Theistic critiques of atheism, part 4January 7, 2009 — Deacon Duncan
We’ve barely scratched the surface of Dr. William Lane Craig’s “Theistic Critiques of Atheism,” but we’re already seeing a pattern develop: in the rarified heights of philosophical theism, they dispense with any obligation to find God in the real world, and instead impress each other with the complexity and subtlety of the characteristics they can imagine that a perfect God might possess. To their credit, they do manage to think some deep thoughts, but without that connection to the real world (which they dismiss as “Verificationism”), they risk ending up like the fellow who grew so obsessed with fantasizing about the Perfect Lover that he lost all interest in real women.
The problem with philosophy is that, if you’re really good at it, you’re tempted to forget that genuine reality is a bit too vast and complex to be contained by a finite human mind. You forget that you are dealing with abstractions—details that have been separated from their real-world context so as to simplify the task of considering their characteristics. You forget that, by isolating the details from their context, you are working with facts that are, to some degree, false. You’ve broken their connection to the rest of reality, and thus interrupted the perfect self-consistency that is the characteristic of truth.
The solution to this particular philosophical problem is to continually refer back to the real world in your philosophizing—to verify your findings against the gold standard of reality itself. Non-Alethian theologians, however, cannot do this, since there is no real-world God for them to observe and check their speculations against. They end up like Craig, building relatively small, locally-coherent thought castles that are amazing for their intricacy and thoughtfulness but that, due to their disconnect with verifiable reality, fail to unite into a broader and more comprehensive coherence. Each sounds good on its own, and is more than enough to consume the entire attention of even the most intelligent individual, but when you try to put them all together and find an overall, real-world consistency, they have conflicts.
With omnipotence, Craig tried to show that atheistic critiques only made theology stronger by causing theologians to refine their concepts. True, they fixed the problems by reducing the scope of what “omnipotence” means, and their fixes only addressed the particular objections that have been raised so far. There’s no particular guarantee that new objections won’t be found, or that the scope of omnipotence won’t ultimately need to be reduced to the point that virtually anyone would qualify. But they did address the particular, specific objections being raised, and it wasn’t until we stepped back and considered that the new definition of “omnipotence” requires God to be able to sin that we noticed this rationalization was merely shifting the problems somewhere else.
The same sort of thing happens when Craig tries to defend omniscience, though in this case, instead of overtly reducing the scope of omniscience, he rationalizes away the problems of omniscience by making the definition of “to know” much broader and more vague.
On the standard account of omniscience, for any person S, S is omniscient if and only if S knows every true proposition and believes no false proposition. On this account God’s cognitive excellence is defined in terms of his propositional knowledge. Some persons have charged that omniscience so-defined is an inherently paradoxical notion, like the set of all truths. But the standard definition does not commit us to any sort of totality of all truths but merely to universal quantification with respect to truths: God knows every truth. Moreover, the standard definition does not purport to give us the mode of God’s knowledge but merely its scope and accuracy. Christian theologians have not typically thought of God’s knowledge as propositional in nature but as an undivided intuition of reality, which we finite knowers represent to ourselves in terms of propositions. We express propositionally what God knows non-propositionally. On this view there do not actually exist an infinite number of propositions but only as many propositions as human beings have cognized. Indeed, if one is a fictionalist with respect to abstract objects like propositions, then propositions are just useful fictions which we employ to describe people’s belief states, and the ground is swept from beneath any objections formulated on the basis of Platonistic assumptions concerning the reality of propositions. Finally, adequate definitions of divine omniscience are possible which make no mention of propositions at all. Charles Taliaferro proposes, for example, that omniscience be understood in terms of maximal cognitive power, towit, a person S is omniscient iff it is metaphysically impossible for there to be a being with greater cognitive power than S and this power is fully exercised.
