said: I’m not aware of any philosophers of religion today who think that the freedom-foreknowledge dilemma is resolved simply by observing that distinction. See, e.g., Zagzebski’s article in SEP.

            Molina himself did at times – and I think Freddoso picked up on it. At least, he embraced a radical form of Occhamism in his first two articles on accidental necessity and then backed away from them for a more simple answer in his intro to Concordia. Different foreknowledge arguments require different responses, so sometimes the consequence/consequent distinction is required and other times Occhamism is needed. Your stress on entailment (with your denial that you were arguing for causal determinism) led me to bring up the consequence/consequent distinction and here’s how I would apply it to the core of Zagzebski’s argument:

            (3) It is now-necessary that yesterday God believed T. [1, 2]
            (4) Necessarily, if yesterday God believed T, then T. [Definition of “infallibility”]
            (5) If p is now-necessary, and necessarily (p → q), then q is now-necessary. [Transfer of Necessity Principle]
            (6) So it is now-necessary that T. [3,4,5]

            http://plato.stanford.edu/entries/free-will-foreknowledge/

            3, 4 and 5 are all true, but 6 does not follow. 3 is based on the necessity of the past, which is based in the denial of retro causation. The past is causally necessary, because no cause now can undo the past.

            4 and 5 are not about causal necessity but rather logical necessity. They deal with logical entailment. They deal with propositions and truths, not events or the basis of truths. So 6 does not follow from 3-5 because it equivocates logical necessity in 4 & 5 (necessity of the consequence) and causal necessity in 3 (necessity of the thing consequent). No Occhamism or Boethianism needed for this one. But Zagzebski’s argument can be reformed in a way that requires Occhamism and Fisher has done the best job I have seen at arguing against Occhamism.

            If you are not arguing for logical necessity or causal necessity, just what type of necessity is God supposedly imposing on us by guaranteeing the future?

            You said: I don’t know what you mean by saying that “S would be the basis of truth of the proposition S nonAed in C.” It sounds like you’re saying that everything is fine so long as S is the truth-maker for counterfactuals about S (remember, we’re talking about S having LFW in the actual world, the world in which S does A in C).

            I had spoken of the ability to do that, not doing it. So in this world, if you hover your mouse over the agent, right click and hit properties, it will list the ability, given all preceding causes, to cause nonA. That’s in this world and at least causal deterministic Calvinists disagree with that.

            You said: But I’m sure you’re familiar with the objections to the notion that creatures are truth-makers for the counterfactuals of freedom which God knows prior to creation.

            Yes, and I am more comfortable with the responses that have been given than I am with the alternatives.

            God be with you,
            Dan

          • Dan,

            I’m perplexed, because 6 follows deductively from 3, 4, and 5; more precisely, 6 follows from 3 and 4 by way of the principle expressed in 5. There’s no equivocation in Zagzebski’s formulation of the argument, because she takes care to distinguish temporal/accidental necessity from logical necessity.

            “If you are not arguing for logical necessity or causal necessity, just what type of necessity is God supposedly imposing on us by guaranteeing the future?”

            It’s not an absolute logical necessity (hence my earlier response) but rather a relative logical necessity, i.e., relative to God’s prior action to arrange circumstances. By way of comparison, think of the initial act of creation. Necessarily, if God says “Let there be light,” there will be light. The existence of light isn’t an absolute logical necessity, but it is a relative logical necessity given God’s creative declaration. Likewise, S doing A in C is a relative logical necessity given God’s action to arrange circumstances. There is no possible world in which the latter is true but the former is not, for the reason I’ve already given: God’s actions include his intentions.

            One reason I think this is distinct from the freedom-foreknowledge problem is that the Ockhamist solution is even less plausible here. How could God’s foreordaining actions be “soft facts”?

            “I had spoken of the ability to do that, not doing it. So in this world, if you hover your mouse over the agent, right click and hit properties, it will list the ability, given all preceding causes, to cause nonA. That’s in this world and at least causal deterministic Calvinists disagree with that.”

