I will try to recast Craig’s argument and give a response. Take the proposition below :

If I’ve read Craig right, then according to him, future contingent propositions like P1 possess a true/false status (a’la the Principle of Bivalence) and if God is omniscient He would know it. But if God believes P1 and a few years later Brian ends up with Judy instead (or if he remains a bachelor), then God would’ve held a false belief in the present. So either we follow the classical view in which future propositions are determinate and thus are either true or false (thus entailing that God would have to know them, contra the open view), or we follow Boyd’s apparent error of implying that God holds false beliefs.

Yet things are not so simple.

A major element in open theology debate is about – to use one of Boyd’s phrases – the ‘ontological status of the future in the present’. It concerns the nature of the future. I suspect arguments like Craig’s tend to short-circuit this investigation by presupposing one version of the future.

Consider two further propositions below:

If P1 above is either true or false, then either P2 or P3 must be false. Because if P1 is true, then P3 is false. If P1 is false, then P2 is false. If it is a determined fact that Brian weds Susan, then saying that Brian might not marry her is clearly false as this is no longer a possibility. If it is already an actuality that Brian does not marry Susan, then saying that Brian might marry her is about as correct as saying that a triangle might have four sides.

However, open theists submit that both P1 and P2 are TRUE in the present and that God believes both to be true.

But if this is the case, then P1 ("Brian will marry Susan in the future") logically CANNOT be either true or false until the future when it is rendered a status accordingly.

The question for Craig (and Hunt and Helm, for that matter) is whether he thinks falsifying either P2 or P3 is acceptable. His position would demand that either one has to be false, but this would betray his bias against the idea that possibilities are real, and thus knowable to God. But if we agree that God’s omniscience includes possibilities (because creation includes them) then the Principle of Bivalence cannot be applied without qualification to future propositions. (I cannot see how phrasing P1 to become P1a, "It is true that Brian will marry Susan in the future", makes any difference to the argument. P1a carries a future proposition which, if my preceding case about the ambiguity of such propositions is right, can only bestow a similar ambiguity upon its carrier. If both P2 and P3 remain true in the present, P1a cannot without qualification be slotted into the true/false grid, or at least not in the same present).

However, Craig seems to affirm the reality of possibilities when he writes:

I suspect some incoherence here. If God knows future propositions to be definitively either true or false, then the future is as fixed as the past (since God cannot hold false beliefs). How then can a phrase like "Brian might not marry Susan" be a true statement about possibility if, to take an example, "Brian will marry Susan" is true? Because if it is 100% certain that Brian and Susan will be married, then the idea that Brian might not marry her becomes a logical im-possibility and hence false i.e. the present possibility of Brian not marrying Susan doesn’t exist and thus cannot be said to be real.

Craig’s idea of possibility would only be meaningful in some eternal logical point prior to God’s decree to create an actual world. Once this world exists, whilst we may concede that creatures still freely choose their actions, there is no longer any possibility of their actions being other than what has been decreed. In a word, subsequent to creation, possibilities cease to be.

Finally, despite the foregoing, the open view does affirm that SOME future propositions follow the Principle of Bivalence and are either true or false, contra Craig:

I think the open view is more nuanced than what Craig’s comments may suggest. Open theory, as I understand it, has not shut the door completely on future propositions. Some CAN be affirmed as either true or false i.e. those (and only those) which God has sovereignly decided to render actual.

In such situations, yes, God would have definite knowledge of these particular aspects of the future, the definiteness of which are grounded in His own decision and therefore determinate. Good examples of these would include Cyrus’s edict and other prophecies, though (and this is important) we need to bear in mind that even such determinate future propositions contain elements of contingency. God can flex the timing, the manner and even the very existence of some decrees based on the free choices of individuals and communities as they unfold in time.

(This could provide yet another way of understanding Peter’s triple denial of Jesus without presupposing an exhaustively settled future - John Sanders posited it as conditional prophecy, whereas Boyd suggested it to be a manifestation of a solidified aspect of Peter’s character which was easily predicted. Alternatively, we could postulate that God unilaterally rendered that aspect of Peter’s future as settled. It wasn’t too hard to predict that Peter would deny Jesus at least once. I propose that God/Jesus intervened to ensure a triple denial, according to His plans for this eventual leader of the church and what He knew a volatile character like Peter required.)

In conclusion, Craig’s argument against the open view from future propositions seems to presuppose the certainty of all future events. This (despite what he says) nullifies the idea of possibility, the very thing open theists insist are real in the present and makes the future in-determinate. Nevertheless, open theists hold that not all the future is so, because God can choose to settle whatever He wishes to. The Bivalence Principle can apply to the future if God so desires.