Books Providence Theology Proper

A Molinist model of God’s providence








In Chapter 2 of  Four Views on Divine Providence, William Lane Craig presents a Molinist perspective.

A restatement of William Lane Craig’s model of divine providence

William Lane Craig begins his presentation by noting that Christian theology has traditionally affirmed God’s knowledge of conditional future contingents, what philosophers often call “counterfactuals.” Craig defines these as “conditional statements in the subjunctive mood, such as “If I were rich, I would buy a Mercedes” (79). The antecedent or consequent clause in such statements is typically contrary to fact, which is why the statements are dubbed “counterfactual,” but sometimes they are true. Craig might become rich and buy a Mercedes, for instance. Such statements are common in our ordinary language, and we make life-and-death decisions on the basis of the presumed truth of counterfactual statements (80).

Historical background

Although Christian theologians traditionally believed that God knows true counterfactuals, they disagreed about when, logically (not chronologically) speaking, God has this hypothetical knowledge. Catholic theologians of the Dominican order held that “God’s hypothetical knowledge is logically subsequent to his decree to create a certain world. They maintained that in decreeing that a particular world exist, God also decreed which counterfactual statements are true” (81).

Luis de Molina and fellow Jesuit theologians maintained that “God’s hypothetical knowledge of creaturely free decisions is logically prior to his creative decree,” and they  charged that the Dominicans effectively “obliterated human freedom by making counterfactual truths about creaturely choices a consequence of God’s decree” (81). Molinists posit that, by knowing how a libertarianly free creature would act in particular circumstances, within a particular world, God is able to bring about his ultimate purposes by decreeing the world in which creatures would do freely (in the libertarian sense) what fits within God’s purpose. Molinists called “middle knowledge” the logical moment at which God knows how a particular creature would act in particular circumstances, in a particular world. There may have been worlds in which Peter would affirm Christ and worlds in which Peter would deny Christ, though the circumstances were the same in all of them. If the possible world in which Peter freely affirms Christ does not fit within God’s ultimate purpose, then choice of that world is not feasible for God.

Craig’s arguments for Molinism

Craig asserts that it is not difficult to show biblically that God possesses hypothetical knowledge but, since this was agreed upon by Christian theologians until recent times, he cites only John 18:36, where Jesus tells Pilate: “If my kingship were of this world, my servants would fight.” What Scripture does not indicate, however, is whether God has this hypothetical knowledge logically prior to or after his creative decree.

Theological arguments for the Molinist perspective are therefore claimed to be the strongest ones, and Craig considers middle knowledge to be “one of the most fruitful theological concepts ever conceived” (84). For the purposes of this essay, its fruitfulness need only be demonstrated in regard to divine foreknowledge and divine providence, though Craig lists numerous doctrines which are illumined by the concept. Middle knowledge offers a conceptualist model of God’s cognition, rather than a perceptualist model. Since the future does not exist, God does not know the future by “foreseeing” it. Rather, “God’s knowledge is self-contained; it is more like a mind’s knowledge of innate ideas” (85). It is based in God’s middle knowledge and his knowledge of his own decree.

The fruitfulness of the Molinist account of providence is cited in reference to Acts, 2:23 and 4:27-28, where God’s sovereignty over human affairs is clearly asserted, the conspiracy to crucify Jesus having “happened by God’s plan based on his foreknowledge and foreordination” (86). On the Molinist account, “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, favoring the release of Barabbas; what Herod, if king, would freely do in reaction to Jesus and to Pilate’s plea to judge him; and what Pilate himself, if holding the prefecture of Palestine in AD 30, would freely do under the pressure of the Jewish leaders and the crowd.” Knowing all this, “God decreed to create just those circumstances and just those people who would freely do what God willed to happen” (86).

Proponents of the openness model admit that a strong doctrine of divine providence is impossible without middle knowledge. They cannot explain the coalescence of human freedom and divine sovereignty. Although “openness theologians are fond of comparing God to a Grand Master in chess,” God is not so brilliant that he could “achieve the universal salvation he desired,” and he “regretted having created man” (87). But, if  Molinism is correct, “it may be that a world having more saved but less damned than the actual world was not feasible for God” (88). Whereas the open God might “churn up a lot of unforeseen, unnecessary, and pointless suffering,” as he plays the chess game of the world, the God of Molinism only permits suffering in light of his ultimate purpose to build the kingdom of God.

Craig asserts that the revised proposal made by some openness theologians, that God uses his knowledge of what creatures might do in particular circumstances, fails if openness theology is correct about the unknowability of counterfactuals of creaturely freedom (i.e., the grounding objection) (89). Furthermore, “knowledge of mere ‘might’ counterfactuals is insufficient to give God the sort of providential control described in the Bible” (90-91).

Craig charges the Augustinian-Calvinist account of providence with making God the author of sin, because his foreknowledge is based on foreordination: “God knows what will happen because he makes it happen” (91). It turns “God into the devil” because God moves people to commit moral evil but then holds them “morally responsible for acts over which they had no control” (91).

