God, Time, and Molinism

Here I am going to tackle another anti-Molinist argument found on the Internet, just for fun! The good news is this critique seems to explain the basics of Molinism and middle knowledge accurately, which is not always a given. However, his arguments against Molinism seem fairly confused with respect to counterfactual semantics.

For a first example, consider his argument that Molinism entails a contradiction in God’s omniscience. He argues that in order for would-counterfactuals to be contingently true, it must also be the case that God knows that these counterfactuals “might not” be true. But, he claims, it is the case that “might-not” and “would” are contradictory; hence, Molinism teaches that God knows contradictions, which is absurd.

Presumably, this author has seen the counterfactual square of opposition (or else holds that these are contradictory for some other reason). Whatever the case may be, he is correct: “Would” and “might not” counterfactuals are contradictories. However, he is confusing might-not counterfactuals with the concept of contingency. As Craig and Moreland point out,  

‘Might’ counterfactuals should not be confused with subjunctive conditionals involving the word ‘could.’ ‘Could’ is taken to express mere possibility and so is a constituent of a modal statement expressing a possible truth. . . . The fact that something could happen . . . does not imply that it might happen under those circumstances. ‘Might’ is more restrictive than ‘could’ and indicates a genuine, live option under the circumstances, not a bare logical possibility.[1]

This shows his confusion. The author wants to say that another way of saying these would counterfactuals are contingent is to say they might not occur, while also admitting that such counterfactuals are contradictory. The former just isn’t so on the technical semantics (even if it’s how we use it in ordinary language).

The next criticism he has is against a “form of Molinism” whereby God exists outside of time. It’s worth noting that someone can hold to God’s timelessness or some other view and believe Molinism. If there were to be a significant problem for holding both Craig’s model and Molinism, one could jettison the model and retain Molinism, perhaps. He writes of Craig’s view of Molinism, “What I do take issue with is God knowing those possibilities based on what we do because this would render God’s knowledge contingent on us and therefore God would be ‘learning.’”

It’s not clear why he says, “therefore,” as no rule of inference will derive this conclusion from that premise. I can repair the argument to be valid: If God’s knowledge is contingent, then he is learning; God’s knowledge is contingent; therefore, he is learning. But we’d still need some argument that the major premise is true, given Molinism. I have no idea what that argument is.

His last set of critiques is that Molinism is problematic because it denies determinism. But of course, anyone who does not already embrace determinism shouldn’t be convinced by this. His argument is that there seem to be certain things that are true in all possible worlds God could actualize regarding supposedly free choices in the libertarian sense, so that really he has to embrace some kind of determinism.

But this is multiply confused. First, if some fact F is necessarily true, it is not theologically causally determined (or, it is so only in light of its logical necessity). Unless he believes in some kind of modal universal possibilism, if something is necessarily true not even God himself could change that truth value. Second, he writes out his modal argument purely in English sentences; if he had symbolized it, he would have seen the argument commits the very same modal fallacy he earlier recognizes as fallacious. One can still see it in the English itself. Here it is:

1.     Simultaneity is transitive (if A happens simultaneously with B, and C happens simultaneously with B, then, necessarily, A happens simultaneously with C).

2.     Apply to middle knowledge: if A takes place at t1, and this is simultaneous with God’s knowledge, then God’s knowledge of A is simultaneous with C.

3.     Therefore, A is necessarily taking place simultaneously with C.

But this is just the well-known modal fallacy. The problem starts with the initial formulation itself. The transitive property is not that if A=B and C=B, necessarily A=C. It’s rather, “Necessarily, if A=B and C=B, then A=C.” Here’s a brief example. Consider height relations. Based on the meaning of the term, they use transitive relations. So, necessarily, if Jim is as tall as Dan, and Brad is as tall as Dan, then Jim and Brad are as tall as Dan. But who wants to thereby infer that it’s a necessary truth that Jim and Brad are as tall as Dan? Who wants to say there is some mysterious truth of logic that dictates the relative heights of these three men?

Finally, this is confused as to Craig’s model of time with respect to middle knowledge. The content of middle knowledge comes logically prior to the creative decree. On Craig’s model (which he is ostensibly critiquing), crucially, there is no time at all; this is why it is referred to as coming logically prior to the creative decree. Thus, there is no simultaneity to speak of anyway.

In conclusion, this author seems at least relatively familiar with Molinism and some issues surrounding Craig’s work, but not familiar enough to avoid some serious misunderstandings.

EDIT: Changed example for Jim, Brad, and Dan to better reflect the transitivity.

---

[1] J.P. Moreland and William Lane Craig, Philosophical Foundations for a Christian Worldview (Downers Grove, IL: IVP Academic, 2003), 53.