            The Calvinist compatibilist can say that there are meaningful senses in which S “has the ability” to do A or not to do A, even given causal determinism. So again, this does nothing to shed light on what Molinism has to offer here. You need to spell out “has the ability” in terms consistent with the infallibility of divine foreordination. What you can’t do is say that there are possible worlds in which God acts so as to guarantee that S does A in C, but S doesn’t do A in C. I want to know what it means for S to have LFW given that constraint. I refer you back to the second part of my post above.

          • Godismyjudge

            Dr. Anderson,

            You said: There’s no equivocation in Zagzebski’s formulation of the argument, because she takes care to distinguish temporal/accidental necessity from logical necessity.

            Here’s the equivocation I see:
            (3) It is now-[accidentally] necessary that yesterday God believed T. [1, 2]
            (4) [Logically] Necessarily, if yesterday God believed T, then T. [Definition of “infallibility”]
            (5) If p is now-[logically] necessary, and [logically] necessarily (p → q), then q is now-[logically] necessary. [Transfer of Necessity Principle]
            (6) So it is now-[accidentally or logically?] necessary that T. [3,4,5]

            Is the conclusion about accidental necessity or logical necessity? If 6 is about logical necessity (i.e. the necessity of the consequence), we grant it, but it does not threaten LFW. The same seems to apply to what you call relative logical necessity. If 6 about accidental necessity (which I argued above is a form of causal necessity), that would be a problem for LFW. But then the syllogism is invalid, because accidental necessity does not pass from 3 to 6, since it’s not in 4 or 5 (and 4 or 5 cannot be reworked to include accidental necessity in a way in which we would accept). So the syllogism commits and equivocation fallacy not that different than: A feather is light. What is light cannot be dark. Therefore, a feather cannot be dark.

            You said: “By way of comparison, think of the initial act of creation. Necessarily, if God says “Let there be light,” there will be light. The existence of light isn’t an absolute logical necessity, but it is a relative logical necessity given God’s creative declaration.”

            Setting aside any type of causation in creation (the type we find in Genesis 2:9), yes, God’s decree implies the events He decrees will happen.

            You said: There is no possible world in which the latter is true but the former is not, for the reason I’ve already given: God’s actions include his intentions.

            Agreed on both fronts – but I will note that on Molinism, God’s intentions (or His decree) does not alter the inbound total causal contribution to a person’s choice. So there are possible worlds where the person experiences the same causal influence and they choose differently. This is in part because the possibility of God’s intentions are within His natural knowledge and in part because God’s decree’s unfolds in accord with His natural and middle knowledge.

            You asked: How could God’s foreordaining actions be “soft facts”?

            As Dr. Craig said, if God’s middle knowledge had been different, His plan (and foreknowledge) would have been different. Fisher criticizes Occhamism’s calling God’s foreknowledge a soft fact as a special pleading, but Molinism provides the justification, since God’s foreknowledge depends on His decree, which depends on middle knowledge.

            You said: The Calvinist compatibilist can say that there are meaningful senses in which S “has the ability” to do A or not to do A, even given causal determinism.

            Classic compatiblism, epistemic compatiblism, and dispositional compatibilism all fall short of libertarian freedom on substance if not semantics. So long as Molinism is consistent with LFW, it has something to offer.

            You said: What you can’t do is say that there are possible worlds in which God acts so as to guarantee that S does A in C, but S doesn’t do A in C. I want to know what it means for S to have LFW given that constraint.

            While Open Theists can say all past facts are hard facts, Occhamist/Molinists cannot. But we can say, pas causal determinism, is that holding the causal influence on the agent the same, the agent can cause A or not. We can also say the logical necessity is only the necessity of the consequence rather than an absolute logical necessity.

            God be with you,
            Dan

          • Dan,

            By “now-necessity” Zagzebski means accidental necessity and by “necessity” she means logical necessity. So again, there is no equivocation in the argument. You’re misinterpreting premise 5.