Finally, in regard to theological alternatives, Craig assesses the “simple foreknowledge account,” and finds it very weak because, “without middle knowledge, God cannot know prior to the creative decree what the world would be like.” But, Craig charges: “Surely there is more substance to the biblical doctrine of foreordination than the triviality that God decrees that what will happen will happen!” (94).

Philosophically, Craig argues that, “since there are counterfactual truths, God must know these.” God cannot know them posterior to his creative decree, because then “it is God who decrees what choices agents shall make in whatever circumstances they find themselves, and human freedom is annihilated.” Consequently, God must know counterfactuals of creaturely freedom logically prior to his creative decree, which is to say that God has middle knowledge” (95).

In conclusion, Craig commends a Molinist account of divine providence “for serious consideration,” because it gives God “sweeping sovereignty over the affairs of men.” Yet it does so in a manner compatible with human freedom because “the circumstances envisioned in counterfactuals of creaturely freedom are nondetermining” (100).

Items not clear to me in Craig’s presentation

In a later post, I will deal with responses to Craig’s proposal and interact with it myself at that point. Here, however, I want to mention a couple of places at which I can not make sense of Craig’s presentation. These may be his problem, or they may be mine, but either way they render response a bit more difficult.

First, I am puzzled by the way in which Craig describes the Dominican view. He states that, at the moment logically prior to God’s decree, “God knows, for example, that there is a possible world in which Peter denies Christ three times and another possible world in which Peter affirms Christ in precisely the same circumstances” (81 [emphasis mine]). Whether or not Peter denies Jesus in such circumstances is determined by God in his choice of a particular world. I find that description coherent. Later, however, Craig says: “Thus, [God] decrees, for example, that if Peter had instead been in such-and-such circumstances, he would have denied Christ two times” (81 [emphasis mine]). There is an important difference between these two ways of speaking, because of the use of “instead” in the second statement, and this muddies the situation. In the first case, Peter’s libertarianly free choice would account for the difference between a world in which he denies Jesus and a world in which he does not, since everything else is the same. In that case, Peter’s choice would be made for him by God, who chooses that such will be the case in the world he creates. In the second case, however, if the change in circumstances accounts for a difference in Peter’s action in the two possible worlds, then Peter need not be libertarianly free in his choice. But then it would seem likely that God knew the counterfactuals naturally or necessarily, logically prior to his decree. The difference is very significant.

My second puzzlement arises in Craig’s critique of the revisionary open theism which attributes significant control to God on account of his knowledge of “might counterfactuals.” Craig argues that the Molinist God “exceeds the God of revised open theism in that he knows not only how creatures might (or could) choose in any set of circumstances but also how they would choose under any such circumstances” (90 [emphasis mine]). Here, the problem I identified in regard to the description of the Dominican model arises again. Craig speaks as though the actions of libertarianly free creatures are relative to the circumstances, but this appears to me to be a form of compatibilism that belies the libertarian contention that a free being could have done otherwise in precisely the same circumstances.

On page 86, Craig argued, concerning the crucifixion of Jesus, that “God decreed to create just those circumstances and just those people who freely do what God willed to happen.” But I could say that, as a soft compatibilist, without affirming libertarian freedom. It seems to me that what Craig should say is that “God decreed to create the world in which just those people freely choose to do what God willed to happen in just those circumstances.” Essential to the libertarian account of freedom is the belief that there was a world in which the same creatures would have acted differently, even though the circumstances were not different. I assume that Craig has simply been insufficiently careful about his way of speaking, because otherwise it is difficult to be certain of the model he is proposing as a Molinist.

We will return to this subject in the future, God willing.

Previous posts in this series: Part 1, Part 2, Part 3.



By Terrance Tiessen

I am Professor Emeritus of Systematic Theology and Ethics at Providence Theological Seminary, Canada.

4 replies on “A Molinist model of God’s providence”

I think I can shed some light on some of your puzzlements. Although, of course, I could just misunderstand either you or Craig.

First, Craig’s concern with the Dominican view is that God decides the would-counterfactuals instead of us. To him, there is no way of affirming this without removing our freedom in these actions. When he says, “God knows, for example, that there is a possible world in which Peter denies Christ three times and another possible world in which Peter affirms Christ in precisely the same circumstances” he’s referring to God’s knowledge of what could be (which everyone affirms). But when he says, “Thus, [God] decrees, for example, that if Peter had instead been in such-and-such circumstances, he would have denied Christ two times” he’s moved to talking about God’s knowledge of what would be. In the latter case, according to the Dominican view, God has decided that Peter would do one of the possible actions instead of the other in those circumstances, even though Peter still could do both. Peter’s two choices always serve to differentiate the two possible worlds, since it never ceases to be that Peter could do either action. What changes is what Peter would do.

