Four (Bad) Philosophical Arguments for Open Theism
Four (Bad) Philosophical Arguments for Open Theism
While the tradition is hardly sacrosanct, one should at least have good reasons for departing from it. I have argued that Scripture does not provide good reasons for departing from the traditional doctrine of divine foreknowledge. At the same time, I don't believe the Scriptures alone absolutely require the classical position. If there were strong nonscriptural reasons for rejecting Classical Theism—say, if Classical Theism just didn't make sense—then I don't believe there would be insuperable obstacles to reading the Scriptures as openists propose. I do think that the case for Openism comes down to reasons for thinking that Classical Theism, in one way or another, just doesn't make sense; that is, it comes down to philosophical objections to Classical Theism. Four are particularly important.9 I cannot hope to do justice to all four in this brief chapter, but I can perhaps say enough about each to persuade the reader that the traditional Christian commitment to exhaustive divine foreknowledge is not in serious jeopardy.
God Can't Know the Contingent Future Because It's Not Yet True
Statements about future contingent events aren't true until the events occur; since the occurrence of these events isn't true in advance, God can't know them in advance.
This reason for embracing Openism flies in the face of both logic and common usage. Let's begin with logic. Either I will call my mother tomorrow, or I won't call my mother tomorrow. One or the other of these statements about the future must be true. The principle that either a given statement or its denial is true is called the “Law of Excluded Middle.” But this first brief on behalf of Openism requires that this law be abrogated. That's a heavy cost, and the vast majority of logicians would decline to pay it.
Ordinary usage and common sense also reject it. We make claims about the contingent future all the time, and we assume that such claims are sometimes true. Consider the following:
1. This coin will land heads on the next toss.
2. My wife will vote for candidate X in tomorrow's election.
3. The U.S. will elect its first female president in 2016.
The openist may object to taking such claims at face value on the grounds that the future is not yet real and that claims about it are therefore not yet true. But this objection would be received with bemusement by anyone engaged in the actual practice of making claims about the future. If I bet that this coin will land heads and it does land heads, then I was right; you can't take my money on the grounds that there are no true future contingents, I bet on one, and so I was wrong.
This attempt to renege on a bet, like the openists' denial of truth-values to future contingents, rests on a misunderstanding of what future-tense statements are all about. This misunderstanding is perhaps best exposed by comparing the future tense with the past tense. If statements about the future are not yet true because the future is not yet real, then statements about the past should be no longer true because the past is no longer real. But that's nonsense. Consider the following:
4. Caesar was assassinated in 44 BC.
5. In 1492 Columbus sailed the ocean blue.
6. Yesterday I promised my wife I'd wash the car.
These statements are true in virtue of what happened in 44 BC, AD 1492, and yesterday; none is true in virtue of anything that is happening now (if Caesar were being assassinated now, then he wasn't assassinated in 44 BC). Yet each of these is true now. (Imagine trying to get out of washing the car on the grounds that, because statement 6 is past, it is no longer true!) These statements are about the past (not the present) but are being made now. That's the nature of statements about the past. Likewise, supposing that my wife does vote for candidate X tomorrow, it is now true that she will vote for X tomorrow. This obviously does not mean that she is now voting for X; the polls haven't opened yet! This statement is about the future (not the present), but it is being made now. That's the nature of statements about the future.10
There is one important difference between statements about the past and statements about the future. The contingent future can only rarely be known by human beings, and even when it can be known (as I might know whom my wife will vote for tomorrow), it cannot be known with certainty. But this is simply irrelevant to the question whether there are future-contingent truths. “This coin will land heads on the next toss” is true just in case that's what happens; its truth does not depend in any way on whether anyone can also know this truth in advance. If God can't know these truths, it must be for some reason other than the absence of any truths there to be known. But what could that reason be, given that God is unsurpassably great?
2. God Can't Know the Contingent Future Because Then It Wouldn't Be Contingent
A contingent event is one that can go either way; but if God already knows which way it will go, it can't go the other way because God can't be mistaken. So this foreknown event is not contingent after all; nor can any infallibly foreknown event be contingent. Consider my wife's vote tomorrow. God knew before she was even born whether (and if so, how) she would vote. When tomorrow arrives, can she do anything other than exactly what God has always believed she would do? Of course not. But then she's not free in what she does.
Entire books have been written about this problem, and I don't think anything I could say within the space available to me here would be adequate. I will therefore restrict myself to two inadequate points—inadequate in that they don't pretend to show what is wrong with this argument, only that there is something wrong with it.
In the first place, readers should know that the vast majority of Christian philosophers, past and present, have rejected this argument. Openists are in a minority here. That doesn't mean they're wrong—the truth isn't determined by majority vote—but it does mean that readers ill-equipped to follow the abstruse twists and turns of the philosophical debate shouldn't endorse the openist position just because an openist apologist claims that it's a good argument.
There is no consensus, among the majority who reject the argument, over the exact point at which the argument goes wrong. I have my own favorite response to the argument, which follows Augustine's analysis of the problem in his On Free Choice of the Will.11 But discussion of the argument over the centuries has turned up many different points at which it can be challenged. Defenders of traditional foreknowledge can regard this as an embarrassment of riches. What are the chances that all of these challenges are mistaken? A further point concerns the historical figures most closely associated with these various critiques. They include Aristotle, Augustine, Boethius, Thomas Aquinas, Duns Scotus, William of Ockham, and Jonathan Edwards. What is interesting about this list is that all of these figures are Christians, except Aristotle. And which step does Aristotle reject? The very step that most openists reject: that there are future-contingent truths. This is at least ironic, given the openist charge that classical theists are the ones corrupted by pagan philosophy!