            You have a habit of changing the subject. For example, I asked a question about God’s foreordaining actions and you gave an (irrelevant) answer about God’s foreknowledge, when I have already emphasized the difference between the two.

            With respect, I don’t see this discussion heading in a productive direction, so I have nothing further to say.

  5. Adam Omelianchuk

    What kind of determinism is it that denies that God cannot get precisely what he wants? I would think Molinists would be comfortable with that, because they are concerned with avoiding theological determinism. Theological determinism is true in a world, w, if and only if God is able to get whatever he wants in w (we are assuming that God never wants things that logically impossible). But Molinism denies precisely this claim: God cannot actualize every world that is possible. That, if anything, is contrary to a view that puts all things under God’s control, which is precisely what theological determinism requires. But Molinists are okay with a high degree of God’s control; they just deny that all things are under God’s control, for example, the truth-value of counterfactuals of creaturely freedoms.

    • Adam Omelianchuk

      Sorry, the previous comment was written hastily. Here’s what I am trying to say:

      What kind of determinism is it that denies that God can get precisely what he wants? I would think Molinists would be comfortable with that, because they are concerned with avoiding theological determinism. Theological determinism is true in a world, w, if and only if God is able to get whatever he wants in w (we are assuming that God never wants things that logically impossible). But Molinism denies that theological determinism is true in every possible world (because of LFW); furthermore, they deny that God can actualize every world that is possible (i.e. worlds where every creature with LFW does what is right without exception). That, if anything, is contrary to a view that puts all things under God’s control, which is precisely what theological determinism requires. But Molinists are okay with a high degree of God’s control; they just deny that all things are under God’s control, for example, the truth-value of counterfactuals of creaturely freedoms.

      • Hi Adam, when you say on theological determinism “God is able to get whatever he wants”, do you mean that each and every thing that exists, occurs, obtains, etc., is something God wanted in itself, or, that God can get whatever plan *taken as a whole* that he wants—that is, there are some thing he “wants” only in the sense that they contribute to the whole God wants. As far as I can tell, TD does not entail the former reading, but the latter reading is consistent with Molinism (indeed, that’s it’s main selling point!), which would make it a species of theological determinism, on your parsing.

        • Adam Omelianchuk

          Hi Paul! I am thinking of the latter sense, though I would disagree with you about it being consistent with Molinism. Presumably, God wants a world that resembles our own in the relevant respects where everyone always and without exception freely obeys his will. But, on Molinism, it is possible that the CCFs don’t allow for the creation of such a world to be feasible. The upshot is that there are worlds that God desires (say, for example where everyone is saved), that are not feasible for him to create. Thus, God cannot get what he wants. That is the main reason a lot of Christians reject it, I gather.

  6. Hi Adam, I would have thought that, at best, on Molinism we’d want to say that God wants a universalist world if he can get one, and if he can’t, he doesn’t want or desire what he can’t get. He might ‘wish’ (as it were) that the actually true CCFs had been difference such that he could have got a universalist world, but given the way they are he doesn’t, as it were, cry over spilt milk. A poker player might wish he had been dealt a different hand, but having been dealt the hand he has, he plays the cards he’s been dealt, and he doesn’t pine away for a hand he can’t get. Do you not see it this way?

    That aside, it may seem to be the case to you that the Molinist God wants a universalist world, but I’m not sure that’s necessarily the case. For all I know, it’s even true on Molinism that the incarnation and redemption of sinners is an outweighing good, such that the Molinist God doesn’t wish everyone always did right. For all I know, the Molinist God thinks displaying the full range of his attributes, and thus exhibiting justice on some truly deserving sinners, is an outweighing good, and thus universalist worlds are morally inferior. So the Molinist God gets everything he wants in this world. Neither of these seem impossible on Molinism, so for all I know, the Molinist God truly does get everything (in our agreed upon sense, i.e., the plan as a whole) he wants in our world. Do you have an argument for why, necessarily, our world isn’t exactly the kind of world (taken as a whole) that God wants?