Second, as far as I know (admittedly, from his other work) Craig’s account does still allow the agents could choose otherwise in the same circumstances. Rather, when he talks about circumstances, he’s referring to what agents would do is certain circumstances. While I could do a number of things in the same circumstance, there’s only one thing I would do. Regardless, Craig typically identifies at least two theses under the banner of libertarianism: (1) the agent could have chosen otherwise in the same circumstances, and (2) the agent wasn’t externally caused to make the decision they made. He denies (1) as being necessary or sufficient for freedom on the basis of the Frankfurt counterexamples but affirms (2) is necessary (and sufficient if I’ve correctly understood him) for freedom. How he distinguishes his view from compatibilism isn’t on the basis of (1) (since both he and the compatilibist deny (1)), but on (2). To affirm the compatibility of external causal determination of a choice and that’s choice’s freedom (as the compatibilist does) is to deny (2).

If I understand correctly, the statement, “God decreed to create the world in which just those people freely choose to do what God willed to happen in just those circumstances.” affirms, along with the Dominican, that God decided the would-counterfactuals of creaturely freedom, which is contrary to the Molinist contention that these were true prior to God’s creative decree. So I’m not too sure Craig should have said that.

I think it’s important to note that while the soft compatilibist can also affirm “God decreed to create just those circumstances and just those people who freely do what God willed to happen.” without affirming libertarian freedom, Craig is attracted to Molinism precisely because he can affirm thestatement while keeping libertarian freedom. Consider what he says on page 92, where he responds to a potential objection from the Augustinian-Calvinist: “…the Molinist account would still enjoy the considerable advantage of making room for creaturely freedom.”

Thank you Roland. Your comments are helpful in giving me a chance to revisit Craig’s work at points which originally puzzled me.

In terms of my first item, you have made helpful observations about the importance of the distinction between “could” and “would” in the Molinist account, but I regret that your comments have not removed my original point of concern. I understand Craig’s concern about the Dominican proposal: if counterfactuals are determined by God, it seems that creatures can not be libertarianly free, as the Dominicans argued posited. So far so good.

My puzzlement arose from Craig’s statement that “if Peter had instead been in such-and-such circumstances, he would have denied Christ two times.” You do well to note that Craig’s earlier statement had to do with possibilities (what could be), this one speaks of what would be (hence counterfactuals). My puzzlement remains, however, regarding Craig’s statement that God might have decreed “if Peter had instead been in such-and-such circumstances, he would have denied Christ two times” (81). I still think that Craig gives the impression here that the circumstances were significant, but I’m going to assume that he spoke inadvertently at this point (as you grant he may have done at the point of my later query). What Craig wants to emphasize, I assume, is that, since Peter could have denied or affirmed Christ three times, in the same circumstances, he could also have denied Christ twice in those same circumstances. The critical thing is that, in the Dominican model, there never were any counterfactuals concerning Peter’s action. There were possibilities before the decree, and there were facts after the decree. But any counterfactuals in unrealized possible worlds would have to be decreed by God. I see no reason why God would have bothered to decree such counterfactuals. Consequently, the Dominicans are left only with possibilities and facts. Disappointingly, Craig chose not to address at all the Reformed soft-compatibilistic alternative that counterfactuals exist prior to the decree (as Molinists assert), but that God knows them naturally or necessarily. I’ll say more of this when I deal with responses to Craig.

I agree with you that libertarian freedom is critical for Craig and other Molinists. They are uniformly opposed to all forms of compatibilism because they reject soft-determinism. As you observe, they reject the hard-compatibilism of Thomism because they deny that it allows creatures to be libertarianly free. They reject the soft-compatibilism of Reformed theologians because they deny that the freedom of volition is sufficiently robust to ground moral responsibility. Only libertarian freedom will suffice, in their view. In a forthcoming post, we’ll examine whether Molinists can have their cake and eat it too, i.e., whether God can know counterfactuals logically prior to his decree, but those counterfactuals be determined by the moral creatures.

Ah I see how I misunderstood your original puzzlement. Yes, admittedly it is puzzling that God would “need” to decree the would-counterfactuals at all. I imagine that Craig is eager to assert that God did indeed decree them because in our experience we perceive these counterfactuals as true and, as you point out in the post, he mentions how Scripture also affirms the truth of these counterfactuals. Of course, there is perhaps another account of how these counterfactuals might be true independent of God knowing them prior to his creative decree or as a result of that decree, which construes them in terms of “nearness in possible worlds” or something like that. This latter option wouldn’t require God to decree them in order for them to be true.

I’m intrigued by your soft-compatibilist account and so I look forward to your later post. Does such an account place the would-counterfactuals in God’s necessary knowledge? On such an account would you say, “An agent A would do freely action S in circumstance C if and only if in every possible world in which A is put in circumstance C, and A is not externally caused to do otherwise, A does S”?

Yes, Roland, I do believe that God knows counterfactuals as part of his natural or necessary knowledge. And yes, I would concur with the syllogism that you stated. I find the grounding objection to Molinism persuasive, but I affirm the usefulness of God’s natural knowledge of counterfactuals, because I think it gives a fuller picture of the compatibility between the agency of God and that of his moral creatures. As my discussion of this book goes on, more of this will doubtless emerge.

Leave a Reply

Your email address will not be published. Required fields are marked *

HTML tags are not allowed.

145,573 Spambots Blocked by Simple Comments