Responding to Craig and Hunt (Part 2 – “The Argument”)

This is the second installment in a series of posts in which I respond to a recent 2013 paper by William Lane Craig and David Hunt (hereafter, C&H). Entitled “Perils of the Open Road,” C&H critique two papers defending open theism: a 2006 paper that I co-wrote with Greg Boyd and Tom Belt entitled “Open Theism, Omniscience, and the Nature of the Future” and a 2007 paper by Dale Tuggy entitled “Three Roads to Open Theism.”

In part 1 of this series I laid some groundwork by defining key terms like “open theism” and “future contingent”, distinguishing between future contingent propositions (FCPs) and their indeterminate (IFCP) and determinate (DFCP) subtypes, and clarifying the notions of truth-at-a-time and truth simpliciter.

In this post I examine C&H’s discussion of what they call “The Argument” (p. 52), by which they mean an argument for the incompatibility of causal openness (i.e., future contingency) with an epistemically settled future. Those who have read part 1 will recognize this as the third component in my definition of “open theism”. By challenging The Argument (hereafter, TA), C&H hope to undermine the motivation for open theism. Given my definition, however, they are potentially doing more than that–if they can successfully refute TA then they will have successfully refuted open theism. I will argue in what follows that C&H’s dismissal of TA is way too quick.

What is “The Argument”?

Despite the emphasis they put on TA, C&H only state a simplified version of it in a footnote and without any commentary. This is partially because they regard it as a “complex” (p. 53) argument and mainly because they want to devote their space in the paper to other matters. Here’s how they state TA (I’ve numbered their statements for ease of reference):

TA:
[1.] Suppose God knew before you were born that you would do A at T (where your doing A at T would otherwise be regarded as paradigmatically free).
[2.] Necessarily, if God knew before you were born that you would do A at T, then you will do A at T.
[3.] But it’s no longer possible for God not to know that you will do A at T (given the supposition that he did know this).
Therefore
[4.] [I]t’s no longer possible for you not to do A at T. (p. 53, fn. 6)

This is, in fact, a passably decent statement of TA, though [1] could have been omitted since it’s parenthetically included in [3]. About this sort of argument, C&H raise two objections.

C&H’s first objection to TA

Their first objection is that “The Argument is strongly counterintuitive” because “[m]ere foreknowledge … just does not seem like the sort of thing which, when added to an action that otherwise satisfies one’s favorite conditions for free agency, could really render that action unfree” (p.53).

This response is one that has been frequently repeated by both Craig and Hunt in various places, but it reflects a serious misunderstanding of TA. The argument is not that divine foreknowledge (of your doing A at T) “renders” the action unfree, but rather that such foreknowledge is only possible if the action is unfree. Jonathan Edwards stated the point very well:

Whether Prescience be the thing that makes the event necessary or no, it alters not the case. Infallible Foreknowledge may prove the Necessity of the event foreknown, and yet not be the thing which causes the Necessity (Freedom of the Will, sec. XII).

C&H are undoubtedly familiar with Edwards’ point, so it is hard to fathom why they would object to TA on the grounds that foreknowledge doesn’t “render” our actions unfree. This is merely objecting to a straw man.

Moreover, Edwards’ point is not particularly counterintuitive given that lots of smart people have worried about it over the past two millennia and have devoted tens of thousands of pages trying to rebut it.

C&H’s second objection to TA

C&H’s second objection is that, when fully stated, TA turns out to be “quite complex”, such that “[t]here are many points at which it could go off the rails” and, indeed, there is “an embarrassment of riches … for The Argument’s critics” (p. 53). They go on briefly to indicate a couple different lines of response, noting that Craig and Hunt diverge rather widely on which response(s) they think might work.

Craig, we are told, is a Molinist who “endorses an Ockhamist solution to the problem of theological fatalism” (p. 49, fn. 2), whereas Hunt is an anti-Molinist and anti-Ockhamist who endorses a Frankfurtian solution to the problem on the grounds that “divine foreknowledge does not cause or explain foreknown actions” (p. 53, fn. 9).

There is irony here in that their respective “solutions” are completely antithetical. It’s hard to see how either of them can say that there is an “embarrassment of riches” for TA’s critics when each of them independently thinks that nearly all of the options, save the one they each prefer, are inadequate. There is additional irony in the fact that Craig’s Molinism and Ockhamism are themselves antithetical, a point Craig fails to realize. Ockhamism renders God’s past beliefs about future contingents explanatorily dependent on the actual occurrences of those events, which dependency is incompatible with the robust providential control over future contingents that Molinism is supposed to provide.

Finally, while many versions of TA (like Nelson Pike’s from his famous 1965 paper, “Divine Omniscience and Voluntary Action”) are fairly complex and are thus contestable at several points, the basic argument for the incompatibility of foreknowledge and future contingency is in fact very simple and is demonstrably valid, as I show in this paper. The first premise is

1. God knows of a unique possible future F that it is (or is going to become) the actual future.

By a “possible future” here I mean a complete, linear, logically possible extension of the causally relevant actual past, compatible with holding fixed the laws of nature and concurrent divine causal contributions to creaturely events. (Dear reader, please don’t worry about all the qualifications. If you must know, they are needed to avoid begging the question against Ockhamism.)

Let F be one such possible future, and let K stand for God’s knowing that F is (or is going to become) the actual future. We can then express (1) as K and, necessarily, if K then F. And since logical necessities are unpreventable, (1) entails K and, unpreventably, if K then F, where “unpreventably” is to be read as “in all causally possible futures”.

The second premise that we need is just this:

2.  Unpreventably, God knows that F is (or is going to become) the actual future.

Or, in short, Unpreventably, K.

Now, given that I’ve defined “unpreventably” as “in all causally possible futures”, it logically follows from (1) and (2) that F is unpreventable, i.e., that F occurs in all causally possible futures. From this it follows that F is the only causally possible future and thus that F is fated to occur.

Since C&H both concede (1), their only option to avoid fatalism is to reject (2), which is exactly what Ockhamists do. (Actually, since Hunt rejects Ockhamism, his strategy must be to accept the soundness of the argument and then to argue Frankfurt-style that unpreventability is compatible with future contingency. In other words, he has to say both that there is a causally possible future in which F does not occur and that it is unpreventable that F occurs. Clearly, either he’s contradicting himself or he’s construing “unpreventability” differently than I do. If the latter, then he’s simply equivocating on the term in rebutting the argument I laid out.) In sum, there is no “embarrassment of riches” for the anti-fatalist. One can reject (1) and be an open futurist or one can reject (2) and be a preventable futurist. There are no other options.