With this post I begin a series in which I respond to a 2013 paper by William Lane Craig and David Hunt. Entitled “Perils of the Open Road,” the paper critiques two papers defending open theism. More specifically, they critique 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.” All three papers were published in the journal Faith and Philosophy.
Craig and Hunt (hereafter, C&H – not to be confused with the pure cane sugar variety) regard the recent trend among Christian philosophers toward open theism to be “lamentable” and “baffling”. Their stated purpose in is to “explain why … the arguments offered by Tuggy and Rhoda et al. should not persuade anyone to venture down the road to open theism” (pp. 49-50).
In the near future I plan to publish a response paper to C&H. This series of posts is intended to facilitate that by helping me to gather my thoughts. I’ll be walking through the C&H paper more-or-less in sequence, focusing on only a few pages at a time. But first we need to concern ourselves with some preliminaries.
What is open theism?
It’s good practice to carefully define one’s key terms, so before I begin fisking C&H I need to define “open theism” and “future contingent”.
In a previous post, I defined”open theism” as follows (slightly edited here):
Open theism =def. a view that affirms
- broadly classical theism =def. there is a unique, personal, metaphysically necessarily being (namely, God) who essentially possesses a maximally excellent compossible set of greatmaking attributes, including maximal power, knowledge, and goodness, to whom all (concrete) non-divine beings owe their existence, and who created the world (i.e., the space-time system of concrete non-divine beings) ex nihilo and can unilaterally intervene in it as he pleases
- causal openness of the future
- incompatibility of a causally open future with an epistemically settled future
In what follows I will take this definition as a given. (2), causal openness, is the claim that there are future contingents, a notion I define below. (3) concerns the relation between causal openness and epistemic openness. The claim is that if the future is causally open, then it must also be epistemically open, by which I mean that there is no unique, complete, linear “story of the future” that can be known to be (or to be going to be) the actual future. If the future is epistemically open for God, then God sees the future as a branching array of possibilities, not as a unique, linear sequence of events.
I define “future contingent” as follows:
Future contingent =def. a causally possible future event that is not also causally necessary.
As such, future contingents both might happen (i.e., there is a non-zero chance of their occurring) and might not happen (i.e., there is a non-zero chance of their not occurring), where a ‘chance’ is a single-case, objective, non-epistemic probability of an event’s occurring. For example, assuming a genuinely indeterministic system, the chance that two fair 6-sided dice will land double-sixes is 1/36. The chance of a future contingent’s occurring must be greater than 0 and less than 1.
One of the key issues separating C&H and myself is whether it can be the case that a given future contingent will happen even though, by definition, it has a chance of not happening, or conversely whether it can be the case that a given future contingent will not happen even though, by definition, it has a chance of happening. To these questions C&H say Yes and I say No.
Incidentally, we can see from the above definitions that C&H’s characterizations of open theism as a view that “restricts” (p. 49) or “abandons” (p. 52) God’s foreknowledge of future contingents is inaccurate. While one can be an open theist and affirm that there are future contingent truths that God does not know on the grounds that they are somehow unknowable, most open theists believe that God has perfect and complete foreknowledge. Where they differ from non-open theists is in believing that the future that God perfectly and completely foreknows is an open-ended one. If, as such open theists believe, there is no fact of the matter to the effect that a given future contingent will occur or will not occur, then it cannot be a defect of or a restriction in God’s knowledge for Him not to know that fact.
Propositions about future contingents
We also need to be careful how we express propositions about future contingents and how we demarcate the relevant class of propositions. First, a future contingent proposition (FCP) is simply a proposition about a future contingent event. Obviously this is a relational and not an intrinsic property of the proposition for it depends on the event’s being both future and causally contingent. As time passes, what was future eventually becomes past. And causal contingencies may be foreclosed by other events, e.g., prior to the admiral’s order it may be causally contingent whether there is going to be a sea battle, but once the admiral orders the attack it is no longer causally contingent (assuming nothing remains to prevent the admiral’s order from going into effect).
There are two general classes of FCPs. I’ll call them representationally determinate future contingent propositions (DFCPs) and representationally indeterminate future contingent propositions (IFCPs), respectively. The latter express might and might not claims about future contingents. Given causal indeterminacy with respect to whether a sea battle occurs tomorrow, that there might be a sea battle tomorrow is an FCP—it is a proposition about the causally contingent future—but it is an IFCP since it represents the future as indeterminate with respect to whether a sea battle occurs tomorrow. The proposition is compatible with both the occurrence and the non-occurrence of a sea battle tomorrow. In contrast, to say that there will (not) be a sea battle tomorrow is to represent the future as determinate in that respect.
C&H and myself are, I believe, largely agreed about IFCPs, even though they largely ignore the category and conflate FCPs with DFCPs. The central debate concerns DFCPs and whether any DFCPs are true. Again, to that C&H say Yes and I say No.
One thing we can see already is that C&H’s way of posing the question, “Does God know future contingents?” (p. 50), is inadequate. They intend this question to separate open from non-open theists, anticipating that the former must answer with a No. But in fact open theists can literally and straightforwardly answer Yes to this question since there are true FCPs (specifically, IFCPs) that God knows to be true. Of course what C&H meant was to ask whether God knows DFCPs. But that’s not what they said. This is why defining terms carefully and making careful distinctions is so important. Similar comments apply to C&H’s follow-up questions, “Are there any future contingent truths?” (p. 50) and “Are future contingent statements then false?”
“True now” versus “true simpliciter“
To say that a proposition is true can be interpreted in two different ways. On one reading it can mean that the proposition is true now (i.e., at the present moment). This is a special case of truth-at-a-time. Some philosophers (like Peter van Inwagen) have complained that the notion of truth-at-a-time is incoherent. To them I say, if the notion of truth-at-a-possible-world makes sense (as I’m sure PvI would concede), then so does truth-at-a-time. In both cases we are employing the notion of truth-at-an-index, where the index specifies the reference frame from the perspective of which truth is to be evaluated. To say that a proposition is true at possible world W is just to say that it would be true simpliciter if W were actual. Likewise, to say that a proposition is true at time T is just to say that it would be true simpliciter if T were present. (Caveat: If put in terms of Robert Adams’ well-known distinction between truth-in-a-world and truth-at-a-world, then what I’m calling “truth-at-an-index” here should be rephrased as “truth-in-an index.” Adams’ distinction, however, isn’t relevant to my present purposes, and since “truth-in” sounds a bit awkward when dealing with time, I’m sticking with “truth-at”.)
Having defined truth-at-a-time in terms of truth simpliciter, let us contrast it with the latter. The difference is this: Whereas truth-at-a-time (and, more broadly, truth-at-an-index) evaluates truth from the relative perspective of a specified reference frame, truth simpliciter evaluates truth from the perspective of an absolute or all-encompassing reference frame. It evaluates truth, as it were, from a God’s eye perspective.
Defenders of open theism who affirm an unrestrictive notion of God’s foreknowledge deny that any DFCPs are true either now or simpliciter. They may affirm the possibility of a DFCPs becoming true, but would insist that when that happens it also ceases to be an FCP (since for a DFCP to become true is for the future to cease to be causally contingent in the relevant respects). For their part, C&H believe (p. 51 fn.) that for any proposition that is ever true-at-a-time there is a corresponding tenseless proposition that is always true simpliciter.