    • Sorry for the addition: You wrote,

      “Presumably, God wants a world that resembles our own in the relevant respects where everyone always and without exception freely obeys his will.”

      And presumably since these worlds are not feasible, then it’s a contingent truth (given the luck of the CCF draw) that a world resembling ours where everyone always and without exception freely obeys his will, isn’t actualizable. Had the CCFs turned out differently, such a world would have been feasible. Call this world,U. Now, you grant that U is exactly what God wants. U is also possible, in that the relevant CCFs needed to actualize U could have been true. Thus, it’s possible that, on Molinism, God gets “exactly what he wants”, viz., U, and U is such a world where both these claims are true:

      1. Theological determinism is true at U (your def. of TD).

      2. Every creature *libertarian* freely does right. (def. of U)

      But the conjunction of (1) and (2) entail a contradiction.

      So, do you find it necessary to endorse TWD? If so, as I’m sure you know, you need to address the arguments that Plantinga has not shown that TWD is logically possible, or arguments by Otte that TWD is logically impossible.

      So it seems to me that either your definition of TD needs to go, since it entails a contradiction, or you are forced to claim worlds where everyone does right are impossible. But then, we have God desiring not just an infeasible state of affairs, but an impossible one. This seems problematic to my mind. Thoughts?

  7. Adam Omelianchuk

    Good thoughts as always, Paul. I think you are right: the way I define TD has to go. Let me try to revise it this way:

    (TD): For every possible world, w, theological determinism is true at w if and only if God gets precisely what he wants in w by virtue of controlling everything in w (God controls everything in w =def: there is no y such that y obtains in w, and y’s obtaining is not up to God, nor is there a proposition P such that P’s truth-value in w is not up to God).

    As for TWD, I am aware of Pruss’s argument against it, and I think it is a good one (if he assumes a modified divine command theory). On his view, God can create a morally perfect world where only one morally significant decision is made, and knowing what Eve will do in the garden, God actualizes a world where he doesn’t forbid the eating of the apple, but commands something else, which she satisfies (dancing a jig, in Pruss’s case). Once she obeys, the world ends. Thus we have a morally perfect world, and God should prefer creating that world to our world, because God should always value a morally perfect world to an imperfect one. I agree with you in rejecting that assumption, because it doesn’t seem that a world like that is better than ours. And I can agree that, for all I know, God chose to make a world like ours, because it contains unsurpassable goods like the Incarnation and Atonement—goods that could not be had in a morally perfect world. I can’t say if such a world is better than one that is minimally different from our world where everyone, like you or I, always obeys God without exception (I doubt that it would be, but for I know, maybe not). Even so, wouldn’t it be better if everyone was saved by the atonement than not? No doubt this cries out for more development, but if this is the case, the Molinist could retain a free will-style theodicy while abandoning the thesis that transworld depravity is possibly true.

    The reason I reject the Edwards-style “display of wrath to the elect” theodicy is that it, to my mind, it contradicts Scripture’s revelation of God genuinely desiring everyone’s salvation and God’s displeasure in the destruction of the wicked.

    I’m sure these thoughts are inadequate, but I await your critique; you always seem to push me to think harder about my position and I am grateful for that.

  8. Godismyjudge

    Dr. Anderson,

    You said:
    —————————–
    Molinists will say that C doesn’t include God’s decree, such that in those possible worlds in which S does not do A in C, while C is the same, God’s decree is different. (As Craig puts it, God’s providential plan is not “firm and fixed” across worlds in which S chooses in C, even though C is fixed.)… …In the first place, it strikes me as special pleading to exclude God’s decree — which is an active decision with causal consequences, not a mere passive knowledge — from the circumstances in which S makes his free choice.