The second reason readers should distrust this argument as a ground for Open Theism has to do with the fact that we are sometimes justified in thinking that a position is mistaken even when we don't know how. An example is Zeno's paradoxes. One of the more famous of these is a “proof” that Achilles cannot win a footrace against a tortoise if the tortoise is given a head start, no matter how small. We know that this can't be right—we see faster things passing slower things all the time—but Zeno gives a clever argument for his conclusion, and it's actually hard to refute. In the face of this argument, one is perfectly within one's rights in saying: “I know that argument is mistaken even though I don't see how.”
The argument that God's knowledge of the future would render the future noncontingent is another case of this sort. Here's why. Take an action that satisfies to the highest degree your favorite criteria for free action, whatever they may be. If anything is a free action, that is. Now add an infallible foreknower to the scenario. How can the introduction of this foreknower, just by itself, change the status of the action so that it is no longer free? There are conditions that clearly would warrant such a reassessment—for example, if it were added that the agent was acting under posthypnotic suggestion or was being controlled like a marionette by tiny invisible wires. But the idea that the mere presence of an infallible foreknower could make this kind of difference is utterly implausible. Readers therefore have good reason to be confident that this argument goes wrong, even if they're unable to determine exactly where it goes wrong.12
3. Even if God Knew the Future, Such Knowledge Could Not Help Him
Many supporters of traditional foreknowledge may fear that abandoning this doctrine would disable God in some way, depriving Him of a crucial providential resource. Openists hold that this fear is baseless. “The doctrine of divine foreknowledge … is of no importance whatever for the religiously significant concerns about prayer, providence, and prophecy,” William Hasker writes.13 “If simple foreknowledge [of future contingents] did exist, it would be useless.”14 There is therefore no real theological cost to giving up the traditional doctrine.
Note that there would still be one significant theological advantage to exhaustive foreknowledge, even if Hasker and other openists are right about its uselessness: it would make God smarter, and thus greater, than if He lacked it. But leave this to one side.
If God already knows what's going to happen, how can He use that knowledge to enhance His providential oversight? He couldn't use it to change the future because the future would then be different than He “knew” it would be, and that's clearly impossible. So what good is it to know the future? Wouldn't such knowledge turn God into a passive spectator of events rather than an active agent?
To refute a universal claim, it is enough to provide a single counterinstance. Suppose then that a warrant will be issued next week for the arrest of an underground church leader. The classical God knows about this decision by the authorities well in advance; the openist God does not. It's also clear that this gives the classical God an advantage. This fact can be obscured if one looks for this advantage to God's ability to prevent the authorities from issuing the warrant in the first place. God can, of course, prevent the issuance of the warrant; but then His foreknowledge would not have included its issuance! Given that He foresaw the warrant being issued, He can't use that knowledge to bring it about that the warrant isn't issued. What He can do, however, is just what you would do if you knew about the arrest order in advance and wanted to help the underground church leader: He can warn him. And this is something the openist God, who lacks knowledge of future contingents, cannot do.
A number of questions can be raised about this case. For example, mightn't God's intervention and the church leader's response to it, coming before the foreknown event, interfere with it in some way? Answer: Whether it can, it won't. Since what God foreknows is what will in fact happen and is consequently the result of everything leading up to it, if God's warning to the underground leader results in his taking action that leads the authorities to act sooner, then their acting sooner is precisely what God foresaw in the first place. It remains the case that whatever God foresaw happening (and what He foresaw is what will in fact happen, given everything leading up to it), He can use that knowledge in ways not available to the openist God.
This is all I can do with the short space available. But I trust this is enough to convince the reader that the prospects for the providential usefulness of divine foreknowledge are more promising than they may have appeared.15
4. If God Knew the Future, the Problem of Evil Would Be Harder to Resolve
The problem of evil is tough enough as it is; we shouldn't take theological positions that make it even tougher. But that's just what the traditional doctrine of exhaustive foreknowledge does. It's harder to explain and justify evils God knows about in advance. John Sanders, for example, speaks movingly of the death of his brother in an auto accident and the false comfort of Christians who assured him that it was all part of God's providential plan. By denying divine foreknowledge, one can avoid making this problem even harder than it has to be.
A number of things are wrong with this move. For one thing, if this were an acceptable way to address the problem of evil, why stop with foreknowledge? Why not reduce God's power and goodness as well? Replace Jehovah with Zeus, and the problem of evil might disappear altogether.
More importantly, I don't believe that the problem of evil is made any easier by denying traditional foreknowledge. (It's an easier problem for Zeus, but that's for reasons other than Zeus's lack of exhaustive foreknowledge!) Many of the concrete cases that try believers' faith, like the death of Sanders's brother or the cancer death of my one-year-old nephew, are not explained by denying divine foreknowledge. The openist God, omniscient with respect to the past and present, had more than enough knowledge to save my nephew, but He didn't do so. Foreknowledge plays no role in generating this puzzle, and adding foreknowledge does not make the puzzle harder.
Why did God allow my nephew, or Sanders's brother, to die? I don't know, and I don't know anyone who does. Since I don't know what the reason is, I'm in no position to say that it's a reason that would be available on the assumption that God lacks exhaustive foreknowledge but unavailable on the assumption that God's omniscience encompasses future contingents. No open theist, to my knowledge, has even begun the task of showing this. Solving the problem of evil for the God of Open Theism is already so difficult that any solution, supposing one to exist, might turn out to be adequate for the God of Classical Theism as well—in which case evil would provide no ground for preferring the former over the latter.