This is an interesting argument, because by moving “knowledge” away from the concept of propositional statements and into the fuzzier area of “intuition,” Craig shifts us just that much closer to the argument that Alethea, and not Jehovah, is indeed the one true God. Knowledge, in this sense, is less of a verbal description of the truth, and more some sort of direct correspondence with the truth. This is a good parallel to our principle that the truth is consistent with itself, since it defines true knowledge in terms of direct consistency with the real world truth.
Speaking somewhat informally, we could propose that there is no more perfect consistency with the truth about X than to be the truth about X. It’s an idea with some theological precedent: there is no better way to know the truth than to be the Truth. But Jehovah cannot be the truth about everything, since that’s pantheism, which Christianity rejects. Alethea can be, and is, the truth about everything, therefore Her omniscience surpasses Jehovah’s, which according to Taliaferro’s definition means Jehovah God is not omniscient.
More than that, as we saw in Vox Day’s discussion of Richard Dawkins’ objection to omniscience, there’s a conflict that Craig’s “improved” definition of omniscience does not address: the only way God can know what our future choices will be is if those choices are already strictly determined. If there is any non-zero chance that any of us could make a choice that differs from what God knows we will choose, then God cannot truly know that we will make the choice He has foreseen. Given current conditions, therefore, all of our future choices must be strictly deterministic in order for God to be able to know infallibly what they will be.
There’s two ways our choices can be deterministic. One way is if we have no free will, but merely possess an illusion of free will due to our inability to perceive all the various causes and other factors that go into determining the only possible choice we can really make. If this is the case, then given any set of initial conditions (including the Big Bang, for example), the entire course of history from beginning to end is fixed and immutable, since our choices are strictly determined by factors outside of our control. This permits an infinitely wise being to anticipate exactly how people will act, as individuals and as groups, before they know themselves, and to be infallibly correct about it, since there is no chance anyone could behave differently than predicted by the mechanistic processes that control them.
The other possibility is if God exists outside of Time and perceives the whole of history as a unified progression corresponding to what Craig called the “hard past” in his discussion of omnipotence. Even if we think we have free will as regards our future choices, there is no question that once our choices become part of the hard past, we are no longer at liberty to change them. We can try and compensate for them, but the past is immutable, since we are looking back on what has already become part of the flow of history. But for a being who perceives all of history from the same vantage point, from beginning to end, it is all “hard past,” and therefore strictly determined, regardless of the perceptions of those bound inside the flow of time.
The reason this is a particular problem for theology is because that means God Himself has no power to behave differently than what He knows His own actions will be. However “omnipotent” He might be, He effectively has no choice at all about what He will do, since He already possesses the knowledge of how He must act throughout all of Time. An omnipotence that leaves God with no choice as to what He will do is hardly worth the effort of spelling out the word.
Additionally, let’s recall that Craig’s defense of the lack of evidence for God is to suppose that God knows how each person would react if given the chance to confront proof of His existence. This presupposes that human behavior is strictly determined, to the point that one can know the exact response given any given set of initial conditions. More to the point, if both human behavior and divine behavior are strictly determined, it makes the whole hypothetical situation meaningless, since there’s no such thing as “if God revealed Himself to unbelievers.” What is, is, end of story. There is no justice, there’s only what “just is.”
So we see the intricacy and subtlety of the philosopher’s snare. With a lot of hard work, deep thought, and a certain amount of genius, the dedicated rationalizer philosopher can construct an abstract, speculative definition that is locally coherent and that effectively addresses individual objections made by skeptics. Without “Verificationism,” however—without continually referring back to the much larger, real-world picture—he is only building on foundational premises that are inescapably flawed, because the process of abstraction has cut them off from the real world context that injects all those implications and consequences that make reality itself too big for the human mind to fully contain. As soon as you try to relate those fancy and even ingenious mental inventions back to the original real-world context, they end up conflicting with each other and with the facts. It’s a fatal and inescapable flaw that can only be addressed by continuous and persistent real-world verification.