    Later you said:
    It isn’t a causal determinism, but it’s still determinism in the sense that God’s decree that S will do A (which is fixed prior to any of S’s choices) guarantees that S will do A. Given that God has decreed that S will do A, S cannot do otherwise than A.
    ———————————–
    We can differentiate God’s decree from His causal impact. He decrees to create and He actually creates. He decrees the events of the cross and He actually arranges the events of the cross. Let’s say God is deciding between world’s A & B. God would have causal influence either in A or B, but let’s say He chooses A over B. Then He exerts the causal influence He planned for A. I make this distinction to specify what Molinists mean when we exclude God’s decree. We exclude God’s decision of A over B, but not God’s causal influence in A.

    This exclusion is not a special pleading to escape your argument – it’s part of the very definition of Molinism. Molinism provides a logical order in that each step helps explain the next. Let’s walk through Molinism’s logical order and how it applies to your argument:

    1 Natural knowledge is God’s pre-volitional knowledge of all necessary truths – it’s the range of what can happen. Natural knowledge arises from God’s very nature rather than His choice or decree. As it applies to your argument, God’s causal influence in any world is part of His natural knowledge, because it’s what God can do. God’s nature defines what God can do, what He can enable us to do and what we can do via His enablement and what He can do about what we can do. So natural knowledge includes the possibility of God’s creation and sustainment of the world, as well as His general and special providence.

    2 Middle knowledge is God’s pre-volitional knowledge of what we would choose in any setting. Middle knowledge limits the range of events God can actualize. Let’s say God can create and sustain Bob and arrange for him to be in an ice cream shop today and God enables Bob to choose chocolate or vanilla. Via middle knowledge God knows Bob would choose chocolate. God still can choose for Bob not to be in the ice cream shop, but He cannot make Bob choose vanilla. A determined free event is a contradiction and logically impossible. As it applies to your argument, this limiting of events God can actualize is strong evidence of God granting us libertarian freedom. Also, given natural knowledge precedes and helps explain middle knowledge, God’s causal activity is included in middle knowledge. Rather, God’s actual decision or decree is excluded.

    Next comes God’s decree which is God’s choice of what will happen and how it will happen. God decides for the events to happen exactly as He knew them via natural and middle knowledge. In natural knowledge, God’s possible causal influence on Bob was such that Bob could still choose chocolate or vanilla, so Bob can still choose chocolate or vanilla. So God’s decree provides no additional causal influence on Bob (or remove any ability from Bob) over and above natural and middle knowledge. As it applies to your argument, you are asking us to explain middle knowledge in light of God’s decree. But that inverts Molinism’s logical order – God’s decree logically follows and is explained by middle knowledge, not the other way around. Nor is the logical order of Molinism a special pleading – it’s not as if Molinism was defined to answer your argument against Molinism. Also, while God’s decree does in some sense guarantee the future, that guarantee does not remove our ability to do otherwise than what God decrees, because God’s guarantee matches how things unfold in His natural and middle knowledge and His guarantee experts no causal influence on the event.

    Finally comes God’s free knowledge which is God’s post-volitional knowledge of what will happen. God not only knows what will happen, but how it will happen (i.e. God knows Bob will choose chocolate). While God’s foreknowledge is certain and infallible, that does not necessitated the events He knows. So Bob can but will not choose vanilla. As it applies to your argument, some of the force of your argument applies to simple foreknowledge as well as middle knowledge. Just as it is wrong to conclude that given God has decreed that S will do A, that S cannot do otherwise than A, it is equally wrong to conclude given God foreknows S will do A, that S cannot do otherwise than A. What follows is that S will not do otherwise than A and that the combination of “God decreed S will do A” and “S does otherwise than A” is logically impossible. But we cannot get from there to S cannot do otherwise than A without a division fallacy similar to reasoning from my lifting two 150 lbs weights at the same time is impossible, therefore I cannot lift one of them. It’s the combination that’s impossible, not each element.

    God be with you,
    Dan